Credit, Money, and Production: An Alternative Post-Keynesian Approach

Citation preview

1

’*\

~

"

,‘M

~

-

J!

pg“

U

N ‘ \ . MA W',_.;'J>‘ I‘: ‘ \ ‘\’I§€“y€*: ,,/',‘y"; ” M , “‘~I,~~~ ,,. ‘W3 v \‘\\( * 3 "H //\ ‘w~Mw:*;J1I,)i“\2‘ J 9% H ‘,,,‘./- j‘_,\,“w

;s¢.~=»‘»?l“éziw@?”“,'w"~ 1 \

\"Y\?‘-“v*\,;“;“,';,“.H‘; "n,;

2“ q»

- ‘

‘ (A.

,»i§ F~,;fW

" WI‘ * ’ \ ~ " ‘ -~H

Q

I

Credi , one and Production " .‘¢a;,‘.-,3 » ,» » ms,\1;=;

| ~\' L ".|w;:” _,l -ix;5'/l'_1‘1‘Y,_.“

‘Q »

An Alfie? A /M Q10, ' Post-Keyneslan Appro

. _

_i- I

‘ Iv: "~'I ‘

-!:v2r-W’

.,

I ;M‘.L

Credit, Money and Production This thought-provokingbook clearly and systematically analyzes the post-Keynesian approaches to endogenous money and, in doing so, provides an informed critique of the development of postKeynesian economics. Using a horizontalist perspective the author offers an historical overview of the post-Keynesian and circuit approaches to endogenous money, starting with a comprehensive survey of the FrancoItalian circuit school. He argues that rather than emphasizing the early writings of Minsky, Kaldor and Tobin in the 1950s, and of Davidson and Rousseas later, postKeynesians ought to have followed the writings of Joan Robinson and Richard Kahn who offered far better theories of credit-money. The author then compares the current post-Keynesian structuralist theory with New Keynesian monetary thought. In conclusion, he develops an innovative theory of banking based on Keynesian uncertainty and consistent with the horizontalist tradition taking into account credit restraints, crunches and creditworthiness. 4 This book will be illuminating to scholars of post-Keynesian economics, macroeconomics, and history of economic thought.

Credit, Money and Production

Credit, Money and Production An Alternative Post-Keynesian Approach

Louis-Philippe Rochon Stephen B. Monroe Assistant Professor ofEconomics and Banking Kalamazoo College, Michigan

Edward Elgar Chelte ham UK

Northampto

MA USA

l l l jl l l l l l l l l l l l l l l l l This

One

© Louis-Philippe Rochon, I999. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited

Glensanda House Montpellier Parade

Cheltenham Glos GL50 IUA UK Edward Elgar Publishing, Inc. I36 West Street Suite 202 Northampton

Massachusetts 01060 USA

A catalogue record for this book is available from the British Library

Library of Congress Cataloguing in Publication Data Rochon, Louis-Philippe. Credit, money, and production: an altemative post-Keynesian approach / Louis-Philippe Rochon, Stephen B. Monroe. l. Money. 2. Credit. 3. Keynesian economics. I. Monroe, Stephen B. ll. Title. HG22l.R637 I999 332.4-—dc2 I 99-27516 CIP

ISBN 1 85898 895 0 Printed and bound in Great Britain by Bookcraft (Bath) Ltd.

Contents List ofFigures Preface Acknowledgements

Introduction The Franco-Italian Circuitists: Credit, Money and Production Credit, Money and Post-Keynesian Theory: Clarifications of Familiar Themes

The Early Views of “Endogenous” Money: Minsky, Kaldor and Tobin The Early Views of “Endogenous” Money Revisited: Davidson and Rousseas versus Robinson and Kahn Horizontalists and Structuralists: Credit and Endogenous “Money” Post-Keynesians and Orthodoxy: “Neo” Post-Keynesians? New Keynesian Monetary Theory and the Transmission Mechanism: A Comparison with Post-Keynesian Theory A Post- Keynesian/Circuitist Theory of Banks: Uncertainty, Creditworthiness, and the Supply of Credit Bibliography ndex

V

Figures 1.1

The circulation of money in a production economy

3.1

Minsky's interest rate / velocity theorem

5.1

The horizontal credit supply curve

5.2

Horizontalism and a change in monetary policy

5.3

Minsky's two-price diagram

5.4

Kalecki’s static principle of increasing risk

5.5

The dynamic two-price Minsky diagram

7.1

The New Keynesian credit rationing model

vi

Preface The study of credit and money in post-Keynesian theory should not necessarily be a consensus building exercise. While convergences exist on many issues and at many levels, differences also arise, important in scope and far reaching in their conclusions and implications. Strangely enough,

both are a sign of strength. Convergence can be interpreted as a sign of paradigm building, and the differences as a sign of a mature body of thought, complex in its attempt to model the real world, successful in its

appeal. Post-Keynesian theory has ofien been described as negative, meaning that post-Keynesians know only what they are not. While this statement might have applied a few decades ago, it is still worth exploring in the sense that it

is worth reminding ourselves of what we are not: neoclassical. In this, there should be no apologies, and this alone should stand as a testimony to Keynes’s on-going struggle of escape. But where struggles exist, self-criticism is inevitable. Keynes showed to

what degree he was hard on himself, and blamed “the weight of the past” for the many mistakes in the General Theory. Criticism should be welcomed as a way of reminding ourselves of the paths we should avoid. lf at times I appear critical of post-Keynesian theory, as at times I am, it is only to remain within the confines of Keynes’s “revolution.” Criticism can take on two very different faces. Negative criticism is aimed at undermining an approach or a school of thought. Post-Keynesians’ criticism of neoclassical theory certainly falls within this category. Positive or constructive criticism, however, is quite diflerent. lts purpose is to question the foundations or stnicture of a given paradigm in an effort at

solidifying it. This book is within this tradition of critical thinking in hopes of steering debate toward making post-Keynesian theory more heterodox. To those who think that criticism has no place in post-Keynesian theory, as I have been told by some, I can only say that it is a sad day when selfexamination is considered wrong. To assume that post-Keynesians are united and that no differences exist is self-defeating. Differences, once again, should be celebrated as a sign of rigor, success, and growth.

vii

viii

Credit, Money and Production

This book is meant to be many things. Foremost, however, it is meant to be an update of Moore’s 1988 book, but also an extension to Lavoie’s 1992 book (chapter 4). It is also meant to be an altemative to Wray’s (I990) book, and to Wray’s (1999) most recent book on Chartalism or what may also be referred to as “funny monetarism.” This book aims at reestablishing post-Keynesian monetary theory on more proper, i.e. heterodox, ground. I believe post-Keynesian theory has strayed in the recent past, and post-

Keynesians must now choose their future course of research. I hope that readers will find its content interesting and useful.

Acknowledgments Many individuals need to be thanked for their various contributions to this dissertation. Such a work is not — and cannot be — the result of isolated existence. I am grateful for the numerous discussions and arguments that I have had with many individuals over an extended period of time. First, I would like to thank my doctoral supervisor at the New Schoolfor

Social Research, Edward Nell, for his continuous support and encouragement. His insights and keen observations were appreciated and, in many respects, followed. Where we disagreed, especially on the nature of the monetary circuit, his criticism was still very much appreciated. His

approach to macroeconomics and credit has allowed me to rethink many of the views I held before. The need to consider time, markets and transformations is evident throughout this dissertation. I would also like to thank in particular Marc Lavoie for his excellent

guidance and generous support over the last dozen years or more. I would especially like to thank him for having taken the time to read the content of

this book from cover to cover. He first introduced me to post-Keynesian economics, and readers will certainly recognize his influence within these pages. It is also through Marc that I was introduced to the theory of the monetary circuit. To him, I owe my biggest debt. My many discussions and lunches with Alain Parguez through the years proved to be of great inspiration. They also offered me the opportunity to clarify many nuances within the theory of the monetary circuit. Alain has the frustrating (l) ability to write only a few words that then translates into the reuniting of entire sections. I am very grateful, and cherish our fiiendship.

My exchanges with Augusto Graziani over the last few years proved not only insightful but also critical. Never refusing to comment on the many various drafts of my many chapters, his kind words of encouragement were welcomed during some discouraging times.

Basil Moore has, of course, played a pivotal role. Since early on, I found his work both stimulating and rich, and I acknowledge his apparent intellectual influence. From Keynes, tluough Robinson and Kaldor, Moore’s views revived the horizontalist tradition of post-Keynesian theory. I also have to acknowledge the generous comments and words of encouragement from a number of post-Keynesians at large who have commented on specific chapters: Riccardo Bellofiore, Henry Garretsen, ix

x

Credit, Money and Production

Peter Howells, Marcello Messori, Emesto Screpanti, Mario Seccareccia, John Smithin, and Bemard Vallageas. I must also thank Robin Rowley of Mcgill Univeristy for his cynicism, kindness, and friendship. My sojoum at the New School was also an enriching experience. While my arguments with John Eatwell helped me put post-Keynesian theory into perspective, those with Tom Palley helped me shape my interpretation of the structuralist approach. Also, I benefited from a healthy atmosphere of critical though, and from some enlightened exchanges with a number of students who contributed, albeit indirectly, to this dissertation: Garrett Bekker, Enrique Delamonica (with whom I wrote my first conference paper), Gonzalo Fonseca, Ray Majewski, Gerard McDonald. I would like to thank in particular my very good friend, Matias Vemengo, with whom I have had numerous exchanges. He is always a great inspiration. I must also acknowledge the encouragement and support I have received from a number of colleagues during the writing of this dissertation: Chidem Kurdas, Gary Mongiovi and Steve Pressman (who, as co-editors of ROPE, published my first paper), and Christine Rider. I would also like to express my gratitude to Tyrone Newhook for editing and proofreading this entire dissertation. A joumalist by trade, he has ofien expressed fiustration at the way academics write, claiming so very often

that we do not know how to do it right. He is perhaps correct. For this advice, I am gratefiil. Professors can always achieve more when they benefit from the help of their students. To this end, I would like to thank my research assistant, Donnita Fowlkes, for her tireless assistance. My only regret is that she is a senior and will be leaving next year. Her shoes will be very difficult to fill. Finally, I would like to thank my father. His tireless support in the face of what must have appeared to him as an endless joumey is the reason why I am here. In my own little world, to my eyes, he embodies truly the notion of an endogenous supply of money. My only regret is that my mother did not live long enough to see all this. This book would not have been possible had it not been for the generous financial support of many organizations. I would like to thank, in particular, the New School, and the Social Sciences and Humanities Research Council, in Canada.

Introduction While the notion of money endogeneity can be regarded as the “coping

stone” of post-Keynesian political economy, the increasing development and refmement of this view in the last decade has contributed to the emergence of a number of controversies within this broadly-defmed paradigm. For instance, post-Keynesians have debated for some time the appropriate role of liquidity preference within endogenous money, and the exogeneity of the base rate of interest. While these issues have been largely

resolved, differences nonetheless remain as to whether interest rates rise with economic activity (the slope of the money supply curve), and as to the appropriate role of the “demand for money.”

In general, the debates have created within post-Keynesian thought two distinct approaches which Pollin (1991) has labeled “accommodationist” and “structuralist,” although the former, following Moore (1988), is more

appropriately referred to as “horizontalist.” Interestingly enough, protagonists on each side claim that the other approach is a special case of each other (see Palley, I996; Lavoie, I996). According to Trautwein (I995), the differences among the various postKeynesian protagonists are important enough that it is debatable whether a summary of the main theoretical ideas within a single line of thought is possible. Similarly, Pollin (I991, p. 368) has argued that “the substantive

distinctions between the two approaches are clear and sufficiently important.” This view is not shared by all post-Keynesians. Some claim that the differences are rather a matter of emphasis than content. According to Lavoie (l995a, p. I), for instance, the controversies among postKeynesians “reflect diflerences of opinion or emphasis as what should be considered primary rather than secondary factors.” The view taken here is somewhere in between. That two different views exist is quite clear. And while they share much common ground, important differences also exist.

The differences are too often the result of interpretation and misinterpretation. There are overall three specific objectives within this book. The first is to clarify some of the concepts which will be used. In particular, there is a great need to shed some light on the differences between two very basic concepts in monetary theory: credit and money. In post-Keynesian theory,

I

2

Credit, Money and Production

it is not always clear what is being discussed, resulting in the many current controversies. The second objective is to trace the “post~Keynes” roots of endogenous money — that is, the contemporary revival of money endogeneity. It will be shown that many of the controversies in post-Keynesian monetary theory emanate from having followed Minsky and Davidson‘ too closely (and Keynes’s General Theory), rather than Robinson and Kahn. Finally, the third objective is to develop a truly heterodox theory of credit and money based on production, distribution and accumulation. Many have argued that in order to do so, it is necessary to cast aside Keynes, relying instead on the insights of Kalecki. It is argued here that this does not need to be the case. Although many parts of the General Theory must be forsaken, many insights into a heterodox theory of credit and money can be found in Keynes’s later writings and correspondence, especially in his Economic Journal articles written between I937 and I939. These can be pieced together to bring about a strong altemative to orthodox theories of money. Of course, there is still a need to tum to Kalecki whose views can be easily reconciled with those of Keynes. To achieve these objectives, this book is divided into eight main chapters. In chapter one, a theory of the monetary circuit is introduced where the emphasis is placed on credit flows as an explanation of endogenous money.

Keeping this chapter in mind, chapter two then develops the themes of credit and money by emphasizing their different nature and roles. It is argued that credit is not a component of money, and the demand for credit camiot be included within the demand for money. Money enters the economy, not arbitrarily, but through production and accumulation, through credit. This differentiation lends itself well to the discussion of endogenous money. If post-Keynesians generally reject the exogeneity of money (although many recognize that a portion of the money supply is exogenous), they do not always agree on how money becomes endogenous. For this purpose, two definitions of endogeneity are proposed. First, the “revolutionary” theory of endogenous money follows the postGeneral Theory Keynes closely. Here, money is endogenous because of the “needs of trade.” Money is created ex nihilo from credit extended by banks to cover production and investment costs. Second, money can also be said to be endogenous in the “portfolio” sense. This approach emphasizes the creation of money as a result of shifts in portfolio preferences and the changing value of the velocity of money. The money supply, in the traditional sense, is still exogenous. But as households exchange assets for money, money is created over and above the preferences of the central bank. Given these two definitions, only the former definition is consistent with Keynes’s “monetary theory of production.”

Introduction

3

Finally, chapter two also stresses the similarities and differences between the horizontalist and structuralist positions.

It is shown that post-

Keynesians agree on many important variables. Whether this suggests that the two positions are necessarily irreconcilable is a question that will remain

open throughout the book. With these two defmitions of endogenous money in hand, chapters three and four proceed to analyze the roots of post-Keynesian monetary theory. The starting point is post-Keynes as opposed to pre-Keynes. The reason for this is two-fold. First, many post-Keynesians have already written on the

similarities between the current post-Keynesian theory and the Banking school. There is therefore no need to repeat this here. Second, and perhaps

more important, the revival of the theory of endogenous money following the publication of the General Theory is not given to close scrutiny. PostKeynesians have not taken the time to closely examine their own early writings, for had they, perhaps many of the current controversies would not

have arisen. Chapter three begins this analysis into the roots of money endogeneity by looking at the early works of Hyman Minsky (l957a, l957b), Nicholas Kaldor (I939, I955, I958, I960, I964) and James Tobin (I960, I963, I965, 1965a, I968). Surprisingly, all three economists shared a similar

view of the financial workings of a modem economy. To claim, however, that these authors had developed a theory of endogenous money is certainly an exaggeration? Rather, it is argued that all three authors accepted many of the orthodox arguments of the Quantity Theory of Money, such as the exogeneity of the money supply, the causality between reserves and money,

and between money and production. These authors adopted a theory of “portfolio” endogeneity although, as argued above, this cannot be deemed a

theory of endogenous money in Keynes’s sense. In particular, it is argued that their views were not a rejection of Friedman, but rather a modification of his views where the velocity of money is allowed to vary. In Minsky’s case, the argument is made that he still argued in tenns of loanable funds, a fiamework within which he developed his views on fmancial fragility.

Chapter four continues this line of reasoning by exploring the early works of two other post-Keynesians: Paul Davidson (I965) and Stephen Rousseas (I960). The conclusions are the same as in the previous chapter. Irrespective of their later contributions, the early works of Davidson and

Rousseas are not consistent with a theory of money endogeneity in the “revolutionary” sense. The same chapter then looks at the early works of Joan Robinson (1933, I951, I952, I956, I961), in particular her Accumulation of Capital, and at Richard Kalm’s Memorandum to the Radcliffe Committee in I957. It is claimed that each author offered interesting insights not only into a

4

Credit, Money and Production

revolutionary theory of endogenous money but also into a theory of the monetary circuit. Robinson and Kahn wrote at the same time as Minsky, Kaldor (albeit the early Kaldor) and Davidson, but their work on money was largely ignored by American post-Keynesians. This is unfortunate since it is argued that the current state of post-Keynesian theory would be less ambiguous and confusing had post~Keynesians followed the writings of Robinson and Kahn rather than those of Minsky, Kaldor and Davidson. In this sense, Robinson and Kahn ought to be considered the founders of the modem revival of endogenous money. Having examined more closely the early beginnings of the post-Keynesian theory of endogenous money, the debate then moves forward to more contemporary settings. In many respects, the many differences between Minsky and Davidson, on the one hand, and Robinson and Kahn, on the other — with Kaldor having been “converted” to the latter group later on in his career — are at the center at the current debates and controversies between the structuralists and the horizontalists. Chapter five examines this claim in greater detail. Although the idea of an exogenously detennined base rate of interest appears now to be accepted by most post-Keynesians, differences still

remain as to the detennination of the loan rate of interest. It is argued here, following Deriet and Seccareccia (I996), that both post-Keynesian camps support the mark-up approach. Although for horizontalists the mark-up varies independently of economic

activity, structuralists argue that liquidity preference influences the premium. In essence, smrcturalists argue that, as economic activity increases, so does the rate of interest given decreases in the liquidity positions of banks and borrowers. The proper graphical representation is therefore an upward sloping money supply curve in money-interest space. Chapter six carries the discussion of the previous chapter further. Many of the differences between the structuralists and the horizontalists emanate from the structuralists’ reliance on orthodox variables, four in particular: the role of uncertainty in explaining the existence of money, the reliance on changes in the velocity of money, the view of money as a stock rather than a flow, and the emphasis on the lack of reserves as determining the quantity of bank loans. It is argued that the inclusion of these more orthodox variables within the structuralist — or neo-post-Keynesian — tradition are a result of the influence of the early post-Keynesians who had not fully emancipated themselves from orthodox monetary thought. It is argued that, in some respects, the structuralist position is reduced to the theory of the money multiplier where reserves (or the lack of reserves) determine the supply of loans. The

Introduction

5

structuralist position can best be described as a money multiplier model in

which a loans market has been inserted. The discussion over the more orthodox elements of post-Keynesian monetary theory lends itself to a comparison with another “Keynesian,” orthodox approach: the New Keynesians. Chapter seven does just this. It begins first with a sormd description of the New Keynesian theory of

money. Despite some claims to the contrary, even within post-Keynesian circles, credit rationing and asymmetric information are not crucial to New

Keynesian theory.

Rather, three key assumptions are made.

New

Keynesian theory of money relies first on the notion that credit is supplydriven, dictated largely by the policy of the central bank. Second, the substitution between loans and other publicly issued (open-market) bonds must be imperfect. Finally, prices are taken as downward-sticky, otherwise a change in nominal reserves initiated by the central bank could be compensated by an equal change in prices, thereby leaving unaffected, in real terms, the banks’ and firms’ balance sheets. In this chapter it is argued that, despite first appearances, New Keynesian and post-Keynesian theories are incompatible. The New Keynesian theory of money is set well within the orthodox tradition. Whereas the money

supply may be credit-driven, the supply of credit is determined by the banks’ liabilities. If some post-Keynesians — in particular the structuralists - argue that they are compatible, it is because their own model at times borders on orthodoxy. This claim is examined in full. Finally, the chapter on New Keynesian theory ends with a comparison with Minsky’s (and Davidson’s) early works, as explored in chapter three. Here it is shown that, given some recent debates in New Keynesian theory, it can be claimed that New Keynesians have, so to speak, rediscovered Minsky forty years later. In fact, the parallels are striking. Having spent a considerable amount of time criticizing post-Keynesian theory, chapter eight spends a great amount of time rebuilding it along more heterodox ground, i.e. without reference to Marshallian or orthodox notions. In order to build such an altemative, it is necessary to abandon the confmes

of the General Theory and tum to Keynes’s later writings on the “fmance motive.” In the Economic Journal, Keynes devotes a number of articles to the topic. It is argued here that, contrary to the prevailing wisdom (see, for instance, Dow, 1997), Keynes’s frnarrce motive articles were written, not as a defense of the General Theory, but rather as a way of leaving it behind

and starting anew. The fu'st part of this exercise was started in the first chapter. The story continues in chapter eight, and takes off where chapter

one ends. In chapter one, the Franco-Italian school of monetary circulation is emphasized. This is done for two reasons. First, because of their close

6

Credit, Money and Production

relation to post-Keynesian theory, and second, because of their proven ability to build a theory of endogenous money, in Keynes’s sense, without reference to orthodoxy. A theory of the monetary circuit is described where money is created by bank credit and destroyed at the end of the circuit through, principally, the repayment of loans. Chapter eight seeks to continue

this work by introducing a post-Keynesian theory of the bank, consistent with Keynes’s analysis of uncertainty and the monetary circuit. PostKeynesians and circuitists alike have paid little attention to the decisionmaking of banks. Although post-Keynesians have devoted more attention to banks recently, this is under the auspices of liquidity preference, not uncertainty. Finally, the chapter ends with a discussion of liquidity preference and the role of the demand for money. It is argued that post-Keynesians are attempting to occupy two extreme positions of the circuit, the beginning and the end, thereby confusing the logical sequence of events. Only when a carefully understood sequential circuit of money is introduced, and only

when credit and money are carefully differentiated, can these issues be properly resolved. The subject of this book is the theory of endogenous money. It is not

about Keynes, although his views will be featured prominently. As such, there is no specific chapter devoted to his views. Rather, they will be presented, discussed and analyzed throughout.

Many have argued that Kalecki is a better starting point than Keynes in developing a theory representative of a true “monetary economy of production.” For instance, Sebastiani (I989, p. xi) has argued that “the widespread rediscovery of Kalecki which we are witnessing today is certainly due, at least in part, to the growing dissatisfaction with Keynes’s theory.” Similarly, Sawyer (I995, p. I48) writes, “Kalecki provides not only a more realistic starting-point for the development of an altemative macroeconomics but fits in well with many widely-accepted features of developed capitalist economies.” For this reason, Sawyer calls for the postKeynesian approach to be rebaptized “post-Kaleckian” in light of the fact that many post-Keynesians “owe more to Kalecki than Keynes.” Although at first glance this appears to be a reasonable argument, one realizes that much of the criticism of Keynes is aimed at his General Theory. While it is argued here that the General Theory represents a considerable obstacle at developing a theory of money, once we leave its confines, one can find a radical Keynes, both before and after the General Theory. Keynes’s Economic Journal articles, for instance, offer us a

window into a new world, quite different than the one presented in the General Theory.

In fact, outside the General Theory, Keynes’s and

Kalecki’s views can be reconciled without great difficulty. Albeit Kalecki’s

Introduction

7

views are clearer — especially since he did not carry with him the burden of orthodoxy — it is nonetheless possible to construct from Keynes a “postKaleckian” theory of macroeconomics and money. This said, Kalecki’s views are presented as well, especially in chapter one

where an attempt is made to construct a theory of endogenous money pinged from the more orthodox elements of post-Keynesian theory. For this, tuming to Kalecki seems to be an obvious choice, not for his views on money — which remained largely underdeveloped — but for his ability at

combining money with production, accumulation and distribution. All references to Keynes in this work are done with respect to the Collected Writings. For instance, Keynes (xiv, p. 201) refers to a passage on page 201 of the fourteenth volume. This makes the reading of the manuscript clearer and also allows readers to recognize immediately the

reference.

NOTES I.

2.

Some post-Keynesians have criticized my discussion of Minsky and Davidson. While my purpose is not to minimize or trivialize their contribution, I also believe it is necessary to understand their early work and to ask whether it was consistent with endogenous money. Some structuralist post-Keynesians have criticized me for my criticism of Minsky. Strangely enough, none have seen my criticism of Kaldor as problematic.

1. The Franco-Italian Circuitists: Credit, Money and Production The objective of the present chapter is to develop a theory of endogenous money based on the circuit approach where bank credit takes center stage. Not only is money created ex nihilo, but endogenous money also becomes a theory of accumulation and of income distribution. The obvious analytical point of departure is Keynes, although it is also necessary to expand the horizons and to recognize the limitations of Keynes’s analysis. While the purpose is to “build upon the shoulders of Keynes” and to pursue his “revolution,” post-Keynesians should also not hesitate to purge from his monetary analysis those elements still trapped within the confines of “classical” economics. In doing so, much of the current post-Keynesian tradition, which follows Keynes too closely, will also fall under scrutiny. But to pursue Keynes’s revolution, it is necessary to not only build upon his shoulders, but in some respect to go beyond Keynes himself, or at least Keynes of the General Theory.' Only then can the existing void be filled. In rebuilding a theory of credit and money appropriate for the real world, that is for a monetized economy of production, an obvious place to begin is with the important work of Kalecki, in particular his theory of distribution. Only when the insights of Keynes and Kalecki are successfully integrated can we obtain a theory of money useful in the analysis of a monetized economy of production. Circuitists,2 as this chapter will show, do precisely this, by integrating the creation of money with a theory of production, distribution and accumulation in a modem theory of a monetized economy. The circuit can be described, as Schmitt (I996, p. I32) argues, as being “founded on the hypothesis of circulating income.” The monetary circuit will show how money is created, how it is circulated and how it is finally destroyed at which point the circuit will close and the stage will be set for another circuit to begin. This analysis will show why the theory of the monetary circuit is also called a theory of the “dynamic circuit,"3 or, as Lavoie (I987, p. 91) calls it, a theory of the “dynamic history of the production process." In this respect, the theory of the monetary circuit is not only a theory of the creation of money, but is also a theory of distribution and accumulation. As Graziani (1996, p. 4) explains, “The very fonnation of profits is explained by the 8

The Franco-Italian Circuitists

9

presence of money, as well as by the way in which money is created and introduced into the market circuit.” 4

KEYNES’S POST-GENERAL THEORY ARTICLES AND THE FATE OF THE GENERAL THEORY From the Treatise of Money to his articles in the Economic Journal (I937 -1939), passing through the General Theory, Keynes’s intellectual development and his attempt at building the foundations of a monetary theory of production were a slow progression toward recognizing credit-

money as the central institution of modem capitalism. The fact that Keynes had not frilly succeeded in completing his “long

struggle of escape” with the publication of the General Theory should not lead by now to many stares of incomprehension. Much has already been

said about Keynes’s acceptance of Marshallian microeconomics (Parrinello, 1985), and of the “vestigial traces” of orthodox thought (Rousseas, 1992, p. 69). In addition, much ink has been poured in discussing Keynes’s decision to accept an exogenous money supply. It is for these reasons that Parguez (I985, p. 258) claims that the General Theory “is not a coherent whole, a

completed cathedral. It shows us a new world, but gives us only a key. Keynes could never fully rid himself of the weight of academic conventions.” The debate over whether Keynes assumed an exogenous supply of money in the General Theory appears to be flaring up once again. Recently, Dow (I997) argued that Keynes was well aware of money endogeneity and in

fact implicitly assumed the money supply to be so in the General Theory. There, Keynes assumed, according to Dow, a given supply of money — not an exogenous one. The problem with this argument is that given the fact that the General Theory studies — in Keynes’s own words — the “forces which determine changes in the scale of output and employment as a whole,” it is difficult to see how the “supply of money” can be given, since production, output, credit and money are linked in post-Keynesian theory. An increase in production or output will require additional finance thereby increasing

endogenously the money supply. Hence, it can not be given.’ If that is the case, then production must also be given. It is perhaps better to agree with

Foster’s (I986) argument, that the theory developed in the General Theory requires a theory of endogenous money. In considering Keynes’s “long struggle of escape fiom the habitual modes of thought and expression,” post-Keynesians must make the assumption that this struggle was an ongoing process which did not end with the publication of the General Theory, especially with respect to

I0

Credit, Money and Production

Keynes’s views on money which were subjected to much criticism. Rather, the General Theory should be seen as one of many elements in a continuing sequence of events which would eventually lead Keynes to reject orthodox theory in its quasi-entirety. Hence, as an inspiration for developing a theory

of money along Keynesian lines, Keynes’s post-General Theory articles offer a better starting point, a fresher beginning.‘ This argument is not

without controversy, and is certainly one of the many points of contention between circuitists and post-Keynesians. Post-Keynesians generally interpret Keynes’s Economic Journal articles

as a clarification of his views developed in the General Theory. The attempt is to reconcile his position developed alter the General Theory with those developed within it. This position is consistent with that of Dow (I997, p. 65) who claims that “none of this interpretation of Keynes’s [General Theory] view is affected in any way by the developments of I937.” Similarly, Cottrell (I994, p. 599) argues that the on-going debate between horizontalists and structuralists extends to Keynes’s I937-I939 papers: “This conflict between the radical endogeneity theory and Keynes’s ideas

extends to his post-General Theory writings.” As Kaldor (1982) observes, however, Keynes’s monetary analysis in the General Theory is still in the “monetarist” tradition, thereby amounting to a modification rather than to the rejection of the Quantity Theory. The position defended here is that Keynes appears to have been aware of this,

and set upon a course to change all that. In this light, his post-General Theory articles therefore cannot be interpreted as his attempts at clarifying his views in the General Theory, but rather as attempts by Keynes to

distance himself fiom certain views which he so carefully developed in the General Theory, especially those on credit and money. Keynes’s views on credit, money and commercial banks alter the General Theory represent a clear and radical departure from the General Theory, to the point of even

contradicting his views developed in the General Theory. Although he did not always succeed, it is argued here that the post-General Theory articles should be interpreted as a progression in Keynes’s “moments of illumination,” and not as a clarification of his thoughts. They do not complement the General Theory but rather replace it. In his post-General Theory articles, Keynes developed a theory of endogenous credit-money. In Keynes’s post-General Theory articles, therefore, we find the necessary ingredients for developing a proper theory of money endogeneity, minus the

more ambiguous arguments of the General Theory, such as liquidity preference, the velocity of money, and the emphasis on stock analysis. The General Theory can be retained certainly, but rather for the insights Keynes offered into other areas, such as for instance the theory of effective demand. Viewed in this light, there should be no attempt by post-Keynesians to interpret them within the Marshallian framework of the General Theory.

The Franco-Italian Circuitists

Il

They are not the “third edition” of the General Theory — as Shackle (1967, p. I36) claims - but rather a whole new book, with new insights and approaches for a theory of a capitalist economy of production, and for a

theory of money. As Graziani (I984) argues, the finance motive which Keynes developed in the Economic Journal articles invalidates the General

Theory in many respects. This is also the position defended here: Keynes’s post-General Theory articles ultimately culminate in the rejection of Keynes’s views on velocity and liquidity preference which he developed in the General Theory. As Rizzo (I992, p. I25) and Graziani (1985, p. I60) have claimed, it is the General Theory itself which should be interpreted in light of Keynes’s Economic Journal articles.

Contrary to Moggridge (see volume vii, p. xviii), who claims that “how [Keynes] would have revised the General Theory if he had remained in good health is impossible to guess,” it is argued that Keynes nonetheless lefi us important clues on how to piece together a coherent theory of credit and money purged of the more orthodox elements of the General Theory. Hence, Keynes’s “leaps of intuition” (Harcourt and Sardoni, I996, p. 6) were an effort to abandon the confines of the General Theory in favor of a more generalized theory of credit and endogenous money.

Hicks (I989, p. I02) argued that “if Marshall was put out of date by Keynes, has not much the same thing happened to Keynes’s own theory [of the rate of interest], at least in the form he gave it in his most famous book.” The answer to Hicks’ question is “yes” - but it was “put out of date” by Keynes himself in his post-General Theory articles where he defends, as we will see, a theory of endogenous money where the rate of interest is exogenous. This position stands in obvious contrast to the position defended by many post-Keynesians. For instance, Dow’s (I997, p. 62) statement to the effect that “had Keynes written a second edition of the General Theory in I938 or

I939, it is not clear that he would have expressed himself very differently with respect to the money supply.”

Yet there is ample evidence to support the claim defended here, which can be divided largely into two large categories. First, it is known that Keynes himself was not entirely happy with the General Theory. Second, Keynes’s

correspondence with Hugh Townshend, ignored by most post-Keynesians, offers interesting insights into Keynes’s transition ti-om the General Theory to the Economic Journal articles. These points are addressed below. There is increasing evidence that Keynes grew tired of the General

Theory. Once published, he engaged in numerous discussions and debates with many of his critics and supporters, as evidenced by the publication of

the correspondence in volumes xiv and xxix of the Collected Writings. These debates gave Keynes the opportunity of clearing some air, and helped him ir1 exposing some of the weaknesses of the General Theory. As time

I2

Credit, Money and Production

wore on, Keynes appeared to become increasingly dissatisfied with the General Theory as a whole, not as much with its central message of

eflective demand, but with some of the other arguments related to the theory of the rate of interest or the supply of money. Just as Keynes showed some remorse over the Treatise, he too showed

some remorse over the General Theory. For instance, we know he intended to produce “footnotes” — “in the course of the next year or so” — to help him in “re-writing and re-casting” the General Theory. In a letter to Townshend (April 23, I936; see xxix, p. 246), Keynes admits to wanting to “revise the

book properly.” In a letter to Joan Robinson, Keynes (xxix, p. I85) goes further and writes

that “I am trying to prevent my mind from crystallizing too much on the precise lines of the General Theory.”

Later, Keynes would also tell

Robinson that he felt himself “gradually getting

into an outside position

towards the book, and am feeling my way to new lines of exposition.”7 Moggridge (see volume vii, p. xvii) also reaches this conclusion, suggesting that “as the debate progressed, his own ideas were changing.” Finally, in the preface to the German edition of the General Theory, Keynes (vii, p. xxv; emphasis added) claims that he regarded the General Theory as a “transition away from the English classical (or orthodox)

tradition,” thereby implying that it is part of an on-going development which did not end with the publication of the General Theory.

This

suggests that just as the General Theory was meant to “extend and correct” the Treatise, Keynes also perceived his post-General Theory articles in the same way. Keynes’s dissatisfaction with the General Theory comes from his

continued “struggle of escape.” As a “transition,” the General Theory was still subject to changes, both minor and drastic. This suggests that Keynes still saw the General Theory as rooted partly in the orthodox tradition, and

decided to revise those parts with which he was dissatisfied. Keynes acknowledged that the General Theory was partly still trapped in orthodoxy. He wrote in the preface to the French edition of the General Theory, that “in that orthodoxy, in that continuous transition, I was brought up. I leamt it, I taught it, I wrote it” (vii, p. xxxi). He then adds that they

are undoubtedly responsible for “certain faults in the _book.” The second point made here is that in some conespondence with Hugh Townshend, Keynes has given us some interesting insights into the transition from the General Theory to the Economic Journal articles. In

this correspondence, Keynes is exposed to a number of arguments which would later retum in his post-General Theory articles. Whether Townshend had any influence on Keynes in helping him develop his post-General Theory ideas is a difficult statement to prove. However, Keynes makes abundantly clear that the discussion he had with Townshend was definitely

The Franco-Italian Circuitists

13

insightfirl, claiming on one occasion (xiv, p. 239) that “once more you have

shown a complete comprehension of what I am driving at.” With respect to money and monetary theory, there are a number of arguments made by Townshend following the General Theory which would later appear in some fonn in Keynes’s Economic Journal articles. For instance, we know that in the General Theory, Keynes essentially adopted the quantitative equation but assumed that velocity of money was rmstable. This notion is absent fi'om his later writings. In fact, the velocity of money plays no role in the finance articles. Coincidentally, on March 26, 1936, Townshend wrote Keynes a long letter (xiv, pp. 239-45) where the author claims that “the equation is thus a statement of a kind of theoretical bookkeeping; significant as an identity, though of no economic importance.” Moreover, Townshend specifically criticizes the quantity equation and the use of velocity for their reliance on stock principles: “The generalised quantity equation does not seem to have any relation to the flow of cash in the industrial circulation (or to changes in the rate of the flow) as a whole. Thus the generalised quantity theory does not apparently have

any positive hearing on monetary policy in the ordinary sense (i.e. roughly, central banking). But I think that perhaps you will agree with this.”

Keynes’s reply to Townshend is interesting. Keynes (xiv, pp. 246-7) writes back claiming that the theory of the velocity of money “amounts to very little, contributes nothing to the understanding of the argument and is simply encouraging the reader to waste his time in a rather futile sort of

way.” A far cry from his views in the General Theory. Keynes associates the use of velocity with his failure to frilly escape neoclassical theory: “I am conscious that this [velocity], like a good deal

else in the book, is largely the product of the old associations of my mind, the result of always trying to see the new theory in its relation to the old one

and to discover more affmities than really exist. When one has entirely sloughed off the old, one no longer feels the need of all that.” The problem in recasting velocity in tenns of a “positive doctrine” is that it would be “more difiicult to clean the thing up except by rather drastic changes. And here again the trouble really arises from my trying to produce a closer analogy between my terms and those previously employed than the circumstances really justify. It would become so tortuous and complicated. Here again the right solution probably lies in simply cutting it all out.” In the post-General Theory articles, Keynes did just that. The velocity of money as a casual argument in his theory of money is absent. If money is

made endogenous, the argument does not rely on the velocity of money,‘ but rather on the financial and industrial needs of production. This point alone should convince many of the merits of the post-General Theory

I4

Credit, Money and Production

articles, and how they represent a better starting point than the General Theory. The position taken here is that for post-Keynesians to develop a theory of

endogenous money, the confines of the General Theory ought to be abandoned, in favor of Keynes’s Economic Journal articles, written between I937 and I939, on fmance and interest rates. 9

CREDIT, MONEY, AND PRODUCTION: THE CIRCUIT APPROACH The dissatisfaction with Keynes’s treatment of credit and money in the General Theory has led some economists to search openly for an “altemative Keynesian approach” to monetary theory. Unlike Minsky’s, KaIdor’s, Davidson’s, and Rousseas’s attempts at building a heterodox approach to monetary theory based on changes in the velocity of money and portfolio changes (see the following chapter), a group of mostly French economists — and later a closely related Italian school would emerge — while Keynesian in every sense of the word, rejected much of the teachings of the General Theory, mostly for reasons already explored. Instead, they turned to Keynes’s post-General Theory articles, as well as to the writings of Kalecki. '°

This does not imply that circulationists and post-Keynesians are necessarily incompatible bedfellows. Rather, circulationists have simply developed a different facet of Keynes, albeit a more radical one. As Messori (I985, p. 216) explains, “Keynes’s (I936) theory is unable to

analyze this part of the monetary circuit since it neglects the financing of production and the role played by banks.” Yet, these views are developed

by Keynes in his Economic Journal articles. With the emphasis placed elsewhere, circuitists were largely shielded from the continental influence

of Friedman and the monetarists. As Nell and Deleplace (I996, p. I0) have argued, “The emphasis put on the conditions of money creation protected French economists for long from the debate between monetarists and standard Keynesians.” Rather than seeing money based on the demand for money balances, circuitists place the emphasis on the creation and the ultimate destruction of “money,” where proceeds (and profits) are generated in order for fnms to pay back debt. In fact, money is created because of finns’ decision to enter into debt. As Schrnitt (I975, p. I60; see also Maricic, I985) says, money is “debt which circulates freely.” Seccareccia (I997, p. 3) explains, “Much like the Sisyphean metaphor in Greek mythology, business enterprises are perpetually engaged during discrete intervals in a circular process of acquiring or extinguishing their debt vis-a-vis the banking system.” Money

The Franco-Italian Circuitists

15

appears in the system only once production has been financed by bank credit. While the creation of bank credit and deposits (money) is simultaneous (Graziani, I990b, p. 52), credit and money are nonetheless

seen as different entities, each playing a different, yet specific, role in the circuit. There is hence no need to extend the definition of money to include other instruments. As Graziani (1996, p. I41) argues, “The extension of the category of money to a multitude of liquid or semi-liquid resources is precluded.” The economy is therefore one where credit dominates. In this sense, circuitists offer a “coherent and logical organization of the

circulation of money” (Barrere, I990, p. ll). This implies naturally that “Keynes’ finance motive has instead little to do with the transactions

motive” (Graziani, I984, p. 485). It is a separate motive altogether, where credit creates money.

The emphasis is placed on a flow approach to credit, and not on portfolio decisions. In this respect, credit and money become endogenous not because of any single decision of households or of the central bank, but rather because of the needs of production. As DeVroey (I984, p. 383)

writes, money is a “social reality within the system: a non-commodity in a universe of commodities.” Money therefore enters with production; but also only once credit has been used. As Graziani (I985, p. I64) writes, “If a bank’s finance, while

guaranteed, has not yet been used, the corresponding liquidity has not yet been created, and it is not possible to talk of the existence of liquidity.” Circuitists generally accept this view. For instance, Mondello (I985, p. 387) writes that the central question becomes “the penetration of bank money in a capitalist economy of production.” Similarly, Goux (1987, p. 601) claims that “fmance becomes a monetary advance necessary for finns to distribute incomes as production begins.” Lavoie (I984, p. 773) holds a similar view: “The integration of money in the economic system must not be done when output is already specified, as in the exchange economy of general equilibrium models or even in models a la Clower-Leijorrlrufvud but rather must be introduced as part of the production process.” While velocity and financial innovations may play a role, these would have to be considered in the ex post and are not causal in the determination of credit availability. They become secondary issues. Lavoie (I984, p. 778) has argued this last point claiming that “money is in some way endogenous whether central banks are dynamic or not.” Lavoie (I996a, p. 533) later retums to this theme, arguing that “accommodation or the lack of it, liability or the lack of it, and financial innovations or the lack of it are second-order phenomena to the crucial causal story that goes fiom

debt to the supply of means of payment.” A good collection of the circuit views can be found in Nell and Deleplace (I996).

I6

Credit, Money and Production

Since first developed by Schmitt (1966, 1971, 1975, 1984) and Parguez (1975, 1984), the theory of the dynamic circuit of money has found many adherents, notably in France with Arena (1985), Barrére (1979), Mondello (1985), Poulon (1980, 1982) and Zerbato (1987, 1989), as well as with some proponents of the Regulation school, for instance, Aglietta (1970) and Di Ruzza ( 1984). Outside France, there are many adherents as well, for instance, Cencini (1984, 1988). More recently, a closely related group has emerged in Italy which can claim to have strong links to the French school. Under the guidance of Graziani (1984, 1987, 1990a, l990b, 1991, 1996), the Italian school consists of many converts, including Messori (1985,

1936, 1991), Bellofiore (1983), and Screpanti (199s).“

Circuitists, however, are not all united in their approach to the monetary circuit. This is largely the result of various influences within the school. Though Keynes of the Treatise and the post-General Theorjy is certainly central to the circuitist arguments, others play an equally important role: for instance, Marx’s reproduction schemes, Kalecki’s investment-profit relationship, Schumpeter’s work on the entrepreneurial role of commercial banks in production and credit decision, and Wicksell’s pure-credit economies,” as well as Keynes’s finance motive, all figure prominently within the circulationist model. But these various influences also lead to divergences of opinions. Indeed, Nell and Deleplace (1996, p. I0) have argued that the various sources of influence explain why there are “extremes” within the circulation approach, and as such, a unified approach remains elusive: “The diversity of sources explains why the circulation approach does not present a unified front, an integrated school.” Despite these differences, circulationists share a great number of points of view. For instance, they all see money as being endogenous through the various phases of capitalist economies, and not tied to specific periods, as in post-Keynesian theory. This is because money does not rest on a specific institution, but on the logical process of production and finance. Hence, money is not the result of inherent uncertainty as in post-Keynesian theories. Through their supply of credit, banks always create the equivalent quantity of money as purchasing power. The circuitist argument can be traced back to Keynes’s Treatise on Money, although Keynes’s post-General Theory articles on the fmance motive may be a better starting point (Graziani, 1984). Many circulationists share Kaldor’s and Moore’s horizontalist argument," although this is seen as a “second-best solution.” References to horizontalism or to a money supply schedule are made only in order to remain within the elegant framework of what Lavoie (I996a, p. 4) calls “standard normalcy.” In fact, many circulationists — by positing that supply is demand (see Parguez, I985, 273) — refuse to even speak of money supply (see Di Ruzza, I984). 4 As Lavoie (1987, p. 91) says, “For many circuitists, the notions of

The Franco-Italian Circuitists

I7

supply and demand (of money) are as such inexistant.” The notion of a supply of money somehow conjures up the notion of a function, thereby suggesting a relationship between the supply of credit and the rate of

interest. Moreover, the separation of the economy into a “real” and a “monetary” sector is seen as unrealistic: in circuit theory, they are integrated, and cannot be separated, even if the objective is to simplify the “real world.”'5 Perhaps the best way to “draw” this supply is to represent it by a single point — a dot so to speak — where the quantity supplied will exactly equal

the quantity demanded. Any position off this “point” makes little sense. There is no “supply” of credit outside the quantity of credit “desired” by

firms. For instance, seeing the notion of a supply of exogenous money as an “a-economic” phenomenon, Di Ruzza (1984, pp. 6-8) writes: It appears to me that the use of supply and demand analysis with respect to money gives rise to more inconveniences than to advantages. To consider supply and demand functions for money, irrespective of their characteristics,

implies foremost the definition and acceptance of a certain separation between money and the “real economy”. If we assume that the supply of money is endogenous, and detemiined by demand, the notion of supply itself loses all

its significance. The only quantity of money foreseeable, possible and normal, would be the “desired” quantity, i.e. those arising from the intemal economic needs of the system.

Circuitists therefore reject the very notion of a “money market”, where a price and a quantity are detennined. Rather, the rate of interest is determined exogenously, by the central bank, outside of the confines of a money market (see below). The central bank sets the real rate of interest (Pivetti, I988; Panico, I988). As for the central bank, its role is to guarantee the stability of the financial

market. The relationship goes from the exogenous rate of interest, to the demand for credit, to its supply by commercial banks, back to the central bank and the supply of reserves in order to keep the system functioning well. But reserves are not emphasized in circuit theory since they are seen

as a notion left over from non-credit economies. In this sense, the role of the central bank is not limited to supplying reserves to prevent a crisis, but

is an integral and permanent component of the circuit. It does not play according to the “fireperson” rules — jumping in whenever the system is

near collapse. The relationship between banks and the central bank is ongoing. Lavoie (1984, p. 780) has described the role of the central bank in these following terms:

18

Credit, Money and Production The central bank must face an insidious “moral suasion”: it must comply with decisions taken and the environment created by commercial banks. Otherwise, the central bank may lose whatever control it has on financial operations, either because (financial or non-financial) economic units would attempt to diminish their needs for banking money and high-powered money, or because financial markets would have been thrown into a pessimistic frenzy, following the bankruptcy of a few banks. Since the govemment cannot let things get out of hand, and since all economic units are aware of this, the central bank is in a rather awkward position.

Before turning to a discussion of the circuit itself, a further clarification is needed with respect to the definition of the short mn. As it well known, Keynes set the General Theory within Marshall’s short period of “a few months to a year,” where the stock of capital is considered fixed. In the General Theory, one could argue that the “circuit” is defined as the time necessary for the full multiplier effect to take effect (Parguez, I980). While this certainly gives a definitive time dimension to economic theory, it becomes problematic within the confines of the circuit. In contrast, circuitists define the time span of the circuit as “merely the consequence of the duration of the credit” (Nell and Deleplace, 1996, p.

13). This suggests that the Keynesian short-period is thus defined as the average interval of time needed for banks to evaluate the ability of firms to generate a profit from an initial spending. Thus, the circuit is not only defmed as the time necessary for money to circulate, i.e. fiom the time it is created by bank credit to the time it is destroyed,“ but also the time needed for firms to reevaluate their production decisions based on the refinancing of all debt at the end of the period with the liquid saving of households. As such, there is little concem, say, for the total value of effective demand in 1997. Rather, emphasis is placed on effective demand which results fi'om production in 1997 irrespective of whether this takes place in I997, 1998, or beyond. It is the tracking of money through time (Lavoie, I984). Short-period Keynesian analysis should be defmed as the time between when entrepreneurs first make their production and investment decisions and when they need to revise these decisions for the next period of production. Implicit in this definition is the notion that entrepreneurs revise their decisions when profits have emerged, and money has been destroyed. The circuit is closed, and a new round of credit demand begins. There is evidence that Keynes himself was moving away from his Marshallian definition of time. In his rough course notes following the publication of the General Theory, Keynes moves away from a concrete definition of time to one where time is an abstraction of reality. Consider the following passage in Keynes (xiv, pp. 179-80):

The Franco-Italian Circuitists

I9

Time relationship between effective demand and income is incapable of being precise. In case of factors other than the entrepreneurs and rentiers the two are

more or less simultaneous. For the latter income becomes determinant and is transferred at varying subsequent dates. No definite relationship between aggregate effective demand at one time and aggregate income at some later time. This does not matter when one is dealing with aggregates, aggregate effective demand at time A has no corresponding aggregate income at time B. All one can compare is the expected and actual income resulting to an entrepreneurfiom a particular decision.

Moreover, we know that Keynes understood the Marxian notion of the production circuit, M-C-M’ (see Keynes, xxix, p. 8|). The circuit must also be understood within the greater context of time. It

is evident that many circuits exist simultaneously, that the economy does not operate in such a neat pattem - that is, circuits do not begin and end at

the same moment, they are constantly overlapping. The Circuit of Credit and Money

The division of society prevalent in Kaleckian theories of distribution and economic growth also forms the foundation of the circulationists’ theory of money, and this in two ways. First, within the context of credit-money, cireulationist macro models of money are traditionally divided according to a specific hierarchy of agents in particular according to agents’ access or control over money. These three macro sectors - or what Graziani (l990a, p. 8) refers to as the “relationship among macro-groups” — consist of banks,

firms and wage eamers. The theory of the circuit is set within what Kalecki has called the “pseudo-monopoly” — that is where firms are further subdivided between goods-producing and investment-producing firms (Vallageas, 1988)." Banks supply the much needed credit to firms who undertake investment and cany out production plans, and as such detennine the allocation of productive resources. Wage-eamers supply the intellectual and physical labor. The monetary flows themselves are “hierarchialized:” bank credit must first be obtained before production can begin; incomes caimot be disbursed before production begins; and firms caimot reimburse

banks until goods are purchased by households. The creation of money is therefore the result of the complexity of the interaction between three specific causal relationships: banks and firms;

finns and workers (or households); banks and the central bank. Hence, banks are not simply treated as another firm. Rather, banks have the responsibility of keeping the engines of the economy well oiled by supplying the needed credit. Banks create money and are distinct from firms. As Graziani (l990a, p. ll) explains, “Banks and firms must be considered as two distinct kinds ofagents. [They] cannot be aggregated

20

Credit. Money and Production

into one single sector." Moreover, it is money as a medium of circulation which explains this hierarchy (Poulon, 1980, p. 384; Ducros, I978). As Parguez explains, “The keynesian circuit is the entire hierarchy of flows corresponding to the global process of production” (quoted in Poulon, 1980, p. 384). Second, this division of society helps in understanding the income distributive role of money, that is the ability of capitalists to determine their own levels of profits through their access to credit, and their investment and consumption plans (Kalecki, 1971). Graziani (1996, p. 142) argues precisely that “money plays the fundamental role of determining the distribution of income.” At heart is the notion that money is part of a circuit, and as such it is first created, then circulated and finally destroyed given the monetary profits resulting from the productive use of money. A monetary economy is

therefore viewed as existing through a specific sequence of irreversible events where the principal factor is the “essentiality of money” (Parguez, 1986, p. 24). Money is defined as “behavioralistic.” The capitalist process as a whole is a monetary sequence starting with the creation of money. The first stage of the circuit: production decisions and prices

The monetary circuit begins with the firms facing two cnicial, albeit related, decisions. At the beginning of the circuit, firms must determine the monetary value of their output, thereby requiring them to determine not only the output or quantity produced, but also the price at which it will be sold. While firms already have an inherited stock of capital from the previous period, as well as a given stock of debt, they must decide whether to continue production at the previous period’s level, or to increase — or even decrease — production. This decision will be based on their expectations of the unknown future. As suggested by Keynes, the level of production is influenced by the

expectations of effective demand, the level of which is detemiined by shortrun expectations of proceeds (Barrére, I990). As for the factors which influence investment (changes in the capital stock), the discussion is more ambiguous; here it is assumed — following Keynes — that investment decisions are influenced by expectations of the long run (the stream of quasi-rents - see Asimakopulos, 1991). In determining their levels of production as well as their accumulation of capital, finns must therefore carefully weigh the expectations of the near future relative to those of the long run. In essence, the finn carefully assesses the future flow of income generated by production. While costs are known a priori, firms must carefully anticipate the state and strength of effective demand later within the period of the circuit. At this point, firms know with certainty the cost of credit (the rate of interest), and the cost of labor — the wage level of the

The Franco-Italian Circuitists

21

workers. These are exogenous variables within the set of variables facing the firm. These are, however, not the only constraints imposed on firms, as will be shown shortly (banks also impose a monetary mark-up on firms sufficient to generate profits to repay loans). Hence, firms typically operate

at less than full capacity; full employment is not assumed. There is thus ample place for uncertainty to influence production and investment decisions within the monetary circuit. Uncertainty plays in fact a central role, as in post-Keynesian theory. Post-Keynesians have at times suggested that circuitists ignore — or at least downplay — the importance of uncertainty on investment decisions. At times, this may appear to be so. In

fact, many circuitists assume that the circuit begins with the granting of credit to firms thereby enabling them to start production (Graziani, 1990a, p. I2). This would appear to suggest that circuitists pay only lip service to the effect of uncertainty on production and investment decisions. This is, however, not the case. Uncertainty remains a central component of the theory of the monetary circuit. It is in this respect that Parguez (I987) has emphasized that the circuit is a “pure gamble” against the future (see also Zerbato, I989). According to Parguez (1996, p. I59), in the short run, firms must “bet on the short-run profits that should be both the outcome of their

current sales and the proof banks need to support the rise in the effective capital value.”" At this stage as well, given their known costs, firms will also set the price

at which their goods will be sold. As in post-Keynesian theory, circuitists accept the mark-up approach to pricing, albeit with a certain difference. In post-Keynesian theory, the firms’ mark-up is generally determined by their desire to raise as much intemal finance required to finance investment, i.e. at the level of the firm, growth objectives detennine the price and hence mark-up. In the context of the monetary circuit, as will become amply clear below, this is not a satisfactory approach, essentially for two reasons. First, banks are money-creating institutions which can finance the investment expenditures of firms. There is therefore no absolute need for firms to raise

the fmance ex ante. Second, the ptupose of firms is to generate sufficient profits in order to pay back their initial finance and cancel their debt. This logic follows from Kalecki (1971) who argued that fimis aim to generate the largest surplus of revenue over cost. This does not imply that retained eamings do not play a role in the fmancing of investment.

However, the primary source of

investment finance are the banks. If retained eamings are used, it is because of what was carried over from the previous period, over and above the

proceeds needed to repay the banks and debt-holders of the firms. In the circuit approach, therefore, the mark-up will be set differently than in post-Keynesian theory, and irrespective, although not necessarily, of the

22

Credit, Money and Production

oligopolistic structure of the market.” If the objective of firms is to generate profits, then the mark-up must be related to profits as well. Given the future is unknown, firms cannot know with certainty their rate of retum, hence, they can only estimate a targeted rate of retum on production costs. In this sense, the mark-up is also a constraint imposed on finns by banks. To obtain credit, firms must prove to banks that the set mark-up is sufficient to generate an adequate amount of profits to pay back loans. The mark-up will therefore enter into the bank’s creditworthiness evaluation of the borrower. The bank’s influence enters into the fim1’s price equation in two ways. First by setting the rate of interest, it imposes on finns a cost on credit which enters the firm’s cost function. Second, it also impose on firms a minimum profit requirement necessary for the reflux principle to guarantee the repayment of bank finance. The second stage of the circuit: initial finance and the endogenous creation of money: Keynes’s finance motive The second step of the monetary circuit is the actual access by firms to bank credit” — Keynes’s finance motive.“ What is important to note is that the financing of production is not tied to prior saving on behalf of finns or

households. On this point, Keynes (xiv, p. 207) is quite clear: “Planned investment — i.e. investment ex ante — may have to secure its ‘financial provision’ before the corresponding saving has taken place.” Moreover, Keynes (xiv, p. 209) adds the following: “‘Finance’ has nothing to do with saving. At the ‘financial’ stage of the proceedings no net saving has taken place on anyone’s part, just as there has been no net investment.” Keynes considers the notion of prior saving — or ex ante saving — as ludicrous. In a footnote, Keynes indicates that “as for the concept of ex ante saving, I can attach no sound sense to it” (xiv, p. 210, n. l). To make his point, Keynes introduces the notion of the revolving fund — which first appeared in the June 1937 Economic Journal article — though Keynes never made clear its underlying mechanics. Of course, the independence of investment was recognized in the General Theory; it was indeed the causal factor of Keynes’s model. But there,

Keynes does not explicitly indicate the importance of the banking system as he had done previously in the Treatise (see for instance, v, p. 96). The need for credit is therefore vital to the system. In his post-General Theory articles, Keynes defines “finance” essentially as an “advance provision of cash required by the current decisions to invest” (xiv, p. 208) and altematively, as the “credit required in the interval between planning and execution” (xiv, p. 216, n. 1).” Given economies exist in historical time, there necessarily exists a period of time between the production of commodities and the revenues

The Franco-Italian Circuitists

23

originating from the sales. In other words, there exists a temporal separation of cash flows between the time costs are incurred and the receipt of revenues. Although this argument is made in the General Theory (Keynes, vii, p. 46), it is also made early on in A Tract on Monetary Reform. According to Keynes (iv, p. 33; emphasis added): “During the lengthy process of production the business world is increasing outgoirigs in tenns of money - paying out in money for wages and other expenses of production — in the expectatiorrs of recouping the outlay by disposing of the product for money at a later date.” This is only logical given the fact that revenues can only come about after the commodities have been produced, brought to markets and sold. During this time, the firm must incur the total costs ofproduction. Keynes returns to this argument in his post-General Theory articles in the

Economic Journal. Because of this “interregnum” between production and proceeds, firms must have some access to a wage-fund, as well as access to credit to fmance part of their investment projects. This was well recognized by Keynes (xiv, p. 219), who argued that there exists an “interregnum

between the intention to invest and its achievement” during which firms incur costs associated with production. As such, they must first fmd appropriate venues to finance this investment: “An investment decision may sometimes involve a temporary demand for money before it is carried out Planned investment — i.e. investment ex ante — may have to secure its ‘financial provision’ before the investment takes place; that is to say, before the corresponding saving has taken p1ace” (xiv, p. 207). Keynes recognizes the clear distinction between investment decisions and investment spending — a distinction which was clearly absent in the General Theory. “Now, at

ante investment is an important, genuine phenomenon, in as much as decisions have to be taken and credit or ‘finance’ provided well in advance

of the actual process of investment” (xiv, p. 216). While credit is surely needed to finance both production and investment expenditures, Keynes assumes that it may be best to anange a line of credit (xiv, pp. 208, 223). This statement concurs as well with the insights provided by Robinson years later. For instance, Robinson (1951, p. 104) assumes in her analysis the existence of overdraft facilities, and writes that “it is more natural to suppose, however, that entrepreneurs take bank advances as required and retire them by the issue of shares alter the investment has been under way for some time.” Robinson (1952, p. 81, n. 1) also writes: “The entrepreneur, to save interest payments, as far as possible avoids borrowing in advance of actual outlays. He prefers to arrange for a ‘line of credit’ (in the simplest case, overdrafi facilities at a bank).” While it is clear that Keynes claims that investment expenditures must be financed, this argument does not find a sympathetic ear among all

24

Credit, Money and Production

circuitists. In fact, most of them would exclude the purchase of investment goods fi'om initial fmance. This represents perhaps the greatest disagreement between circuitists. According to Seccareccia (1997, p. 3), “With the exception of such writers as Parguez most circuit theorists adopt a model that is akin to the old-line nineteenth-century ‘Real Bills’

doctrine of financing activities.” Hence, according to some, investment is never fmanced through the use of bank credit (see Graziani, 1996; Messori,

1985, p. 211, n. 5).”

For instance, Graziani (I990a, pp. 12-14; see also 1994, p. 277), claims

that credit is needed only to fmance the incomes of workers, such that total bank credit equals the wages of all workers: “If we consider firms as a whole, their only extemal purchase is labour force.

Finance requirements

depend on the monetary cost of output in general, and are not specifically connected with the investment activity.”24 Moreover, Nell (1998) assumes that only the wages of the workers in the capital-goods industry need fmancing. As wages of workers in that sector are paid out, their consumption will finance the wages of workers in the

consumption-goods sector as money is tumed over and over. In this sense, Nell relies on the velocity of money as a way of financing production.

The question of investment financing arises only at the end of the circuit, allowing for growth in the subsequent period. They are financed as “intemal transactions” (Graziani, 1990, p. 12) within the non-financial business sector,” that is through retained eamings and the new issues market (Vallageas, 1988; Graziani, 1996). Graziani (1990a, p. 12) argues this point: “Only at the end of the production process firms buy capital goods to be used in the following period.”

Some circuitists even argue that the assumption of using bank credit to finance investment is akin to assuming that saving is used to finance

investment. This is the argument put forth by Bailly (1992, pp. 110-12). In criticizing Parguez in particular, the author claims that, “We see by reading

A. Parguez that the break with the quantity theory is far from complete. His theory keeps its essence.

In defining ‘finance’ in this way

A.

Parguez reinforces OhIin’s thesis [of the determination of the rate of interest].” There are a number of problems, however, with exempting investment

purchases from bank credit. As Seccareccia (1996, p. 402) notes, the “asstunption regarding bank credit going exclusively towards the financing of working capital, is highly questionable.” First, if bank credit is used exclusively for the financing of wages, then firms in the investment-goods market make no profits. This therefore poses a considerable problem for the closure of the circuit (Seccareccia, 1997; Lavoie, 1987). Profits of the consumption-goods firms are generally unproblematic (Lavoie, I987), and are generally equal to, assuming the normal classical hypotheses, the wages

The Franco-Italian Circuitists

25

of the workers in the investment-goods market. If this is the case, then the profits of investment-goods firms fall to zero, and can even be negative as soon as we assume household saving. This is precisely the argument of Nell and Deleplace (1996, p. 14): “If bank credit is limited to wage costs, the proceeds of the capital goods sector cannot be superior to the invested profit of the consumer goods sector, hence to its own wage bill: profit in the capital goods sector is nil.” The easiest way of solving this issue is to assume that part of the investment purchases, of firms in the consumptiongoods and investment-goods sectors, is financed through bank credit. This is the conclusion reached, notably, by Seccareccia (1997, p. 5), who writes: “Without some other net debt in the system, the closure of the monetary circuit can ensure only if firms’ purchases of A (p,,K) are fmanced by the creation of credit-money at the beginning of the circuit,” otherwise investment-goods firms would not be able to reimburse their debt with the banks. This is the familiar debate over the existence of profits, or in Marx's tenninology, how can M become M’. As Seccareccia (1996, pp. 405-6) asks, “How can firms collect more money revenues that the amount that economic agents are willing to borrow collectively from the banking system?” Similarly, Nell and Deleplace (1996, p. 14) also ask, “I-Iow firms who borrow a given sum can reimburse it and pay the interest, if all money comes fi'om bank credit?” Even Schumpeter (1934, p. 189) asked this question. Consider the following reference: “Within the circular flow it is impossible with a given money sum to obtain a greater money sum.”2‘ Second, as Seccareccia (1997, p. 6) points out, Graziani’s argument also assumes an asymmetry in financing: “Firms in the [capital-goods] sector cannot somehow be assumed idly to await orders and be paid in advance of production (by all those fums wishing to purchase capital goods via creditmoney), while firms in the consumption-goods sector are presumed to behave difi'erently by initiating production with at nihilo bank credit!”

Finally, in response to Bailly’s argument, using bank credit to finance investment is not equivalent to assuming the natural rate hypothesis, as post-Keynesians well know. First, the rate of interest is considered exogenous, thus unrelated to the forces of supply and demand, of money or saving. Second, as long as circuitists accept the reversed causality between deposits and loans - which they do - there are no grounds to Bailly’s arguments. According to the theory of revolutionary endogeneity, the assets of the banks determine the liabilities: loans create deposits. Money is created ex nihilo. Hence, even though bank credit is used to finance investment, these credits do not come from prior saving. This is the very foundation of a revolutionary theory of endogenous money. For these reasons, it is assumed here that investments are fmanced, albeit

not entirely, through bank credit. This does not imply that retained earnings

26

Credit, Money and Production

or the new issue market are not important sources of financing investment. As Lavoie (1987, p. 69) contends, investment is partly financed by all three sources, although bank credit (both medium and long-terrn fmancing) remains at the core of the financing. As Seccareccia (1997, p. 10) has

shown using Canadian data, “A notable proportion of corporate investment is done via debt creation rather than through the build-up of corporate equity.n27

By assuming, however, that investment expenditure is also financed by bank credit along with circulating capital, one thing is clear: there exists in the circuit a struggle between short-run financing (wages and raw materials) and longer-nin financing (capital goods). As will be shown below, the

short-run financing is always reimbursed in full to the banks by the end of the circuit. Graziani (l990a, p. 15) refers justly to this financing as “temporary finance.”

The issue of investment fmancing is, however,

somewhat more complicated. Finns will typically negotiate with banks in order to reimbtuse within a given circuit only a traction of their total

investment debt. This implies of course that firms are perpetually in debt toward the banks. Hence, the reimbursement of investment finance is stretched over several periods. Following this discussion, two issues now arise. First, if credit is used, who supplies this credit? Does it matter whether credit arises out of bank loans or fi'om other financial intermediaries? Second, is credit used for the production of investment-goods alone, or can it cover the production of consumption-goods as well? These issues are treated in tum, where Keynes’s discussion of the finance motive is at the heart of the following analysis.” It is in his post-General Theory that Keynes introduces his fmance

motive, to better explain in what way investment is never determined by a prior saving. Despite Davidson’s (I965) best efforts, the finance motive remains largely ignored — even by post-Keynesians. Rousseas (1992, pp. 39, 47) refers to the fmance motive as “transient in nature” and is “much ado about very little.” Moreover, post-Keynesians believe that the fmance motive is needed only when the economy is growing. Consider the following statement by Davidson (1986, p. 101; emphasis added): “The role of the banking system is to create additional short-tenn finance whenever entrepreneurs wish to

increase the flow of real investment.” Similarly, according to Rousseas (1992, p. 41), “The finance motive comes into play primarily during a charged expansion of the economy leading to abnormal increases in

plarmed investment.” These statements by Davidson and Rousseas are quite representative of the post-Keynesian position. Circulationists, on the other hand, consider the finance motive as integral to economic activity irrespective of whether the economy is growing. ln

The Franco-Italian Circuitists

27

fact, Seccareccia (1997, p. 2) associates the need to finance production only when the economy is growing with the “neoclassical theorists of loanable funds. Even in a stationary environment, albeit highly uncharacteristic of capitalist economies, in which the flows of income and production are perceived to remain unchanged over time, the flow of production would require financing.” _ Moreover, as Graziani (1985, p. 167) writes: If the level of production is constant, finns receive constant revenues fi'om the sale of their output But the liquidity they receive must be reimbursed to the banks, which means that immediately after, even if fimis desire to keep output constant, they must first obtain new credit. All that can be said is that a constant level of output requires a constant level of finance, but in no cases can it be suggested that in a stationary state the need for finance disappears.

Graziani (1996, p. 147) would later write that “in stationary conditions, the amount of finance needed by firms is constant - something which by no means implies that firms become financially independent fi'om the banks.” The argument of utilizing credit even if production or costs remain constant will become clearer below. It is, however, related to the need for firms to reimburse banks the value of their initial credit. At the end of the circuit, firms must first reimburse the banks, they cannot use the same proceeds to continue production. Banks need to be paid back. Even though production remains constant, once banks are reimbursed their loan, firms must once again borrow funds to cover circulating capital and their investment expenditure for this new period. This is called by circuitists the “reflux” principle, and is discussed more at length below. Tuming now to the two questions above, the answer to the first one is that although banks become the foremost source of credit fimds, circulationists do not necessarily difierentiate banks from other financial intermediaries. For them, it is sufiicient to argue that the finance needed does not originate from the securities market. Bank fmance is defined as the opposite of financing through the financial markets, which have, as will be shown below, a completely difierent role to play in the circuit. Overall, therefore, what is emphasized is that production is financed, irrespective of prior saving. In this context, it becomes immaterial whether this finance is provided by banks or by other financial intermediaries. As long as one breaks the causal relationship between deposits and loans which exists in mainstream banking theories, and as long as one also recognizes that the central bank has no influence in determining the supply of loanable funds, then the difference between banks and other financial intermediaries is blurred.

What is of primary importance is that production is financed by credit, originating foremost fi'om commercial banks, where the assets of the banks

28

Credit, Money and Production

are causal in the relationship. Hence, in contrast to New Keynesians and some post-Keynesians, deposits are a result of a prior loan. In fact, Parguez (1996, p. I58) explains that “in every state of the economy, the amount of

liquid resources initiated by credit contracts is thus identical to the amormt of newly created money.” In the General Theory, Keynes virtually ignores commercial banks in his

analysis, despite having put them at the center of his analysis in the Treatise on Money. For instance, in the Treatise, Keynes (v, p. 182) writes, “In

order that producers may be able, as well as willing, to produce at a higher cost of production

they must be able to get command over an appropriate

quantity of money. Thus the first link in the causal sequence is the behaviour of the banking system.” Keynes retums to the role of banks afier the General Theory. There appears to be initially some confusion regarding the fmancing of investment

which is, however, later made clear. Initially, Keynes argues that finance “may be provided either by the new issue market or by the banks; — which it is makes no difference,” while later suggesting that fmance is forthcoming

“more likely

if he is fmancing himself by a new market issue than if he is

depending on his batik” (xiv, p. 208).

Keynes therefore seems to be initially suggesting that finance is most probably made available, not through banks, but by the “new issue market.” At the very least, Keynes is arguing that banks are only one of the possible sources of finance. But in the context of a monetary circuit, “which it is” does “make a difference.” In arguing that the new issue market is a source of fresh finance, Keynes is remaking the argument that ex ante saving determines at ante investment since the corresponding saving is produced only once the investment has taken place. Keynes, however, later clears up his position. In his December article — following criticism raised in particular by Robertson — Keynes (xiv, p. 217) clarifies this ambiguity suggesting “firstly, that [the entrepreneur] can obtain sufficient short-tenn fmance during the period of producing the investment; and secondly, that he can eventually fiind his short-term obligations by a long-tenn issue on satisfactory conditions.” Keynes sees

this dual-sequential process as “the characteristic one.”

This suggests

rightly the proper way of seeing a monetary circuit. Bank credit is used for

production, and saving may be used as a way of reimbursing the initial debt by firms selling new securities - what can be labeled financial saving. 2° Further, Keynes (xiv, p. 210) argues that the “finance required during the interregnum between the intention to invest and its achievement is mainly supplied by specialists, in particular by the banks.” In the following

paragraph, Keynes claims that the finance is “wholly supplied by the banks,” a situation which he claims is “substantially representative of real life” (xiv, p. 219). Keynes (xiv, p. 285) reaffirms this position a year and a

The Franco-Italian Circuitists

29

half later in his 1939 article in the Economic Journal: “It is the role of the credit system to provide the liquid fimds which are required first of all by entrepreneurs during the period before his actual expenditure, and then by the recipients of this expenditure during the period before they have decided to employ it.” In this last statement, it is clear that by “credit system,” Keynes has in mind commercial banks as the primary source of finance. Moreover, with respect to this fmance, the banking system is usually not

constrained. Circuitists will generally argue that as long as customers meet the banks’ collateral requirements (creditworthiness conditions), then banks are willing and able to extend credit.” Keynes also recognized the banks’ ability to increase their “financial cormnitments” on demand. The key rests with the banks’ “willingness” to determine the money supply. This does not suggest that commercial banks will grant all requests for credit; Keynes (v, pp. 212-3; vi, pp. 364-7) recognizes that there will always be a “fringe of unsatisfied customers.” Generally, however, banks can increase the supply of credit to those demanding it, albeit as long as they meet the banks’ collateral requirements. This does not mean, as pointed out by DeVroey (I984, p. 385), that “the credit granting is well founded.”3' For instance, Keynes (xiv, p. 210; emphasis added) writes “the banking system chooses to make the fmance available.” Also Keynes states that “unless the banking system is prepared to augment the supply of money, lack of finance may prove an important obstacle to more than a certain amount of investment decisions being on the tapis at the same time.” Clearly, Keynes rests the primary responsibility for economic activity squarely on the banks. In this sense, Keynes establishes two important points. First, he recognizes that commercial banks have a role to play in

determining the quantity of money. In essence, it is a tacit recognition that the supply of credit-money is demand determined: “Too great a press of imcompleted investment decisions is quite capable of exhausting the available fmance, ifthe banking system is unwilling to increase the supply of money. Yet this is only another way of expressing the power of the

banks through their control over the supply of money” (xiv, pp. 210-11; emphasis added). In fact, Keynes argues that banks are never resourceconstrained and that they can lend as much as they wish. However, they may choose not to lend for a variety of reasons. Also, Keynes recognizes the importance of the commercial banks in determining the “pace of economic activityz” “The banks hold the key position fi'om a lower to a higher scale of activity The investment market can never become congested through shortage of saving. This is the most fundamental of my conclusions within this field” (xiv, p. 222). Also, “This is the explanation of why [the baiiks’] policy is so important in determining

the pace at which new investment can proceed” (xiv, p. 219). The emphasis

30

Credit, Money and Production

is therefore put on banks rather than entrepreneurs in determining the “pace of economic activity.” Keynes also makes clear his views on the endogeneity of money, and the

role of banks in determining and influencing the supply of money. Keynes (xiv, p. 210), writes about the “power of the banks through their control

over the supply of money.” Since banks appear to “choose” to augment the supply of money, and since the central bank is virtually absent from his analysis in his post-General Theory articles, then there can be little doubt that Keynes’s analysis is closer to the post-Keynesian, cireulationist tradition than to the New Keynesian views.” As for the second question, circuitists claim that fmance is required in order to produce investment goods as well as consumption goods, as made amply clear by Graziani (1985, p. 161). This also corresponds to Keynes’s own position on the issue - a point which he made on numerous occasions. For instance, Keynes (xiv, p. 208) writes, “Investment fmance in this sense [as an ‘advance provision of cash’] is, of course, only a special case of the finance required by any productive process.” Also, he writes (xiv, p. 283; emphasis added): “The production of consumption goods requires the prior provision of funds just as much as does the production of capital goods. It is not an increase of investment as such which requires an immediate increase in “available funds,” but an increase of output whether for investment orfor consumption, or more strictly an increase in the turnover of transactions for any purpose. [Finance] covers equally the use of the revolving pool of funds to finance the production of capital goods with the production of consumption goods.” Finally, Keynes (xiv, p. 283) writes -

while referring to the “funds available for investment” - that credit “covers the use of the revolving pool of funds to fmance the production of capital

goods or the production of consumption goods.” For circuitists therefore, money is endogenous as a result of debt

formation. The constant flow of production requires firms to obtain credit from banks, where deposits are created in the process. Money, in this sense, is a result of debt and production, not a result of uncertainty. Production can never be financed by changes in the portfolio decisions of households. Money is created ex nihilo (Aglietta, 1979, p. 335) as soon as credit is awarded to firms.“ Keynes also had this notion of the monetary circuit in mind, when in his Economic Journal articles, he discusses the revolving ftmd of fmance. Consider, for instance, the following statement made by Keynes (xiv, p. 209; emphasis added): “If investment is proceeding at a steady rate, the fmance required can be supplied from a revolving fund of a more or less

constant amount, one entrepreneur having his finance replenished for the purpose of a projected investment as another exhausts his on paying for his completed investment [it] looks afier a flow of investment. It is a

The Franco-Italian Circuitists

31

revolving fund which can be used over and over again.” Then, Keynes (xiv, p. 230; emphasis added) adds: “A given stock of cash can provide a revolving ftmd for a steaay flow of activity; but an increased rate offlaw needs an increased stock to keep the channels filled.” Keynes is describing the process of the monetary circuit with the causality running fi'om investment decisions (expected or desired income), to finance (demand for, and supply of, credit), to investment, to effective income and saving — hoarded and financial. The discussion of Keynes’s revolving frmd is nonetheless problematic. Keynes appears to be talking about a stock of finance financing a flow of investment. This would, however, go against the meaning of the monetary circuit. But as Graziani (1996, p. 147) notes, “The fact of its amount being constant does not deprive the revolving fund of its being by nature a flow of liquidity supplied by the banks?” At this stage ofthe circuit, as banks extend credit to firms, the firms’ bank accounts increase by the exact amount of the credit demanded and supplied. Thus, initially, the “supply of money” in conventional terms has also increased by that same account and will remain so through the various transactions of the economy until the end of the circuit. Moreover, the banks’ balance sheets remain in equilibrium: their liabilities (firms’ money deposits) increase by the same amount as the assets (loans). Money thus appears because of the various debt-relationships between the various “macro-groups.” If credit is extended indeed by lines of credit, then the same logic appears. As firms “borrow” on their lines of credit, only the used portion appears as a loan on the banks’ asset side of their balance sheets, and only that portion appears as money. One thing is clear, however: until firms use their lines of credit, no money is created. The third stage of the circuit: the creation of income flows The third step of the circuit is the disbursement of hicomes (wages and dividends) and the purchase of investment goods.“ This is a vital step of the circuit since it exemplifies the notion that firms must spend in order to make money and profits. This point was well put by Parguez (1997, p. 5): “This stage depicts the paramount characteristic of the capitalist economy: firms must spend to be able in the future to recoup money from the sale of their output.” Both of these actions are outlays for the firms (the efllux principle), but are also revenues for households and for investmentgoods firms. Bank credit must be put into productive use in order for fnms to generate income. It is the creation of incomes which allows money to circulate. If there were no workers to remunerate, then money could not circulate and hence exist.

32

Credit, Money and Production

The fourth stage of the circuit: monetary reflux and the creation of profits Once incomes have been disbursed through the circuit, the fourth step of the circuit, the reflux mechanism, begins and firms, in both sectors, start to receive proceeds from the sales of their products: the aggregate receipts. As mentioned above, as soon as consumption-goods firms purchase capital goods — and also when investment-goods firms purchase these goods —

investment-goods finns receive revenues from these sales."

Similarly,

when households purchase consumption goods, these purchases represent

proceeds for the consumption-goods firms. Thus, the monetary reflux can be defuied as the liquidity which firms get back during this stage of the monetary circuit. With the proceeds fums receive from their sales, they will pay back their initial finance which allowed the circuit to begin: firms reimburse the banks and extinguish their debt. Money is destroyed, the

circuit is closed, and a new circuit is ready to begin. Provided of course households do not save, then fums will have succeeded in extolling from

the circuit the exact amount of money they injected into the economy at the begirming of the circuit. Thus at the end of the circuit, with no household saving, firms will have no debt toward the banks, and there will thus be no change in the money supply, since no deposits, i.e. household saving, exist.

At this stage, the discussion over the importance of the revolving fund of finance comes into question. Post-Keynesians assume that the finance motive is operative only when the economy is growing. In contradiction to this position, circuitists argue that the fmance motive is always important, even in situations of constant production or costs. The reflux principle makes this important point clear. Once firms receive these proceeds, they

can do one of two things. First, they can decide to hold on to the proceeds received from the sale of the goods and begin anew the production process.

In this sense, as post-Keynesians argue, there would be no need for bank credit provided the level and costs of production have remained unchanged.

Second, they can use the proceeds to pay back the banks the exact amount of the initial finance (provided household hoarded saving is nil). The first scenario is highly unlikely, both fiom the point of view of firms

and the banks. For firms, they will want to reimburse the banks since their initial finance is costing them dearly in terms of interest payments. Reimbursing the banks therefore allows firms to avoid any unnecessary costs.

Similarly, banks will insist on getting reimbursed since this

represents for them an income. Firms will need to reapply for credit, even if their level of production has remained constant. The finance motive is therefore far from being “temporary.” This point was well developed by Poulon (1980, p. 281) who argues that “when the date of the cancellation of money arrives, firms use their sales

proceeds to reimburse banks.” This is the “final finance.” LeBourva (1992,

The Franco-Italian Circuitists

33

p. 454) has called this process “alternating movements of creation and cancellation of money.”

Once this has been achieved, firms must renew their demand for credit to start production at the following period. This is, however, not an automatic mechanism: banks may decide for a number of reasons not to renew their financial commitment to the firms (see chapter 8). Once again, Poulon (1980, p. 281) makes this argument: “[Firms] must ask for a new credit in order to continue their production activities.” Hence, even if production remains constant or costs do not change, the revolving fimd is always operative. Keynes’s finance motive, far from being “transient in nature” and “much ado about very little,” is in fact a permanent feature of production and the creation of endogenous money. It is constant when production is constant, and increases when production and costs rise. The assumption of households not saving is unrealistic. Generally, households will save a portion of their income. While saving generally represents a drain on firms’ proceeds and profits, it is important to clearly differentiate between the different types of saving. Just as income can be divided into two categories, consumption and saving, so can saving. On the one hand, households will use a portion of their saving to increase their liquid balances - hoarded savings deposited in households’ savings accounts. On the other hand, they will use the remaining portion to purchase debts on firms: new secmities. This can be labeled financial savmg. ln the first case, the decision of households to “hold on” to money will consist in their demand for money as an asset: this is liquidity preference. Money is a perfectly liquid asset which is demanded to meet uncertainties. Liquidity preference and uncertainty are integrated within a theory of the demand for money. The decision by households to demand money as an asset will be met by the banks’ decision to supply the money. Hence, both at the beginning of the circuit as at the end of the circuit, the demand for “money” - both in tenns of credit and money proper — will create the necessary supply: the endogeneity principle operates in each case (Parguez, 1997). What is clear is that the demand for money is an ex post notion, that is at post to credit demand and production decisions. The demand for money therefore carries no real significance: it does not determine the firms’ investment decisions, although it impacts on their realized profits." Apart from their demand for money, households will also use their saving to purchase new securities (financial saving). This consists in their demand for money to be spent on speculative purposes, and will increase their holdings of securities in their portfolios. From the point of view of firms, this will consist in an additional source of reflux: money which firms would have succeeded in recapturing fi'om the market. Firms thus try to recapture

34

Credit, Money and Production

a portion of household saving through the mechanism of the financial market. Consumption is therefore not the only source of reflux.

Keynes

concurred. For instance, interpreting “saving” in the following passages as “financial saving,” Keynes (xiv, p. 221) writes in his December 1937 article that “consumption is just as effective in liquidating the short-term fmance as saving is. There is no difference between the two.” Then, in a March

25, I938 letter to Shaw, Keynes (xxix, p. 276) writes: “For the purpose of restoring liquidity consumption is just as good as saving. When income is created by the expenditure of the ‘finance,’ it is a matter of indifference in this connection whether the recipients save or spend it. In either case liquid

resources are increased by the release of the ‘finance’ which has been temporarily held up.” Also, Keynes (xiv, p. 282) argues that “spending

releases funds just as much as saving does.” The conclusion which can be drawn here is that Keynes - and circuitists — does not see the role of financial markets as supplying fresh finance or financing investment, but rather as closing the circuit. The fmancial market allows firms to recapture some household saving, and supplies them with additional funding — fmal finance — with which to pay back bank loans. As such, the consumer goods market and the financial market play similar roles. Liquidity preference — that is households’ decisions to hold part of their saving as idle balances and the rest as new securities — will depend largely on the discrepancy between the rate of return on money and the rate of retum on other assets (for simplicity, let us assume new securities), as well

as uncertainty, either of the future or of the future value of the retum on other assets. For convenience, assume the following relationship: r,/rd

which can be referred to as the “liquidity preference ratio,” where r, is the

rate of retum on securities, and rd, the rate of retum (interest) on bank saving deposits, i.e. the own-rate of retum on money. The decision to allocate saving between money and securities can be reduced roughly to a struggle between the short-tenn rate and the long-tenn rate. As Graziani (1996, p. 144) explains, the decision to allocate saving between deposits

and securities will be based, in part, on “the rate of interest that firms offer on the securities they issue [which] must be high enough to balance the liquidity preference of savers.” If the liquidity preference ratio is equal to one, households are indifferent

between holding money and holding other assets. If it is greater than one, households prefer securities over money, and finns enjoy a reflux from the

The Franco-Italian Circuitists

35

securities market. Finally, only when the ratio is below one does this pose an obstacle for the closure ofthe circuit. From this analysis, we can see once again to what degree uncertainty plays an important role. For instance, there is nothing incompatible at this stage of the circuit between circuitists and post-Keynesians. In an ex post

situation, that is once money has been created ex nihilo from credit, the post-Keynesian analysis of money as wealth based on uncertainty fits into the picture. It is the uncertainty of the future course of the various rates of interest — as well as the fear of the future value of capital assets — which will influence households in their decision to part with hoarded saving. Households will thus be influenced partly by the actual — or “present” — value of the liquidity preference ratio, but also by its future, anticipated, course. For instance, if they are particularly “bullish” in their perception of financial markets, then they will anticipate the liquidity preference ratio to increase, and hence reduce their demand for money as a liquid asset in order to buy a greater number of securities. However, one point is clear: in cireulationist theory, never can portfolio decisions be used to finance investment. As Arena (1996, p. l8) claims, “For circuit theory neither the liquidity preference, nor the financial rate of interest may affect investment decisions.” Rather, portfolio decisions are used to reimburse banks their initial finance, and arise only at the very end of the circuit.” One of the conclusions which can be drawn from the above analysis is that from the point of view of the firms, both consumption-goods and investment-goods firms, consumption and fmancial saving (new issues market) play a similar role. At the very limit, if firms are able to fully recapture household saving, then with hoarded saving equal to zero, the monetary circuit is closed. As Seccareccia (1996, p. 413) argues, If finns are able to attract fully household savings in the capital market so that the increment in the capital stock would be matched by an equivalent

increment in the stock of corporate securities held by the public, firms would still be in a position to retire their debts entirely vis-a-vis the banking system.

Terzi (1986-7, p. 190) made the same argument: “Thus, firms repay the Bank with their sale proceeds and fimd the remaining debt by issuing debt titles to economic units whose cash receipts exceed their expenditure.” The problem arises, however, when households decide to keep part of their

saving as bank deposits. It is therefore the increase in bank deposits which poses a problem for the closure of the monetary circuit: hoarded saving represents a leakage. Liquidity preference represents a problem for the closure ofthe system. If this occurs, then the closure of the system is problematic. For one thing, firms will be unable to reimburse their short-tenn finance, and will be

36

Credit. Money and Production

therefore placed into a position to renegotiate their fmancial commitments toward the banks.“ The above discussion exemplifies to what degree money demand is a

residual of the system. The stock of money which is held by households makes up the increase in the stock of money.

What is clear is that

households’ demand for money, the debt of firms, and the final increase in the supply of money are all equivalent notions." As Lavoie (1992, p. 69) has made clear, “The portfolio decision of the households leads to residual

stock of credit and of money." Seccareccia (I997, p. I6) holds the same position: It is only when households choose to withhold their savings from the financial capital markets and seek to hold a significant proportion of their saving in the

form of bank deposits that difiiculties of reimbursement appear. This forces banks into uneasy “intermediation” role of re-financing debt or acquiring by default business assets equivalent to the net accumulated household bank deposits.

The last phase of the circuit: banks as financial intermediaries. The final increase in the supply of money therefore represents the inability of firms to fully rid themselves of their short-term debt. Their debt coincides with the portion of saving which households have hoarded, that is the aggregate debt of the economy will be equal to the demand for money. An important and logical question to ask is what then happens to the unpaid portion of firms’ debt? At the end of the monetary circuit, firms owe banks the exact amount that

households have decided to entrust to banks as saving deposits. Banks will use these funds to refinance firms’ debt under longer tenn conditions. Parguez (1997, p. 7) has summarized this scenario in these terms: “When there is an increase in the demand for money as an asset, firms cannot repay their whole short-term debt to banks. Banks are thus pledged to refund this

debt by granting firms long-term credits.” At the end of the circuit, therefore, these saving may be channeled back into circulation. As Parguez (1996, p. 190) also writes, “They substitute a long-run claim on firms for their remaining deficit.” Banks are thus fulfilling their role as financial intermediaries, channeling hoarded saving to fu-ms enabling the closure of the circuit. This position is defended notably

by Nell and Deleplace (1996, p. 15) who write, “[Hoarded saving] creates a leakage for the firms but not for the economy as a whole since, if money is

permanent, cash balances may be used to finance firms’ deficits. Banks are then acting as financial intennediaries which can complement the financial markets to channel households’ savings towards the firms.” The claim that banks act as “financial intermediaries” may not be the best

description of the process.

Although this role resembles the orthodox

The Franco-Italian Circuitists

37

argument of deposits and loans, it is important to mention that it shares nothing with the more mainstream monetary theories. To begin with, this role of banks arises only at the end of the period, and is, so to speak, an exception to the rule. Second, whereas in orthodox thought bank deposits determine the supply of loans, here, deposits determine the demand fi'om firms to refinance their loans. Hence, the causality is also different (Graziani, l990a). As long as banks agree to play their role of “fmancial intermediaries,” they would have avoided a possible financial crisis by financing the firms’ long-term liabilities.

CONCLUSION The theory of the monetary circuit emphasizes many of the important elements of the theory of endogenous money without relying on portfolio theory or money as a stock. Rather it emphasizes the creation of money

starting with credit demand and the production needs of firms, its circulation, and its ultimate destruction. Money becomes a theory of distribution and accumulation. Flow and hierarchical analyses play a central role.

NOTES l.

2. 3. 4. 5.

6.

This does not imply that the General Theory is inadequate. References to it were made in previous chapters, despite the harsh criticism of it at times. Poulon (I980) and Messori (l99l) for instance both argue that the General 1heory is to some degree compatible with the monetary circuit, although as Messori (I991) also points out, the General Theory poses a problem with respect to endogenous money. However, such an analysis would require some contonions. This is why reference to the Treatise and in particular to Keynes’s post-General Theory articles are the best route to follow. This observation concurs with the point made by Zerbato (I989, p. 93) - that if Keynes referred to the notion of the monetary circuit, this was not in the General Theory, but rather during his “radical phase,” corresponding, it is argued here, to his post-General Theory years and his Treatise on Money. The words “circuitists” and “circulationists” will be used interchangeably to refer to those economists who adhere to the theory of monetary circulation. The expression circuit dynanrique was coined by Parguez (I984). While profits appear at the end of the circuit, that is afler the reflux mechanism, they may appear at the beginning of the circuit provided they were “inherited” from the previous period following govemment debt or debt fiom other countries. It also can not be exogenous. This was precisely Keynes’s mistake. Conventional wisdom (see Robinson, I970) therefore suggests that Keynes assumed an exogenous money supply in order to receive a better hearing from his fellow economists. This was a "strategic" decision, or a “tactical” one (see Dow and Dow, I989, p. I49). ln reading the early proofs of the General Theory one can see that Keynes initially wanted to retain many of his insights of the Treatise on Money. For instance, in a chapter

Credit, Money and Production contained within the first proofs of the Genera! Theory but which did not make it to final print, Keynes (xxix, pp. 67—8) describes his endeavor as follows: “It is to the theory of a generalised monetary economy that this book will attempt to make a contribution." Elsewhere, Keynes (xiii, p. 4l l) writes, “Accordingly I believe that the next task is to work out in some detail a monetary theory of production, to supplement the real-

exchange theories which we already possess. At any rate that is the task on which I am not wasting my time." Unfortunately, as will be argued later, Keynes did not develop, within the General Theory a theory of a monetary theory of production, as opposed to a theory of a monetized economy. Quoted by Moggridge. See Keynes (vii, p. xviii). Recall that many post-Keynesians have argued that in Rousseas, Davidson, Minsky and Kaldor, money is endogenous largely because of changes in the velocity of money. Two main schools in France exist. The Dijon school, under Bemard Schmitt and Alvaro Cencini, are interested foremost in the analysis of intemational payments. In Paris, Alain

Parguez and Frederic Poulon are interested in unemployment, economic levels, and more recently, in scarcity and govemment deficits. Moreover, a closely related group in Italy, under Augusto Graziani, although perhaps closer to the Paris group, nonetheless hold important differences with them. For circuitists in general, the theory of the monetary

circuit is a short-period, equilibrium theory. Keynes’s Treatise on Money also plays an important role in the following analysis.

Given the orthodox climate in France at the present time, the Italian circuitists, under the influence of Graziani, represent the more dynamic of the two groups, contributing significantly to the development of the theory of the monetary circuit. The list of proponents is not exhaustive, but also indicates the various differences. For instance, while Bellofiore may be classified a Marxist, Screpanti and Messori are closer to the structural ist post-Keynesians. A waming on Wicksell. Although banks and bank credit play a central role in the endogenous creation of money, this remains nonetheless a disequilibrium phenomenon a position obviously rejected by circuitists and post-Keynesians. This is not always the case. Messori (I986) rejects the horizontalist position. Even Graziani (I990b, p. 57) appears to adopt the upward sloping money supply curve: “Consider a bank with a regular supply curve which defines the supply of loans as a rising fiinction of the rate of interest." Some post-Keynesians also accept this position (see Fonnan, Groves and Eichner, I985).

While many would undoubtedly classify circuitists as horizontalists, this is not always the case. For instance, while circuitists accept the exogeneity of the rate of interest, Screpanti draws the money supply curve as upward sloping based on Kalecki‘s principle of increasing risk.

As will be seen later, this definition is not necessarily difierent from Keynes’s own definition of the multiplier circuit, as long as one postulates, as do circuitists, that the

multiplier is necessarily equal to one, as in Moore (I988) for instance. The inclusion of a Govemment sector, as demonstrated by Parguez (I997) changes very little to the analysis. Although investment is the causa causarr of the system, this in no way implies that consumption plays no role in anticipating future levels of effective demand. As Parguez (I996, p. I68) writes, “Investment being initiated by bets on the future course of demand, aggregate investment is the outcome of the ways of betting on future consumption, directly or indirectly.” It is in this sense that Parguez (I997) uses the expression “monetary mark-up" as opposed to the more traditional term, “Kaleckian mark-up." Of course, banks also extend credit to households. Within the monetary circuit, these can be taken as advances on salaries (Vallageas, I988, p, 356, fii. 2) and are thus ignored for the purposes of this analysis.

The Franco-Italian Circuitists

39

Hence, contrary to Rousseas's (I992) criticism, circulationists claim to successfiilly reconcile Keynes's finance motive with a “horizontalist” position, provided money is seen primarily as a medium of exchange and circulation, rather than as an asset. This view has been developed at length by Graziani (I984, I985) and Parguez (I975). Keynes is therefore aware of a certain distinction between credit and money. If he insists on referring to credit as money, it is perhaps because, at this time, Keynes prefers staying as close as possible to the tenninology of the General Theory. Dividend payments on past debt are also included in credit demand. According to l.avoie (I987, p. 84), this is an acceptable component of credit demand: "The payments made to households include the salaries and the interests and dividends.” Similarly, according to Lavoie (I992, p. I08), “The fact remains that payment of dividends is a de facro obligation of the firm, similar to its dejure obligation to make interest payment." Among the post-Keynesians, Moore (I988) has also adopted a model where bank credit goes toward circulating capital. See Seccareccia (I997, p. 4) for a criticism of t.liis point There have been several attempts at explaining the existence of profits. Most rely on some extemal factors, such as govemment, a foreign sector or, as in Robinson (I956), on the wages of workers in the banking sector. Although the topic deserves greater exploration, it is not discussed here. Parguez (I996, p. I61) has, however, given a more realistic suggestion, relying on the specific hierarchical relationship between banks and their customers. Finns negotiate the reimbursement of their debt over several periods. Individual credit contracts cannot always insist that current investment must be paid back or refiurded entirely by current net income generated by current sales. The contracts arranges that a share of new investment should be refunded by current profit, the other share being refianded later by a share of those firture increasedprofits, whose expectation is the support of the rise in capital value which is the stake of current investment. Parguez’s position (which is supported here) is in contrast with Zebato‘s (I989, p.96) view that profits cannot be generated within a closed system. Zebato (I989, p. 96) argues that “we must admit that the circuit can only be imagined as open. Closed it would be impossible to account for profits." If greater retained eamings allow firms to finance a greater portion of their purchases of capital-goods, it should not be forgotten that the causality is also reversed. As creditdemand increases, retained eamings or profits at the macro level increase. The following discussion on Keynes's finance motive is largely taken from Rochon (I997). Financial saving is defined as the component of saving which households use to purchase financial assets. Hoarded saving is what households keep in their bank accoimts. This is their demand for money. In chapter eight, a post-Keynesian theory of banks consistent with the monetary circuit is proposed. Banks may incorrectly evaluate the creditworthiness of customers, or economic events may make even the best customers unable to repay their loans. With these views and Keynes's later conversion to the exogeneity of the rate of interest, Keynes is making evident his slow progression toward the horizontalist tradition of postKeynesian economics. Post-Keynesian generally assume that bank credit is needed only for the production of investment-goods. For instance, Davidson (I985, p. I05) writes, that credit is needed only for the production of “real investment goods which cannot be used by households for consumption purposes." Not all circuitists agree that money is created simply as a result of credit. For instance, Schmitt (I984) would claim that money exists only at the moment that a payment is

Credit, Money and Production made, at which point "money" would be transferred from one household's account to another. When money stays idle too long, Schmitt argues that this constitutes wealth which should not be confused with money. For Bailly (I992, p. I09), Keynes's revolving fi.|nd may indeed be a stock, but it looks afier a flow of credit. This stage can be incorporated within the second stage as the disbursement of loans — or the use of lines of credit - is used at the same amount as the disbursement of incomes. As Lavoie (I987, p. 9|) points out, the decision to purchase investment-goods implies a long-tenn commitment from finns. It is clear that this demand for money may have a feedback effect on firms. If hoarding is too great that it reduces the firms’ proceeds to the point of not being able to meet their contractual obligations, banks may see the firms as less creditworthy and deny them credit in the subsequent period.

Of course, we acknowledge the existence of a feedback effect although this is not automatic. Saving and financial decisions will affect rates of retum which will then become part of the stock of knowledge for the following period. Idle balances also represent a redistribution of profits fiom consumption-goods firms to investment-goods finns. As hoarded savings increase, the proceeds of consumptiongoods firms decrease, and the relative profit ratio decreases as well: Flcl Fl, The repercussions do not cnd there. Seccareccia (I984) and Graziani (l990a) contend that as idle balances increase and proceeds fall, prices of consumption goods also fall.

This conclusion was also reached by Lavoie (I995, p. 6): “The stock of deposits willingly detained by households is necessarily equal to the consolidated amount of loans to the finns." “There is thus no difference between the outstanding amount of loans and the stock of money" (Lavoie, I992, p. I56). Graziani (l990b) argues along identical lines claming that the existence of a stock of money at the end of the monetary circuit is due to the fact that one or more agents decide on keeping liquid assets.“ Hence, the stock of money is essentially the sum of hoarded saving through time (Lavoie, I987). Parguez (I996, p. 189) agrees: “Wage earners’ saving accounts for the gross deficit of finns as a whole which is that of initial credit finns cannot pay back by their sales-generated gross income." As a result, “the outstanding stock of bank advances and the demand for bank deposits cannot be considered to be independent variables (Lavoie, 1995, p. 4).

T‘M

MM

_”

I

QWMB

mE Z__ §_ AIL‘m°M§"LEE Q°u§ S fmanced by bank credit (Seccareccia, 1994).“ Post-Keynesians must also claim that money is essential and can never be

neutral. In this sense, they must also reject the existence of a natural rate of interest (see below). In the following chapter, it will be argued that the early “post-Keynesian” critique of orthodox monetary theory was essentially done in the context of the first scenario; this implies that they largely accepted the multiplier model as described above, but argued instead that while its causality held up, the parameters of the model were unreliable.

As such, the early

architects of post-Keynesian theory proposed a variation of orthodox monetary theory, as opposed to a rejection of it. They accepted much of the orthodox implications of the multiplier model. In this sense, the early postKeynesian views on money were not very “post-Keynesian,” but rather fairly orthodox. As Lavoie (1992, p. 204) reminds us, “Ironically, the initial

post-Keynesian opposition to the Friedmanian causal interpretation of the Quantity Theory of Money relied primarily on the concept of banking innovation rather than on reverse causation.” Where we disagree with Lavoie (1992, p. 204) is when he claims that “there is no inconsistency

between these views [the portfolio endogeneity approach] and those based on the reverse causation of the money supply.” The Post-Keynesian “Revolutionary” Endogeneity Approach

The above “portfolio endogeneity” approach does not lead us very far. By seeing credit still as emanating from money, it retains much of the orthodox notion of scarcity, especially since the causality between reserves and loans is maintained. Also, it does not address the difference between portfolio decisions versus bank credit. As well, the above simple definition does not attempt to clarify the difference between a stock approach versus a flow

approach to money. The second approach to money endogeneity is largely different. According to this approach, “money” is “credit-money,” but is not the result of portfolio decisions. Rather, money is “created” through the demand for real credit by firms who wish to carry out their production and investment

decisions. It is created ex nihilo by commercial banks, and not simply “introduced” back into the economy though changes in the velocity of money. In this context, money is “credit-driven and demand-determined,” and does not arise from an existing pool of money. The issue of scarcity does not exist. Credit is the result of the real needs of production of finns. In keeping with the spirit of Keynes’s “revolution” and attempt to break

Credit, Money and Post-Keynesian Theory

63

with orthodox thought, it is here labeled the “revolutionary” endogeneity approach. While it was argued above that post-Keynesians have at times flirted with the “portfolio” approach (perhaps as a result of being under the influence of Minsky), they have nonetheless traditionally stressed the second approach. If post-Keynesians claim to be presenting an altemative to neoclassical or orthodox thought, however, they need to embrace fully the second approach. There are a ntunber of disputes among post-Keynesians with respect to the precise definition of money endogeneity — or even on how money becomes endogenous (see Pollin, 1991, 1996; Palley, 1991, 1994).

Here, money

endogeneity is defined according to the second approach, since postKeynesians need to measure themselves according to a non-orthodox fiamework. It is also not sufficient to simply argue that money is linked to production since, as argued above, the portfolio approach can be amended

to include a loans market. As such, post-Keynesians must go further. As such, the defmition of money endogeneity given below keeps within the spirit of Keynes’s “monetary economy of production,” his Treatise on Money, and his post-General Theory articles on finance. The following five characteristics are believed to be consistent with the second, “revolutionary,” approach to endogenous money. 1) reverse causality between money and income, specifying that the causality runs from expected (or desired) income of firms, to the demand for credit, to money and efi°ective income (Robinson, 1956, 1971);” 2) reverse causality between reserves, deposits and loans (Pollin, 1991; Hohnes, 1969; Lavoie, 1992; Eichner, 1987), where reserves are endogenous and have no causal influence on loans; 3) reverse causality between savings and investment (Kregel, 1973; Davidson, 1972; Shapiro, 1977), thereby implying that firms must finance production before any saving is generated; 4) an exogenous base rate of interest (Lavoie, 1996; Hewitson, 1995; Smithin, 1994; Wray, 1995); the base rate of interest is not determined by any market mechanism where demand and supply schedules interact; 5) the money supply is “demand-determined and credit-driven” — “money”

is created ex nihilo, and is not a result of portfolio decisions. In this sense, “money” exists in a continuous circular flow, and is a result of the demand for credit allowing fmns to fulfill their expenditure plans. The supply of credit is endogenous, based on the decisions of commercial banks. “Money” is primarily a flow, created by credit, and extinguished through the reimbursement of loans (Eichner, 1987; Lavoie, 1984, 1992; Parguez, 1984, 1987).

64

Credit, Money and Production

Perhaps the most important argument which can be made in favor of the “revolutionary endogeneity” approach is that money does not enter the economy arbitrarily, or when production and income have already been specified (Lavoie, 1984) — as in the portfolio endogeneity approach. “Money” — or “credit-money” — becomes part and parcel of the production

process, keeping in mind a number of reverse causalities. Thus, for a postKeynesian theory of money claiming to be not only endogenous but also an alternative to orthodox theories of money, all five characteristics listed above must be present; otherwise, what may appear to be a “demanddetermined” theory of money, or a theory of money linked to production,

remains trapped in orthodoxy. This said, it is important to emphasize that credit-money is endogenous because of the real needs of production — money is not endogenous because of some reaction function of the central bank, or because of a behavioral

function defining the choices of the banking system (as a result of an exogenous rate of interest). Credit-money is a flow, created and destroyed, with its stock (the liabilities of commercial banks) fluctuating in the process.

For instance, while New Keynesians argue that production takes time and fmance is needed in the “interregnum,” they nonetheless argue that the pool of fmance is scarce, and must be rationed. In contrast, post-Keynesians emphasize the ex nihilo ability of banks to create credit. In the post-

Keynesian sense, credit is not scarce. POST-KEYNESIANS: SOME CONSENSUS The above analysis has served the purpose of identifying two interpretations of endogenous money, different in as much that one approach is still rooted in the orthodox view of scarcity. It was further argued that credit cannot be included as one of the components of the demand for money function largely because it does not serve the same functions as money. Credit, in this sense, is neither a medium of exchange, a unit of account, nor a store of value. Moreover, credit is not demanded for transactions purposes, nor is it

demanded for precautionary for speculative purposes.”

Rather, credit

serves a more important fimction: it allows production to begin, and money

to be disbursed through the economy. Whether at this stage money is used for any of its functions or purposes is secondary to the issue of production. Having thus defined the properties of money and credit, we now tum to a different purpose: that of identifying the common ideas in post-Keynesian theory, that is, themes which transcend the divisions between horizontalists and structuralists.

Credit, Money and Post-Keynesian Theory

65

As will be seen, there exists, as Deriet and Seccareccia (1996, p. 140) claim, a “high degree of overlapping” between the two approaches. The five characteristics of money endogeneity which were presented in the first chapter will be kept, but other characteristics will be added. It will be shown that structuralists and horizontalists agree on many arguments and aspects of endogeneity. While the differences will also be highlighted, it is argued here that there is more which unites post-Keynesians, than divides them. This is not to say that the difierences are minor or secondary. In fact, it is argued here that while there is perhaps more that unites postKeynesians, the differences are nonetheless important and cany consequences for post-Keynesians. It is now acknowledged that there exist within post-Keynesian thought a nmnber of influences, which in many ways are at the root of the difficulty regarding the discussion of a unified post-Keynesian paradigm. As pointed out by Hamouda and Harcourt (1989), we can identify the lines of argumentation running from Marshall to Keynes (the so-called American post-Keynesians), from Marx to Kalecki (the Kaleckians and in some respects, the circulationists), and from Marx once again through Sraffa (the Neo-Ricardians). This is in line with Kaldor’s (1989, p. 8) observation that Keynes “is a deviationist from Marshall whereas [Kalecki] was a

deviationist fi-om Marx.” These divergent approaches have resulted in many debates on whether these approaches can be successfully combined under the common labeling of “post-Keynesian theory?” For instance, Dow (1990, p. 346) claims that “there is a sense of confirsion, even among some post-Keynesians, as to what post-Keynesianism is,” referring to it as “Babylonian” (see Dow, 1985). Hamouda and I-larcom-t (1989) referred to it as a “horses for courses” and a “portmanteau tenn”; Rao (1987) called it “a curious blend,” and Dutt and Amadeo (1990) claim that “their contributors do not comprise a homogeneous group.” Sawyer (1988, p. 1) for his part claims that postKeynesian theory covers “a considerable assortment of approaches,” while Arestis (1992, p. 86) has said that “its bormdaries have not been precisely defined.” Sebastiani (1994, p. 135) moreover claims that “this state of things renders the post-Keynesian approach a source of stimulating ideas and impulses but, at the same time, makes it difficult to set up a unique theoretical corpus, a new paradigm to match the neoclassical one.” Hamouda and Harcourt (1989, p. 52) have claimed that “attempts to synthesize the strands to see whether a coherent whole emerges [are] a misplaced exercise.”

66

Credit, Money and Production

Moreover, everyone has an opinion on specific areas of comparison between Keynes, Kalecki and Srafi'a. For instance, Sawyer (1995, p. 178) has argued that Keynes’s and Kalecki’s “approaches should be separately

developed and not conflated together,” although Kaldor (1989, p. 8) argues that “there were indeed very strong similarities between Keynes and

Kalecki.” Dow (1988, p. 13) has claimed that Sraffian economics is “methodologically incompatible with the rest of Post-Keynesianism,” this despite Robinson’s (1978, pp. 14-15) argument to the fact that it is “the task of post-Keynesians to reconcile the two. Post-Keynesians must make use of Sraffa to build a type of long-period analysis which will prevent neoclassical equilibrium from oozing back into the General Theory.”

Others (see Arena, 1982, 1987, 1992; Lavoie, 1982, 1992; Hemy, 1982) share Robinson’s enthusiasm, and argue that a “coherent whole” can emerge as long as emphasis is not placed on the “constituted” versions of post-

Keynesian and Sraffian economics, that is the “extremes” of each school of thought. It is therefore suggested that we exclude these versions and concentrate rather on the more “flexible” representatives, or those Arena (1992, p. 587) has called the “dissenting voices” within each school. And

still others, for instance Hewitson (1995), insist that despite these differences, post-Keynesian views are “reconcilable,” and that references to a “school of thought” would not be an exaggeration. While it is not the purpose of this chapter to address the possible

reconciliation between the various “strands” of post-Keynesianism, this chapter — as well as chapters four and five — will nonetheless look at the contributions of some Kaleckians and Srafiians to show that on the issue of money, their views are closely related to those of post-Keynesians. The question of any attempt at reconciling the various schools of thought aside, the tasks of summarizing post-Keynesian theory and identifying its main elements have been numerous. Of course, there is basic agreement on

a variety of issues, such as the importance of uncertainty, the creation of money, and the importance of effective demand (see Brown, 1981; Lavoie, 1992; Arestis, 1992; Dutt and Amadeo, 1990) — although important differences exist on these very same issues. Many — both within and outside the confines of post-Keynesianism - have

argued that post-Keynesians are united primarily in their opposition to neoclassical theory. For instance, Eiclmer (1985, p. 51) once claimed that

“it is less controversial to say what post-Keynesian theory is not than to say what it is. Post-Keynesian theory is not neoclassical theory.”

Arestis (1990, p. 222) claims similarly that “post-Keynesians tend to define their program in a negative way as a reaction to neo-classical economics,” while Sawyer (1988, p. 1) stated that “the unifying feature of

Credit, Money and Post-Keynesian Theory

67

post-Keynesians is the dislike of neoclassical economics.” Minsky (1996, p.

84) refers to post-Keynesian theory simply as “anti laissez-faire.” More recently, Hewitson (1995, p. 285) has made the claim that “there are many points of consensus between post-Keynesian theorists, the most obvious being their united opposition to orthodox monetary theory.” Others who have made this claim are Chick (1973), Davidson (1972), Kregel (1973), Minsky (1975), and Weintraub (1966) (see also a recent article by Walters and Young (1997) and the rebuttal by Arestis, Dunn and Sawyer, 1998). This has also been the criticism advanced by orthodox economists — of both the neoclassical and Keynesian types.” For instance, Solow (1984, p. 137) claims that I don’t see an intellectual connection between a Hyman Minsky

and

someone like Alfred Eichner except that they are all against the same thing, namely the mainstream whatever that is [Post-Keynesians] seems to be mostly a community which knows what it is against, but doesn’t offer anything very systematic that could be described as a positive theory.

Backhouse (1988, p. 40), while claiming that post-Keynesians act as the “conscience of neoclassical economics,” has also stated that it is thus quite reasonable for neoclassical economists to conclude that whilst there may be problems with neoclassical economics, and whilst Post Keynesians may have many interesting things to say, they have not yet managed either to provide a suitable altemative to the neoclassical programme, or to show that the methodology underlying neoclassical economics is misconceived.

Dombusch and Fisher (1987, p. 688) argue that “Post Keynesian economics remains an eclectic collection of ideas, not a systematic challenge.” Finally, Felderer and Homburg (1987, pp. 71-2) have made the claim that “Postkeynesian authors are far from fonning a homogeneous crew

[and

are] somewhat removed from the prevailing orthodoxies and [do] not offer a coherent theory.” This “negativity” can certainly be understood in terms of the critical contributions of Cambridge economics to the capital critique. According to Eichner and Kregel (1975, p. 1294), Part of the problem has been the diversion created by the “Cambridge controversy” over the theory of capital. While it is true that some of the elements of post-Keynesian theory became better known through the criticisms by Cambridge, England, of the treatment of capital in the neoclassical growth models favored by Cambridge, Mass., the debate has

68

Credit, Money and Production

nonetheless lefi the misleading impression that the adversaries fi-om across the Atlantic had only a negative critique to offer.

As for monetary theory, the “negative” aspect of post-Keynesian theory is rooted perhaps in the fact that post-Keynesian theory of money emerged out of theoretical debates between Friedman and post-Keynesians — as suggested in the first chapter. Post-Keynesians — except perhaps for Robinson and Kalm, as was discussed earlier — were indeed part of the antiFriedman response, instead of developing a positive contribution to the

field. It is indeed a fact that post-Keynesians reject neoclassical theory — both in

methodology and theory. But to claim that there is little consensus within this school (or the schools) is surely to exaggerate the divisions. There is in

fact much agreement on many aspects of macro and micro economics: mark-up pricing, distribution, fmance, endogenous money, the workings of

a capitalist economy, and uncertainty — just to name a few. There are also agreements on methodological questions, as emphasized by Dow (1990, 1992) and Arestis (1990). The differences arise more when dealing with the specifics of each of these elements. But surely this is a valid criticism of neoclassical — and New Keynesian — theory as well. Overall therefore, Arestis (1990, p. 221) has claimed that post-Keynesians

offer “a positive statement of methodology, ideology and content [and] provide an altemative paradigm to orthodox economics.” In the specific area of monetary theory, it has generally been accepted that the theory of the endogeneity of money is a central component of postKeynesian theory. In fact, many post-Keynesians would argue that it forms the “coping stone” of post-Keynesian theory, since post-Keynesians attempt to continue Keynes’s work of dealing with a magnetized capitalist economy of production. On this issue, a vast consensus exists. For instance, Cottrell

(1994, p. 1 l 1) calls money endogeneity a “central theme” of post-Keynesian economics, while for Gedeon (l985—6, p. 208), it is a “major theme.” For

Dalziel (1995, p. 311), it is a “major distinguishing characteristic,” and for Dow (1984), it is a “central element.”

For Davidson (1988, p. 370),

endogenous money plays “a unique and important ro1e,” while for Screpanti (1995, p. 1), it is a “fundamental tenet,” and for Palley (1996, p. 13), it

fonns the “comerstone” of post-Keynesian thought. There are a number of important convergences between the postKeynesian protagonists when it comes to money endogeneity. Overall, post-Keynesians set their theory within Keynes’s “monetary theory of production” keeping in mind Keynes’s long struggle of escape. As such,

post-Keynesian theory — especially in the area of monetary theory — seeks to “push Keynes beyond himself [by developing] the radical elements in Keynes and to discard the conservative” (Cottrell, 1994, p. 588).“

Credit, Money and Post-Keynesian Theory

69

It is argued here that the following eight elements are shared by both the structuralists and the horizontalists. The Exogeneity of the Base Rate of Interest

There is now wide-ranging acceptance in post-Keynesian theory around the notion of the base rate of interest as an exogenously determined variable, set by the central bank. In this sense, the control variable of central bankers is the rate of interest as opposed to the money supply - a price and not a quantity.” Although Wray (1990) did once argue that the base rate of interest was endogenous, he now endorses the exogeneity issue. In Wray (1995, p. 279), the author now claims that “I will agree that the central bank can ‘administer’ any particular interest rate.”

Similarly, Hewitson (1995, p. 290) argues that “it is the discount rate rather than base money which is the control variable.” Rogers (1989, p. 253) claims that “the interest rate reflects psychological, institutional, and other historical factors which cannot be specified a priori.” This is the position of Pasinetti (1974, p. 44) who argues that the base rate is “detennined exogenously with respect to the income generation process. Whether, in particular, liquidity preference, or anything else detennines it, is

entirely immaterial.” The base rate therefore is seen as a “non-market conventional phenomenon” (Deriet and Seccareccia, 1996, p. 140). While this is clearly a rejection of the arguments put forth by Keynes in the General Theory, it is nonetheless the position which Keynes later defended, following the publication of the General Theory, as is argued below (see Rochon, 1997). Smithin (1994, p. 109) summarized the post-Keynesian position on

interest rates as follows: “In the end, it does seem to be agreed by economists from a wide spectrum of opinion that the endogenous money approach with policy-determined short rates does reflect the contemporary institutional realities.” The Rejection of the Natural Rate of Interest While money is endogenous, horizontalists and structuralists also reject the existence of a “natural” rate of interest. Following Keynes, post-Keynesians reject the theory of loanable funds according to which the rate of interest is determined by the forces of productivity and thrifi, that is investment and saving. For this reason, Shackle (1983, p. 366) calls this contribution of the General Theory, the “most momentous, radical and decisive service.” It also undermines the economy’s lack of tendency toward a final position of long-run equilibrium.

70

Credit, Money and Production

The rejection of the “natural rate hypothesis” is shared by Smithin (1994),

Lavoie (1996), Rogers (1989), and Seccareccia (1990, 1994). This is an attempt to differentiate post-Keynesian theory of money endogeneity from

other theories of endogenous money developed by orthodox economists. For instance, both Wicksell and Hayek recognized the endogenous nature of

the credit supply, while, as Lavoie (1996a) makes the point, money endogeneity is also shared by orthodox French economists of the overdraft school. As outlined in the previous chapter, “portfolio endogeneity” is also recognized by orthodox economists of the multiplier type.

This is analogous to Seccareccia’s (1994, p. 71, n. ll) assessment, to the effect that contrary to Moore (1988) who has argued that it is money-supply endogeneity

which fundamentally distinguishes the neoclassical from the post-Keynesian conception of money, one would like to think that there is something substantially more than merely the endogeneity/exogeneity issue that separates

them. Alter all, in addition to Hayek and other Austrian writers, modem neoclassical theorists of the real business cycle, as well as contemporary “free banking” advocates, also rely on a specific endogeneity of the money supply and, in that sense, are indirectly espousing a view of the money system that is

not monetarist and is more like that of Hayek.

The rejection of the natural rate can be traced back to Keynes (vii, pp. 1834) who, in the General Theory, argues that “Saving and Investment are the determinates of the system, not the determinants. They are the twin results of the system’s determinants, namely, the propensity to consume, the

schedule of the marginal efiiciency of capital and the rate of interest.” Later, Keynes (vii, pp. 242-4) specifically rejects the natural rate hypothesis: “I am no longer of the opinion that the concept of a ‘natural’ rate of interest, which previously seemed to me a most promising idea, has

anything very useful or significant to contribute to our analysis.” To this statement, Rogers (1995, p. 2) adds, “The existence of money is therefore not sufficient to generate post Keynesian results — although it is

necessary.

The classic and market forces required to determine the natural

rate of interest simply do not exist and/or cannot interact in such a way as to

generate the optimum or natural rate.” As Kohn (1986, p. 1224) claims, All the neoclassicals, at least since Wicksell, had held a monetary theory of the rate of interest, believing the market rate of interest to be proximately detennined by the supply and the demand for loans of money (the “loanable

funds” theory). What was new about Keynes’s liquidity preference theory was not, therefore, that it was a monetary theory, but rather that it denied the existence of any “normal” or “natural” rate relative to which the market rate

moved.

Credit, Money and Post-Keynesian Theory

71

As such, “A comprehensive critique ofthe classical or loanable funds theory of the rate of interest therefore requires a compelling critique of the concept of the natural rate of interest” (Rogers, 1995, p. 12). This is best done, as argued by Lavoie (1996) by positing an exogenously determined base rate of interest. According to Lavoie (1996, p. 6) (see also Smithin, 1997), Those post-Keynesian or lnstitutionalist authors who claim to be describing monetary production economies are implicitly defining models where the natural rate of interest is irrelevant. To define the proper foundations of a post-Keynesian monetary analysis, one must thus simultaneously and explicitly adopt an endogenous money approach (with generalized liquidity preference), and discard the notion of the natural rate hypothesis.

The Interest Rate Mark-Up The loan rate of interest can be thought of as a mark-up over the cost of

fimds — Moore (l988c, p. 292) for instance, claims that the rate is a mark-up over the deposit rate. Moreover, both horizontalists and structuralists agree

that the mark-up need not be fixed.” In fact, we can write:

i, =(l+m,,)ic where i, is the loan rate of interest, ic is the cost of obtaining additional ftmds, and m,, is a mark-up reflecting the Kaleckian degree of monopoly

(see Seccareccia, 1996), which can also be seen as resulting from corrmrercial banks’ target rate of retum.“' 3’ There are still nonetheless a number of debates surrounding the mark-up approach. For instance, there is disagreement on what precisely influences

the mark-up (Palley, 1991; Deriet and Seccareccia, 1996), and whether it is positively related to the demand for loans. Accordingly, Smithin (1994, p. 110) argues, “The remaining debating point is, then, whether the setting of short rates by the central bank uniquely determines the entire structure of interest rates including long rates, or whether there is still room for an altemative, but still monetary, principle by which interest rates other than those directly controlled by the central bank can be detennined.” However, it will be argued in this chapter that a varying mark-up is perfectly compatible with a horizontalist position. Hence, the mark-up approach used here is neither in support nor in defense of either postKeynesian positions (Hewitson, 1995, p. 292). The mark-up will be influenced largely by the banks’ own target rate of profit, an imposed creditworthy criteria on firms, and the need for firms to generate profits in

72

Credit, Money and Production

order to pay back existing loans. There is no doubt that the mark-up can vary as the economy heats up and cools down. The Role of Banks and Credit-Money

Post-Keynesians reject the notion of conunodity money in favor of a system of credit-money. They also necessarily reject the notion of an exogenous credit supply. As explained by Moore (l988c, p. 291), “The mainstream paradigm of a money stock exogenously determined by the central bank

is appropriate only to a world of commodity of fiat money. The supply of credit is different. The supply of credit money does not exist independently

of the demand for it” (see also Moore, 1986). As Wray (1996, pp. 445-6) has insisted, “ln reality, credit-money is the ‘natural’ form of money

commodity money was developed, at best in part, to deal with the problem of the possibility of default on credit money.” In a credit-money system, the supply and the demand for credit-money are interdependent. Whether the best way of representing this is by using a

supply and demand curve analysis is debatable. The whole approach is lefi over from commodity money, and especially the reference to a vertical supply curve. By using the supply and demand apparatus, we end up falling on to the orthodox terrain, and debating over the shapes and positions of the

curves. While retaining this analysis has the obvious advantage of framing the argument in a vocabulary which is clearly understood, post-Keynesians should be reminded of its origins and limitations. Moreover, post-Keynesians should argue in terms of the demand and supply of bank credit, not money. Credit is the foremost important argument, since, as Arestis and Eiclmer (1988, p. 1005) tell us, the “amount of fimds in circulation depends on the amount of loans made by banks and

thus on the demand for credit.” For Wray (l989a, p. 154), banks “give purchasing power to entrepreneurs, financing irmovations and enabling economic growth and development.” As a result, Arestis and Driver (1988, p. 128) have emphasized the notion that money is “essentially an output

variable responding to changes in the behaviour of private economic units [banks] rather than the behaviour of the monetary authorities.” The Reverse Causality among Reserves, Deposits and Loans

As a result of the supply of credit being demand-determined, the general causality between reserves, deposits and lending ability of banks is reversed in post-Keynesian theory. As Holmes (1969, p. 73) claims, “ln the real world banks extend credit, creating deposits in the process, and look for the reserves later.” Similarly, Hewitson (1995, p. 287) has argued that “loans

Credit, Money and Post-Keynesian Theory

73

are made, deposits are created, and banks only later seek the reserve assets required to support these deposits and meet reserve requirements.” Eichner (1987, p. 849) argues, “An increase in bank loans, by increasing the amount of bank deposits, will add to the banking system’s required reserves.” Also, banks tum either to the central bank for reserves, or “use the wholesale markets, where they are price-takers and quantity-setters, to lend or borrow any net excess sources or uses of funds” (Moore, 1985, p. 106; see also Pollin, 1991, 1996; and Dymski and Pollin, 1992). As an obvious result, post-Keynesians see the supply of high-powered money - and hence the monetary base — as an endogenous variable (Moore, 1988c), adjusting to the needs of trade. According to Forman et al. (1985, p. 30), “The amount of bank reserves, and thus the monetary base, is not the exogenously determined variable assumed in both orthodox Keynesian and monetarist models but instead depends on the level of nominal income.” Moreover, “Since reserves are ordinarily supplied endogenously on demand, they have no causal role in the money supply process” (Moore,

1989!», p. 12)."

As such, post-Keynesians rank highly the role of lender-of-last-resort of the central bank in its attempt to prevent a financial crisis. The point was well put by Forman et al. (1985, p. 30): “The central bank, in order to maintain the liquidity of the financial system, is forced to purchase govemment securities in the open market so as to accommodate, at least in part, the need for additional credit as the pace of economic activity quickens.” The Dynamic Credit-Supply Curve Differences between the slope of the credit supply curve can be reduced to differences regarding the “static” versus “dynamic” nature of the supply curve (Palley, 1991, p. 400). It can be argued (see Lavoie, 1996) that both horizontalists and structuralists agree that the short-term - or static - credit supply curve is horizontal. For instance, Palley (1991, p. 400) claims that “at any moment the static loan supply schedule is horizontal.” This is the horizontalist position. On the other hand, horizontalists concede the possibility of a longer-term — or dynamic - supply curve, upward sloping in credit-interest rate space, although the specific slope can also well be downward-sloping. For instance, Lavoie (1996, p. 280) states that “horizontalists would capture this upward sloping curve by positing the existence of a set of horizontal supply curves.

[which] when money demand is taken into

consideration, will constitute a positively-sloped money supply curve. The upward sloping curve is thus a special case, based on a particularfeedback

74

Credit, Money and Production

rule, ofthe more general horizontal depiction.” Moore (1988, p. 265, fn. 9) talks of a “longer-terrn ‘money supply function’ upward sloping with respect

to interest rates.” Arestis and Howells (1996, p. 541) have also stated that “such a curve may be horizontal but it may not; it all depends on the interaction between the supply curve and the demand curve.”

But as Lavoie (1996) is careful to point out, this upward sloping curve is not an endogenous feature of the system, but rather a result of the non accommodating behavior of the central bank.” In fact, the dynamic supply curve can take several fonns. It can be perfectly horizontal if the central bank decides to peg the rate of interest, or it can even be downward-sloping if the rate of interest is falling. This specific point is made by Lavoie (1996), but also by Palley (1987-8, p. 296): “However, expansionary [openmarket operations] may be interpreted as a downward shift of the loan supply schedule and generate a negative correlation between nominal output

and interest rates.” One point which must be emphasized is that the dynamic supply curve is not the same upward sloping curve posited by structuralists such as Pollin and Wray. It is a dynamic curve, and can take various shapes, depending on the behavior of the central bank and the interaction between the supply and

demand curves for credit. As such, Moore (1988, p. 265) has argued that it is “in general strictly undefmed, since it is not independent of credit and money demand forces?” The Importance of Realism For post-Keynesians, the starting point of economic analysis and monetary theory, is realism. In this respect, post-Keynesians have generally sought to

construct theories based on realistic assumptions and hypotheses, i.e, theories which begin with a “vigilant observation of the real world” — or as Kaldor suggests — with “stylized facts.” Realism thus sets post-Keynesians apart fiom their orthodox counterpart. As Dow (1990, p. 347) has repeatedly argued, realism is “a particular philosophy of science, i.e., it defines the foundation for theorizing.” This leads naturally to Caldwell’s

observation that “post Keynesians value explanation in economics more than they do predictions” — referring to post-Keynesians’ preference for inductive, as opposed to deductive reasoning. As such, Sawyer (1995, p. 66), argues that “Post-Keynesian economists have generally sought to construct theories that are built on realistic assumptions derived fi'om observations of the world.”

This is why post-Keynesians argue that the existence and types of institutions will greatly influence the way in which markets operate (Nell,

1992). Accordingly, Hodgson (1989, p. 110) claims that “if we make the

Credit, Money and Post-Keynesian Theory

75

less rigid assumption that individual tastes and preferences are malleable and will change or adapt, then the objectives and behaviours of agents can be moulded or reinforced by institutions.” This is also the observation of Fazzari and Minsky (1984, p. 106) who claim that post-Keynesian theory is “institutionally specific.” This is what Hewitson (1995) has called the “legacy of Keynes.” As such, the study of monetary policy and the behavior of money cannot be analyzed without referring to the current monetary institutions (Davidson, 1972). Post-Keynesian theory therefore stands in contrast to orthodox theory which integrates money as a simple afierthought once production has taken place, irrespective ofthe institutional environment. The Emphasis on Uncertainty In analyzing Keynes’s “monetary theory of production,” post-Keynesians have emphasized the fact that production takes time, and that firms need access to finance in order to bridge the time between the outlays of costs and the inflow of revenues (Arestis and Eichner, 1988, p. 1009). However, production implies also the presence of “fimdarnental uncertainty” — in Keynes’s sense - although Wray (1996, p. 772) prefers talking about “existential uncertainty,” and Dymski (1996, p. 381) uses the expression “exogenous uncertainty.” The post-Keynesian emphasis on uncertainty is ftmdamentally different than the orthodox views on risk, or probability distributions. According to Keynes (xiv, p. 114), fundamental uncertainty arises because we “simply do not know” the firture. In essence, this implies that in historical time analysis (see Asimakopulos, 1991), “The past is given and cannot be changed, and that its future is uncertain and cannot be known” (Moore, 1979, p. 121). The emphasis on uncertainty however has also served to reinforce the existing differences between structuralists and horizontalists. This is due to different views on how uncertainty ought to be introduced — or dealt with — in monetary theory. Both approaches therefore can be said to emphasize difi'erent aspects of Keynes’s monetary writings. In essence, structuralists have placed the importance of uncertainty squarely on firms and decisions to invest, and hence the need to keep money as “a link between the past and the present and also between the present and the future” (Arestis, 1987, p. 1). As a result, money is seen as a stock and foremost a store of value. This view is verified by Wray (1996, p. 440) who

criticizes post-Keynesians for having too much “emphasized money as a stock.” As Dymski (1996, p. 381) claims, “Exogenous uncertainty creates a need for a store of valuei agents’ willingness to undertake tirne-using

76

Credit, Money and Production

activities may entail a desire simply to hold resources in reserve. This desire was termed liquidity preference by Keynes.” Money thus becomes a “means of storing value in the face of exogenous uncertainty” (ibid., p. 383). In the horizontalist tradition, uncertainty is placed rather on banks’ decisions to lend. This is keeping in line with Keynes’s emphasis on the role of banks as holding the “key” to the expansion of the economy. As a result, the emphasis is placed on bankers’ decisions to extend credit, based on their “animal spirits.” As a result, credit is thus seen as a flow and a medium of exchange. This distinction allows horizontalists to discuss a number of institutional characteristics of banking. By placing uncertainty in the context of the behavior of commercial banks, horizontalists are thus able to discuss the reasons behind changes in the target rate of retum, changing collateral requirements through the cycle, and why banks refuse credit to some and not to others. This different emphasis leads also to divergent views on the causal versus residual role of liquidity preference, although this issue will be dealt with in more detail in chapter four. It suffices to mention here that despite Dow and Dow’s (1989) observation, horizontalism and liquidity are not incompatible. While the above list of characteristics shows that there exists consensus on many aspects of monetary policy between the horizontalist and structuralist strands of post-Keynesian monetary theory, there exist nonetheless important differences as well, many of which come from the same issues discussed above. These differences were put in perspective succinctly in the following

passage by Rousseas (1985, pp. 59-60): As endogenous as the money supply may be, however, it does not mean that accommodation to the “needs of trade” takes place smoothly or equitably or that it is without cost in terms of distortions in the flow of credit. The problem of controlling the paths it takes by controlling the flow of credit through the economy remains a problem most post-Keynesian monetary theorists have ignored.

Many post-Keynesians argue that the differences between structuralists and horizontalists “are not mere squibbles” (Pollin, 1996, p. 511), leading one post-Keynesian to argue that “the substantive distinctions between the two approaches are clear and sufiiciently important” (Pollin, 1991, p. 368), and

that “the two approaches cannot therefore be reconciled theoretically" (Pollin, 1996, p. 505). According to Trautwein (1995), the differences among the various postKeynesian protagonists are important enough that it is debatable whether a

Credit, Money and Post-Keynesian Theory

77

surmnary of the main theoretical ideas within a single line of thought is possible. Others however claim that the differences are rather a matter of emphasis than content. According to Lavoie (1995, p. 1), the controversies among post-Keynesians “reflect differences of opinion or emphasis as what should be considered primary rather than secondary factors.” Similarly, Moore (1991, p. 405) has claimed that “the disagreement between the two views is in no sense profound. It must be stressed that the areas of common agreement are much broader than the areas of disagreement.” As such, in the words of Eichner and Kregel (I975, p. 1294), “Post Keynesian theory has the potential for becoming a comprehensive, positive altemative to the prevailing neoclassical paradigm” in the Kulmian sense. Overall, it is argued here that horizontalists and structuralists do share common elements, sufficiently to claim that they form an “integral whole.” What unites them is foremost a view on the disequilibrium nature of markets, on how they operate, on how “tranquillity breeds instability” — to paraphrase Minsky, and also on how money enters the economy through the production process. Hence, before claiming that the two approaches are beyond reconciliation, it is perhaps best to recall Dow’s (1988, p. 2) claim that “methodology then helps to define the boundaries of post Keynesian thought, since it explains the nature of unity underlying apparently disparate theoretical approaches. It then defines the cohesion among the varied content of the school. [Post Keynesians] share a view of the proper way in which to approach economic theories.” A particular school of thought therefore, can be defined “by its unique world view, and the unique methodology by which it is expressed” (Dow and Earl, 1984, p. 149).

CONCLUSION The purpose of this chapter was two-fold. First, it was to bring much needed clarity to the notions of credit and money. This was done essentially because there exists in the post-Keynesian literature considerable confusion as to what credit and money really are. The chapter presented two theories of endogenous money, one based on the endogenous creation of credit flows, the other on portfolio analysis and the changing value of the velocity of money. While both have in common the fact that the quantity of money in circulation is greater than the desired level set by the central bank, the fact remains that only the first one can be deemed post-Keynesian, hence “revolutionary.”

78

Credit. Money and Production

The chapter then went on to argue that post-Keynesians have at various times emphasized both approaches thereby leading to considerable confusion. This is because post-Keynesians have been unable to clearly

diflerentiate between credit and money. For many, credit is a fourth motive for holding money - either an addendum to the transaction motive or an entirely independent motive. This suggests, however, that fmance as money finances investment. This poses obvious problems. Rather, it was claimed that credit is entirely different than money: it creates money and has purposes and properties of its own. Credit — as opposed to money - is the sine qua non of the system. Finally, this chapter argued that there is much cohesion within the postKeynesian school. Post-Keynesians are fairly united in their view of market

economies and the real world.

Where they differ — as will be shown

throughout - is on particulars. But surely this is just as true for orthodox or

neoclassical economics, and not just on how many times one should brush his or her teeth. This chapter has also attempted to show that while criticism may have been a characteristic of the earlier contributions, this is no longer the case. As Ferri and Minsky (1989, p. 123; emphasis added) claim, “The initial

unifying element in the thinking of post-Keynesian economists was negative.” Hence, contrary to claims made by many of its critics, postKeynesian monetary theory evolved considerably over the years, so much that it is no longer appropriate to refer to it as a negative theory. This chapter - as well as the rest of this work — has hopefully demonstrated that post-Keynesian theory has become — in Solow’s words - a “systematic” theory that can indeed be described as “positive.”

NOTES l.

2. 3. 4.

American post-Keynesians, following Davidson (1965) have remedied this problem by adopting the IS-LM model in two incarnations. First by recognizing the interdependence of the IS and LM curves; second by drawing the LM curve as horizontal. Neither approach is, however, satisfactory. The IS-LM model has no place in post-Keynesian macroeconomics. Keynes defines the finance motive as a motive for “holding cash." This is a misleading label and leads to confusion. It is best to interpret the finance motive as a demand for bank credit, as Keynes came to realize (see Rochon, 1997). While the structuralists emphasize the General Theory, the horizontalists emphasize rather the Treatise on Money and Keynes's post-General Theory articles and contributions. Although it is often claimed that post-Keynesians ought to tum to Kalecki for inspiration on this issue, it does not mean that Keynes regarded the issue as unimportant. In fact, while income distribution is not frilly developed in the General Theory, Keynes was nonetheless aware of its importance, as is made clear in some of his comments regarding

Credit, Money and Post-Keynesian Theory

79

reviews of the General Theory. For instance, on the policy side, Keynes (xiv, pp. 1617) wrote to Hawtrey claiming that he favored “a scheme of direct taxation in order to redistribute incomes in such a way as to increase the propensity to consume.” In a brief article published in the Review of Economic Statistics, in 1939, (see xiv, pp. 270-1), Keynes brings to task a reviewer of the General Theory for claiming that he does not deal with distribution. Claiming that he may deal with an aggregate propensity to consume, Keynes claims that he does recognize that this aggregate value is determined “partly [by] the subjective needs and the psychological propensities and habits of the individuals composing it and the principles on which the income is divided between them (which may sufier modifications as output is increased" — a passage taken directly out of the General Theory (vii, pp. 90-1). Keynes was in fact so bothered by this point that he wrote to Pigou for advice, arguing that the reviews were all based on “obvious misunderstandings of what 1 say." Moreover, in drafts to the General Theory, Keynes (xiii, p. 404) argues that “the determination of 0 [output] is our goal. For the maintenance of O at an optimum level should be one of the twin objects of the management of an economic system (the other object being the maintenance of an optimum distribution of income." Here, Keynes had not realized, as is made clear in Kaleckian models of growth and distribution, that an “optimum” distribution of income leads to higher levels of output. Keynes falls short of linking the theory of distribution to the theory of endogenous money. Many post-Keynesians, while correctly emphasizing banks, do not realize that the existence of banks alone does not guarantee a theory of money endogeneity and reversed causation between saving and investment. As will be seen in chapter seven, New Keynesians have adopted a theory placing “special” bank credit at the center of the economic process. However, they still claim that saving determines investment, and deposits create loans. While Arestis (1992, p. 179) argues that endogenous money “can be seen as a step towards satisfying one of the important aims of post-Keynesian economics which is to complete the unfinished General Theory," it is argued in firrther chapters that Keynes himself finished his “unfinished” General Theory by accepting an exogenous rate of interest and money endogeneity after the publication of the General Theory. See Rochon (1997). As Harcourt and Sardoni (1996, p. 4) say, Davidson “should receive considerable credit for putting the finance motive centrally back on the agenda." The finance motive is treated here as a demand for bank credit. Initially, Keynes may have defined it as a motive for “holding cash," he clearly came to see it as bank credit, as is discussed later. This does not mean that he accepted Robertson's criticism. However, in this context of reverse causality between deposits and loans, there is no problem in seeing the fmance motive as back credit. It is not seen in terms of loanable funds. In many respects, Keynes shared this opinion as well. See also Shackle (1967). The expression “non bank public” is also ambiguous since it does not differentiate between firms and households who play very different roles in the economic process and the creation of credit versus money. Similarly, St-Hill (I995, p. 37, emphasis added) claims that “endogeneity of the money stock might be regarded as the quintessence of post Keynesian monetary theory." Here again, while the distinction is a welcomed one, it should be emphasized that it remains rather imprecise. “Credit” is not a medium of exchange. Rather, it is created in order to allow production to proceed. The expression “medium of production” may be a better choice. On an interesting note, while post-Keynesians consider credit as a component of money, New Keynesians clearly differentiate between the two concepts. In fact, Gertler (I988,

Credit, Money and Production p. 560), in a criticism aimed at Keynesians but which could nonetheless apply to some degree to post-Keynesians, criticizes Keynesians for devoting too much “attention to Keynes’s liquidity preference theory which emphasized the importance of money, as opposed to credit." Despite this conclusion, New Keynesians nonetheless argue that money determines credit. The New Keynesian view will be fully explored later. This view is taken because most post-Keynesians do not include wages in the finance motive. In this respect, since credit is used exclusively for the purchase of capital goods, then it is needed to carry out a transaction. Hewitson (I995) is perhaps one of the only structuralists who sees the finance motive as an important argument even when production is constant or the economy is in steadystate equilibrium. Bibow (I995, p. 659) also claims that “the cnix of the matter is that the finance motive is a concept of change and we end up in utter confusion when we do not fully embrace the implications of the changes involved." Credit in this sense is “productive” credit, i.e. it is used for production. “Household” credit, on the other hand, is used for the purchase of goods by households, not fimis. While household credit is ignored here, the issue will arise again in chapter seven in exploring the debate between Howells, Moore and Lavoie on the demand for money. Another example of confiision is Hewitson’s (I995, p. 286) claim that “money or credit must precede production.“ Here, it is clear that only credit precedes production, while money follows it. A theory of banking activity is developed in chapter eight. Dow (I997, p. 7l) also identifies “two senses of endogeneity" in the post-Keynesian literature. Her dichotimization, however, is based on a misunderstanding of the horizontalist position which she assumes to mean that banks meet all demand for credit. According to Dow (I997, p. 7|), horizontalism applies to situations where the “money supply is endogenous to the private sector as a whole (detennined by the non-bank public through the banks)," whereas the structuralist, liquidity preference approach applies when the “money supply is endogenous to (i.e. determined by) only the banks." As will be seen in chapters five and eight, this is inconect. Horizontalism is perfectly compatible with a “fringe of unsatisfied customers." ln this sense, horizontalism becomes compatible with both of Dow's definition. ln this sense, a better differentiation between the two theories of endogenous money in the post-Keynesian literature is the one proposed here, between credit and portfolio approaches. As we will also see in chapter seven, this is also the view of New Keynesians. ln this sense, we disagree with Dow (I997, p. 63) who claims that “for the money supply to be regarded as exogenous to the private sector, there must be a belief in the capacity

of the monetary authorities to control it." As will be argued in chapter seven, New Keynesians essentially agree that the money supply is exogenous, but given financial innovations, the central bank does not have full control over it. Rather, they claim the central bank can regulate credit. Dow’s definition here is too vague. This view is adopted by monetarists, who argue that volatile changes in the quantity of money in circulation can be attributed to changes in the multiplier in the short run. ln the long run, the central bank retains full control over the supply of money. On a related note, it is interesting to see that Dow ( I997) also argues that in the General Theory, Keynes did not make the money supply exogenous. Rather, she claims that in the General Theory, Keynes chose to make the money supply “given.” But how can money be given within a post-Keynesian framework if output is expanding? According to Trautwein (I995, p. l4), Hayek insisted “that credit expansion must result in a crisis and that the crisis is a re-equilibraling process, a ‘cure‘ of distortions in the structures of prices and production that has to be waited out.“ On this issue, Cottrell (I986, p.

I8) claims that “taking a position on the

exogeneity/endogeneity issue does not thereby resolve the issue of the causal relations

Credit, Money and Post-Keynesian Theory

81

between money and income, since endogenous money can cause income." Cottrell does not differentiate, as it is here, between the expected (or desired) income and efiecttve income. It is recognized that credit may be obtained for speculative purposes. This issue is not addressed here. Moreover, some have attempted to reconcile post-Keynesians with Austrians, with lnstitutionalists (Hodgson, I989), and with Evolutionists (Foster, I989). These issues are, however, beyond the scope of the present research. Overall, as Chick (I995, pp. 20—l) observes, orthodox criticism usually boils down to the notion that post-Keynesian theory is neither coherent, scientific, moderrn theory, nor economics. Cottrell's reminder will become important in chapter five when the more orthodox elements of post-Keynesian monetary theory are examined. Increasingly, Sraffians, long regarded as ignoring the monetary side of their analysis, are recognizing the exogeneity of the rate of interest. For instance, Pivetti (l99l, p. l6) argues that “it is the level of the rates of interest, rather than a given quantity of money, which constitutes the primary object of monetary policy.” Rousseas (l985a, p. I42), however, claims that the mark-up is “relatively stable.” Palley (l99l, p. 398) wrongly suggests that Moore assumes a fixed mark-up. This is not so, and a rejection of a fixed mark-up in no way implies a rejection of Moore’s horizontalism. In fact, it implies a rejection only of Rousseas's approach — himself a proponent of an upward-sloping credit supply curve. We find a similar interpretation in Panico (I985), where ii = i, + 1| , where ti is a liquidity premium. ln Epstein (I990, p. 6l9), the mark-up can vary because of “economic and political reasons:” “The banking system [is] oligopolistically structured, where banks attempt to coordinate prime rate changes for both economic and political reasons. To do so, they use the discount rate as a signal to coordinate their changes.” As will be argued in chapters five and six, this is not quite exact as many structuralists argue that the central bank does have the ability of constraining bank lending, thereby acknowledging a causal role for reserves. lt is argued in chapter five that a non-accommodating behavior of the central bank is also compatible with horizontalism On this point, however, there is not complete agreement. For some, the horizontalist and structuralist positions would coincide in the long run, except that the money supply curve would be horizontal, not upward-sloping. For Screpanti (I997, p. 2), while agreeing that in the very short run the supply schedule is horizontal, argues that “the money supply can be considered to be accommodating in the very long run, since monetary authorities are usually reluctant to severally stifle the banking system and the economy.”

3. The Early Views of “Endogenous” Money: Minsky, Kaldor and Tobin In the previous chapter, it was argued that there exists in the post-

Keynesian literature a considerable amount of confusion surrounding the issues of credit and money. One immediate consequence of this confusion is that two very different defmitions of endogenous money have arisen. A necessary question is where does this confusion arise? The objective of this chapter is precisely to examine in greater detail the origins of the post-Keynesian theory of endogenous money. In doing so, there will be no attempt to analyze the views of the “Real Bills” doctrine and its similarities with post-Keynesian theory. This has already been done (for instance, see Wray, 1990). The scope of the analysis will be limited to the post-Keynes revival of endogenous money.

This chapter therefore will focus its attention on the more modern roots of post-Keynesian monetary theory. Its purpose is to analyze the early works

of two monetary post-Keynesians most often associated with a theory of endogenous money: Hyman Minsky (l957a, 1957b) and Nicholas Kaldor (I955, I958, 1964), and to compare them with those of another “Keynesian”: James Tobin. The standard interpretation of Minsky’s and Kaldor’s respective contributions is that both post-Keynesians adopted fairly early a theory of money endogeneity which deviated little from the views they would hold many years later. As for Tobin, Wray (1990) also includes him as an advocate of endogenous money. As this chapter will show, this traditional interpretation of Minsky, Kaldor and Tobin needs to be reconsidered. Far from having a theory of endogenous money in the post-Keynesian or revolutionary sense, these authors adopted a weakened monetarist view. In Minsky’s case, the conclusion is more far reaching, and it is argued that early in his career, Minsky was a proponent of the loanable funds approach.

82

The Early Views ofEndogenous Money

83

THE CHOICE OF MINSKY, KALDOR AND TOBIN Post-Keynesians generally agree that the roots of the post-Keynes revival of endogenous money can be traced back to Hyman Minsky and Nicholas Kaldor. For instance, Niggle (I991, p. 140) claims that “Hyman Minsky in the United States, and Nicholas Kaldor in the United Kingdom presented formal models in which the money supply was seen as endogenous.” Similarly, Wray ( 1992b, p. 161) claims that “Minsky’s vision requires a sophisticated analysis of the endogenous creation of money to finance spending Minsky adopted an endogenous approach to money in his early work, and that this approach is consistent with his fmancial instability hypothesis.” Dymski and Pollin (1992, p. 41) have also argued along these lines, claiming that money endogeneity “originated with one of his first published papers (Minsky, l957)” - referring to Minsky’s article in the Quarterly Journal ofEconomics.

As for Kaldor, Musella and Panico (1993), and Targetti (I992) all argue that he introduced money endogeneity as early as 1939 - views which are consistent, according to these authors, with those which he would develop fi'om 1970 onwards. This stands in contrast to the views of others, for instance Zarmoni and McKenna (1981, p. 479), who claim that Kaldor introduced endogenous money only in I970. In this chapter, it is argued rather that both of these standard interpretations about Minsky and Kaldor should be revised. This is not to say that these authors did not change their views later on in their respective writings. Indeed, Minsky and Kaldor (the latter perhaps more than the former) did in fact develop respective theories of money endogeneity. However, it is ascertained below that neither author’s early views are

consistent with a revolutionary theory of endogenous money. The choice of Minsky, Kaldor and Tobin is done largely for two reasons. First, all three authors are considered by many post-Keynesians as part of the “post-Keynesian response to Friedman” (Hewitson, I995, p. 285). Tobin, as argued by Cottrell (1994, p. 587), is considered a “‘postKeynesian’ in a more extended sense.” By their own admission, the postKeynesian revival of money was tailored as a response to Friedman’s brand

of monetarism. All three authors published important papers criticizing various components of Friedman’s analysis of money and inflation. Yet, being a “response” to Friedman, their analysis still remained in the Friedmanian framework, choosing to criticize the direct link between money and income, rather than the exogenous nature of the money supply. As King (1996, p. 68) correctly asserts in referring to Minsky, “Minsky’s objections to Milton Friedman are, at this early stage, moderate in tone and technical in substance; they do not involve the endogeneity of the

84

Credit, Money and Production

money supply.” It is for this reason that the “critical” analyses offered by

Minsky and Kaldor are claimed to be in fact modifications of Friedman’s work, and not a complete rejection of it. Had Minsky and Kaldor been directly inspired in their attack on Friedman by the works of Wicksell or even Hayek (as were Kahn and Robinson, see the following chapter), postKeynesian monetary theory today would probably look much different. Indeed, while neither Wicksell nor Hayek broke away fi'om the loanable ftmds approach (they still recognized the existence of a natural rate of interest), they were nonetheless staunch opponents of the Quantity Theory of Money. For instance, Hayek (1935, p. I27) referred to the Quantity

Theory of Money as a “too narrow conception” of monetary theory. More importantly, however, both espoused views on money which recognized the endogenous nature of credit and the role of commercial banks as ex nihilo

creators of money. The second reason why Minsky, Kaldor and Tobin were chosen as the

focus of this analysis is that while these “architects” of post-Keynesian theory shared similar views early on, they nonetheless also contributed in establishing three separate bodies of thought regarding monetary theory. Minsky inspired the so~called structuralist view, Kaldor led to the emergence of the horizontalist position as an altemative to orthodoxy, while Tobin inspired the New Keynesian school of credit rationing. The conclusions which will be drawn from this exercise are two-fold: first,

both Minsky and Kaldor indeed held similar views with respect to money, as was recognized by Targetti (1992, p. 268, n. 54). Second, these early views were established within the orthodox framework of the money multiplier model with causality running from an exogenous supply of high-

powered money to the money supply. If money was thought to be endogenous, it was not in the post-Keynesian sense, but rather in a more orthodox fashion, i.e. in the portfolio sense.

In Minsky’s case, his early views on fmancial fragility will be shown to have been developed within the loanable fimds approach - defined as the

determination of the rate of interest by ex ante investment and saving, that is by the forces of productivity and thrifl. This claim was made by Lavoie

(I996), yet stands in stark contrast to the views adopted, in particular, by Wray (l992b). Finally, it will also be argued that both Minsky and Kaldor have more in common with Tobin than with the current proponents of post-

Keynesian endogenous money theory. Given the central role that both Minsky and Kaldor play in the current

literature, the analysis presented below is not done so as to divide postKeynesians, but rather to unite them behind an agreed upon defmition of endogenous credit. If post-Keynesian theory wants to be a credible altemative to the current orthodox thinking - both in theory and policy -

The Early Views ofEndogenous Money

85

then it must be rooted in Keynesian analysis, and abandon any vestigial traces of orthodox thought. If the analysis is correct, who then can claim the mantel of founders of the post-Keynesian revival of credit-money endogeneity? Many postKeynesians would argue that Davidson or Rousseas may be possible contenders. This interpretation is also rejected. It will be argued in the following chapter that Davidson (I965, I972), and Davidson and Weintraub (I973), and Rousseas (I960) are still far away fi'om a theory of endogenous credit-money. Again, this is not to say that Davidson’s or Rousseas’ later contributions in the sphere of monetary theory are in any way minimized. Rather, it merely suggests that their earlier work does not amotmt to a theory of money endogeneity in the post-Keynesian sense. Having therefore rejected Minsky, Kaldor, Tobin, Davidson and Rousseas as potential founders of the post-Keynes revival of money endogeneity, chapter four will argue that post-Keynesians should turn their attention to the writings of two other Cambridge economists: Joan Robinson and Richard Kahn, two “Keynesian” economists whose views on money are seldom recognized if even acknowledged.

MINSKY’S EARLY VIEWS Considered by many as one of the modem founders of post-Keynesian monetary theory, Hyman Minsky has influenced a generation of postKeynesian economists who continue his work on financial fragility, and

incorporate it within a generalized model of endogenous money. These economists, as will be discussed in chapter five, view the money supply schedule as an upward sloping curve in interest rate-money space. This is a result of the lack of accommodation fi'om the central bank (Pollin, I991), but also chiefly due to the increased illiquidity of fmns and commercial

banks as economic activity expands (Wray, I990, 1992b). As a result, while the base rate of interest may be treated as an exogenous variable, other rates rise with economic activity. The implication is clear: increases in investment, through the demand for credit, lead to increases in the rate of interest. This view - labeled the ‘structuralist” view by Pollin (I991) - stands in contradistinction with the horizontalist or acconrrnodationist position held notably by Moore (I988), Lavoie (I992), and Eichner (I987) - as well as by a number of French and Italian circuitists - who view the credit-money supply curve as horizontal, at least in the short run.‘ The rate of interest is detennined exogenously by the central bank within a given range. Although

86

Credit, Money and Production

it may vary through the cycle, it is not correlated with economic activity,

but may vary with other economic, and even political, variables. While these two positions will be firlly discussed in chapter five, it is worth asking here where the post-Keynesian structuralist notion of an upward sloping curve originates? The answer to this question is to be found in Minsky’s early views on money, which can be traced back to two articles published in I957, one in the American Economic Review (l957a) and the other in the Quarterly Joumal ofEconomics (l957b).

The objective of this section is to take a closer look at these earlier papers, and to identify within these two articles the “vestigial traces” of orthodox thought. Discussion and criticism of Minsky’s views on fmancial fragility per se will be left for chapter five. Minsky’s early contributions are of importance given the fact that they contain within them the early seeds of his fmancial fragility hypothesis which he would carefully develop in the following years — and hence the intellectual inspiration of the upward sloping money supply curve. At the same time, however, his American Economic Review article, perhaps more than the other, contains evidence of Minsky’s acceptance of the loanable ftmds theorem. The position developed here is neither new nor

original. It was first made by Lavoie (I985), but then largely ignored. Lavoie (l996a) retumed to the issue, and expounded carefully the links between Minsky, the loanable funds approach, and the upward sloping money supply curve. This criticism has been, however, largely ignored given post-Keynesians’ refusal to engage in a debate over Minsky’s early views, although the issue was discussed by King (1996). While King (I996, p. 61) claims to fmd “merit” in both positions, the author certainly appears to be leaning toward Lavoie’s interpretation.’ While King (1996, p. 68) does claim that “there is no shortage of references to endogenous money in his early writings. [Minsky] can even be interpreted as a horizontalist,” the author (I996, p. 69) later claims that “it would have been very difficult to recognize in Minsky, circa I964, the future Post

Keynesian?” Moreover, Minsky’s AER article is rarely cited by Minskian authors — as Lavoie (l996a) is careful to indicate. For instance, in discussing Minsky’s 1957 views on fmancial fragility and their links to endogenous money, Wray (I992, 1992b) omits any discussion of the AER article, and does not address Lavoie’s specific criticism. This section will argue that Minsky’s early views on money are not consistent with a theory of endogenous money in the revolutionary, or postKeynesian, sense. As a result, his upward sloping money supply curve is best identified with orthodoxy rather than with post-Keynesian theory,

The Early Views ofEndogenous Money

87

which is meant to be an altemative to the loanable frmds approach, in the spirit of Keynes. This position is, however, at odds with most accepted interpretations of

Minsky. For instance, Wray (I990, p. 135; see also Wray, 1992b) argues precisely the contrary, that “Minsky’s view of money is consistent with the endogenous money approach. In I957, Minsky related institutional innovation to profit opporttmities, and showed how innovation allows

business activity to expand even without an expansion of bank reserves.” But a post-Keynesian theory of endogenous money is more than whether businesses can expand output despite central bank restrictions on bank reserves. It implies, as explained above, a number of reversed causalities, and the acceptance of money as a flow rather than a stock. Each of Minsky’s early articles - though related - focuses on difierent arguments. The QJE article questions the ability of the central bank to meet its monetary policy objectives given financial innovations and their impact

on the velocity of circulation of money. The conduct of monetary policy is influenced by the specific and existing structure of financial institutions. Hence, “the efiicacy of any particular technique of monetary policy depends upon the fmancial institutions and usages that exist” ( 1957b, p. I71). The

problem arises when fmancial institutions change, ofien due to “profit possibilities” existing in the money market resulting fi'om high or rising interest rates. This then encourages commercial banks to find more “efficient” ways of meeting the demand for loans. As a result, the velocity of circulation of money changes, thereby increasing the quantity of money

in circulation. Minsky’s QJE article also discusses the notion that increased innovations

lead to increased debt loads, and thereby decreased liquidity as the debt to net worth ratio rises. Here we also find the early thoughts on fmancial

fragility. The velocity of money also plays a crucial role in Minsky’s less celebrated article in the American Economic Review. In this article, Minsky makes use of a multiplier-accelerator model, and shows that in a number of altemative monetary systems, an excess of ex ante investment over ex ante saving

generally leads to increases in the rate of interest. This is accompanied by increased activity and increased debt loads, unless household liquidity

preference decreases, which Minsky claims is equivalent to an increase in velocity. In both articles, however, increases in the rate of interest - tight monetary policy - induce increased velocity. Throughout these two articles, Minsky rejects all five characteristics of money endogeneity. The only possibility of an endogenous money theory

appears in his discussion of an “infmitely elastic money supply curve” which he then appears to rule out as a possibility, as will be argued below.

88

Credit, Money and Production

Saving, Investment, and the Rate of Interest While Minsky (l957a) offers a glimpse of the modem circuit school approach by appearing to capture the essence of the flux—reflux notion,‘ this remains at odds with the rest of his article which is rooted in the Swedish concepts of ex ante saving and ex ante investment. In fact, while one of the tenets of the circuit school - as it is in post-Keynesian theory following Keynes and other Cambridge economists - is to reverse the causality between saving and investment, Minsky argues the contrary. Nowhere is the standard causality between saving and investment more

obvious than in Minsky’s American Economic Review article, where he argues that investment is primarily financed from an existing, exogenouslydetennined, supply of ex ante saving. This is made evident in numerous places. For instance, Minsky (l957a, p. 864) claims that “ex ante saving and decreases in the liquidity of households can be used for either debt or equity financing of investment.” While Minsky does claim that banks can finance investment, their role is limited to any surplus demand, over and above ex ante saving. It is, in fact, in this context that Minsky (l957a, pp. 865-6) introduces his

accelerator model, arguing that “a necessary condition for the frmctioning of an accelerator process during an expansion is that a source of financing of

investment in addition to ex ante saving should exist.” The fact remains, however, that ex ante saving is the main source of financing investment. The causality between saving and investment which post-Keynesians steadfastly condemn is maintained: saving determines investment. In no instance does Minsky argue that investment creates saving. Moreover, as in the loanable funds theorem, the interaction between investment and saving determines the rate of interest - a purely endogenous variable. References to loanable funds are pervasive in both I957 articles, especially in the American Economic Review article, where the expression is, however, nowhere to be found. For instance, Minsky (l957a, p. 864) claims that “the interest rate is determined by the demand curve for investment, ex ante saving, and the

tenns upon which holders of liquidity are willing to substitute eaming assets for money.” Moreover, he (l957a, p. 865) argues that “if ex ante investment is greater than ex ante saving, the ex ante saving has to be rationed among investors [resulting] in a rise in interest rates.” In the QJE article — where the expression appears - the concept of loanable funds also takes center stage. There, Minsky (l957b, pp. l72~82; emphasis added) argues that high interest rates “are evidence of a vigorous demand for financing relative to the available supply. Since at present interest rates the demand for loans is greater than the supply, the central

The Early Views ofEndogenous Money

89

bank constraints result in higher interest rates.” The idea in both articles is the same: at present rates, loanable funds must be rationed among potential borrowers (the notion of scarcity of funds) hence pushing up the rate of interest.

Minsky (l957b, p. I83; emphasis added) refers to the loanable fl1IldS approach once again while describing the “velocity-interest rate step

ftmction” (see below). Here, Minsky argues that during the time it takes fmancial innovations to work themselves out (the horizontal section of the

schedule), there may be “increases in the rate above r,, if the short-run demand for financing increases by more than the increase in financing implicit in the rate at which the institutional framework is changing.” Shortly before, Minsky (l957b, p. 182) also says that if “at present rates the demand for loans is greater than the supply,” then interest rates will rise. The Exogeneity of the Money Supply

While it is clear in Minsky’s analysis that the rate of interest is endogenous, the author also argues that the money supply is exogenous, and under the control of the central bank. Granted, the central bank may not have full control over the quantity ofmoney in existence due to fmancial innovations, but Minsky does not doubt the ability of the central bank to set exogenously the growth rate of the money supply. References to money exogeneity are prevalent in Minsky. In his QJE article, Minsky argues that the central bank has the ability to control the supply of money. The problem only arises when interest rates are high (or rising) as to encourage banks to develop new ways of fmancing investment which may be outside of the control of the central bank. Hence, the “ability

of the central bank to achieve its objectives” depends on existing institutions. If these institutions do not change ‘significantly,” however,

then “once the efficacy of the various central bank operations is established, financial institutions can be ignored in discussions of monetary policy”

(Minsky, 1957b, p. I71). The problems with this statement are two-fold. First, Minsky offers no clarification of what consists of “high” or “rising” interest rates, or whether these changes in interest rates need to be — or need to be perceived as -

temporary or permanent. Second, does Minsky’s position imply that with lower or falling rates of interest, money approaches more closely its exogenous, base-controlled level? What is of a definite nature is that in periods where there are no fmancial innovations, the central bank controls the money supply.

Money is also seen as exogenous in what Minsky labels an “infmitely elastic money supply.” In this scenario, Minsky (l957a, p. 870) argues that

90

Credit, Money and Production

any discrepancy between ex ante investment and ex ante saving can be accommodated by the banking sector. This would be where money is “created” endogenously by commercial banks. ln arguing this point, however, Minsky introduces the following condition, AM=exantel-exanteS=AY While this equation shows causality rutming from money to income, two arguments must also be kept in mind. First, the main source of financing investment is still saving. Second, even if banks finance the difference between ex ante investment and saving, Minsky’s analysis makes clear that this is done through the standard mechanism that banks lend their liabilities. Lavoie (l996a) also argues that this equation can be found “in many accounts of the loanable funds approach or of Wicksell’s economics.”5 Seccareccia (1990, p. l4l) also claims that this equation is in the works of

Hayek. Recall that in Hayek, the source of the disequilibrium is that I > S. This is the same scenario present in Minsky. Here, we can see the roots of the boom that will engineer the fmancial crisis. ln a state of tranquillity, investment is constrained by saving. However, as we leave the state of tranquillity, we have the excess of investment over saving equaling the change in the stock of money - that is, we have the same result as in Hayek: I > S = A M, and the economic crisis begins. Minsky’s views are therefore

far from a theory of endogenous money. While money appears to be “created” when ex ante investment is greater than ex ante saving, if the growth of the money supply if sufficient to cover the difference between ex ante investment and ex ante saving, then the money supply is entirely exogenous. This is Minsky’s (l957a, p. 871) argument: “When the money supply increases at an independently given rate, the autonomous increase in the money supply is not necessarily equal to the difference between ex ante investment and ex ante saving.” Minsky (l957a, p. 875) in fact reduces the concept of an infinitely elastic money supply curve to that of money exogeneity: “A monetary system in which the rate of growth of the money supply is exogenously given, for example a fractional reserve banking system based upon a gold standard, is equivalent to an infinitely elastic money supply if the difference between ex ante investment and ex ante saving does not exceed the per-period growth of the money supply.” As a result, “horizontalism” and money exogeneity are no longer incompatible concepts. lt then becomes possible in Minsky to

have exogenous money, exogenous reserves, and a horizontal money supply curve. Moreover, the possibility that the exogenous rate of growth of the money supply may not be sufficiently high cannot be sustained since eventually this

The Early Views ofEndogenous Money

9|

would lead to a dampening of the accelerator process. It can therefore only be a short-run phenomenon; in the medium and long run, money is perfectly

exogenous. As firms add to their debt loads, “A cmnulative explosive expansion on the basis of the creation of money will (ceteris paribus) result in a fall in the ratio of equity to debt in the balance sheets of firms.” In a footnote (l957a, pp. 872-3, n. 16), Minsky adds, “The ratio of equity financing to total investment decreases as the accelerator process continues.”

And while borrowers’ risk will rise, this will lower “the

accelerator coefficient which in tum lowers the rate of increase of income. This continues until the accelerator coefficient falls sufiiciently to replace the explosive by a cyclical time series, in which there eventually occurs a fall in income.” Minsky (l957a, p. 883) also appears to understand the notion of money endogeneity when claiming that “to keep interest rates at a given level, the central bank must be willing to supply reserves to commercial banks, in response to commercial banks’ demands, without limit at a fixed rediscount rate.” But Minsky’s claim that “commercial banks create money by lending to business finns” does not amount to a theory of endogenous money. Minskian scholars today - in particular Wray (l992b, p. 172) - argue precisely that even with full accommodation, increased debt loads and increased borrowers’ risk increase lending rates in times of cyclical expansion, despite an exogenous base rate of interest. Minsky’s argument in I957 is thus closely related to the later views developed by his followers (Lavoie, 1996a). Finally, Minsky (l957a) argues that to counteract the dampening effect of

a high rate of increase in the money supply (supposedly greater than the difference between ex ante investment and ex ante saving), it is necessary to keep interest rates “high enough” to discourage investment. If this is the case, however, financial innovations will take place (Minsky, 1957b) and

the velocity of money will increase ‘and be greater than unity. If this is the case, then the money supply cm've camiot be horizontal, and must be upward sloping.

Elsewhere, however, Minsky offers us a view of the money supply process which is completely different than the one presented before. While in his I957 articles Minsky could be seen as a proponent of an upward sloping supply schedule of money, in 1960, Minsky defends the position of a verticalist. Take for instance the following passage in Minsky (I964, p. 190): The amount of money in existence is not determined by the public; it is

determined within the monetary system. Given the reserves and the reserve ratio, the amount of money and also the value of the eaming assets owned by the monetary system are in principle detennined. The reserves and reserve

92

Credit, Money and Production ratio are both determined within the monetary system. Hence in a very significant way it is not true that the monetary system adjusts to the behavior of the economy; what happens is that within wide limits the economy adjusts to the behavior of the monetary system.

Later, Minsky (I964, p. 190) adds, “within an economy money is an indestructible asset; there is nothing that can be done by a household or a business firm to change the quantity of money.” These strong and unequivocal statements make clear the fact that Minsky

accepted the exogeneity of money, and adopted the orthodox multiplier model in which a loans market has simply been added, almost as a simple afterthought. The Exogeneity of Reserves Discussion of money exogeneity naturally leads us into discussion over the exogenous nature of reserves, and into the relationship between reserves and the money supply. Post-Keynesians generally agree that reserves are an endogenous variable, that commercial banks first make loans, creating deposits in the process and only afterwards find the necessary reserves to support the creation of deposits. While Minsky appears at time to suggest that reserves are indeed endogenous and that the central bank ought to be prepared to supply them to commercial banks “without limit,” there are two reasons why this statement is perhaps incompatible with the rest of Minsky’s theory. First, as argued above, Minsky rejects the possibility of an infinitelyelastic money supply curve. Second, elsewhere in these articles, Minsky argues precisely the opposite, that reserves are exogenous and under the control of the central bank. For instance, Minsky (l957b, pp. I81-2) clearly identifies the orthodox causality when arguing that with institutional changes, “(I) a given volume of reserves now supports more deposits; (2) a given volume of deposits now supports more bank loans to business.” Hence, Minsky establishes that the causality runs from reserves to deposits to loans, which is the exact opposite of post-Keynesian theory. In fact, Minsky claims financial imiovations are “equivalent to an increase in bank reserves” (l957b, p. I80). The difference here is that the relationship may be somewhat unreliable, but exists nonetheless. Minsky therefore accepts the money multiplier model, where values of the multiplier may vary and become imprecise. Moreover, Minsky (l957b, p. I84) then claims that despite increases in velocity which represent “permanent increases” in lending abilities of banks, the central bank can still restrict bank loans directly provided it is ready to act “strongly.” In this respect, it is worth quoting Minsky extensively:

The Early Views ofEndogenous Money

93

Whenever such an institutional change in the money market is working its way through the economy, restrictive monetary policy, to be effective, must

offset the rise in velocity by decreasing the quantity of reserves. Purely passive constraint which operates by not allowing the quantity of money to increase will not be effective in preventing inflation. Therefore, unless the central bank acts strongly to decrease the money supply, monetary policy has only a very limited domain of effectiveness in controlling inflationary

pressures. Minsky alluded to the money multiplier model elsewhere as well, for instance by quoting approvingly Warren L. Smith, where the latter claims that the “basic frmctioning of financial institutions is the mobilization of the fmancial resources of the economy in support of economic activity" (see Minsky, 1957b, p. I92, n. l). Hence, banks are not seen as “creators of credit” but as fmancial intermediaries channeling ex ante savings of households to bank deposits (the “mobilization” of the financial resources), to potential borrowers. Minsky’s references to the central bank as “lender of last resort” should therefore not be interpreted in terms of a theory of money endogeneity. Rather, it is of the firefighter’s type, which rushes in only when the economy is in need of it — that is to prevent a financial crisis. Endogenous

reserves are an example of what the central bank is capable — or willing of doing. They are not an institutional component of his model. Money and Income Once the exogeneity of the money supply and the causality between reserves and money have been established, the causality between money and income can be looked at more closely. Here we fmd that in many instances, Minsky appears to have no problem with the orthodox moneyincome causality.‘ For instance, Minsky ( 1957a, p. 882) claims that “a too rapidly growing money supply results in rapid price inflation and that a too slowly growing money supply results in a downturn of income.” In both instances, the causality runs from MV to PY. In other instances, Minsky recognizes that while the causality may still run from money to income, it may also be partially frustrated by changes in velocity resulting from changes in the stock of money. For each level of money income Y, there exists a minimum quantity of money M, which is necessary to sustain the volume of payments associated with Y. If

M, is the total quantity of money in existence then there is no money available for portfolio use; we have a maximum income velocity of money V,,, for each

Y, so that M,. V,,, = Y. If M is greater than M, then the actual velocity, V, is

94

Credit, Money and Production less than V,,,. The difference between M and MT is ML, the amount of money which is held as a liquid asset.

This passage is important for several reasons. The causality is clearly from money to income. Moreover, while income is constant (exogenous), any changes or adjustments are done through either the money supply or the

velocity of money. Finally, money used for portfolio decisions appears at the same time as the rest of the money supply. In this respect, we can link Minsky’s discussion of an infinitely elastic money supply curve with the loanable funds theorem. Structuralists argue

that Moore’s horizontalist position necessarily implies that the velocity of money is equal to unity (see chapter four). Keeping this in mind, then we have the following logic. According to Minsky’s analysis argues that if M > MT, then ML > 0. The velocity of money will fall, and could possibly equal unity if ML is sufiiciently high; V = l < V,,,. However, “With ML > 0, the interest rate is detennined by the demand curve for investment, ex ante saving, and terms upon which holders of liquidity are willing to substitute

eaming assets for money” (Minsky, 1957a, p. 864). Hence, even if the velocity of money could be unity, the rate of interest would still be detennined by the interaction of the supply and demand of saving. Elsewhere, Minsky (1963, p. 68) criticizes Friedman and Schwartz’s

money-income link claiming that “the connections they specify

have a

long and rather variable lead over the business cycle peaks and troughs Any process that results in such long and variable leads can also result in considerable slippage between monetary changes and the presumably resultant changes in income.” In other words, Minsky’s criticism is that the link between money and income according to Friedman and Schwartz is not sufficiently direct. Minsky (1963, p. 68; emphasis added) then asks, The relevant question in regard to the significance of monetary changes is whether the lag (and slippage) of income changes following changes in the money stock (or its rate of change) is so large and variable that it is not feasible, under all circumstances, to operate on the money stock in order to achieve desired changes in income. But the feasibility of using changes in the money stock to affect income depends upon the channels through which a monetary change affects income.

Minsky then proceeds to suggest and specify two transmission mechanisms where the link between and income is more direct - what he calls the “open market” and the “commercial loan” models. The former is equivalent to “changes in the money supply due to a substitution of money for incomeeaming assets,” while the latter results in “changes in the money supply due

The Early Views ofEndogenous Money

95

to acquisition by the banks and authorities of newly created fmancial liabilities.” Minsky (1963, p. 69) then states that “the link between changes in the money supply by these teclmiques and changes in demand is direct.”

A final thought regarding Minsky’s loanable ftmds beliefs: Minsky links together the money supply schedule with a loanable frmds curve — which he calls a “velocity-interest rate relation” curve. However, the latter is really a loanable funds curve in interest rate-velocity space. The relationship is depicted in Figure 3. I. The upward sloping curves in the right-hand side of the graph are shortrun curves showing the relationship between higher interest rates and increased velocity (Minsky compares them to short-run cost curves). Within a stable institutional framework, a given upward sloping curve

suggests that as interest rates rise, banks will be able to offset “at least in part” the effects of the central bank’s tight money policy. Increased velocity

represents a “pennanent” increase in loanable fllI1ClS. But, Minsky argues that increased interest rates feed back into the system encouraging fmancial innovations of various types, hence “shitting” the curves. This gives us Minsky’s long run “step function” where the horizontal sections represent the time period during which fmancial irmovations work themselves out. What must be understood is that Minsky’s graph is clearly a loanable funds graph - despite claims by Pollin (I996, p. 499) that it is a creditmoney supply curve. Hewitson (I995, p. 299) also argues that this position is consistent with a theory of endogenous money: “ln the 1950s, Minsky argued that the velocity function is a positively sloped step fimction of the interest rate which shifts to the right because of financial innovations, which are in tum induced by high interest rates. It implies that general liquidity

is endogenous.” The problems regarding the two theories of endogenous money in post-Keynesian theory are evident here. But, as Minsky himself claims ( 1957b, p. 182), this is not so. Minsky is not referring to “endogenous” money, but to loanable fimds: “An increase in velocity increases loanable funds this can be represented by a positively sloped curve between velocity and the interest rate,” where ro is the “liquidity trap” portion of the money supply curve. Minsky then argues that if the central bank wants to counteract the effects

of increased velocity, it needs to decrease the quantity of reserves, thereby once again suggesting that causality runs from reserves to money. Minsky

is not referring to a non accormnodating behavior on the part of the central bank, but rather in exogenous reductions in reserves. The only caveat is that

to be effective, the central bank needs to act “strongly to decrease the money supply.” Minsky’s graph must be understood within the overall framework of his money model which is orthodox.

96

Credit, Money and Production

r/\

M’

I

I

ll/fl. I

II

III

A/fl

+

a

/ 7'»

TW0ney

0

Vaocitj?

Figure 3.1: Minsky ‘s interest rate / velocity theorem

Loanable Funds and Financial Fragility It has now been established that Minsky’s early papers are rooted in the loanable funds approach. It remains now to show whether Minsky’s financial fragility hypothesis was developed within the Austro-Wicksellian tradition. This is Lavoie’s (I996a, p. 2) position: “The origins of many aspects of Minsky’s financial fragility hypothesis can be found in an earlier algebraic model of his based on the loanable funds theory.” References to Minsky’s financial fragility hypothesis are numerous. For instance, Minsky (l957b, p. I85) makes the argument that, associated with the ability of the money market to finance an inflationary expansion is a decline in the liquidity of households and firms. To the extent that either the most liquid assets leave the banking system for the portfolios of other financial institutions or the debts of the newly grown and developed financial institutions enter the portfolios of banks, the liquidity of the banking system declines. Declining liquidity of banks, households, and business firms

The Early Views ofEndogenous Money

97

has two attributes. One is that the debt M“,net worth ratio rises. The other is that the vulnerability of money-market assets to fall in value increases. The two attributes of declining liquidity reinforce each other so that the chances of insolvency and illiquidity increase simultaneously.

On other occasions, he argues that increased debt loads of fn'ms (increases in the debt to equity ratio) accompany increased economic activity. In Minsky (l957a, p. 864), he argues that households, business firms, and banks are sensitive to the composition of the

balance sheets of firms; in particular an increase in the ratio of debt to equity or a decrease in the ratio of cash to other assets in firms‘ balance sheets will make business firms less willing to borrow and households and banks less willing to lend. I-lence if investment is financed in such a way as either to increase the ratio of debt to total liabilities or to decrease the liquidity of business firms, the amount of investment induced by a given change of income will fall. In the AER article, Minsky constructs a table which shows that using money

to fmance debt (savings are used for equity fmance), the debt to equity ratio increases with economic activity. Minsky (l957a, pp. 872-3) argues that a cumulative explosive expansion on the basis of the creation of money will (ceteris paribus) result in a fall in the ratio of equity to debt in the balance sheet of firms. Even if the terms upon which firms can borrow are unchanged by the deterioration of their balance sheets, borrowers’ risk will rise. Minsky ( 1957a, p. 873) then adds that “even with a monetary system that permits all of ex ante investment to be realized, the fmancing of investment

by bank debt can result in lowering the accelerator coefficient which in tum lowers the rate of increase in income.” Moreover, with fmancial fragility, the possibility of a “crisis” increases. As Minsky (l957b, p. I84) argues, The reverse side of the coin to the increase in velocity is that the liquidity of the economy. [This implies] that risks to increase In time, these compounded changes will result in unstable money market so that a slight reversal of prosperity

[it] decreases the economy an inherently can trigger a

financial crisis.

Excess Saving and Debt Finally, Minsky (l957a, p. 873) also claims that with a fall in income, “The excess of ex ante saving over induced investment will be utilized to reduce

98

Credit, Money and Production

bank debt.”

As Lavoie (l996b, p. ll) argues, Minsky appears to be

unaware of Keynes’s banana parable. In the Kaleckian-Keynesian framework, any excess of savings over investment will lead to unsold production - the profit realization problem - and hence to increases in debt for producers. The excess savings will be lent by the banks to businesses to make up their losses. Minsky says exactly the opposite. He claims that excess savings will help to reduce debt!

Robinson (1956, p. 402) argues along the same lines: “An excess of saving over the value of investment is therefore a loss to firms.” The hoarded savings of households will be used by banks — in their role as financial intennediaries — to extend long-tenn loans to firms in order to prevent bankruptcies. In Minsky, any excess savings cannot have any negative consequences on the goods market. These differences can be summarized in the following way: Kaleckian framework S > I —> H < 0 2 Losses 2 Increase in Debt Minsky’s framework S > I —> A Supply of Capital :> Decrease in Debt It was argued above that Minsky’s early views on money show a strong acceptance of orthodox economics. Minsky’s objective — to show that capitalist economies are essentially unstable — is rooted not only in orthodoxy but within the framework of loanable funds. From this, he develops and elaborates his theory of financial fragility, and accepts a “mildly sloping” money supply curve. Hence, Minsky argues that the base rate of interest is not an exogenous

rate of interest — it is endogenous. The problem is that Minsky is far from adopting Keynes’s argument of liquidity preference couched in tenns of

demand and supply of money. Keynes went to great lengths to criticize loanable funds. As Kohn (I986, p. I224) argues, All the neoclassicals, at least since Wicksell, had held a monetary theory of the rate of interest, believing the market rate of interest to be proximately determined by the supply and demand for loans of money (the “loanable funds" theory). What was new about Keynes’s liquidity preference theory was not, therefore, that it was a monetary theory, but rather that it denied the existence of any “nomial” or “natural” rate relative to which the market rate moved.

The Early Views ofEndogenous Money

99

It would appear from the discussion above that Minsky is at odds with Keynes’s theory of money, at least in regards to Keynes’s rejection of the loanable funds theory. In Minsky, the rate of interest is still determined by the forces of productivity and thrift. The above discussion does not imply that Minsky’s later views -

discussed in chapter four - did not succeed in escaping the vices of orthodox thought. The above analysis however did attempt to show that despite numerous claims, Minsky did not develop a theory of endogenous money in his 1957 articles consistent with post-Keynesian analysis. KALDOR’S EARLY VIEWS Having taken a closer look at Minsky’s early views on monetary and

fmancial theory and their links to mainstream or orthodox theory, the task now turns to another “foremost architect” of post-Keynesian thought, Nicholas Kaldor.

While there is little doubt that Kaldor has greatly influenced postKeynesian economists in areas such as monetary theory, and theories of distribution and growth, he remains nonetheless, “in a group by himself’ (Arestis and Skouras, 1985, p. 130), and “defies classification” (Hamouda and Harcourt, I989, p. 3). Kaldor’s writings on monetarism are few and far between, and consist of three self-contained periods, each showing important differences from one another, as well as standing as a testimony to Kaldor’s on-going rethinking of his views on money. They show a progression toward Kaldor’s fmal horizontalist position, but having as a starting point an acceptance of the

basic orthodox model with an exogenous money supply, while criticizing the direct link between money and income. Nonetheless, while spread over more than four decades, Kaldor’s writings

show a consistent attempt to criticize orthodox monetary theory, as embodied in the Quantity Theory of Money, and later personified by Milton Friedman.

Kaldor’s first phase consists in fact of two periods, separated by ahnost twenty years. His first articles on monetary theory are not explicitly an

“attack” on monetarism or the Quantity Theory of Money, but represent a clarification of Keynes’s theory of liquidity preference in the General Theory. It is here that Kaldor introduces his now famous elastic money supply curve. This was followed by a more direct criticism of the Quantity Theory of Money, between I955 andl960 which includes his memorandum

to the Radcliffe Committee.

I00

Credit, Money and Production

Kaldor’s last two phases coincide with his New Monetarism article published in 1970, in the Lloydr Bank Review, and 1980-2, which culminated in his forceful Scourge of Monetarism. Kaldor’s contributions during these last two phases will not be analyzed in this chapter since they are outside its immediate concem, as was also Minsky’s later work.

While Kaldor’s writings in I970 onward are clear as to his position regarding monetarism, the same cannot be said conceming his early writings, including his Radcliffe memorandum. As is the case with Minsky, post-Keynesians are reluctant to discuss, debate and acknowledge any vestigial traces of orthodoxy in Kaldor. As a result, very little has been discussed with respect to his own ties with orthodox theory. For many postKeynesians, linking Kaldor with orthodox monetary thought sounds certainly like heresy, especially given his virulent attacks on monetarism which characterized his later writings, having called it at various times a ‘scourge,” a “terrible curse,” and a “visitation of evil spirits.” Yet, this is precisely what this section will attempt to demonstrate.

While it is clear that Kaldor had no links with the loanable funds approach, he nonetheless shared many elements with orthodox thought. This can be witnessed by the changing argrnnentation Kaldor held against

monetarism. His early views are perhaps the most at odds with the modem post~

Keynesian endogenous money approach of reversed causality, credit and production which he later helped develop.

His early work, as will be

argued, must thus be interpreted as a modification of orthodox thought rather than an altemative to it, or an outright rejection of it. As such, just as Kaldor (1982, p. 2l) had claimed that Keynes’s theory of the rate of interest was “a modification of the quantity theory of money, not its abandomnent,” so is his own early criticism. In retracing the various stages of Kaldor’s views on money, it is useful to keep in mind two specific points: first, Kaldor was not a monetary economist, on the same level as Keynes and Robertson, and money certainly

did not occupy the major part of his work. Second, Kaldor was himself a past advocate of orthodoxy, and as such, had to go through his own “long struggle of escape.” Unlike Pasinetti (I986, p. 30l) who claims that by the mid-1940s Kaldor’s “break with orthodoxy was definite and complete,” it will be shown here that Kaldor only succeeded in ridding himself of the vestiges of orthodox thought later in life.7 This seems to be Kaldor’s own assessment, claiming that his own struggle spanned “several decades,” was a “succession of steps,” and

occurred “sporadically” (Kaldor, 1978a, p. viii). As was the case with Minsky, the five characteristics of money endogeneity defined earlier will be used in order to detennine whether

The Early Views ofEndogenous Money

l0l

Kaldor espoused a theory of endogenous money as early as I939 and I958. It will be argued that he did not. Rather, it will be shown that Kaldor’s break with orthodox monetary thought occurred only in I970, with his “New Monetarism,” and not in I939 nor 1958 where his writings still contained important concessions to orthodox macroeconomics and monetary thought. This argument is consistent with Desai (I989), Dow and Dow (I989, p. I63, n. 2), Lavoie (I991), Moore (I989, I991), and Smithin

(1994). For instance, in his review of Kaldor’s monetary writings, Desai (I989, p. I71) omits discussion of Kaldor’s I939 papers because “nothing in his subsequent writing on monetarism refers to that paper.” This position is nonetheless at odds with that defended by Musella and

Panico (I993) and Targetti (I992) who all argue that Kaldor’s views on money are consistent throughout, and that in particular, he had adopted a theory of money endogeneity in I939. For instance, Targetti (I992, p. 265)

claims that Kaldor “first began to develop [endogenous money] during the 1930s and then expanded in detail fi'om I970 onwards.” On this point alone, Kaldor himself gave several conflicting accounts. First, Kaldor claims that “I had already explained my ideas on the

endogeneity of the money supply in I939 in ‘Speculation and Economic Stability’ and in the article ‘Own Rates on Interest”’ (quoted in Targetti, I992, p.265). Second, Kaldor (I982, p. 22) also claims that when in the early l950’s I first heard of Friedman's empirical findings, I received the news with some incredulity, until it suddenly dawned on me that Friedman’s results must be read in reverse; the causation must run from Y to

M, and not from M to Y. And the longer I thought about it, the more convinced I became that a theory of the value of money based on a

commodity-money economy is not applicable to a credit money economy.

Finally, in his rejoinder to Friedman, Kaldor (l978a, p. 26, n. I) claimed that “all the points I made in my paper in criticism of the ‘Chicago School’ were put forward by me in a succession of meetings at the Merril Center in South Hampton, Long Island, in I958. These views were also implicit in my oral and written evidence to the Radcliffe Committee in l958.” Hence, by Kaldor’s own admissions, money endogeneity was introduced in 1939,

and rediscovered in the early l950s. Yet, elsewhere, as will be argued below, Kaldor suggests that this “sudden dawning” only occurred in I970, where — as an interesting footnote — the expression “money endogeneity” first appears, as does the notion of reverse causation. It is therefore appropriate to question when reverse causation and money endogeneity “suddenly dawned” on Kaldor.

I02

Credit, Money and Production

Before we tum to the discussion of Kaldor’s papers, it is important to address an argument which is at the center of the debate surrounding Kaldor’s endogeneity views. It concems whether Kaldor’s (I982) horizontalist position is an authentic position taken by the author, or

whether it was used simply as a ‘simplifying analytical device” (Musella and Panico, I993, p. 37). This claim is important, for if it can be shown that Kaldor’s 1982 position is a “shorthand explanatory device” (Targetti, I992, p. 276), then that would indicate that his actual position was to propose an upward sloping money

supply curve, making his later position more consistent with that defended in I939. Targetti (I992, p. 275) argues this position exactly, as do Musella and Panico (I993), although it contradicts the position taken by Thirlwall

(I994, p. 77), according to whom, Kaldor’s later position is best described by assuming that at whatever the rate of interest the monetary authorities want to maintain, the

supply of money is elastic with respect to demand, so it’s completely horizontal at whatever the rate of interest that is set. The schedule moves up and down at the price at which the monetary authorities want to maintain the rate of interest. In that sense, he’s a horizontalist.

According to Targetti (I992), the horizontalist position adopted by Kaldor

was “extreme,” and “what Kaldor had in mind was a more general case.” He then proceeds to show why an upward sloping money supply curve is best representative of Kaldor’s later views, hence the consistency between his I939 paper and his I982 book. Indeed, as Musella and Panico (I993, p.

37) claim there are clear “links between his work in the l930s and the 1950s and the subsequent polemic against monetarism.“

But the non-horizontalist proponents adopt a weakened Kaldorian money model, by postulating “three basic propositions” to defme endogeneity (Targetti, I992, p. 273): (I) the interest rate is a “datmn”; (2) there is credit rationing; and (3) there are variations in the velocity of money. Taken either separately or together, these propositions are not inconsistent with a theory of endogenous money. However, they are certainly not sufficient to claim

that they amount, on their own, to a theory of endogenous money. While they even could be made compatible with a theory of portfolio endogeneity (and hence with exogenous money), what is fruther needed to prove revolutionary endogeneity is reverse causality between money and income, and saving and investment, and the need to finance the “needs of trade” through bank credit. In I939 and I958, Kaldor had none of these arguments. It will be argued below that all of these “basic propositions” — while indeed advocated by Kaldor in 1939 or I958 — are nonetheless consistent

The Early Views ofEndogenous Money

I03

with a theory of exogenous money, as adopted by Kaldor, and hence do not lead to a revolutionary theory of money endogeneity. The push to weaken Kaldor’s horizontalist position stems mainly fi'om the

thought that a purely elastic money supply curye implies passivity on behalf of commercial banks and the central bank. Kaldor himself may have been the culprit in this affair having referred to an endogenous supply of reserves as characteristically a “‘passive’ monetary policy” (Kaldor, l978a, p. 14). While this issue will be lefl for fruther discussion in chapter five, it is perhaps appropriate to quote Musella and Panico (1993, p. 39) to this effect. Referring specifically to Kaldor, the authors argue that horizontalism implies that banks cannot modify, on their own initiative, the volume of outstanding loans. The latter is varied unilaterally by the borrower, owing to the existence of overdraft facilities. It does not depend on a discretionary choice of the banking sector, which is seen as “price setter” and “quantity taker.”

Kaldor’s 1939 Views

Kaldor’s writings in I939 consist of two articles (see Kaldor, 1960) of which only one was published at the time. “Keynes’s Theory of the OwnRates of Interest” was originally written as an appendix to “Speculation and Economic Activity” published in The Review of Economic Studies.’ The latter article aimed at generalizing Keynes’s theory of interest found in the General Theory, by taking into consideration fmancial assets and interest rates of different maturity, which Keynes ignored in the General Theory.

This led Kaldor (I960, pp. 3-4) to claim that Keynes had invented “some short-cuts through the maze of complications of a multi-market analysis and

reduced the essential aspects of the problem to manageable dimensions.” The objectives of Kaldor’s paper were manifold. First, Kaldor was chiefly interested in identifying the hypotheses under which speculation would make it necessary for Keynes’s assertion that saving and investment are equilibrated through variations in income (rather than through the rate of interest) to hold (Targetti, I992, p. 25l).'° At first glance, this suggests that Kaldor did not fully rmderstand the point which Keynes was making in the General Theory. According to Keynes, investment and saving were equilibrated ex post through variations in income, irrespective of substitution of assets and speculative behavior. Secondly, Kaldor analyzed the relationship between the short and the long rate of interest, as well as the conditions under which the rate of money

interest dominated other rates. Kaldor’s interest was not the detennination of the money supply, despite chastisirrg Keynes many years later for having

104

Credit, Money and Production

made too many concessions to orthodoxy by making money exogenous in the General Theory (see Kaldor, I983). In a now famous footnote on page 39, Kaldor (I960, p. 39) draws a slightly upward rising money supply curve, and an inelastic money demand curve, both dependent on the short-tenn rate of interest. Kaldor justifies these curves claiming that the demand curve is drawn on the assumption that the volume of money transactions (i.e. the level of income) is given. This demand curve is inelastic, since the marginal yield of money declines fairly rapidly with an increase in the proportion of stock to tumover. The elasticity of the supply of money in a modem system is ensured partly by the open market operations of the central bank, partly by the commercial banks not holding to a strict reserve ratio in the face of fluctuation in the demand for loans, and partly it is a consequence of the fact that under present banking practices a switch-over from current deposits to savings deposits automatically reduces the amount of deposit money in existence, and vice versa.

There are a number of important elements in the above quote. First, Kaldor assumes, as did proponents of the Quantity Theory of Money, that the demand crrrve is fixed. It is in fact in this respect that Kaldor “modified his views” in his memorandum to the Radcliffe Committee."

Second, the demand curve is drawn solely in terms of the transaction demand for money. Hence, there is no discussion of the fmance motive (despite Keynes’s Economic Journal articles), credit or the need to fmance production. Kaldor was foremost interested in speculation, that is “the purchase (or sale) of goods with a view to re-sale (re-purchase) at a later date, where the motive behind such action is the expectation of a change in the relevant prices relatively to the ruling price." Third, Kaldor clearly argues along the lines of a multiplier model, causality nmning from an exogenous monetary base to the quantity of money in circulation, although the components of the multiplier are not fixed (the behavior of households and commercial banks, as well as

exogenous central bank decisions to engage in open-market operations). This is strangely the position of Musella and Panico (I993, p. 42) as well, arguing that “in dealing with the supply of money, Kaldor made use of the

deposit multiplier identity.” In this sense, the money supply curve cannot be strictly vertical. Foruth, Kaldor argues that “a switch-over from cunent deposits to savings deposits automatically reduces the amount of deposit money in existence.” Presumably, however, these portfolio changes are done more so with increasing rates of interest. In other words, as the rate of interest rises, money should shift away from current deposits to saving deposits. As the

The Early Views ofEndogenous Money

I05

rate of interest rises, the quantity of money in existence should fall, thereby leading to a negatively-sloped supply of money. As can be observed, Kaldor’s analysis in I939 is fairly conventional. The supply of money, defined narrowly, is strictly under the control of the central bank given its ability to control the supply of high powered money. But does an upward sloping curve imply money endogeneity? Or can an

exogenous money supply be compatible with an upward sloping money supply curve? Robinson (I970, p. 508) argues that the notion of an upward sloping LM curve is associated, rightly so, with an exogenous money supply. As such, a positive relationship between economic activity and interest rates is rooted in the orthodox position of scarcity: “Thus a higher output, requiring more money in active circulation, leaves less available to satisfy liquidity preference and so is associated with a higher rate of interest.”

Similarly, Lavoie (I991, p. 9) has argued repeatedly that to claim that a growing economy will eventually trigger rising (real) interest rates, presumably as a consequence of excess credit demand or lack of savings, or that the supply of money has a rising schedule in the income — interest rate plane (as in any ordinary demand and supply analysis), is to reintroduce scarcity analysis through the back door.

As will be seen below, Kaldor introduces in I955 the notion of the scarcity of ftmds. In Kaldor, there is little doubt that the base is exogenous and that the central bank has, in that way, direct control over the supply of money. Yet, the value of the money multiplier is not given, such that the power of the central bank may not be absolute; that is some part of the “quantity of

money” in circulation may be over and above that which is created by the central bank.

But this “extra” money is not the result of the needs of

production, it is not credit-money. Indeed, Kaldor is careful in differentiating between an exogenous money supply, and the quantity of money in circulation arising out of household, asset-holders and bank activities, and not from central bank behavior. The difference rests in Kaldor’s use of the multiplier model with a non-constant

multiplier.

In this case, while the money supply can be exogenous,

changing values of key components of the multiplier can lead to increased money in circulation. In this sense, the central bank’s monetary policy

becomes imprecise and thus ineffective. Targetti (I992, p. 276) is thus correct is arguing that Kaldor rejects the notion that “Central Banks control the money supply through the base multiplier model,” but not because the money supply is endogenous, but

because the multiplier is beyond the control of the central bank.

For

I06

Credit, Money and Production

instance, commercial banks may not hold to specific values of reserve

requirements, and a reduction in bank deposits reduces the amount of money a bank can extend as credit. The position developed above is supported by other post-Keynesians, such as Moore (I991, p. 236) who states that in the early Kaldor, “The broad monetary aggregates and the quantity of the base were still viewed as if they

were exogenously under central bank control.” Fruther, what this implies is the notion that causality runs fi'om bank deposits to bank loans.

Proponents of early endogeneity in Kaldor refer to two specific quotes to make their point. The frrst, made by Musella and Panico (I993, p. 42), is that “the high elasticity assumption [of the money supply curve] makes the money supply endogenous.” Accordingly, they suggest that Kaldor’s (I960,

p. 39) claim that “since the elasticity of supply of cash with respect to the short-tenn rate is nonrrally much larger than the elasticity of demand, the current short-tenn rate can be treated simply as a datum, detemrined by the policy of the central bank,” is an important indication that Kaldor introduced endogenous money as early as I939, by advocating that “the level of the short-term interest rate is set by monetary authorities” (Musella and Panico, I993, p. 45). But this argument is not without its weaknesses. The “exogeneity” of the rate of interest can be explained in two different ways, neither related to a theory of money endogeneity. On the one hand, Kaldor argued in favor of a given and stable money demand curve. Once it is fixed, the monetary authorities can easily have the rate of interest of their choice, by initiating exogenous changes in the money supply, through openmarket operations. Hence, the rate of interest can be simultaneously endogenous and a datum, but dependent on the monetary policy of the central bank, that is through open market operations, which are exogenous. Kaldor (I960, p. 56; emphasis added) referred to “the power of the

monetary authorities to determine the rate (by varying the quantity of money).” Once Kaldor abandons the stable money demand argument, in 1958, the rate of interest no longer appears as a “datum.”

On the other hand, and this is a more plausible explanation, the “exogeneity” of the rate of interest may not be related to the shape of the

money supply curve or the power of the central bank. Rather, Kaldor argues that it can be a result of the shape and elasticity of the demand for money schedule. Kaldor (I989, p. III) offered this explanation in a l98l article co-written with Trevithick, A Keynesian Perspective on Money,

reproduced in this long, but pertinent, quote: The extent to which the central bank can control the supply of money through

open market operations (by buying or selling govemment securities and thereby increasing or decreasing the banks’ vault cash) is inversely related to the interest elasticity of the demand for money of individuals and businesses.

The Early Views ofEndogenous Money

I07

The more elastic the demand the easier it is for the central bank to induce wealth holders to substitute bonds for money in their portfolios by offering a higher price for bonds. In the opposite extreme, small changes in the amount of money in circulation can be brought about only at the cost of enonnous changes in interest rates. Hence, as a general rule, the more interest inelastic the demand for money is [as in I939], the easier it is for the central bank to control the rate of interest and the more difficult it is to control the supply of money.

This quote can be supplemented by another one, given in the introduction to Kaldor’s (I960, p. 5) second volume of the Collected Papers: The nature of that curve [money demand], and its elasticity, will be different according as they choose to regulate the quantity of money directly (through varying the note issue, or the clearing banks’ reserve requirements, or by putting a ceiling on bank advances, etc.) or whether they wish to regulate it indirectly through varying the Bank Rate, and the money market rate.

For Moore (I991, p. 237), this was an indication that Kaldor was “already halfway down the road to monetary endogeneity by I960.” But as can be seen by the former quote above, varying the bank rate is not incompatible with an exogenous money supply, provided the money demand ftmction is inelastic. Hence Kaldor (I960, p. 5) writes “in the case of the second type of policy, the monetary authorities may well fmd that they are confronted by an inelastic demand curve (as suggested in note 2, p. 39).” Both situations can be compatible with an exogenous money supply. The reason given by Kaldor is that an inelastic money demand curve encourages the creation of money substinrtes. Hence, with an inelastic money demand curve, the short-term rate of interest can be treated as a “datum.” Given this, the quantity of “money” in circulation — as opposed to the money supply — is seen as out of the reach of central banks. This rests on Kaldor’s emphasis on portfolio adjustments by asset-holders. Given the explanations given above -- the differentiation between the

money supply and the quantity of money in circulation, and the role of the money demand curve — are thus in a position to better understand Kaldor (I960, p. 6; emphasis added) who claims that “indeed, it is only through their power to control the whole range of short-term interest rates that the monetary authorities can be said to ‘control’ the supply of money in its broader sense.”

The second argument pertaining to whether Kaldor introduced endogeneity early in I939 is that of Targetti (I992). According to the author, Kaldor made explicit reference to money endogeneity in his

I08

Credit, Money and Production

appendix to the I939 article, published only in I960, in volume 2 of The Collected Papers. In that paper, Kaldor claims that “in a modem community it is best to regard the short rate of interest q, (rather than the quantity of money) as being fixed by the policy of the monetary authorities (as was implied by the

slope of the supply curve in the diagram on page 39 above) and the quantity of currency in circulation as being determined by the demand for cash balances by the public.” For Targetti (I992, p. 253), “This constitutes a clear statement of what Kaldor subsequently called the endogenous theory

of money supply.” This view is consistent with Musella and Panico (1993, p. 47).

In

referring to Kaldor, they claim that “the quantity of money in circulation is determined, instead, by the demand for it.” There are however a number of problems with this statement. Having dealt with Kaldor’s reference to the rate of interest as a “datum,” let us proceed directly to the second part, that is whether the quantity of money is “demand” determined. First, it is important to recall that Kaldor spoke only in terms of the transaction demand for money (currency), and not in terms of credit or the need to finance circulating capital (needs of trade), as he would in I970. Second, the demand for “cash” in this article is

made in tenns of portfolio decisions, hence only once income has been detemrined, and afier the production process has started. This suggests that Kaldor sees money as a stock, rather than a flow. Indeed, in comparing the own-rate of money-interest of money with other

assets, Kaldor argues that the “equilibrium” condition occurs when the marginal net advantages of holding various assets are equalized. Using Kaldor’s notation, the own-rate of money-interest of a given asset is defined as the following: a,+q,-c,-r,;

i=l,...,n

where a is the expected appreciation of the asset in terms of money, q is the yield of the asset which can be considered a convenience yield or a money return, c represents the carrying cost of the asset, and r is the

negative of the marginal risk premium, that is “instead of regarding liquidity as an addition to the yield [as in Keynes], we shall represent it as a deduction from the yield of those assets which, on account of uncertainty of future value (or retum) in tenns of money, or on account of their imperfect marketability, carry a risk premium for which this yield must compensate” (Kaldor, I960, p. 60). Once households and asset-holders receive their income, they allocate their money holdings accordingly, thereby maximizing their net retums. For

The Early Views ofEndogenous Money

I09

money, a and r are always zero since there is no uncertainty about its future value in terms of money. Furthemrore, c is also considered to be zero. “Hence in the case of money the own-rate of own-interest is necessarily equal to the own-rate of money-interest and consists simply of q, the yield

of money” (Kaldor, I960, p. 62). This is because money is an accepted medium for discharging debt, and its demand is directly related to the

transaction demand, and hence negatively related to a given income. As for other assets, such as short-tenn securities, a and c are also zero

since they are considered close substitutes of money.

Therefore, the

equilibrium for money and short-term assets is reduced to q|=q2-F2; q2 Credit-money C. Figure 5.1: The horizontal credit supply curve By positing an exogenous rate of interest and a horizontal credit supply curve, horizontalists break the relationship between the rate of interest and economic activity, as stipulated by the orthodox approach. Lavoie (1996, p. 276) makes this point precisely, in his criticism of the structuralist approach, arguing that if the link between the interest rate and the level of output is not broken, some of the standard beliefs imbedded in the neoclassical paradigm are recovered: an increase in investment leads to an increase in the rate of interest, as the loanable funds theory would predict. We are back to the neoclassical world of scarcity, with crowding out effects and the like.

The horizontalist argument implies two important arguments, often misunderstood by post-Keynesians. First, at r.,, commercial banks will stand ready to supply credit to all those borrowers who meet their criteria for loans. Clearly, if a bank customer does not meet banks’ requirements, they will not receive a loan (see chapter eight). Horizontalism does not imply that banks are passive (see below). Second, the exogeneity of the rate of interest does not imply that central banks have the ability to set the rate at levels which are deemed “inconsistent” with world markets, or irrespective of current economic conditions, as Wray (1992) suggests. ln fact, Moore (1991, p. 406) has

160

Credit, Money and Production

argued that “‘exogenous’ does not denote that central banks are flee to vary interest rates between minus and plus infmity. But it does imply that central banks always have a substantial range of discretion over which they can vary rates” (see also Moore, I988, p. 266, n. 1 I). Lavoie (1996, p. 278) has also made this claim, arguing that the central bank sets the rate of interest within some “boundaries of existing conventional fmancial wisdom.”7 This argument also does not imply that central banks are in the habit of pegging short-term rates for any lengthy period of time. This is Wray’s (1995) interpretation of the horizontalist position. Yet, according to Lavoie (1996, p. 279), “To describe the money supply curve as a flat curve does not imply that central banks forever peg the base interest rate.” Rather, monetary authorities will change the rate of interest whenever they deem it necessary. Central banks continuously respond to market changes and the general state of the macroeconomic economy, as they are intent on meeting certain macroeconomic targets. In other words, central banks will respond with policies of “leaning against the wind” or with a reaction function. As the economy approaches full employment, central banks may perceive this, rightly or wrongly, as inflationary. They may think that increases in the rate of interest will dampen inflationary pressures. Graphically, as Kaldor (I983, p. 22, n. 36) later argues, changes in the base

rate of interest can be represented by a “set of horizontal lines, representing different stances of monetary policy” (see also Lavoie, 1992, 1996), rather than an upward-sloping money supply curve, as in Figure 5.2. The

difference is clear. While the former does not imply a necessary correlation between the rate of interest and economic activity, the latter does. As the central bank changes its monetary policy, the rate of interest will rise, shifting the credit supply curve upwards, which may then affect the demand for bank credit. What is important is to note that at that new rate, banks will stand ready once again to supply the necessary credit to firms who meet their criteria. This argument therefore implies that the rate of interest is no longer an equilibrium price between the demand for and supply of loanable funds, but neither is it an equilibrium price between the demand and supply of either money or credit. A “money market” does not properly exist where price and quantity are simultaneously detennined, or where the rate of interest is

brought into equilibrium by excesses. Equilibrium exists, but it does not refer to an adjustment mechanism. ln fact, as Kaldor argues, equilibrium always exists since there can never exist an “oversupply” of credit or money. Supply always exactly equals demand. According to Kaldor (1983, p. 21), “Since credit (and hence bank money) varies in response to the demand for bank loans, the ‘money supply’ camiot be assumed to vary relatively to the money demand; the supply of

Horizontalism and Structuralists

161

money can never be in excess of demand for it” (see also Moore, 1988; Lavoie, 1992).“ Elsewhere, Kaldor (1982, p. 46) claims that “excess amounts of money could never come into existence.” This is the argument put forth also by Arestis and Eichner (1988).

"A

r,

I

r.,

C“. C“,

be

C?edit-money

Figure 5.2: Horizontalism and a change in monetary policy Post-Keynesians also reject, as was argued earlier, any notion of a “natural” rate of interest. In fact, Smithin (1994) has convincingly argued that the proper post-Keynesian view should not only reject the exogeneity of the money supply, but must also reject the natural rate of interest hypothesis. As he points out, what differentiates post-Keynesians from Wicksell’s theory of money endogeneity is precisely the fact that post-Keynesians negate the existence of a natural rate.9 What is also rejected is the arbitrage which may exist when the natural and monetary rates are different. The exogeneity of the rate of interest — while rejecting the loanable funds theory — also rejects Keynes?s own theory of liquidity preference as developed in the General Theory.” As Moore (1994, p. 122) argues, “In consequence, Keynes’ ‘liquidity preference’ and the classical ‘loanable fiinds’ theories of interest are both invalidated.” While this is so, it does not invalidate Keynes’s views expressed following the publication of the General Theory, which remain largely ignored. This is precisely the position argued by Pivetti (1996, p. 76) who argues that following the General Theory, Keynes accepted the view that the rate of interest was an exogenously-determined, policy variable.

I62

Credit, Money and Production

Keynes and the exogenous rate of interest Keynes gave several indications that the rate of interest is exogenous, and that it does not rise with economic activity. In fact, Keynes expressively argued in the latter years of his life that the rate of interest was typically an exogenous variable. In this respect, Keynes’s liquidity preference theory of the rate of interest must be seen as a two-stage process, as Keynes (xiv, p. 218) himself recognizes shortly afier the General Theory. First, Keynes argues that an increase in the finance required may impose pressure on the rate of interest. However, the actual movement in the rate depends not on

the demand curve, but on the supply curve. In essence, Keynes argues that the rate of interest will rise only if banks refuse “to relax.” In a May 28, I936 letter to Hubert Henderson, Keynes (xxix, p. 222) makes the argument as follows: “If the supply of money is suitably adjusted, then there is no necessary reason why interest rates need rise during a boom or fall during a depression.” This is the same argument which Keynes

makes several times in his post-General Theory articles in the Economic Journal, where he argues that the “appropriate” monetary policy would keep the rate of interest constant given a rise in the demand for credit.

In a I937 letter to Hicks, Keynes (xiv, p. 80) repeats this argument, adding that “it is important to insist” that “an increase in the inducement to invest need not raise the rate of interest. I should agree that, unless the

monetary policy is appropriate, it is quite likely to.” While this reference is often quoted, the rest of the quote is seldom mentioned. This is unfortunate, since Keynes (xiv, p. 80, emphasis added) specifically argues that the difference between himself and the “cIassicals” precisely lies “in the fact that they regard the rate of interest as a nonmonetary phenomenon, so that an increase in the inducement to invest would raise the rate of interest irrespective of monetary policy.” In other words, Keynes argues that the “classicals” necessarily exclude the possibility of a horizontal supply curve, while he recognizes the existence of such a curve. This is precisely the position defended by Lavoie (I996, p. 276) who argues that there exists very little difference between a vertical supply curve and one which is positively sloped: “If the link between the interest rate and

the level of output is not broken, some of the beliefs embedded in the neoclassical paradigm are recovered: an increase in investment leads to an increase in the rate of interest, as the loanable funds theory would predict.” Keynes made even more precise and conclusive points on the exogeneity of the rate of interest. For instance, Keynes (xxv, p. I49) argues specifically that the rate of interest is exogenous, a “political” variable. In a letter to Harrod (dated April I9, I942), Keynes (xxv, p. I49) writes that “in my view the whole management of the domestic economy depends upon being free to

Horizontalists and Structuralists

I63

have the appropriate rate of interest without reference to the rates prevailing elsewhere in the world.” In a speech in the House of Lords on May 23, I944, Keynes (xxvi, p. l6)

says, “We intend to retain control of our domestic rate of interest, so that we can keep it as low as suits our own purposes, without interference from the ebb and flow of intemational capital movements or flights of hot money.” Finally, in March 1945, Keynes (xxvii, pp. 390-2, emphasis added) makes the following decisive statement on the issue: The monetary authorities can have any rate of interest they like.

They can

make both the short and long-tenn [rate] whatever they like, or rather

whatever they feel to be right.

Historically the authorities have always

determined the rate at their own sweet will and have been influenced almost

entirely by balance of trade reasons.

In this statement, Keynes makes two distinct arguments. First, he clearly suggests that central banks have the ability to determine the rate of interest, that is, they are capable of determining the terms on which they “are ready to become more illiquid” (xiv, p. 2l9). If as a result of an increase in the inducement to invest the rate of interest “need not rise,” then this must imply that the money supply curve is horizontal. This would be —- in Keynes’s own words — the “appropriate” monetary policy. Keynes goes even further in claiming that even the long-term rate of interest is exogenous. Hence, we are far from the arguments of the General Theory where monetary details “had fallen into the background.” The bottom line is that Keynes no longer viewed the rate of interest as being determined by the interplay of the demand and the supply of money. As such, contrary to Dow and Dow (I989, p. I62), horizontalism and an exogenous rate of interest are

not a “complete break” with Keynes. The supply of reserves The second argument related to the behavior of the central bank concems

the supply of reserves. According to horizontalism, money becomes endogenous largely because of the lender-of-last-resort behavior of the central bank.

This represents the emphasis horizontalists attach to the

accommodative nature of the central bank.

Hence, accommodationists

“rank the supportive responsibilities of central banks above their control

duties” (Moore, 1979, p. I26). Indeed, central banks do not have the ability to lower or constrain the money supply. Central banks are “impotent in their allrlility to restrict the rate of growth of the money stock” (Moore, I985, p. I2).

I64

Credit. Money and Production

Post-Keynesians agree that as loans are extended and create deposits in the process, reserves necessarily follow, thereby reversing the causality between liabilities and loans in orthodox theory. The implication of reverse causality is that post—Keynesians generally agree that the supply of reserves is endogenous (Fonnan et al., I987, p. 690), based on the demand for reserves by commercial banks. According to Eichner (I987, p. 850), The change in the Fed’s holdings of govemment securities and thus its open market operations, instead of being strictly a policy variable, is for the most part endogenously detennined by the need to maintain the liquidity of the banking system. Indeed, this is why it is an error to assume, as macroeconomic theory normally does, that the monetary base, or highpowered money is an exogenously determined policy variable.

Eichner (I987, p. 847) has identified two types of reserve endogeneity behavior: the defensive and accommodating behaviors. The “accommodative” behavior of the central bank is the traditional postKeynesian role attributed to central banks. In this context, the central bank agrees to supply additional reserves incurred as a result of increases in commercial bank loans. The “defensive” behavior, however, is defined by Eichner (I987, p. 847) as the “component of the Fed’s open market

operations [consisting] of buying or selling government securities so that, on net balance, it offsets these flows into or out of the monetary-financial system.” This is the result of changes in portfolio decisions, and increases or decreases in bank (demand) deposits. Moore (I989, p. 26, emphasis added) has also made this argument: Once it is recognised that loans are made at borrower initiative, and that loans create deposits, it logically follows that the money supply, bank reserves, and the high-powered base all vary endogenously in response to changes in the demand for money and credit.

Moore therefore differentiates here between commercial banks’ demand for reserves as a result of changes in the demand for credit (i.e. the demand for loans), and that resulting from changes in the demand for money, as analogous to portfolio decisions. The former (credit) is Eichner’s “accommodative” supply of reserves, while the later (money) is his “defensive” behavior. The end result, however, is that irrespective of whether the behavior is accommodative or defensive, the supply of reserves is still seen as endogenous, or as a “passive” behavior (Forman et al., I987, p. 689), with no causal influence on economic activity. This stands in stark contrast, as

Horizontalists and Structuralists

I65

we shall see in the next section, with the position held by some structuralists. Many horizontalists finther argue that the central bank must meet all the

demand for reserves, or risk the destabilizing of the financial market. In other words, any refusal of the central bank to meet the full demand for reserves will trigger a fmancial crisis. Thus, according to Forman et al. (I987, p. 698), “The Fed’s foremost responsibility, at least as viewed from within the monetary-financial system, is to maintain the orderliness of the money and credit markets.” Horizontalists have ofien argued that the supply of reserves by the central

bank is complete, that is the central bank fully accommodates the needsafor reserves.

As a result, commercial banks will obtain, either through

borrowed or non-borrowed resen/es (the discount window), the necessary reserves. For instance, Forman et al., ( I987, pp. 689-90) claim that any expansion of credit will, as a result of the additional deposits created, require that the banking system maintain larger reserves, and the Fed can be said to accommodate this rise in the demand for loans only to the extent that, through its open market operations, it provides the banking system with additional reserves equal to the increase in required reserves.

Similarly, Arestis (1987, p. 5) has argued that Central Banks simply cannot refuse to provide “discount window” facilities for they cannot afford to jeopardize the solvency of the banking system. The Central Bank’s main function of lender of last resort must be preserved religiously and continuously so that the liquidity of financial institutions and markets should be guaranteed utterly. Monetary authorities must, therefore, accommodate for the reserve needs of the financial markets.

The lender-of-last-resort argument is therefore emphasized considerably. Arestis (I987) argues that this role is paramount since it is only through the

lender-of-last-resort function that financial assets can ultimately be exchanged quickly, easily and cheaply into cash thus ensuring the liquidity.

The central bank is therefore the pillar of the liquidity-creating process. Other horizontalists have argued the importance of the central bank to fully accommodate the reserve needs of commercial banks. For instance, Moore (I996, p. 9l) claims similarly that “borrowed reserves may be provided freely in unlimited amounts at the discount window at the central

bank’s official discount rate,” such that “commercial banks can never be constrained in their reserves” (Arestis and Eichner, I988, p. I018).

Kaldor (I982, pp. 25, 47) makes the same argument.

I66

Credit, Money and Production The Central Bank cannot close the “discount window” without endangering the solvency of the banking system; they must maintain their function as lender of last resort.

Precisely because the monetary authorities cannot afford the disastrous consequence of a collapse of the banking system, while the banks in tum cannot allow themselves to get into a position of being “frilly stretched”, the

“money supply” in a credit—money economy is endogenous.

Lavoie (I985, p. 71) has summarized the horizontalist position as follows: In normal times commercial banks are ready to provide all loans, and central banks are ready to provide all reserves or advances which are demanded at the existing rate. Loans make deposits and deposits make reserves. The money supply is endogenous at the rate of interest fixed by the central bank or the Banking System. It can be represented by an horizontal supply-curve at the given rate of interest.

The fully accommodative role of the central bank has not gone unchallenged. In fact, Wray (I992, pp. 300-1) claims precisely that “even if we all accept that the money supply is endogenous in the control, theoretical, and statistical sense, there is still plenty of room for

disagreement.” Rousseas (1992, p. 48) has written that “to argue that the central bank fully accommodates any and all increases in the demand for money not only overstates the case but eliminates banks as a barrier to

increased investment.” Accordingly, Rousseas (I992, p. 47) argues that Keynes’s finance motive becomes a “trivially ephemeral and unimportant novelty,” and is reduced to “Say’s Law in reverse” (Rousseas, I992, p. 77) where demand creates its own supply, and Keynes’s novel discussion of banks as holding the key to economic expansion is “reduced to nil.” In this

way, Rousseas (I992, p. 79) has called the horizontalist approach “political, not economic.” Cottrell (I988, I994) has named it “radical endogeneity,” while Coghlan (I978, p. 75), Musella and Panico (I993, p. 38), Rousseas (I989, p. 474), and Wray (1992, p. 1160) have all referred to it as an

“extreme” post-Keynesian theory of money. The lack of full accommodation forms one of the main criticisms of the structuralist position. According to this position,-the central bank may refuse to fully accommodate the reserve needs of commercial banks. The

following two consequences may arise, to be explored more fully in the following section: a rise in the rate of interest, and a reduction in the quantity of loans extended by banks. In certain ways, horizontalists have already acknowledged that the central bank may refuse to fully accommodate banks’ reserve needs. While appearing to argue differently elsewhere, Eichner (I987, p. 851) has

claimed that the central bank only accommodates eighty percent of the

Horizontalists and Structuralists

I67

accommodative and defensive demands for reserves. In Fonnan et al. (I987, p. 692), the author claims this can be “a matter of policy or inadvertently.” Either way, Forman et al. (I987, pp. 692-3) argue that as a lack of full accommodation, “banks will be forced to curtail their lending activity for lack of adequate reserves,” unless they can enter the Federal funds market in order to make up for the shortfall in reserves leading to an increase in the other short- and long-term interest rates. Indeed, this is the principal way in which the Fed’s non accommodating behavior will translate into higher interest rates.

This is in fact analogous to Moore’s (I996) “flown costs” associated with increased discount window borrowing. As a result, the loan/deposit ratio

will increase - what Eiclmer (I987) has called “liquidity pressure.” According to Arestis and Eichner (I988, p. 1007), the central bank’s failure to accommodate will, in the short run, “Give rise to a rise in the level of interest rates.

A reduction in the banking system’s excess reserves will

lead to a rise in the federal ftmds (if not also the rediscormt) rate.” However, as Lavoie (I996) argues, this rise in the rate of increase is not the result of increased economic activity, but rather the result of a lack of accommodation. This is an important point since it still allows the credit supply curve to be drawn as horizontal. Structuralists disagree. For example, Palley (I994, pp. 73-4) assrunes that the lack of accommodation of the central bank implies that the “loan supply schedule would also be positively sloped.” In this respect, Palley claims that the lack of reserves invalidates the horizontalist position, and that as a

result, horizontalism becomes a special case of the more general structuralist position.

Despite the lack of full accommodation, however, the money supply schedule should still be seen as horizontal in credit/interest rate space. Such a decision by the central bank should be interpreted as a change in central bank monetary policy, or interest rate regime. The “new” horizontal creditmoney supply schedule should now be drawn at the higher exogenous rate of interest. At this new, higher rate of interest, demand for credit and

changes in portfolio decisions of households will vary accordingly as to affect the needs for reserves.

This position was argued specifically by Eiclmer (1987, p. 858). The difference is that this change in interest rate regime should not be seen as

depicting an upward-sloping supply schedule. Thus the perfectly elastic supply curve for additional credit

may shifl up or

down, depending on whether the Fed is pursuing a less or a more than fully

I63

Credit, Money and Production accommodating policy. However, this shifl of the supply curve for additional credit should not be confiised with a movement along the curve, with the quantity supplied assumed to vary positively with the interest rate. That would involve the fallacy of posing a supply curve for additional credit

that was separate from, and therefore independent of, the demand for additional credit. This is also the exact position taken by Lavoie (I996, p. 280): “While at every moment of time the supply of credit-money is fully horizontal, shifts in the demand curve for loans induce the creation of a set of temporallyordered horizontal supply curves of credit-money, because of the non accommodating behaviour of the central bank.” Structuralist Criticism and the Horizontalist Response: Setting the Record Straight Despite many attempts by horizontalists to clarify their position regarding a number of issues which have arisen in recent years, some post-Keynesians still see horizontalism in tenns of arguments which have never been made by horizontalists. The above section tried to clarify the horizontalist

position with respect to the slope of the credit-supply schedule and central bank accommodation of reserves. The following section addresses specific

arguments which have been made against the horizontalist view. As will be shown, in many respects, horizontalists never took the position which is claimed by structuralists raising the important question of where and how the argument first came to light. The structuralist criticism of horizontalism has generally centered on five specific points, which will_be discussed in length below. First, the general post-Keynesian interpretation of Moore’s horizontalist position is that banks simply accommodate all demand for loans, and as such are passive players in the production process (Goodhart, I989, p. 30). Second, it has been suggested that while the base rate of interest is exogenous, horizontalists have failed to explain or justify its level (see Heise, I992, p. 29]). Third, many structuralists have argued that a variable mark-up somehow invalidates the horizontalist approach. As is argued below, however, horizontalism is perfectly compatible with a mark-up which varies through the cycle.

Fourth, some have criticized Moore’s use of liability management as a way for banks to meet reserve requirements. Reliance on liability management would somehow go against the notion of a horizontalist position.

Horizontalists and Structuralists

I69

Finally, some have argued that horizontalism necessarily implies a constant velocity of money. Overall, many of the structuralist arguments

against horizontalism stem both flom a misunderstanding of the theory and over a confusion regarding the difference between credit and money, and their respective role in the monetary circuit. Passivity of commercial banks Many post-Keynesians have interpreted the horizontalist position as implying that cormnercial banks are passive agents accommodating all demand for credit. In this sense, Mongiovi and Rllhl (I993, p. 89) argue that money carmot play a causal role. Similarly, Cottrell (I994, p. 599) argues that “on the Kaldor-Moore view

the banks are mere ciphers in this process, passively accommodating whatever demands they happen to experience.” Minsky (I996, p. 77) refers to them as “simpletons.” Others who have criticized Moore’s “non-discretionary” views are Messori (l99l), Musella and Panico (I993), and Gedeon (I985, p. 837), the latter having argued that “the money supply curve cannot be infmitely

elastic if commercial banks - the suppliers of money credit — are accorded their own independent “animal spirits” and are alleged to be a precipitating source of instability in their financial markets.” Dow and Dow (I989, p. 164, n. I2) claim that “it surely cannot be suggested that credit is fully accommodating to the extent that no potential borrower is ever refused credit.”

Interestingly enough, however, Dow

(I996, p. 498) has accepted the position that this is not an entirely correct interpretation of the horizontalist position. In fact, Dow claims that “neither Kaldor nor Moore would seem to argue that literally all demand for credit is accommodated.” Nonetheless, this has not prevented the author flom

building her subsequent critique of horizontalism around the fact that banks ofien do restrict credit. Similarly, Jarsulic (I989, p. 37) has claimed that “the money supply [is] a passive, demand-driven magnitude,” thereby referring to Moore’s defmition of endogeneity as “passive endogeneity” (I989, p. 38). Wray (I992, p. I72) has also made this claim. According to his interpretation of horizontalism, private banks passively supply credit money at any short-term interest rate

established by the central bank’s discount and open market policies. They are able to meet any level of credit demand since they can always obtain reserves from the central bank at the discount rate. There is no room in [Moore’s] model for liquidity preference, for entrepreneurial financial institutions, for market power, or for credit rationing and quantity restraints.

I70

Credit, Money and Production

As a result of all this, Wray (I992, p. I72) concludes that in Moore’s

model, commercial banks “do not care about their balance sheet position [and] are indifferent to liquidity ratios.” As will be shown below, this is not the case, and in fact has never been the case. How structuralists, like Wray for instance, have arrived at this conclusion is puzzling. Horizontalists must share some of the blame for this criticism having at times perhaps exaggerated this position. For instance, Eiclmer has himself

claimed that banks meet all loan demand. In Forman et al. (I985), the authors claim that banks will have all the reserves they need to meet

whatever may be the demand for credit, and they will have “no reason to stint on loans.” But horizontalists have readily acknowledged this problem. For instance, Lavoie (I996, p. 283) claims that “the reader is sometimes given the impression that those who support horizontalism are unaware of those quantitative constraints.” Horizontalists acknowledge and accept the notion that banks often do restrict credit. Banks do not meet all the demand for loans. Despite this, the slope of the money supply curve is unaffected. Lavoie is perhaps the better-known advocate of this position, although Moore has repeatedly

made the claim. Lavoie (I996, p. 284) explains that “the claim, quite legitimate, that banks have some restrictions on their lending does not call

into question the validity of horizontalism. Banks oflen choose not to lend.” Keynes’s (v, pp. 2l2—3; vi, pp. 364-7) statement to the effect that there will always be a “fringe of unsatisfied customers,” becomes perfectly compatible with horizontalism. Earlier, Lavoie (I993, p. I0) also claimed that “to argue that the money

supply is horizontal is not to argue that there are no constraints on credit.” Horizontalism does not negate the fact that banks are price - and quantity -

setters. Similarly, as early as I986, Parker Foster (I986, p. 954) argued that banks simply “respond to this demand in greater or less degree.” Even Moore has made this point, on several occasions. For instance, Moore (I996, p. 90) argues that “provided borrowers have sufiicient asset

and/or income collateral, they will be granted formal lines of credit up to some predetermined amount.” Earlier, Moore (I995, p. 264) claims that “in an overdrafi system, bank accommodation of increased demand for funds by credit-worthy borrowers is in no sense extreme. It is instead completely normal, so long as borrowers remain within their allocated

credit limits.” Moore (1994, p. I23) made the argument elsewhere. This is not to deny that many small borrowers are effectively credit-

constrained.

New businesses and poor households in particular do not

possess the income, assets, and credit records criteria that banks require in

order to make profitable and financially sound loans (the banks’ three C’s: credit, collateral and character.

Horizontalists and Structuralists

l7l

If these references to Moore were the only ones, then perhaps the structuralist criticism could be well taken. However, this is not the case, since Moore even took this position clearly in Horizontalists and Verticalists. There, Moore (1988, p. 62) argues, although apparently ignored, that commercial bank loan officers must assure that loan requests meet the bank’s income and asset collateral requirements. They must in general satisfy themselves as to the credit-worthiness of the project and the character of the borrower. It is precisely for these reasons that banks develop client relationships with their borrowers. Arestis and Eichner (1988, p. I006) have also made this claim, where the firm’s “collateral” becomes the contract it has signed with a customer: With a contract in hand for the delivery of goods sometime in the future, a business firm will, under ordinary circumstances, have no trouble obtaining a loan from any one of the banks with which it maintains an account. Gedeon (1985-86, p. 209) while claiming that “the supply of credit in the

economy can be characterized with reasonable accuracy by an infinitely elastic supply schedule of money," (although recall her previous quote) has

also claimed (1985-86, pp. 210-211) that even though the money supply schedule may be infinitely elastic in the short rim, the money supply will not continue to increase without limit even though enterprises may be willing to take on additional investment projects. Eventually bankers, who are also profit maximizers, will refuse to lend for the more risky ventures it is the responsibility of both the monetary authorities to judge the quality of the credits being granted and the commercial bankers to judge the ability of firms to repay loans.

And despite Musella and Panico’s (1993, p. 30) claim to the contrary, Kaldor (I981) also emphasized this point. According to Kaldor (1981, p. 15), “At any one time the volume of bank lending or its rate of expansion is limited only by the availability of credit-worthy borrowers“ (see also Arestis, 1987). Moreover, in a recent article, Wolfson (1996) has made this point precisely arguing that banks will meet all requests of the “effective” demand for credit, as opposed to the “notional" demand: “Given the effective demand curve, then, we assume that the bank does its best to accormnodate all demands for loans.”'2 As can be seen from the above numerous quotes, horizontalists have repeatedly made the argument that banks constrain credit, and that indeed they have large discretionary powers at their disposal to either accept or

I72

Credit, Money and Production

reject individual credit demand. The emphasis is placed primarily on the creditworthiness of the borrower. It is therefore still surprising to see the claim that horizontalists ignore the restrictions on credit imposed by banks." The argument therefore rests on the creditworthiness of the borrowers. Ironically, as will be seen in chapter eight, this argument is now even being made by a number of structuralists. It was even put forth by Minsky (1991, p. 208) who claims that “those proposals that meet the banker's standard of the time will be financed.” Horizontalists have been making this same argument for over a decade, although largely ignored by structuralists. While horizontalists accept the notion of credit constraints,“ the emphasis is placed on non-price bank-imposed constraints, as done by Wolfson (1996). The author (1996, pp. 455, 459) asks at first glance, the two concepts of endogenous money and credit rationing seem incompatible. If bankers create money by accommodating the credit demands by bank borrowers, how can bankers also refuse to accommodate

these demands by rationing? The answer is simply that bankers accommodate all creditworthy demands for credit, and ration all those demands not deemed credit worthy. Banks have standards in the form of minimum income and cash flow requirements, maximum loan-to-value ratios, and so on. Those borrowers below the minimum will not obtain the desired loan.

These are all non-price mechanisms. Hence there is no need to increase the rate of interest as the demand for loans increases. The horizontalist argument is meant to emphasize two facts. First, banks cannot lend without a prior demand for loans. Robinson (I952, p. 29) makes this claim precisely: “The amount of advances the banks can make is limited by the demand from good borrowers." Second, horizontalists, as argued by Deriet and Seccareccia (I996, p. 140), choose to emphasize the fact that “banks acquire the strategic role of ex nihilo purveyors of endogenous credit money. The emphasis is therefore

on the causal role of bank lending which goes primarily towards satisfying the financing requirements of firms." Hence, banks “hold the key" to expansion. Keynes’s finance motive can never be seen as “nil.” The horizontalist argument therefore does not negate the possibility of credit constraints. What is rejected, however, is the notion that constraints occur because of some exogenous quantity of “loanable funds" which must be allocated efficiently, that is they reject the notion of the scarcity of funds. Hence, horizontalists argue that there can never be an excess demand for loans, relative to supply.“ As was seen in chapter four, this is precisely the argument of Robinson and Kahn.“

Horizontalists and Structuralists

I73

If horizontalists do not ignore the credit constraints imposed by banks, they have yet to develop a consistent theory of bank lending. There are certain passages in the horizontalist literature which allude to this idea, but it remains largely under-developed. For instance, Lavoie (I984, p. 791) has argued that when bankers begin losing some of their high “animal spirits” though they are aware of the fact that their new behavior will harm the economy, they prefer to

restrain the creation of credit-money. They know that those banks that are the least affected by the recession are those banks that show the most moderation.

For this reason, it is quite possible for the banking system to start reducing its credit lines just when firms needs extended loans.

Moore (I986) discusses - without developing the issue - the fact that banks face three types of risk: the credit risk (risk of default by borrowers), the interest rate risk (due to the fact that banks borrow short and lend long), and the liquidity risk (the risk of not meeting demand for withdrawals). These references suggest two elements. First, even with horizontalism, banks are the key element to economic expansion. Portfolio decisions are secondary. What is required, as stated above, is a clear theory of bank behavior, one which combines together microeconomic elements without forgetting the fact that there are also macroeconomic elements which dictate

bank behavior: macroeconomic laws exist independently of microeconomic behavior. Such an approach must rest on some of Keynes’s most important insights. In particular, unlike the position taken by some structuralists, banks must be seen as profit-seeking firms facing an uncertain future. Their decision to lend must not be dictated by automatic increases or decreases in liquidity, but rather by betting on the future in the same manner as firms do. Second, the notion of uncertainty should be introduced within the theory of endogenous money, but not exclusively with respect to firms and their decisions to produce and invest, as is usually the case in post-Keynesian theory. Instead, uncertainty should be introduced within the context of dynamic banks, as they decide to validate or not the requests for loans. If this is the case, the emphasis is somewhat shifted away from effective demand to banks‘ expectations of the future and their expectations of meeting a target rate of retum. This is not to say that uncertainty does not affect firms’ decisions to produce and invest, as it clearly does. But once these “bets” have been placed, firms face another obstacle in requiring extemal financing. The analysis requires the introduction of another level of uncertainty; that of banks. '7 The implications of this shift of emphasis regarding uncertainty away flom firms toward banks minimize the role played by portfolio decisions, and

I74

Credit, Money and Production

hence the stock component of the money supply. Rather, credit is seen as a circular flow, responding to the needs of trade.

The base rate of interest The second criticism made against the horizontalist argument is that it fails to explain the level of the base rate of interest as determined by the central bank (Heise, I992), or that the rate of interest is pegged (Palley, I987-8, p. 283). The argument has been made by Wray (I992) who says that interest rate exogeneity fails to take into consideration current economic conditions. Wray (I990, p. I85) has claimed that an exogenous rate of interest goes

essentially against the spirit of Keynes’s writings: If the interest rate is actually set by the central bank, then there is nothing to prevent the central bank flom establishing any interest rate at which firll

employment will be achieved. Apart flom the fact that, as argued above, Keynes later accepted the exogeneity of the rate of interest, horizontalists have also explained the level at which central banks set the rate of interest, although such an explanation

does not rely on any endogenous reasons. ls For instance, Eichner (I987, p. 860) has argued that the rate of interest is a “politically determined distributional variable,” and any decision to vary

the rate is because “central banks have consciously taken the political decision to increase them” (Lavoie, I993, p. 9). The rate is determined by the central bank based on its own “reaction function.” It can be based on market and non-market phenomena, such as the growth of the economy, inflation, the level of unemployment, or even purely political reasons. Moore has repeatedly made this argument. For instance, in Moore (I988, p. 266), the nominal rate of interest is made dependent on the techniques of monetary policy, the sensitivity of economic behavior to interest rate changes, the size of openness of the economy, the degree of capital mobility, the extent to which the central bank is willing to allow foreign exchange reserves and exchange rates to fluctuate, the expected

domestic and foreign inflation rate, the willingness of the govemment to regulate and impose controls on the economy, and the extent to which policy

is coordinated across countries. Moore (I989, p. 487) repeats the argument, emphasizing clearly that the central bank reaction function with short-term interest rates as the dependent variable, includes the authorities’ estimates of

Horizontalists and Structuralists

I75

(I) the future state of the domestic economy (demand factors), (2) the responsiveness of system behavior to interest rate changes, (3) their ultimate goals (full employment, price stability, growth, balance of

payments, terms of trade, exchange rates, the distribution of income), (4) the effects of interest rate changes on the viability, prosperity, and liquidity of the financial system and, (5) in democracies at least the implication of interest rate change for the goveming party in the next election.

As such, “In ptusuit of their macroeconomic stabilization goals central banks ordinarily vary interest rates procyclically, in response to the perceived state of the economy” (Moore, I994, p. I23).

Similarly, Lavoie (1992, p. 163) has argued that the base rate “is fixed by the central bank, in accordance with its political or economic objectives, for instance the rate of unemployment, the distributive issues, or the extemal constraints on the balance of payments.” Horizontalists have therefore explained the level of the nominal base rate of interest. Arguments to the contrary are misplaced. A variable mark-up

Palley (I991, I994) has argued that given the variability of the interest rate mark-up (over the base rate) through the cycle, the horizontalist position is necessarily invalidated. In his interpretation, and in that of several other structuralists, horizontalism implies a constant mark-up. In the Kaldor-Moore-Lavoie horizontalist version, however, it is the base rate of interest which is exogenous not the loan rate of interest. The latter is set by a mark-up, but it does not follow that it must remain constant through

the cycle. What the horizontalist position does suggest, however, is that changes in the mark-up, and presumably variations in the loan rate of interest, are not the result of either increases in the demand for credit, or of economic activity.

The loan rate of interest can be illustrated using the following equation (Deriet and Seccareccia, I996): ii =[l+mlc

where ii is the loan rate of interest, ic is the cost of obtaining additional funds, and mk is the markup reflecting the Kaleckian degree of monopoly

(the bank’s expected rate of retum). The crucial element is e, which represents a risk premium - i.e. a measure of banks’ liquidity preference. For horizontalists, e varies independently of economic activity.”

For

instance, Eiclmer (I987, p. 858) accepts this approach, claiming that the mark-up depends on the type of borrower, the nature of the assets serving as

I76

Credit, Money and Production

collateral, and the duration of the loan. As such, “lt is the combination of all three factors [which] will determine the risk premimn.” As Lavoie (I996, p. 280, n. I9) suggests, “Empirical data showing the variability or the flexibility of the mark-up over base rates only disproves Rousseas’s (I986) version of interest determination. lt has no bearing on Moore’s theory or the validity of horizontalism.” Furthemrore, horizontalists accept Kalecki’s principle of increasing risk, that is that the mark-up can increase if banks believe fim1s are relying too heavily on extemal finance, thereby becoming a riskier factor. The risk component may reflect other factors as well, although they can loosely be grouped under the expression “creditworthiness.” As the creditworthiness of a particular bank borrower decreases, the risk premium, and hence the

loan rate of interest, increases. As Lavoie (I996, p. 276) claims, “Horizontalism is fully compatible with the Kaleckian principle of increasing risk.” Seccareccia (I996, pp. l44—5) argues that such a component does not

“pose problems for the advocates of the horizontalist perspective since [the risk component] could simply be considered an additional element that would be covered by the markup as long as [the risk component] does not

fluctuate with demand, and would depend on factors extraneous to business cycle activity.”

The acceptance of Kalecki’s principle of increasing risk does not imply an upward-sloping credit supply curve. The rate of interest on a loan is judged on an individual basis, and is not related to the overall, or macro, activity of

the economy. Liability management Another area of disagreement between the horizontalists and the structuralists is the role of liability management within the overall structure of the economic banking system. It has been suggested by horizontalists that if commercial banks cannot meet their reserve requirements, liability management can be seen as a way for them to meet their goal. Putting aside the fact that reserves are becoming less and less an important feature of our systems, Moore has suggested that banks can always find needed reserves through the practice of liability management. This argument has, however, been criticized by structuralists. For instance, Pollin (l99l, p. 37l) has wondered why banks would engage persistently in liability management if the central bank is ready to supply all required reserves. At best, liability management should be viewed as a short-term, or even transitory, phenomenon. Pollin (I996, p. 502) asks, “One must ask why private intermediaries would systematically engage in liability management if they were flee to draw at the discount

Horizontalists and Structuralists

window.

l77

In the aggregate, an institutional framework where no quantity

constraints exist would not encourage the systematic practice of liability management?” Goodhart (1989, pp. 30—l), sympathetic to the horizontalist position, has similarly pointed out that “Moore puts just a bit too much emphasis on the

importance of liability management.

Afier all, as Moore noted

central

banks have always been prepared to provide commercial banks with the

reserves that they need to maintain convertibility.” But while some post-Keynesians appear to claim that horizontalism and liability management are incompatible, a proper understanding of both concepts reveals the opposite conclusion. Liability management has traditionally been discussed in terms of nonaccommodation from the central bank. After all, when reserves fall short of their needs, commercial banks develop imiovative ways to either fmd additional reserves or economize on reserves. It is perhaps true that in a

non-accormnodating environment commercial banks may engage in liability management in order to meet their reserve needs. This would be the position of Arestis (1987, p. 7) who claims that “it follows, therefore, that

fmancial irmovations have enabled banks to become increasingly immune fi'om Central Bank control and consequently dependent upon it as lender of last resort.” But liability management must also be seen as a permanent practice of banks to reduce their current reserves in the face of a given amount of aggregate deposits — not necessarily because the central bank does not fully accommodate. Banks may also practice liability management because reserves represent an implicit tax to banks and a loss of potential earnings. Even in a fully-accommodating environment, banks would still practice liability management. In banking systems where the practice of reserve requirements has been abandoned — as is the case in Canada — banks still actively engage in asset and liability management.

Variable velocity of money Many post-Keynesians have interpreted the horizontalist position as implying that the velocity of money must necessarily be constant. For instance, Hewitson (l995, p. 293) argues precisely that “a perfectly elastic,

horizontal supply of money curve will generate a constant income velocity of money.” Similarly, Rousseas (l986, p. 85) argues that “in a very imKeynesian way [horizontalism] divorces the income velocity of money from changes in the rate of interest.” Cottrell (1994, pp. l 12-3) also argues that horizontalism implies a constant velocity of money, thereby rendering it “in line with Milton Friedman’s monetarism.”

178

Credit. Money and Production

Horizontalists have never claimed that the velocity of money needs to be fixed. ln fact, horizontalists are perfectly comfortable with a variable velocity of money and in no way does it affect Moore’s horizontalist position. In his reply to Cottrell, Moore (l994, p. l24) argues precisely that the “velocity [of money], like many other macro- and microeconomic time series, closely approximates a random walk.” Velocity is an ex post variable, the value of which is “impossible” to predict. As Seccareccia (I988, p. 56) has argued, in a non-accommodating setting,as banks extend loans, they may also reshuffle their liabilities in order to attract money into less liquid deposits, with lower reserve requirements. However, in their attempt to economize on reserves, banks may have to offer a higher rate of interest as an incentive for money holders to switch their money fi'om the liquid deposits to term deposits. Changing values of velocity are a result of liability management, but in an ex post situation, that is only when loans have been extended, incomes and money created. As money is shifled into and out of liquid deposits, the velocity of money will necessarily vary. As Lavoie (I996, p. 282) claims, “It was never denied by horizontalists that higher interest rates would generate attempts to economize on cash or demand deposits.” Reminding readers of a passage he wrote earlier in 1984, Lavoie (I996, p. 282, n. 27) then adds, “Banks would encourage the transfomiation of demand deposits to term deposits;

large finns would start acting as banking institutions and credits between companies would be extended. As a consequence the velocity of money as defined by the authorities would be on the rise.” lf horizontalists see the velocity of money as anything but constant, they do argue though that the notion of velocity of money is not causal, but is an endogenous, ex post, variable. As such, it becomes a “passive factor,” as claimed by Kahn (l972).2° What must be clearly acknowledged is that changes in velocity occur afier production and income have been detennined. In other words, as Lavoie (I985, p. 845) claims, as long as income still determines money, “Post-Keynesians can feel comfortable with either stable or unstable velocity.”2' As Lavoie (1996, p. 282) further argues, “These portfolio adjustments, however important they may be, rank second to the expansion of balance sheets required by economic expansion — the income effects highlighted by horizontalists.” These views were fully endorsed by Robinson (1956, pp. 403—4). As she claims, the concept of velocity is of “limited significance. It merely shows that some M fonnerly lying in an inactive account (where it represented the title to someone’s quasi-permanent wealth) has moved into an active account (where it represents a temporary balance lodged for a short period between the receipt of income and its disbursement).”22

Horizontalists and Structuralists

179

Similarly, Kahn (1972) adopts this same view in his memorandum to the Radclifie Committee: Bank deposits are held not only as a means of exchange but also as a way of holding wealth which for the moment the owners do not desire to put into

other forms. The quantity of money held in the latter “inactive” form is closely related to the level and structure of rates of interest, but the relationship is not at all a stable one, so much depending on expectations

about the future. The velocity of circulation, as normally conceived, is an entirely bogus concept. It is a weighted average of the velocity of the “active” circulation and of the velocity of the “inactive” circulation — the latter happens to be zero. The weights are the quantities of money held “actively” and “inactively” respectively.

The exact same view was adopted by Lavoie (1996, p. 282), “The liabilities of the banks become more heavily weighted towards temi deposits which

carry higher rates of interest than demand deposits.” Moreover, Moore (1989, p. 486) argues along the same lines: “A horizontal money supply function does not imply that V is stable. But changes in the supply and demand for transactions balances will cause V to vary even when interest

rates remain unchanged. Financial innovation, disintermediation, and the development of near-money substitutes are simply forces which affect money demand and so cause V to vary.” The above discussion has shown that horizontalism is compatible with a variable mark-up, a variable velocity of money, non-accommodating behavior of the central bank, credit constraints, and liability management. What must be kept in mind is that many of the activities described above are seen by horizontalists as ax post, that is occurring once credit has allowed production to begin. Liability management occurs with or without

accommodation, but occurs nonetheless after the initial extension of credit. The velocity of money varies accordingly. This says nothing about the economic system in which we operate except that after production, households move money constantly between active (liquid) and inactive

(illiquid) bank accounts. THE STRUCTURALISTS The structuralist approach to post-Keynesian monetary theory is rooted in the analysis of a capitalist economy with a complex, sophisticated fmancial

system, where the focus is placed on capital accumulation, and the possibility of a financial crisis. The primary intellectual influence of structuralism originates in the work of Minsky ( 1957a, 1957b, 1975, 1982).

I80

Credit, Money and Production

Since Pollin (l99l, p. 368) asserted that structural endogeneity is “less well developed than that of accommodative endogeneity,” its ranks have increased in recent years, attracting the most converts. Its proponents do not portray it as an “extreme” position, but rather as a theory “rich in

institutional detail” (witty, 1995, p. 273).”

The emphasis is placed on the firm’s desire to accumulate capital, and its decision to fmance this accumulation through extemal debt, where banks are the primary way of fmancing economic expansion. Hence, Minsky’s financial instability hypothesis is a theory of the impact of debt on system behavior and the way debt is validated. But the “way debt is validated” will have important consequences on the economy as a whole. The primary result will be on the liquidity of both borrowers and lenders, both becoming less liquid. Hence, as Wray (I995, p. 279) claims, the Minskian-structuralist position is primarily interested in “balance sheet positions and changes in balance sheet positions of lenders and borrowers.” The argument is that in a boom, the cash and liquid-asset holdings diminish and portfolios are burdened with debts. The positions become more and more risky, both for households, firms and banks. The increasingly illiquid position of banks and finns will have repercussions. For instance, while credit-money is still seen as an endogenous component of the system, the increased riskiness of extending loans will lead to increases in the loan rate of interest. Hence, the loan rate

of interest no longer becomes an exogenous variable with respect to the growth of the economy, but rather an endogenous variable which varies positively with a rise in the demand for loans. As Hewitson (I995, p. 294) states, “The partial endogenisation of the interest rate through liquidity preference does not invalidate the claim that the money supply is endogenous.” The endogeneity of both credit-money and interest rates in the structuralist tradition has led Moore (I995, p. 261) to claim that this “can hardly be viewed as Keynesian. If both money and interest rates were in fact really endogenous, monetary policy could have no effect on the economy, since the monetary authority could directly affect neither money nor interest rates.” Structuralism is associated with four general arguments. These are: (l) the lack of central bank accommodation; (2) the relationship between firms’ leverage ratio and the loan rate of interest (liquidity of firms); (3) the relationship between bank lending and the loan rate of interest (the liquidity

of banks); and (4) the degree of liquidity pressure. The first argument is concemed with central bank behavior, while the others deal with economic activity (Musella and Panico, I993). For structuralists, these arguments justify an upward sloping credit supply curve, irrespective of whether the central bank is accommodative.

Horizontalists and Structuralists

181

The structuralist, upward sloping curve has not gone unchallenged by horizontalists. For instance, Lavoie (1996) claims it is an attempt to reintroduce scarcity “through the back door” (see also Deriet and

Seccareccia, I996). Moore (I989, p. 486) argues that it suggests a “longrun secular increase in interest rates over time” - as also suggested by an upward sloping LM curve ofien adopted by structuralists.

Moreover, the upward sloping demand curve for credit also reestablishes the notion that the rate of interest is determined by supply and demand. In fact, as argued by Pollin (1995, p. 10), “The variability of the loan demand curve [establishes] an equilibrium quantity and price of lending?“ The analysis below will take a closer look at the first three propositions of

structuralism listed above. In each case, it will be shown either that some flaws or contradictions exist within the structuralist argumentation, or that they are perfectly compatible with the horizontalist position. The conclusion which will be drawn is that horizontalism is the more general

case while structuralism can be considered a special case of horizontalism. The Lack of Accommodation As was argued in chapter two, structuralists agree with the general horizontalist position that loans create deposits, which in tum creates reserves. This position can be observed in Pollin (I996, p. 495) where the author claims that the two [post-Keynesian] approaches share the basic Tookean insight that the rate of money supply growth and, more importantly, broad credit availability are fundamentally determined by demand-side pressures within the financial markets. More specifically, both approaches accept the assessment by the

former New York Federal Reserve Bank senior vice president Alan Holmes (I969, p. 79) that “in the real world banks extend credit, creating deposits in the process, and look for the reserves later”.

However, when the central bank is non accommodating, structural endogenists argue that the necessary reserves may not always be forthcoming due to the central bank’s decision to restrict either their expansionary open market operations, or banks’ access to the discount window. As a result, as pointed out by Pollin (l99l), deposits will not create the necessary reserves in the process.

According to Pollin (1991, p. 367), this becomes the crucial point separating horizontalists from structuralists: “Where the two approaches diverge is in moving beyond Holmes’ point to explaining the process of ‘looking for reserves later’.”

I82

Credit, Money and Production

There are two structuralist arguments regarding the lack of reserves. The first is the argument put forth by Pollin (I991) pertaining to the role of the central bank, as outlined above.

Pollin (I991, p. 373) offers three reasons why the central bank may refuse banks access to the discount window. First, the central bank may fear the “inflationary dangers associated with monetary ease.” Whether excess money is inflationary is irrelevant if the central bank believes it is. Second, excess reserves may exert downward pressures on the exchange rate value of the national currency. Third, the central bank may simply be unable to supply the necessary reserves due to “imperfect information.” In this case, as was pointed out by Forman et al. (I985, I987), the central bank would “miscalculate” the reserve needs of banks. The second argtunent for “aggregate reserve deficiencies” is advanced by Palley (I987-8, p. 287) who claims that while “individually each bank believes it has access to an unlirriited supply of funds at the going price therefore willing to lend as much as possible, subject to standard credit worthiness considerations,” at the aggregate level, however, banks face important reserve deficiencies. In this respect, Palley emphasizes the variations of deposits due to households’ portfolio decisions. For structrualists, a fall in deposits

necessarily implies a fall in reserves. Also, Palley claims that the existence of lines of credit (which is Moore’s principal argument for the horizontalist position) implies that banks over-lend, and therefore have little control over reserves: “The now widespread practice of granting lines of credit makes banks especially vulnerable to excessive lending in the event of unexpected positive shocks to aggregate loan demand. This second source of reserve deficiency has been largely overlooked in the determination of the money supply” — although as argued above, this argument corresponds to Eichner’s (1987) “defensive” central bank behavior. Irrespective of the reason, the structuralist argument suggests that the central bank, unlike in horizontalism as it is claimed, possesses real and efiective quantity controls. As such, the lack of accommodation from the central bank threatens the full endogeneity of the ‘money supply. This position can be traced back to Chick ( I977) — who argued that central banks pursue dynamic operations. As Pollin (I991, p. 373) argues, “Structural endogeneity proponents believe that central bank efforts to restrict the growth of nonborrowed reserves will impose significant quantity constraints on the total reserves available to financial markets.” Screpanti (I995, p. I) also claims that “it remains true that sometimes, when they want to restrain the growth of base money, central banks do succeed in doing so.” Similarly, the central bank can increase the money supply exogenously. As Pollin (I995, p. ll) claims, the central bank can also substantially

Horizontalists and Structuralists

I83

encourage credit growth by increasing the supply of reserves to the private system, thereby raising the banks’ liquidity, an argument also advanced by Palley (I994).

Hence, the creation of excess reserves allows banks to

increase their lending.” The lack-of-accommodation argument put forth by structuralists therefore

implies that discount window borrowing is not a close substitute for nonborrowed reserves since the central bank will impose “real and effective” restrictions on discount window borrowing (Pollin, I991). Banks must look elsewhere for reserves, which are generated within the financial structure itself. If the central bank is neither willing nor able to supply the needed

reserves, structm'alists argue that banks will turn to the practice of liability management to find the needed reserves. As Pollin (1991, p. 374) makes the point, banks will attempt in “attracting funds out of demand deposit accounts, which have relatively high reserve requirements, into CDs, the federal funds market, Eurodollars, and similar instruments within the shortterm money market.” This implies therefore that reserves are generated flom within the system. Liability management thus becomes a way for banks of avoiding the power of the central bank, and a way of reducing the needed reserves in the face of the central bank’s refusal to meet the demand for bank reserves. The lack of accommodation leads, according to structuralists, to two important consequences. First, banks may not succeed in obtaining all needed reserves such that a reserve shortage arises. As Pollin (l99l, p. 376) argues, “Liability management will not necessarily create an adequate supply of reserves to meet demand” (see also Pollin, I996, p. 495).

Rousseas (1989, p. 478) agrees, claiming that “the money supply is endogenous to a marked degree without being perfectly so. Demand does create, to a greater or lesser extend, but never entirely so its own supply. The degree to which the supply of money is positively sloped depends on

the discretionary policy of the Federal Reserve.” Palley (1996, p. 520) similarly argues that there is an “incipient aggregate scarcity of reserves. This view can be traced directly back to Minsky (I982, p. I7l): “These institutional changes may not lead to a suflicient increase in fmancing ability to effect the same increase in financing as would have occurred if

there had been no central bank constraint.” In that respect, banks are faced with the following three restrictions: they will have to “call in loans,” sell

their assets, and extend fewer new loans.“ Second, the growth of liability management will lead to inevitable rises in the rate of interest within a given institutional structtue. This can be attributed, as Pollin (I996, p. 498) assumes, to the notion that money is considered a financial asset: “The reward of higher interest yield will be

184

Credit, Money and Production

necessary to induce asset-holders to shift their holdings into less liquid forms.” Palley (I996, p. 520) shares similar views: “Unless the monetary authority is being fully accommodative, banks will engage in asset and liability management to obtain liquidity, and this then affects interest rates.” Structuralists admit, however, that the rate of interest need not rise if accompanied with financial imovations, presumably, as argued by Wray (1992, p. 305) due to exogenous shifts of the money supply curve to the right. This notion is attributed to Minsky’s early work, as was analyzed in

the second chapter.

According to Pollin (I991, p. 365), “The upward

interest rate pressure can lead to fmancial irmovations, producing a new institutional environment whereby rates on managed liabilities need not continue to rise, even if the central bank continues to constrain reserve growth. Managed liabilities can be marketed at lower costs. The upward pressure on intennediaries’ lending rates will also diminish.” The difference with horizontalism is that interest rate determination becomes a two-way process. The horizontalist position regarding the lack of accommodation was presented in the previous section, there is thus no need to repeat it here. However, there are nonetheless certain criticisms which can be made here

regarding the structuralist position. First and foremost is the different role attributed to liability management in the context of horizontalism. For horizontalists, liability management is part and parcel of banking activity, irrespective of whether banks need to fmd reserves, and irrespective of whether the central bank is fully acconunodating, despite Palley’s (I996) contention. This answers Pollin’s query of why Moore has emphasized liability management even in a perfectly accommodative setting. For structuralists such as Pollin (I991 , p. 376), “Liability management [is] seen as an altemative means of generating needed reserves.” Palley (I996, p. 523) calls this the “critical difference” between accommodationists and

structuralists, that is the fact that banks actively engage in liability management following a tightening of the federal funds market. But this is not so. Liability management exists even in a perfectly accommodating environment. In this respect, post-Keynesians can draw on the literature regarding reserves as an implicit tax to cormnercial banks (Johnson, 1968). Reserve requirements impose a cost on depository institutions and their customers. Because reserves typically do not earn interest, reserve requirements force depository institutions to forgo potential interest income. The impact of reserve requirements on bank profits in Canada has been examined by Johnson (I968, pp. 977-8) who writes that the convention of non-payment of interest by the central bank on commercial bank reserves constitutes in effect a tax on the creation and use of deposit

Horizontalists and Structuralists

I85

money to the detriment of efiiciency in the long-run allocation of the economy’s resources between the provision of the payment mechanism and other uses. The burden of this tax is increased by the stipulation of minimum or average cash reserve ratios, to the extent that such stipulation

obliges the banks to hold a larger volume of non-interest-earning reserves than they would voluntarily choose to hold for the efficient conduct of their business. Legal minimum requirements impose an additional burden, since the banks must guard against violation of the requirement by holding excess cash reserves or by keeping their non-cash assets sufficiently liquid to be able to meet unexpected reserve drains.

Similarly, Fama (I980, p. 47) writes that “such a reserve requirement is a direct tax on the deposit retums since it lowers the retum on deposits by the fraction of deposits that must be held as reserves. Deposits now involve opportunity costs, that is, lower returns than non-deposit assets with the

same risk.” Hence, precisely because reserve requirements are considered an implicit tax, fmancial institutions have a strong incentive to avoid them, either by taking advantage of changes in rules or by irmovating arormd existing rules (Palley (I991) has even acknowledged this point).

As such, liability

management has grown substantially in recent years.” Hence, horizontalism and liability management become perfectly compatible since the latter is not made dependent on the lack of reserves argument of structuralists, but rather on profit seeking bank behavior. Commercial banks thus attempt at all times to maximize their holdings of

“revenue-generating assets.” Moore has made this point precisely, although no one has picked it up. According to Moore (l99la, p. 407), banks “continuously seek to reduce their effective reserve requirements, since reserves are nonearning assets.” Hence, Moore emphasizes the importance of liability management, but not necessarily in the way structuralists have done so.” A second criticism which can be brought forth concems the role assigned to reserves. At a more general level, horizontalism can be interpreted in terms of the passive role played by reserves — passive in the sense that horizontalists break with orthodox thought in severing the causal link between reserves and bank loans. Horizontalists do not recognize the causality nmning fi‘om the liability side to the asset side of the bank’s balance sheet. As Deriet and Seccareccia (I996, p. I40) claim, “The

liability side of banks’ balance sheets is presumed to play a largely passive role in accommodating firms’ demand for credit. In this fiamework, moreover, the portfolio decisions of the public in holding deposits has no causal consequence on the decision of the banks in both satisfying the

186

Credit, Money and Production

demand for credit money and providing the desired fimds at a given rate of interest.” In that respect, excess reserves have no role to play either. Even if banks have excess reserves, credit is granted only when there is a demand for

loans, and subsequent to the creditworthiness test banks impose on potential borrowers. In this sense, what is also questioned is the central bank’s ability to increase the money supply. Hence, it is one thing to argue that the central bank may not fully acconuriodate, it is quite another to claim that reserves have a causal role in determining the quantity of loans a bank can make. Screpanti (I995, p. I) recognizes this point claiming that structuralists believe that the “base money is not fully endogenous,” thereby not fully breaking with the orthodox tradition. Moreover, Screpanti (I995, p. I) — while a proponent of an upward sloping credit supply curve — believes that for this reason, structuralists “in order to explain why the supply curve is rising resort, more or less knowingly, to the neoclassical argument of maximizing behaviour in portfolio management.”2° Rather, he argues, the structuralist argument should rely on the liquidity preference argument, to which we now tum.

Liquidity Preference of Banks

The “lack of accommodation” argmnent has become somewhat of a moot argument, and this for two reasons. First, as argued in the previous section, horizontalists have admitted that the central bank may refuse to fully acconunodate the demand for reserves, although they still do not claim this makes the supply of credit schedule positively sloped. The other reason why central bank behavior may not be reflective of a deep division between the two camps is due to the fact that many structuralists are now claiming that even with a fully accommodative central bank, the supply curve of credit is still upward sloping. For instance, Dymski and Pollin (I992, p. 42) argue that “the increased loan riskiness associated with rising cash commitments relative to cash flows may cause interest rates to rise even in an accommodating environment.” Palley ( I 996, p. 530, n. 6) similarly argues that “even if the monetary authority is being fully acconunodative, there are still differences between accommodationists and structuralists.” Wray (I990, pp. 166-7) also argues this point: “The money supply is upward sloping relative to interest rates even though there is no strict quantity constraint on bank lending.” Wray then adds, it is “the ‘liquidity preference’ of the banks which causes the short term interest rate to rise.” This has become Wray’s (1992, p. I72; see also l992c) principal argument

Horizontalists and Structuralists

187

against horizontalists, that “if banks are concemed with the ratio of loans to equity, then discount window borrowing will not eliminate pressure on interest rates as banks expand loans.” Structuralists claim that economic expansion itself makes the rate of interest rise given the fact that as firms increasingly rely on extemal borrowing — and as banks meet the credit demand of firms — both firms and banks will become more illiquid. As a result, banks will have to charge higher rates to compensate for the risk of increased illiquidity. Wray (1995, p. 278) has sunuriarized the argument as follows: Liquidity preference of both banks and borrowers plays a fundamental role in determining the price and the quantity of credit. Given a state of liquidity preference. expansion of balance sheets, leveraging capital and reserves, and exceeding prudent margins of safety can occur only at rising interest rates.

These arguments are attributed to Minsky’s hypothesis of “financial fragility,” according to which expansions are in effect destabilizing. Sustained economic growth, business cycle booms, and the accompanying financial developments generate conditions conducive to crises for the entire economic system.

At the start of an expansion, fim1s are in a position of “hedge” finance, but as expectations rise through the early stages of the cycle, firms are encouraged to seek greater extemal debt by borrowing credit from conunercial banks. As this occurs. firms will generally move from hedge to “speculative” finance, and finally to “ponzi” finance. This suggests that at the start, the expected cash flows are greater than the cash payments in all periods. Increasingly, however, expected cash flows fall short of the firm’s cash needs. First, they are sufficient to cover interest payments in future periods, but not the initial principal. Finally, fim1s are not even capable of covering the interest payments themselves. Hewitson (1995, p. 302) thus argues “financial units are characterized by the relationship between cash payment commitments on debt and the expected cash flows due to the quasi-rents on capital assets or the sale." As Dow and Dow (1989, p. 158) have observed, the economic cycle is “characterized by falling liquidity preference in upswings and rising liquidity preference in downswings.” As stated by Isenberg (1988, p. 1048), “Financial fragility is defined by a movement into a riskier financial position." For Minsky and structuralists alike, increased indebtness — high leverage ratios — itself leads to increased fragility and the possibility of crisis, as economic growth increases firms’ confidence and willingness to borrow. This is made clear in the following quote by Carter (1989, p. 281), “An increasing proportion of financing will be short term as firms grow more

I88

Credil, Money and Production

confident of their ability to roll over debt as necessary for refinancing and seek to take advantage of the typically lower yields on short-term debt instruments." Minsky’s views are represented in the following well-known graph: A

prices v

Demand price for capital lenders ‘s risk

borrowers’ risk p

Supply price of capital

_

ii) Retained earnings

V

IF

>

Figure 5.3: Minsky 's two-price diagram

Where v is the present value of a set of capital goods (the price of capital assets reflecting the expected gross profits after taxes), and p is the supply price of investment goods, where the difference between these two prices determines investment decisions. The IF curve — the “internal finance” schedule — acts as a constraint on firms, and is last period’s ex post profits. If firms want to invest in capital whose value is greater than IF, then they need to go through banks.3° The role of external — or bank — finance, as argued by Delli Gatti and Gallegati (1990, p. 368) is to speed up growth, income and accumulation, while also increasing debt (leverage ratios). As this occurs, firms become over-leveraged and the possibility of a crisis arises: “Increasing financial fragility can be approximately captured by an increasing leverage ratio.” As a firm’s leverage ratio increases, banks will compensate the riskiness of the loan by charging a higher rate of interest. As Wray ( 1992a, p. 305, n. 24) writes: “As a bank increases such leverage ratios [as loans increase], the

Horizontalists and Structuralists

I89

probability that income flows will fall short of payment commitments increases, as does the probability that losses on assets will exceed bank equity.” The Minskian-structuralist position is shared by a great many postKeynesians. Dow and Dow (1989, p. 154) for instance argue that liquidity preference affects the rate of interest in two ways: in an “objective” way (firm liquidity) and in a “subjective” way (bank liquidity). They further argue that “in particular. changing macroeconomic conditions will alter bankers’ perceptions of risk, and their confidence in these perceptions. To the extent, then, that risk assessment is subject to systematic variations, there is likely to be systematic adjustment of risk premia (regardless of monetary policy), and thus to the mark-up on the rate set by the monetary authorities.” Similarly, Wray (1995, p. 279) asserts that bank mark-ups will vary “with balance sheet positions and changes in balance sheet positions of lenders and borrowers.” As such, liquidity preference, Keynes’s views on uncertainty and Kalecki’s “principle of increasing risk” become related. Wray (I990, I992) is perhaps best associated with this view. Wray (I992, p. 305) in fact argues that “given a degree of liquidity preference, as banks‘

balance sheets expand, perceived lender's and borrower's risk rises so that the banks raise retail loan rates of interest.

In consequence the money

supply curve is upward sloping with respect to the interest rate." This is because banks “require higher interest rates to compensate for greater perceived risk as balance sheets expand” (Wray, 1992, p. I160). Dymski and Pollin (I992, p. 38) make a similar statement: “Bankers (or other lenders) extract compensation for their lenders‘ risk by imposing harsher terms on borrowers: i.e., higher loan rates, shorter terms of maturity, collateral, and restrictions on dividend payouts." However, structuralists make an even stronger argument, in claiming, as does Wray (I992, p. 1160), that “even where additional loans can be made which are no more risky than loans which have been made previously, banks may require higher interest rates to compensate for riskier leverage ratios". It is not only the risk component of lending to high leveraged firms that requires a higher interest rate, but it is also the mere extension of (safe) loans which requires banks to seek higher rates. As soon as banks become less liquid, interest rates must increase, irrespective of whether individual loans are risky or not. This is because lenders’ risk increases. According to Wray (1992, p. 305, n. 24), Lender's risk is a function of the leverage of equity, reserve, safe assets, and prospective income flows. As a bank increases such leverage ratios, the possibility that income flows will fall short of payment commitments increases, as does the probability that losses on assets will exceed bank equity.

I90

Credit. Money and Production

According to structuralists, the liquidity of banks is measured by the following two ratios: the capital—asset and the reserve—asset ratios. Structuralists argue that during an expansion financed with bank credit, banks will become increasingly less liquid as the capital—asset ratio of banks falls, and, more generally, moves countercyclically (Minsky, I986). Wray (1995) has interpreted the horizontalist approach as implying that the reserve—asset ratio must remain fixed through the cycle. Minsky’s views on- financial fragility have not gone unchallenged. The criticism comes from various venues. First, it is not explained why the intersection of the p and the IF schedules determines the firms’ financial constraint. It is evident that the diagram is seen as applying to the micro level, not the macro level. It is also worth mentioning that a structuralist has criticized the Minskian view as “inadequate.” According to Pollin (I987, p. 148), “We would expect increases in debt financing to be accompanied by increases in their fixed investment growth rate. But in fact what has accompanied the rise in corporate debt financing since the mid-1960s has been a decline in investment growth.” Moreover, Mongiovi and Riihl (1993, p. 92) ask why firms would finance their investment externally to the point of becoming over-leveraged: Why, first of all. do firms allow themselves to become over-leveraged? Surely from past experience management knows the maximum gearing ratio

consistent with sound business practice. Textbooks on finance management discuss at length the appropriate levels — generally rules of thumb — of various financial ratios. Any systematic tendency of firms to accumulate levels of debt which threaten their own survival would be a phenomenon which requires explanation.

They then wonder (I993, p. 92) why increasing loan rates through the cycle should not discourage further increases in extemal financing. The horizontalist criticism of this argument has been that in an accommodative environment, there is no reason to believe that in the aggregate, reserves and assets will not move pari passu. According to Moore (I995, p. 265), Providing bank reserves and bank capital are replenished pari passu as their balance sheets expand, as ordinarily occurs in the real world due to in part to central bank accommodation, banks experience no increase in lender's risk. They therefore do not require greater compensation to induce them to expand their loans and assets.

But Moore does not clearly differentiate between what applies to individual banks and what applies to the banking industry. Granted, there may be

Horizontalists and Structuralists

I91

some banks who experience a shortage of reserves and thus find themselves in a deficit situation. Simultaneously, however, there must be other banks who will be in a surplus position. Overall, reserves should move in tandem with loans. What may apply to individual banks would not apply to the

aggregate economy. Macro dynamics exist irrespective of the conditions which govem the micro level. Deriet and Seccareccia (1996, p. 144) have made this point precisely (see also Lavoie, I996), Undoubtedly, on an individual basis, not all banks would be seeing their liabilities rise proportionally with a growth in their loans. As the overall volume of loans expands, there would emerge surplus banks and deficit banks, and the latter would rely either on central bank accommodation or the interbank funds market to meet their short-term liquidity needs. In the

aggregate, however, the net position of banks would not much change with the expansion of their balance sheets.

To this specific charge, Wray (I995) has responded in three ways. First, he claims that even if after some time the capital—asset ratio returns to its original level, “Expansionary activity will initially lower capital—asset

ratios.” Second, Wray also claims that a fixed capital—asset ratio can occur only under specific — or special - circumstances, that is when the rate of interest on assets is equal to that on liabilities, whereby the mark-up would zero. According to Wray (1995), it is more usual to claim that the growth rate of assets is smaller than the growth rate on liabilities. During an expansion therefore, the capital—asset ratio must necessarily fall. Third, Wray (1990) argues that banks must nonetheless be sensitive to the composition of their assets and liabilities as loans expand. a question raised as well by Deriet and Seccareccia (1996). But as argued above, horizontalism is not weakened by the recognition that the reserve—asset ratio falls necessarily during an expansion. This is

perfectly compatible with the accomrnodationist interpretation of liability management. As loans create deposits, reserves become a function of the deposits, not the loans, as Moore (199la) reminds us in his response to Pollin (1991). As such, as banks attempt to economize on reserves — which

is not the same as attempting to find reserves in the face of a lack of accommodation by the central bank — liability management will free banks from the “reserve-tax.” Hence, there is no reason to believe that horizontalism must imply a constant reserve—asset ratio. This point cannot be emphasized enough. Moreover, according to Wray (I995. p. 277),

I92

Credit, Money and Production The bank experiencing rapid growth of its balance sheet will be forced to tum to expensive “wholesale” sources of reserves by issuing large denomination certificates of deposits and repurchase agreements. It is certainly not true that banks face the same wholesale rates — those thought to be engaging in risky behaviour must pay a premium on wholesale deposits.

On this issue, Wray is somewhat confused. On the one hand, the author claims that faster growing banks will be “punished” by facing rising costs. Hence, Wray (I995, p. 277) writes, “No bank can long afford to grow at a rate greatly in excess of the growth rates of other banks because it will be punished by higher average costs of issuing liabilities due to reliance on wholesale funds to higher interest rates required on liabilities it is thought to be risky due to rapid growth.“ Wray therefore admits that banks cannot grow faster than other banks, and

banks must grow at the same pace. If this is the case, the reserve—asset ratio may be constant — a fact which Wray readily concedes, claiming that “none lose reserves.” Wray’s defense is that banks growing at a same rate is a “special case" (Wray, I995, p. 277). Yet, Wray just claimed that banks cannot grow at a different rate for very long. The key is for banks “to keep in pace,” as Keynes reminds us — as do horizontalists. Similar rates of growth cannot therefore be a special case. For instance, Lavoie (I996, pp. 288-9) argues this point precisely: If a single bank is pursuing aggressive lending policies. as it could in the present environment of generalized liability management, its credit-worthiness may become questionable and other financial institutions may refuse to lend to it. The main objective limit to the expansion of a bank is the rate of expansion of the other banks and the norms of prudent and acceptable behaviour self imposed by the banking industry. If banks have a higher rate of retum on their loans than on their safe assets, the volume of their equity could grow at a fast pace, so that the loans to equity ratio need not rise during the expansion.

The first part of this quote is reminiscent of Hicks (I974, p. 54), the liquidity of an individual bank is based on “the extent of the funds which it thinks itself to be able to borrow.” The point is the same as in Keynes (v, p. 23), so long as banks “move forward in step." Moore (l989b, p. I5) claims also that “consequently, providing a bank just keeps pace with the rate of loan extension of its competitors, no net outflow of funds at clearing need result.” On a final note, apart from the illiquid position of banks, structuralists have not explained why the extension of more loans in itself represents a riskier position for banks. Surely, if structuralists emphasize the

Horizontalists and Structuralists

I93

creditworthiness of bank borrowers, then the extension of more loans would imply an extension of additional credit to a larger amount of creditworthy borrowers. The extension of more credit does not necessarily imply a riskier position. Moreover, whether banks extend more or less credit, they can never isolate themselves from the risk of default which is due to factors outside the control of banks. As long as banks protect themselves by establishing criteria for evaluating the creditworthiness of their customers, then it is difficult to see how the extension of additional credit necessarily represents a riskier position for banks. Liquidity Preference of Bank Borrowers If the increased lending risk is not a sufficient argument in explaining the upward sloping credit supply curve, structuralists further argue that the increased illiquidity of bank customers also contribute to increasing the rate of interest during an economic expansion. Wray (1995, p. 278) claims that, Similarly, borrower’s risk increases as equity and prospective income flows are leveraged. Since most loans are made primarily on the basis of expected

income flows of the borrower, rather than on the basis of collateral, bank risk increases as borrower’s risk rises. Thus, loan rates of interest should rise as either lender’s risk or borrower's risk rises.

As firms increase their reliance on external debt, they necessarily represent a greater risk to the bank, irrespective of its creditworthiness or the realization of profits. This reasoning is based on Kalecki’s principle of increasing risk. Kalecki (1937, p. 442) argues that there are two reasons why borrower’s risk increases with increased investment: The first is the fact that the greater is the investment of an entrepreneur the more is his wealth position endangered in the event of unsuccessful business. The second reason making the marginal risk rise with the size of investment is the danger of “illiquidity.” The amount invested must be considered as a fully illiquid asset in the case of sudden need for “capital.” If, however, the entrepreneur is not cautious in his investment activity it is the creditor who imposes on his calculation the burden of increasing risk charging the successive portions of credit above a certain amount with rising rate of interest.

Kalecki’s principle of increasing risk can be represented in the following (Figure 5.4). As the graph depicts, if firm B has a higher debt—equity ratio, then it would be reasonable to impose a higher rate of interest. This is due

I94

Credit, Money and Production

to the fact that it is a “riskier” — i.e. less creditworthy — borrower. Dymski and Pollin (I992, p. 30) make the same argument, “The more a firm is leveraged — that is, the more debt obligation it carries relative to its equity base — the more likely that default on any project will require liquidation.” A r B

A >

debt/equity Figure 5.4: Kalecki 's static principle of increasing risk

Kalecki’s principle of increasing risk is a notion accepted by horizontalists — and even by circulationists — as a reasonable argument to make in a static context. Indeed, if a firm is leveraging itself more at any point in time, then it is reasonable for a bank to impose a higher interest rate on that firm. As Lavoie (1996, p. 285) says, “What we have here is a cross-section of firms

at a moment in time. If there are two otherwise identical firms, A and B, with different leverage ratios, one is expected to be charged a higher rate of interest than the other. This is a reasonable assumption to make when dealing with a single firm taken in isolation." This is how Kalecki’s quote should be interpreted, that is “the wealth position” of the entrepreneur is “endangered” only during the actual time when finance is being expanded. This argument is related to Minsky’s (I975) notion of financial fragility. As stated above, Minsky argues that as firms finance their expanding

Horizontalists and Structuralists

I95

production and investment with bank credit, they will likely move toward riskier positions, until the system is brought near a financial crisis. But Minsky — and structuralists — make an additional argument. They argue that Kalecki’s principle of increasing risk not only holds in a static environment, but that it also applies to a growing economy. Minsky (1975) argues that the leverage ratio of firms increases over the cycle, that is through time. This view is expressed by Hewitson (1995, p. 296): “During a boom, when liquidity preference is relatively low and assets are perceived to be highly liquid, the interest rate rises from a low level as the debt to equity ratio rises." Hewitson (I995, p. 296, note) claims however that “an implicit assumption underlying [this reasoning] is that the debt to equity ratio rises functionally with increased lending.” Dow (1996, p. 502, emphasis added) — a leading proponent of this approach — has also claimed that “bank loans charges are conventionally set within a rate structure (base rate plus risk premium). The effective supply of credit curve is thus in fact [upward sloping], which carries over the cycle with the degree of perceived risk. [There is a] systematic movement within the structure over the cycle which should be made explicit.” This view is also clearly expressed by Wray (I995, p. 278, emphasis added): ls there any difference between lending, say, 80 percent of the value of a development project, and, say, I05 percent? At any point in time, would a bank charge the same interest rate for these two very different loans? Over time, would a transition from an economy in which the nomial rule of thumb set a maximum development loan at 80 percent to an economy in which I05 percent becatne the norm generate higher interest rates?

Horizontalists accept the first part of Wray’s statement, that “at any point in

time” the risk premium associated with a higher leverage ratio is a reasonable assumption. But horizontalists reject the notion that “over time,” the leverage ratio would increase. This argument relies on the existence of macroeconomic laws quite independent of microeconomic ones, a fact, as pointed out by Isenberg (1988, p. I047), which is absent in the structuralist argument: “Minsky’s financial fragility hypothesis builds on a microeconomic foundation to produce macroeconomic effects." In a Kaleckian macroeconomic setting, increased lending/borrowing leads to increased profits, since increases in investment and capitalist consumption lead to greater profits. As profits increase through the cycle, the leverage position of firms is not threatened. In fact, the realization of profits leads to decreasing leverage ratios. The essence of the argument has

I96

Credit, Money and Production

been put forth by Lavoie (1996, pp. 285-6) who precisely questions the validity of extending Kalecki’s principle of increasing risk to any dynamic, or macro, level: [ls it a] relevant assumption at the macroeconomic level? Does the leverage ratio of industrial firms rise when economic activity increases? When investment increases, profits increase as well, unless other elements induce reduced profits, such as a higher rate of savings by households, or a deficient trade balance. Individual efforts to increase leverage ratios may lead to lower aggregate debt to equity ratios.

This is what Lavoie calls the “paradox of debt." In Lavoie’s growth models, increased accumulation can even be associated with falling leverage ratios (see Lavoie, 1995a). Seccareccia (I988, p. 56) argued this point over a decade ago: “Overindebtedness [high debt-equity] becomes a generalized problem only when firms, as a group, reduce their rate of spending.” This is also Crotty’s (1996, p. 337) criticism of the structuralist approach: “High sales and capacity utilization raise profits, cash flow, and the value of equity, and thus lower the debt to equity ratio.” What this represents is a rightward, or dynamic, shift of Minsky’s IF curve, as shown in figure 5.5 below. Moreover, as Lavoie (1996, p. 287) claims Kalecki’s principle of increasing risk may not hold even in a static setting due to the fact that most established firms have lines of credit. Fimis can thus increase their borrowing needs at will without facing increased rates each time they borrow: Any decision on the pan of the firm to draw on its credit line would be automatically made good, at the ruling prime rate of interest. Such loans are thus automatically created. without a discretionary decision on the part of the managers of the bank. The principle of increasing risk, as initially presented by Kalecki (I937), does not quite hold for the analysis at hand.

As pointed out by Lavoie (I996), Minsky (1975, p. 114) has at times recognized the paradox of debt. More recently, however, Minsky (I992, p. 32) clearly recognizes the dynamic shift in the IF curve: In the aggregate the profit flows that are available to service the liabilities of finns (including dividends or common stock) and to be the source of retained earnings that may be used as the base for the financing of investment demand are determined by the composition of aggregate demand. The basic framework is the Kalecki equation, I = Pfts.

Horizontalists and Structuralists

I97

Minsky then adds the Kaleckian “addendum” that investment determines profits. As a result, “In the aggregate firms are finding their debt burdens lightened even as their capacity to carry debt increases [since] in the aggregate profit flows are not determined by the action of individual firms” (Minsky, 1992, pp. 33-4). Also, Minsky (1992, p. 36) stipulates further that _ A Prices

v

Demand pricefor capital lenders’ risk borrowers’ risk

p

Supply price ofcapital

IF, 4

.

V

gt’

IF»

>

New Retained earnings

Figure 5.5: The dynamic two-price Minsky diagram

During a period characterized by robust finance, the effect of the leveraging of intemal funds with extemal funds leads to profit flows exceeding those that went into the negotiations that led to investment demand: as a result the

burden of debts in terms of the cash flow allocated to debt validating is lower than expected and the amount of leveraging necessary for the financing of investment is also smaller than expected. A run of such pleasant surprises increases the willingness to lever by both sides of the bargaining table, further increasing investment and aggregate profits.

These references to Minsky are important for the fact that they question the structuralist argument of an upward-sloping money supply curve, and his own as well, that increased investment necessarily implies a higher debt—equity ratio for finns. As Minsky claims, increased investment

I98

Credit, Money and Production

implies greater profits. If this is the case, then increased credit demand would not translate into a more risky position, and there is thus reason for the rate of interest on loans to increase to reflect this fact.

CONCLUSION This chapter explored in greater detail some of the debates between the horizontalists and the structuralists with respect to monetary policy in general, but more specifically, to the slope of the credit supply curve. The conclusions which can be drawn from this exercise are as follows: 1) There is greater consensus between post-Keynesians than there are differences. As such, Pollin’s assessment is rejected. Overall, postKeynesians accept the validity that credit is largely credit — and demand — determined, and that markets may not reach a full employment position. 2) Despite the presence of similarities and consensus, a “synthesis” is nonetheless some distance away. The differences, while dealing with perhaps details of post-Keynesian monetary policy, are still indicative of a certain malaise. Yet, post-Keynesians are closer today than they were a decade ago. The remaining differences can be summarized as whether horizontalism is a special case of structuralism (Palley, 1994), or whether structuralism is a special case of horizontalism (Lavoie, I996). Which it is will depend on how one perceives the world, and how sufficiently one has broken his or her ties with the orthodox tradition. 3) The chapter attempted to show some of the shortcomings of the structuralist arguments, indicating that many of their criticisms of horizontalism have already been noted by accommodationists, and taken into consideration. Nonetheless, they do not amount to a rejection of the horizontalist position, but rather have shown that horizontalism is sufficiently general as to accommodate structuralism as a special case. This is true in the case of a refusal by the central bank to fully accommodate the reserve needs of banks, as it is in the case of the importance of liability management. In no sense therefore is horizontalism an “extreme” endogeneity position. 4) On another level, it was shown that the structuralist arguments justifying an upward sloping, credit supply curve are not necessarily validated at the macroeconomic level, especially when taking into consideration the Kaleckian causality between investment and profits. This has led Lavoie (I996, pp. 286-7) to claim that “what must be concluded on this issue is that neither Kalecki’s principle of increasing risk nor Minsky’s financial fragility hypothesis, whatever their virtues within their domain of

Horizontalists and Structuralists

I99

validity, can sustain the hypothetic construction of an upward sloping creditmoney supply curve.” 5) It has been shown that Keynes came close to endorsing the horizontalist position, in contrast to Wray’s (1995, p. 281, n. l) conclusion that Keynes

“would not have accepted horizontalism.” There is hence no reason to believe that the rate of interest will necessarily rise through the cycle. As argued by Keynes, this is akin to the “classical” theory of loanable funds. The following chapter will continue the debate on post-Keynesian theory by examining more closely some of the arguments of the structuralist postKeynesians. It will be shown that many post-Keynesians have yet to make a complete break with orthodoxy. In their approach,.there is still a great amount of baggage from the past. This does not suggest that the whole of the structuralist position is invalidated. It does suggest, however, that post-

Keynesians should take a closer look at their views and purge the more orthodox elements. It is in fact these, as will be argued later, that contribute largely to the many debates between the structuralists and the horizontalists. NOTES I.

The second of these issues will be dealt with in greater detail in chapter eight, along with the discussion of the appropriate role for liquidity preference within the circuitist framework.

2.

A more fundamental disagreement is whether increases in economic activity lead to increases in the rate of interest. For structuralists, this implies that they accept an

3. 4. 5. 6.

upward-sloping money supply curve, an idea shared with neoclassical theory. Note that for Lavoie (I996), it is structuralism which is a “special” case of horizontalism. In this chapter, we will show that this is not the case. Chapter eight will answer many of Palley’s criticisms. Arestis appears to have considerably changed his views on this issue. See Arestis and Howells (I996). For Pasinetfi (I974, p. 47), the rate of interest is “determined exogenously with respect

to the income generation process.

Whether, in particular, liquidity preference, or

anything else determines it, is entirely immaterial."

7.

8.

9.

Pivetti (I991, p. I5) makes the same argument. While claiming that the rate of interest is purely exogenous, the author argues that “we do not think that interest-rate policies are unconstrained ~ that is that monetary authorities can always bring the rate of interest to whatever level they seem desirable." Moore (I988, p. xiii) argues that “the quantity of nominal money demanded is thus

always and necessarily equal to the quantity of nominal money supplied.“ The supply of credit is not independent of the demand for it. According to Pivetti (I996. p. 75). in I930, Keynes “still regarded Wicksell’s ‘natural rate of interest’ as a very useful and significant concept.” At the end of chapter I7 of the General Theory. however, Keynes (vii, p. 243) writes, “I am no longer of the opinion

Credit, Money and Production that the concept of a ‘natural’ rate of interest, which previously seemed to me a most promising idea, has anything very useful to contribute to our analysis." Following the publication of the General Theory, Keynes (xiv, p. 213) himself claimed that the liquidity preference theory of the rate of interest was “exceedingly simple" and “by itself does not carry us very far." He also noted (xiv. p. 215) that “to speak of the ‘liquidity-preference theory’ of the rate of interest is, indeed, to dignify it too much." This argument is true only in the short run. ln the medium to long run, a high or increasing rate of interest may slow dovm the growth rate of the money supply as it affects the demand for bank credit. This argument is also made by Lavoie (I992, I996). ln chapter eight, a micro theory of bank lending is developed where horizontalism and credit constraints are compatible. The expression "rationing" implies the neoclassical notion of scarcity. lt is therefore not “a matter of semantics whether or not the resulting availability of credit is termed ‘rationing’ as argued by Dow (I996, p. 499). Nor can there be any excess supply. Lavoie (I996, p. I5) has proposed that “we may, if we want, consider the demand for credit curve to represent the demand for credit arising from credit-worthy borrowers, i.e., those borrowers who fulfill the norms imposed by the banking system. We may also want to represent the limits on credit lines by a truncated supply of credit: beyond some point, the supply simply vanishes." Current attempts at developing a theory of banking behavior rest uneasily with postKeynesian theory. Dymski (I992, I994) and Neal (I996) develop theories based on a combination of Keynesian uncertainty, risk, asymmetric information, and probability distribution. The argument that central banks can set the interest rate at the full employment level has been discussed above. Structuralists, however, must explain why their views reject the natural rate hypothesis. If the rate rises and falls with economic activity, then surely

there must exist a rate at which the economy is in a market-clearing position. If the rate rises and falls with economic activity, then it must also be gravitating towards some

other rate. Structuralists, on the other hand, argue that liquidity preference influences the premium. In essence, the latter argue that as economic activity increases, so does the rate of interest given decreases in the liquidity positions of banks and borrowers. The proper graphical representation is therefore an upward-sloping money supply curve in money-interest space. Keynes came to accept this view shortly after the publication of the General Theory, resulting from his correspondence with Townshend. Even in the General Theory, Keynes (vii, p. 299) argues that “the ‘income-velocity of money‘ is, in itself, merely a name which explains nothing." This argument holds only when discussion of the velocity is limited to the ex post, that is when “the crucial direction of causality [between loans and deposits] is not challenged by the possibility of unstable velocity." For the velocity of money to increase, it is not sufficient to argue that money is transferred from inactive to active accounts. Money must be put into circulation, and thus circulated. As will be seen in chapter eight, the horizontalist tradition along the lines of the circulation approach is more Keynesian than the structuralist approach, and “richer” in institutionalist details. Set in these tenns, Pollin's position also contradicts Keynes’s position in the General Theory. The argument is also made in Moore (I988).

Horizontalists and Structuralists

201

As Moore claims (I985, p. ll0), “Banks have little ability to reduce their earning assets at their initiative," especially regarding the notion of “calling back" loans. The recent abandonment of reserve requirements in Canada is attributed to the fact that reserves represent substantial losses to banks. As a result, structuralists misinterpret horizontalism when claiming, as does Pollin (I991), that the reserve—loan ratio needs to remain fixed. As loans expand, it is quite normal for the ratio to fall. For a full discussion of this point, see the next chapter. In this sense, the argument is similar to the one introduced in l957.

6. Post-Keynesians and Orthodoxy: “Neo” Post-Keynesians? While examining in the previous chapter the various controversies surrounding the horizontalist and structuralist approaches to post-Keynesian monetary theory, it was suggested, although perhaps just in passing, that some aspects of the structuralist post-Keynesian theory insufficiently break away from orthodox monetary thought. The purpose of this chapter is to discuss this statement in greater detail by exploring the more orthodox elements of post-Keynesian monetary theory. It should be mentioned that the emphasis is placed primarily on the structuralist argument for it is this approach which shares the most with orthodox theory. In this way perhaps the expression “neo-post-Keynesians” is perhaps better suited.‘ This is not to say that the horizontalist position — especially in Moore’s version - does not similarly suffer from some of the

same weaknesses, but if it does, it does so to a much lesser degree. The conclusion which is reached is that ironically post-Keynesians — and structuralists in particular - have failed to sufficiently “push Keynes beyond himself [by developing] the radical elements in Keynes and to discard the conservative” (Cottrell, 1994, p. 588). While Keynes appears to have succeeded in his “long struggle of escape” - albeit only after the publication of the General Theory - post-Keynesian structuralists ought to reconsider some key elements of their approach to money. While stnicturalists have a well developed theory of a monetized economy, where the existence of money and its specific properties affect the decisions of economic agents, they have not yet sufficiently developed a theory of “money,” defined as a theory which explains the creation, circulation and destruction of credit and credit-money. This is a crucial initial step, for without explaining how money is created, as opposed to how it affects economic decision making once it does exist, post-Keynesians are left with an incomplete theory of modem, production economies of credit. It is argued here that the structuralist view can become compatible with the money multiplier model largely — although not exclusively — as a result of the insistence on the “bi-variate” causality between reserves and loans. The obvious reason for this bifurcation is due to post-Keynesians following too closely in the footsteps of Keynes’s General Theory, and then in those of Minsky whose early work, still a force in post-Keynesian theory, was in 202

Post-Keynesians and Onhodoxy

203

many respects anti-Keynes. Keynes’s objective was to break away from “classical” thought, not to bridge the divide which appears to be Minsky’s

intention. The emphasis on the causal role of reserves in post-Keynesian theory poses a number of serious problems, both in terms of policy and in temis of

theory. In a sense, as soon as one accepts that reserves impact on loans,'no matter whether the reserves are exogenous or endogenous, then there is a danger of slipping back into arguments advocated by mainstream theorists, namely in ICITHS of causalities between assets and liabilities. In this case, post-Keynesian theory runs the risk of becoming a “special” case of

orthodoxy. It should now be restated that the multiplier model needs to be interpreted not just as a theory of causality between exogenous reserves and the “money” supply, but also as a theory of the importance of portfolio decisions. In both of these instances, whether reserves are exogenous or

even endogenous, they are both causal in the determination of the supply of credit. In each case, reserves act as constraints given portfolio decisions of

households. Structuralists, as is argued below, attribute to reserves (the lack of reserves) a causal role in determining the supply of bank credit, largely

based on portfolio choice decisions. This view, however, contradicts all that from which post-Keynesians have tried to escape: the scarcity of funds. Overall, it appears as if post-Keynesian monetary theory does not sufficiently differentiate itself from more mainstream theories of money.’ As a result, this has led some post-Keynesians to adopt a position which at times is not unlike that depicted by the orthodox multiplier model. In the process, the post-Keynesian theory of money endogeneity is also jeopardized. Before proceeding, let us be clear that it is not argued here that postKeynesian monetary theory in its entirety is equivalent to orthodoxy. As explored in chapter two, there are plenty of reasons to show why postKeynesian theory is in many respects revolutionary and a definite challenge to orthodox thought, even in its New Keynesian style. However, some elements of post-Keynesian theory, especially when dealing with money, are not as revolutionary as they could - or should - be. The purpose of this chapter is to point to these weaknesses and purge them from post-Keynesian theory. Chapter one sought, in this way, to rebuild post-Keynesian theory on more solid, and heterodox, grounds.

204

Credit, Money and Production

POST-KEYNESIANS AND ORTHODOX THOUGHT The starting point of the post-Keynesian, structuralist theory of endogenous money is the notion that the demand for money includes four motives to hold money as defined by Keynes: the transactions motive, the precautionary and speculative motives, but also the finance motive. Credit, in the sense of the finance motive, is money proper. It is a transfer of purchasing power from the future to the present (Minsky, 1991), or as Davidson (1980, p. 297) would have it, money, and thus credit, is a “time machine.” As argued in chapter two, while credit is money in post-Keynesian theory, this can take one of two forms. Either the finance motive is a separate motive altogether (Smith, 1979), or it is considered an “addendum” to the transactions motive (see Davidson, 1965; Chick, 1983; Bibow, 1995). Either way, the finance motive and the demand for credit is a component of the demand for money (see also Rousseas, 1989; Knodell, 1995; Wray, 1995). This in itself is not orthodox, but the analysis and implications which structuralists draw from this are at times problematic. However, once these two concepts are separated — as they were in chapter one — much

unnecessary confusion is clarified. Credit creates money instantaneously as a quantum of purchasing power. There is no money (or new money) without credit. This is the meaning of the expression “loans create deposits." Discussions over the definition of what constitutes money, its functions and purposes, become secondary to the discussion over the creation of credit-money. Before proceeding, two points need to be addressed. First, failure to clearly differentiate between credit and money can lead post-Keynesians to adopt either one of the theories of endogenous money described in chapter two. However, recall that only the revolutionary theory of endogenous money is consistent with post-Keynesian theory. The second approach is consistent with the money multiplier model where changes in households’ (and banks’) portfolios affect the supply of loans. Second, the inability of post-Keynesians to distinguish between credit and money suggests that while they have a theory of a monetized economy, they do not have a theory of money. In post-Keynesian theory, the emphasis is on the “essential properties of money" — Keynes’s famous chapter 17 — where these properties can influence the economy in various ways. The emphasis is not on the creation of money through credit. As Keynes (vii, p. vii) claims in the Preface of the General Theory, “A monetary economy is essentially one in which changing views about the future are capable of influencing the quantity of employment and not merely its direction." In such a setting, a post-Keynesian view of the real world is claimed to be

Post-Keynesians and Orthodoxy

205

more “general” than that of neoclassical theory essentially because it incorporates money.’ But this is not a satisfactory approach. A theory of a monetized economy must also include a theory of money - that is, a theory which explains how money is created, how it is circulated, how it is destroyed, and how its circulation is interrupted or affected by various macro elements. Granted, money may affect the economy in various ways. There is little doubt that changes in portfolio may affect households’ and firms’ decision making, and may even have other ramifications. Households may indeed decide to hold on to money for a number of reasons, and saving may well depress effective demand. However, there is a great difference between what money does once it is created, and how it is created. A theory of a monetized economy must be accompanied by a theory of money creation. Ironically, this is also the criticism raised by de Largentaye (1979, p. l2) against Keynes’s General Theory. According to the author, the General Theory is not “in a subsidiary way a theory of money, but rather a theory of a monetized economy [where] the theory of employment and the theory of money are merely two facets of one and the same analysis." This is one example of how closely post-Keynesians have followed Keynes.

In Davidson (1980, p. 297), one also gets the impression that the author is interested in describing the role and functions of money, as opposed to how it enters the economy: “Thus the first definitional feature is: (1) money is the means of contractual settlement. Money is also (2) capable of serving as an

instrument to transport generalized (nonspecific) purchasing power over time, i.e., money can act as a one-way (present to future) time machine.” Here, despite emphasizing the irreversibility of time, Davidson nonetheless talks of money as a unit of account, and then as a store of value.‘ In other words, post-Keynesians are primarily interested in arguing why money is not neutral - even in the long run — as opposed to how it is created. According to Davidson (1996, p. 62), “The existence of the societal institution of legally enforceable forward contracts denominated in nominal (not real!) terms [creates] a monetary environment that is not neutral, even in the long run.” Post-Keynesians argue that their approach to economics stands in stark contrast to neoclassical theory, in essence fomnng a coherent altemative to mainstream theory. It contends to be “returning to the age-long tradition of common sense” — as Keynes (xiii, p. 552) would have it. Yet, their reliance on the General Theory in order to develop an altemative monetary theory is somewhat misplaced, especially given Keynes’s admission in the preface of the General Theory, that he had left the “technical details" of money “in the background.” And while the General Theory remains an excellent contribution to the theory of effective

206

Credit, Money and Production

demand, it remains lacking in the area of money. Keynes, of course, knew this, and many claim that it was intentional — in order to “get a better hearing” as Robinson reminds us. Yet, this has not prevented many post— Keynesians from tuming to Keynes and the General Theory to develop their views, to the point, unfortunately, of ignoring Keynes's other works, despite Davidson’s (1965) effort to introduce the finance motive into the literature. Many heterodox economists — for instance, proponents of the circulation approach — have criticized this canonical emphasis on the General Theory. For instance, Fontana (1997, p. 3, n. 3) claims, “In the Anglo-American Post-Keynesian tradition there has been an obsessive analysis of the General Theory with almost oblivious [reference] to Keynes’ works.” This echoes a statement made earlier by Lavoie (1985, p. 63) that “one danger when starting to study money matters is to follow Keynes too closely, in particular his writings in the General Theory.” Such criticism of the General Theory, usually relies on the following five observations. First, the General Theory is not about monetary theory, but rather about effective demand. Second, as Keynes himself would explain in the preface to the Gennan edition of the General Theory, the book was only a “transition” from the “classical” theory to new lines of exposition. As such, his views in the General Theory were only partially developed, and therefore subject to changes and clarification. Third, Keynes still made use of Marshallian analytical tools, thereby placing him within the orthodox tradition despite his best efforts to escape it. Fourth, the supply of money is exogenous. Whether this was for strategic reasons or not, it still poses some problems. The early postKeynesian response to Friedman, for example, followed the General Theory too closely, with repercussions which can still be seen today. Finally, the General Theory does not contain a sufficiently developed sequential analysis, as Keynes would later develop in his Economic Journal articles. In many respects therefore, the General Theory remains too static for a sufficient theory of money. The General Theory itself does not therefore constitute a sufficient break with orthodoxy. This was precisely the criticism, for instance, raised by Ohlin in an exchange between himself and Keynes (see xiv, p. 196) following an article which Ohlin sent to Keynes in March 1937: “In this as in some other respects Keynes does not seem to me to have been radical enough in freeing himself from the conventional assumptions." This was Keynes’s assessment as well. For instance, in an August 30, 1936 letter to Harrod, Keynes (xiv, pp. 84-5) writes, “What some people treat as unnecessarily controversial is really due to the importance in my

Post-Keynesians and Orthodoxy

mind of what I used to believe.

207

You don’t feel the weight of the past as I

do. One cannot shake off a pack one has never wom.” Similarly, in his notes to a Stockholm lecture that same year, Keynes (xiv, p. 100) wrote, “I also make it more difficult than I need because I myself am so much bound by the past. I am extremely anxious to emphasize and bring to a head my difference from orthodox theory precisely because I was brought up in this faith."5

Many heterodox economists have made the same assumptions regarding post-Keynesian monetary theory. For instance, Winnett (1992, pp. 47-8) claims that: Some versions of the endogeneity assumptions draw post-Keynesians into diversionary disputes with their neo-classical colleagues and inhibit the development of their own distinctive approach. By this I mean that the phrasing of the assumptions is such that they are already operating within the confines of a basically neo-classical model, arguing about the shape or independence of the money supply function. They are then trapped, an

opportune word, within the confines of a special case. In posing the problem in temis of demand and supply functions for money, they have already made very large concessions to the neo-classical point of view, and

they have denied themselves many possibilities for showing how their assumptions might form a coherent altemative. What this suggests is the inability of post-Keynesian theorists to fully break away from neoclassical theory. Nell and Deleplace (1996, p. 6) argue precisely that in the context of monetary theory, the “debate [over monetarism] took place on grounds defined by orthodoxy.” Vestiges of neoclassical theory are evident in the use of Marshallian methodology and marginalism, prompting Moore ( 1991, p. 404) to claim that “the structural position makes less of a break with the mainstream view.” This is evident for two reasons. First, as we have shown above, both neoclassical and structuralist endogenists posit an upward-sloping money supply curve thereby implying that the money supply is partly exogenous and partly endogenous. The debate then concems the degree of endogeneity. Second, as Lavoie (1996, p. 2) has demonstrated, an upward sloping curve borders on the loanable funds theory. According to the author, Those who believe that the credit-money supply curve should be drawn upward sloping cannot rely upon the fact that for more than thirty years

Minsky has been asserting that interest rates and leverage ratios should rise when activity increases, since this assertion is ultimately based on a neoclassical model of the economy, and hence is meaningless from a postKeynesian perspective.

208

Credit, Money and Production

In this approach, the rate of interest becomes clearly the price of scarcity the fact that demand is greater than the available supply. Carvalho (1996, p. 73; emphasis added) has admitted as much. According to the author, the interest rate rises following an increase in demand “not because of the necessary stimulus to consumers to save more in order to finance the investment, but because the demand for money would rise above the existing supply." Also, increases in investment financed by credit imply an increase in the rate of interest as posted by the loanable funds approach. In fact, some would argue that the structuralist post-Keynesian view is wholly compatible with neoclassical views. For instance, Screpanti (1995, p. 1) argues that structural endogenists - by positing an upward sloping money supply curve and resting their argument on the lack-ofaccommodation premise — resort “more or less knowingly, to the neoclassical argument of maximizing behavior in portfolio management." In some respect, post-Keynesians have accepted thiscriticism — and to some degree have embraced it. For instance, Cottrell (I988, p. 245) readily acknowledges that his own approach to post-Keynesian theory is an “attempt to mark out a middle position between the orthodox money supply/demand paradigm and the position Kaldor espouses." Such a statement therefore appears to argue that structuralism is not fully an alternative to orthodox thought, but rather a variation of it. At another level, post-Keynesians do not reject the notion of an exogenous money supply. Rather, they argue two points. First, a fully exogenous money supply applies to a different historical period. Second, even with the existence of credit, money is only partially endogenous. With respect to the first point, Minsky (1991, p. 208) claims that “there are periods in history and economic conditions where the money supply was mainly endogenous and other periods and conditions where the money supply was largely exogenous.” Guttmann (1990, p. 820) also argues that “since the supply of gold was exogenously fixed, it was not capable of responding to the credit-and liquidity-needs of the economy with sufficient elasticity.” This is also the argument of Niggle (1991), Dow (1984), and even Moore (1996, p. 89) who claims that the exogenous—endogenous debate is somewhat misleading since “both views are correct, but that each is applicable to a difierent historical period“ (see also Davidson, 1972; Davidson and Weintraub, l973).6 Cottrell (1988, p. 296) claims that “the case that the stock of outstanding credit is purely demand-determined is overstated." Cottrell (1986, p. 2) has also argued that the money supply can be either “causally active or passively endogenous,” therefore suggesting that “it is possible for the money stock to be greater or less than the aggregate of desired money balances.”

Post-Keynesians and Orthodoxy

209

In fact, Cottrell (1986, 1988) does not refute Moore’s criticism that his analysis suggests that “the central bank can initiate exogenous increases or decreases in the economy’s stock of money.” This, Cottrell (1988, p. 295) refers to simply as a “weaker version of monetary endogeneity.” As for the argument that the “money supply” is only partly endogenous, post-Keynesians accept the view that the central bank can increase the “money supply” at will, simply by expansionary open-market operations. The argument of asymmetry in monetary policy is made by most post-

Keynesians who claim that the supply of money is independent of the demand of money, but only to some degree. This argument appears to be shared by structuralists and Moore. For instance, Moore (1996, pp. 48-9) claims that, During periods of deflation, when loan demand is weak, the mainstream view

is correct. Central bank initiated open market purchases result in a multiple expansion of bank deposits, providing only that banks invest their excess reserves in marketable securities.

The monetary authorities are always able

to increase the money supply at their initiative by the purchase of securities in the open market. The money supply can always be increased, by some multiple, of the growth of bank reserves, at the initiative of the central bank.

Davidson (1972) also makes this argument. For instance, Davidson (1972, p. 303) has argued that “the money supply can be expanded exogenously

(i.e., by the deliberations of the Central Bank) or endogenously when the banking system responds to an increase demand for money.” Elsewhere, Davidson (1989, p. 489) claims that “as early as 1972 Post Keynesian monetary theory had not limited itself to the concept of an endogenous money supply. Observed increases in the money supply could be either exogenous or endogenous" - what he labeled the “portfolio change process.” Similarly, Minsky (l99l, p. 208) claims that “typically, the money supply is in part endogenous and in part exogenous.” Weintraub (1978, p. 75) also argues that “there can be a good deal of both ‘endogeneity’ and ‘exogeneity’ in the [monetary authorities’] actions.” Rousseas (1989, p. 478) has argued along similar lines: “Verticalists notwithstanding, the money supply is not perfectly inelastic (exogenous), nor are the horizontalists right in arguing that it is perfectly elastic (fully endogenous). It is partly both. The degree to which the supply of money is positively sloped depends on the discretionary policies of the Federal Reserve.”-

Niggle (1991, p. 143) claims that7

210

Credit, Money and Production the money supply is seen as only partially endogenous; the elasticity of credit (and money) with respect to the rate of interest is neither perfectly elastic (horizontal when graphed) nor perfectly inelastic (vertical when graphed), but rather somewhere in between (the money supply curve should be drawn with an upward slope against the rate of interest).

For these reasons — and reasons outlined below — Lavoie (1996) has claimed that structuralism and other aspects of post-Keynesian theory does not sufficiently break with orthodox monetary thought. The debate simply becomes about the degree of endogeneity. Is the money supply schedule more or less inelastic? Is it more or less independent? The attempt to demonstrate the more orthodox elements of postKeynesian theory rests on the following five characteristics — although some of them are necessarily a consequence of another: 1) uncertainty; 2) portfolio behavior; 3) the stock of money; 4) velocity; 5) reserves.

UNCERTAINTY8 At first glance, it may appear somewhat misleading to include uncertainty in the list of the more orthodox elements of post-Keynesian theory. After all, post-Keynesians have repeatedly claimed that uncertainty is what differentiates them from other schools of thought. It is safe to claim that the notion of uncertainty has not been taken seriously by those other schools (Lavoie, 1992, p. 43).9 Such a statement therefore demands clarification. The argument presented below is not that uncertainty in itself is a mitigating factor, but rather the use post-Keynesians make of it. Whether the presence of a theory of uncertainty in post-Keynesian monetary theory represents a necessary step in developing an altemative to orthodoxy is a question which is not discussed here. It is quite clear that orthodox thought, from New Keynesian economics to neoclassical and Walrasian general equilibrium models, makes no room for the incomplete knowledge of the future. However, it can be claimed that it is not a sufficient prerequisite. It requires that the notion of uncertainty be used in specific ways. The importance of uncertainty — or of what Keynes (xiv, p. 114) calls an “awkward fact” — in post-Keynesian theory has been discussed ad nauseum. It is emphasized in virtually all aspects of post-Keynesian thought: financial fragility, production, investment, and monetary theory. In fact, most postKeynesian authors would rank it among the defining characteristics of postKeynesian thought. Although Davidson (1972) is perhaps best associated with the post-Keynesian position on uncertainty, it is safe to assume that most - if not all — post-Keynesians — and other heterodox economists — have

Post-Keynesians and Orthodoxy

211

emphasized it, at one time or another (Shackle, 1974; Vickers, 1987; Asimakopulos, 1991). As Robinson (1979, p. xi) argues, “It is from this point that post-Keynesian theory takes off.” While the use of uncertainty in post-Keynesian theories of production and investment is largely unproblematic, it is the way post-Keynesians link it to money which causes problems. Before addressing this issue, however, it is necessary to clarify the concept of uncertainty, what it is and how it enters into economic analysis. The role which uncertainty plays in post-Keynesian theory cannot be sufficiently emphasized. It is certainly true to say that uncertainty — as postKeynesians interpret it - does not play a role in orthodox or neoclassical theory. Rightly so, Keynes argued that his emphasis on this concept was

sufficient - in his mind — to be one of the “main grounds” of his departure from orthodoxy. Similarly, Shackle (1982, p. 438) claims that uncertainty is the “elemental core of Keynes’ conception of economic society," as well as the “ultimate meaning” of Keynes’s ideas (see Shackle, 1967, p. 129). Uncertainty thus represents a radical departure with the prevailing body of thought. Many post~Keynesians have in fact interpreted the emphasis on uncertainty in similar ways, claiming that it is a sufficient condition to make their approach to theory different from neoclassical theory. The emphasis on uncertainty nattually follows the importance post-

Keynesians attach to the notion of time. As Carvalho (1984, pp. 265-7) claims, “Until one deals explicitly with the concept of time one cannot analyze the concept of changes in the economic system. If the PostKeynesian approach is to become a viable altemative to neoclassical economics, it must explicitly develop the importance of [historical] time to economic processes.” Post-Keynesians and other heterodox economists have rightly emphasized the notion of logical time versus that of expectational time (see Dow, 1990, p. 12; Termini, 1981; and Carvalho, 1984), or that of historical versus mechanistic time (Georgescu-Roegen, 1971). It therefore follows that postKeynesians are primarily interested in situations of transition, as opposed to static equilibrium analysis. Because production, investment, and other related actions take place in time, and because such decisions are necessarily based on expectations, uncertainty will influence the choices we make, especially in light of disappointed expectations. As it has now been argued extensively, Keynes fully endorsed the notion of uncertainty. According to O’Donnell (1992), Keynes developed his views as early as 1910, in teaching (from 1910 to 1914) a course entitled Principles of Economics, covering the theory of value and distribution. At this time, Keynes expressed his views on an uncertain future in the following way (quoted in O’Donnell, 1992):

212

Credit, Money and Production It is this element of ignorance which complicates the analysis of the economist. His method is to assume a certain fairly well defined amount of intelligence and persistence. and to regard cases where the course of events upsets anticipations as abnormal. But to the extent to which he assumes that the course of events can be accurately foreseen, he is abstracting from the real order of affairs.

Keynes developed these views extensively through the years, and while the notion of uncertainty is not sufficiently developed in the General Theory, it is nonetheless implicitly recognized. For instance, on page 148 (footnote l), Keynes had this to say: “By ‘very uncertain’ I do not mean the same thing as ‘very improbable’.“'° In his post-General Theory articles — which Shackle (1967, p. 136) calls the “third edition“ of the General Theory — Keynes (xiv, pp. ll3—14) returns to the concept of uncertainty clearly emphasizing its importance. Consider for instance the following well-quoted reference: By “uncertain” knowledge, let me explain, 1 do not mean merely to distinguish what is known for certain from what is only probable. The game of roulette is not subject, in this sense, to uncertainty; nor is the prospect of a Victory bond being drawn. Or, again, the expectation of life is only slightly

uncertain. Even the weather is only moderately certain. The sense in which l am using the term is that in which the prospect of a European war is uncertain, or the price of copper and the rate of interest twenty years hence, or the

obsolescence of a new invention, or the position of private wealth owners in the social system in 1970. About these matters there is no scientific basis on which to fomr any calculable probability whatever.

Keynes (xiv. pp. 113-14, 124) then adds that about the future, “We simply do not know. Actually, however, we have, as a rule, only the vaguest idea of any but the most direct consequences of our acts. The future never resembles the past. We do not know what the future holds." Keynes then specifically refers to "classical" (i.e. neoclassical) theory as “one of those pretty polite techniques” precisely because it overlooks uncertainty. The discussion over uncertainty is undertaken in relation to risk. PostKeynesians have steadfastly argued that uncertainty can never be reduced to risk (Lavoie, 1992). Post-Keynesians therefore differentiate themselves from their orthodox counterparts by emphasizing that probability distributions or standard errors cannot be calculated, or if values are assigned to them, they are meaningless. Risk is described as a situation where the probability distribution is “numerically measurable and describes the relative frequency of a given event in the course of time” (Hoogduin, 1987, p. 53). According to Lavoie (1992, p. 43), “There is risk, or certainty equivalence, when each choice

Post-Keynesians and Orthodoxy

213

leads to a set of possible specific outcomes, the value of which is known, each outcome being associated with a specific probability.” For mainstream economists, risk is how they define uncertainty. As Davidson (1996, p. 22) tells us, “The mainstream view [is] that the concept of ‘uncertainty’ is synonymous with that of probabilistic risk. In this view, a certain event has a probability of unity. Any event whose probability is less than unity is uncertain." Orthodox economists have always confused these issues. As Knight (1921, pp. 19, 232) remarked more than three-quarters of a century ago, “Uncertainty must be taken in a sense radically distinct from the familiar notion of Risk from which it has never been properly separated. The word ‘risk’ is ordinarily used in a loose way to refer to any sort of uncertainty viewed from the standpoint of an unfavorable contingency?" The problem arises of course when dealing with theories of choice and decision making. As Davidson (1996, p. 57; see also 1972, 1982-3, 1988) has made clear on numerous occasions, “Probability distributions are not the basis for comprehending real world behavior under uncertainty."'2 Keynes (xiv, pp. 112, 122) recognized this as well: The calculus of probability was supposed to be capable of reducing uncertainty to the same calculable status as that of certainty itself. This false rationalisation follows the lines of a Bentharnite calculus. The hypothesis of a calculable future leads to a wrong interpretation of the principles of behavior which the need for action compels us to adopt. On the other hand, uncertainty is described as the complete lack of knowledge regarding either the specific outcome, the probability of that outcome occurring, or both. This distinction was also made earlier by Knight (1921, p. 233): The practical difference between the two categories, risk and uncertainty, is that in the former the distribution of the outcome in a group of instances is known (either through calculation a priori or from statistics of past experience), while in the case of uncertainty this is not true, the reason being in general that it is impossible to form a group of instances,'because the situation dealt with is in a high degree unique.

There is another argument which enters into the discussion of uncertainty, and that is the weight of the argument. This can best be described as the quality or the quantity of information we have when making decisions. Keynes (vii, p. 84) believed that increasing the amount of information translated into increased weight, that is greater confidence in the decisions we take. Shackle (1958, p. 33) has argued that “the word uncertainty

214

Credit, Money and Production

however suggests an objectively-existing future about which we lack knowledge, rather than a void to be filled by new creation." Here, the author talks in terms of lack of knowledge implying that additional information can always be obtained. Post-Keynesians, however, have generally claimed — although not all — that uncertainty is independent of the weight of an argument, or the quantity of information held. Increased information does not necessarily lead to less uncertainty. The acquisition of more information in no way implies that uncertainty would diminish, in fact, it may precisely imply the opposite. For instance, obtaining more information may have the negative effect of questioning the pre-existing stock of information, such that the two stocks of knowledge are mutually exclusive rather than complementary. This argument was made, among others, by Lavoie (1992, p. 46): “Further information might reduce the degree of confidence without necessarily changing the assessed probability. There is new information, but this information has destroyed part of the past accumulated knowledge, or has uncovered new aspects of unsuspected ignorance." Similarly, no quantity of information regarding the past can make the future less unknown. “We simply do not know," as Keynes claimed. Uncertainty, in Keynes's sense, should imply therefore two things: first, the information is not available or does not exist; and second, even if agents acquired it, it would not imply increased confidence and reduced uncertainty. Lachmann (1943, p. 12) emphasized this point a long time ago: Economic action concemed with the future, so far from being strictly detemtined by a set of objective “data,” is often decided upon in a penumbra of doubt and uncertainty, vague hopes and inarticulate fears, in which ultimate decisions may well depend on mental alertness, ability to read the signs of a changing world, and readiness to face the unknown.

But the existence of uncertainty does not mean that households or other “economic units” do not act upon the information they hold. They possess nonetheless what Knight (1921, p. 229) calls the “subjective feeling of confidence” where they can rely on the “deliberative confidence” of their decisions. Likewise, Keynes has emphasized a number of rules of thumb which may guide “practical men" in their decisions. This discussion does not suggest that the weight — i.e. the “state of confidence" or the “degree of certainty“ — does not play an important role, but that uncertainty does not diminish with increased information, and neither does the confidence increase. Uncertainty in this sense is exogenous.

Post-Keynesians and Orthodoxy

215

PORTFOLIO BEHAVIOR AND THE STORE OF VALUE While the above concept of uncertainty is certainly a welcomed addition to post-Keynesian theory, two arguments are made here, although the second reason is somewhat a result of the first. First, although perhaps a necessary condition, uncertainty is not in itself sufficient to claim that post-Keynesian monetary theory is an altemative to neoclassical theory. Thisris directly related to the use post-Keynesians make of uncertainty, which they link to the existence, and behavior, of money. The debate has often been reduced to whether households use money for speculative purposes (O’Donnell, 1992), or whether it is used for precautionary reasons (Dow, 1996; Chick, 1983). Either way, money is seen as a result of the pen/asiveness of uncertainty. Second, post-Keynesians do not generally develop the notion of uncertainty in the context of banking — and lending — decisions. This is

strange, since post-Keynesians have placed considerable emphasis on the credit creation role of banks, and on the fact that commercial banks “hold the key” to economic expansion. Yet, they have largely ignored the microeconomic behavior of banks. When post-Keynesians have attempted

to develop a theory of banking, it has been along the lines of New Keynesian theory of asymmetric information (Dymski, 1990). Yet, just as banks are important players in the overall system, they also suffer the consequences of an uncertain future. The amount they lend (or in other words, the degree to which they restrain credit) can therefore be linked, not to a given supply relative to demand, but rather to the effect of uncertainty. The notion of uncertainty in post-Keynesian theory is largely - although not exclusively — related to the role which money plays as a store of value. And it is this role of money which post-Keynesians have emphasized, as opposed to money’s other roles, such as a means of circulation or a unit of account. Post-Keynesians argue that money exists because of uncertainty. Take for instance the following passage in Dow (1996, p. 36): “Money’s role itself is the product of uncertainty.” In this sense, economic agents may want to hold on to money in case of unforeseen events. Money bridges time by allowing economic agents to keep the value of their wealth unchanged. Dymski (1996, p. 381) claims that “exogenous uncertainty creates a need for a store of value: agents’ willingness to undertake time-using activities may entail a desire simply to hold resources in reserve.” This position is traced back to Keynes (xiv, p. 116) who maintained that households keep money balances because of “calculations and conventions conceming the future. The possession of actual money lulls our disquietude; and the premium which we require to make us part with money

216

Credit, Money and Production

is the measure of the degree of our disquietude.” This is why “anyone outside a lunatic asylum" would hold on to money. Rousseas (1992, pp. 12, l5) also makes this claim, arguing that “money

is what binds the present and the future in a world of uncertainty. It is the capitalist response to uncertainty that gives money its peculiarity." Moreover, according to the author (1992, p. 22), “In a non-barter monetary economy, it is the continued presence of uncertainty in all markets, to a lesser or greater extent, that permits money to serve as a store of wealth or value, that is to be used as a hedge against uncertainty.” Davidson’s notion of money is also related to uncertainty — it is in fact, as Dymski (1996, p. 85) claims, the “entry point” of Davidson’s postKeynesianism. According to Davidson (1978, p. 144), “It is only in a world of uncertainty and disappointment that money comes into its own.” Another post-Keynesian who has emphasized this point is Arestis. Claiming that uncertainty is “one of the central elements” of post-Keynesian theory, Arestis (1990, p. 228) argues that money becomes a link between the irreversible past and the unknown future, as Keynes argued in the General Theory (vii, p. 294). According to Arestis (1990, p. 229), “It is precisely the kind of uncertainty inherent in historical time that is both necessary and sufficient conditions for the existence of money.” This is also the interpretation Dutt and Amadeo (1990, p. 112) have of post-Keynesian theory: “The institution of money can be thought of as a response to uncertainty, as a way of postponing the making of actual decisions, without uncertainty there would be no need to hold money except for normal transactions purposes.” Linking the existence of money to uncertainty suffers from an obvious problem: as a stock, it undermines the role of banks in financing production through credit flows. Once production is financed through the extension of credit, and provided that supply of credit is demand-determined, then the analysis shifts to flows, and it becomes clear that money exists irrespective of uncertainty. As money is created, in part by the payment of wages, money appears even before households make their portfolio decisions under conditions of uncertainty. Granted, uncertainty will influence credit demand, but again, in this case, it influences credit, not money. Moreover, post-Keynesians seem to imply that the absence of uncertainty would reduce monetary economies to that of barter. In fact, this is not the case. Irrespective of uncertainty, that is even if entrepreneurs possessed full and perfect information, they would still have to have their production plans financed by bank credit, out of which wages would be disbursed. Money would still exist. 1t is post-Keynesians’ inability to emphasize the different roles of credit and money which leads to this confusion. After all, workers are not paid in kind, as in Ricardo’s corn model.”

Post-Keynesians and Orthodoxy

217

In the structuralist theory therefore, uncertainty in the context of monetary theory is discussed only in terms of money as a store of value, hence of money as an asset and a component of portfolio decisions of households, or

what Nell (1967, p. 386) has called the “theory of choice.” Accordingly, money becomes an asset among others, which enters households’ portfolio choices. The difference between money and other assets is that the value of money is more certain. According to Dow (1996, p. 37), “Liquidity preference thus follows from money’s central economic role as the asset whose value is most certain.” We are now able to link together the discussion of uncertainty as applied to money, with the demand for money as an asset. This structuralist position was nicely summarized by Dow (1996, p. 38) who believes, contrary to the discussion above, that increased information leads to

increased weight of an argument. This then translates into a fall in the demand for money as an asset: Weight is derived from the relative amount of relevant evidence brought to bear on the (generally unquantifiable) estimation of probability. The greater the amount of relevant evidence, the greater the degree of confidence with which expectations may be held as to the relative retums on altemative assets,

and thus the less the demand for money which eams no monetary retum. New evidence, therefore, which increases confidence in expectations thus reduces money’s liquidity relative to the expected retum on altemative assets, and may also reduce the discount on the expected retum on altemative assets, both encouraging a fall in the demand for money.

The meaning of the above reference is ambiguous. Is Dow in fact saying that as the stock of “relevant” information increases, uncertainty decreases? Or are post-Keynesians saying that increased information makes entrepreneurs more confident in their decisions, regardless of uncertainty? Either way, the argument is flawed. If the argument is the former, then strangely enough, this is also the position taken by early Walrasians, such as Clower and Leijonhufvud. As the stock of information increases, we tend toward a Walrasian model and the quantity of money tends toward zero, hence to a com economy. Post-Keynesians, however, claim that there can never be perfect certainty. Money will therefore always exist. If the argument is the latter, then as argued before, the weight of the argument is not derived from additional information. There is more. Conflating credit and money, and tying money to uncertainty raises a number of other issues. For instance, can bank credit be financed by additional saving, the result of increased uncertainty? Some post-Keynesians argue that investment is wholly financed by portfolio reshuffling. For instance, in a more extreme post-Keynesian position,

218

Credit, Money and Production

Knodell (1995, p. 271) argues that bank credit is wholly financed by portfolio reshuffling, with changes in credit having little impact on the quantity of money. Moreover, Knodell (1995) criticizes Wray (1995) for paying too much attention to flow analysis. Consider the following statement by Knodell (1995, p. 271; emphasis added): A vertical money supply world is not necessarily a commodity-money world; to the contrary, Keynes’s model is fully consistent with a credit-money world where the volume of credit-money depends on the volume of bank lending. None of the four shift factors of Keynes’s money demand schedule necessarily imply a change in the volume of bank lending; hence it would be incorrect, in a credit-money world, for changes in any of these factors [the transactions, precautionary, speculative and finance motives] to be met with changes in the quantity of credit money supplied by the economy.

This suggests that the finance motive is financed entirely through portfolio changes, and hence by pre-existing money. Once again, according to Knodell (1995, p. 271): “In Keynes’s world, income growth is financed through a stock adjustment and money balances are transferred from speculative use to transactions use." Cottrell (1986, p. 2) claims that “the money stock may be either causally active or passively endogenous, depending on the behavior of both private sector agents and the monetary authorities," thereby emphasizing the portfolio decisions of “private agents.” As seen in chapter two, this is equivalent to claiming that changes in the velocity of money are sources of investment financing, as in the early Minsky and Kaldor papers. Structuralists place the emphasis of uncertainty on households (and firms) — as opposed to banks. This tradition can be traced back to Knight (1921, p. 199) who argues that “if we are to understand the workings of the economic system, we must examine the meaning and significance of uncertainty; as well as how uncertainty relates to individual and firm behavior.” By doing so, however, post-Keynesians tend to undermine the importance of finance,

and the fact that banks also face an uncertain future. Firms can take all the relevant decisions pertaining to employment and production but if the finance is not forthcoming, all that amounts to very little. In this case, banks may have more pessimistic “animal spirits” than firms. Banks must not only make expectations regarding the future course of the economy, but also on how this unknown future will impact on firms’ profits. In other words, banks must estimate the impact of uncertainty on two levels. It is in this context that post-Keynesians have not sufficiently emphasized the uncertainty of commercial banks‘ decision-making. Where post-Keynesians have linked bank behavior with uncertainty, it is through liquidity preference and the composition of banks’ portfolios (Dow, 1996, Wray,

Post-Keynesians and Orthodoxy

219

1990, 1992). Yet, this is not uncertainty with respect to the future unknown behavior of markets. Uncertainty in itself does not explain why credit is endogenous, and neither does it begin to address the issue of reverse causality between the liability and asset sides of banks’ balance sheets. It is in this way that uncertainty is not a sufficient argument because in and by itself, uncertainty does not negate the causality between reserves and deposits, and loans. This can be done rather by emphasizing uncertainty and credit, and the role of banks as ex nihilo purveyors of credit. This does not mean that uncertainty is not a useful concept for the purposes of monetary theory, since undeniably, the stock of money will have some influence. But in terms of monetary theory, it is a secondary issue. The primary concern must be the creation of money through credit. As such, uncertainty will influence decisions in two ways: on firms in making their investment and production decisions, but also on banks in deciding how much and to whom credit will be expended. There is therefore a fundamental incoherence in post-Keynesian theory. If money is defined as a store of value and hence as a stock which is held for reasons of uncertainty, then we cannot say that credit creates money, which then renders incomprehensible the role of banks, unless they are interpreted

as mere financial intermediaries (which is how some post-Keynesians. have interpreted them). Here, however, there is another problem. In post-Keynesian theory, the issues of money and uncertainty are linked symbiotically. If money exists because there is uncertainty, the existence of money also reduces uncertainty by introducing stability in the economy. This is achieved mostly by entering into long-term money-wage contracts (see Lemer, 1952; Davidson, 1972; Wells, 1978). This symbiotic relationship, however, still does not address the issue of the creation of money. The reasoning appears to be circular: the more uncertainty there is, the more money is required, hence the more stability and less uncertainty.

THE STOCK OF MONEY The discussion over money as a store of value and uncertainty regarding the future leads naturally into a discussion of money as a stock. This is a consistent analysis, and post-Keynesians have emphasized the stock aspect of money, as opposed to its credit-flow aspect. For instance, Cottrell (1994) argues that it is the money stock which is causal in post-Keynesian theory. This is precisely Wray’s (1996, p. 440) conclusion as well: “In contrast [to circulationists], most Post Keynesians have emphasized money as a stock.”

220

Credit, Money and Production

In his post-General Theory articles, Keynes talks about the causal role of the “stock of money as a revolving fund taking care of a flow of credit." While a definite step in the right direction, Keynes's analysis still does not get us very far. Keynes’s emphasis was placed on the causal role of the stock of money. Hence, post-Keynesians have to first place the emphasis on the flow of credit, and then as a secondary issue, discuss the consequences of changing stocks of money (Sawyer, 1996; Lavoie, 1992).“ The notion of a stock of money ties in well with the use in post-Keynesian theory of demand and supply analysis, and of the IS-LM diagram — either with a shifting upward sloping LM curve (Davidson, 1972), or a horizontal schedule. But the use of supply and demand analysis within monetary theory is logically and methodologically derived from Quantity Theory — as was argued by Nell and Deleplace ( 1996). At first glance, the use of demand and supply curves in the context of a credit-economy may not appear to many post-Keynesians as a fallacy. Yet, for others, it is a definite proof of the “vestigial traces” of orthodoxy in postKeynesian thought. This reflects a reliance on the notion of money as an asset. As Nell (1996, p. 245) argues, Explaining the role of money in the economy by means of supply and demand has often seemed inadequate, even awkward. Money is evidently an institution — not just another good whose excess demand equation is to be added to those of the other goods, to fonn a monetized general equilibrium where before a barter system prevailed.

The very concept of a supply of a good implies some sort of scarcity. Moreover, according to Nell and Deleplace (1996, p. 25): Most post-Keynesian authors agree there is a supply curve of money, but hold that it is not vertical (hence the supply is not exogenous) and that it is unstable. The supply of money is seen as a function of the interest rate and the aggregate level of income, in parallel to the demand for money. In short, post-Keynesians accept the basic ideas of the LM curve, which means that post-Keynesian authors rely on a supply and demand framework to analyze money, the same framework used by the textbooks.

The debate is then reduced to discussing the slope of the money supply curve, and the response of the short-term rate of interest to changes in the demand for money, credit or even investment. Moreover, post-Keynesians argue that the response of the short rate to changes in the money supply also is diminished as close money substitutes become available. Hence, the money supply schedule is less inelastic. The upward sloping supply curve of money leads to other issues, for once we posit a positive sloped curve, then it is explicitly acknowledged that the

Post-Keynesians and Orthodoxy

221

money supply carries both endogenous and exogenous components. This view has been expressed by leading post-Keynesians. The emphasis post-Keynesians give to money as an asset, and its causal importance and significance, brings post-Keynesians back to the orthodox multiplier model. If money is an asset, then as households adjust their portfolios to changing economic conditions — for instance due to changes in the rate of interest — then the extra money will have a causal influence in determining the quantity of loans banks are allowed to make. This is for instance precisely the argument made by Knodell (1995, p. 270). Cottrell (1986, p. 17) also makes this argument: “The actual money stock is jointly determined by the authorities and private agents. The portfolio behavior of the latter is crucial in mediating between the authorities’ policy stance and the actual outcome for money stock.” But the problem with the emphasis on money as a store of value and on portfolio decisions of households is two-fold: first, money is not introduced per se out of the needs of production, but rather by household decisions. Second, if changes in portfolio go toward financing production, then it is a recognition that the liabilities of banks are causal in the determination of bank loans. Either way, neither argument is related to post-Keynesian

theory.

VELOCITY OF MONEY The above discussion leads into the discussion of the role of the velocity of money, and more particularly, about the causal role changes in the velocity of money play in much of the post-Keynesian discussion of money endogeneity theory. In this analysis, as was shown in chapters two and three, changes in the velocity of money brought about by households’ portfolio changes flood banks with additional funds with which they can extend loans. The problem at hand is that such increases in velocity and portfolio decisions to fund loans necessarily come from an existing pool of funds. Structuralists accept this argument. For instance, Pollin (1996, p. 499) claims that “clearly, the notion of an increase in the velocity of narrowly defined money is exactly equivalent to a supply increase in broadly defined credit money." This, as argued in chapter two, is equivalent to arguing in terms of a variable multiplier, where changes in the household demand for money balances, from illiquid to more liquid bank deposits, increase velocity. In Cottrell (1986), the emphasis on exogenous reserves leads the author to argue that adjustment in the stock of money following increased government expenditures is entirely done through household decisions, and “if velocity

222

Credit, Money and Production

does not have this specific value then the monetary expansion indicated above will be either ‘deficient’ or ‘excessive‘ in relation to the rise in real income and the assumed unit-elastic demand for money." According to Cottrell (1986, p. 23), “It is possible for the money stock to be greater or less than the aggregate of desired money balances.” Hence, Cottrell’s theory of money endogeneity is a result of portfolio adjustments.

RESERVES As argued in the introduction to this chapter, one of its purposes was to examine the more orthodox elements of the structuralist theory of money endogeneity and then to link it with the orthodox multiplier model. This statement is nowhere clearer than in the treatment of reserves and in their causal role in determining the supply of credit, and money. This statement must certainly come as a surprise to many post-Keynesians. In fact, postKeynesians, for instance Niggle (1991, p. 143), would associate the multiplier model with an exogenous money supply, and Palley (1994, p. 67) claims that the post-Keynesian theory of money is an attempt at “re-focusing attention away from the money multiplier.” Yet, some aspects of the post-Keynesian approach are compatible with this same multiplier model. This said, it should be understood, as suggested earlier, that the multiplier model is not simply a model where reserves and the money supply are exogenous. In fact, the model can easily incorporate endogenous reserves, as long as these reserves are then given some causal role in the determination of the availability of credit by banks. Moreover, as argued in chapter two (and as argued by many post-Keynesians), the multiplier model does lead back to a version of endogenous money (in the portfolio sense), although it is not compatible with the post-Keynesian revolutionary version. In this sense then, we are back to arguing that reserves determine loans, and hence the multiplier model.

Some structuralists fully embrace this view. For instance, Pollin (1996, p. 498) adopts a model which is not unlike the multiplier model, a model which he calls the “credit money multiplier.” Despite the small name change, it remains the orthodox model. The model (see Pollin, 1996. p. 497) is the following:

(l+n+f+c)

AM= A—-—-———i.B (rd+rnn+e+c)

Post-Keynesians and Orthodoxy

223

where n is the ratio of non-transaction deposit funds to transaction deposits; f is the ratio of non-deposit liabilities, such as money market mutual accounts, to transaction deposits; c is the ratio of the total currency held by

the non-intermediary public to transaction deposits; rd and r,, are respectively the reserve requirements on transaction and non-transaction deposits; and finally e is the ratio of excess reserves to transaction deposits. This approach is then linked with the rest of the structuralist theory of financial innovations. If central banks refuse to fully accommodate the

needs of banks, then banks will be forced to tum elsewhere for reserves, and introduce some financial innovations. Yet, “The reserve-generation process through liability management will not necessarily create an adequate supply of reserves to meet demand” (Pollin, 1996, p. 499). This is an argument made earlier by Minsky (1982, p. 171): “The institutional changes may not lead to a sufficient increase in financing ability to effect the same increase in financing as would have occurred if there had been no central bank constraint.” At this point, according to structuralists, reserves will then directly impact on the supply of credit. According to Pollin (1996, p. 498), “A liquidity shortage will then emerge: intermediaries may be forced to call in loans and sell assets to meet their reserve needs, and the extension of new loans will diminish.”

In this sense, reserves play a causal role in the determination of both the supply of money and the supply of available credit. This amounts, however, to a recognition that if reserves are not forthcoming — for whatever reason — then the lack of reserves will cause a fall in loans, thereby reestablishing the orthodox causality between reserves and credit. This in fact re-introduces scarcity into the post-Keynesian analysis of endogenous money. It is little wonder that in referring to “reserve-constrained” banks, Wray (1989, p. 154) defines them as “neoclassical.” Dymski’s banks are “consistent with the loanable funds in which depositors provide the funds which, through bank intermediation, enable investment." Moreover, Pollin (1996, p. 498) claims that “how much upward interest rate pressure will be exerted will depend on particular institutional configurations. The more liquid N and F can become through financial irmovation, the lower will be the interest rate differentials between these, D

and the sterile C.”'5 In Pollin’s model, changing values of the parameters of the multiplier will induce changes in the liability side of the commercial banks’ balance sheets thereby impacting on the supply of money. In turn, these liability changes will impact on banks’ assets. This is perhaps the best example of the current confusion in the structuralist literature. Of course, Pollin recognizes the causal role of reserves in determining deposits, yet, he also clearly

224

Credit, Money and Production

accepts the orthodox notion that excess deposits (reserves) will push the financing barrier outward, thereby allowing banks to extend their credit supply. This analysis of the determination of the supply of money can also be found elsewhere in the structuralist literature. For instance, in Cottrell’s (1986) analysis of central bank behavior, the initial shock is an exogenous increase in the supply of reserves, leading to increases in the ability of banks to extend new credit. As the author (1986, p. 8) argues, “The success of expansionary open market operations does not depend on an ex ante increase in money demand." Hence, his examination of a “series of possible cases of monetary expansion and restriction" begins with exogenous increases or decreases in reserves. In Cottrell (1988, p. 295). the author even argues that “the case that the stock of outstanding credit is purely demand determined is overstated" — referring even to “reserve-constrained banks” and to the “asset purchase/sale decisions of private agents play[ing] a role” in determining the quantity of credit supplied by banks. This is the portfolio model which is another way of presenting the multiplier model where bank liabilities determine the assets of the banks. Desai (1995, p. 112) is another non-orthodox author, in the postKeynesian tradition, who claims that the supply of reserves contains an exogenous element: “The monetary base is exogenous provided by the State.” Dymski also offers some passages which are rather confusing and ambiguous. For instance, in Dymski (1996, p. 383), the author argues that “the volume of liquidity available to finance the activities of deficit units varies inversely with surplus units’ demand for money as a store of value.” Moreover, Dymski (1996, p. 382) also argues that “expenditures by deficit units depends [sic] on whether they can secure credit or external financing from surplus units."'° Dymski (1996, p. 382) also argues that credit rationing occurs as a result of the scarcity of funds: “Some units seeking to undertake time-phased activities will be finance constrained — the volume of their planned expenditures will exceed their available finance.” The author attributes this to a “coordination” failure “because ex ante deficit units‘ demand for purchasing power exceeds the supply which ex ante surplus units possess.” What these passages suggest is that banks appear to be playing the role of financial intermediaries, redirecting the deposits of “surplus units" to bank borrowers. Hence, the causality between deposits and loans is reestablished according to orthodox tradition. Moreover, Dymski accepts the view that funds are scarce, and that irrespective of scarcity, ex ante saving is a source of financing of expenditures.

Post-Keynesians and Orthodoxy

225

In the structuralist approach, the monetary base, while perhaps endogenous, also has an exogenous, i.e. causal, influence on the quantity of loans being granted. And while the causation may run from loans to deposits, it is also correct to say that according to structuralists, there is a “feedback” effect from deposits to loans—or as Palley (1994, p. 82) argues, a “bivariate causality.” Pollin (1991) and Palley (1991) both claim that when banks fall short of their reserve needs, they must either call back existing loans or reduce the amount of new loans. This is therefore a tacit recognition that the quantity of reserves detennines the amount of loans which banks can grant. In this sense, reserves are seen as a source of “funds for purposes of lending” (Palley, 1991, p. 399). To develop this argument further, structuralists assume that in a situation of less than full accommodation, banks will - via liability management — attempt to lure depositors to place their money into those deposits carrying lower reserve requirements, such as CDs or other liabilities. Typically, these deposits are less liquid and offer a higher rate of interest. This suggests therefore that a lack of reserves leads to changes in the composition of bank liabilities (with varying reserve requirements). As a result, two conclusions can be drawn. First, it is the liability side of banks’ balance sheets which dictates the slope of the money supply curve (Lavoie,

1996). Second, we can argue that liability management changes the values of the money multiplier parameters thereby leading to “endogenous”

changes in the money supply. Hence, interpreting the central bank’s refusal to frilly accommodate as an exogenous policy decision, we obtain exactly the orthodox argument: causality runs from reserves (lack of reserves) to the money supply via endogenous changes in the money multiplier brought about by market forces. Hence structuralism and the money multiplier model become compatible. This can be seen in Pollin (1996, p. 497) where he makes use of the money multiplier model, and acknowledges that liability management will change the values of the parameters of the multiplier." Pollin (1995, pp. 16-7) argues that “of course, if depository institutions have no excess reserves, their lending capacity will be at its maximum, and can only increase if the total level of reserves itself increases through either government intervention or their own asset and liability management interactions with non-depository sources of funds." Also, implied in structuralist arguments is the notion of the scarcity of funds. There is the notion that there exists only a given pool of funds, out of which banks can lend. According to Pollin (1995, p. 17), the outstanding loans by depository institutions are not the only “loans” in the system, since all financial assets constitute loans of some sort. The reason for distinguishing [loans by depository institutions] is that the depository

226

Credit, Money and Production institutions - commercial banks and thrifts - are far more heavily regulated than other entities, through reserve requirements, capital adequacy requirements and general regulatory surveillance. Thus, as the proportion of private sector loans originating from depository institutions diminishes, the lending capacity of the system will rise.

This is the scarcity of funds. We would assume that the opposite also exists, that is as the loans of the depository institutions increase, the lending capacity of the system falls because of strict controls. Hence, given this reintroduction of the scarcity principle in post-Keynesian monetary analysis, the determination of the rate of interest by this demand and scarce supply of funds becomes problematic. '8 Moreover, according to structuralist post-Keynesians, a rise in economic activity (production or investment) will necessarily lead to a rise in the rate of interest. This is represented in their portrayal of the money supply curve as upward sloping (as well as the upward sloping LM curve). As such, as firms finance their capital outlays, the demand for money rises thereby leading to a rise in the rate of interest as in loanable funds theory. As such, we reinstate the notion of scarcity and crowding-out. This is specifically contrary to the arguments put forth by Keynes, as shown above. By positing a horizontal money supply curve, horizontalists at least avoid this trap altogether.

TI-IE MONEY MULTIPLIER MODEL Given the discussion above, it is now perhaps appropriate to question the relationship between the structuralist post-Keynesian approach to monetary theory and the orthodox multiplier model. A legitimate question to ask is whether, given the more orthodox elements of the structuralist position, post-Keynesian monetary theory is compatible with the multiplier model of the determination of the money supply. As stated on numerous occasions, post-Keynesians appear to be arguing simultaneously from both sides of the coin. Loans create deposits and reserves, but reserves also cause loans. By giving reserves such a dual role, post-Keynesians are in fact reinstating the money multiplier model by introducing into it a loans market. This does not prevent structuralists from adopting a theory of endogenous money. However, theirs is some sort of hybrid theory where both theories of money endogeneity are present. By emphasizing portfolio decisions, however, post-Keynesians risk reducing their theory of money to that of “portfolio endogeneity”.

Post-Keynesians and Orthodoxy

227

The nuances between the multiplier model and the structuralist model can best be depicted below by the following equations: M= m~B

where M is the money supply, m is the multiplier, and B is the supply of

high-powered money. In a simplified model, the multiplier is equal to [(1+c) / (c+r)]. Given traditions, conventions and the monetary autlrority’s control, the value of the multiplier is assumed constant. In this way, an exogenous supply of highpowered money will have a predictable and causal effect on the money supply, and thus prices. In Pollin (1996), the author adopts a model which is quite similar to the multiplier model. For him, reserves, while endogenous, exist in a specific relation to deposits, according to the following equation:

AB=(—--—-—il)D rd +r,,n+e+c where the values of the parameters of Pollin’s “credit-money” multiplier

vary and generate funds for investment purposes. The multiplier is, however, not taken as given. Its value will change according to a changing financial structure, in particular, financial innovations and portfolio reshuffling. As the value of the multiplier changes, this then creates opportunities for increased lending. This is exemplified in the following statement: “Pressures emerging endogenously within financial markets are the basic determinant of credit availability.”

(l+n+f+c) AM= A————i.B (rd +r,,n+e+c) Hence, endogenous changes in the term suucture of interest rates will generate shifts among assets (including money), and supply banks with the necessary funds to meet credit demand. In this context, as Pollin (1996, p. 448) argues, “Intennediaries will aim to increase M through increasing n and f and decreasing e and c." Credit is therefore not created ex nihilo, but rather by portfolio decisions and hence ex ante saving, since any monies shifted around have already been created before. According to Pollin (1996, p. 498), “Liability management means seeking to shift the values of n, f, c and e, specifically to raise n and f and lower c and e. [This will]

228

Credit, Money and Production

put upward pressure on interest rates the reward of higher interest yield will be necessary to induce asset-holders to shift holdings into less liquid forms.” The frequent use and acceptance of the money multiplier model by structuralists is attributed to the fact that they believe that it can successfully incorporate a theory of endogenous money, explained by changing values of the multiplier. This is obvious when considering Cottrell’s (1994, p. 597) interpretation of Keynes’s General Theory. According to Cottrell, while in the General Theory, money “as a first approximation” is exogenous, Keynes nonetheless made implicit the notion that money is endogenous through “the formula, money supply equals money multiplier times monetary base [because] it is recognized that the magnitude of the money multiplier is in part determined by the portfolio decisions of the private sector.” As argued in chapter two, however, this cannot be considered a theory of endogenous money since it not only undermines the role of banks, but also ignores the important link between production and credit, and credit and money. For structuralists therefore, the causality from reserves to loans is clearly established, and given particular attention in their attempt to argue that “loans must be financed.” For instance, Palley (1994, p. 79) suggests that “if there is an increase in loan demand, individual banks sell secondary reserves to fund additional lending. The modeling of bank asset and liability choices provides banks with an incentive to seek the cheapest sources of financing.” Hence, portfolio decisions by households (and banks) are a key in “financing loans.” Others speak of the “availability of credit” (Dow, 1996, p. 499). Moreover, Palley (I987-8, p. 281) reminds us that “only 1 - k of each dollar deposited may be lent,” a passage reminiscent of the orthodox multiplier model, where k is the reserve ratio. Structuralists must rely on the lack of reserves argument in establishing a maximum to the quantity of loans banks can make precisely because they have not developed a theory of banking behavior where uncertainty plays an important role and where credit constraints can be introduced without reference to a multiplier of any sort. The above discussion helps therefore to better make sense of Dickens’s (1990, pp. 1-3) argument that “both the monetarist claim and the postKeynesian retort presuppose a multiplier relationship between the supply of bank reserves and the stock of money. In short, the debate between the monetarists and the post-Keynesians comes down to a simple polarity.”

Post-Keynesians and Orthodoxy

229

CONCLUSION Overall, as argued in chapter five, post-Keynesians accept the reversed causalities between loans, reserves and deposits, and also accept the view that money is an endogenous feature of production economies. Yet, as was shown in this chapter, there remain some ambiguous elements in postKeynesian theory which not only beg clarification, but also repudiation. In these cases, it is difficult to see which side of the fence post-Keynesians lie. Money is endogenous, but the mechanics of this endogeneity are not always clear. While seeing credit as a central component of their analysis, they nonetheless see it as a component of money thereby undermining the role of banks in financing production and investment. In some respects, post-

Keyrresians argue, as was shown above, that production is financed “wholly” by portfolio decisions. This is not too surprising given their insistence that money arises out of uncertainty. Hence, an increase in uncertainty will deflate the attractiveness of securities and other assets. Households will therefore increase their demand for liquidity, swelling up bank deposits. This, at least according to Knodell, Dymski and Palley, can lead to increased lending by the banking sector.

The aim of this chapter was to show how the structuralist, [neo] postKeynesian, approach of endogenous money appears at times to be bordering on the orthodox multiplier model. This argument was primarily related — although not limited — to the bi-variate causality between reserves and loans, as advocated notably by Palley and Pollin. However, other aspects of the structuralist view are equally problematic, especially the treatment of uncertainty and how it is related to money, the causal role of velocity, and the notion of money as a stock of money and a store of value. One of the main arguments which can be drawn is that the reliance by many post-Keynesians on the lack of reserves assumption, and subsequently on the bi-causal role of reserves, is akin to the multiplier model and the orthodox view that reserves cause loans. This is also the conclusion reached by Lavoie (1992, p. 212) who claims that “some authors even go so far as to claim that, as a result of liability management by the banks, the expansion of monetary and credit aggregates is supply determined. In this extreme form, there is little difference between liability management and the monetarist story." In a sense then, perhaps the best way to differentiate between horizontalists and structuralists is to claim that for the former, endogenous money is always demand-determined, whereas for structuralists, it can also be supply-determined. The post-Keynesian lack-of-reserves hypothesis is a problematic element of their overall theory. And while Lavoie argues that it is analogous to the monetarist approach, other post-Keynesians have emphasized that banks can

230

Credit, Money and Production

never be constrained. For instance, in claiming that banks can never be “reserve constrained,” Wray (1989, p. 154) precisely associates the lack of reserves to the “neoclassical bank.” This is why Wray, as argued in chapter five, prefers emphasizing liquidity of preference rather than the lack of reserves in defending the upward-sloping money supply curve. Some post-Keynesians have emphasized the portfolio/velocity argument to explain endogeneity, as is the case of the early views of Minsky and Kaldor, and others. Yet, if this is the case, one of two possibilities arise: either money is not created by the needs of production, or if it is, then past savings are shuffled around to finance current investment. In either case, the conclusions can hardly be considered post-Keynesian. Also, if money is tied to the uncertain future, then by including credit in the demand for money, then it too must logically become a store of value. This makes little sense, and while post-Keynesians have not made this argument, it is a logical conclusion to their treatment of credit and money. Hence, post-Keynesians must either differentiate between money or credit and hence face importance implications - or accept the view that credit serves the same functions as money, such as being demanded because of an unknown future. This makes, however, little sense in a credit economy. The “holding of money” by entrepreneurs facing an uncertain future is not compatible with the role which credit plays. Whether the future is known or not is irrelevant. The entrepreneur will still demand credit to pursue his production plans. The uncertain future may influence, however, the amount of credit which he is demanding, but not the need to demand credit. It is in this sense that the demand for “money” may be nil; but it is never true that the demand for credit may be nil. In the end, it can be asked whether the structuralist theory of money endogeneity is reduced to the orthodox multiplier model. This is in fact Moore’s (199la, pp. 404-5; emphasis added) interpretation of the structuralist position: “The structural position makes less of a break with the mainstream view that central banks control monetary aggregates exogenously by varying the supply of reserves. As a result, although bank reserves can be to some degree at least exogenously controlled by the authorities, the supply of credit money is endogenous, due to endogenous variations in the money multiplier.”

NOTES l. 2.

I am grateful to Alain Parguez for this expression. Interestingly enough, in a recent article on the “Legacy of Keynes," Davidson (l996a) compares post— and New Keynesian in the area of price flexibility, unemployment and

Post-Keynesians and Orthodoxy

231

effective demand. Apart from the issue of the non-neutrality of money in the long-run, Davidson does not discuss money or monetary theory.

As Hahn (1965) reminds us, there is no place for money in neoclassical theory or general equilibrium models. It is true that in his later writings, Davidson has developed views which errrphasize the initial financing of investment through bank credit. Recall also Kaldor’s assessment that Keynes’s views were still “too monetarist.” Presumably, the argument is that under a gold or commodity-rrroney regime, the supply

is given by the physical stock. See also Dow (I997, p. 64) for a more recent argument: “As long as the supply of credit is not fully demand-determined, that is as long as the supply is independent of demand to any extent, then Keynes’s monetary theory retains its essentials." We should allow for different types of uncertainty. We should differentiate between “infomrational” (epistemological) uncertainty and “market” (ontological) uncertainty.

The former deals with behavior resulting from a lack of infomration where additional infomration will not eliminate uncertainty, although it may change the behavior and decisions of economic agents. In the latter, markets have a life of their own, and that in the aggregate, it is impossible to foresee how markets will behave.

As Lavoie (1992) reminds us, E.R. Weintraub (1975, p. 530) considers uncertainty in Keynes “an innovation of sublime importance.” Of course, uncertainty takes center stage in chapter twelve.

Whether Knight and Keynes held similar or different views with respect to uncertainty is ignored here. For a detailed discussion, see Davidson (1996) and Hoogduin (1987). Neal (1996, p. 400) calls “extreme” the view that probability distributions cannot be calculated. If there is no uncertainty, households may still also want to increase their wealth position by accumulating idle balances. At the end of the circuit, money would still exist. In fact, we can not even say that with no uncertainty, the demand for money would be nil. Not all post-Keynesians emphasize the stock element of money; see Wray (1990).. This is the standard Minskian argument of financial innovations. The problem here,

however, is that the theory of financial innovations is one of the very long run. According to Pollin (1996, p. 498), “if financial innovations can reduce the transaction costs associated with moving reserves between institutions with excess and deficient supplies — that is lowering e - this will reduce the general upward interest rate pressure associated with liability management.” Dymski (1996, pp. 395-6, n. 2) also argues that “credit (extemal finance) exists when one economic unit borrows other’s idle resources in exchange for a promise to repay." It appears that Dymski comes close to arguing that investment is financed by “idle” savings. These passages from Dymski make Wray’s (1989) comparison of Dymski’s model to the “loanable funds" approach understandable. Other structuralists make a similar argument. According to, Guttman (1990, p. 83; emphasis added), “Most credit-money is nowadays issued by private banks tuming their non-earning excess reserves into income-yielding loan assets.” In this context, excess

reserves create bank loans. Guttman (1990, p. 83) also argues in terms of a multiplier model: “Imposition of reserve requirements, discount loans for reserve-deficient banks, and trading of govemment securities in so called ‘open market operations‘

allow the

central bank to manipulate bank reserves and thereby determine the ability of private banks to create money.” For Hewitson (1995, p. 290), “lf private agents are not willing to go into debt, because of low profit expectations, or if banks are unwilling to expand their balance sheets, because they doubt that future business revenues will finance the payments due on debt, no amount of excess reserve holdings will raise the money supply."

7. New Keynesian Monetary Theory and the Transmission Mechanism: A Comparison with Post-Keynesian Theory This chapter will explore the monetary contributions of New Keynesian economics for two specific reasons. First, post— and New Keynesian theories should, at first glance, share in common Keynesian roots. This would imply that they should, to a greater or lesser degree, have some common interests. If the resemblances are only cosmetic, then this fact should also be brought

to light.‘

Second, many post-Keynesians have already suggested that there exist a number of similarities between post— and New Keynesian economics. Palley (1994, p. 68; 1996) even calls his own post-Keynesian approach the “mixed portfolio-loan demand approach,” and argues that it is “very much in the spirit of the earlier ‘New View’.” Whether post— and New Keynesians are intellectually and theoretically reconcilable depends on two important arguments. First, it depends on how each school is respectively interpreted, and second which arguments from each school are emphasized. Two points can be made, however. First, if post— and New Keynesians share common arguments, this cannot be very gratifying since, following the conclusions of the previous chapter, this must be a result of post-Keynesian structuralists’ inability to fully break away from orthodox thought. A reconciliation between post— and New Keynesian theories would imply sacrificing some of the key, and radical, insights of Keynes and postKeynesians. Second, both schools are, at some level, worlds apart with respect to their respective theoretical intent. While post-Keynesians intend, but do not always succeed, in pursuing Keynes’s revolution, New Keynesians do not hide the fact that this is of little consequential importance. The overall conclusion to this chapter is that New Keynesian theory is neither new, nor very Keynesian. On the one hand, their current model is

very similar to the one Minsky developed in his 1957 articles. On the other hand, there is no place for demand: the driving force of monetary policy is 232

New Keynesian Monetary Theory and the Transmission Mechanism

233

the supply of loanable funds. In Minsky (l957a, b), there is also no role for demand.

THE FOUNDATIONS OF NEW KEYNESIAN THEORY New Keynesian theory has not been properly surveyed by post-Keynesians. Post-Keynesian analysis of New Keynesian monetary economics is limited

to essentially two arguments. First, it addresses the relevance of asymmetric information and credit rationing; second, it raises the possibility of their reconciliation with post-Keynesian theory. For instance, Wolfson (1996, p. 39) has argued that “much of the edifice of the credit view has been built upon a specific assumption, that of

asymmetric infonnation." Similarly, Fazzari and Variato (1996, p. 360) also claim that “asyrrmietric information plays a key role in new Keynesian investment theory.” Crotty (1996, p. 348) believes that “perhaps the most important contribution New Keynesian theory can make to Post Keynesianism is to focus its attention on the theoretical implications of the informational and expectational asyrmnetries inherent in its assumption set

yet relatively unexplored by Post Keynesian theorists” (see also Dymski, 1988, 1993). Yet, while these notions are obviously important components of the New

Keynesian literature, they are nonetheless secondary elements. They are in fact not necessary to generate New Keynesian results. This position is shared by New Keynesians themselves. For instance, Kashyap and Stein (1994, pp. 222, 252) argue that the “lending view does not hinge critically on whether or not there is quantity rationing in the loans market. As a matter of practical reality, shifts in bank loan supply may well be accompanied by variations in the degree of rationing, but this is not necessary for there to be a meaningful lending channel. Quantity rationing in the loan market is not necessary for there to be a meaningful lending channel, although in practice such rationing is likely to be present to

some degree." Bemanke (1993, p. 56) also argues that “credit rationing is certainly consistent with the existence of a credit channel. However, credit rationing is not at all necessary for the credit channel to exist. All that is required for a credit channel is that bank credit and other fonns of credit be imperfect substitutes for borrowers." According to Gertler (1988, p. 570), “Basic insights from this [New Keynesian] literature need not be tied to particular forms of credit rationing.”

234

Credit, Money and Production

Among post-Keynesians, this was recognized only by van Ees and Garretsen (1993, p. 39) who claim that “in this respect the issue of quantity rationing and interest rate rigidity is only a side issue." This section aims at filling a void by presenting a comprehensive survey of New Keynesian monetary economics. It is hoped that following this exercise, post-Keynesians will be in a better position to evaluate New Keynesian monetary economics, and to draw from it the proper conclusions regarding a possible reconciliation. First, it may be useful to clearly state that while New Keynesian theory claims to counter some of the assumptions and conclusions of the New Classical approach, it nonetheless shares many of the New Classical hypotheses. For instance, despite the existence of asymmetric information, it is clear that New Keynesian theory relies heavily on the Walrasian framework. Markets essentially clear in the long run. In the short-run, however, some sub-sectors of the economy may not have full information thereby leading to some transitory effects in financial and labor markets.2 In the short-run, prices (wages, rate of interest or the prices of commodities) may be sticky, thereby preventing markets (due to imperfections) to adjust. New Keynesians have sought to give a micro foundational argument to price stickiness, first argued but then only assumed by their earlier (Old) Keynesian cousins. Moreover, New Keynesians do not hide their support for the notion of money neutrality, although this holds in the long run only. According to Mankiw (1992, p. 563), money is “neutral in the long run but not [in] the short run. This is exactly the position held by new Keynesians." While critical of New Classical theory, two claims are made. First, New Keynesian theory shares essentially the same frame of analysis. Second, proponents of the New Keynesian credit view do not see their approach as a repudiation of the more orthodox money view, but rather as an additional transmission mechanism. Both approaches, that is the money view and the credit channel, share the same initial shock: an exogenous fall in the supply of high-powered reserves by the central bank. From this point, however, their respective analyses diverge. Whereas the money view concentrates its attention on the liability (or money) side of the banks’ balance sheets, the credit view looks at what happens to the banks’ assets (or loans), although the latter group does not deny the influence of monetary policy on banks’ liabilities. New Keynesians do not claim that both approaches are “polar views" — as argued by Romer and Romer (1990, p. 150) — but rather that both channels can operate simultaneously. As Kashyap and Stein (1994, p. 252) have claimed, “The lending view need not imply that the more traditional money channel of policy transmission is inoperative; clearly, the two channels can

New Keynesian Monetary Theory and the Transmission Mechanism

235

co-exist and can complement each other.” This position is shared as well by Bemanke and Gertler (1995, p. 28), who claim that “we don’t think of the credit channel as a distinct, free-standing altemative to the traditional monetary transmission mechanism, but rather as a set of factors that amplify and propagate conventional interest rate effects. For this reason, the term ‘credit channel’ is something of a misnomer; the credit channel is an enhancement mechanism, not a truly independent or parallel channel.” In this sense, the credit channel “fills the gap“ in explaining the relationship and causality between changes in monetary policy and variations in output. New Keynesians have argued that the traditional money view is merely not sufficient to fully explain the relationship between central bank policy and movements in real output. Following a tightening of monetary policy, as interest rates rise, there is not an immediate fall in output — rather, output begins falling roughly after a slight lag. For New Keynesians, this raises questions regarding the legitimacy of the money view which predicts that changes in output are instantaneous. According to the money view, the effects of tightening monetary policy take place in two stages. In the first stage, the central bank pushes up market interest rates by reducing the supply of money. This stage begins

when the central bank exogenously sells govemment securities to the public in exchange for checks drawn on private banks in the economy. As the

central bank debits the reserve accounts of these banks, reserves in the banking system fall relative to deposits. If reserves fall below the central bank’s legal reserve requirements, the banking system as a whole must reduce its holding of deposits. As a result, the supply of money in the form of bank deposits falls. Moreover, because the demand for money has not "changed, market interest rates rise to allocate the smaller supply of money. Interest rates continue to rise until households are satisfied holding fewer deposits and more govemment securities. 3 In the second stage of the transmission mechanism, higher interest rates reduce spending in the economy. This effect is felt immediately. Business investment on plant and equipment declines in the face of higher borrowing costs. Consumer spending also falls, particularly spending on housing and durable goods, that households often buy on credit. According to the money view, the transmission mechanism ends here. Spending declines only because market interest rates rise, and market interest rates rise only because the central bank reduces its supply of deposits when it tightens policy. The monetary adjustments are done entirely through the banks’ liabilities, with little attention given to the asset side of their balance sheet. Hence, whether banks reduce lending after their deposits decline is irrelevant to the money view. If banks cut back on loans,

236

Credit, Money and Production

firms who borrowed from banks before the tightening of the money supply will instead borrow at similar terms in the bond market or commercial paper market. In other words, the money view assumes that firms can maintain their desired level of spending simply by switching from bank loans to other sources of credit. Bank credit is not “special.” Firms have access to perfect credit substitutes. The economy is “money-led.” It is essentially against the money view explanation of the transmission mechanism that New Keynesian economics begins, as a further explanation of the mechanisms inside the “black box” of monetary transmission. Bemanke and Gertler (1995, p. 34) offer three important reasons why the traditional interest rate channel may not hold. First, many New Keynesians have shown that even small rises in the rate of interest can lead to large decreases in economic activity. Second, the timing between changes in monetary policy and changes in output is inadequately explained by the money view. According to New Keynesians, while the effect on interest rates is transitory (it retums to trend within nine months), some components of GDP do not begin to change until after the interest rate retums to its trend value. For instance, according to Bemanke and Gertler (1995), inflation only begins to move twelve months after the initial monetary shock, while the larger part of the response of business fixed investment occurs between six and twenty-four months. Moreover, some studies (see Blinder and Maccini, 1991) have concluded that despite increases in the rate of interest, inventories rise initially. As Bemanke and Gertler (1995, p. 34) conclude, “The poor correspondence in timing between changes in interest rates and movements in some components of spending no doubt helps to explain why robust effects of interest rates on spending have been hard to pin down empirically." Finally, the composition of the spending effects is difficult to reconcile with the interest rate channel. According to New Keynesians, while monetary authorities control short-term interest rates, it would be expected that much of the impact would be felt on “short-lived” components of GDP. Yet, the component of “final demand" which responds the most to rises in interest rates is residential investment, while “structure investment” — another “long-lived" investment — is barely affected. For these reasons, New Keynesians argue that there must be additional explanations of the monetary transmission — it remains largely a “black box.” According to them, changes in the supply of credit, most often the result of exogenous changes in the supply of high-powered money, are a more accurate explanation of changes in output. It is the fall in the supply of credit, irrespective of credit rationing, which explains the transmission mechanism of monetary policy.

New Keynesian Monetary Theory and the Transmission Mechanism

237

While asymmetric information and credit rationing will be reviewed in this section, they play a less prominent role in New Keynesian theory than post-Keynesians have claimed. Their impact, although recognized, is not necessary to reach New Keynesian results. Accordingly, New Keynesian economics is seen through a different light, one which emphasizes the credit-driven nature of the economy, but also recognizes the “supplydetermined” nature of credit — as opposed to the “demand-determined" characteristic of credit in post-Keynesian economics. In this respect, money is credit-driven, but supply-determined. It is the first part of this statement which may lead some post-Keynesians to see possible common links between post— and New Keynesians. Following Bemanke and Blinder (1988), and Kashyap and Stein (1994), New Keynesian economics is defined according to the following three propositions: Pl. While the money supply is credit-driven, it remains supplydetermined, dictated largely by the policies of the central bank. Banks can only lend what they have at their disposal, either supplied by the deposits of the savers or the supply of high-powered money by the central bank. P2. Bank assets must be to some degree imperfect substitutes. This statement has two distinct dimensions and applies similarly to banks and their customers. On the one hand, bank customers must not be able to substitute bank credit for other sources of credit. On the other hand, banks

themselves must not be able to circumvent the central bank’s control over the supply of reserves. The overall result must be that a fall in reserves must lead to a fall in the supply of batik credit, and overall credit, and subsequently a fall in output. For instance, Bemanke and Blinder (1992, p. 901, n. l) have claimed that “an assumption of imperfect substitutability of loans for securities in bank portfolios is needed to ensure that a decline in reserves leads to a decline in loans.” Banks and bank loans are “special.” The Modigliani—Miller capital—structure invariance proposition must break down so that firms are unable to offset a decline in the supply of loans by borrowing directly from the household sector in public markets (commercial paper) or by turning to finance companies. As Morgan (I992, p. 32) says, “Banklending matter very much indeed. Bank loans are special.” P3. Prices are taken as sticky downward in the short run. Otherwise, if a change in nominal reserves initiated by the central bank is met by an equal change in prices, then banks’ and firms’ balance sheets will be left unaffected in real terms. In this case, there are no real effects of monetary policy through either the lending channel or the more orthodox money channel. Money is neutral only in the long run. These three propositions have been well summarized by Kashyap and Stein. According to the authors (1994, p. 223), “By changing the quantity

238

Credit, Money and Production

of reserves available to the banking system, the Federal Reserve must be able to affect the supply of intennediated loans. That is, the intermediary sector as a whole must not be able to completely insulate its lending activities from shocks to reserves, either by switching from deposits to less reserve-intensive forms of finance (for example, certificates of deposits [CDs], commercial paper, equity, etc.) or by paring its net holdings of bonds." This suggests that in addition to its traditional interest rate channel, monetary policy can affect aggregate activity independently through its effect on the supply of loans. Having made these claims, let us now tum to more specific issues in New Keynesian theory. The next sections will cover some of the traditional elements of New Keynesian economics, such as credit rationing and asymmetric information, since although secondary, these issues nonetheless play an important role. Once this is done, we will turn to the more important elements of New Keynesian economics: the importance of the balance sheet channel, the bank lending channel, and the impact of liability and asset management — financial innovations — on the bank lending channel. Only then will we be in a position to compare New Keynesian economics with post-Keynesian theory as it is today, and with the early views of Minsky, Kaldor and Davidson.

A BRIEF HISTORY OF CREDIT RATIONING The original underlying theoretical construct of the New Keynesian school lies in the principle that commercial banks typically face an excess demand for credit at a given short-run rate of interest. As a result, lenders must ration the supply of credit available to them at any point in time (Morgan, 1990). This suggests that credit rationing affects investment independently of the level of the rate of interest. The subject of credit rationing was the focus of considerable attention in New Keynesian economics and remains an important issue still to this day. It is rooted in the theoretical analysis of Gurley and Shaw who recognized at an early stage the importance of credit in the monetary transmission mechanism, but also in Akerlof's famous “lemons” problem. The credit — or lending — view can be traced back to the early 1950s in an interesting paper by Karecken (1957) who emphasized that monetary policy may operate through credit channels rather than directly through interest rates. These early works on credit rationing rested, however, on “Keynesian” price rigidities, rather than asymmetric information. Later, Jaffee and Russell (1976) and Stiglitz and Weiss (1991) argued that credit rationing may occur in equilibrium given asymmetry in the possession

New Keynesian Monetary Theory and the Transmission Mechanism

239

of knowledge and infonnation, as opposed to sticky prices. In particular, Jaffee and Russell (1976) sought to apply Akerlofs “lemons” problem where asymmetric information between buyers and sellers with respect to product quality can cause markets to malfunction. Credit rationing arises, as described by Jaffee and Stiglitz (1990, p. 849) in “instances in which some individuals obtain loans, while apparently identical individuals, who are willing to borrow at precisely the same terms do not.” Bester (1985, p. 850) has the same definition: “Credit rationing is said to occur when some borrowers receive a loan and others do not.“ But there is also another dimension to credit rationing which is that loan applicants who see their credit demand rejected would not be able to secure a loan even if they offered to pay a higher rate of interest to compensate for their perceived higher risk. According to Stiglitz and Weiss (1991, p. 249), credit rationing is in these terms: “Among loan applicants who appear to be identical, some receive a loan and others do not, and the rejected applicants would not receive a loan even if they offered to pay a higher interest rate. There are identifiable groups of individuals in the population who, with a given supply of credit, are unable to obtain loans at any interest rate, even

though with a larger supply of credit, they would.” Typically, therefore, at an interest rate, r*, banks will face a demand for

bank credit which exceeds the supply of credit, dictated by the central bank’s supply of high-powered money. As Stiglitz and Weiss (1991, p. 393) claim, there is at a given rate of interest an “excess demand for loanable funds." This is still, however, an equilibrium, profit-maximization. This suggests that in the New Keynesian model, the loan rate of interest charged by banks is not determined by supply and demand conditions, but rather by profit maximizing conditions of banks, subject to the availability of the supply of loanable funds. As Stiglitz and Weiss (1991, p. 249) claim, there are no competitive forces forcing demand to equal supply. Credit is rationed. According to Greenwald and Stiglitz (1993, p. 31), “The interest rate charged [is] that which maximizes the expected retum to lenders, and at that interest rate, there is an excess demand for credit.” Moreover, Mankiw (1986) has shown that a small rise in the rate of interest can lead to large decreases in lending, possibly even to a collapse. In other words, the quantity or supply of bank lending is very sensitive to exogenous disturbances. The situation can be depicted by the following figure.

240

Credit, Money and Production

r/\ M‘

H

r9

....................................................


Profits

Figure 7.1: The New Keynesian credit rationing model The supply of “money” function in figure 7.1 is drawn as backward sloping at higher rates because as rates rise, the average borrower quality falls since good borrowers drop out. As a result, the expected profitability of the banks also diminishes. According to Gertler (1988, p. 569), “After a point, further increases in the interest rate may lower lenders’ expected retums, making the loan supply curve bend backward.” The rate, r* is the rate which maximizes profits, while r is the rate which would prevail if the “money market” was allowed to work.‘ The market does not clear (Stiglitz and Weiss, 1991). As the authors (1991, p. 248) claim, Both the demand for loans and the supply of funds are functions of the interest rate. ...Clearly, it is conceivable that at [r‘] the demand for funds exceeds the supply of funds. Traditional analysis would argue that, in the presence of an excess demand for loans, unsatisfied borrowers would offer to pay a higher interest rate to the bank, bidding up the interest rate until demand equals supply. But although supply does not equal demand at [r*], it is the equilibrium interest rate! There are no competitive forces leading supply to equal demand, and credit is rationed.

As a result, “It may not be profitable to raise the interest rate or collateral requirements when a bank has an excess demand for credit.” According to Gertler (1988, p. 569), at a given interest rate, lenders eam less because “an

New Keynesian Monetary Theory and the Transmission Mechanism

241

unobserved spread in a borrower’s project return distribution reduces the expected payment to lenders under default.”

MORAL HAZARD AND ADVERSE SELECTION Moral hazard occurs when changes in the rate of interest influence borrowers’ behavior, such that borrowers may choose to pursue riskier projects as the rate of interest rises. As the demand for credit increases, banks become reluctant to increase the rate of interest since changes in the rate of interest will impact on the behavior of bank borrowers in ways which are not always compatible with the overall macro, profit-maximization objective of banks. I-Ience, in New Keynesian economics, a rise in the rate of interest may increase the demand for credit. Adverse selection (or adverse incentive) occurs when increases in the rate of interest affect the overall riskiness of the applicant pool, either by encouraging riskier applicants to demand credit, or by making existing projects riskier at higher rates. This argument is made by Bemanke (I981,

p. 157): “If borrowers have better knowledge of their own default risk than the bank has, then the familiar adverse selection mechanism makes the

quality of the applicant pool dependent on the loan terms offered.” Banks therefore have a profit incentive motive to attempt to identify those bonowers who have a higher probability of paying back their bank loans. According to Jaffee and Stiglitz (1990, p. 857), “The adverse selection effect can easily outweigh the direct effect. There is a critical interest rate, r*, at which the safer borrowers stop applying. Thus, the retum to the bank declines precipitously at r*." Banks would be happy to charge the higher rate of interest to the “bad” borrowers, but because of the problem of asymmetrical information, the banks cannot differentiate the potentially “good” borrowers from potentially “bad" ones. Ex ante, borrowers are indistinguishable. As a result, banks are forced to ration the pool of credit, and r* does not adjust to r.’ While higher loan rates increase the lender’s expected retums for any given project, higher rates may create moral hazard and adverse selection effects that reduce the lender’s expected retums for all borrowers.

ASYMMETRIC INFORMATION Underlying the credit-rationing literature, as stated above, is the fact that there exists an asymmetrical relationship between commercial banks and potential credit borrowers. This is manifested primarily by the existence of

242

Credit, Money and Production

differences in the amount of knowledge banks and borrowers possess. Borrowers will typically not divulge — i.e. volunteer — information to banks which they believe may hamper their probability of obtaining a loan. This point was made by Stiglitz and Weiss (I991, p. 249) who claim that each investment carries an a priori probability distribution of returns known to the borrower, but not the bank. Hence, “The banks cannot ascertain the riskiness of a project." At the initial stage of a loan application, banks must carefully examine the proposed projects, evaluate them, and then decide which receive funding. However, banks cannot differentiate those projects which are “good” from those which are “bad" as to the probability of the investment project failing. All projects are identical ex ante (Bemanke and Gertler, I990). According to Stiglitz and Weiss (1991, p. 249), “The bank cannot ascertain the riskiness of a project.” Since borrowers are “observationally identical”

(Stiglitz and Weiss, I992, p. 695), the theory of credit rationing relies therefore on the “representative agent." Asymmetric information, defined as discrepancies between probability distributions over uncertain project outcomes, is therefore seen as a “flaw” and an “imperfection” which once removed would reinstall the Walrasian system. It inserts a “wedge” between the costs of funds raised extemally and the opportunity cost of using internal funds (Bemanke and Gertler, I995, p. 35). This is called the “external finance premium,” and relates to the costs to the institution of evaluating, monitoring and collecting information. The size of this wedge will vary according to the borrower’s creditworthiness, but also with monetary policy — as will be shown below, and will compensate the lender for these additional costs (see Gale and Hellwig, 1985; Williamson, I987; and Bemanke and Gertler, 1989). Because of asymmetric information, markets are characterized by under lending, since lending is sensitive to exogenous disturbances (Gertler, 1988, p. 570). Similarly, credit can be misallocated, since banks cannot, ex ante, differentiate between the good and the bad borrowers. In this sense, some portion of the supply of credit will be awarded to borrowers who would be considered, if banks had the appropriate information, bad borrowers. Overall, then, the performance of the economy does not only rely on credit, but also on the distribution of the stock of credit-funds. Hence, the level of investment is a function not only of the level of the rate of interest, but also of the availability of bank credit.“ 7

New Keynesian Monetary Theory and the Transmission Mechanism

243

SCREENING MECHANISMS Reliance on issues of moral hazard and adverse selection as an explanation of credit rationing has been criticized from within the New Keynesian camp as being unnecessary to the rationing literature. Some New Keynesians have argued that the use by banks of screening devices helps in forcing potential bank borrowers to reveal their true intentions, and hence eliminates the need for banks to ration credit. Screening devices thus become an important way for banks to circumvent the problem of asymmetrical information.

For instance, banks may impose on different borrowers a premium over and above the lending rate as a way to screen out the bad borrowers. The argument suggests that only bad borrowers would agree to take out a loan at a higher rate of interest, since higher interest rates, as suggested above, induce borrowers to choose riskier projects, and thus projects with a lower probability of repayment by lowering the profitability of the projects. According to Stiglitz and Weiss (1981, p. 393), bad borrowers “are willing to borrow at high interest rates because they perceive their probability of repaying the loan to below.” Bester (1985) - see also Bemanke and Gertler (1989, 1990), and Besanko and Thakor (1987) - has argued that changing collateral requirements could in a large measure reduce the need for credit rationing. According to Bester (1985), if lenders are free to use different interest rates and collateral requirements across borrowers, then there is no need for credit rationing in equilibrium. In essence, according to Bester (1985), good borrowers would be willing to accept higher collateral conditions as long as they are accompanied by a reduction in the rate of interest. The money — or loan — market clears in the Walrasian sense. The combination of varying rates of interest and collateral conditions implies that banks are able to apply effective screening mechanisms in such a way that they will be able to separate good from bad borrowers. Hence, Bester (1985) argues that there are in fact two separate influences operating simultaneously on borrowers, and hence the need for screening devices based both on rates of interest and collateral requirements. As Bester (1985, p. 850) explains, For a given collateral, an increase in the rate of interest causes adverse selection, since only borrowers with riskier investments will apply for a loan at a higher interest rate. Similarly, higher interest payments create an incentive for investors to choose projects with a higher probability of bankruptcy.

244

Credit, Money and Production

Moreover, according to Bester (1985, p. 850), “Equilibrium is characterized by separation of borrowers of different risk. Borrowers with high probability of default choose a contract with a higher interest rate and a lower collateral than borrowers with low probability of default.” In response to this argument, three points have been raised. First, as pointed out by Calomiris and Hubbard (1990, p. 92), borrowers in Bester’s model have an unlimited access to collateral — an observation also acknowledged by Bester (1985, pp. 854-5) who says that “only partial screening may be possible and adverse selection could still arise.” Second, and perhaps more importantly, some New Keynesians maintain that variations in collateral requirements would not help in reducing problems of moral hazard and adverse selection, once wealth considerations are taken into account. For instance, in an interesting rebuttal, Stiglitz and Weiss (1992, p. 694) claim that “collateral and other non-price rationing devices would not eliminate the possibility of credit rationing [since] increasing collateral requirements may adversely affect the mix of applicants. Wealthier individuals would both be willing to put up more collateral and would undertake riskier projects than would less wealthy individuals." Third, some have argued that increasing collateral would also jeopardize bank profits by encouraging good borrowers to drop out. According to Stiglitz and Weiss (I991), increasing collateral requirements may result in lowering bank profits since more risk-adverse borrowers (presumably, “good” borrowers) would simply drop out of the market. Since then, Wette (1983) has shown that this also applies to risk-neutral borrowers.

BALANCE SHEET AND BANK SHEET CHANNELS Having laid out the foundational arguments of New Keynesian monetary economics with respect to credit rationing, asymmetric infonnation, moral hazard, adverse selection, and screening mechanisms, it is now possible to examine recent debates in the New Keynesian literature regarding the precise mechanisms through which these issues arise. This will show also in what respect the above arguments are not necessary to generate New Keynesian results. They are interesting, but secondary, arguments. What does matter are the factors which determine or influence the supply of bank credit, and subsequently, the effects of a shortage of credit by banks. New Keynesian monetary theory does not stand or fall with credit rationing, as claimed by many post-Keynesians. As said earlier, the New Keynesian theory can best be summarized by claiming that the money supply is “credit-driven but supply-detemiined.”

New Keynesian Monetary Theory and the Transmission Mechanism

245

What is of primary importance are the three propositions discussed earlier. From these propositions, New Keynesian economics can be properly surveyed by examining two transmission mechanisms of monetary theory: the balance sheet and bank lending channels. The Balance Sheet Channel The first of the transmission channels of the lending or credit view is the balance sheet channel. In what follows, post-Keynesians should recognize some interesting similarities with Minsky’s early (and later) views on financial fragility. According to the money view, banks raise the rate of interest on loans only sufficiently to cover the increase in their cost of funds resulting from higher market (base) interest rates. Because of the perfect substitutability between loans and other types of credit, if banks raised their loan rate above the “equilibrium” rate, firms would quickly switch to other credit markets. New Keynesians, however, see things differently since they assume imperfect substitutability such that banks can raise their rate above the equilibrium rate. New Keynesians see this in terms of a non-perfectly competitive market. According to Ng and Schaller (1996, p. 375), “Finance constraints induced by capital market and information imperfections raise the extemal cost of borrowing above the risk-free market interest rate. This may

provide an additional transmission mechanism for monetary policy which is not present under perfect capital market.” The balance sheet channel claims that monetary policy, through its high interest rate policy, will influence the interest rate on loans, over and above what is necessary to cover the costs of searching for funds. This point was well put by Gertler and Gilchrist (I993, p. 49), claiming that “regardless of whether there is rationing, credit market imperfections distort the real investment decisions, since the premium for extemal funds affect the overall price of funds that the borrower faces.“ Hence, the premium on extemal finance, defined as a “wedge” between the cost of raising funds extemally versus that of using retained earnings, rises endogenously. The extemal finance premium includes, among other things, the costs to banks of collecting, evaluating and monitoring information. Although the extemal finance premium may be affected by asymmetric information, the former operates irrespective of the latter. The rate on bank loans increases more than in the money view of the monetary transmission mechanism for another reason. The rise of the extemal finance premium raises the debt of firms thereby making them less

246

creditworthy.

Credit, Money and Production

As a result, banks will impose a “risk premium” to

compensate. This last statement requires some explanation.

Shifts in monetary policy, i.e. exogenous decreases in the supply of money raising the rate of interest, impact on the financial condition of business firms, directly and indirectly, through the net worth or balance sheets of firms. Fluctuations in a borrower’s balance sheet or financial position will affect the firm’s investment and production decisions, and its ability to secure new or on-going credit, thereby impacting on the economic cycle. As Bemanke and Gertler (1995, p. 35) recently state, “Endogenous procyclical movements in borrower balance sheets can amplify and propagate business cycles" — a phenomenon New Keynesians refer to as the “financial accelerator” (see Bemanke, Gertler and Gilchrist, 1996). And as emphasized by Bernanke and Gertler (1995), there is no reason to believe that this channel has become less important in recent years. The balance sheet channel operates directly and indirectly on credit. The direct way in which the balance sheet channel operates can be divided into two different transmission mechanisms. First, if borrowers have outstanding short-term debt, or even if they have a floating-rate debt, then a rise in the rate of interest following a restrictive monetary policy, will raise the interest payments on their debt, thereby lowering the firms’ cash flow and weakening the borrowers‘ financial position. As the loan rate of interest increases therefore, default risk increases as well because higher loan rates increase a borrower’s total burden of debt, which, all else equal, increases the likelihood that the borrower’s profits will not cover its debt payments.“ To compensate for the higher default risk, banks must raise loan rates more than is required simply to cover the higher cost of funds.9 The second direct way in which the balance sheet channel operates is through asset prices and the collateral of firms. A rise in interest rates will also lower asset prices, which will then contribute in lowering the value of borrowers’ collateral (see Borio, Kennedy and Prowse, 1994). The balance sheet channel operates also indirectly through its effects on capital-goods firms. After the implementation of a restrictive monetary policy, the rise in the rate of interest will discourage investment spending by non-capital goods producing firms.’° This will translate into lower net worth for capital-goods firms since their revenues would have fallen with their costs remaining the same in the short-run. A fall in collateral once again implies a rise in the loan rate to compensate for the increased riskiness of the loan. In all cases, firms have less collateral for their loans. As a result, they will be less likely to secure a loan (or to renew an existing one. Bemanke and Gertler (1989), for instance, have shown that investment is positively linked to the firm’s balance sheet position, defined as the ratio of net worth to liabilities. If the ratio is high, then there are more resources to

New Keynesian Monetary Theory and the Transmission Mechanism

247

use either as intemal funds or as collateral to obtain outside funds. Hence, the market equilibrium level of investment will vary positively with the borrowers’ balance sheet position, as the net worth of firms increases, and negatively with the rate of interest. As net worth increases, firms have more retained earnings at their disposal, and can hence rely less on extemal bank loans (Calomiris and Hubbard, 1990; Fazzari, Hubbard and Peterson,

1988)."

In their contribution to the debate, Bemanke and Gertler (1995, p. 37) have suggested a way of measuring this loss of net worth, by computing what they call the “coverage ratio”: the ratio between the interest payments over the interest payments plus profits. According to the authors (I995, p. 37), “Increases in the funds rate translate almost immediately into increases in the coverage ratio, and hence, ultimately, into weaker balance sheet positions." Increases in the rate of interest will impact on both components of the coverage ratio: it increases interest payments and also reduces profits by impacting on costs and revenues.” To this literature, Gertler and Gilchrist (I993, 1994) have found that Bemanke and Gertler’s results are different for small and large firms since the former generally have more limited access to short-term credit.

On a related issue, as the central bank raises the base rate of interest following a restrictive monetary policy, the spread between the loan rate of interest and the base rate may also rise. Irrespective of this, however,,is the fact that the spread between the loan rate and the base rate is a measurement

of the borrowers’ financial position. This spread will typically vary along with changes in monetary policy. The higher the borrowers’ net worth - liquid assets and marketable collateral, for instance — the lower the external finance premium. Hence, changes in monetary policy affect borrowers’ net worth, thereby impacting also to a large degree access to credit. This will in tum impact on investment and production (spending) decisions. This is the same argument put forth by post-Keynesians, both horizontalists and structuralists, with respect to Kalecki’s theory of increasing risk. New Keynesians see this phenomenon — the financial accelerator — as the source of the business cycle. In other words, as carefully explained by Bemanke and Gertler (1995, p. 35), “Endogenous pro-cyclical movements in borrower balance sheets can amplify and propagate business cycles." Monetary policy can therefore affect the growth cycle independently of credit rationing. The New Keynesian model does not stand or fall with credit rationing. Increases in the rate of interest will endogenously affect the net worth of firms, and hence the loan premium (see Bemanke and Gertler, 1989, 1990; Calomiris and Hubbard, I990; and Gertler, 1992).

248

Credit, Money and Production

The Bank Lending Channel Beyond the impact a restrictive monetary policy may have on the balance sheets of firms, firms may also face a shift in the supply of intermediated credit, in particular loans by commercial banks. This suggests that the central bank is able to control the availability of loanable funds. As the supply falls, firms who depend on bank credit may see their access to such credit eliminated, with the obvious impact on production and investment — unless they are able to find other sources of credit-funds, either through commercial paper or through nonbank intermediaries. Either way, such moves will likely translate into higher costs in establishing new relations with other credit suppliers. The message here is that the economy is credit~ driven, but that loanable funds are supply-determined by the central bank. The central argument of the bank lending channel therefore is that a fall in the supply of credit will translate itself into lower economic activity and production, over and above the traditional interest rate, money channel. This is the result of firms adjusting or abandoning their investment and production decisions. An important element in this analysis is that other forms of credit are imperfect substitutes for bank loans. Otherwise, a fall in credit would have little impact on economic activity. Credit controls can originate from either a central bank decision to restrict reserves or by administered (political) controls. Either way, a fall in supply will impact on loans (see Schredt, I990; Morgan, I990, 1992). The bank lending channel is perhaps the more orthodox of the two transmission mechanisms, and affects the supply of loans by commercial banks. It assumes that banks cannot replace lost deposits with other sources of loanable funds. Today, however, they argue that since regulation Q was removed, reserve requirements are less of a burden, and markets for bank liabilities are now more liquid and more developed. However, as argued by Kashyap and Stein (I994), the existence of a bank lending channel does not require banks to be totally incapable of replacing bank deposits. It is sufficient that banks do not face a perfectly elastic demand for their open-market liabilities, so that an open-market sale by the Federal Reserve also increases the banks‘ cost of funds, which will then shift the supply of loans inward, squeezing out bank-dependent borrowers and raising the external finance premium.

New Keynesian Monetary Theory and the Transmission Mechanism

249

NEW KEYNESIANS AND THE MONEY MULTIPLIER MODEL As was done in the previous chapter regarding the structuralist postKeynesians, an important issue which must be addressed is the general framework within which New Keynesians operate. It is contended here that as in the case of the structuralists, New Keynesians also make extensive use of the orthodox multiplier model. However, unlike post-Keynesians, New Keynesians do not recognize the endogenous nature of reserves — or if they do, it is not a permanent or institutional feature of the system. Rather, it is a result of the “leaning against the wind” policy of the central bank. As Feldstein and Stock (1994) argue, it is merely an indication of “what the Fed can do” — which they quickly criticize as ineffective and “eclectic judgmentalism.” In their model, there are only exogenous reserves operating on loans and thus the money supply. This is in fact consistent with one of the three characteristics of New Keynesian theory described earlier. New Keynesians readily embrace the exogeneity of high-powered money — on this there is complete consensus. Nonetheless establishing the New Keynesian model within the orthodox money multiplier model is important for two reasons. First, it will allow for a more proper discussion of the methodological framework within which takes place the discussion of credit supply and of credit rationing. Second, it will also allow for a better comparison between New Keynesian theory and the post-Keynesian altemative. In the New Keynesian literature, loans are rationed because banks cannot lend more than their deposits and reserves. Moreover, Gertler and Gilchrist (1993, p. 47) make clear that banks need to ration credit given the central bank’s ability to directly control the supply of bank credit via reserves. Hence, as Jaffee (1987, p. 720) argues, there exists at most times an “excess demand for loans and thereby the need for credit rationing” of the “available supply of funds.” Credit supply is scarce, and must be carefully rationed. The causalities between deposits and loans, and money and income are preserved. New Keynesians see the central bank as having “considerable leverage” (Blinder, 1987, p. 330) over the availability of bank funds for credit. In this context, deposits are “generated endogenously from the available reserves” (Blinder, 1987, p. 330). Deposits determine the stock of loanable funds. As economic activity expands, higher transaction balances are required so bank deposits rise. As funds flow into the banking system, the supply of bank credit expands further (Blinder, 1987, p. 328). As Bemanke and Blinder (1992, p. 901) have expressed, “When the Federal Reserve reduces the volume of reserves, and therefore of loans, spending by customers who depend on bank credit must fall, and therefore

250

Credit, Money and Production

so must aggregate demand” (see also Bernanke and Blinder, I988; and Blinder and Stiglitz, 1983). The central bank holds a key role in influencing the “availability of funds”, as argued by Blinder (1987, p. 33): “In a system of fractional reserve banking the central bank has considerable leverage over the latter. For the banking system as a whole, however, the supply of deposits is generated endogenously from the available reserves and required reserve ratio. Thus for the system as a whole, reserves not deposits, are the binding constraints.” Similarly, Fuerst (1992, p. 4) claims that “the loanable funds available to intermediaries consists of previously deposited balances by savers, and the newly injected currency.” Mishkin (1995, p. 7) describes his model as starting first with an exogenous decline in the high-powered money which translates into a fall in bank deposits, followed by a decrease in bank loans, investment and finally income. Kahn (I991, p. 25; emphasis added) explains how “banks largely determine how they will allocate an increase in deposits between loans and securities."’3 Similarly, Kashyap et al. (I993, p. 82) clearly state that the supply of money is exogenous. Bernanke and Gertler (1995, p. 29) acknowledge that the central bank can influence the “supply of loans by depository institutions.” Romer and Romer (1990, p. I50) argue precisely the fact that “a reduction in the stock of reserves will necessarily reduce the quantity of such loans. Competition among banks for the scarce reserves needed to make the loans will then bid up interest rates paid by banks to depositors." The multiplier principle is made amply clear by Gertler and Gilchrist (1993, pp. 45-6) in the following, albeit long, passage: According to the [New Keynesian] credit view, monetary policy works at least in part by altering the flow of bank credit. An important step in the argument is the contention that legal reserve requirements on deposits provide the Federal Reserve with considerable direct leverage over the quantity of funds that banks may obtain. An open market sale reduces the real quantity of deposits banks can issue. This in tum induces banks to contract lending, which ultimately constrains the spending of borrowers who rely primarily on bank credit. It is important to reiterate that two distinct hypotheses underlie the credit view: first, that bank credit is special; second, that monetary policy directly constrains bank lending.

In a similar way, Mankiw (1992, p. 562) has argued that increased bank deposits through a fall of consumption and increased saving will increase the commercial banks‘ ability to extend credit, and hence create additional investment: “Almost all economists now believe that additional saving will,

New Keynesian Monetary Theory and the Transmission Mechanism

251

in the long run, lead to additional investment rather than inadequate aggregate demand.” Recalling the discussion in chapter two conceming the two versions of endogenous money, the New Keynesian theory of money can be associated with the “portfolio endogeneity" version which, as argued before, cannot be associated to a theory of endogenous money in the post-Keynesian, i.e. revolutionary, sense. In this context, money becomes “endogenous” due to changes in the value of the multiplier — either through liability or asset management, or by households’ decisions to hold fewer securities and more demand deposits. Credit is financed from an existing stock of saving, or from a stock of reserves. When New Keynesians speak of endogenous money, it is in the portfolio sense. For instance, the later Tobin (1991, p. 223) talks openly about endogenous money: “Ml, M2, M3, or $GNP are endogenous in the sense that the paths of their values depend not only on operations the Federal Reserve does control but also on the responses and behavior of banks, other financial market participants, and multitudes of other agents throughout the world.” This view can be traced back to Burger (1971, p. I; emphasis added) who

then argued that “the money stock is no longer considered to be completely exogenously determined, but is viewed as a quantity whose magnitude is partly determined within the economic system by rational portfolio decisions ofthe commercial banks and the public.""’ This view is also offered by Blinder (I987, p. 328): “As economic activity

expands, higher transactions balances are required; so bank deposits rise. As funds flow into the banking system, the supply of bank credit [through portfolio reshuffling] expands further." As was seen in the last chapter, this is also the view of many neo-post-Keynesians, especially Knodell (I995). Bemanke and Gertler (1995) have admitted that New Keynesians question neither the loanable funds approach nor the money-income link, but are merely interested in “what happens in between” — what they have called the

“black box.”’5

The discussion above has hopefully established the fact the New Keynesians adopt a theory of credit rationing because of the general limited supply of loans. In this sense, since the demand for loans is at all times greater than the supply, rationing must occur. This intuitive argument runs counter to the overall post-Keynesian framework.

252

Credit, Money and Production

SOME RECENT DEBATES IN NEW KEYNESIAN THEORY As outlined in the beginning of this chapter, New Keynesians do not critically rely on credit rationing in order to achieve their results. The main characteristic of the theory is that the central bank can directly control the supply of credit from which banks can lend, due to its control over the supply of high-powered money and the non-substitutability between loans and other assets. Recently, however, these assertions have been undermined within the New Keynesian literature by a number of debates on the importance and impact of financial innovations. The recent emphasis by some New Keynesians in considering financial innovations is an attempt to inject into their otherwise orthodox framework some elements of the real world. It is a recognition that financial markets innovate following central bank policies, which in Minsky resulted from high interest rate policies. This suggests that two of the three characteristics of New Keynesian theory described at the beginning of this chapter break down. Hence, the Modigliani—Miller condition rejected by New Keynesians is reinstated, implying that bank loans and other credit instruments become near perfect substitutes. Second, as a result, the central bank loses much of its control over the supply of bank credit, and the lending channel of monetary policy breaks down. Take for instance, the following reference by Kashyap and Stein (1994, p. 223): The quantitative importance of the lending channel is likely to be sensitive to a number of institutional characteristics of the financial markets. Thus understanding the lending channel is a prerequisite to understanding how innovation in financial institutions might influence the potency of monetary policy.

Bemanke and Gertler (1995, p. 41) have also claimed that “the Bemanke and Blinder (1988) model is a poorer description of reality than it used to be, at least in the United States.” In particular, they point to the abolishment of Regulation Q, lower reserve requirements, and the elimination, in some respects, of requirements on many types of deposits. The debate was initiated in 1990, in an influential article by Romer and Romer, in which the authors claim that since banks can “fund” loans at the margin using managed liabilities, the direct impact of open-market operations on bank lending is minimal. The idea is that banks can mitigate the effects of tight money on lending by issuing CDs and other kinds of managed liabilities to offset any drop in deposits. Much of their analysis

New Keynesian Monetary Theory and the Transmission Mechanism

253

concems the fact that in the US, long-term CDs do not require any reserve requirements. Hence, according to the authors (1990, p. I56), “Reductions in the stock of reserves caused by restrictive monetary policy would reduce the quantity of transactions balances. But this would have no impact on banks’ ability to lend: banks could simply issue more CDs." Moreover, the authors (1990, p. 157) claim that “banks greatly mitigate any direct impact of tight monetary policy on their lending by issuing CDs subject to low reserve requirements in response to a decline in the quantity of transactions balances caused by a reduction in reserves." Edwards and Mishkin (1995) argue that as a result banks play a less prominent role in credit markets today precisely because banks are able to circumvent the restrictive powers of the central bank (see also Mishkin, I995). Attention has also focused on asset management. Recall that one of the characteristics of New Keynesian economics, as listed above, is that bank loans are special, and that substitutability between loans and other credit instruments is imperfect. Emphasis has focused on the following innovations: l) the rise of nonbank — or nondeposit-taking — intermediaries (finance companies); 2) the increasing use by firms of commercial paper to meet their financial needs; and 3) the increasing use by firms of pre~ determined and established lines of credit. Combined, these elements

imply, according to the critics, an important weakening of the link between central bank policy and bank loans. Hence, tightening by the central bank would have little impact on the economy except through the traditional money channel. Let us take a closer look at each of the arguments. According to Becketti and Morris (1992, p. 71), “The rapid growth [in recent years] of nonbank sources of business credit has led others to believe that bank loans have become less special. If these other sources of business credit are in fact good substitutes for bank loans, then a slowdown in bank lending is not as damaging to the economy.” The authors then argue that the use of finance companies and commercial paper has increased considerably and, as such, have become good substitutes for bank loans, especially since rating firms like Moody's and Standard and Poor’s can provide viable and credible information gathering, monitoring and credit evaluation - activities once reserved to banks. These critics argue that banks have lost their advantage over those other sources of credit. With respect to non-bank financial intermediaries, two specific arguments have been presented, both undermining the bank lending channel. First, if non-bank financial intermediaries control a large part of the loans market, then obviously any restrictive monetary policy would have little affect. Second, even if non-bank financial intermediaries are marginal players, but succeed in picking-up the slack resulting from an “effective” central bank

254

Credit, Money and Production

attempt in reducing reserves, the overall impact on loans and therefore on money supply will be minimal (Gorton and Pennachi, 1990). As argued by Kashyap and Stein (1994, p. 229), “If nonbank intermediaries are responsible for most of the lending volume in the economy, the Fed will be unable to have much of an impact on the overall supply of intermediated loans, even if they can influence bank supply.” Another area of interest within financial innovations is asset management. The argument is that if deposits or high-powered money fall banks may sell other assets, such as T-Bills, rather than cut back on loans thereby rendering central bank policy once again ineffective. With respect to lines of credit, Morgan (1990) claims that that in developed capitalist economies, banks enter into formal “credit commitments” with their borrowers, agreeing to supply them with lines of credit at agreed upon rates of interest. Morgan (1992, p. 40) distinguishes between “revolving credit agreements” and “lines of credit.” The former refers to “formal, long-term contracts committing the banks to lend to the holder for several years,” where the latter are typically shorter term. Either way, both instruments may effectively isolate borrowers against the possibility of a credit crunch. According to Morgan (1990, p. 51), “Commitments obligate banks to lend at predetermined terms, commitment holders are shielded from such rationing.”'° Since the Romer and Romer (I990) article, many New Keynesians have recognized the impact of financial innovations on the bank lending view. For instance, Blinder (I987, p. 327) has stated that “if there are ready substitutes for money, control of money will not give the authorities much leverage over the real economy.” Gertler and Gilchrist (1993, p. 44; emphasis added) claim similarly that the credit view maintains that owing to reserve requirements on deposits, monetary policy directly regulates the availability of bank credit. [But do]

reserve requirements in fact enable monetary policy to directly constrain bank lending. As we discuss, whether this prerrrise is reasonable in the current regulatory/institutional environment is the subject of current controversy.

Accordingly, Bernanke and Gertler (I995, p. 29) recently concluded that “institutional changes during the past 15 years have rendered the bank lending channel, at least as traditionally conceived, somewhat less plausible." The recognition that central banks may have little control over the money supply and economic activity implies that much of the New Keynesian theoretical foundation crumbles. It is little wonder that in recent years New Keynesians, while recognizing the importance of financial innovations, have attempted to limit the impact financial innovations have on their theory by

New Keynesian Monetary Theory and the Transmission Mechanism

255

imposing constraints and limits, which bolster their claim that central banks still have influence over credit, albeit to a lesser degree. New Keynesians conclude that while financial innovations have perhaps reduced the power of the central bank over the supply of reserves and loans, they have not eradicated it. As Kashyap and Stein (1994, p. 229) conclude, “Thus it seems reasonable to believe that shocks to the supply of intermediated loans might have important aggregate implications, even in today’s [financial innovations] enviromnent.” Hence, despite recognizing these institutional changes and the importance

of financial innovations, Bemanke and Gertler (1995, p. 40) nonetheless argue that these effects, at the very least, warrant more research, and that conclusions should not be hastily arrived at. The New Keynesian rebuttal has essentially concentrated on market discrimination between small and large firms, and smallversus large banks. For instance, with respect first to certificates of deposits, New Keynesians have claimed that smaller banks do not generally have access to the CD market. According to Morgan (1992, p. 38), “Smaller, under capitalized banks may not be able to issue all the CDs they need at the market interest rate.” This argument was echoed by Kashyap and Stein (1994), and Gertler

and Gilchrist (1993). This counter-criticism was also raised by Bemanke and Gertler (1995, p. 41): Despite financial reforms and innovations, it remains likely that the demand for banks’ managed liabilities is not perfectly elastic: For example, large CDs are incompletely protected by deposit insurance, which raises the usual issues

of evaluation and monitoring by lenders. Further, the market for CDs is still not especially deep or liquid relative to some other public markets — as shown by the fact that many CDs are nonnegotiable or difficult to trade on secondary

markets — so that new investors can be called forth only by paying higher rates of interest. Gertler and Gilchrist (1993, p. 45) add the following: First, in periods of tight money, the Fed has often raised the legal reserve requirements on managed liabilities.

Thus, even though the Fed recently

eliminated reserve requirements on managed liabilities, it is quite conceivable, at least inferring from past experience [the 1973-74 and 1979-80 periods of monetary tightening] that they would not hesitate then if they felt the need to

seriously contract the growth of bank credit. A third factor, according to Gertler and Gilchrist (1993, p. 46) is whether banks can effectively issue large CDs elastically at the margin. The authors argue that since they are not federally insured, large CDs raise the risk of banks’ portfolios.

256

Credit, Money and Production

With respect to asset management, Bernanke and Gertler (1987), as well as Kashyap and Stein (1994) and Greenwald and Stiglitz (I991) have all argued that while commercial banks do indeed practice some sort of asset

management by selling off T-Bills rather than reducing their supply of loans, they claim nonetheless that this does not imply that commercial banks are indifferent to the composition of their assets. In particular, they claim there is a clear trade-off in terms of rates of return between loans and TBills to which banks are not indifferent. Since T-Bills generally offer a lower rate of retum, banks will strive to maintain a certain optimum balance between the two. On the question of non-bank intermediaries, New Keynesians have argued two important points. First, there are definite costs involved in switching from banks to non-bank intermediaries which cannot be ignored. Kashyap and Stein (I994) have shown that there exists an important degree of “locking-in” — that is borrowers tend to stay with the same financial institution. As the authors (1994, p. 228) state, “The very fact that a bank does monitoring creates the potential for lock-in. In the course of a relationship, a bank will acquire an informational monopoly with respect to its client, a monopoly which puts other potential lenders at a comparative disadvantage.” There have also been some empirical papers supporting this view (Rajan, 1992; Sharpe, 1990). Hence, non-bank intermediaries do not “pick up the slack" when the central bank reduces its supply of highpowered money. . With respect to lines of credit, Morgan (I990, p. 58) pertinently asks, “How broad is this shield?” Typically, smaller investors and borrowers will not be able to obtain lines of credit, primarily because they do not possess adequate assets or creditworthiness.” The degree to which lines of credit can successfully protect the economy from a fall in the supply of credit initiated from an exogenous decision by the central bank to decrease reserves will depend on the degree of discrimination. As Morgan (I990, p. 58) argues, “Bank commitments cannot protect the entire economy because the smaller firms most likely to be rationed in a crunch are the least likely to own commitments. Finally, the case of commercial paper. Here again, New Keynesians argue that markets discriminate against smaller firms. According to Becketti and Morris (1992, p. 75), “Only the largest and most creditworthy firms can issue it.” Similarly, Morgan (1992, p. 38) argues that “small and mediumsized firms may be unable to raise funds directly from investors in public

debt and equity markets.” In some respects, asymmetric information also applies to commercial paper. Households and firms have different information conceming the “creditworthiness” of firms, and households must determine whether firms issue commercial paper for financially sound

New Keynesian Monetary Theory and the Transmission Mechanism

257

reasons, or whether they are attempting to pass on bad debts. Hence, according to Morgan (1992, p. 39), “Even large firms that issue publicly traded stock and debt might be financially constrained to some extent. Investors may have lingering uncertainty about the prospects for even publicly traded firms.” Gertler (1988, p. 570) concludes that “asymmetric information about the value of a firm’s existing assets can restrict its ability to issue new shares. Outside lenders must discuss whether the share issue is a legitimate effort to either obtain new financing or diversify risk, or is instead simply an attempt to pass off bad assets (see also Greenwald, Stiglitz and Weiss, 1984). To conclude, while Bemanke and Gertler (1995, p. 40) remind us that “Bemanke and Blinder’s (1988) model of the bank lending channel suggested that open market sales by the Fed, which drain reserves and hence deposits from the banking system, would limit the supply of bank loans by reducing banks’ access to loanable funds,” they nonetheless argue that the bank lending channel remains an important element, and that the reserveloans causality is preserved, despite the rise of financial innovations. According to the authors (1995, p. 41), “An open-market sale by the Fed shrinks banks’ core deposit base and forces them to rely more on managed liabilities.” Asset and liability management, the rise of commercial paper and non-bank intermediaries aside, as long as there is discrimination in banking and markets, the bank lending channel will remain an important explanation of the monetary transmission mechanism. Before turning to a comparison of New and post-Keynesian theories, two last arguments are necessary. First, a discussion on what appears at first

glance to be interesting passages in the New Keynesian literature is briefly examined below. Second, some criticism of the New Keynesian literature from the post-Keynesian point of view will better help us to judge the merits of the discussion between the conflation of New and post-Keynesian theories.

SOME INTERESTING PASSAGES IN THE NEW KEYNESIAN LITERATURE Advocates of post-Keynesian theory would have noticed in reading the above section that there appear to be some parallels between New Keynesians and post-Keynesians. Indeed, as will be argued below, there does appear to be some room to compare both approaches, especially at the policy level. Moreover, in reading the literature, one comes across some firrther passages which appear to be of a certain post-Keynesian flavor.

258

Credit, Money and Production

For instance, Bernanke and Blinder have argued convincingly that the supply of reserves by the central bank can not only be endogenous, but also perfectly elastic. According to the authors (1992, p. 903), “The supply of nonborrowed reserves between Federal Open Market Committee (FOMC) meetings is extremely elastic as the target funds rate. The funds rate or the funds-rate spread though not statistically exogenous, is at least predetermined within the month.” Later, the authors (1992, p. 914) claim that “for the funds rate to be a good measure of monetary-policy actions, it must be essentially unresponsive to changes in reserve demand within a given month. This amounts to saying that the Federal Reserve supplies reserves completely elastically, or nearly so, at its target funds rate."“’ In other words, “We think of the fed as setting a supply curve for nonborrowed reserves for the month or week with a horizontal curve.” Hence, the rate of interest is seen as a quasi-exogenous variable. As Bemanke and Blinder (1992, p. 903) argue, “Movements in the funds rate are genuine policy changes, not simply endogenous responses of the Federal funds market to changes in the economy. We consider it reasonable to treat either the funds rate or the funds rate spread as a measure of Federal Reserve policy.”’° Elsewhere, post-Keynesians can read that credit is needed because of the existence of time, that firms need access to credit in order to finance not only investment, but also working capital, essentially due to the time discrepancy between incurring costs and receiving revenues from the sale of their products. This argument is made by Blinder (1987, p. 329) who claims that “firms need credit for working capital before they receive revenues from sales, and must borrow in order to do so. Credit [is] an essential ingredient in the production process." Credit is wholly supplied by the banks since “there is no auction market for credit. [As a result] banks have primacy in the credit market in a very stark way.” At first glance, it is little wonder that post-Keynesians see in New Keynesian theory a possibility of dialogue.” It is indeed a puzzle to reconcile the above statements with the overall New Keynesian position. For instance, as is described above, New Keynesian theory relies primarily on the central bank’s ability to control reserves and to influence batik loans. The only possible reconciliation of the above statements with the overall New Keynesian position is to argue, as do post-Keynesian structuralists, that the central bank may not fully accommodate the banks’ demand for reserves, in which case, they would have to resort to either lowering their supply of loans or practicing asset management.“ Another possible answer to this riddle is one given by Bemanke (1993, p. 56) and quoted above: “Credit rationing is certainty consistent with the

New Keynesian Monetary Theory and the Transmission Mechanism

existence of a credit channel.

259

However, credit rationing is not at all

necessary for the credit channel to exist. All that is required for a credit channel is that bank credit and other forms of credit be imperfect substitutes.” Here, Bernanke is downplaying the first of the characteristics of New Keynesian theory outlined earlier: it appears that for the author, the control of reserves is not a necessary condition, as long as loans- and other credit instruments remain imperfect substitutes, thereby implying that if banks refuse some borrowers, these cannot receive finance elsewhere. Reserves can be exogenous or endogenous, but as long as bank loans are special, New Keynesian theory retains its essentiality.

POST-KEYNESIAN CRITIQUE OF NEW KEYNESIAN THEORY Before proceeding to a comparison of post— and New Keynesian theories, it is worth dwelling on some of the criticism which has been raised by postKeynesians conceming the merits of some of the theoretical arguments of New Keynesian theory. This section will address four specific points. Interestingly enough, while

post-Keynesians have criticized specific features of New Keynesian theory, they have overlooked its overall framework, with perhaps the exception of van Ees and Garretsen (1993). This point will, however, be addressed more fully in the conclusion to this chapter. 1) The notion of asymmetric information relies on the argument that ex ante, all borrowers are perceived by banks as identical. Indeed, banks cannot tell the good from the bad borrowers, and hence must ration the given supply of credit. Wolfson (1996, p. 446, n. 2) has criticized this position claiming that while New Keynesians argue that borrowers are identical, they do not explain how the rationing occurs. They do not explain how banks finally decide on how to divide the available supply — that is who among the potential borrowers receive credit. Ex post, New Keynesians admit that the distribution of credit may not be the most optimal one, yet they cannot explain how ex ante, banks divide the available supply if bank borrowers are identical. Moreover, New Keynesians do not offer any criteria used to determine which borrowers received loans. 2) Credit rationing rests on the notion that banks are able to calculate the rationed-equilibrium rate of interest, r*, at which level it will ration credit according to its supply constraints. Banks therefore refuse to raise r* since it corresponds to the profit maximizing rate.

260

Credit, Money and Production

Following Crotty (1994, 1996), Wolfson (1996, p. 445) has argued that in order to calculate r*, banks must have a tremendous amount of information: “The bank has to know an enormous amount about its borrowers in order to calculate r*. The bank needs to know precisely how these borrowers will react to increases in the interest rate.” Crotty has gone even further. According to Crotty (1996), asymmetric information does not address the issue of the quantity of information needed to calculate r*. According to the author (1996, pp. 336—7), The [New Keynesian] models are often game-theoretic in structure. Borrowers optimize given the market rate of interest; their “reaction firnctions“ describe credit demand and project choice as a function of the interest rate. Lenders — the market’s “leaders” — then select the equilibrium interest rate so as to maximize their expected utility given borrowers’ reaction functions. But in order to be able to construct these reaction functions, lenders have to know not only the population of potential investment projects, but also borrower preference functions. In other words, lenders must possess all the information required to solve the borrowers’ optimal choice problem in order to solve their own.

Moreover, Crotty (1996, p. 337) claims that “if the Neoclassical lender can be said to have ‘perfect’ (though stochastic) information, then the New Keynesian lender must possess information that is in some sense beyond

perfect, yet inadequate.” 3) The assumption of asymmetric information, while also discussed below, has been criticized for not dealing with Keynesian-type fundamental uncertainty. Post-Keynesians present one of five arguments: 1) asymmetric information and uncertainty, while different, are complementary (Dymski, 1993, 1994); 2) asyrrimetric information necessarily implies an environment of fundamental uncertainty (Crotty, 1996; Fazzari and Variato, 1994, 1996); 3) according to Davidson (1994, p. 550), asymmetric information is an issue of law rather than of economics: “Misrepresenting oneself in signing a loan contract is, after all, fraud.” 4) asyrrimetric information ignores fundamental uncertainty and both concepts cannot be reconciled (Wolfson, 1996; van Ees and Garretsen, 1993); finally, 5) there is no such thing as asymmetric information — it cannot exist (Screpanti, 1997). As a direct criticism of New Keynesian theory, Screpanti’s (1997) and Cr0tty’s (1996) arguments are the most important. According to Crotty (1996, pp. 336, 347—8), Al does imply FU. Since Al implies FU and there is no logically consistent way to integrate FU in a New Keynesian model, there is no logically consistent way to incorporate Al either. These models must be misspecified.

New Keynesian Monetary Theory and the Transmission Mechanism

261

While FU is logically incompatible with New Keynesian theory. Al seems perfectly consistent with Post Keynesian core assumption. Screpanti’s (1997, p. 3) argument is that banks spend a considerable amount of time gathering information and as a result, have developed the necessary skills to eliminate asymmetric information: “Banks contribute to overcoming the potential information asymmetry with any specific risk, and do so by dealing with the asymmetry themselves, by practically getting rid of them.”22 4) The overall framework of the New Keynesian monetary theory is supply-determined. As opposed to the credit-demand views of postKeynesians, van Ees and Garretsen (1993, p. 43) have argued that “from a macroeconomic point of view the action is on the supply side of the economy and aggregate demand is typically exogenous. Changes in, for example, the degree of credit or equity rationing imply that the aggregate supply schedule shifts and hence that supply shocks cause changes in output and employment.” This point was amply emphasized throughout the chapter. It was argued above that if credit is rationed in New Keynesian economics, it is because

the demand for credit is typically greater than the available supply of credit, as determined by deposits and savings. Moreover, if we argue that credit rationing is not an essential element of New Keynesian macroeconomics,

then we know that the supply of loanable funds is still determined exogenously by the supply of high-powered money and bank deposits. Either way, the supply is exogenous, and any resemblance to money endogeneity must be understood in terms of a supply-determined theory.

POST-KEYNESIANS AND NEW KEYNESIANS: CLOSE SIBLINGS OR DISTANT COUSINS? Having now carefully analyzed in this chapter the framework and details of New Keynesian economics, we are now in a better position to compare it with post-Keynesian theory. One possible problem quickly comes to mind. As was argued in previous chapters, post-Keynesian theory is not a unified theory. The last chapter even argued that some elements of post-Keynesian monetary theory border on orthodoxy. This certainly stands in contrast to the post-Keynesian horizontalist tradition which, as was argued in chapter four, is more heterodox in many more respects. In what follows, the expression “post-Keynesian" is used in its more general nature — that is as defined in chapter one where elements common to both post-Keynesian approaches were identified - unless specified otherwise.

262

Credit, Money and Production

It has been suggested now by many post-Keynesians that it is perhaps possible to forge an alliance with the New Keynesians to create a general Keynesian synthesis. For instance, Dymski (1996, p. 378) asks, “Can New and Post Keynesian insights be combined in synthetic economic models?” to which the author answers in the affirmative. Similarly, Neal (1996, p. 398) also believes it is possible to reconcile post— and New Keynesians based on uses of uncertainty and risk. Structuralist post-Keynesians — more so than the horizontalists — have generally borrowed some aspects of the New Keynesian monetary model without, so it appears, realizing that they are attached to an overall framework and model which are incompatible with the overall postKeynesian model. Hence, attempts by many post-Keynesians to pick and choose must be seen as a limited understanding of the New Keynesian model. This chapter has attempted to deconstruct New Keynesian economics. While it is certainly acceptable to pick and choose elements of New Keynesian theory to construct a post-Keynesian theory of banking, this must only be done as long as the elements are understood in relation to the overall framework. For instance, it makes little sense for post-Keynesians to borrow from New Keynesians the notion of credit rationing, since this concept is set within the overall theme of a given, supply~detennined, stock of loanable funds. The comparison between New Keynesian and post~Keynesian economics can be done at the policy and theoretical levels. It is argued here that many of the comparisons which can be drawn are done primarily at the policy level. At the theoretical level, however, it is difficult to see many common elements between the New and the post-Keynesian approaches to macroeconomics and monetary theory. At the level of policy and the role of the central bank, New Keynesians and post-Keynesians have a few arguments in common. This has prompted some post-Keynesians to agree with many of the policy recommendations of the New Keynesians. In general, this is not a misplaced statement, as can be seen from the list below. In policy and theory therefore, there may appear to be some similarities between post— and New Keynesians. I) Both theories emphasize the powerful role of banks in determining the “pace of economic activity”. In this sense, post— and New Keynesians recognize that monetary theory affects the real economy through the credit channel, and that banks are “special” in the sense that they can "create" money (Kashyap and Stein, 1994). Although there are important differences on the precise transmission mechanism, both approaches nonetheless accept the fact that credit and the financing of production and investment are central to the analysis (Seccareccia, 1997; Bemanke, 1993).

New Keynesian Monetary Theory and the Transmission Mechanism

263

In this sense, both approaches emphasize “monetary nonneutrality” (Kashyap and Stein, 1994, p. 221; Davidson, 1972). 2) New and post-Keynesians recognize the volatility of investment decisions, albeit for different reasons (see below). Morgan (1992, p. 38) argues that “by making investment spending more volatile, financial constraints make the economy more volatile. And by slowing investment spending on plant and machinery, financial constraints slow the economy’s long-term growth.” In this context, credit plays an important role in the growth and fluctuations of output (Kaufman, 1987; Eckstein and Sinai, 1986; Asimakopulos, 1991; Davidson, 1972; Lavoie, 1992). 3) Given the importance of banks and credit, New and post-Keynesians each prefer a monetary policy of low interest rates as a stimulus to investment. 4) Both approaches have emphasized the recent institutional innovations, in particular the impact of asset and liability management, and its impact on bank behavior. This implies that in both approaches, the emphasis is placed on the development of markets and the importance of institutions. For instance, Gertler (1988, p. 559) argues that “financial markets and institutions deserve serious attention” (see Minsky, 1975). 5) New Keynesians claim that “the Fed [is] able to affect - by

manipulating the amount of reserves available to the banking sector - the aggregate supply of loans by intermediaries” (Kashyap and Stein, 1994, p. 228). This argument applies irrespective of whether reserves are endogenous or exogenous to the system.” This is precisely the same

argument made by some neo-post-Keynesians whereby the lack of accommodation on behalf of the central bank will impact on the supply of reserves — and hence on the supply of loans - of commercial banks. 6) Both approaches argue that banks are lirriited in the amount of credit they can extend to creditworthy borrowers. Post- and New Keynesians emphasize the “fringe of unsatisfied customers”. While horizontalists also emphasize credit constraints, their approach is still nonetheless firmly rooted in the demand side. However, by emphasizing the lack of accommodation, neo-post-Keynesians also focus on the supply side of the loans market, as do New Keynesians. As a result, credit rationing, as a supply problem, becomes important. Dymski (1992, 1993) accepts the credit rationing literature as valuable. 7) Both approaches reject the perfect or complete information assumption of New Classical models. This was recognized by Dymski (1996, p. 378), Both New Keynesian and Post Keynesian models proceed by postulating some important deviation from the complete-information assumption of Walrasian theory. New Keynesians have explored how informational asymmetries may induce credit rationing, price stickiness and unemployment;

264

Credit, Money and Production Post Keynesians have explored how real time and "Keynesian uncertainty“ undermine agents’ ability to make longer term plans with confidence

For Dymski (1992), these two approaches are hence “complementary”. For the author, in tranquil economic times, bank behavior is better described according to the New Keynesian theory of decision making under asymmetric information, while in more turbulent times, Keynesian uncertainty is a better description of the real world. 8) Both approaches, that is the New Keynesians and the structuralist postKeynesians, have in common an upward sloping money supply curve. Referring to Henderson’s criticism of Keynes (xxix, pp. 224-6), Lavoie (1996, p. 7) claims that, These various statements of the loanable funds theory are undoubtedly compatible with an upward-rising credit-money supply curve. This is why it is important to make sure that the reasons supporting such a construct are not linked in any way to a theory of the natural rate or to some form of loanable funds theory.

9) Both approaches share in common the fact that investment is partly financed by retained earnings and partly by bank credit (Seccareccia, 1997; Bemanke and Gertler, 1987). Despite these corrimon arguments, however, it is quite another thing to claim that post-Keynesians and New Keynesians have similar approaches to macroeconomics and monetary theory. Important differences exist, and it is these differences, in particular methodological in nature, that makes the union of post-Keynesian and New Keynesian theories unlikely. While perhaps related, it is best to refer to New and post-Keynesians as distant cousins, each rooted in Keynes, albeit a very different Keynes. 1) Perhaps the greatest difference between the New Keynesians and the post-Keynesians is the very theoretical foundation of their respective approaches to economics, macroeconomics and monetary theory in particular. Post-Keynesians, it is well known and accepted, reject the assumptions of neoclassical theory and the Walrasian equilibrium conditions. In contrast, New Keynesian economics accepts most of the neoclassical theoretical apparatus. As noted by Dymski and Pollin (1992, p. 36), “The New Keynesians accept the new classicals’ premises of long-run equilibrium and of the need to microfound macroeconomic relations using the postulate of individual rationality. The financial factors which they have identified as affecting real outcomes, then, make no reference to Keynesian

New Keynesian Monetary Theory and the Transmission Mechanism

265

uncertainty.” This echoes a comment made by Dymski (1993, p. 49) where the author states that “new Keynesian economists have almost invariably developed models that deviate as little as possible from standard orthodox premises.” Crotty (1996, p. 334) makes a similar conclusion. According to the author, “There are few significant methodological differences between New Keynesian and Neoclassical theories,” although he also adds that “an undeniable advantage to New Keynesians of the close affinity of their assumptions and methodology to Neoclassical theory is that it has permitted them to be taken seriously by mainstream economists and in mainstream journals, a status denied to those who base their works on the assumption of [fundamental uncertaintyl.” For their part, Fazzari and Variato (1994, p. 354) claim that New Keynesian theory “rests on formal optimization models derived from Neoclassical first principles.” Similarly, van Ees and Garretsen (1993, pp. 38, 43) also claim that “the new Keynesian economics of financial markets introduces asyrmnetric information into an otherwise neoclassical framework. In fact, the macroeconomic framework is often cast in terms

of a standard general equilibrium model in which only the number of financial assets is enlarged.”

2) While it is correct to assume that post-Keynesians and New Keynesians agree on the importance of banking, there are nonetheless linrits to this assertion. First, it is worth mentioning that New Keynesians generally build their banking theory on standard microfoundations. As Romer (1993) argues, New Keynesians made progress in understanding the microeconomics of price rigidities, asymmetric information and credit rationing. In this sense, banks are financial intermediaries facilitating the transfer of funds from savers to borrowers, thereby reducing the cost of monitoring fil'l’l'lS and gathering information. In contrast, while post-Keynesians have not sufficiently developed microfoundations to banking and endogenous money, one thing which can be emphasized is that such rriicrofoundations would not be independent of macroeconomic laws, and vice versa. Post-Keynesians would not build their theory of banking based on rational expectations, for instance. 3) On the issue of the determination of the rate of interest, several issues arise. On the one hand, it is clear that New Keynesians and post-Keynesians differ on how the base rate of interest is detennined. As argued in chapter five, there is now wide-ranging consensus among post-Keynesians that this rate is determined exogenously by the central bank. On the other hand, New Keynesians still accept the view that this rate is endogenously determined by market forces, the supply and demand for money.

266

Credit, Money and Production

The determination of the loans rate of interest is another issue. Interestingly enough, both schools accept the mark-up view, where the mark-up may or may not be given. However, the determination of this mark-up is a source of a certain confusion. In post-Keynesian theory, it can be claimed that this mark-up is determined by a target rate of retum set by the banks. The post-Keynesian view is not necessarily incompatible with the profit-maximizing approach of New Keynesians, where the mark-up is foremost a result of costly monitoring and information-gathering activities the extemal risk premium. Moreover, as was argued above in the context of the balance sheet channel of monetary policy, the external risk premium is directly linked to the coverage ratio of firms. As the central bank raises its base rate, the net worth of firms may fall given a decline in cash flow. As a result, banks may increase the extemal risk premium since the firms now are seen as more risky. Another way of looking at this is to say, as does Wray (1990), that firms become more illiquid. In this way, both the New Keynesians and the post-Keynesians (structuralists) see the loan rate of interest as endogenously linked to the liquidity of firms. 4) The issue of the complementarity between asyrrrrnetric information and

uncertainty has not found a sympathetic ear in post-Keynesian circles, despite Dymski’s (1992, p. 314) claim that both approaches share in common the “relaxing [of] the Walrasian informational assumptions.” As van Ees and Garretsen (1993, pp. 38-40) argue, “The idea of the complementarity of the post and new Keynesian approach rests upon a mistaken interpretation of both the new and the Post Keynesian literature. The new Keynesian concept of asyrrimetric information and the Post Keynesian reliance on fundamental uncertainty imply that the analytical framework of both theories is radically different.” In other words, New Keynesians rely on “unequal” information, whereas post-Keynesians rely on the “lack” of knowledge. On this issue, two criticisms can be raised.

First, the post-Keynesian

emphasis on uncertainty would suggest that whether the economy is in tranquil or turbulent times, uncertainty must necessarily prevail. This is somewhat the argument put forth by Fazzari and Variato (1994) and Crotty (1996). For them, asymmetric information, while perhaps technically

correct since lenders may not always have full information, nonetheless implies that it is equivalent to an environment characterized by Keynesian fundamental uncertainty. Crotty (1996, p. 450) claims that despite asymmetric information, both lenders and borrowers “are subject to fundamental uncertainty about the future.”

New Keynesian Monetary Theory and the Transmission Mechanism

267

In this sense, Fazzari and Variato (1994) and Crotty (1996) claim that the concepts of asyrrunetric information and fundamental uncertainty are complementary, although not in the same way advocated by Dymski. Second, Screpanti (1997, p. 3) has taken another route to criticize New Keynesians regarding asymmetric information. For the author, there is no such thing as asymmetric information. It does not exist precisely because banks ensure that it does not: “If there is an industry in which asymmetric infonnation is scarcely relevant, it is precisely the banking sector." This is because banks have developed “an expertise and a technical skill” in acquiring information thereby reducing the “imperfection of the lender’s perception of the borrower’s risk.” 5) While the New Keynesians recognize the fact that credit drives the economy, this statement is not entirely compatible with the post-Keynesian approach. Whereas in post-Keynesian theory, credit is essentially “demanddetermined” — although it does not negate the possibility that banks will refuse credit to some borrowers - in New Keynesian theory, credit is entirely supply-driven. In fact, this characteristic is one of the three described at the beginning of this section. While many post-Keynesians would be tempted to conclude that both they and New Keynesians

emphasize credit, this would be an inappropriate conclusion. Post-Keynesians emphasize, correctly, the importance of credit demand and the volatility of investment based on the impact of uncertainty on investment decisions of entrepreneurs. In this way, post-Keynesians place the volatility of investment decisions at the center of their analysis. New Keynesians, however, put little interest on the importance of the demand for credit, and even less on investrnent decisions. This is made clear in Morgan (1992, p. 38): “Financial constraints affect both the stability and growth of the economy. By making investment spending more volatile, financial constraints make the economy more volatile.” In fact, we may conclude that there is little room for demand in New Keynesian economics. Moreover, Bemanke and Gertler (1995, p. 41) argue that “an increase in the cost of funds to banks should shift the supply of loans inward.” This suggests that investment is volatile because of supply conditions, not demand or the impact of uncertainty on investment decisions. Hence, any impact credit conditions have on unemployment and other macro variables is a result of changes in the supply of credit. This implies that if the central bank made available more reserves, then banks could conceivably meet a greater portion of demand. The effect of central bank decisions is foremost on the aggregate supply curve, not demand. As van Ees and Garretsen (1993, pp. 43-4) argue, “The idea that frictions on the supply side are responsible for fluctuations in output and employment and that aggregate supply and demand are independent imply

268

Credit, Money and Production

that a [sic] from a Post Keynesian point of view the new Keynesian analysis just provides a microfoundation for pre-Keynesian classical macroeconomics.” Hence, the difference between post-Keynesians and New Keynesians is in trying to fill in the “black box” of monetary transmission. For postKeynesians, the causality is from the exogenous rate of interest to the demand for credit. For Neo Keynesians, it is through the supply of reserves to the supply of credit. This means that the central bank can affect the supply of loans (Kashyap and Stein, 1994). In fact, there is no role for demand in New Keynesian economics. Since money is treated like other goods in Walrasian systems, the following reference to Gordon (1990, p. 1117) applies to money as well: “The entire demand side of the economy is omitted [in New Keynesian economics]. Topics on the demand side can be omitted simply because they are not at the heart of the conflict between new-Keynesian and new Classical macroeconomics.” 6) While post-Keynesians do argue, in a secondary manner, that reserves

impact on loans, they nonetheless overwhelmingly accept the reverse causality argument, that loans determine deposits, which then determine reserves. For New Keynesians, this is never the case. Their model is strictly set within the orthodox causality framework. In their model, reserves always determine loans, and the supply of loans is endogenous to monetary policy. 7) The two previous points are important arguments with respect to the theories of endogenous money as described earlier. Post-Keynesians, while at times flirting with the portfolio approach, overwhelmingly support the revolutionary endogeneity approach. In contrast, however, New Keynesians categorically reject this endogeneity approach in favor of the portfolio approach, that is a multiplier model with a loans market. In fact, this is precisely how Kashyap and Stein (1994) describe New Keynesian theory. On a related issue, while both approaches, as indicated above, do

concentrate on the role of credit, their respective views on endogeneity and exogeneity are evident when in post-Keynesian theory, money is created ex nihilo, and credit has an independent role to play. In contrast, New Keynesians claim that credit is not “an independent causal factor affecting the economy. it is not a primitive driving force” (see Bemanke and Gertler, 1995, p. 45). 8) Another important difference between post— and New Keynesians concerns the extension of loans to creditworthy customers. Some postKeynesians - more notably the horizontalists and the circuit theorists argue that while banks are not reserve-constrained in their lending ability, they will always provide credit to customers as long as they are

New Keynesian Monetary Theory and the Transmission Mechanism

269

creditworthy. This is in the best interest - i.e. profit interest - of the banks. If credit is “constrained” in the horizontalist tradition, it is not because

credit is scarce, but rather because banks are unable to find creditworthy customers. There is no room in the post-Keynesian vocabulary for such expressions as “excess demand” for credit or the “availability of supply.” As Levratto (1992, p. 333) argues, “The New Keynesian notion of credit rationing is linked to a discrepancy between the supply and the demand for loanable funds. Market analyses based on rationing - the rejection of part of the demand for funds - attempt to show that decisions by banks to refuse credit are linked to profit maximizing strategies.” Seccareccia (1996, p. 140) further reminds us that for post-Keynesians, “The portfolio decision of the public in holding deposits has no causal consequence on the decision of the banks in both satisfying the demand for credit money and providing the desired funds at a given rate of interest.” In New Keynesian economics, creditworthy customers may not receive funding. According to van Ees and Garretsen (1993, p. 46), post— and New Keynesian approaches to credit are “built upon fundamentally different notions of real-world behavior.” In post-Keynesian theory, credit may be constrained by banks, but not because of supply conditions. 9) For post-Keynesians, an increase in interest rates will likely translate itself into a fall in the demand for credit for production and investment decisions. For New Keynesians, the issue is not as clear. While the balance sheet channel suggests a fall in credit supply, New Keynesians also claim that an increase in the rate of interest may lead to increases in demand because of moral hazard and adverse selection. 10) In New Keynesian theory, loans are financed from money, or from portfolio decisions. While this is also true of some post-Keynesians, as was shown earlier (see for instance Cottrell, 1994; Knodell, 1995), the primary emphasis is on the ability of banks to create money ex nihilo. Hence, New Keynesians are still to some extent rooted in the loanable funds approach. This is also the position of Levratto (1992, p. 333): “Credit rationing is linked to a uneven adjustment between the supply and the demand of loanable funds.” 11) As was argued in the previous section, one of the assumptions of New Keynesians - as with neo-Keynesians before them - is their reliance on price rigidities to achieve their results. In contrast, post-Keynesians argue that although prices are “sticky” they are not central to their conclusions. Even with perfectly flexible prices, post-Keynesians do not believe that markets would achieve full employment equilibrium. Moreover, postKeynesians argue that contracts and rigid real prices or nominal prices are a

270

Credit, Money and Production

way for entrepreneurs to deal with the uncertainty of markets. The postKeynesian model does not rely on imperfections. 12) For post-Keynesians, the reverse causality between saving and investment, and Keynes’s “banana parable” are important and vital elements to the whole theoretical structure. New Keynesians, on the other hand, readily admit that “few economists today believe that excessive saving threatens the economy” (Mankiw, 1992, p. 560). 13) Finally, a last point of difference may be brought to the surface. This concems the substitutability between bank credit and other credit assets. As was shown in this chapter, the imperfect substitutability between bank credit and other assets is an important element in the New Keynesian analysis. This allows the central bank to have a direct control over the supply of commercial banks’ supply of loanable funds. This in fact guarantees that the velocity of money is relatively stable. In contrast, post-Keynesians have emphasized the variability of the velocity of money, albeit ex-post. And while changes in the velocity of money ought never to be confused with the financing of production, the substitutability between credit and other assets is of no (or little) relevance in post-Keynesian theory. The central argument in post-Keynesian theory is the financing of production, independently of

prior saving, irrespective of its source. This explains why post-Keynesians have generally downplayed the difference between commercial banks and other financial institutions (with the exception of Pollin, 1996, and Palley, 1996).

NEW KEYNESIANS AND THE EARLY POSTKEYNESIANS Let us end this chapter with an interesting extension of the discussion, by arguing that while a reconciliation of post— and New Keynesian theories is perhaps not the best of the possible scenarios, it is worth exploring the parallels between the two approaches, not in terms of contemporary theory, but rather with a look back to the early post-Keynesians. Recall that in chapter three, a detailed analysis of the views of Minsky and Kaldor was offered. In chapter four, a similar treatment of Davidson’s and Rousseas’ contributions was presented. This exercise can now be linked to the current discussion by way of a discussion of the similarities between New Keynesian theory and the early post-Keynesians. It is argued here that New Keynesians have adopted, albeit four decades later, the same model as did Minsky, Kaldor, Davidson and Rousseas. First, let us recall the essential elements of the early post-Keynesian model, listed below:

New Keynesian Monetary Theory and the Transmission Mechanism

271

1) Reserves are exogenous, under the control of the central bank. The money supply is also considered exogenous. 2) Credit is supply-determined by the banks’ liabilities; household portfolio decisions determine largely the availability of firnds. 3) Because of financial innovations and the creation of “near monies,” central bank control over the supply of bank credit, and hence the total quantity of money in circulation, is diminished. This is equivalent to an autonomous increase in financial intermediation and is expansionary. 4) Financial innovations are generated by restrictive monetary policies. 5) The money multiplier is accepted, to which a loans market is added. 6) Implicitly, the early post-Keynesians were arguing that loans and other credit instruments are imperfect substitutes. This is evident when claiming that changes in the supply of credit drive the cycle. 7) In the case of Davidson’s (1965) exogenous money model, the IS and LM curves are made interdependent because of the inclusion of credit

demand. We can see that in fact the three New Keynesian propositions from the beginning of this chapter are found in the early post-Keynesians. As for the

third proposition, that is the one about sticky prices, Minsky (1963, p. 70; emphasis added) adopted that assumption as well. In his reply to Friedman

and Schwartz, Minsky states that “as we are considering a net increase in assets, unless the monetary change is wholly absorbed by a rise in asset prices, additional real or financial assets must be created in exchange for the money spent by the additional borrowing unit.” In light of the analysis of New Keynesian theory in this chapter, it should now be apparent that the current New Keynesian model of financial intermediation and credit is almost identical to that of the early postKeynesiatis, especially Minsky. This has now become more apparent with the New Keynesian emphasis on liability management. Early postKeynesians anticipated the New Keynesian theory of credit rationing and the batik lending channel. With the New Keynesian emphasis on liability and asset management, and the creation of credit substitutes, New Keynesians have come full circle, back to the late 1950s. If Minsky’s and Kaldor’s early views rest primarily on a changing value of the velocity of money because of financial innovations, then consider the following passage

in Blinder (1987, p. 327): “If there are ready substitutes for money, control of money will not give the authorities much leverage over the real economy.” In some ways, it is understandable why some post-Keytiesians see close affinities between themselves and New Keynesians. After all, if these postKeynesians see their own views closely related to those of the early Minsky,

272

Credit, Money and Production

Kaldor and Davidson (which is usually the case), then by arguing here that today's New Keynesians have the same model as the early post-Keynesians, then logically, it should all add up. However, recall that in chapters three and four, it was argued that Minsky’s and Kaldor’s early views were not “post-Keynesian" in the sense of endogenous money and reverse causalities. However, the post-Keynesian view has evolved considerably since then. Post-Keynesian claims therefore of a reconciliation between themselves and the New Keynesians are misplaced. It is a result of an important misreading of the New Keynesian literature. Yet, it is worth exploring the issue further. To date, the post-Keynesian discussion regarding a synthesis of New and post-Keynesian views has concentrated quasi-exclusively on asymmetric information and credit rationing. This, as stated earlier, is a misplaced exercise. We have, however, established that New Keynesians and the early post-Keynesians have the same model, and this is a definite beginning. The New Keynesian theory, however, is not limited to the bank lending channel. As shown above, there is also the balance sheet channel, where increases in the rate of interest affect the financial burden of firms, their debt burden, as well as their ability to borrow additional credit. This, it is argued here, comes very close to many ideas developed by Minsky and structuralist post-Keynesians, and many may be acceptable to horizontalists. Recall that New Keynesians argue that when central banks raise the rate of interest, this impacts directly on the cash flow of firms, measured by the coverage ratio - expressed as the ratio of interest payments to interest payments and profits. As the ratio increases, the ability of firms to pay back their loans is diminished. This is akin to Minsky’s analysis of pottzi finance, although Minsky’s analysis is richer in context and details by providing a better account of how the mechanism works. New Keynesians have called this the “financial accelerator."2’ Essentially, however, the reasoning and conclusions are the same. In both cases, bartks may impose a risk premium, based on Kalecki’s increasing risk hypothesis, identical to the view of Bernanke and Gertler (1995, p. 35) who claim that the “external finance premium facing a borrower should depend on borrower’s financial position." The greater the borrower's net worth, the lower the premium, presumably because, in post-Keynesian parlance, the borrower is more liquid and hence less risky. Moreover, in Bernanke and Blinder (1988), we find the same model which Davidson developed almost twenty-five years before, notably the interdependence of the IS and the LM curves, the real and monetary sectors. In their model the authors draw a downward sloping schedule in interest rate/income space, which they label “in deference to Don Patinkin” the CC curve - for cotrmiodities and consumption. It is clear that this is the IS

New Keynesian Monetary Theory and the Transmission Mechanism

273

schedule with the addition of the demand for credit, i.e. Keynes’s finance motive. Consider the following statement by the authors (1988, p. 436): “It is easy to see that the CC curve is negatively sloped like the IS curve, and for much the same reasons." The authors define this curve as,

y= Yli.¢(i.y. R)! where y is GNP, i is the rate of interest on bonds, and R is reserves. It is

solved by incorporating the demand for credit into the standard IS curve. However, Bemanke and Blinder (1988, pp. 436-7) write that, in contradistinction to the standard IS curve, the CC curve “is shifted by monetary policy (R) and by credit-market shocks that affect either the [loan demand] or [loan supply] functions, while the IS curve is not. An increase in R shifts both the CC and the LM curves outward.” Of course, perfect substitutability between loans and bonds is ruled out, as in post~Keynesian theory, since this would reduce the CC curve to the IS curve. The loan demand includes p, the rate of interest on loans, the rate of interest on bonds (for substitutability reasons), and income. In this same article, consider the following statement by Bemanke and Blinder (1988, p. 435): “We assume that lenders choose between bonds

and loans according to the interest rates on the two credit instruments. If p is the interest rate on loans and i is the interest rate on bonds [then] the desired portfolio proportions depend on rates of retum of both assets.” Assumingly, the central bank can, through its open-market operations, influence the relative rates, and hence dictate to the banks the appropriate allocations of assets between loans and other assets with respect to the profit maximizing assumptions and constraints of banks. Contrast this statement to a similar one made by Kaldor in 1955, and referred to in chapter three (see Kaldor, 1955, p. 106): Commercial banks thus regulate the volume of borrowing not through interest variations but through credit rationing; and stricter control depends on

inducing banks to reduce the borrowing limits granted to their regular customers. A rise in the money market rates is important in this connection only that it enables the batiks to earn higher rates of interest on their liquid investments and thus reduces the differential advantage, from a banking point

of view, of employing their funds in the form of advances to customers. Here, Kaldor emphasizes the credit channel through the supply of loans, not through demand as he would later argue. The argument is the same as Bemanke and Blinder’s: the central bank can influence the supply of loans either directly by reducing the supply of reserves, or through influencing the

274

Credit, Money and Production

relative rates of interest in such a way as to make lending a less attractive proposition for banks relative to ‘other investment’, i.e. the purchase of other assets. New Keynesians have attributed the development of their theory, along the lines argued above, to two points. First, the “progress in theory” and, second, to the theory’s “rigor” (see Gertler, 1988). Interestingly enough, it has taken them nearly four decades to achieve the same “progress” and “rigor” post-Keynesians reached in the late 1950s. Post-Keynesians have since developed their theory along new and interesting lines of argument, abandoning the confines of scarcity.

CONCLUSION The objectives of this chapter were essentially two-fold. First, it was to offer a critical and exhaustive examination of the contributions of New Keynesians in the field of money and monetary theory, and in particular with respect to its more current debates on financial innovations. Second, the chapter considered the feasibility of a grand synthesis of post— and New Keynesian ideas, as advocated lately by a number of post-Keynesians. The conclusions which can be drawn from the analysis above are the following. The emergence of New Keynesian economics appears, perhaps at first glance, as an interesting venue for post-Keynesians to explore further. However, claims that a synthesis is feasible arise out of a mistaken interpretation both of New Keynesian and post-Keynesian economics. Moreover, it was argued that while post-Keynesians attach great importance to the New Keynesian notions of asymmetric information and credit rationing, these are not essential elements in the derivation of New Keynesian conclusions. Lastly, it was noted that the notion of credit rationing aside, there are additional elements in New Keynesian economics which should warrant a closer examination by structuralist post-Keynesians. In particular, the whole discussion over the balance sheet channel and its close relations to Minsky (l957a, b), as well as the discussion over the merits of the bank lending channel given financial innovations, are all areas in which postKeynesians have a definite advantage, given the fact that these views have been part of the post-Keynesian literature for almost forty years now. It is important to note the historical differences as well. Post-Keynesians have pledged allegiance to Keynes, while New Keynesians share a greater affinity with Friedman. While post-Keynesians want to continue Keynes’s “revolution,” New Keynesians are more interested in the “reincarnation” of

New Keynesian Monetary Theory and the Transmission Mechanism

275

Keynes’s ideas, and their “rebirth into another body” (Mankiw, 1992, p. S60). Hence, Mankiw (1992, p. 560) declares unapologetically: To some old Keynesiatis, new Keynesians may be hard to recognize as Keynesian at all. New Keynesian econotnics may appear more similar to the classical economics of David Hume, or even to the monetarist economics of Milton Friedman. If new Keynesian economics is not a true representation of Keynes’s views, then so much the worse for Keynes.

On a final note, Mankiw (1992, p. 565) suggests that “perhaps we should conclude that the term ‘Keynesian’ has out-lived its usefulness.” It is difficult to make sense of claims by post-Keynesians that a dialogue between post— and New Keynesians is a worthwhile endeavor. For instance, how should the following quote by Fazzari and Variato (1994, p. 366) be interpreted by post-Keynesians: “The ‘new Keynesian’ approach shares methodological features with mainstream neoclassical research. [Nonetheless], the new models are more consistent with the views of post Keynesians than anything in mainstream macro for decades.” Hopefully, this chapter has put to rest this issue. And despite Pollin's (1995, p. 12) claim that there exist “significant similarities” between postKeynesians and New Keynesians, there appear to be far more differences, in methodology, theory, content, and policy which separate post-Keynesians from New Keynesians. At best, perhaps the only thing the two schools share in common is the fact that they are both united in their objective of criticizing New Classical economics and advocates of the real business cycle - although one approach rejects most of the foundations of New Classical economics, while the other embraces many of its assumptions. The bottom line is well summarized by Levratto (1992, p. 341): “These characteristics should not lead us to conclude hastily that these works are Keynesian since New Keynesian have also been influenced by the new classical macroeconomics and modem microeconomics as suggested by the uses of moral hazard, imperfect information or adverse selection.” If the stated objective of post-Keynesians is to pursue the more radical elements in Keynes thereby discarding the more orthodox elements, they must also abandon the more orthodox elements of their current approach to macroeconomics and monetary theory.

NOTES 1.

Davidson (1996) in fact does this brilliantly, comparing the three Keynesian schools. Unfortunately — and surprisingly — Davidson does not compare their views on money.

Credit, Money and Production In contrast, New classical econotnists would typically argue that New Keynesian econottiics, like Keynesian and even monetarist models, rests on an imperfectionist argument to generate short-run fluctuations in output. Note, however, that as households draw down their deposits to purchase securities, reserves may well fall, but deposits also fall, pari passu. Hence, the rate of interest need not rise since the demand for money. in the neoclassical sense, has also fallen: it has not remained unchanged. While a price, the rate of interest is rigid upward in New Keynesian theory, contrary to other prices which are rigid downward. According to Greenwald and Stiglitz (I993, p. 31): “Increasing interest rates may have adverse affects both on the mix of loan applicants and on the incentives of borrowers to undertake risky activities. and that these adverse incentive and selection effects can be so strong that lenders‘ expected retums may actually decrease as the interest rate change increases." Similarly, Stiglitz and Weiss (l99l, p. 247) argue that “as the interest rate rises, the average ‘riskiness’ of those who borrow increases, possibly lowering the banks‘ profits." Some New Keynesians have also argued that credit rationing can arise out of asymmetric information when borrowers also attempt to raise funds outside the banking industry, through the emission of new loans. For instance, according to Greenwald, Stiglitz and Weiss (I984), potential non-bank lenders must figure out whether the issue of new shares is a legitimate way of raising funds or an attempt to pass on to the public bad debts. The New Keynesian argument of moral hazard and adverse selection is very much in conflict with the overall argument of “identifiable” groups (Stiglitz and Weiss, I991. p. 249). According to New Keynesians, moral hazard occurs because banks charge a rate of interest which attracts "bad" lenders. To avoid this, they must set the rate at r*, and then ration credit. However, according to Stiglitz and Weiss (1991, p. 249), there exist “identifiable groups." If they are “identiftable“, then so must be all other groups, and the argument of credit rationing no longer holds. In other words, what was, in Minsky’s terminology, Hedge finance, now becomes Ponzi finance. Banks may also tighten other tertns of lending, such as collateral requirements and the size and maturity of loans. We could also assume that the argument holds for the purchase of capital goods by other capital-goods firms. As will be seen in the section below, the balance sheet channel is closely related to the work of Minsky on financial fragility. See van Ees, Kupet and Sterken, I996. Bemanke and Get1ler(l995) estimate that forty per cent of the decrease in profits can be attributed to increases in the short-run rate of interest. In this sense, it is important to keep in mind that according to New Keynesians (see Kahn, I991 — not to be confused with Richard Kahn), an increase in the supply of highpoweted money may not necessarily lead to an increase in loans, since banks may choose to allocate all — or part of — the excess reserves to the purchase of securities. Moore (I988) mistakenly assumes this to be consistent with the post-Keynesian tradition. Given recent financial innovations in the banking industry, Romer and Romer (I990) argue that the “availability” of funds is of less importance today given the removal of Regulation Q. large denomination Certificates of Deposits, and other “managed liabilities“. However, open market operations still have an effect by raising the cost of acquiring additional funds (Kashyap and Stein, 1994). As an interesting footnote, it should have become clear to the reader that these arguments are used by many post-Keynesians in defense of the endogeneity of the money supply. For New Keynesians and their “money” critics, however, money is still very much

New Keynesian Monetary Theory and the Transmission Mechanism

277

exogenous. Moore (I988) makes lines of credit the focus of his argument. Yet, the differences with New Keynesian monetary economics are not sufficient. That is why the

theory of the monetary circuit developed in chapter one makes for a better foundation. Morgan (1990) argues that only a third of loan commitments under $100,000 were approved, whereas loan commitments over $500,000 had a success rate of over seventy per cent. For Morgan. the size of the loan commitment is a proxy for the size of the firm. In correspondence, Bemanke argues that this fact does not alter, in any way, the overall framework of his model and still sees the money supply as essentially exogenous. This same approach is shared by Berger and Udell ( I992), and Stiglitz and Weiss (1991. p. 247), the latter having argued that “one means by which [banks] compete is their choice of a price (interest rate) which maximizes their profits.” Once again, however, post-Keynesians have emphasized credit rationing and asymmetric information as opposed to financial innovations or the endogenous nature of reserves. As argued in the previous chapter, the lack of accommodation does not necessarily imply a reduction in loans. However, it would be the only appropriate way of reconciling the endogenous nature of reserves with the New Keynesian position.

The position taken here is that asymmetric infomtation and fundamental uncertainty are not only different but also incompatible. However. it can be argued quite convincingly that banks and bank borrowers have different stocks of knowledge, different expectations and different behaviors when confronted with fundamental uncertainty. ln correspondence, Bemanke makes this claim precisely. According to him, the important argument is that the control of reserves. either exogenous and hence ex ante, will affect the supply of loans, but the decision of the central bank to refuse to fully accommodate reserves when they are endogenous will have the same effect.

Gertler (1988, p. 564) briefly mentions Minsky’s (1975) work on financial crises.

8. A Post-Keynesian/Circuitist Theory of Banks: Uncertainty, Creditwoithiness, and the Supply of

Credit The theory of the monetary circuit developed in chapter one is still incomplete. While it is perhaps the better foundation for a general theory of modem capitalist economies of production, in Keynes’s sense, there are nonetheless important constraints. The purpose of this chapter is to attempt to fill one of the voids of this approach by addressing the issue of a microeconomic theory of banks.

A CIRCUITIST THEORY OF BANKS Parguez (1996, 1997) is among the few circuit theorists who has analyzed the decision—making process of banks. Indeed, most circuit theorists have ignored this important element, a strange event given their emphasis on the hierarchical relationship inherent in the economic structure. In chapter six, it was argued that while the discussion of uncertainty in post-Keynesian theory is a welcomed addition to economic theory, the way in which post-Keynesians use the notion of uncertainty with respect to the theory of a monetized economy is limited to investment decisions and the demand for money, which includes, for many post-Keynesian, credit demand. There is very little discussion of uncertainty in the context of banking behavior. In fact, Screpanti (1997, p. 2) argues that “PostKeynesians, in this field, are more lacking than the neoclassical economists.” While post-Keynesians indeed lack in this area, so do circuitists. In this chapter, uncertainty will be placed at the very heart of the analysis of banks thereby undermining the role played by the price mechanism. Instead, it is the quantity which will adapt itself to the needs of trade. While this approach is consistent with the overall circuit approach, it will also be made consistent with the horizontalist approach. Also, while post-Keynesians do have a theory of banks in the form of the liquidity theory, it nonetheless poses serious problems, as it was examined 278

A Post-Keynesian/Circuitisr flreory ofBanks

279

in earlier chapters. In essence, it is in its structuralist incarnation that the post-Keynesian bank is passive, not in its horizontalist nature. According to the structuralist position, banks have no discretionary decision—making

abilities. They simply lend until a given, but never explained, liquidity ratio is met. The task of developing a circuitist theory of banks is, however, not an easy one since there is no systematic discussion of banking behavior in circuit theory. In doing so, this section will “borrow” some aspects of other approaches which can fit within the general framework of the monetary circuit. It is here, it is argued, that many of the insights of post-Keynesian

theory and indeed of New Keynesian economics can come together within the context of banking analysis under Keynesian — or what Screpanti (1997, p. 4) calls “generic” — uncertainty. Emphasizing uncertainty and bank behavior also serves to understand that the limitations on production and

investment come from the restrictions placed upon credit rather than technology, endowment and preferences (Fazzari and Variato, 1994, p. 351).‘ Two issues will be analyzed below. The first concerns the supply of credit; the second the rate of interest on loans. These issues are not unrelated. This does not imply that there exists a positive, causal relationship between the two, as advocated by some proponents of orthodox economics and post-Keynesian theory. In fact, the rate of interest and the supply of credit will be determined by difierent factors altogether. Whereas the quantity of credit a bank is prepared to supply is determined by the rules of creditworthiness and uncertainty, the rate of interest will be determined as a mark-up over the exogenous, central bank rate, with

respect to the robustness of the creditworthiness of bank borrowers. Here, Kalecki’s principle of increasing risk may be added to the analysis.

This explanation is in many respects more Keynesian than the explanation offered by some post-Keynesians: it is related to the role uncertainty plays in banking theory, and how it affects the decision of banks to determine the interest rate on loans and the quantity of credit which they supply. It is thus not related to the liquidity preference of banks or to profit maximizing

behavior. It is assumed that banks will seek to grant credit to all creditworthy customers. One of the elements at the core of circuit analysis is the relationship between the various macro—groups. Among these, the relationship between the firms and the banks features prominently. While firms are responsible

for production, banks are responsible for financing this production. Without this financing, firms are unable to begin production, and the flow of income will be interrupted. As Arena (1996, p. 20) writes, “The most important economic constraint is obviously the nature of the contractual

relations between banks and firms.”

280

Credit, Money and Production

But just as firms face an uncertain future, banks must defacto face this same future, unknown and unknowable. Post-Keynesians attribute uncertainty to fim1s and households, not to banks. But Keynes’s definition of uncertainty must apply to all economic agents, but in particular, it must apply to banks above all since they are the oil by which production is kept well—greased. Typically, banks will face two different sources of uncertainty. On the one hand, banks face uncertainty as to their evaluation of the borrower based on the presence of asymmetric information, although this is perhaps not the best expression to use. Have the banks correctly evaluated the potential borrower with respect to its current creditworthiness and to its ability to generate an appropriate level of profits? This exemplifies the crucial relationship between banks and firms. Although banks will spend a considerable amount of time and energy in developing close relationships with their customers in order to accumulate as much information as they can, banks can never be certain. To overcome any possible uncertainty, banks will in fact typically assign an account manager, loans officer or branch manager to specific accounts such that each borrower is carefully looked after. In these situations, banks are able to collect infomration regarding the borrower, but also about the on-going success of the firm. The stability of the relationship between the bank and its customers is paramount. It is conceivable for fmns and banks to possess different infomration regarding investment projects, although banks do invest a considerable amount of time and energy in evaluating the projects. In this respect, banks will try to determine the creditworthiness of individual firms. Keynes (vi, p. 365) argues along very similar lines: “The amount lent to any individual [is] govemed not solely by the security and the rate of interest offered, but also by reference to the borrower’s purposes and his standing with the bank as a valuable or influential client.” Because of uncertainty, this will only be a “bet” — to borrow Parguez’s expression - which banks place on individual fim1s. In evaluating frms, banks will impose a set of constraints on them, such as norms of profitability (the mark-up of firms analyzed before), and the rate of interest. Bank constraints are therefore binding on firms and will affect the pricing of the firm’s products since the rate of interest banks charge will affect the costs of production of the firm. On the other hand, banks face their own uncertainty regarding the unknown future. Banks, just like firms, cannot know what the future will look like, and how it will impact on firms. In other words, banks must in a first instance place a “bet” on their evaluation of the borrower, and then place a second “bet” on the future of the business cycle. Both “bets” are unrelated: irrespective of the creditworthiness of bank borrowers, the unknown future may play havoc on firms and effective demand. Lavoie

A Post-Keynesian/Circuitist Theory ofBanks

28l

(1992, p. 106) emphasizes these two sources of uncertainty facing banks, and writes that “the uncertainty about the future, as well as the lack of relevant knowledge about the competence of the managerial team and about the profitability of the project, forces bankers to rely on the performance record of the past, that is the profits generated in the past by the frnn.” These two forms of uncertainty can be labeled, in post— Keynesian language, as borrower’s and lender’s risk. It becomes clear, therefore, that in a proper theory of banking behavior, both borrowers and banks are under the influence of uncertainty. According to Wolfson (1996, p. 450), “The most appropriate post Keynesian assumption is that both borrower and lender are subject to fundamental uncertainty about the future” (see also Heise, 1992, p. 295). The unknown future must affect all agents alike. A theory of banks should therefore include the role of uncertainty. The role of banks therefore is not to “pool risk" but rather to decide whether to lend to a potential borrower given the unknown future. Faced with uncertainty, banks will generally rely on a number of rules of thumb and conventions. As Keynes (vii, pp. 148-52) emphasized in the General Theory, conventions and rules of thumb are a good way of dealing with uncertainty. In the face of an unknowable future, for instance, relying on others may be an appropriate behavior. Uncertain how to act, banks may

well tum to their competitors and observe how they set rates, for example, or how they determine how much credit to supply. They may well adjust their behavior accordingly. For instance, if a bank observes that it is supplying too much credit, it may decide to curtail its activities in order to “move in step” with the other banks. This may also be done because banks

which are perceived to be extending too much credit may face their own creditworthiness being questioned. This point was made by Lavoie (I996, pp. 288—9): If a single bank is pursuing aggressive lending policies, as it could in the present environment of generalized liability management, its creditworthiness

may become questionable and other financial institutions may refuse to lend to it.

The main objective limit to the expansion of a bank is the rate of

expansion of the other banks and the norms of prudent and acceptable behaviour self imposed by the banking industry.

In this context, as Keynes (v, p. 26) tells us, “There is no limit to the amount of bank—money which the banks can safely create, provided they move forward in step.”2 This position is defended as well by Moore (l989b, p. 15) who writes that “consequently, providing a bank just keeps pace with the rate of loan extension of its competitors, no net outflow of funds at clearing need result.” This behavior, as pointed out by Bamerjee (1992, p. 798) is perfectly rational: “Paying heed to what everyone else is

282

Credit, Money and Production

doing is rational because their decisions may reflect information that they have and we do not.” This aspect of bank behavior can easily be reconciled with Keynesian uncertainty. As Lavoie (1984, p. 791) explains: A bank gets into trouble only if its spirits are too high, that is if it lends more than its rivals. When bankers begin losing some of their high “animal spirits," though they are aware of the fact that their new behavior will harm the economy, they prefer to restrain their creation of credit money. But “moving in step” or relying on the behavior of other banks tells us nothing about how banks, in toto, deal with uncertainty. Even if we claim that banks follow the lead of another bank, this herd behavior, although perhaps explaining the behavior of all other banks, tells us little about the

way the “industry leader” behaves and sets policy. Herd behavior is, however, not incompatible with the analysis presented here. There is still a need to explain the role of uncertainty on the individual bank’s decision making. How does uncertainty affect the supply of bank credit of an individual bank? Overall, it is argued here that the best way banks have at their disposal is

not to restrain credit arbitrarily, or to observe how much credit is supplied by other banks, but rather to set up guidelines by which credit will be

supplied. This, banks can do by determining a set of creditworthiness criteria. Hence, in the face of Keynesian uncertainty, the banks must decide not on how much credit to extend, but rather on the minimum conditions under which they will accept to supply credit. Once these criteria have been determined, banks will lend to anybody who meets these criteria. It is these criteria that will fluctuate with the banks’ “animal spirits.” In circuit theory, cormnercial banks are credit—creating institutions aiming to make profits. Their decisions to meet credit demand will be influenced by a target — i.e., anticipated — rate of retum, but also by their desire to meet their own creditworthiness imposed by the stockholders of the banks. Because of uncertainty, banks cannot simply “ration” credit at r"' corresponding to profit maximization, as in New Keynesian economics. Not only because it is impossible to calculate r* (Wolfson, 1996), but also because with such an unknown future, the bank cannot predict how the cycle will affect the ability of firms to generate profits and reimburse their debt. Hence, if r"' represents the ex ante rate at which the bank will supply q* worth of credit, the bank can still not be certain that q* will ex post be the amount which will maximize profits. Some borrowers will default, others will not be able to fully honor their agreement. In this case, profits will fall. Such a policy is therefore not only impractical, but also unworkable.

A Post—Keynesiarr/Circuitist Theory ofBanks

283

Having established the fact that uncertainty is just as important to banks as it is to frrrns (investment decisions) and households (portfolio theory), we are still left with the best way of incorporating it into model building. How do banks deal with uncertainty, and how do they tum their expectations of the future into banking policy? While Keynes (xiv, p. 124, 114) argues that “peace and comfort of mind require that we should hide fi'om ourselves how little we foresee,” he also writes that “practical men” must “overlook this awkward fact” and “save face.” Banks have at their disposal essentially two tools to deal with uncertainty:

creditworthiness ratings and the rate of interest which they charge on bank loans. These issues, as discussed above, are related — both in static analysis as in a dynamic setting, but are not causal. This suggests that banks can extend as much credit without it affecting the rate of interest. Banks will operate according to an expected target rate of retum. But they will only be able to find out whether they have met this targeted rate at the end of the period of circulation. Only then will banks be able to compare their effective rate of return with their targeted, or expected, rate of retum. This comparison, as will be seen below, will contain important information for the bank in setting policy for the next period.

In behaving according to a target rate of return, banks will therefore extend as much credit as possible, hoping that in the process, they will meet their targeted rate, although they will have no guarantee. The bank will therefore not stop supplying credit once it believes it has reached its expected target rate of retum - i.e., arbitrarily. There can therefore never be “too much” credit supplied, or “too little.” There can only be too many “bad borrowers,” or similarly, too few good borrowers. This analysis does not imply that banks will extend credit to anybody who demands credit. It is perfectly within their profit-seeking activities to curtail their supply. Banks will indeed turn down applicants. This procedure requires two separate steps and different uses of creditworthiness. Banks will set up a set of minimum creditworthiness criteria according to which they will judge potential borrowers. As long as applicants meet these criteria, banks will find it profitable to extend loans. -As Wolfson

(1996, p. 455) argues, “Bankers accommodate all creditworthy demands for credit, and ration those demands not deemed creditworthy” — which the author calls the “efl‘ective" demand for credit.’ Lavoie (1996) has called this demand, “solvent demand,” although perhaps a better term would simply be to call this the “creditworthy demand.” Howells (1995, p. 90) argues that “banks set their collateral standards and their lending rates (the latter as a markup on central bank lender of last resort) and then meet all loan requests that are forthcoming."

Graphically, we may want to assume that the demand curve drawn in interest rate/credit space is the creditworthy demand for credit which

284

Credit, Money and Production

excluded those bank borrowers who would not meet the bank’s creditworthy criteria. Two issues need to be differentiated. On the one hand, what determines the “base” set of creditworthiness criteria established by the banks; and second, how do banks determine the creditworthiness of the bank customer. These are two separate steps. The first is influenced largely by uncertainty. Given the unknown future, banks will determine what they believe are the minimum requirements needed for frrrns to receive a loan. These may be different depending on the type of the loan (for instance, a new loan versus a renewal; working capital versus fixed capital; the reputation of the borrower and his relationship with the bank), but will generally be determined by the banks’ expectations of the future. This is

the hanks’ “bet” on the future. As the banks become more pessimistic about the future, they may fear that firms will have difficulty in meeting their fmancial agreements with the banks and thus not be able to pay back their initial fmance. Banks will raise their minimum creditworthiness criteria while still wanting to extend credit to those who meet their new, stricter, guidelines. In more pessimistic times, firms will therefore need to meet these new criteria, possibly by attempting to raise more capital or collateral. In a changing “macrofmancial environment” - to use Keynes’s (vii, p. 154) expression — banks’ bets on the future will necessarily change, along with their creditworthiness criteria. As the future becomes bleaker, banks

will require, for instance, greater collateral to secure a loan. The second step will be the banks’ “bet” on the frm by detemiining the creditworthiness of the frrrn itself. This issue was examined previously and does not require much elaboration. The creditworthiness of any particular firm can be evaluated in particular by the past performance of the firm and its on-going relationship with the bank. Provided the firm’s creditworthiness is greater than this “base” set of creditworthiness criteria — i.e., the robustness of the creditworthiness of the firm — then the firm is guaranteed a loan. This simply means that from the point of view of the bank, the fimi is thought to be able to repay its loan, based on the bank’s expectation of the unknown future. But how will banks judge the future? This is a difficult question where

there are no simple answers. Post-Keynesians have been asking for many years now how best to represent uncertainty in model building.

A possible solution is the one presented here. Banks may be able to gauge the future by comparing their effective and target rates of retum. By the end of the period, banks have the ability to compare their targeted rate of return with their effective realized rate. If the two rates coincide, then banks interpret this to mean that the economy is perfonning as predicted. The banks’ bets would have therefore been correct. This will suggest that they will “continue as usual."

A Post-Keynesian/Circuitist Theory ofBanks

285

If the effective rate is lower, however, probably as a result of firms being unable to completely reimburse their loans, then banks may interpret this to mean that their bets on both the firms and the future were not accurate and

must now be revised. They will interpret this to further mean that the economy is not performing as well as predicted, and may therefore change their expectations of the future accordingly. Their initial bets were too optimistic. Banks will become more pessimistic, and as a result raise their creditworthiness criteria. Banks will change their base set of creditworthiness criteria to compensate for changing animal spirits. As uncertainty grows, banks will typically raise the base set of criteria, thereby eliminating a number of potential borrowers in the process who no longer meet their new creditworthiness criteria. This is quite natural: in the face of greater uncertainty, banks will require their borrowers to be more creditworthy.

This could be achieved notably by requesting higher collateral requirements. Through the cycle, then, creditworthiness criteria will vary. For instance, at the beginning of a cycle when animal spirits are high, creditworthiness criteria will be lower and banks will seek out potential borrowers. As the economy grows, banks may fear an eventual downturn, and may become more pessimistic about the future, and raise the minimum required criteria. In this way, as Wolfson (I996, p. 459) claims, “Tighter standards imply a direct fonn of credit rationing.” This thus constitutes a

non-price based theory of credit constraints. As banks become more pessimistic and raise their creditworthiness criteria, this in tum will translate into a decrease in the amount of credit supplied, but not because of a fall in the “supply of credit” due to an inward shifi of the supply frmction, but rather because of a fall in the number of firms qualifying for credit at the banks’ new, higher, criteria. The fmancing of production which was extended to previous borrowers

may not be renewed at these higher standards. As uncertainty grows and creditworthiness criteria are raised, existing clients of the banks may fmd themselves, once reviewed, without fmancing. Hence, fmancing is not guaranteed from one period to another. When uncertainty is high, banks will begin to constrain credit, not because of a given supply of “loanable frmds,” but rather out of fear of a generalized fall in profits. Firms can guard themselves against the animal spirits of banks by raising their collateral to meet the banks’ changing creditworthiness criteria. Firms will also want to meet the minimum rate of return imposed by banks as an indication of their creditworthiness. As Parguez (1997, p. 12) explains, “r“ is the proof required by banks of firms’ forecasted ability to eam an own profit out of monetary costs expenditures.” This requires not only a theory of profits, but also a theory of the growth in demand thereby enabling fums to increase their collateral and capital value through time. This point is, however, beyond the scope of the present work.

286

Credit, Money and Production

One can fmd in the post-Keynesian literature some bits and pieces of the views presented above, although these are few and not very developed. For instance, Dow (1996, p. 499) writes: The “moods” of the financial institutions may become pessimistic, or display reduced confidence in prediction. Then borrowers who had previously been acceptable may find the value of their collateral and projections of future income streams reduced accordingly to the assessment by financial institutions. The borrowers themselves may perceive no change in their own assessment of their creditworthiness. Here, Dow speaks of several points: the role of banks in credit-supply, their importance in tenns of evaluating firms’ creditworthiness, but also of the possibility of “asymmetric expectations.” There is thus no need for the rate of interest to rise with the demand for

credit. The rate of interest may vary, although this is not necessary, or even automatic. By raising the creditworthiness criteria, banks feel less of a

need to raise the rate of interest. In this way, the causality between increased demand for credit and the rate of interest, invest in orthodox theory and some versions of post-Keynesian theory is broken. If it does rise, it is best represented as an upward shift in the horizontal supply curve.

At the new interest rate, given the new established criteria for creditworthiness, banks will meet all credit demands of those borrowers who meet the criteria. The position taken here is therefore in contrast to the position held by Heise (1992) and Wray (1992), and other post-Keynesians. In general, they would argue that banks will stop lending when the ratio of assets to liabilities (or loans to deposits) reaches a certain level. For instance, Heise (1992, p. 292) writes: “Hence, it is the unwillingness of commercial banks to get indebted over and above this ratio [reserves to liabilities], rather than the lack of profit promising investment projects, that will ultimately govem

the quantity of money.” Similarly, Wray (I992, p. 303) writes that “even if some agents are still willing to borrow and some banks are willing to meet this by expanding their balance sheets, these banks will find it increasingly difficult to meet this demand for credit.”‘

Two criticisms are raised here. First, these authors never explain what the actual level of the ratio should be. Banks stop lending at a certain ratio, but this ratio remains unexplained. This makes the liquidity theory approach highly impractical. Second, this post-Keynesian position undennines the role of uncertainty. Households and frms, in their respective decisions to hold on to money or to invest, are influenced by Keynesian uncertainty. The situation appears to be different for banks, who are assumed to operate within a certain

A Post-Keynesian/Circuitist Theory ofBanks

287

enviromnent where uncertainty appears to play no role. Banks do not “bet” on the future as do firms. In a way, this post-Keynesian analysis of banks is “unKeynesian” — banks operate given the structure of the balance sheets with no regard to the future.

Credit “constraining” can therefore arise not when the supply of credit is constrained, but rather when banks’ animal spirits are weaker than those of the borrowers. It is conceivable, and in fact very probable, that banks and their borrowers do not see eye to eye on their respective expectations of the

future. Hence, instead of talking about asymmetric infornration, between banks and fums with respect to the creditworthiness of firms, it is perhaps better to emphasize, as do Wolfson (1996) and van Ees and Garretsen (1993), the existence of “asymmetric expectations.” According to Wolfson (1996, p. 451), “All that is necessary for a theory of credit rationing based on Keynesian rmcertainty is that borrower and lender evaluate the future difierently — that is, that they have asymmetric expectations.” When banks and borrowers share the same expectations, then no problems arise. The amormt of credit demanded will be equal to what the

bank is willing to supply. When expectations of both fall in face of rising uncertainty, supply will fall pari passu with demand. There is thus no credit constraint. However, the most likely situation is when expectations

difier, where there may arise a situation of credit constraint, provided of course that the banks’ animal spirits are weaker than those of the firms.

Hence, banks fear that if their expectations of the future are more accurate than their borrowers’, then supplying greater credit, especially when only one bank does it, can translate into borrowers not being able to reimburse the banks.

How important are credit constraints?

Based on Canadian data, it

appears, however, that credit constraints play a less prominent role in banking decisions. In fact, banks are eager to meet credit demand. Overall,

in Canada, more than 84% of credit demand has been met by commercial banks, with more than three—quarters of these in full. The story, however, does not end there. Of those applicants who were tumed down initially, a third were asked to resubmit their application for credit. Initial rejection might have tumed on a lack of information. Banks will ask for additional information and ask applicants to resubmit when the information is forthcoming. This is especially the case of small and medium businesses,

and those who apply for the first time. In general, as argued above, banks will rely on the past performance of firms.

But if the bank has no

information on the frnns, then granting credit may become a problem. The conclusion, however, is clear: banks want to meet credit demand. Of those applicants who resubmitted, 87% had their request accepted.

The report (Canadian Bankers Association, I996, p. I6) concludes that “chances of being approved are excellent. The vast majority of these loan requests are approved on initial application. Not only is the initial

288

Credit, Money and Production

acceptance rate high, but when the bank asks the applicant to provide more information or to re—work the loan, the approval rate is even better.” Liquidity ratios appear to play a role. Moreover, the accumulated debts of fimis at the end of each period may also play a role not only by influencing the banks’ animal spirits, but also by contributing to help change the banks’ set of creditworthiness criteria. As discussed in earlier chapters, at the end of each period, the fimis’ remaining debt is refinanced using the money hoarded by households and placed in banks as deposits.’ This may work in two ways. At the micro level, if a firm becomes increasingly indebted toward the bank, the latter may become fearful that the firm may soon become unable to continue its operations. In this way, the bank’s bet on the firm becomes less hopeful. This would be consistent, for instance, with Minsky’s analysis of Ponzi finance. At the macro level, the accumulated debt of all firms may also trigger banks to become more pessimistic, and hence raise their creditworthiness criteria. As these grow, banks may become fearful that too much accumulated debt will slow down the economy. Hence, the banks’ bet on the future, as opposed to on an individual firm, will become pessimistic. Moreover, the central bank may also view the macr0—accumulation of debt as inflationary, as firms may raise prices in an effort to recapture some proceeds. The central bank may hence decide to increase the rate of

interest. Banks would then raise their own rates on loans. The discussion above emphasized the creditworthiness argument of banks and firms, and had little to say about how the rate of interest on loans was determined. The discussion now tums to this issue. To begin, circuitists generally accept the exogeneity of the rate of interest, determined by the central bank, thereby breaking with the natural rate hypothesis, but also with Keynes’s theory of liquidity preference as a detemiinant of the rate of interest. The supply of credit is neither vertical nor upward-sloping. For instance, Parguez (1996, p. 183) writes: Being perfectly exogenous relative to the requirements of full employment, the rate of interest is not the outcome of a pure whimsical power. Central bankers are spontaneously led to believe that the consistency of the credit network requires the perfect stability of the value of the money.

Once established, commercial banks will set their loan rate within a given range, based on a mark-up formula over the base rate. This mark-up will be influenced notably by the robustness of the creditworthiness of bank borrowers, that is how much “risk” does the borrower represent to the bank. It is the robustness of this creditworthiness which will determine the rate of interest at which banks will extend the credit demanded: how much

A Post-Keynesian/Circuitist Theory ofBanks

289

collateral they have, and other fmancial infomiation related to the past performance of the firm. In particular, the size of the firm is important (the size of its capital value). Banks do not set the rate of interest in order to discriminate or to exclude potential borrowers. Rather, banks determine the rate of interest on loans as a mark-up over the base rate in order to generate their own profits in order to satisfy their “creditors,” the shareholders of the banks.

There is not a single rate of interest on loans, but rather a certain range within which banks have the discretion to set the rate. Where within that range the rate will be set will depend on a number of other factors, or the robustness of an individual borrower’s creditworthiness. In other words,

the highest rate within the given range will correspond to the strictest minimum banks are willing to accept as creditworthy.

The greater the

value of a borrower’s collateral, the lower will be the rate within the given range. This is an application of Kalecki’s principle of increasing risk as applied to an individual firm, analyzed earlier. There is, however, no proof that as the supply of loans in toto increases, the economy becomes more fragile.

To evaluate the creditworthiness of potential borrowers, banks will consider foremost the collateral of credit demanders, keeping in mind that credit also creates its own collateral. In this respect, banks will typically

estimate the value — or rather the increase in the value — of the firms’ capital. Other factors will also come into play, however. Among the most important ones, banks will also look at the management and business experience of the owners, the number of full—tirne employees, the past sales volume of the firm and how they stand relative to given target sales figures,

and the education of owners and managers. These elements are the best way for banks to acquire some of the vital information regarding the creditworthiness of the borrowers. In particular, past sales volume (and revenue) will allow banks to detemiine whether firms are able to meet their

credit payments on a monthly basis. Overall, the report of the Canadian Bankers Association (1996) indicates that the acceptance rate of loan demand generally increases with all of these criteria.

Banks may also evaluate the creditworthiness of their borrowers by looking at their past perfomiances, in terms of their ability to make payments, but also in terms of their ability to meet their sales objectives. In this respect, Wolfson (1996, p. 453) claims that “borrowers who have a history of repaying loans on time and maintaining a strong fmancial condition will be preferred.” Dow (1996, p. 500) argues along similar lines: “Under conditions of uncertainty (unquantifiable risk), the potential lender must fall back on habit, conventional judgment and assuming the future replicates the past.” Among circulationists, Parguez (1996, p. 159) elaborates on this point:

290

Credit, Money and Production

Past performances relative to the firms’ commitments are data that could influence banks’ new set of bets on the future of finn x. For the same firm there is a level of the capital value it should reach because this would convince the banks of the perfect ability of the fim1’s management to meet their commitments whatever the increase in their current demand for liquid fund. Banks would thus consider this required capital value V,' as the required collateral for their loans; firms are thus endeavouring to reach their target capital values that should protect them from any credit rationing or increase in the cost of credit.

Another criterion in the evaluation of creditworthiness of borrowers is the

reliance of the firm on bank credit, what Kashyap and Stein (1994) have called “locking in.” As a firm grows dependent on the bank, the bank is then able to better monitor the financial condition of the borrower. If finns are fmanced by a number of institutions, individual banks will be less able to collect the necessary information, although the borrowers themselves may have more information. This could be interpreted as a situation of “asymmetric infomiation.” The Canadian Bankers Association report (1996, p. l0) indicates, “The greater the concentration of business debt with banks the higher the approval rate." As the “macrofmancial enviromnent” changes, this may also affect the

creditworthiness of bank borrowers, i.e., of firms. For instance, if the central bank changes its policy and raises interest rates because of inflationary fears, this will force banks to raise all of their rates. In turn, an increased rate may force frnns to move into Ponzi fmancing, thereby lowering their creditworthiness. This part would be consistent with Minsky’s analysis but also, as seen in the previous chapter, with New Keynesian economics. As Parguez (1985, p. 272) writes, “The more they [banks] fear the future, the more they mistrust the industry, the more they worry over the ‘public’s’ disapproval, the more they raise the rent on contracts to obtain the money to spend." Hence, the rate of interest on loans will rise with increased uncertainty. Interest rates, however, do not rise because of increased demand, but rather because of increased uncertainty on the part of banks. The best graphical representation — if one wishes to stay within the confmes of “standard nomialcy” (Lavoie, 1992) — is to represent the supply of credit as horizontal, or rather, as Kaldor (1982) explained, as a “set of horizontal lines” moving up and down through the cycle. At every rate of interest, once the creditworthiness criteria have been detemiined, banks will supply as much credit as is demanded as long as all creditworthiness conditions are mCt.

A Post—Keynesiarr/Circuitist Theory ofBanks

291

MONETARY CIRCUIT AND LIQUIDITY PREFERENCE Before closing this chapter, it is perhaps appropriate to discuss one more issue: the appropriate role of liquidity preference in the theory of the monetary circuit. Some post-Keynesians argue that there is no place for liquidity preference in the circuit approach. Liquidity preference is perfectly compatible with the circuitists’ interpretation of credit, money and endogeneity. Also, in as much as horizontalism shares a great many insights with the circulation approach, this chapter will show ipso facto that horizontalism is compatible with liquidity preference (Lavoie 1992, 1996). Dow and Dow (I989, p. 147) have expressed this concern in the

following way: “implicitly or explicitly, adoption of an endogenous credit analysis is ofien taken to preclude the significance of liquidity preference.” Dow (1996, p. 41; see also Chick, 1995) has also later claimed that “there has been some divergence of interest between those who focused on the theory of liquidity preference and those who focused on endogenous money." Cottrell (I994, p. 598) has claimed that horizontalism “entails the

abandonment of Keynes’s liquidity preference theory of interest.” In each case, however, these statements prove to be misleading. It will be shown here, and indeed it has been demonstrated earlier, that liquidity preference and endogenous money can certainly be “reconciled.” Both approaches really address the same concems and issues. As for Cottrell’s position, it is rather an odd one. That horizontalism, the

theory of the monetary circuit, or any theory which sees the base rate of interest as exogenous, is inconsistent with Keynes’s defmition of the determination of the rate of interest has been aclmowledged earlier. It was shown that properly interpreted, Keynes’s finance motive, as Keynes

himself slowly came to realize, is a repudiation of his liquidity preference theory of the (base) rate of interest (see Rochon, 1997). This position should not surprise many post-Keynesians who no longer see Keynes’s narrow approach (in the General Theory) as valid. For instance, Dow and

Dow (1989, p. 188) took this position almost a decade ago: This apparent rejection by post-Keynesians of liquidity preference theory is understandable if “liquidity preference” refers to the demand for non—interest bearing money only. Then liquidity preference determines the rate of interest on interest—-bearing money. This narrow notion of liquidity preference is not of much interest, since money as a store of value falls outside its scope. As they (1989, p. 162) further explain, Keynes “intended much more by his theory of liquidity preference than the simple, separable demand for money

fimction.” Post—Keynesians have for some time now abandoned Keynes’s theory of liquidity preference as defmed in the General Theory, in favor of

292

Credit, Money and Production

a more “generalized” version. This does not imply that liquidity preference is not a useful concept, but rather that it must be reinterpreted in a way which is consistent with the analysis presented here. Before discussing the nature of this more “generalized” version, a few points are needed in order to clarify the issue at hand. We need to discuss liquidity preference from the point of view of both banks and households; And while banks’ liquidity preference arises at the beginning of the circuit, households’ liquidity preference arises at the very

end of the circuit, and concems a decision by households on how best to allocate their saving, or to reallocate their overall portfolio. Below, both facets of liquidity preference are addressed, beginning with the liquidity preference of households. The decision of households with respect to liquidity preference is not between saving and consumption, but rather on how to divide their saving between hoards and other financial assets. Households either consume or save. But, out of their saving, they either hold hoarded saving or fmancial saving. Their liquidity preference, expressed at the end of the monetary circuit, is a decision to hold money or other financial assets in their portfolios. Arena (1996, p. l8) explains incorrectly that “if wage—eamers spend entirely their income on consumer goods and/or assets, they reveal

no liquidity preference.” Liquidity preference arises therefore out of the creation of income. It is thus a decision on how best to allocate saving between hoards and other assets. It is a portfolio decision. Should households hold more liquid money, or should they hold instead more securities, bonds, shares, or other instruments of saving? Eiclmer (1987, pp. 819-20) also held this position, in what is called here a “lexicographic” theory of saving: Each nonfinancial entity, it can be assumed, will determine from experience

how large a cash balance it needs to maintain either as a checkable deposit in some bank or as currency in hand, so that its ability to continue making

payments will not be impaired. This minimal cash balance will be the entity’s first priority, insofar as holding financial assets is concemed. However, once

this minimal cash balance is established (or restored, following a net outflow of funds), the entity will prefer to hold other types of assets. This is because little or no interest can be eamed on whatever funds are held as a checkable deposit or as currency. Any net increase in' savings, then, assuming the nonfinancial entity already has the minimal cash balance it needs, can be expected to lead, after the initial increase in its bank balance, to a portfolio

shift, with the nonfinancial entity using the excess cash to acquire other types of financial assets.

Irrespective of the factors motivating these decisions, it is clear that liquidity preference is a decision with respect to saving. It is not a decision to consmne less in order to save more. This view was expressed by Mott

A Post-Keynesian/Circuitisl Theory ofBanks

293

(l985—6, p. 224): “Liquidity preference here refers to the desire to hold bills versus bonds and shares.” Oddly, however, post-Keynesians have at many times interpreted liquidity preference as synonymous with decreased consumption. For instance, Dow and Dow (1989, p. 151) argue that “the dire consequence for frrrns of a rise in liquidity preference which discouraged consumption, then,

would be an unplanned increase in stocks." Wray (1992, p. 303) holds a similar position: “Rising liquidity preference is very nearly the opposite of money demand. When liquidity preference rises, most agents are trying to reduce outstanding debt by cutting spending and selling assets. In contrast, rising money demand represents a willingness to expand balance sheets by issuing new debt.” In these passages, the authors confuse liquidity preference with a fall in the marginal propensity to consume, i.e., “cutting spending.” They argue that a decision to increase liquidity preference necessarily implies a fall in consumption. But while liquidity preference can indeed rise afier a fall in consumption, this in no way is certain. In fact, afier a fall in consumption, households may decide to hold on to more illiquid assets (for whatever

reasons). In this case, a fall in consumption would be accompanied by a fall in liquidity preference. Liquidity preference has nothing to do with

consmnption. Increased liquidity preference does imply, however, that households stand ready to sell existing assets in order to get into money or

less liquid assets. Unlike Wray, therefore, rising liquidity preference is taken here to mean precisely an increase in the demand for money as households prefer holding hoarded money to other forms of saving. Wray’s opposite position is the result of confusion between credit and money, as discussed in chapter two. If credit is not a component of money — i.e., if credit and money are clearly differentiated, as they should be — then a rise in liquidity preference is meant as a preference for liquid assets, of which money is the most liquid. Households would sell off their less liquid assets (bonds, shares,

securities, etc.) in order to increase their hoards or reduce their debt. This was shown in chapter one where it was argued that liquidity preference is equivalent to the stock demand for money, which in tum represents the outstanding debt of firms in the economy. Also, a rise in the liquidity preference of households will be met by an increase in the supply of money, although it will not be met by an increase

in the flow—supply of credit. This position, once again, is contrary to Wray (I996, p. 458): “When the preference for liquidity rises, it is unlikely that this will be met by an increase in the flow-supply of money." It is therefore little wonder that confusion arises in post-Keynesian theory

since there is substantial confusion on the very basic concepts used in the analysis and discussion. It corresponds to the outstanding debt of firms,

294

Credit, Money and Production

that part which firms will have to renegotiate with the banks along more long—terrn conditions. With the liquidity preference of households now clearly defmed, it is perhaps appropriate to ask what role it plays within the circulation. As

stated in chapter one, it cannot determine or influence the base rate of interest, as is the case in the standard IS-LM model. This rate is set exogenously by the central bank. However, if liquidity preference is

understood to be a theory of saving and assets, then it could be understood to be a theory of asset pricing. As households decide to spend a greater portion of their saving in acquiring assets, the price of these various assets will increase thereby lowering their yield. The theory of liquidity preference thus becomes a “mark—up” theory determining the “spread” between the base rate and all other rates (Wells, 1983). As Lavoie (1994, p. 20) explains, “Rather liquidity preference detennines the differential between the base rate and all other rates.” Mott (l985—6, p. 224) concurs:

“Liquidity preference

would have to do with the level of the liquidity

premium in the term structure of interest rates.” As Carvalho (I995, p. 24)

reminds us, Robinson (1951) was one of the first to attempt to extend Keynes’s two—asset theory of liquidity preference to a theory of asset pricing. On this issue, there now appears to be a large consensus among

circuitists, horizontalists and structuralists. For instance, take the following passage by Dow and Dow (1989, p. 148): Liquidity preference, then, in practice detennines the difference between the interest rate on liquid deposits [money] and on less liquid substitutes [securities]. The monetary authorities set the rate at the short-term end of the spectrum; liquidity preference (along with other considerations) detennines the mark-up to long-term rates. Indeed the liquidity preference concept can be expressed in its broadest fomi as a preference for a liquid asset over any illiquid assets.

In this sense, and contrary to Cottrell’s (l994, p. 598) views about horizontalism, liquidity preference does not become “at best a complicated means of determining the value of a mere residual devoid of any causal significance — the nominal stock of money.” ° There is, however, an important problem with this approach to liquidity preference - despite the rather large consensus. As households purchase

assets, it is argued that the price of these assets will rise, thereby reducing their yield. This assertion only holds of course if the supply of these other

assets is less than the required demand. If firms increase fully their supply to match pari passu the demand for securities, then the price of new securities will remain fixed. Therefore, an important assumption is then that the endogenous supply of securities is less than perfectly elastic.

A Post—Keynesiarr/Circuitist flreory ofBanks

295

With respect to the other dimension to liquidity preference — that of banks — post-Keynesians have argued in virtually all of their attempts to discuss banking behavior that the liquidity preference of banks must be taken into consideration (Wray, 1992). If post—Keynesians reduce the decision of households to a choice between consumption and saving, they reduce the decision of banks to choosing between lending to potential borrowers or purchasing other assets.

The banks’ decision is made with respect to

profit-seeking (maximizing or ‘satisficing’) behavior. According to this argument, when banks are pessimistic, their liquidity preference increases. As a result, they will reduce their supply of loans — since the “risk” of

banlauptcy is too great — in order to increase their purchases of other assets (such as bonds or securities). Here, too, there is a certain problem with the post-Keynesian approach to banks’ liquidity preference. The problem with this logic, as is explained below, is that post-Keynesians are attempting to occupy simultaneously two extreme positions of the circuit, that is the beginning and the end of the period of the circuit. Time must be respected, even for banks. As was seen in chapter three, the reduction of banking behavior to extending loans or to purchasing other assets was also the position adopted by Minsky and even Kaldor, from where it was canied over into the more contemporary post-Keynesian literature. The following passage in Dow (1997, pp. 66, 74-5) represents well the standard post-Keynesian interpretation of banks: Banks would express liquidity preference by curtailing credit creation (loan being their least liquid asset) and placing any free resources in investments. Suppose that there is an increase in liquidity preference in the sense of a wish to switch from long—tenn assets. Households will be unwilling to undertake debt, and will withhold consumption demand, as they attempt to build up stocks of liquid assets. Fimis will find their profits squeezed, will revise downward their investment plans, and will likewise be unwilling to undertake new debt while trying to build up liquid resources. Banks will find the riskiness of their loan portfolio increasing, will perceive increased lender's risk attached to new loan requests, and will for both reasons favour holding placements over new lending. This passage, however, is problematic in several ways.’ The opening and

closing statements assume, as discussed earlier, that banks face a decision either to lend or to purchase other assets. The question of where banks take these “free resources” from is a legitimate concem. If it is argued that banks have at their disposal a given - ex ante — pool of loanable fimds which they divide into loans and other assets, then what does this imply? Would this mean, in fact, that reserves (or deposits) create loans, as in the

more orthodox theories of money? If not, then post-Keynesians must better explain this statement. In particular, they must explain where this

296

Credit, Money and Production

pool of “free resources” comes from, if not from a prior supply of loanable funds. The issue of banks’ liquidity preference also poses a certain problem. As long as liquidity preference is understood to mean that banks are reluctant to lend given increased pessimism, uncertainty or animal spirits, then the expression is a sound one. However, if it is understood to mean, as argued above, a decision between lending and buying other assets, then no sense can be made of it. Lending is also an “investment” for the bank — an investment which generates profits. Banks do not have a preexisting pool of funds to invest, and hence their decision is not one of “either or.” Banks can only invest in other assets from their profits, or only once they have generated sufficient profits to do so at the end of the circuit, since profits arise only after banks have made loans to entrepreneurs, when these loans have been paid back, with interest. This point was well made by Parguez (l996, p. 185), claiming that “banks spend their whole profit in acquiring a part of the new shares. Shares can be sold because there are already available profits needing to be spent." The correct sequence of irreversible events therefore is the following. Ex ante, banks agree to extend credit to their creditworthy borrowers, at which point money is created ex nihilo. The supply of credit will exactly equal the creditworthy demand for bank credit from firms If barn‘-:3 express a preference for liquidity. they will do so at the beginning of the period as well, but this will be a function of their expectation"-2 of ti»: future course of the cycle, i.e., their animal spirits - not from their desire to “invest” in other assets. From their lending activities, provided banks’ bets are realized, profits will be generated, ex post. At this point, and this point only, banks will be able to invest their profits by purchasing securities or other assets. Hence, their decision to lend is made at the beginning of the circuit, while their decision to purchase other assets is made at the end of the circuit. There is therefore no decision to be made between the two. In this sense, post— Keynesians have proven unable to correctly differentiate between the two stages of the monetary circuit, 8 and to explain the generation of profits out of lending activities. Having discussed liquidity preference thoroughly, the real question is once the spectrum of rates has been detemiined, will there be a “feedback” effect on investment decisions? In other words, is liquidity preference causal? The answer is not an easy one. Within a given circuit, it is clear that it is not, since decisions relative to liquidity preference arise only at the end of the period. But can it be assumed that these decisions taken at the end of a given period will have repercussions for the next period? As post— Keynesians argue, as liquidity preference falls, households will purchase securities, thereby raising the prices of assets, and reducing their yield,

A Post—Keynesiarr/Circuitist Theory ofBanks

297

provided, of course, the supply of new securities is not perfectly elastic. In this sense, household and firm decisions with respect to financial instr'uments will impact on the spectrum of interest rates which will certainly have feedback effects.

At this point, as relative interest rates change, will this impact on the investment decision of firms in the next period? In this sense, the analysis

presented above can be seen as a bridge between two consecutive periods. Increased liquidity preference of households in this period will affect the profits of firms in the same period. The fmns’ outstanding debt will be

equal to households’ demand for money, i.e., their hoarded saving. This stock of money will become the inherited stock in the following period. In the process, as relative rates change, decisions of firms and banks may be affected. Both the resulting stock of money and the-changes in relative rates become the inherited stock of knowledge at the beginning of the next period.

CONCLUSION Credit becomes the focus of analysis of a modern capitalist system. Credit

is advanced to fmns ex ante, before production takes place and incomes are created. Money appears ex post, as do profits. Money is created ex nihilo by the banks as fmns’ investment and production plans are financed. A theory of banking practices based on uncertainty was developed which was consistent with the overall frame of analysis of the monetary circuit.

This was an important step since circulationists have traditionally ignored this aspect of the circuit. Uncertainty, however, affects banks’ decisions to grant credit, just like it affects fmns’ decisions to produce and invest. To deal with uncertainty, banks establish creditworthiness guidelines which they then impose on firms. Provided frrrns meet these criteria, then banks

will meet their demand for credit. Through the cycle, as animal spirits change given the banks’ expectations of the unknown and unknowable

future, the creditworthiness criteria will change accordingly. As banks become more pessimistic, it will become more difficult for firms to receive

credit, or even to have their existing credit renewed. Any credit constraints operate through the creditworthiness guidelines, not through the rate of interest. Despite Wray’s (1996) claim, there is ample place in circuit theory both for uncertainty and liquidity preference. Also, banks were seen as being able to impose a series of fmancial constraints on fu'ms, especially the rate of interest, but also the mark-up

firms incorporate in their pricing decisions. This was interpreted to mean that banks can effectively impose norms of profitability on firms. Finally, the analysis of the monetary circuit helped to put in proper context the issue of liquidity preference, both from the point of view of

298

Credit, Money and Production

banks and households. While banks’ desire to remain more liquid is expressed in terms of uncertainty, it is expressed at the beginning of the circuit. Households, however, express their liquidity preference at the end of the circuit, it is not causal in the determination of investrnent, nor in

influencing banks into lending in the same period. Households’ demand for money is a stock which is carried over to the next period, where the

circuit begins again. It may have an influence on real activity, but only in the subsequent period. It is households’ liquidity preference which bridges consecutive periods of circulation. Overall, the circuitists theory of endogenous money represents a clear break with mainstream orthodox theories of money and credit. This chapter, along with chapter one, has shown to what degree the dynamic

circuit is removed from orthodoxy, to a greater degree than post-Keynesian and New Keynesian theories of credit and money. As Lavoie (1996, p. 6) writes, “Circuit theory

constitutes the proper foundations to a non-

orthodox monetary theory, which itself must be part of a larger nonorthodox research programme encompassing effective demand as well as

value theory.”

NOTES l. 2. 3.

Blinder (I987, p. 328) makes a similar statement: “Firms may have a desired or ‘notional’ supply based on relative prices, expectations, and other variables. But they rmiy need credit to produce the goods.” This is another example of how close Keynes came to defending a horizontalist position. See Rochon (1997). Moreover, the extension of credit to creditworthy borrowers does not guarantee the bank profits. While creditworthiness is a guideline for banks to approve credit, and is in fact determined with respect to banks’ expectations of the future, it in no means protects banks and indeed bank borrowers from the ravages of the uncertain future. Banks’ ‘bet’ on firrm and the future may be wrong, and firms may default and not be able to

4.

reimburse the banks. It is nearly impossible to reconcile much of Wray’s (I999) new views on chartalism and statism with his previous post-Keynesian views developed in Wray (I990). lf Wray was post-Keynesian then, “chartalist” or even “funny monetarist" may better apply today.

5.

As discussed in chapter one, at the end of the circuit, banks become financial

6.

There have been interesting attempts to extend the theory of liquidity preference to the long period (see Rogers, 1989, p. 252; Pivetti, 1985; Panico, I988; and Lavoie, I996). In this context, according to Lavoie (1996, pp. 6-7), “Any change in interest rates orchestrated by the monetary authorities will be considered transitory as long as banks and rentiers have not adjusted to it. The rate of interest set by the monetary authorities becomes permanent when the liquidity preference of the public stops playing a role. Then, imperfections excluded, the central bank discount rate, the money market rate and the long-term rate of interest will all be equal." This was the sarne position held by

intermediaries.

Robinson (l95l, quoted in 1979, p. I43): “Equilibrium in the market is attained when the interest rates are such that no wealth is moving across any frontier. Prices are then

A Post-Keynesian/Circuitist Theory ofBanks

299

such that the nurket is content to hold just that quantity of each type of asset which is available at the moment." This issue is, however, beyond the scope of this work. Once again, notice the choice of households is between consumption and saving. On another note, it is interesting to see that Dow makes the same mistake that she accuses her horizontalist colleagues of nuking. Dow (I997, p. 76) clairm that “initially, firms facing a profit squeeze may prefer not to bonow, but may be forced to do so to maintain working capital." Dow assumes that a profit—squeezed firm will be able to borrow from banks in order to nuintain capital. However, a bank will probably not want to lend to such a firm, since its creditworthiness my be too low. The only way a firm may continue to borrow automatically is if it had a line of credit — a fact which Dow goes through great lengths to discredit.

Bibliography Aglietta, M. (1970), Régulation et crise du capitalisme, Paris: Calman— Lévy. Aglietta, M. (1979), A Theory of Capitalist Regulation: the US Experience, London: New Left Books. Amadeo, E.J. (1986), “The Role of Capacity Utilization in Long Period Analysis’, Political Economy, 2, 147-60. Anderson, L. and J. Jordan (1968), “Monetary and Fiscal Actions: a Test of their Relative Importance in Economic Stabilization”, Federal Reserve Bank ofSt—Louis Monthly Review, November, 50 (1 1), 11-2.

Arena, R. (I982), “Réflexions sur la compatibilité des approaches ricardienne ey keynésienne du fonctiomiement de l’activité économique”, Economie Appliquée, 35(3), 405—48. Arena, R. (1985), “Circulation, revenu ct capital: théorie monétaire et

tradition non—quantitative", in R. Arena and A. Graziani (eds), Production, circulation et monnaie, Paris: Presses universitaires de

France. Arena, R. (1987), “L’école intemationale d’été de Trieste (1981-1985): vers une synthése classico-keynésienne”, Economies et Sociétés, Oeconomia, March, 7, 205-38. Arena, R. (1992), “Une synthése entre post—keynésiens et néo-ricardiens

est—elle encore possible?”, L 'Actualité économique, December, 68 (4), 587-606. Arena, R. (1996), “Industrial and Financial Factors in Investment Decision: a Significant Difference Between Circuit and Post-Keynesian Approaches’, in E. Nell and G. Deleplace (eds), Money in Motion,

London: Macmillan Press. Arestis, P. (1987), “Post-Keynesian Theory of Money, Credit and Finance”, Thames Papers on Political Economy, Spring. Reprinted in P. Arestis (ed.), Post-Keynesian Monetary Economics: New Approaches to Financial Modeling, Aldershot, UK: Edward Elgar. Arestis, P. (1990), “Post-Keynesianism: a New Approach to Economics”, Review ofSocial Economy, Fall, 48 (3), 222—46. Arestis, P. (1992), The Post-Keynesian Approach to Economics: An Altemative Analysis of Economic Theory and Policy, Aldershot, UK: Edward Elgar.

301

302

Credit, Money and Production

Arestis, P. and C. Driver (1988), “The Endogeneity of UK Money Supply:

a Political Economy Perspective”, Economies et Sociétés, Monnaie et Production, 9, 121-38. Arestis, P., S. Dunn and M. Sawyer (1998), “On the Coherence of PostKeynesian Economics: a Commentary on Walters and Young”, Unpublished.

Arestis, P. and A. Eichner (1988), “The Post-Keynesian and lnstitutionalist Theory of Money and Credit”, Journal of Economic Issues, December, 22 (4), 1003-21. Arestis, P. and P.G.A. Howells (1996), “Theoretical Reflections on Endogenous Money: the Problem with “Convenience Lending”, Cambridge Journal ofEconomics, 20, 539-51. Arestis, P. and T. Skouras (1985), Post Keynesian Economic Theory: a Challenge to Neo Classical Economics, Sussex: Wheatsheaf Books. Asimakopulos, A. (1985), “The Role of Finance in Keynes’s General Theory”, Economic Notes, 3, 5-16. Asimakopulos, A. (1991), Keynes ’s General Theory and Accumulation,

Cambridge: Cambridge University Press. Backhouse, R.E. (1988), “The Value of Post Keynesian Economics: a

Neoclassical Response to Harcourt and Hamouda”, Bulletin of Economic Research, January, 40 (1), 35-42. Bailly, J.L (1992), “Nouvelles considerations sur le motif de “fmance” de

John Maynard Keynes’, Economie Appliquée, 65 (1), 105-127. Bailly, J.L. (1994), “Monnaie, “finance” et intérét”, Economie Appliquée, 66(1), 151-74. Balbach, A. B. (1981), “How Controllable Is Money Growth?”, Federal Reserve Bank ofSt-Louis Review, April, 63 (4), 3-12. Barker, W. and L.-R. Lafleur (1994), “Business Cycles and the CreditAllocation Process: an Institutional Perspective”, in Credit, Interest Rate Spreads and the Monetary Policy Transmission Mechanism, Proceedings of a conference held at the Bank of Canada, November. Bamerjee, A. (1992), “A Simple Model of Herd Behavior”, Quarterly Joumal ofEconomics, August, 107 (3), 797-817. Barrére, A. (1979, Déséquilibre économique et contre—rév0lution keynésienne, Paris: Economica. Barrere, A. (1990), ‘Signification générale du circuit : une interpretation”, Economies et Sociétés, Monnaie et Production, 6, 9-34. Barro, R. (1976), “The Loan Market, Collateral, and the Rate of Interest”, Journal ofMoney, Credit and Banking, May, 10 (2), 170-83. Becketti, S. and C. Morris (1992), “Are Bank Loans Still Special?”, Federal Reserve Bank ofKansas City Economic Review, 77 (3), 71-84. Bellofiore, R. (1983), “Moneta e' sviluppo nell’ana1ysi di Schumpeter”, Quaderni di Storia dell 'Economia Politico, 2, 39-99.

Bibliography

303

Berger, A.N. and G.F. Udell (1992), “Some Evidence on the Empirical Significance of Credit Rationing”, Joumal of Political Economy, 100

(5), 1047-77. Bemanke, B.S. (1981), “Bankruptcy, Liquidity, and Recession", American Economic Review, Papers and Proceedings, 71 (2), 155-9.

Bemanke, B.S. (1993), “Credit in the Macroeconomy”, Federal Reserve Bank ofNew York Quarterly Review, Spring, 18, 50-70.

Bemanke, B.S. and A. Blinder (1988), “Credit, Money, and Aggregate Demand”, American Economic Review, Papers and Proceedings, May, 78, 435-9. Bemanke, B.S. and A. Blinder (1992), “The Federal Funds Rate and the

Channels of Monetary Transmission", American Economic Review, September, 82 (4), 901-21. Bemanke, B. and M. Gertler (1987), “Banking and Macroeconomic Equilibrium”, in W.A. Bamett and K.J. Singleton (eds), New Approaches in Monetary Economics, New York: Cambridge University

Press, pp. 89-111. Bemanke, B. and M. Gertler (1989), “Agency Costs, Net Worth, and Business Fluctuations’, American Economic Review, March, 79 (1), 14-

31. Bemanke, B. and M. Gertler (1990), “Financial Fragility and Economic Performance”, Quarterly Joumal ofEconomics, February, 105 (1), 87114. Bemanke, B. and M. Gertler (1995), “Inside the Black Box: the Credit Channel of Monetary Policy Transmission”, Journal of Economic Perspectives, Fall, 9(4), 27-48. Bemanke, B., M. Gertler and S. Gilchrist. (1996), “The Financial Accelerator and the Flight in Quality”, Review of Economics and Statistics, February, 78(1), 1-15. Besanko, D. and A.V. Thakor (1987), “Competitive Equilibrium in the Credit Market under Asymmetric Infomiation”, Journal of Economic Theory, June, 42, 167-82.

Bester, H. (1985), ‘Screening versus Rationing in Credit Markets with Imperfect Information”, American Economic Review, 75, 850-5. Bibow, J. (1995), ‘Some Reflections on Keynes's “Finance Motive” for the Demand for Money”, Cambridge Journal ofEconomics, 19, 647-66.

Blinder, A. (1987), “Credit Rationing and Effective Supply Failures’, Economic Joumal, June, 97, 327-52. Blinder, A. and L. Maccini (1991), “Taking Stock: a Critical Assessment of Recent Research on Inventories’, Joumal of Economic Perspectives,

Winter, 5, 73-96. Blinder, A. and J. Stiglitz (1983), “Money, Credit Constraints, and Economic Activity”, American Economic Review, 73, 297-302.

304

Credit, Money and Production

Bomhoff, E.J. (1977), “Predicting the Money Multiplier”, Journal of Monetary Economics, July, 3 (3), 325-45. Borio, C.E.V., N. Ketmedy and S. Prowse (1994), “Exploring Aggregate Asset Price Fluctuations across Countries’, BIS Economic Papers, April, 40. Brown, E.K. (1981), “The Neoclassical and Post-Keynesian Research Programs: the Methodological Issues’, Review of Social Economy, October, 39 (2), 111-32. Burger, A. (1971), The Money Supply Process, Belmont, CA: Wadsworth. Cagan, P. (1965), Determinant and Eflects of Changes in the Stock of Money, 1867-1960, New York: National Bureau of Economic Research. Calomiris, C. and R.G. Hubbard (1990), “Firm Heterogeneity, Intemal Finance and Credit Rationing”, Economic Journal, 100, 90-104. Canadian Bankers Association (1996),. Small and Medium Businesses in Canada: Their Perspective of Financial Institutions and Access to Financing. Volume 1, Ottawa: Queen’s Printer. Cartellier, J. (1985), “Théorie de la valeur ou hétérodoxie monétaire: les terrnes d‘un choix", Economie Appliquée, 38 (1), 63-82. Carter, M. (1989), “Financial Innovation and Financial Fragility”, Journal ofEconomic Issues, September, 23 (3), 779-93. Carvalho, F.C. (1984), “On the Concept of Time in Shacklean and Sraffian Economics‘, Journal ofPost Keynesian Economics, Whiter, 6 (2), 26580. Carvalho, F.C. (1995), “Post-Keynesian Developments of Liquidity Preference Theory", in P. Wells (ed.), Post-Keynesian Economic Theory, Boston: Kluwer Academic Publishers. Carvalho, F.C (1996) “Paul Davidson’s Rediscovery of Keynes’s Finance Motive and the Liquidity Preference versus Loanable Funds Debate”, in P. Arestis (ed.), Keynes, Money and the Open Economy: Essays in Honour ofPaul Davidson, Volume 1, Cheltenham: Edward Elgar. Cencini, A. (1984), Time and the Macroeconomic Analysis of Income, London: Pinter. Cencini, A. (1988), Money, Income and Time: a Quantum Theoretical Approach, London: Pinter. Chick, V. (1977), The Theory of Monetary Policy, London: Gray-Mills (First edition, 1973). Chick, V. (1983), Macroeconomics Afier Keynes: a Reconsideration of the General Theory, Cambridge, MA: MIT Press. Chick, V. (1986), “The Evolution of the Banking System and the Theory of Saving, Investment and Interest”, Economies et Sociétés, Monnaie et Production, 3, 95-110. Chick, V. (1995), “Is There a Case for Post Keynesian Economics?”, Scottish Journal ofPolitical Economy, Febniary, 42 (1), 20-36.

Bibliography

30$

Clinton, K. and D. Howard (1994), “From Monetary Policy Instruments to Administered Interest Rates: The Transmission Mechanism in Canada”, Ottawa: Bank of Canada, Technical Report, 69. Coghlan, R. (1978), “A New View of Money”, Lloyds Bank Review, July, 129, 12-27. Cottrell, A. (1986), “The Endogeneity of Money and Money—Income Causality”, Scottish Journal of Political Economy, February, 33 (1), 227. Cottrell, A. (1988), “The Endogeneity of Money: a Reply”, Scottish Journal ofPolitical Economy, August, 35 (3), 295-97. Cottrell, A. (1994), “Post-Keynesian Monetary Theory”, Cambridge Joumal ofEconomics, December, 18(6), 587-605. Cottrell, A. (1994a), “Endogenous Money and the Multiplier", Journal of Post Keynesian Economics, Fall, 17 (1), 111-20. Cramp, A.B. (1962), “Two Views on Money", Lloyds Bank Review, July, 1-15. Cramp, A.B. (1970), “Does Money Matter?", Lloyds Bank Review, October, 23-37. Crotty, J. (1994), “Are Keynesian Uncertainty and Macrotheory Compatible? Conventional Decision Making, Institutional Structures, and Conditional Stability in Keynesian Macromodels’, in G. Dymski and R. Pollin (eds), New Perspectives in Monetary Macroeconomics: Explorations in the Tradition of Hyman P. Minsky, Ann Arbor, MI: University of Michigan Press. Crotty, J. (1996), “ls New Keynesian Investment Theory Really “Keynesian”? Reflections on Fazzari and Variato”, Journal of Post Keynesian Economics, Spring, 18 (3), 333-57. Dalziel, P. (1995), “The Keynesian Multiplier, Liquidity Preference, and Endogenous Money”, Journal ofPost Keynesian Economics, Spring, 18 (3), 311-31. Dardi, M. (1994), “Kahn’s Theory of Liquidity Preference and Monetary Policy”, Cambridge Journal ofEconomics, February, 18(1), 91-106. Davidson, P. (1965), “Keynes’s Finance Motive”, Oxford Economic Papers, March, 17 (1), 47-65.

Davidson, P. (1972), Money and the Real World, New York: John Wiley and Sons. Second edition, 1978.

Davidson, P. (1980), “The Dual-Faceted Nature of the Keynesian Revolution: Money and Money Wages in Unemployment and Production Flow Prices’, Journal ofPost Keynesian Economics, Spring, 2 (3), 291-307. Davidson, P. (1982-3), “Rational Expectations: a Fallacious Foundation for Studying Crucial Decision-Making Processes’, Journal of Post Keynesian Economics, Winter, 5, 182-97.

306

Credit, Money and Production

Davidson, P. (1986), “Finance, Funding, Saving, and Investment”, Journal ofPost Keynesian Economics, Fall, 9 (1), 101-10. Davidson, P. (1988), “Introduction: Three Views on the Endogenous Money Supply”, Journal ofPost Keynesian Economics, Spring, 10 (3), 370-1.

Davidson, P. (1988a), “A Technical Definition of Uncertainty and the Long Run Non—Neutrality of Money”, Cambridge Journal of Economics,

September, 12 (3), 329-7. Davidson, P. (1989), “On the Endogeneity of Money Once More”, Journal ofPost Keynesian Economics, Spring, 11 (3), 488-90. Davidson, P. (1991), “Is Probability Theory Relevant for Uncertainty? A Post Keynesian Perspective”, Journal of Economic Perspectives, 5 (1), 129-44. Davidson, P. (1994), “Comment on “Do Infonnational Functions Justify Federal Credit Programs?"’, Journal of Money, Credit and Banking, August, 26 (3), 545-51. Davidson, P. (1996), “What Are the Essential Elements of Post Keynesian Monetary Theory?”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Davidson, P. (l996a), “What Revolution? The Legacy of Keynes’, Journal

ofPost Keynesian Economics, Fall, 19 (1), 47-60. Davidson, P. and S. Weintraub (1973) “Money as Cause and Effect”,

Economic Journal, December, 83, 1117-32. De Largentaye, J. (1979), “A Note on the General Theory of Employment, Interest and Money”, Journal of Post Keynesian Economics, Spring, 1 (3), 6-15. Delli Gatti, D. and M. Gallegati (1990), “Financial Instability, Income Distribution, and the Stock Market”, Joumal of Post Keynesian Economics, Spring, 12 (3), 356-74. Deriet, M. and M. Seccareccia (1996), “Bank Markups, Horizontalism and the Significance of Banks’ Liquidity Preference: An Empirical Assessment”, Economies et Sociétés. Monnaie et Production, 10 (2-3), 139-63.

Desai, M. (1989), “The Scourge of the Monetarists: Kaldor on Monetarism and on Money”, Cambridge Journal ofEconomics, March, 13 (1), 17182. Desai, M. (1995), Macroeconomics and Monetary Theory: The Selected Essays ofMeghnad Desai, Volume 1, Aldershot, UK: Edward Elgar.

DeVroey,

M.

(1984),

“Inflation:

a

Non-Monetarist

Monetary

Interpretation”, Cambridge Journal of Economics, December, 8 (4), 381-99. DeVroey, M. (1988), “I1 circuito della moneta: due interpretazioni”, A. Graziani and M. Messori (eds), Moneta e Produzione, Turin: Giulio Einaudi, pp. 215-45.

Bibliography

307

Diamond, D.W. (1984), “Financial Intemrediation and Delegated Monitoring", Review ofEconomics Studies, July, 51 (166), 393-414. Dickens, E. (1990), “Financial Crises, Innovations and Federal Reserve

Control of the Stock of Money”, Contributions to Political Economy, 9, 1-23. Di Ruzza, A. (1984), “Quelques commentaires sur la communication du Professeur Davidson”, mitneo, University of Ottawa Conference,

October. Dombusch, R. and S. Fisher (1987), Macroeconomics. 4th edition. New York: McGraw-Hill. Dow, S. (1984), “Money Supply Endogeneity”, Unpublished Manuscript,

University of Ottawa - ISMEA Conference, Ottawa. Dow, S. (1985), Macroeconomic Thought: A Methodological Approach, Oxford? Basil Blackwell.

Dow, S. (1988), “Post-Keynesian Economics: Conceptual Underpinnings’, British Review ofEconomic Studies, Autumn, 10 (23), 1-18.

Dow, S. (1990), “Post-Keynesianism as Political Economics: A Methodological Discussion”, Review ofPolitical Economy, 2 (3), 345-

58. Dow, S. (1992), “Post-Keynesian Methodology: A Comment”, Review of Political Economy, 4 (1), 111-13. Dow, S. (1996), “Horizontalism: A Critique”, Cambridge Journal of

Economics, 20, 497-508. Dow, S. (1997), “Endogenous Money", in G. Harcourt and P. Riach (eds), The Second Edition ofKeynes ‘s General Theory, London: Routledge. Dow, A.C. and S.C. Dow (1989), “Endogenous Money Creation and Idle Balances’, in J. Pheby (ed.), New Directions in Post-Keynesian Economics, Aldershot, UK: Edward Elgar.

Dow, S. and P. Earl (1984), “Methodology and Orthodox Monetary Policy", Economie Appliquée, 37(1), 143-63.

Ducros, B. (1978), “Les relations entre prix, flux et stocks en analyse du

déséquilibre”, Economie Appliquée, 31 (1-2), 15-34.

Dutt, A.K. (1990-1991), “Interest Rate Policy in LDCs: A Post Keynesian View”, Journal ofPost Keynesian Economics, Winter, 13 (2), 210-32. Dutt, A.K. (1990), Growth, Distribution and Uneven Development, Cambridge: Cambridge University Press.

Dutt, A.K. and E.J. Amadeo (1990), Keynes ‘s Third Altemative: the NeoRicardians and the Post Keynesians, Aldershot, UK: Edward Elgar.

Dymski, G. (1988), “A Keynesian Theory of Bank Behavior”, Journal of Post Keynesian Economics, Fall, 10 (4), 499-526.

Dymski, G. (1989), “Keynesian versus Credit Theories of Money and Banking: A Reply to Wray”, Journal of Post Keynesian Economics,

Fall, 12(1), 157-63.

308

Credit, Money and Production

Dymski, G (1990), “Money and Credit in Radical Political Economy: A Survey of Contemporary Perspectives’, Review of Radical Political Economy. Sumrner-Fall, 22 (2-3), 38-65. Dymski, G. (1992), “A “New View” of the Role of Banking Firm in Keynesian Theory”, Journal of Post Keynesian Economics, Spring, 14 (3), 311-20. Dymski, G. (1993), “Keynesian Uncertainty and Asymmetric Information: Complementary or Contradictory?”, Joumal of Post Keynesian Economics, Fall, 16 (1), 49-54. Dymski, G. (1994), “Fundamental Uncertainty, Asymmetric Infomration, and Financial Structure: “Post” versus “New” Keynesian Microfoundations’, in G. Dymski and R. Pollin (eds), New Perspectives in Monetary Macroeconomics: Essays in the Tradition of Hyman P. Minsky, Ann Harbor, MI: University of Michigan Press. Dymski, G. (1996), “Basic Choices in Keynesian Models of Credit”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan Press. Dymski, G. (l996a), “Money as a “Time Machine” in the New Financial World”, in P. Arestis (ed.), Keynes, Money, and Exchange Rates: Essays in Honour ofPaul Davidson, Aldershot, UK: Edward Elgar. Dymski, G. and R. Pollin (1992), “Hyman Minsky as Hedgehog: The Power of the Wall Street Paradigm”, in S. Fazzari and D. Papadimitriou (eds), Financial Conditions and Macroeconomics Performance: Essays in Honor ofHyman P. Minsky, Armonk, NY: M.E. Sharpe. Early, J. and G. Evans (1982), “The Problem is Bank Liability Management”, Challenge, January-February, 54-6. Eckstein, O. and A. Sinai (1986), “The Mechanisms of the Business Cycle in the Postwar Era”, in R.J. Gordon (ed.), The American Business Cycle, Chicago: University of Chicago Press for NBER, pp. 39-132. Edward, F. and F.R.S. Mishkin (1995), “The Decline in Traditional Banking: Implications for Financial Stability and Regulatory Policy", Federal Reserve Bank of New York Economic Policy Review, July, 1 (2), 27-45. Eichner, A. (1982), “La théorie post-keynésienne et la recherche empirique”, Actualité Economique, 58 (1), 223-45. Eichner, A. (1985), “Towards an Empirically Valid Economics’, Eastem Economic Journal, October-December, 11 (4), 437-49. Eichner, A. (1986), Toward a New Economics, Armonk, NY: M.E. Sharpe. Eichner, A. (1987), The Macrodynamics of Advanced Market Economies, Amionk, NY: M.E. Sharpe. Eichner, A. and J. Kregel (1975), “An Essay on Post-Keynesian Theory: A New Paradigm in Economics’, Journal of Economic Literature, December, 13 (4), 1293-311.

Bibliography

309

Epstein, G. (1987), “Federal Reserve Behavior and the Limits of Monetary Policy in the Cmrent Economic Crisis’, in R. Cherry and T. Michl (eds), The Imperiled Economy, New York, NY: Union for Radical Political

Economics. Epstein, G. (1990), “Prime Rates, Federal Reserve Signaling, and Financial Instability”, Journal of Post Keynesian Economics, Summer, 12 (4), 618-34. Fama, E. ( 1980), “Banking in the Theory of Finance”, Joumal ofMonetary Economics, January, 6 (1), 39-57.

Fand, D.I. (1980), “Can the Central Bank Control the Money Stock?”, in T. Havrilesky and J. Boorman (eds), Current Issues in Monetary Theory and Policy. Second edition. Illinois: Harlan Davidson. Fazzari, S. (1992), “Keynesian Theories of Investment and Finance: Neo, Post, and New”, in S. Fazzari and D. Papadimitirou (eds), Financial Conditions and Macroeconomic Performance: Essays in Honor of Hyman P. Minsky, Armonk, NY: M.E. Sharpe. Fazzari, S. R.G. Hubbard and B. Peterson (1988), “Financing Constraints

and Corporate Investment”, Brookings Papers on Economic Activity, 1, 141-95. Fazzari, S. and H. Minsky (1984), “Domestic Monetary Policy: If Not Monetarism, What?”, Joumal ofEconomic Issues, March, 28 (1), 10116.

Fazzari, S. and A.M. Variato (1994), ‘Symmetric Infonnation and Keynesian Theories of Investment”, Journal of Post Keynesian Economics, Spring, 16(3), 351-69. Fazzari, S. and A.M. Variato (1996), “Varieties of Keynesian Investment Theories: Further Reflections’, Journal of Post Keynesian Economics, Spring, 18(3), 359-68. Felderer, B. and S. Homburg (1987), Macroeconomics and New Macroeconomics. Berlin: Springer—Verlag. Feldstein, M. and J. Stock (1994), “The Use of a Monetary Aggregate to Target Nominal GDP", in G. Mankiw (ed.), Monetary Policy. Chicago: University of Chicago Press. Ferri, P. and H. Minsky»(1989), “The Breakdown of the IS-LM Synthesis: Implications _for Post-Keynesian Economic Theory”, Review ofPolitical Economy, March, 1 (1), 123-43.

Fontana, G. (1997), “Post Keynesian Approaches to Monetary Theory”, in S. Dow and J. Hillard (eds), Beyond Keynes, Cheltenham, UK: Edward Elgar.

Forman, L., M. Groves and A. Eichner (1985), “The Demand Curve for Money Further Reconsidered”, in M. Jarsulic (ed.), Money and the Macro Policy, Boston: Kluwer Nijkoff.

310

Credit, Money and Production

Forman, L., M. Groves, and A. Eichner (1987), “The Critical Dynamics of

the American Economy: Preliminary Results from an Econometric

Model’, Economie Appliquée, 40 (4), est-70s.

Foster, G.P. (1986), “The Endogeneity of Money and Keynes’s General Theory”, Journal ofEconomic Issues, December, 20 (4), 953-68. Foster, J. (1989), “The Macroeconomics of Keynes: an Evolutionary Perspective”, in J. Pheby (ed.), New Directions in Post-Keynesian Economics. Aldershot, UK: Edward Elgar.

Friedman, B. and K. Kuttner (1992), “Money, Income, Prices, and Interest Rates’, American Economic Review, June, 82 (3), 472-92. Friedman, M. and A. Schwartz (1963), A Monetary History of the United States, I867-1960, Princeton: Princeton University Press. Fuerst, T. (1992), “Liquidity, Loanable Funds and Real Activity”, Journal ofMonetary Economics, February, 29, 3-24. Gale, D. and M. Hellwig (1985), “Incentive-Compatible Debt Contracts 1: The One Period Problem”, Review ofEconomic Studies, October, 64764. Gedeon, S. (1985), “A Comment on and Extension of Lavoie’s ‘The Endogenous Flow of Credit and the Post Keynesian Theory of Money”’, Journal ofEconomic Issues, September, 29 (3), 837-42. Gedeon, S. (1985-86), “The Post Keynesian Theory of Money: A Summary and an Eastem European Example”, Journal of Post Keynesian Economics, Winter, 8(2), 208-21.

Georgescu-Roegen, N. (1971), The Entropy Law and the Economic Process, Cambridge: Cambridge University Press. Gertler, M. (1988), “Financial Structure and Aggregate Economic Activity: an Overview”, Journal of Money, Credit and Banking, August, 20 (3), 559-88. Gertler, M. (1992), “Financial Capacity and Output Fluctuations in an Economy with Multi-Period Financial Relationships’, Review of Economic Studies, 59, 455-72. Gertler, M. and S. Gilchrist (1993), “The Role of Credit Market Imperfections in the Transmissions of Monetary Policy: Arguments and Evidence”, Scandinavian Journal ofEconomics, January, 95(1), 43-64.

Gertler, M. and S. Gilchrist (1994), “Monetary Policy, Business Cycles, and the Behavior of Small Manufacturing Firms’, Quarterly Journal of Economics, May, 109, 309-40. Godley, W. and F. Cripps (1983), Macroeconomics, London: Fontana. Goodhart, C. (1975), Money, Interest and Uncertainty, London: Macmillan. Goodhart, C. (1989), “Has Moore Become Too Horizontal?” Journal of Post Keynesian Economics, Fall, 12 (1), 29-34. Goodhart, C. (1989a), “The Conduct of Monetary Policy”, Economic Journal, June, 99 (396), 293-346.

Bibliography

31 1

Goodhart, C. (1994), “What Should Central Banks Do? What Should Be Their Macroeconomic Objectives and Operations?", Economic Journal, November, 104, 1424-36. Gordon, R.J. (1990), “What is New Keynesian Economics?”, Joumal of Economic Literature, September, 28 (3), 1115-71. Gorton, G. and G. Permachi (1990), “Financial Intermediaries and Liquidity

Creation”, Joumal ofFinance, 45, 49-71. Goux, J.F. (1987), “La théorie monétaire de la “fmance” chez Keynes: une réinterprétation”, Revue d "Economie Politique, September-October, , 97 (5), 592-612. Graziani, A. (1984), “The Debate on Keynes’s Finance Motive”, Economic Notes, 1 (1), 15-33. Graziani, A. (1985), “Le débat sur le “motif de fmancement” de Keynes’, Economie Appliquée, 38, 159-75. Graziani, A. (1987), “Keynes’s Finance Motive”, Economies et Sociétés. Monnaie et Production, 7, 7-36. Graziani, A. (1988), “Il circuito monetario”, in A. Graziani and M. Messori (ed.), Moneta e Produzione, Milan: Einaudi. Graziani, A. (1989), “Money and Finance in Joan Robinson’s Works’, in G.R. Feiwel (ed.), The Economics of Imperfect Competition and Employment: Joan Robinson and Beyond, London: Macmillan. Graziani, A. (1990a), “The Theory of the Monetary Circuit”, Economies et

Sociétés. Monnaie et Production, July, 24 (7), 7-36. Graziani, A. (l990b), “La théorie du circuit et la théorie macroéconomique de la banque”, Economies et Sociétés, Monnaie et Production, 6, 51-62. Graziani, A. (1991), “La théorie keynésienne de la mormaie et le fmancement de l’économie”, Economie Appliquée, 44 (1), 25-41. Graziani, A. (1996), Money as Purchasing Power and Money as a Stock of

Wealth in Keynesian Economic Thought”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Greenwald, B. and J.E. Stiglitz (1987), “Keynesian, New Keynesian and New Classical Economics’, Oxford Economic Papers, 39, 119-32. Greenwald, B. and J.E. Stiglitz (1991), “Toward a Reformulation of Monetary Theory: Competitive Banking”, Economic and Social Review, October, 23 (1), 1-34. Greenwald, B. and J.E. Stiglitz (1993), “New and Old Keynesians’, Joumal ofEconomic Perspectives, Winter, 7 (1), 23-44. Greenwald, B., J.E. Stiglitz, and A. Weiss (1984), “Informational

Imperfections in the Capital Market and Macroeconomic Fluctuations’, American Economic Review, May, 74, 194-200. Gurley, J. and E. Shaw (1959), Money in a Theory ofFinance, Washington: Brookings Institute. Guttmann, R. (1990), “The Regime of Credit-Money and its Current Transition, Economies et Sociétés, Monnaie et Production, 7, 81-105.

312

Credit, Money and Production

Hahn, F. (1965), “On Some Problems of Proving the Existence of an Equilibrium in a Monetary Economy”, in F. Hahn and F.P.R. Brechling (eds), The Theory ofInterest Rates, London: Macmillan Hahn, F. (1989), “Robinson-Halm Love-Hate Relationship: An Interview”, in G.R. Feiwel (ed.), Joan Robinson and Modern Economic Theory, New York: New York University Press. Hamouda, O. and G. Harcourt (1989), “Post-Keynesianism: From Criticism to Coherence?”, in J. Pheby (ed.), New Directions in Post-Keynesian Economics. Aldershot, UK: Edward Elgar. Harcourt, G. and C. Sardoni (1996), “The General Theory of Employment, Interest and Money: Three Views’, in P. Arestis (ed.), Keynes, Money and the Open Economy: Essays in Honour of Paul Davidson, Volume 1, Cheltenham: Edward Elgar. Hayek, F.A. ( 1935), Prices and Production, London: Routledge. Heise, A. (1992), “Commercial Banks in Macroeconomic Theory”, Journal ofPost Keynesian Economics, Spring, 14 (3), 285-96. Henry, J. (1982), “Les méthodes “post-keynésiennes’ et l’approche post— classique”, Actualité Economique, January—June, 58 (1-2), 17-60. Hewitson, G. (1995), “Post-Keynesian Monetary Theory: Some Issues’, Journal ofEconomic Surveys, 9 (3), 285-310. Hicks, J.R. (1974), The Crisis in Keynesian Economics, Oxford: Basil Blackwell. Hicks, J.R. (1989), A Market Theory of Money I, Oxford: Oxford University Press. Hodgman, D.R. (1960), “Credit Risk and Credit Rationing”, Quarterly Journal ofEconomics, May, 258-78. Hodgson, G. (1989), “Post-Keynesian and Institutionalism: The Missing Link”, in J. Pheby (ed.), New Directions in Post-Keynesian Economics, Aldershot, UK: Edward Elgar. Holmes, A. (1969), “Operational Constraints on the Stabilization of Money Supply Growth", Federal Reserve Bank of Boston, Controlling Monetary Aggregates, June, 65-77. Hoogduin, L. (1987), “On the Difference Between the Keynesian, Knightian and the Classical Analysis of Uncertainty and the Development of a More General Monetary Theory”, De Economist, 135, 52-65. Howells, P.G.A. (1995), “The Demand for Endogenous Money”, Journal of Post Keynesian Economics, Fall, 18 (1), 89-106. Howells, P.G.A. (1996), “Endogenous Money and the State of Trade”, in P. Arestis (ed.), Keynes, Money and the Open Economy. Essays in Honour ofPaul Davidson, Volume 1, Cheltenham: Edward Elgar. Howells, P.G.A. (1997), “The Demand for Money: A Rejoinder”, Journal ofPost Keynesian Economics, Spring, 19 (3), 429-35.

Bibliography

3 13

Howells, P.G.A. and K. Bain (1983), “The Derivation of the LM Schedule: A Pedagogical Note”, British Review ofEconomic Issues, Spring, 5, 5765. Isenberg, D.L (1988), “Is There a Case for Minsky’s Financial Fragility Hypothesis in the 1920’s?”, Joumal ofEconomic Issues, December, 22 (4), 1045-68. Jaffee, D. (1987), “Credit Rationing”, in J. Eatwell, M. Milgate and P. Newman (eds), The New Palgrave, London: Macmillan. Jaffee, D. and T. Russell (1976), “Imperfect Information, Uncertainty, and Credit Rationing”, Quarterly Joumal ofEconomics, November, 90 (4). Jaffee, D. and J. Stiglitz (1990), “Credit Rationing”, in B. Friedman and F. Hahn (eds), Handbook ofMonetary Economics, Volume 2, New York:

Elsevier Science Publishers, 837-88. Jarsulic, M. (1988), Eflective Demand and Income Distribution: Issues in Alternative Economic Theory, Oxford: Polity Press. Jarsulic, M. (1989), “Endogenous Credit and Homogeneous Business Cycles’, Joumal ofPost Keynesian Economics, Fall, 12 (1), 35-48. Johannes, J.M. and R.H. Rasche (1979), “Predicting the Money Multiplier", Joumal ofMonetary Economics, July, 5, 401-13. Johannes, J.M. and R.H. Rasche (1981), “Can the Reserve Approach to

Monetary Control Really Work”, Journal of Money, Credit and Banking, August, 13, 298-313.

Johnson, H. (1968), “Problems of Efficiency in Monetary Management”, Journal ofPolitical Economy, July—August, 76(5), 971-90. Johnson, H. (1976), “Reserve Requirements and Monetary Control’, Discussion Paper 66, Economic Council ofCanada, Ottawa. Kalm, G.A. (1991), “Does More Money Mean More Bank Loans’, Federal Reserve Board ofKansas City Economic Review, July—August, 21-31. Kahn, R. (1972), “Memorandum of Evidence Submitted to the Radcliffe Committee”. Reprinted in Selected Essays on Employment and Growth, Cambridge: Cambridge University Press. Kahn, R. (1984), The Making of Keynes’ General Theory, Cambridge: Cambridge University Press. Kaldor, N. (1939), ‘Speculation and Economic Stability”, Review of Economic Studies, 7, October, 1-27. Reprinted in Collected Economic Essays, Volume 2, Essays on Economic Stability and Growth, New York: Holmes and Muier, 1960. Kaldor, N. (1955), “The Lessons of the British Experiment since the War: Full Employment and the Welfare State”, Paper presented at the Centenary Congress of the Société Royale d’Economie Politique de Belgique, Brussels. Reprinted in Collected Economic Papers, Volume 3, Essays on Economic Policy I , London: Duckworth, 1964. Kaldor, N. (1958), “Monetary Policy, Economic Stability and Growth.” A Memorandmn Submitted to the Cormnittee of the Working of the

314

Credit, Money and Production

Monetary System (Radcliffe Committee), June 23. Reprinted in Collected Economic Papers, Volume 3, Essays on Economic Policy I, London: Duckworth, 1964. Kaldor, N. (1960), Collected Economic Essays, Volume 2, Essays on Economic Stability and Growth, London: Duckworth. Kaldor, N. (1964), Collected Economic Essays, Volume 3, Essays on Economic Policy I, London: Duckworth Kaldor, N. (1970), “The New Monetarism”, Lloyds Bank Review, July, 1-7. Kaldor, N. (1978), Collected Economic Essays, Volume 5, Further Essays on Economic Theory, London: Duckworth. Kaldor, N. (1978a), Collected Economic Essays, Volume 6, Further Essays on Applied Economics, London: Duckworth. Kaldor, N. (1980), Origins ofNew Monetarism, Cardiff: University College Cardiff Press. Kaldor, N. (1982), The Scourge ofMonetarism, Oxford: Oxford University Press. Kaldor, N. (1983), “Keynesian Economics after Fifty Years’, in J. Trevithick and G.N.D. Worswick (eds), Keynes and the Modem World, Cambridge: Cambridge University Press. Kaldor, N. (1989), “Limitations of the General Theory”, in F. Targetti and A.P. Thirwall (eds), Further Essays on Economic Theory and Policy, London: Gerald Duckworth. Kaldor, N. and J. Trevithick (1982), “A Keynesian Perspective on Money”, Lloyds Bank Review, January, 1-20. Kalecki, M. (1937), “The Principle of Increasing Risk”, Economica, 4, 440-7. Kalecki, M. (1954), Theory ofEconomic Dynamics, New York: Rinehart. Kalecki, M. (1971), Selected Essays in the Dynamics of the Capitalist Economy, Cambridge: Cambridge University Press. Karecken, J.H. (1957), “Lenders’ Preferences, Credit Rationing, and the Effectiveness of Monetary Policy”, Review ofEconomics and Statistics, August, 39, 292-302. Kashyap, A. and J. Stein (1994), “Monetary Policy and Bank Lending”, in N.G. Mankiw (ed.), Monetary Policy, Chicago: Chicago University Press. Kashyap, A.K., J.C. Stein and D.W. Wilcox (1993), “Monetary Policy and Credit Conditions: Evidence from the Composition of Extemal Finance”, American Economic Review, March, 53 (1), 78-95. Kaufman, H. (1987), “Debt: The Threat to Economic and Financial Stability”, Federal Reserve Bank of Kansas City Economic Review, December, 3-11. Keynes, J.M. (1971a), The Collected Writings of John Maynard Keynes. Volume iv: A Tract on Monetary Reform, D. Moggridge (ed.), London: Macmillan and St-Martin’s Press.

Bibliography

3 15

Keynes, J.M. (1971b), The Collected Writings of John Maynard Keynes. Volume v: A Treatise on Money. The Pure Theory of Money, D.

Moggridge (ed.), London: Macmillan and Cambridge University Press. Keynes, J.M. (1971c), The Collected Writings of John Maynard Keynes. Volume vi: A Treatise on Money. The Applied Theory of Money, D. Moggridge (ed.), London: Macmillan and Cambridge University Press. Keynes, J.M. (1973a), The Collected Writings of John Maynard Keynes. Volume vii: The General Theory ofEmployment, Money and Interest, D. Moggridge (ed.), London: Macmillan and St-Martin’s Press. Keynes, J.M. (1973b), The Collected Writings of John Maynard Keynes, Volume xiii: The General Theory and After, Preparation, D. Moggridge (ed.), London: Macmillan and Cambridge University Press. Keynes, J.M. (19730), The Collected Writings of John Maynard Keynes. Volume xiv: The General Theory and Afier, Defence and Development, D. Moggridge (ed.), London: Macmillan and Cambridge University Press. Keynes, J.M. (1980a), The Collected Writings of John Maynard Keynes. Volume xxv: Activities 1940-44. Shaping the post—War World: The Clearing Union, D. Moggridge (ed.), London: Macmillan and Cambridge University Press. Keynes, J.M. (1980b), The Collected Writings of John Maynard Keynes. Volume xxvi: Activities 1940-46. Shaping the post-War World: Bretton

Woods, D. Moggridge (ed.), London: Macmillan and Cambridge University Press. Keynes, J.M. (l980c), The Collected Writings of John Maynard Keynes. Volume xxvii: Activities 1940-46: Shaping the post-War World, D. Moggridge (ed.), London: Macmillan and Cambridge University Press. Keynes, J.M. (1979), The Collected Writings of John Maynard Keynes. Volume The General Theory and After: A Supplement, D. Moggridge (ed.), London: Macmillan and Cambridge University Press. King, J.E. (1996), “Hyman Minsky: The Making of a Post Keynesian”, in S. Pressman (ed.), Interactions in Political Economy Malvern after Ten Years, London; Routledge. Knight, F. (1921), Risk, Uncertainty and Profit, London: The London School of Economics and Political Science. Knodell, J. (1995), “Altemative Approaches to Money and Interest Rates: A Comment”, Journal ofEconomic Issues, March, 29(1), 266-73. Kohn, M. (1986), “Monetary Analysis, the Equilibrium Method, and Keynes’s General Theory”, Journal of Political Economy, 91, 1191224. Kregel, J. (1973), The Reconstruction of Political Economy: An Introduction to Post-Keynesian Theory, London: Macmillan.

316

Credit, Money and Production

Kregel, J. (1988), “l1 fmanziamento in Keynes: dal Trattato all Teoria Generale”, in A. Graziani and M. Messori (eds), Moneta e produzione, Turin: Einaudi. Kregel, J. (1992), “Minsky’s “Two Price” Theory of Financial Instability and Monetary Policy: Discounting versus Open Market Intervention”, in S. Fazzari and D. Papadimitriou (eds) Financial Conditions and Macroeconomic Performance: Essays in Honor of Hyman P. Minsky, Amionk, NY: M.E. Sharpe. Kurz, H. (1991), “Technical Change, Growth and Distribution: A Steady State Approach to Unsteady Growth on Kaleckian Lines’, in E. Nell and W. Semrnler (eds), Nicholas Kaldor and Mainstream Economics: Confi-ontation or Convergence? New York: St-Martin’s Press. Lachmann, L.M. (1943), “The Role and Expectations in Economics as a Social Science”, Economica, 10, 10-23. Lavoie, M. ( 1982), “Les post-keynésiens et la monnaie endogene”, Actualité Economique, 58 (1), 191-221. Lavoie, M. (1984), “Un modele post-keynésien d’économie monétaire fondé sur la théorie du circuit”, Economies et Sociétés, 18 (2), 233-58. Lavoie, M. (1985), “Credit and Money: The Dynamic Circuit, Overdraft Economies, and Post-Keynesian Economics’, in M. Jarsulic (ed.), Money and Macro Policy, Hingham, MA: Kluwer Academic Publishers. Lavoie, M. (1986), “L’endogeneité de la monnaie chez Keynes’, Recherches économiques de Louvain, March, 52 (1), 67-84. Lavoie, M. (1987), “Monnaie et production: une synthese de la théorie du circuit”, Economies et Sociétés, September, 9, 65-101. Lavoie, M. (1990), “Le circuit dans la pensée post-keynésienne américaine”, Economie, 6, 105-18. Lavoie, M. (1991), “Change, Continuity and Originality in Kaldor’s Monetary Theory”, in E. Nell and W. Semmler (eds), Nicholas Kaldor and Mainstream Economics: Conflontation or Convergence?, New

York: St-Martin’s Press. Lavoie, M. (1992), Foundations of Post-Keynesian Economic Analysis, Aldershot, UK: Edward Elgar. Lavoie, M. (1992a), “Towards a New Research Programme for PostKeynesianism and Neo-Ricardianism”, Review of Political Economy, 4 (1), 37-78. Lavoie, M. (1993), “A Post—Classical View of Money, Interest, Growth and Distribution”, in G. Mongiovi and C. Riihl (eds), Macroeconomic Theory: Diversity and Convergence, Aldershot, UK: Edward Elgar. Lavoie, M. (1994), “Money and Credit”, in P. Arestis and M. Sawyer (eds), The Elgar Companion to Radical Political Economy, Aldershot, UK: Edward Elgar.

Bibliography

317

Lavoie, M. (1995), “The Demand for Loan Creates the Supply of Deposits: What Then is the Role of the Demand for Money”, University of

Ottawa, Working Paper 955017E. Lavoie, M. (1995a), “Interest Rates in Post-Keynesian Models of Growth and Distribution”, Metroeconomica, June, 46 (2), 146-77. Lavoie, M. (1996), “The Endogenous Supply of Credit-Money, Liquidity Preference and the Principle of Increasing Risk: Horizontalism versus the Loanable Funds Approach”, Scottish Journal ofPolitical Economy, August, 43 (3), 275-300. Lavoie, M. (l996a), “Loanab1e Funds, Endogenous Money, and Minsky’s Financial Fragility Hypothesis’, in A. Cohen, H. Hagemann and J. Smithin (eds), Money, Financial Institutions and Macroeconomics, Boston: Kluwer Academic. Lavoie, M. (1996b), “Monetary Policy in an Economy with Endogenous Credit Money”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Lavoie, M. (1998), “The Credit-Led Supply of Deposits and the Demand for Money: Kaldor’s Reflux Mechanism as Endorsed by Joan Robinson”, Cambridge Journal ofEconomics, 1998 (forthcoming). LeBourva, J.(1953), L 'inflation francaise d 'aprés guerre 1945-1949: Essai sur la théorie et la mesure du processus inflationniste, Paris: Imprimerie nationale. LeBourva, J. (1959), “La théorie de l’inflation, le rapport des experts et l’opération de décembre”, Revue Economique, September, 10 (5), 71354.

LeBourva, J. (1992), “Money Creation and Credit Multipliers’, Review of Political Economy, 4(4), 447-66. Lerner, A.P. (1952), “The Essential Properties of Interest and Money”, Quarterly Journal ofEconomics, 46, 172-93. Levratto, N. (1992), “Une analyse du statut théorique de la notion de rationnement du crédit”, in R. Arena and D. Torre (eds), Keynes et les nouveaux keynésiens, Paris: Presses universitaires dc France. Mankiw, N.G. (1986), “The Allocation of Credit and Financial Collapse”, Quarterly Journal ofEconomics, August, 101 (3), 455-71. Reprinted in N.G. Mankiw and D.F. Romer (eds) (1991), New Keynesian Economics, Cambridge, MA: MIT Press. Mankiw, N.G. (1992), “The Reincarnation of Keynesian Economics’, European Economic Review, April, 36 (2-3), 559-65. Mankiw, N.G. and D. Romer (1991), “Introduction”, in N.G. Mankiw and D.F. Romer (eds), New Keynesian Economics, Cambridge, MA: MIT Press. Maricic, A. (1985), “L’offre de monnaie dans la théorie keynésienne: une recherche empirique de défmitions’, in R. Arena and A. Graziani (eds),

318

Credit, Money and Production

Production, circulation et monnaie, Paris: Presses universitaires de France. Meltzer, A. (1969), “Controlling Money", Federal Reserve Board of StLouis Review, May, 51 (5), 16-24. Messori, M. (1985), “Le circuit dc la monnaie: acquis et problemcs nonrésolus’, in R. Arena and A. Graziani (eds), Production, circulation et monnaie, Paris: Presses universitaires de France. Messori, M. (1986), “Financement bancaire et décisions de production”, Economies et Sociétés, Monnaie et Production, 3, 127-58. Messori, M. (1991), “Keynes’s General Theory and the Endogenous Money Supp1y”, Economie Appliquée, 44 (1), 125-52. Minsky, H. (l957a) “Monetary Systems and Accelerator Models’, American Economic Review, 47 (6), 859-83. Minsky, H. (l957b), “Central Banking and Money Market Changes’, Quarterly Journal ofEconomics, 71 (2), 171-87. Minsky, H. (1963), “Comment”, Review of Economics and Statistics, Supplement, 45, 64-72. Minsky, H. (1964), “Financial Crisis, Financial Systems, and the Perfonrrance of the Economy”, Private Capital Markets, Commission on Money and Credit Research Study. Englewood, NJ: Prentice Hall. Minsky, H. ( 1975), John Maynard Keynes, New York: Colmnbia University Press. Minsky, H. (1982), Can “it" Happen Again? Essays on Instability and Finance, Armonk, NY: M.E. Sharpe. Minsky, H. (1986), Stabilizing an Unstable Economy, New Haven: Yale University Press. Minsky, H. (1991), “The Endogeneity of Money”, in E. Nell and W. Semmler (eds), Nicholas Kaldor and Mainstream Economics: Confrontation or Convergence? New York: St. Martin’s Press. Minsky, H. (1992), Market Structure and Investment Behavior,

Unpublished manuscript. Jerome Levy Economics Institute. Minsky, H. (1996), “The Essential Characteristics of Post Keynesian Economics’, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Mishkin, F.S. (1995), “Symposium on the Monetary Transmission Mechanism”, Joumal ofEconomic Perspectives, Fall, 9 (4), 3-10. Mondello, G. (1985), “Etude de la liaison monnaie-revenu. Demande effective : les théories de Thomas Tooke et J.M. Keynes’, Revue Economique, Mai, 14, 509-55. Mongiovi, G. and C. Riihl (1993), “Monetary Theory after Sraffa”, in G. Mongiovi and C. Riihl (eds), Macroeconomic Theory: Diversity and Convergence, Aldershot, UK: Edward Elgar. Moore, B. (1979), “Monetary Factors’, in A. Eichner (ed.), A Guide to Post Keynesian Economics, White Plains, NY: M.E. Sharpe.

Bibliography

3 19

Moore, B. (1979a), “The Endogenous Money Stock”, Joumal of Post Keynesian Economics, Fall, 2 (1), 49-70. Moore, B. (1983), “Unpacking the Post-Keynesian Black Box: Bank Lending and the Money Supply”, Joumal of Post Keynesian Economics, Summer, 5(4), 537-56. Moore, B. (1984), “Keynes and the Endogeneity of the Money Stock”, Studi Economici, 22, 23-69. Moore, B. (1985), “Contemporaneous Reserve Accotmting: Can Reserves Be Quantity Constrained?”, Joumal ofPost Keynesian Economics, Fall, 7 (1), 103-13. Moore, B. (1985a), “Wages, Bank Lending, and the Endogeneity of Credit Money”, in M. Jarsulic (ed.), Money and Macro Policy, Hingham, MA: Kluwer Academic Publishers. Moore, B. (1986), “How Credit Drives the Money Supply: The Significance of Institutional Developments’, Journal of Economic Issues, June, 20 (2), 445-52. Moore, B. (1988), Horizontalists and Verticalists: The Macroeconomics of Credit-Money, Cambridge: Cambridge University Press. Moore, B. (1988a), “The Endogenous Money Supply”, Joumal of Post Keynesian Economics, Spring, 10 (3), 372-85.

Moore, B. (1988b), “Keynes’s Treatment of Interest”, in O.F. Hamouda and J.N. Smithin (eds), Keynes and Public Policy After Fifty Years, Volume II: Theories and Method, Aldershot, UK: Edward Elgar. Moore, B. (1988c), “The Endogeneity of Money: a Comment”, Scottish Joumal ofPolitical Economy, August, 35(3), 291-94. Moore, B. (1989), “On the Endogeneity of Money Once More”, Joumal of Post Keynesian Economics, Spring, ll (3), 479-87. Moore, B. (1989a), “Does Money Supply Endogeneity Matter? A Comment”, South African Joumal ofEconomics, June, 57 (2), 194-202. Moore, B. (1989b), “A Simple Model of Bank Intermediation”, Joumal of Post Keynesian Economics, Fall, 12 (1), 10-28. Moore, B. (1991), “Marx, Keynes, Kalecki and Kaldor on the Rate of Interest as a Monetary Phenomenon”, in E. Nell and W. Semmler (eds), Nicholas Kaldor and Mainstream Economics: Confrontation or Convergence?, New York: St. Ma1'tin’s Press. Moore, B. (l99la), “Money Supply Endogeneity: ‘Reserve Price Setting’ or ‘Reserve Quantity Setting?”’, Joumal ofPost Keynesian Economics, Spring, 13 (3), 404-13. Moore, B. (1994), “The Demise of the Keynesian Multiplier: A Reply to Cottrell”, Journal ofPost Keynesian Economics, Fall, 17 (1), 121-33. Moore, B. (1995), “The Exogeneity of Short-Term Interest Rates: A Reply to Wray”, Joumal ofEconomic Issues, March, 29 (1), 258-66.

320

Credit, Money and Production

Moore, B. (1996), “The Money Supply Process: A Historical Reinterpretation”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Moore, B. ( 1997), “Reconciliation of the Supply and Demand for Endogenous Money”, Joumal ofPost Keynesian Economics, Spring, 19 (3), 423-28. Morgan, D.P. (1990), “Bank Credit Commitments: Protection from a Credit Crunch?”, Federal Reserve Bank of Kansas City Economic Review, September-October, 51-9. Morgan, D.P. (1992), “Are Bank Loans a Force in Monetary Policy”, Federal Reserve Bank of Kansas City Economic Review, Second Quarter, 31-41. Mott, T. (1985-6), “Toward a post-Keynesian Formulation of Liquidity Preference”, Joumal ofPost Keynesian Economics, Winter, 8 (2), 22231. Musella, M. and C. Panico (1993), “Kaldor on Endogenous Money and Interest Rates”, in G. Mongiovi and C. Ruhl (eds), Macroeconomic Theory: Diversity and Convergence, Aldershot, UK: Edward Elgar. Neal, P. (1996), “Keynesian Uncertainty in Credit Markets”, Journal of Post Keynesian Economics, Spring, 18(3), 397-418. Nell, E. (1967), “Wicksell’s Theory of Circulation”, Journal of Political Economy, August, 4, 386-394. Nell, E. (1986), “On Monetary Circulation and the Rate of Exploitation”, Thames Papers in Political Economy, Summer. Nell, E. (1988), “Does the Rate of Interest Determine the Rate of Profit”, Political Economy, 4, 263-67. Nell, E. (1990), “Keynes and Sraffa: Marshallian or Classical Foundations?”, in K. Bharadwaj and B. Schefold (eds), Essays on Pierro Sraffa, London: Unwin Hyman. Nell, E. (1992), Transformational Growth and Eflective Demand, New York, NY: New York University Press. Nell, E. (1996), “The Circuit of Money in a Production Economy”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Ng, S. and H. Schaller (1996), “The Risky Spread, Investment, and Monetary Policy Transmission: Evidence on the Role of Asymmetric Information”, Review of Economics and Statistics, August, 78 (3), 37583. Niggle, C.J. (1991), “The Endogenous Money Supply Theory: An lnstitutionalist Appraisal”, Journal of Economic Issues, March, 25 (1), 137-51. O’Donnell, R. (1992), “The Unwritten Books and Papers of John Maynard Keynes”, Review ofPolitical Economy, Winter, 24 (4), 767-817.

Bibliography

321

Palley, T. (1987-8), “Bank Lending, Discount Window Borrowing, and the Endogenous Money Supply: A Theoretical Framework”, Journal of Post Keynesian Economics, Winter, 10 (2), 282-304. Palley, T. (1991), “The Endogenous Money Supply: Consensus and Disagreement”, Journal ofPost Keynesian Economics, Spring, 13 (3), 397-403. _ Palley, T. (1994), “Competing Views of the Money Supply Process: Theory and Evidence”, Metroeconomica, 45 (1), 67-88. Palley, T. (1996), “Beyond Endogenous Money: Toward Endogenous Finance”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Palley, T. (l996a), Post Keynesian Economics: Debt, Distribution and the Macroeconomics Economy, New York, NY: St-Martin’s Press. Panico, C. (1985), “Market Forces and the Relation Between the Rates of Interest and Profits’, Contribution to Political Economy, 4, 37-60. Panico, C. (1988), Interest and Profit in Theories of Value and Distribution, London: Macmillan. Parguez, A. (1975), Monnaie et macroéconomie, Paris: Economica. Parguez, A. (1980), “Profit, épargne, investissements: Eléments pour une théorie monétaire du profit”, Economie Appliquée, 2, 424-56.

Parguez, A. (1981), “Keynes et la revolution”, Cahiers d 'économie politique, 6, 171-87.

Parguez, A. (1984), “La dynamique de la monnaie”, Economies et Sociétés, Monnaie et Production, 2, 83-118. Parguez, A. (1985), “La théorie générale: la revolution inachevée dans la théorie de la monnaie et du capital”, in Barrere, A. (ed.), Keynes Aujourd’hui. Paris: Economica Parguez, A. (1986), “Au coeur du circuit, ou quelques réponses aux énigmes du circuit”, Economies et Sociétés, Monnaie et Production, 20 (8-9), 23-39. Parguez, A. (1987), “La crise dans le circuit, ou l’intégration de la fmance

et de la production”, Economies Appliquée, 40 (4), 755-70. Parguez, A. (1996), “Beyond Scarcity: A Reappraisal of the Theory of the Monetary Circuit”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Parguez, A. (1997), “Govemment Deficits within the Monetary Production Economy or the Tragedy of the Race to Balance Budgets’, Unpublished. Parrinello, S. (1985), “Le noyau marshallien de la Théorie Générale”, in A. Barrére (ed.), Keynes aujourd"hui: théories et politiques, Paris: Economica. Pasinetti, L. (1974), Growth and Income Distribution: Essays in Economic Theory, Cambridge: Cambridge University Press.

322

Credit, Money and Production

Pasinetti, L. (1986), “Nicholas Kaldor: An Appraisal’, Cambridge Journal ofEconomics, December, 10 (4), 301-03. Perasan, M.H. (1987), The Limits of Potential Expectations, Oxford: Blackwell. Piccolo, F. (1983), “L’analisi monetaria di Wicksell’, Annali delle Fondazione Lui Einaudi, 17- 34. Pivetti, M. (1985), “On the Monetary Explanation of Distribution”, Political Economy, 1, 73-103. Pivetti, M. (1988), “On the Monetary Explanation of Distribution: A Rejoinder to Nell and Wray”, Political Economy, 4, 275-83. Pivetti, M. (1991), An Essay on Money and Distribution, London: Macmillan. Pivetti, M. (1996), “Joan Robinson and the Rate of Interest: An Important Change of View on a Topical Issue”, in M. Marcuzzo, L. Pasinetti and A. Roncaglia (eds), The Economics of Joan Robinson, London: Routledge. Pollin, R. (1987), ‘Structural Change and Increasing Fragility in the US Financial System”, in R. Chery and T. Michl (eds), The Imperiled Economy, New York, NY: Union for Radical Political Economics. Pollin, R. (1991), “Two Theories of Money Supply Endogeneity: Some Empirical Evidence”, Journal of Post Keynesian Economics, Spring, 13 (3), 366-95.

Pollin, R. (1995), “Financial Interrnediation and the Variability of the Saving Constraint”, Unpublished Manuscript, University of Califomia-Riverside. Pollin, R. (1996), “Money Supply Endogeneity: What Are the Questions and Why Do They Matter?”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Poulon, F. (1980), “Graphe, crise et circuit keynésien”, Revue d’économie politique, 90 (4), 371-409. Poulon, F. (1982), “Théorie des économies d’endettement et théorie du circuit”, in A. Barrere (ed.), Keynes aujourd'hui: théories et politiques, Paris: Economica. Rajan, R. (1992), “Insiders and Outsiders: the Choice between Relationship and Arrn’s Length Debt”, Journal ofFinance, 47, 1367-1400. Rao, S.K. (1987), Contributions to Post-Keynesian Economics: A Resume and a Critique, India: Sterling. Rasche, R. (1972), “A Review of Empirical Studies of the Money Supply Mechanism”, Federal Reserve Bank of St-Louis Review, July, 54, 1 1-9. Riley, J.C. (1987), “Credit Rationing: A Further Remark”, American Economic Review, March, 77, 224-27. Rizzo, B. (1992), “Keynes et les deux motifs de finance bancaire”, Economie Appliquée, 45 (4), 123-59.

Bibliography

323

Robinson, J. (1933), “The Theory of Money and the Analysis of Output", Review of Economic Studies, 1 (1), 22-6, Reprinted in Collected Economic Papers, Volume 1, 52-8. Robinson, J. (1951), “The Rate of Interest”, Econometrica, April, 19 (2), 92-111. Robinson, J. (1952), Ute Rate of Interest and Other Essays, London: Macmillan. Robinson, J. (1956), The Accumulation ofCapital, London: Macmillan. Robinson, J. (1961), “Own Rates of Interest”, Economic Joumal, 71, 596— 600. Robinson, J. (1970), “Quantity Theories Old and New: A Comment”, Journal ofMoney, Banking and Credit, November, 2, 504-512. Robinson, J. (1971), Economic Heresies, New York, NY: Basic Books. Robinson, J. (1978), “Keynes and Ricardo”, Journal of Post Keynesian Economics, Fall, 1 (1), 12-8. Robinson, J. (1979), “Foreword”, in A. Eichner (ed.), A Guide to PostKeynesian Economics, White Plains, New York, NY: M.E. Sharpe. Robinson, J. (1980), Collected Economic Papers, vol. III, Cambridge, MA:

MIT Press. Robinson, J. (1980), Collected Economic Papers, vol. V, Cambridge, MA:

MIT Press. Robinson, J. (1980), Further Contributions to Modem Economics, London: Blackwell. Rochon, L.—P. (1997), “Keynes’s Finance Motive: A Re—Assessment. Credit, Liquidity Preference and the Rate of Interest", Review of Political Economy, 9 (3), 277-93. Rochon, L.—P. (1999), “The Creation and Circulation of Endogenous Money: A Circuit Dynamique”, Joumal of Economic Issues, 33 (l), 1-21. Rochon, L.—P. (l999a) “A Neoclassical Interpretation of Minsky’s Early Work on Credit and Financial Innovations: Minsky as a ‘Special Case”’, Working Paper 99-03, Kalamazoo College. Rogers, C. (1989), Money, Interest and Capital: A Study in the Foundations ofMonetary Theory, Cambridge: Cambridge University Press. Rogers, C. (1998), “Post Keynesian and Strong New Keynesian Macroeconomics: Compatible Bedfe1lows?”, in R. Rotheim (ed.), New Keynesian Economics, Post Keynesian Altematives, London:

Routledge. Rogers, C. (l995a), “Monetary Theory and the Principle of Effective Demand”, Unpublished. Romer, C. and D. Romer. (1990), “New Evidence on the Monetary Transmission Mechanism”, Brookings Papers on Economic Activity, 1, 149-8.

324

Credit, Money and Production

Romer, D. (1993), “The New Keynesian Synthesis’, Journal of Economic Perspectives, Winter, 7 (1), 5-22. Rotheim, R. (1998), New Keynesian Economics/Post Keynesian Alternatives, London: Routledge. Rousseas, S. (1960), “Velocity Changes and the Effectiveness of Monetary Policy, 1951-1957”, Review ofEconomics and Statistics, February, 42 (1), 27-36. Rousseas, S. (1960a), “The Response of Income Velocity to Interest Rate Changes: A Rejoinder”, Review of Economics and Statistics, November, 42 (4), 455-57. Rousseas, S. (1985), “Financial Innovation and Control of the Money Supply”, in M. Jarsulic (ed.), Money and the Macro Policy, Boston: Kluwer-Nijhoff. Rousseas, S. (l985a), “A Markup Theory of Bank Loans Rate”, Journal of Post Keynesian Economics, Fall, 8 (1), 135-45. Rousseas, S. (1986), Post Keynesian Monetary Economics, Armonk, NY: M.E. Sharpe, (1992), Second edition. Rousseas, S. (1989), “On the Endogeneity of Money Once More”, Journal ofPost Keynesian Economics, Spring, ll (3), 474-8. Rowthom, B. (1981), “Demand, Real Wages and Economic Growth”, Thames Papers in Political Economy, Autumn. Sawyer, M. (1988), “Introduction”, in M. Sawyer (ed.), Post-Keynesian Economics, Aldershot,UK: Edward Elgar. Sawyer, M. (1995), “Post-Keynesian Economics: The State of the Art”, Unemployment, Imperfect Competition and Macroeconomics, Aldershot: Edward Elgar. Sawyer, M. (1996), “Money, Finance and Interest Rates: Some Post Keynesian Reflections’, in P. Arestis (ed.), Keynes, Money and the Open Economy. Essays in Honour of Paul Davidson, Volume I, Cheltenham: Edward Elgar. Schmitt, B. (1966), “La monnaie au centre de la répartition”, Revue d "économie politique, 76 (1) , 92-114. Schmitt, B. (1971), L 'analyse macroéconomique des revenus, Paris: Dalloz. Schmitt, B. (1975), Monnaie, salaires et profits, Paris: Presses universitaires de France. Schmitt, B. (1984), Inflation, chomage et malformations du capital, Paris: Economica. Schmitt, B. (1996), “A New Paradigm for the Determination of Money Prices”, in E. Nell and G. Deleplace (eds), Money in Motion, London: Macmillan. Schredt, S. (1990), “Credit Controls: 1980”, Federal Bank of Richmond Economic Review, 76 (6), 25-55. Schumpeter, J.A. (1934), The Theory of Economic Development, Oxford: Oxford University Press.

Bibliography

325

Screpanti, E. (1995), “Banks, Increasing Risk, and the Endogenous Money Supply”, Paper Presented at York University Conference, April. Screpanti, E. (1997), “Banks, Increasing Risk and the Endogenous Money Supply”, Revised, Unpublished. Sebastiani, M. (1989), “Introduction”, in M. Sebastiani (ed.), Kalecki’s Relevance Today, New York, NY: St. Martin’s Press. Sebastiani, M. (1994), Kalecki and Unemployment Equilibrium, New York,

NY: St. Martin’s Press. Seccareccia, M. (1988), “Systematic Viability and Credit Crunches: An Examination of Recent Canadian Cyclical Fluctuations”, Joumal of Economic Issues, March, 22 (1), 49-77.

Seccareccia, M. (1990), “The Two Faces of Neo-Wicksellianism During the 1930s: The Austrians and the Swedes”, in D.E. Moggridge (ed.), Perspectives on the History of Economic Thought, Volume IV, Keynes, Macroeconomics and Method, Aldershot, UK: Edward Elgar.

Seccareccia, M. (1994), “Credit Money and Cyclical Crises: The Views of Hayek and Fisher Compared", in M. Collona and H. Hagemarm (eds), Money and Business Cycles: The Economics of F.A. Hayek, Aldershot, UK: Edward Elgar.

Seccareccia, M. (1996), “Post-Keynesian Fundism and Monetary Circulation”, in E. Nell and G. Deleplace (eds), Money in Motion,

London: Macmillan. Seccareccia, M. (1997), “Pricing, Investment and the Financing of Production within the Framework of the Monetary Circuiti Some Preliminary Evidence”, Paper Presented at Eastern Economic Conference, Washington. Shackle, G.L.S. (1958), Time in Economics, Amsterdam: North-Holland. Shackle, G.L.S. (1961), Decision, Order and Time in Human Aflairs, Cambridge: Cambridge University Press. Shackle, G.L.S. (1967) The Years ofHigh Theory: Invention and Tradition in Economic Thought, 1926-1939, Cambridge: Cambridge University Press. Shackle, G.L.S. (1974), Keynesian Kaleidics, Edinburgh: E.U.P. Shackle, G.L.S. (1982), “Sir John Hick’s ‘IS-LM: An Explanation:' A Comment”, Joumal ofPost Keynesian Economics, Spring, 4, 435-8. Shackle, G.L.S. (1983), “Levels of Simplicity in Keynes’s Theory of Money and Employment”, South African Journal of Economics, September, 51 (3), 357-67. Shapiro, N. (1977), “The Revolutionary Character of Post-Keynesian Economies’, Journal ofEconomic Issues, September, 11 (3), 541-60. Sharpe, S. (1990), “Asymmetric Infomiation, Bank Lending and Implicit Contracts: A Stylized Model of Consumer Relationships”, Journal of Finance, 45, 1069-87.

326

Credit, Money and Production

Smith, P. (1979), “Keynes’ Finance Motive: Some Empirical Evidence”, Journal ofPost Keynesian Economics, Spring, 1 (3), 55-68. Smith, W.L. (1956), “On the Effectiveness of Monetary Policy”, American Economic Review, June, 46 (3), 588-606. Smithin, J. (1984), “Robert M. Solow", in A. Klamer (ed.), The New Classical Macroeconomics: Conversations with New Classical Economists and their Opponents, Sussex: Wheatsheaf Books. Smithin, J. (1993), “La pensée monétaire de Milton Friedman face aux théories contemporaines’, in M. Lavoie and M. Seccareccia (eds), Milton Friedman et son oeuvre, Montréal: Presses de l’université de Montreal. Smithin, J. (1994), Controversies in Monetary Economics: Ideas, Issues and Policy, Aldershot, UK: Edward Elgar. Smithin, J. (1997), “Fundamental Issues in Monetary Economics”, Unpublished. Solow, R.M. (1979), “Altemative Approaches to Macroeconomic Theory: A Partial View”, Canadian Joumal of Economics, August, 12 (3), 339-54. Sotl, I.O. (1957), “The Availability Doctrine: Theoretical Underpinnings”, Review ofEconomics and Statistics, October, 25, 41-8. Sraffa, P. (1960), Production of Commodities by Means of Commodities: Prelude to a Critique of Economic Theory, Cambridge: Cambridge University Press. St-Hill, R.L. (1995), “A Post Keynesian Perspective on Commercial Bank Behavior and Regulation”, Metroeconomica, February, 46 (1), 35-62. Stiglitz, J.E. and A. Weiss (1984), “Informational Imperfections in the Capital Market and Macroeconomic Fluctuations’, American Economic Review, May, 74 (2), 194-9. Stiglitz, J.E. (1987), “Credit Rationing: A Reply”, American Economic Review, March, 77, 228-31. Stiglitz, J.E. (1991), “Credit Rationing in Markets with Imperfect Information”, in G. Mankiw and D. Romer (eds), New Keynesian

Economics, Cambridge, MA: MIT Press, First published in American Economic Review, June,l98l, 71 (3), 393-410. Stiglitz, J.E. (1992), “Asyrrmietric Information in Credit Markets and its Implications for Macro—Economics", Oxford Economic Papers, 44 (4), 694-724. Sylos-Labini, P. (1949), “The Keynesians", Banca Nazionale del Lavoro Quarterly Review, 238-42. Targetti, F. (1992), Nicholas Kaldor: The Economics and Politics of Capitalism as a Dynamic System, Oxford: Clarendon Press. TC1Tl'l1I'1l, V. (1981), “Logical, Mechanical and Historical Time in Economics’, Economic Notes, Monte dei Paschi di Siena, 10 (3), 58104

Bibliography

327

Terzi, A. (1986-7), “The Independence of Finance from Saving: A Flow of Funds Interpretation”, Joumal of Post Keynesian Economics, Winter, 9(2), 188-97. Thirlwall, A.P. (1987), Nicholas Kaldor, London: Wheatsheaf Books. Thirlwall, A.P. (1991), “Nicholas Kaldor 1908-86”, in E. Nell and W. Semmler (eds), Nicholas Kaldor and Mainstream Economics: Confrontation or Convergence?, New York, NY: St-Martin’s Press. Thirlwall, A.P. (1994), “Talking About Kaldor”, in E. King (ed.), Economic Growth in Theory and Practice: A Kaldorian Perspective, Aldershot, UK: Edward Elgar. Tobin, J. (1960), “Towards Improving the Efficiency of the Monetary Mechanism”, Review ofEconomics and Statistics, 42, 276-9.

Tobin, J. (1963) “Commercial Banks as Creators of ‘Money”’, in Essays in Economics: Macroeconomics, Volume 1, London: North Holland. Tobin, J. (1965), “The Monetary Interpretation of History - A Review Article”, American Economic Review, June, 55 (3), 464-85, Reprinted in J. Cunningham Wood and R.N. Woods (eds), Milton Friedman: Critical Assessments, 1990, New York, NY: Routledge. Tobin, J. (l965a), “Money and Economic Growth”, Econometrica, October, 33 (4), 671-84. Tobin, J. (1968), “Notes on Optimal Monetary Growth”, Journal of Political Economy, July, 76(2), 833-59.

Tobin, J. (1970), “Money and Income: Post Hoc Ergo Propter Hoc?” Quarterly Joumal ofEconomics, 84, 301-17. Tobin, J. (1980), Asset Accumulation and Economic Activity, Chicago, IL: University of Chicago Press. Tobin, J. (1991), “On the Endogeneity of Money Supply”, in E. Nell and W. Semmler (eds), Nicholas Kaldor and Mainstream Economics: Confrontation or Convergence?, New York, NY: St. Martin’s Press.

Tobin, J. and W.C. Brainard (1963), “Financial Intermediaries and the Effectiveness of Monetary Controls”, The American Economic

Review, Papers and Proceedings of the Seventy—Fifth Armual Meeting of the American Economic Association, May, 53, (2), 383-400. Trautwein, H.M. (1995), “Pure Credit Economies in Monetary Theory”, Paper Presented at the Conference on Money, Financial Institutions

and Macroeconomics, York University, Canada. Tsiang, S.C. (1980) “Keynes’s ‘Finance’ Demand for Liquidity, Robertson’s Loanable Funds Theory, and Friedman’s Monetarism”, Quarterly Joumal ofEconomics, May, 94, (3), 467-91. Turner, M. (1989), Joan Robinson and the Americans, Annonk, N.Y.: M.E. Sharpe. Vallageas, B. (1985), “Les circuits dans les analyses de Marx, Boehm— Bawerk, Hayek et Keynes”, Economies et Sociétés. Monnaie et Production, August, 2, 47-68.

328

Credit, Money and Production

Vallageas, B. (1988), “Intérét, repartition et théorie des circuits", Economie Appliquée, 41 (2), 355-88. van Ees, H. and H. Garretsen (1993), “Financial Markets and the Complementarity of Asymmetric Information and Fundamental Uncertainty", Journal of Post Keynesian Economics, Fall, 16 (1), 3748. van Ees, H., G.H. Kuper, and E. Sterken. (1996), “Profits, Investment and Business Fluctuations”, Metroeconomica, February, 47(1), 19-33. Vickers, D. (1987), Money and Capital in the Theory of the Firm, Cambridge: Cambridge University Press. Walters, B. and D. Young (1997), “On the Coherence of Post-Keynesian Economics”, Scottish Journal ofPolitical Economy, 44 (3), 329-49. Weintraub, E.R. (1975), “Uncertainty and the Keynesian Revolution”, History ofPolitical Economy, Winter, 7 (4), 530-48. Weintraub, S. (1966), A Keynesian Theory of Employment, Growth and Income Distribution, Philadelphia, PA: Chilton. Weintraub, S. (1978), Capitalism 's Inflation and Unemployment Crisis, Reading, MA: Addison-Welsey. Weintraub, S. (1980), “Money Supply and Demand Interdependence”, Journal ofPost Keynesian Economics, Summer, 2 (4), 566-75. Wells, P. (1978), “In Review of Keynes”, Cambridge Journal of Economics, 2, (3), 315-25. Wells, P. (1983), “A Post Keynesian View of Liquidity Preference and the Demand for Money”, Journal ofPost Keynesian Economics, Summer, 5 (4), 523-36. Wicksell, K. (1936), Interest Rates, London: Macmillan. Wette, H.C. (1983), “Collateral in Credit Rationing in Markets with Imperfect Information: Note”, American Economic Review, June, 73, 442-5. Williamson, S.D. (1987), “Costly Monitoring, Optimal Contracts, and Equilibrium Credit Ratio”, Quarterly Journal ofEconomics, February, 102, 135-45. Winnett, A. (1992), “Some Semantics of Endogeneity”, in P. Arestis and V. Chick (eds), Recent Developments in Post-Keynesian Economics, Aldershot, UK: Edward Elgar. Wolfson, M.H. (1996), “A Post Keynesian Theory of Credit Rationing”, Journal ofPost Keynesian Economics, Spring, 18- (3), 443-70. Wray, R. (1988), “Profit Expectations and the Investrnent-Saving Relation”, Journal ofPost Keynesian Economics, Fall, 11 (1), 131-47. Wray, R. (1989), “Horizontalists and Verticalists: The Macroeconomics of Credit Money: A Review”, Journal ofEconomic Issues, December, 23 (4), 1185-9. Wray, R. (1989a), “A Keynesian Theory of Banking: A Comment on Dymski”, Journal ofPost Keynesian Economics, Fall, 12 (1), 152-6.

Bibliography

329

Wray, R. (1990), Money and Credit in Capitalist Economies: The Endogenous Money Approach, Aldershot, UK: Edward Elgar. Wray, R. (1992), “Commercial Banks, the Central Bank, and Endogenous Money”, Journal of Post Keynesian Economics, Spring, 14 (3), 297310.

Wray, R. (1992a), “Alternative Approaches to Money and Interest Rates”, Journal ofEconomic Issues, December, 26 (4), 1145-78. Wray, R. (1992b), “Minsky’s Financial Instability Hypothesis and the

Endogeneity of Money”, in S. Fazzari and D. Papadimitriou (eds), Financial Conditions and Macroeconomic Performance: Essays in Honor ofHyman P. Minsky, Armonk, NY: M.E. Sharpe. Wray, R. (l992c), “Altemative Theories of the Rate of Interest”, Cambridge Journal ofEconomics, 16, 69-89. Wray, R. (1995), “Keynesian Monetary Theory: Liquidity Preference or Black Box Horizontalism”, Journal of Economic Issues, March, 29 (1), 273-82. Wray, R. ( 1996), “Money in the Circular Flow”, in E. Nell and G. Deleplace (eds), Money and Motion, London: Macmillan. Wray, R. (1999), Understanding Modern Money: The Key to Full Employment and Price Stability, Cheltenham, UK: Edward Elgar.

Zannoni, D. and E. McKerma. (1981), “A Test of the Monetarist “Puzzle”, Joumal ofPost Keynesian Economics, Summer, 3 (4), 479-90. Zerbato, M. (1987), Keynésianisme et sortie de crise, Paris: Dunod.

Zerbato, M. (1989), “Bouc1age monétaire du circuit et austérité”, Economie Appliquée, 42 (1), 91-113.

Index accommodation and banks 225 lack of 181-6

role of central banks 166-8 accommodationist approach 1, 43, I63-4 adverse selection 241 Aglietta, M. 16, 30

post-Keynesian theory 270

bank lending 170-73, 272-4 in New Keynesian and post-Keynesian theory 268-9 bank lending channel 248, 274 arguments against 253-4 banking

in New Keynesian theory 265 in post-Keynesian theory 215, 265

Akerlof’ s “lemons” problem 237,

238

banks 28-9

and accommodation 225

Amadeo, EJ. 44, 65, 66, 216 animal spirits 76, 169, 173, 218, 282, 285, 286, 287, 288, 298 Arena, R. 16, 35, 66, 280, 292 Arestis, P. 44, 49, 53, 54, 55, 57, 65, 66, 67, 68, 72, 75, 99, 156, 161. 165, 167,171, 177, 216

circuitist theory 278-91 and credit 282-3 and credit-money 72

as financial intermediaries 36-7, 224 and firms contractual relations 280 liquidity preference 186-93 and money supply 29-30 and production and investment 147 risks 173 target rate of retum 283, 285

Asimakopulos, A. 20, 47, 48, 75, 211, 263 asset management 253, 254, 256, 263 asymmetric information 233, 241-3, 259

and screening mechanisms 243 and uncertainty 260-61, 266-7

asymmetric infonnation theory 215 Backhouse, R.E. 67 Bailly, J.L. 24, 56 Bain, K. 156 balance sheet channel 244-7, 272

and capital-goods finns 246-7 and credit 246

banana parable 98, 270 bank advances 151-2 bank assets 237

bank credit 122, 150, 157 and investment 147 Kaldor on 110-1 1

bank credit and other credit assets, in New Keynesian and

and uncertainty 281-7 Bamerjee, A. 282 Banere, A. 15, 20

base rate of interest 174-5 exogeneity 69 Becketti, S. 253, 256 Bellofiore, R. 16

Bemanke, B.S. 233, 235, 236, 237, 241, 242, 243, 246, 247, 249, 250, 251, 252, 254, 255, 256. 257, 258, 262, 264, 267, 268, 272, 273 Besanko, D. 243 Bester, H. 239, 243, 244 Bibow, J. 50, 52, 204 black box of monetary transmission

236, 251, 268 Blinder, A. 236, 237, 249, 250, 251, 331

332

Index

252, 254, 257, 258, 271, 272, 273 Borio, C.E.V. 246 borrower's risk 281

production 14-37 circuit school approach 88 circuitist theory of banks 278-91 Coghlan, R. 122, 166

Brainard, W.C. 125

collective bargaining 136

Brown, E.K. 44, 66 Burger, A. 251

commercial banks 45

and demand for credit 156-8

Calomiris, C. 244, 247

passivity 169-74 “commercial loan" model 94

Cambridge economists 143

commercial paper 256

Canadian Bankers Association 288, 290 capital-goods firms, and the balance

commodities and consumption curve 272-3 complete infonnation 263-4 Cottrell, A. 10, 49, 60, 6|, 68, 83, I32, 166, 169, I77, 202, 208, 209, 218, 219, 221, 222, 224. 228, 291, 295 “coverage ratio” 247 creation of income flows, circuit 31 credit I5, 43, 45-6, 48-9, 51, 64, 258, 298 and balance sheet channel 246 and banks 282-3 and central banks 249, 250

sheet channel 246-7 Carter, M. 187 Carvalho, F.C. 50. 208, 21 1, 294 CD market 253, 255

Cencini, A. 16 central banks 17-18, 47, 57, 72, 73, 85, 89, 92, 105-6, I20, 128, 137, 163, 223, 224 accommodative role 166-8 lack of accommodation I77, 181-6 and credit 249, 250

and interest rates 158-61 base rate 174 and monetary policy 160, 235, 262 and money in circulation 139-40 and money supply 58, I39, 237, 254-5 control over 121 , Open Market Operations 58 and reserves 165, 258, 263 supply of reserves 158-61 Chick, V. 48, 52, 67, 182, 204, 215, 291 circuit banks as financial intermediaries 36-7

creation of income flows 31 of credit and money 19-20 initial finance and endogenous creation of money 22-3| monetary reflux and profits 32-6 production decisions and prices 20-22 circuit approach, credit and

commercial banks, and demand for credit 156-8 demand for 48

and endogenous money 155-201 money and production, the circuit approach 14-37 in New Keynesian theory 267-8 in post-Keynesian theory 46, 267-8 and production 27 and uncertainty 216, 219 within demand for money 49 see also lines of credit credit channel 58, 235 credit constraints 287-8 credit and money 57, 164, 204 an altemative Keynesian view 53-6 circuit 19-20 demarcation between 46-6 differences 1-2, 151 post-Keynesian theory 42-8| credit money multiplier model 222 credit rationing 1 1 1, 224, 233, 247,

Index 252, 258-60, 262, 263 history 238-41 credit supply 279 credit-money 64, 204 and banks 72

endogeneity 85 credit-supply curve 73-4 creditworthiness 283-4, 289-90 Crotty, J. 196, 233, 260, 265, 266 Dalziel, P. 68 Dardi, M. 150, 151 Davidson, P. 2, 3, 4, 5, 26, 47, 48, 49, 51, 61, 67, 68, 75, 85, 204, 205, 206, 208, 209, 210, 213. 216. 219, 220. 260, 263, 271 early views on money 132-9

on finance motive I33 Oxford Economic Papers article

133 and velocity of money 134 de Largentaye, J . 205 debt, and excess saving 97-9 Deleplace, G. 14, 15, 16, 18, 25, 36, 207, 220 Delli Gatti, D. 49, 51, 188

demand 44-5, 66 for finance 51 for money 204 for money stock 155 demand curve for credit, upward sloping 181 Deriet, M. 4, 65,69, 71, 172, 181, 185, 191

Desai, M. 51,101,117,129, 224 DeVroey, M. 15 Di Ruzza, A. 16, 17 Dickens, E. 228 discount window borrowing 183 distribution 45 distribution theory 8 Dombusch, R. 67 Dow, A.C.76, 101, 163, 169, 187, 189, 291, 292, 293, 295 Dow, S. 5, 9, 10, 11,44, 47, 65, 66,

68,74, 76, 101, 163, 169, 187, 189,195, 208,211,215, 217, 218, 228, 286, 290, 296

333

Dow, S.C. 76, 101, 163, 169, 187, 189, 291, 292, 293, 295 Driver, C. 72 Ducros, B. 20 Dunn, S. 67 Dutt, A.K. 44, 65, 66, 216 Dymski, G. 47, 55, 73, 75, 83, 186, 189, 194, 215, 216, 223, 224. 233, 260, 262, 263, 264, 265, 266 “Dynamic circuit" see monetary circuit dynamic credit-supply curve 73-4 dynamic two-price Minsky diagram 197

Earl, P. 77 Eckstein, 0. 263 Edward, F. 253 Eichner, A. 44, 47, 63, 66, 67, 72, 73, 75,77, 85, 128, 156, 161, I64, 165.166, I67, 170,171, 174, I75, 182, 293 endogeneity, definitions 2 endogeneity of money 30, 128-9 endogeneity of money supply, and

Kaldor 101 endogeneity of rate of interest 116 endogenous money 1, 2, 3, 54, 62, 251 and credit 155-201 date of introduction 83 defined 57, 63 early views Davidson and Rousseas versus Robinson and Kahn 132-54 Minsky, Kaldor and Tobin 82- 131 links to financial fragility 86 meaning 42-3 and Minsky 89 in New Keynesian 268 “portfolio” approach 2 and post-Keynesianism 68, 268 two theories and the

post-Keynesians 56-64 "revolutionary theory" 2 structuralist approach 229

33 4

Index

endogenous money theory 6

ex ante saving and investment 88, 90 excess saving, and debt 97-9 exogeneity

circulation 5-6, 8-41 Friedman, M. 3, 68, 83-4, 119, 124 “frown costs" 167 Fuerst, T. 250

of base rate of interest 69, 158, 168 exogenous money supply 9, 89-92, 110, 116,127,134, 208

exogenous rate of interest 106, 149, 161 and Keynes 162-3

of reserves 92-3 extemal finance premium 242 Fama, E. 185 Fand, D.1. 60 Fazzari, S. 75, 233, 247, 260, 265, 266, 275, 279 Federal Open Market Committee (FOMC) 258 Federal Reserve Banks 121, I22-3 Feiwel, G.R. 144 Felderer, B. 67 Feldstein, M. 249 Ferri , P. 78 finance, and demand for money 51 finance motive 26-7, 32, 33, 48-53, 204, 292

Davidson on I33 “financial accelerator" 246, 272 and the business cycle 247 financial fragility 190 links to endogenous money 86

and loanable funds 96-7 financial intermediaries 1 18-19 banks as 224 financial saving 33, 34 firms, price setting 21-2 firms’ contractual relations, and banks 280 firms’ debt 36 Fisher, S. 67 flux-reflux notion 88 Fontana, G. 47, 133, 206 Fomran, L. 54, 73, 156, 163, 165, 167, 170, 182 Foster, G.P. 9, 129 Foster, J. 45 Franco-Italian school of monetary

Gale, D. 242 Gallegati. M. 49, 51, 188 Garretsen, H. 234, 259, 261, 265, 266, 267, 269, 287 Gedeon, S. 51,68, 169, 171

General Theory see Keynes, J.M. Georgescu-Roegen, N. 211 Gertler, M. 233, 235, 236, 240, 242. 243, 245, 246, 247, 249, 250, 251, 252, 254, 255, 256, 257. 263, 264, 267, 268, 272, 274 Gilchrist, S. 245, 246, 247, 249, 250. 254, 255 Goodhart, C. 156, 168, 177 Gordon, R.J. 268 Gorton, G. 254 Goux, .l.F. 15 Graziani, A. 8,11,15,16, I9, 20, 21,

24, 25, 26, 27, 30, 31, 34. 37. 56, 145, 147

Greenwald, B. 239. 256, 257

Groves, M. 156 Guttmann, R. 208 Hahn, F. 144 Hamouda, O. 65, 99 Harcourt, G. 11, 50, 65, 99 Harrod 162, 206 Hayek, F.A. 62, 70, 84, 90 Heise, A. 168, 174, 281, 286 Hellwig, M. 242 Henderson, H. 162 Henry, .1. 66 Hewitson, G. 63, 66, 67, 69, 71, 72, 75, 83, 95, 177, 180, 187, 195 Hicks, J.R. 11, 162, 192 Hodgson, G. 74 Holmes, A. 72 Homburg, S. 67 horizontal credit supply curve 159 horizontal supply curve 162 horizontalism 136-79 and change in monetary policy 161

Index

“horizontalist” approach 1, 3, 4, 43, 52-3, 84, 102 horizontalist response, to structuralist criticism 168-9 horizontalists, and structuralists 155201 household saving 33-4, 35-6 Howells, P.G.A. 155, 156, 284 Hubbard, R.G. 244, 247 illiquidity 180 income, and money 93-6 inflation, Kaldor on 1 10 initial finance

and the endogenous creation of money 22-31 and purchase of investment goods 24-5 interest rate channel 236 interest rate mark-up 71-2, 149, 289 variable 175-6 interest rate/velocity theorem 96 interest rates 1, 4, 48, 95, 239-41, 263

and demand for credit 241 Kaldor on 1 15 and money supply 9|

in New Keynesian and post-Keynesian theory 269 see also rate of interest investment 23, 264 and prior saving 48 investment decisions 283 volatility 263 investment financing 24-6 investment goods, purchase, and initial finance 24-5 IS-LM diagram 220 IS-LM Keynesian model 42, 48 Isenberg, D.L. 187, 195 Jaffee, D. 238, 239, 249 Jarsulic, M. 169 Johnson, H. 184 Kahn, G.A. 250 Kahn, R. 2, 4, 144, 145 “Memorandum to the Radcliffe

335 Committee” 3, 144, 150, 179 monetary views 150-52 rejection of Quantity Theory of

Money 150 on velocity of money 150 Kaldor, N. 3, 4, 10, 42, 54, 66, 123, 124, 125, 129, 148, 156, 158, 160, 161, 165, 171,273, 291 A Keynesian Perspective on Money 106-7

on bank credit 1 10-11 Collected Papers 107, 108, 110 early views 99-117 1939 views 103-9 and endogeneity of money supply I01 on inflation 110 on interest rates 115 “Keynes Theory of the Own-Rates of Interest” 103-4 The Lessons of the British Experiment since the War: Full Employment and the Welfare State 109 New Monetarism article 100, 101, 1 17

and orthodoxy 100-101 and the Radcliffe Committee 104, 109-17 and scarcity of funds 105 views on endogenous money 82 Kalecki, M. 6, 8, 14, 19, 20, 21, 145, 193, 196 investment-profit relationship 16 Kalecki’s principle of increasing risk 176, 193-4, 195, 196 Karecken, J.H. 238 Kashyap, A.K. 58, 233, 234, 237, 248, 250, 252, 254, 255, 256, 262, 263, 268, 290 Kaufman, H. 263 Kennedy, N. 246 Keynes, J .M. 170, 207, 210, 213, 264 Collected Writings 7, 1 1, 29, 30, 31,34, 75, 215, 280 Economic Journal articles 2, 5, 6, 9, 10, 14, 22, 23, 29, 48, 49, 50, 51, I62

336

Index

and the exogenous rate of interest 162-3 finance motive 22-31, 292 The Financial Motive 146 General Theory 2, 5, 6, 8, 9-14, 18, 22, 27, 45,47, 48, 49, 58,

69, 70, 99, 103, 104, 129, 133, 150, 162, 202, 205-6, 212, 216,228, 281, 292 and the horizontalist position 199 speech to House of Lords 163 A Tract on Monetary Reform 23, 27 Treatise on Money 1 1, 16, 45, 63, 145, 147 on uncertainty 21 1- 12 King, J.E. 83, 86 Knight, F. 213, 214, 218 Knodell, J. 49, 52, 128, 129, 204, 218, 221, 251 Kohn, M. 70, 98 Kregel, J. 44, 51,63, 67,77 Lachmann, L.M. 214 Lavoie, M. 1, 8, 15, 16, 17-18, 24, 26, 36, 42, 44, 47, 53, 54, 55, 57, 60, 61, 62, 63, 64, 66, 70, 71, 73, 77, 84, 85, 86, 90, 91, 96, 98, 101, 105, 115, 117, 124, 125, 126, 128, 129, 134, 143, 152, 155, 156, 158, 159, 160, 161, 162,166, 167, 168, 170, 173,l75,l76,178,179, 181, 191, 192, 194, 196, 198, 206, 207, 210,212, 214, 220, 225, 263, 264, 281, 283, 291, 294, 298

LeBourva, J. 32 “lemons” problem 237, 238 lender-of-last-resort argument 165-6 lender’s risk 193, 281 Lemer, A.P. 219 Levratto, N. 269, 275 liability management 168, 176-7, 183-4, 263 liability-asset causality, and Tobin 119-22 lines of credit 23, 54, 157, 256

and revolving credit agreements 254 liquidity preference 34-5, 98, 148, 150, 151, 161 of bank borrowers 193-8 of banks 186-93, 295-6 and horizontalism 291 and the monetary circuit 291-7 liquidity theory 279 loan rate of interest equation 175 loanable funds 149, 161, 172, 286 and financial fragility 96-7 in New Keynesian and post-Keynesian theory 269 loanable funds approach 82, 83 loanable funds theory 207 and Minsky 86, 88, 89 and money supply curve 94 loans market, in multiplier model 60 “locking in" 290 long-tenn investment 114-15 Maccini, L. 236, 237 Mankiw, N.G. 234, 239, 250, 270, 275

Maricic, A. 14 mark-up 21-2, 168 Marshallian methodology 207 Marshallian microeconomics, Keynes’ acceptance 9 Marx, reproduction schemes 16 Messori, M. 14, 16, 24, 169 Minsky, H. 2, 3, 4, 5, 48, 49, 67, 75, 78, 122, 169, 172, 179, 183, 190, 194, 195, 197, 204, 208, 209, 223, 263, 271, 274 American Economic Review article

86, 87, 88, 97 criticism by Rousseas 142 early views 85-99 financial fragility hypothesis 96-7, 187, 195 interest rate/velocity theorem 96 Quarterly Journal of Economics article 86, 87,88, 89 views on endogenous money 82 Minsky’s two-price diagram 188 Mishkin, F.S. 250, 253

Index

Modigliani-Miller condition 252 Moggridge, D. 11, 12 Mondello, G. 15, 16

337 98, 103, 106, 116 and loanable funds theorem 94 slope 155

monetary circuit and liquidity preference 291-7

upward sloping 264 money supply schedule 42, 127

see also circuit approach monetary circuit process 144, 145 monetary circuit theory 4, 6, 8-9 monetary policy changes, and horizontalism 161 and price stability 1 12-13 and velocity of money 123, 126 monetary reflux 145, I46, 147 and profits, in circuit 32-6 monetary theory of production 75

Mongiovi, G. 129, 169, 190 Moody's 253 Moore, B. 1, 43, 49, 54, 55, 57, 59, 70, 71, 72, 73, 75, 77, 85, 101,

“monetized production economy” 42 money circulation in a production economy 41 creation 19-20, 55, 62, 66, 205

French and ltalian literature 56

106, 107, 111, 112,117, 127,

155, 156, 157, 158, 159, 160, l61,163,165,167,170,173, 174,175, 178, 179,180, 181. 185, 190, 192, 207, 208, 209. 230, 282 Horizontalists and Verticalists 171 moral hazard 241 Morgan, D.P. 237, 238, 248, 254, 255, 256, 257, 263, 267 Morris, C. 253, 256

motives, for holding money 204

and credit demarcation between 46-6

Mott, T. 293, 294 multiplier model I26, 203

differences I-2 credit and production, the circuit

Musella, M. 83, 101, 102, 103, 104.

approach 14-37 dynamics 47

and income 93-6, 116, 127-8 Tobin on 122-4 Robinson on 145, 146 role and functions 205 as a stock 216-17, 219-21 and uncertainty 215-19

money market 17 money multiplier model 5, 43, 59. 92, 93, 104, 116, 120, 124, 126,

226-8 and New Keynesians 249-51 money neutrality 334

money supply 17, 36, 209 and banks 29-30 and central banks 237, 254-5

changes 137 exogeneity 89-92 and interest rates 91 and quantity of money in circulation 116

money supply curve 85-6, 87, 91, 92,

I06, 108, 109, 166, 169, 171. 180

natural rate of interest, rejection by post-Keynesians 69-71, 157, 161

Neal, P. 262 Nell, E. 14, 15, 16, 18, 24, 25, 36, 49, 74, 156, 207, 217, 220 neoclassical theory 66-7, 68, 207, 264-5 New Classical approach 234 New Keynesian credit rationing model 240 New Keynesian economics, defined 237-8 New Keynesian literature 257-9

New Keynesian monetary theory 5 compared with post-Keynesian theory 232-77 and the transmission mechanism 232-77 New Keynesian school of credit rationing 84

338

Index

New Keynesian theory foundations 233-8 post-Keynesian cn'tique 259-6|

recent debates 252-7 New Keynesianism, and post-Keynesianism compared 261-70 New Keynesians and early post-Keynesians 270-74 and the money multiplier model 249-51 “New Scarcity” view of endogenous money 60 Ng, S. 245 Niggle, C.J. 83, 208, 209, 222 non-bank financial intermediaries 253-4, 256 O'Donnell, R. 2| I, 2l5 Ohlin, B.G. 206 “open market" model 94 orthodoxy, and post-Keynesians 2023|

overdraft system I35-6 Palley, T. I, 42, 49, 53, 58, 60, 63, 68, 7l, 73, I26, I55, I67, I74, I75, I82, I83, I84, I85, I86, I98, 222, 225, 228», 232, 270

Panico, C. I7, 83, I0l, I02, I03, I04, I06, I08, I09, I66, I69, I71, I80 “paradox of debt” I96 Parguez, A. 9, I6, I8, 20, 21, 28, 3l, 33, 36, 63, 278, 286, 289, 290, 29], 296 Parker Foster I70 Parrinello, S. 9 Pasinetti, L. 69, I00 passivity, of commercial banks I69-74 Pennachi, G. 254 Peterson, B. 247 Pivetti, M. I7, I44, l6l Pollin, R. I, 42, 43, 44, 63, 73, 76, 83, 85, 95, I27, I28, I76, I80, I81, I82, I83, I84, I86, I89, I90, l9l, I94, 22], 222, 223.

225, 227, 264, 270, 275 Ponzi finance 288, 290 “portfolio” approach 268 to endogenous money 2, 3 “portfolio” approach to endogenous money 56-7, 58-62 portfolio behaviour, and the store of value 215-19 portfolio decision 35 portfolio reshuffling l4l, I42 portfolio theory 283 portfolios, changes 204 post-Keynesian critique, of New Keynesian theory 259-6| post-Keynesian macroeconomics, theoretical foundation 44-6 post-Keynesian political economy I post-Keynesian theory 64-78 compared with New Keynesian monetary theory 232-77 credit and money 42-8| post-Keynesianism, compared to New Keynsianism 261-70 post-Keynesians and banking 215

credit and money 46-56 early post-Keynesians, and New Keynesians 270-74 and orthodoxy 202-3| and the two theories of endogenous money 56-64 Poulon, F. I6, 20, 32, 33 precautionary motive 204 price stability, and monetary policy I 12- I3

price stickiness 234, 237, 27l prices, in New Keynesian and post-Keynesian theory 269-70 production and credit 27 money and credit, the circuit approach I4-37 production decisions and prices, in circuit 20-22 profits and monetary reflux 32-6 Robinson on I48 Prowse, S. 246

Index Quantity Theory of Money 3, 6I, 62, 84,99, IIO, II6-I7, I40, I49

Kahn’s rejection I50 Radcliffe Committee 3, I44 and Kaldor I09-I7 Rajan, R. 256 Rao, S.K. 65 rate of interest 85, I03, 279, 288-9 central banks I58-6l exogeneity I06 in New Keynesian and post-Keynesian theory 265-6 and saving and investment 88-9 see also interest rates “Real Bills” doctrine 82 real and monetary sectors, interdependence I34 realism, importance in postKeynesianism 74-5 Regulation Q, abolition 248, 252

Regulation school I6 reserve requirements I84-5

reserves 222-6 and central banks 258

339

risk premium 246, 272 Rizzo, B. ll Robinson, J. 2, 3, 4, I2, 23, 42, 54, 63, 66, 98, I05, I72, I78, 2| I, 294 Accumulation of Capital 3, I4-4 monetary contribution I43-50 “Notes on Various Topics” I45, I49 “Own Rates of Interest" I46 on profits I48 “The Rate of lnterest" I44 Rochon, L.-P. 47, 69, 292 Rogers, C. 69, 70, 7|, I56 Romer, C. 234, 250, 252, 254, 265 Romer, D. 234, 250, 252, 254 Rousseas, S. 3, 9, 26, 49, 52, 76, 85, II6, I66, I76, I77, 183, 204. 209, 2l6 criticism of Minsky I42 early contribution to monetary theory I39-42 on velocity of money I39 Ruhl, C. I29, I69, I90 Russell, T. 238, 239

deposits and loans, reverse causality 72-3

excess reserves I86 exogeneity 92-3 and loans, in New Keynesian and post-Keynesian theory 268 role I85-6

role in post-Keynesian theory 203 and the structuralists 225 supply of I63-8 reserves to liabilities ratio 286-7 reverse causality, among reserves, deposits and loans 72-3 “revolutionary” approach to

endogenous money 2, 56-7, 624, I44, I56, 268 revolving credit agreements, and lines of credit 254 revolving fund of finance 22, 30-3l, 32, 48 risk 2I2-I3 and lending I89 and uncertainty 2 I 3

Sardoni, C. ll, 50 saving 33 excess saving, and debt 97-9

investment, and the rate of interest 88-9 savings and investrnent, reverse causality I49-50, 270 Sawyer, M. 6, 53, 55, 65, 66, 74, 220, 67 scarcity 57 scarcity of funds I I I, 203, 225-6 and Kaldor I05 Schaller, H. 245

Schmitt, B. 8, I4, I6 Schredt, S. 248 Schumpeter, J.A. I6, 25 Schwartz, A. II9 screening mechanisms 243-4 Screpanti, E. 16,68, I82, I86, 208, 260, 26|, 267, 278, 279 Sebastiani, M. 6, 65 Seccareccia, M. 4, I4, 24, 25, 26, 27,

340

Index 35, 36, S2, 55, 65, 69, 70, 7l, 90, I72, I76, 178, 181, I85, I91, I96, 262, 264, 269

Shackle, G.L.S. I l, 50, 69, 21 I, 212, 213 Shapiro, N. 63 Sharpe, S. 256 short run, definition I8 Sinai, A. 263 Skouras, T. 99 Smith, P. 49, 50, 51, 204 Smith, W.L. 93 Smithin, J. 58, 63, 69, 70, 71, I01, I56, I58, I61 Solow, R.M. 67 speculative motive 204 Sraffa, P. 66, I44 Standard and Poor‘s 253 Stein, J. 58, 290 Stein, J.C. 233, 234, 237, 248, 252, 254, 255, 256, 262, 263, 268 Stiglitz, J.E. 238, 239, 240, 241, 242,

243, 244, 250, 256, 257 Stock, J. 249 store of value, and portfolio behaviour 215-I9 “structuralist” approach to endogenous money I, 3, 4, 43, 85, 229 structuralist criticism, and the horizontalist response I68-9 stnicturalists 179-98 and horizontalists 155-201 and reserves 225 and uncertainty 218 Sylos-Labini, P. 129

logical time versus expectational time 2| 1 Tobin, J. 3, I26, 251 “Commercial Banks as Creators of ‘ Money”’ I18 early views II7-25 and liability-asset causality I 19-22 on money and income I22-4 “Money and Income: Post Hoc Ergo Propter Hoc?" 1 18, I24-5 and the “Old View" I I8-19 and orthodoxy I I9 views on endogenous money 82 Tookean tradition, revival I43 Townshend, H. I2, I3 transactions motive 204 transmission mechanism 235-6, 262 and New Keynesian monetary theory 232-77 Trautwein, H.M. 1, 76 Trivithick, J. I06 Tumer, M. 151 uncertainty 35, 43, 66, 265 and asymmetric information 260-61, 266-7 and banking behaviour 278-81 and banks 298 and credit 216, 219 and endogenous money 173-4 and investment decisions 21 Keynes on 211-12 and money 215-17, 219 in post-Keynesian theory 75-7, 210-14 and risk 213

target rate of retum, banks 283, 285 Targetti, F. 49, 83, 84, IOI, 102, 103 l05,l07,108,109,1l5,ll6 Termini, V. 2| I Terzi, A. 35 Thakor, A.V. 243 theory of monetary creation 205 theory of monetized economy 205 Thirlwall, A.P. 110, 112 time 44, 21 I historical time 47, I56

and the structuralists 218 Vallageas, B. 19, 24 van Ees, H. 234, 259, 26|, 265, 266, 267, 269, 287 variable interest-rate mark up I75-6 variable velocity of money 177-9 Variato, A.M. 233, 260, 265, 266, 275, 279 velocity of money 13, I9, 24, 58, 87, 92-3, 93-4, II3-I4, II8, I28,

Index 169, 221-2 Davidson on 134 Kahn on I50 and monetary policy I23, I26 and the money multiplier model 127 Rousseas on 139 variable I77-9 “velocity-interest rate relation” curve 95. Vickers, D. 21 1 wages 24 Walrasian equilibrium conditions 264 Walters, B. 67 Weintraub, S. 51, 67, I32, 136, I37, I38, I39, 208, 209 Weiss, A. 238, 239, 240, 242, 243, 244, 257

341

Wells, P. 5|, 219, 294 Wette, 1-I.C. 244 Wicksell, K. 16, 70, I44, I50 Winnett, A. 207 Wolfson, M.H. 171, 172, 233, 259. 260, 281, 282, 283, 285, 287, 290 Wray, R. 11,48, 49, 52, 63, 69, 72, 75, 82, 83, 85, 87, 91, I22, I23, 128, 159, I60, I66, 169, 170, I74, I80, I84, 186, I88, 189, I90, 191, 192, I93, I95, I99, 204, 218, 219, 223, 230, 266, 286, 287, 293, 294, 295, 298

Young, D. 67

Zerbato, M. 16,21

Louis-Philippe Rochon is the Stephen B. Monroe Assistant Professor of Economics and Banking at Kalamazoo College, USA; and co-director of the Center for Macroeconomics and International Trade

Advance Acclaim ‘Rochon provides a clear and unambiguous presentation of the key elements that comprise a theory of endogenous credit-money, showing in particular the similarities and the dissimilarities with the ideas recently espoused by New Keynesian economists. Rochon takes a clear stand on all the debates that he tackles. l-le usually puts forth a dissident view, going against conventional wisdom as accepted by most of his fellow post~Keynesiaii colleagues. l expect the book to generate a large amount of controversy among heterodox economists. lt is an important addition to the literature, as Rochon provides a coherent alternative framework.’ |\/larc Lavoie, University of Ottawa, Canada

‘This book must be read by all those who despair of the capacity of heterodox economists to go beyond the eternal critique of neo-classical economics and escape from the temptation of impossible bridges between approaches that cannot be reconciled . . . Needless to say that the author is to be praised for his beautiful style. his scholarship and his methodology. /\lo proposition is spelled out without being rooted in a remarkable critique of the literature and scientific logic. The reader is now convinced that at last heterodox economics is born.‘ A Alain Parguez, University of Franche-Comte, France and University of Ottawa, Canada