Guns and Rubles: The Defense Industry in the Stalinist State 9780300151701

Citation preview

The Yale-Hoover Series on Stalin, Stalinism, and the Cold War

This page intentionally left blank

Edited by M A R K H A R R I S O N

Guns and Rubles T H E D E F E N S E I N D U S T RY I N T H E S TA L I N I S T S TAT E

Published in cooperation with the Hoover Institution on War, Revolution and Peace, Stanford University Yale University Press New Haven & London

Copyright ∫ 2008 by Yale University. All rights reserved. This book may not be reproduced, in whole or in part, including illustrations, in any form (beyond that copying permitted by Sections 107 and 108 of the U.S. Copyright Law and except by reviewers for the public press), without written permission from the publishers. Set in Sabon by Keystone Typesetting, Inc. Printed in the United States of America by Sheridan Books, Ann Arbor, Michigan. Library of Congress Cataloging-in-Publication Data Guns and rubles : the defense industry in the Stalinist state / edited by Mark Harrison. p. cm. — (The Yale-Hoover series on Stalin, Stalinism, and the cold war) Includes bibliographical references and index. isbn 978-0-300-12524-5 (cloth : alk. paper) 1. Defense industries—Soviet Union. 2. Weapons industry—Soviet Union. 3. Soviet Union—Military policy. 4. Soviet Union—History—1925–1953. I. Harrison, Mark, 1949– hd9743.s672g86 2008 338.4%73550094709043—dc22 2007051391 A catalogue record for this book is available from the British Library. The paper in this book meets the guidelines for permanence and durability of the Committee on Production Guidelines for Book Longevity of the Council on Library Resources. 10 9 8 7 6 5 4 3 2 1

To Anne

This page intentionally left blank

Contents

Foreword by Paul R. Gregory ix Preface xiii Acknowledgments xix Abbreviations, Acronyms, Technical Terms, and Conventions xxi Note on References xxv 1

The Dictator and Defense 1 Mark Harrison

2

Before Stalinism: The Early 1920s 31 Andrei Sokolov

3

Hierarchies and Markets: The Defense Industry Under Stalin 50 Mark Harrison and Andrei Markevich

4

Planning the Supply of Weapons: The 1930s 78 Andrei Markevich

5

Planning for Mobilization: The 1930s 118 R. W. Davies

viii

Contents

6

The Soviet Market for Weapons 156 Mark Harrison and Andrei Markevich

7

The Market for Labor in the 1930s: The Aircraft Industry 180 Mikhail Mukhin

8

The Market for Inventions: Experimental Aircraft Engines 210 Mark Harrison

9

Secrecy 230 Mark Harrison Afterword 255 Contributors 261 Index 263

Foreword paul r. gregory

The collapse of Soviet-style communism suggests that it had little to recommend itself. At its peak, one-third of the world’s population lived in countries claiming to be communist or part of the ‘‘socialist world.’’ Now trivial numbers of people live in the communist hold-out states of Cuba and North Korea. Although there appears to be an upsurge in leftist ideology in Latin America, no national leaders are proposing to create a Soviet-style economic or political system. We still have to come to terms with what appears to be the greatest achievement of the administrative-command system—the victory of the relatively backward Soviet Union over the Nazi forces in World War II. Is the ability of the Soviet-style system to prioritize and to concentrate resources under a centralized command an inherent strength of the system? In more practical terms, it raises the counterfactual question: Could the Soviet Union have withstood the onslaught of German military might in 1941 with an alternate politicaleconomic system? Did Stalin’s brutal policies—forced industrialization, liquidation of class enemies, extreme centralization—in a sense pay off in the military sphere? Was the payoff a longer-run one? Was the Soviet economy, as many indeed hypothesized, segregated into two separate worlds—an inefficient civilian sector hobbled by all the problems of a planned economy and a highly efficient military sector that was able to compete with or even surpass

ix

x

Paul R. Gregory

the most advanced capitalist countries? In effect, we are asking: Does the administrative-command system work well for defense while working poorly for most everything else? All societies resort to some degree of command in organizing their defense sectors. The military is the primary or sole consumer of defense production; the level of expenditures is determined by political rather than market forces; and there are inherent problems of organizing production with one buyer and a few large sellers. No country, democratic or authoritarian, appears to have found good solutions to these problems. Most countries are plagued by scandalridden and wasteful defense sectors in which incentives to produce high-quality products at low prices are distorted. Does Stalin’s Soviet Union provide some positive guidance even for the problems of democratic capitalist countries in meeting their defense needs? Soviet secrecy prevented scholars from seeking answers to such questions. Secrecy was especially intense with respect to the two most hidden areas of Soviet life, namely, defense and repression, manifested most vividly in the Gulag system. With the opening of the Soviet state and party archives, we are now able to penetrate these veils of secrecy and study the actual working arrangements of the Soviet military-industrial complex and of the Gulag itself. This book represents the deepest and most complete effort yet to describe the actual workings of the Soviet military-industrial complex in its formative years under Stalin. Written as a joint undertaking by a group of highly qualified and influential scholars of Soviet economic and military history, Guns and Rubles describes how the Soviet military-industrial complex was created, how it was planned, how producers and consumers of defense goods interacted, and how they innovated. This examination of the reality of the Soviet military-industrial complex is based not on stereotypes or theory but on the actual documents and orders that ran the military system. The picture that is painted is complex. It shows a leadership, headed by Stalin and his Politburo and aided by the military commanders, struggling to organize the planning, production, and distribution of defense goods and their innovation for the purpose of defending themselves against foreign and even domestic enemies. The mature form of organization came about only after a ‘‘failed’’ experiment with a market-oriented system during the New Economic Policy of the 1920s. The result was a bewildering organizational maze, which kept the military at arm’s length from industry, according to the dictates of Stalin. Stalin was particularly intent on preventing military control of defense industry, although he welcomed the support of both groups for his policy of forced industrialization. The organizational structure of military production was more complicated than that of the civil-

Foreword

xi

ian sector because its high priority required extra controls by third parties, including the secret police. One might expect that defense’s high priority and intense scrutiny would have prevented the abuses of low quality, manipulation of statistics, and waste that characterized civilian production. But the story the authors tell is similar to that of the civilian sector: defense industry leaders and managers also behaved opportunistically, sought a wide degree of independence, and followed their own narrow interests. All societies must plan for the eventuality of war. It would be highly wasteful to have an economy in a constant state of full readiness for war. Instead, societies should use their resources for civilian needs during peacetime, while being able to mobilize these resources for defense in time of war. ‘‘Mobilization planning’’ might indeed be an area where a command system might outperform a market economy because planning for war should be a simple extension of planning everything else. Although there was mobilization planning from the first days of the Soviet planning system, it appears not to have prepared the Soviet Union for the ultimate test of World War II, which caught the USSR ill prepared. Mobilization planning, like planning in the civilian sector, appears to have been more on paper than in fact. History has shown that countries cannot remain militarily competitive without innovation. A nation that has propeller-driven planes cannot compete with one that has jet aircraft. Under the Stalinist system, innovation was a planned activity. Proposals from competing labs were examined by central commissions. The progress of innovation was tracked. Some innovators proposed promising leads and others used society’s resources to pursue scientific dead ends, such as the case study of the steam-powered aircraft discussed in this book. As in other societies, scientists and engineers were driven by both material incentives and the prospect of fame. In fact, this book describes a scientific community that was willing to work largely for scientific reputation but required higher material rewards when work became routine. Military matters are subject to more secrecy than other economic matters. It is correct that military secrets should be kept from enemies, real and imagined. However, planning and production require information, and if information is unduly repressed by secrecy requirements, the task of planning, production, and distribution is complicated. In Stalin’s Russia, secrecy rules were pressed to their limits. The rules concerning secrecy were themselves secret! The topsecret defense branch of the State Planning Commission was not allowed to see the top-secret plans for war. If so, how could they rationally direct the economy to prepare it for war? The final chapter of Guns and Rubles examines these fundamental issues to determine what happens when secrecy is taken beyond its optimal boundaries.

xii

Paul R. Gregory

The reader of Guns and Rubles will find much that is strange and exotic at first sight. As a scholar who has studied the working arrangements of the Soviet civilian economy under Stalin, I also find an underlying story that is familiar: here are individuals and organizations engaged in work of national importance, under close scrutiny, hedged around by rules and restrictions, motivated by private goals and personal interests, trying to make the best of a difficult situation for themselves. Despite much greater scrutiny and a higher probability of repression than in the civilian economy, there remained a sphere of informal activity and room for personal maneuver. The Soviet defense industry was at the focus of continuous interplay between the ruler and his subjects, state interests and private interests, finance and politics, and foreign policy and domestic policy. Guns and Rubles lets us watch these interactions as they evolved through Stalin’s quarter-century.

Preface

This book aims to show students, specialists, and interested lay readers what has been learned from the formerly secret Soviet archives about the economics of the Soviet defense industry under Stalin. It is the result of a threeyear collaboration between Russian, American, and British scholars. Today, the Soviet economy is generally considered to have failed. Relative to the rest of the Soviet economy, however, the Soviet defense industry was an outstanding success: it produced the weapons that defeated the armies of Germany and Japan in World War II and kept the Western world under the threat of ‘‘mutual assured destruction’’ for the half-century that followed. Stalin was notoriously obsessed with guns and their technical detail. With his lieutenants, he subjected the production and procurement of military equipment to intense, frequent scrutiny. One would perhaps expect no less of a dictator in a ‘‘planned economy.’’ But this was a mechanism that ensured scrutiny, not effective allocation. The system that actually steered resources and determined outcomes was quite different. Every year, Stalin gave the Red Army a purse of rubles and told it to order the guns it needed from the defense industry suppliers in an internal market for weapons and military equipment. Our book’s title, Guns and Rubles, conveys this simple truth: although surrounded and shot through with bureaucracy, the fundamental process of weapon procurement was one of market-like exchange. Of course, people do

xiii

xiv

Preface

not necessarily behave in the same way in ‘‘market-like’’ settings as in real markets; the differences and similarities between markets and ‘‘market-like’’ exchange are a problem that we will explore and seek to understand. In short, our book uses the case of the defense industry to direct a clear searchlight into the workings of Stalin’s dictatorship, military and economic planning, and the industrial organization of the Soviet economy. It is an investigation in both economics and history. For historians, it seeks to shed light on questions such as these: ∞ Why did Stalin want a large-scale defense industry? How did its promotion help him to achieve and retain power? ∞ How did Stalin’s generals and economic leaders plan for war? How did they organize the defense industry to supply the weapons they wanted? ∞ What was it like to work in the defense industry for the managers and workers engaged in the production of secret weapons, and the inventors responsible for secretly developing new ones? ∞ Why was secrecy so pervasive, and how was it exploited for private benefit? This book is also an economic investigation into public procurement and the organization and development of a major industry in a command system ruled by a dictator. We learn that in the command system most people were able to pursue their own private goals while appearing to carry out the commands they were given. Formally there was little scope for market behavior, yet informal and internal markets were everywhere. Much of the book is given over to examining how reality differed from appearance and how informal relationships supported or disrupted those that were formally laid down by decrees and statutes. For economists, the book asks: ∞ In what ways did dictatorship and a command system frame the industrial organization of defense? What difference did it make to public procurement that the dictator could, and occasionally did, shoot the contracting parties? ∞ What were the personal goals of the workers, managers, and designers in the defense industry? How did they evade the dictator’s scrutiny, and exploit the workings of the command system, to pursue these private interests? ∞ Given the universal pursuit of self-interest, how did the command system work at all? The nine chapters of the book are organized as follows. Three chapters provide an introduction. The first two provide analytical and historical context. Chapter 1 examines why Stalin wanted a defense industry; Chapter 2 considers how he was able to build one. On that basis, Chapter 3 goes on to investigate how he organized it. In more detail, Chapter 1, ‘‘The Dictator and Defense,’’ analyzes Stalin’s choices over military power and political repression as instruments for holding

Preface

xv

political power in the face of foreign and domestic threats. Because the threats were interactive, the policy combination had to be determined simultaneously. One problem was that although military power was the more efficient instrument for countering a foreign threat, it could be adjusted less rapidly than repression. The chapter shows that the empirical pattern of the 1930s conforms to the predicted result. In some years Stalin responded to an unanticipated increase in perceived foreign threats by imposing a substantial excess burden of repression on Soviet society. Chapter 2, ‘‘Before Stalinism: The Early 1920s,’’ goes back to the formative stage of the Soviet defense industry. Competition and profit-seeking were never strong features of the Russian defense industry before the Revolution. World War I and the Russian Civil War profoundly influenced interwar perspectives on the Soviet defense industry and accentuated this characteristic in the process. The defense industry failed to adapt to market conditions under the New Economic Policy: it produced at a loss, depended heavily on budgetary subsidies, and still failed to meet the demands of the armed forces in virtually every field of armament. The blame, at first laid on those in charge of the defense industry, was directed more and more specifically against its ‘‘bourgeois’’ specialists. In the process the Red Army staff became enthusiastic advocates of forced industrialization under a command system through which they hoped to gain direct influence over defense industry personnel and allocations. Chapter 3, ‘‘Hierarchies and Markets: The Defense Industry Under Stalin,’’ summarizes the often bewildering organization of the supply of weapons and military goods in the Soviet economy, and the institutional channels through which demands on it were voiced and made effective. The general setting was one of a hierarchical command system. There were also relatively formalized market-like structures within the system, which are distinguished from formal and informal outside markets. The most important of these internal markets organized the exchange of guns for rubles. Finally, there were overlapping systems of third-party regulation by an array of agencies responsible for planning, arbitration, audit, and state security. Three chapters that follow examine how the real practices of the administrative command system diverged in various ways from appearances. Chapter 4, ‘‘Planning the Supply of Weapons: The 1930s,’’ describes the procedures for planning the Soviet defense industry in the interwar period and analyzes its limitations. Three features differentiated the planning of the defense industry from that of the civilian economy. First, the supply of national defense in the broadest sense had high priority and received the close attention of the country’s top leadership. Second, the process of military and economic planning for defense that created the context of defense industry plans had a strongly forward-

xvi

Preface

looking character and generated increasing demands through the interwar years. Third, the detailed planning of defense industry was carried on simultaneously in two separate bureaucracies, a military hierarchy preoccupied with formulating demands and an industrial hierarchy the task of which was to organize supply; planners made strenuous efforts to reconcile supplies and demands in the defense industry, and met with limited success. Chapter 5 deals with ‘‘Planning for Mobilization: The 1930s.’’ In every society planning for the contingency of war is conducted under conditions of great uncertainty, and requires competing interests to overcome mutual suspicion and resistance. The Soviet Union under Stalin was no exception. This chapter examines the conduct of planning for wartime economic mobilization in the 1930s. It suggests that the attention devoted to mobilization planning followed a U-shaped curve, being greater in the first and last years of the decade. In the early 1930s there was much grandiose talk that was quite important in focusing attention on the long term. In the mid-1930s the increasing complexities of planning current defense production, already discussed in Chapter 4, somewhat distracted the authorities from forward-looking mobilization tasks. In 1938 mobilization planning was reorganized and was conducted much more seriously in the run-up to World War II. Chapter 6 looks at ‘‘The Soviet Market for Weapons.’’ Military marketplaces display obvious inefficiencies under most arrangements, but that of the Soviet Union was unusual for its degree of monopoly and exclusive relationships between buyer and seller. This presented a particular problem for the quality of weapons. This chapter analyzes the problem of quality in terms of an issue that is well known in market economies, the hold-up problem. When A has had to make a prior commitment to a relationship with B, B can ‘‘hold up’’ A for the value of that commitment. This roughly describes the power of Industry over the Army in the Soviet defense market. The normal use that Industry made of this power was to default on quality. The Army’s counteraction took the form of deploying agents through industry with the authority to verify quality and reject substandard goods. The struggle ended not in victory for one side but in a compromise that, being illegal, had to be hidden from the dictator. The last third of the book examines three special topics in the development and organization of the defense industry. Chapter 7, ‘‘The Market for Labor in the 1930s: The Aircraft Industry,’’ is about working in a priority branch of the Soviet economy. It describes trends in employment, recruitment and turnover, the composition of the workforce, the ‘‘working culture,’’ and the impact of policy. Policy pulled the industry in different directions: demands for higherquality aircraft built to more exacting standards lifted it up, but urgent re-

Preface

xvii

quirements for rising quantities dragged it down. The high demand for labor sucked new recruits into the industry in ever-increasing numbers that lacked experience, skills, and work discipline. The authorities tried to resolve their dilemma using a varying mix of repression, regimentation, and rewards. Despite the context of a command system and intense scrutiny from above, it was not possible to improve the quality of effort or working practices by threats and punishments alone. Previous chapters deal with various aspects of the production and procurement of weapons. But the secret of the Soviet defense industry’s successes began with research and design. Chapter 8 looks at ‘‘The Market for Inventions: Experimental Aircraft Engines.’’ The institutions of a command system were poorly equipped to direct the research process. The officials who had to fund and manage research started from a set of unknowns: they did not know how technology would evolve, or where were the best places to look for breakthroughs; worse still, they knew less about science and technology than the designers and so were not well placed to evaluate the personal and team qualities of the people they were supposed to manage. Despite this the process worked. This chapter looks at the Soviet search for new techniques of aircraft propulsion in the 1930s and 1940s, which created a ‘‘market for inventions.’’ This book is about what was once a most secret aspect of the Stalinist state. In Chapter 9, ‘‘Secrecy,’’ we consider why and how the state guarded its secrets, using the defense industry as an example. Stalinist secretiveness was allembracing, but it did not only keep secrets from society or the foreign enemy. Soviet officials also became adept at keeping some secrets from one another. But they were curiously careless about upholding secrecy in other cases. One of the most interesting aspects of secrecy is that although it had an obvious rationale for the dictator, it is less clear why his individual agents should have chosen to uphold it. The prevalence of leaks, after all, is one of the things that makes some societies more open. The Soviet Union kept some secrets only by threatening dire punishments for disclosing them. In other cases it turns out that secrecy was one of those structures of Soviet life to which everyone could adapt and from which anyone could learn to turn a ruble. We close the book with an afterword that draws some lessons and offers some concluding reflections.

This page intentionally left blank

Acknowledgments

I thank the Hoover Institution on War, Revolution, and Peace for financial support that made possible the collaboration underlying this book. The Hoover Archive and its director, Elena Danielson, opened its facilities to our group, provided essential moral and technical support, tolerated our individual and collective requirements, and put up with our noisy conversations and other disruptive behavior with adult patience. I will long remember the good humor of our annual meetings at Hoover, and particularly the evenings that we spent together at the Stanford Linear Accelerator Center guesthouse. These nightly gatherings, chaired by our leader, Paul R. Gregory, quickly became known as the kolkhoz (collective farm). In the kolkhoz we pooled our findings of the day with the produce of each person’s private foraging in a local capitalist supermarket. We shared everything on the communist principle ‘‘from each according to ability, to each according to need.’’ In the years over which this book has been prepared, my co-authors have shared their documents and findings and cooperated with editorial requirements far beyond the norm. I have benefited greatly from the advice and comradeship of R. W. Davies, Andrei Sokolov, and Andrei Markevich, contributors to this book, and from the selfless collegiality of Leonid Borodkin, Simon Ertz, and James Heinzen. Paul R. Gregory has inspired and guided our inquiries in an extraordinary way.

xix

xx

Acknowledgments

I thank the University of Warwick for supporting my research with generous study leave. The final stages of preparation of this book were greatly eased by the European Commission’s award of a Marie Curie Incoming International Fellowship to Andrei Markevich in December 2005, bringing him to Warwick for two years to undertake research on ‘‘Dictatorship, Hierarchy, and the Stalinist Economy: The Soviet Defence Industry, 1929–1953’’ (MIF1CT-2005-021656). I thank Richard Sousa from Hoover and Jonathan Brent and Vadim Staklo at Yale for their concerted efforts to bring this book to publication. I also thank two anonymous referees for their thoughtful and constructive criticisms. I thank the editors of Comparative Economic Studies for permission to reproduce Andrei Sokolov’s article ‘‘Before Stalinism: The Defence Industry of Soviet Russia in the 1920s,’’ vol. 47, no. 2, pp. 437–55, with minor amendments, as Chapter 2. Figure 9.2 is reproduced by kind permission of United Media. Finally, I thank Anne Harrison for her love and support, and for helping to keep me laughing. This book is dedicated to her.

Abbreviations, Acronyms, Technical Terms, and Conventions

Much of this book is about the triangular relationship of three significant actors: Stalin, his army, and the defense industry. To simplify a complex reality the authors of this book often write ‘‘the Army’’ to stand for the Soviet ministry of defense, including the Red Army and Navy command and supply staffs. They also write ‘‘Industry’’ to stand for the industrial ministries, and their factories, that supplied the Army’s main equipment. Both Industry and the Army were subject to frequent administrative reorganizations that are illustrated in figures and analyzed further in Chapters 2 and 3. Even to refer to ‘‘ministries’’ involves some simplification. Soviet government departments were called ‘‘people’s commissariats’’ from 1917 until they were renamed ministries in 1946; we call them ministries throughout. Specifically, we translate NKTP as the ‘‘heavy industry ministry,’’ NKOP as the ‘‘defense industry ministry,’’ and so on. In conformity with this we translate VSNKh (Supreme Council of the National Economy), which managed state industry from 1918 to 1931, as the ‘‘industry ministry.’’ From 1927 most Soviet defense factories were designated by number, for secrecy, and name of the ministry that ‘‘owned’’ them. In place of ‘‘factory no. 24 of the people’s commissariat (i.e., ministry) of the aircraft industry’’ we write ‘‘aircraft factory no. 24.’’ We avoid the use of Russian and Soviet technical terms in the text as far as

xxi

xxii

Abbreviations, Acronyms, & Conventions

possible, but a few are relatively well known, unavoidable for lack of a convenient English equivalent, or required for precision. Some refer to the agencies described in this book and others denote the archives used by the authors. A summary follows.

Agencies chief administration. The chief administration ( glavnoe upravlenie) was the first-level specialized subdivision in a Soviet ministry or people’s commissariat. Glavnoe upravlenie was usually abbreviated to GU, so GUGB was chief administration of state security of the NKVD (see below), Gulag was its chief administration of labor camps, and so on. commissar, commissariat. The first government following the October 1917 revolution was made up of ‘‘people’s commissars’’ rather than ‘‘ministers’’ because the Bolsheviks wished to underline the break with the past. Each people’s commissar headed a commissariat—that is, ministry. This practice continued until 1946, when Stalin ordered all the commissariats to be renamed ministries. Gosarbitrazh. The state arbitration service (Gosudarstvennyi arbitrazh), responsible for quasi-judicial settlement of pre- and post-contract disputes among government departments. Gosplan. The state planning commission (Gosudarstvennaia planovaia komissiia) of the Sovnarkom (see below), later Council of Ministers. Gulag. Chief administration of corrective labor camps (Glavnoe upravlenie trudovo-ispravitel’nykh lagerei) of the NKVD (see below), later MVD. KPK. The party control (i.e., audit) commission (Komitet partiinogo kontrolia), established in 1934. MVD. Ministry of internal affairs; see NKVD. NEP. The New Economic Policy (novaia ekonomicheskaia politika) of a mixed economy pursued from 1921 to 1929. NKVD. The people’s commissariat (until 1946, then ministry) of internal affairs (Narodnyi komissariat vnutrennykh del), responsible for internal security and forced labor from 1934. OGPU. The unified chief political administration (Ob’’edinennoe glavnoe politicheskoe upravlenie), responsible for internal security and forced labor (until 1934). Political Bureau, Politburo. The Politburo was the executive core of the Central Committee of the Bolshevik, later Communist, Party. In practice most decisions that were attributed to the Central Committee and the Soviet

Abbreviations, Acronyms, & Conventions

xxiii

government in public had previously been issued from the Politburo. From 1932 until his death in 1953 Stalin personally took many of these decisions, and no decisions were taken at this level without his approval. secret or security police. A Soviet innovation of the Russian civil war was the Cheka, the All-Russian Extraordinary Commission for combating counterrevolution and sabotage; che and ka are the Russian initials of ‘‘extraordinary commission.’’ Its successor organizations were, in order, the GPU (chief political administration), OGPU (‘‘unified’’ GPU), the chief administration for state security within the NKVD (see above), the NKGB and MGB (people’s commissariat until 1946, then ministry for state security), the chief administration for state security of the MVD (ministry of internal affairs), the KGB (committee for state security), and, in post-Soviet Russia, the FSB (federal security service). Sovnarkom. The council of people’s commissars (sovet narodnykh komissarov)—that is, the government of the USSR, renamed the council of ministers in 1946. The chairman of Sovnarkom or the council of ministers was the prime minister of the USSR. STO. The Council of Labor and Defense (Sovet truda i oborony) (until 1937). Voenprom. The Russian equivalent of ‘‘Industry’’: literally, the military industry (voennaia promyshlennost’). Voenved. The Russian equivalent of ‘‘the Army’’: literally, the military department (voennoe vedomstvo) or defense ministry.

Archives APRF. Archive of the President of the Russian Federation (Arkhiv Prezidenta Rossiiskoi Federatsii), Moscow. GARF. State Archive of the Russian Federation (Gosudarstvennyi Arkhiv Rossiiskoi Federatsii), Moscow. Hoover/RGANI. Hoover Institution (Stanford, California), documents from the Russian State Archive of Recent History (Rossiiskii Gosudarstvennyi Arkhiv Noveishei Istorii), Moscow. Hoover/RGASPI. Hoover Institution (Stanford, California), ‘‘Archives of the Former Soviet State and Communist Party’’ from the Russian State Archive of Social and Political History, Moscow. RGAE. Russian State Economic Archive (Rossiiskii Gosudarstvennyi Arkhiv Ekonomiki), Moscow. RGASPI. Russian State Archive of Social and Political History (Rossiiskii Gosudarstvennyi Arkhiv Sotsial’no-Politicheskoi Istorii), Moscow.

xxiv

Abbreviations, Acronyms, & Conventions

RGVA. Russian State Military Archive (Rossiiskii Gosudarstvennyi Voennyi Arkhiv), Moscow. TsAODM. Central Archive of Social Movements of Moscow (Tsentral’nyi Arkhiv Obshchestvennykh Dvizhenii Moskvy), Moscow. TsMAM. Central Municipal Archive of Moscow (Tsentral’nyi Munitsipal’nyi Arkhiv Moskvy), Moscow.

Note on References

In the case of published works the reader will find citations worked into the text, giving author and date of publication. Fuller bibliographic details are then found in the list of references at the end of the chapter. The reader should note carefully the publication date of Russian sources; those published in 1985 or before were subject to the full rigor of Soviet censorship and secrecy legislation, and although the censorship weakened somewhat in the late 1980s, it was not until the end of 1991 that Russian publications on defense-related matters ceased to be tightly controlled. Citations from the Russian archives either contain little information of use to the nonspecialist reader, or else are too lengthy to retain within the text. Therefore, all such citations appear in endnotes to each chapter. These are typically of the form ‘‘RGAE, 2097/1/64: 8–24’’ where RGAE denotes the archive (Russian State Economic Archive), 2097 is the departmental collection number, 1 is the catalog number, 64 is the file number, and 8–24 are the folio numbers of the first and last pages. A reference to ‘‘ff. 69–69ob’’ means that the folio is written on both sides (‘‘ob’’ stands for obverse); a reference to folios ‘‘132–131’’ means that the archivist has paginated the file in reverse, starting from the back. Standards of best practice vary with regard to the provision of additional information about the document cited. Most Russian-trained his-

xxv

xxvi

Note on References

torians tend to parsimony. Some Western historians prefer to give full details of the document’s date, authors, and addressees. A middle course, sometimes but not uniformly preferred in this book, is to give at least the date of the document that is cited.

1

The Dictator and Defense mark harrison

Why did Stalin spend as much as he did on defense? Given their cost, why did he want a big army or large specialized defense industries? At first sight these questions seem too trivial to deserve much thought. There are several obvious answers. The trouble is that the answers that are obvious are not necessarily consistent, and none provides a satisfying explanation of the stylized facts of Soviet military-economic development. The economic and historical literatures on this subject offer several different explanations of Soviet policies with regard to defense, which I will summarize under three headings: Soviet Defense Against External Threats? According to informed historians such as Lennart Samuelson (2000), Soviet defense spending responded to perceived external threats. In the hands of an experienced economist such as Paul R. Gregory (1974) this hypothesis comes to down to supply and demand: there was a demand for national security as a public good and the economy supplied the inputs. Their supply price was likely to fall with economic growth, which increased supply, while the demand price varied directly with the number and intensity of external threats. This approach suggests that we may look at the motives driving Soviet allocations to defense and the defense industry in much the same way as for any state. But this implication is also the main problem: it seems unsafe to

1

2

Mark Harrison

assume anything of the kind. The Soviet state was not like other states, judged by its extreme propensities to secrecy and dictatorship, or at least it lay at one extreme of the spectrum of states. At the most superficial level, we know that dictators like guns and uniforms and rule by force. This makes an explanation that owes nothing to the facts about Stalin unsatisfying to say the least. Soviet Aggression: The Export of Revolution? According to Viktor Suvorov (1990), Stalin built up his army and defense industry because he planned an aggressive war to conquer western Europe; his plans were frustrated a first time in 1941 because Hitler struck first, and a second time in 1944 when the western Allies opened a Second Front against Germany by landing in France. This view has considerable support among Russian historians, many of whom are generally disposed to believe the worst about Stalin. A few Western scholars share this view (Raack 1995; Weeks 2002). But there are also some problems with the aggressive-war hypothesis. One problem is its historical origins, which lie in the self-serving rationalizations of Hitler’s policies offered after the war by former National Socialists in Germany. Another is that it is based largely on memoir and hearsay; the evidence found in the archives does not support it; for these reasons most qualified Western scholars now reject it (for example, Gorodetsky 1999; Uldricks 1999; Mawdsley 2003). In short, Stalin’s generals may have planned for the contingency of an offensive war, but Stalin did not seek to create the opportunity for one. A Soviet Military-Industrial Complex? Sooner or later most Russian scholars who have written about Soviet defense industry (for example Simonov 1996; Bystrova 2000) have ended up using the term ‘‘military-industrial complex’’ to describe the simple fact that the Soviet armed forces and defense industry were large and interrelated. In the West, however, this term has often been loaded with a further implication (discussed by Rosen 1973; Aspaturian 1973; Agursky and Adomeit 1978; Holloway 1982): the idea that powerful military leaders might be colluding with the leaders of the defense industry to boost the military budget and so extract vast sums of money and resources from the economy to build their power and prestige. Applied to a Soviet context, this suggested that the Soviet Union ended up spending vast sums on its army and defense industry mainly because they were there. Historically, there is evidence for and against. On the plus side, the next chapter by Andrei Sokolov will show that the military were already demanding forced industrialization in the mid-1920s as a precondition for developing the defense industry on the scale they claimed was necessary to counter external threats; it may be that Stalin bought their loyalty by subscribing to their program. Once he had made himself dictator, however, the idea loses its fit to the

The Dictator and Defense

3

facts; it requires, for example, that Stalin, although dictator, was not a very good one since he could not stop his subordinates from combining to make him spend more than he would have chosen on defense. On the evidence now available this did not happen. He did not leave defense policies or outlays to his underlings and took nothing for granted. Formally the Politburo decided how much would be spent on defense (Davies and Harrison 1997: 385–92) but Stalin’s word was decisive on all major issues. Stalin also took a much more active interest in the detail of foreign policy than previous historians would have thought (Davies, Khlevniuk, et al. 2003: 14). Well versed in divide-andrule, he expertly prevented military and industrial lobbyists from colluding (Harrison 2003). Possibly a military-industrial complex formed after his death (Bystrova 2000)—although even here traditional Western scholarship would often take a skeptical view (Holloway 1983: 159–60). In this chapter I will set out a view of the uses of Soviet military force in Stalin’s hands that is intellectually straightforward and also sits well with the facts that we know. It starts from simple premises: Stalin had gained power, and wanted to continue to hold it, but to do so he had to ward off multiple threats. These threats were both external and internal. Military force was one of the means at his disposal for holding power, but not the only one. The other instrument of his regime that will share center stage in this chapter is repression. Stalin’s use of repression is puzzling in its own way. The origins of the Great Terror of 1937/38, when 700,000 were executed for ‘‘counter-revolutionary crimes,’’ have been much debated. Did Stalin intend it, or was it an unintended consequence of the structure of his regime? Some historians always saw Stalin’s hand at work, but its sheer scale and often revolting cruelty made his motivations hard for them to penetrate (for example, Conquest 1971; Medvedev 1971). The terror appeared excessive, counterproductive, or just plain irrational. This led other scholars (for example, Getty 1985) to try to explain how the terror might have run out of Stalin’s control because of bureaucratic rivalries and social tensions. Now that the official documentation of the terror has been made available (Getty and Naumov 1999; Kozlov et al. 2004) the ‘‘unintentional’’ explanation seems less tenable. The terror did not run out of control. Stalin managed it personally; he switched repression on and off and fine-tuned it to a surprising degree. His motivation has also become clearer: he perceived the threat of a fifth column of the embittered and fainthearted who would turn against him in time of war, and he set out to destroy it beforehand (Khlevniuk 1995).∞ In short, the scale and timing of the terror were largely the result of the conscious decisions of a small group of people, and very largely of one man. But why Stalin perceived the problem as he did, and why he regarded the scale of the solution as appropriate and timed it as he did, remain unclear.

4

Mark Harrison

To summarize: we would like to understand why Stalin wanted a large army and defense industry. To do this we will set them in the context of his system of rule and his other means of ruling. In this chapter I will pursue the idea of military force and repression as parallel instruments in the joint equilibrium between Stalin and his potential opponents abroad and at home. The most efficient use of military force was to counter external threats. However, I will provide evidence that the balance of military power also had the capacity to influence the level of domestic discontent and opposition. Likewise, repression at home could influence the conduct of foreign adversaries as well as of internal opposition. In other words, Stalin’s foreign and domestic policies were intimately related. Stalin himself had no doubt of their connections, and particularly feared the consequences of internal opposition in the context of an external crisis, when foreign enemies might exploit collaborators to challenge his regime from within. Arguably, a coalition of internal and external enemies was his worst nightmare. His changing choices over repression and rearmament can be fully understood only in this context.

What Stalin Feared THE SOVIET UNION FACED REAL FOREIGN THREATS

Given the well-known history of the twentieth century it is not difficult to show that the Soviet Union faced real external threats. Russia’s separate peace with Germany in January 1918 prompted her former wartime allies to turn on the new Soviet government and make common cause with the antiBolshevik forces. Soviet Russia was invaded on all sides. Despite eventually prevailing over the White and Allied armies the Bolshevik regime remained understandably fearful of renewed intervention through the 1920s. Russia was now neighbored by countries such as Finland and Poland, lesser powers it is true, but all more or less hostile; the neighbors’ hostility was exacerbated by the memory of actions on the Soviet side such as the failed attempt to export revolution to Poland by force in 1920. Behind the neighbors, it was generally thought, stood the armies of much more powerful countries, including Britain, France, the United States, and Japan (see Chapter 2). Only Germany was a friend at this time, motivated chiefly by convenience. In the 1920s, as a result, Moscow was subject to periodic outbreaks of nerves that were sometimes exaggerated for dramatic effect. In the most famous case, the ‘‘war scare’’ of 1927, we do know that no one was actually planning to attack the Soviet Union. The great powers generally lacked either the popular will for war or the means to wage it. The United Kingdom cer-

The Dictator and Defense

5

tainly lacked both will and means; when the United Kingdom broke off diplomatic relations with the USSR in May of that year, the rupture was symbolic rather than truly threatening. In the Far East there was a local dispute over the Chinese Eastern Railroad, but the fact is that none of Russia’s neighbors had immediate plans for war and Moscow knew it (Simonov 1996; Samuelson 2000: 34–36). The true significance of the war scare was somewhat different and we will return to it. The long-term danger to the Soviet Union did not lie with the neighbors but with Germany and Japan. In September 1931 Japan invaded Manchuria; this was the first step along her path to World War II. Although strong Soviet defenses eventually deflected Japanese ambitions to easier pickings in the British and Dutch colonies in southeast Asia, there can be no doubt that Japan originally intended to make gains at the expense of Soviet territory. Within a year a ‘‘Russian National Union’’ had been formed in Manchuria with the aim of forming an independent state in Siberia allied to Japan (Davies 1996: 278). Japan continued to eye Soviet territory until the border war of 1939. As for Germany, on taking power in 1933 Hitler began secretly to prepare the military means to carve out a colonial living space for ethnic Germans in eastern Europe and European Russia. By 1935 the scale of his preparations could no longer be concealed. On March 31 of that year the Red Army commander Mikhail Tukhachevskii published his article in Pravda attacking ‘‘The War Plans of Contemporary Germany’’; on December 3, Stalin was warned that one of Hitler’s ministers had told a French banker that Germany intended to divide the Ukraine with Poland (Davies and Harrison 1997: 390). Hitler risked war with Britain and France over Austria in 1936 and Czechoslovakia in March and October 1938, went to war with them over Poland in September 1939, and finally took the opportunity to go to war with the USSR in June 1941. FOREIGN THREATS STIMULATED DOMESTIC OPPOSITION

The capacity of external threats to stimulate internal opposition is a pattern deeply rooted in Russian history. Russia’s military confrontation with Britain, France, and Turkey over the Crimea in 1854 brought simmering peasant discontent to the boil and eventually forced the abolition of serfdom in 1861. What Marxist could forget that the uprising of the Parisian communards in 1871 had been sparked by France’s defeat by Prussia the previous year? And what Russian could forget that it was Russia’s defeats at the hands of Japan and Germany that set the scene for the insurrections of December 1905 and February and October 1917? Oleg Khlevniuk (1995: 174) has noted: ‘‘The complex relationship between war and revolution, which had almost seen the tsarist regime toppled in 1905 and which finally brought its

6

Mark Harrison

demise in 1917, was a relationship of which Stalin was acutely aware. The lessons of history had to be learnt lest history repeat itself.’’ Soviet leaders read secret police reports of the volatile mood among the peasants and workers before and during the ‘‘war scare’’ of 1927, therefore, with trepidation. At the beginning of 1927 the president of the Soviet Republic Mikhail Kalinin told the Politburo: ‘‘I have talked with many peasants and can say straight out that in the event of a conflict with foreign states a significant stratum of peasants will not defend Soviet power with any enthusiasm, and this is also reported in the army.’’≤ From August 20 of that year an OGPU (security police) report summarizes workers’ reactions to the prospect of war (cited by Simonov 1996: 1358): ‘‘Kill all the communists and Komsomols [party youth members] who want a war.’’ ‘‘If there’s a war we’ll kill the administration first, then we’ll fight.’’ ‘‘If you give us war we’ll get weapons and make a second revolution.’’ If this was the mood at a time when any prospect of a real war lay far in the future, it was more serious when social tensions deepened further and the threat of war became still more actual. An OGPU summary dated January 19, 1932, from the time of deepening difficulties over grain sowings and procurements from the new collective farms, claimed the threat of war with Japan had ‘‘enlivened ‘kulak’ activities. In the Moscow region, for example, the kulaks were alleged to assert that ‘the kolkhozy are a second serfdom . . . , but we must put up with it for a time, soon Japan will attack Soviet power and we shall free ourselves’ ’’ (Davies and Wheatcroft 2003: 15–16). Similar summaries of the popular mood in 1936 reported such remarks as: ‘‘Soon there will be war and the Soviet regime will collapse’’; ‘‘Germany and Japan . . . will begin the war, and we will help them’’ (Fitzpatrick 1999: 205). The idea that internal and external threats may feed each other is not new. Comparing terror in the French and Russian revolutions, Arno Mayer has described the revolutionaries’ belief in the ‘‘interpenetration of internal and external counterrevolution.’’ In France as in Russia the terror was, in part, a response to military weakness and external threat. In 1793 Robespierre condemned ‘‘the moralists who sought to protect internal enemies ‘from the sword of national justice,’ insisting that by doing so they ‘blunted the bayonets of our soldiers’ who were risking their lives fighting the armies of foreign tyrants’’ (Mayer 2000: 205, 208). In turn, the rise of counter-revolutionary disaffection in the French provinces was often the consequence of the huge military levies ordered by the revolutionary government to fight off the foreign armies gathering to crush it. An understanding that domestic opposition may be highly responsive to foreign threats is evident in the earliest historical records. In the Peloponnesian

The Dictator and Defense

7

wars among the Greeks, for example, it was not uncommon for the armies or navies of one side to parade past the cities allied to the other in the hope that the demonstration would provoke a rebellion within the walls, causing the city to change sides. When the Athenians attacked Spartolus in Chalcidice in 429 b.c. they counted on assistance from the democratic faction within the city; on this occasion they were frustrated because the ruling oligarchs called up forces from an allied neighbor that defeated them in battle (Thucydides 2:79). But the tactic worked often enough that both Athenian expeditions to Sicily, the second of which ruined Athenian power, were raised not to occupy the island militarily but in the expectation that they would weaken the influence of Syracuse, a colony of Athens’ enemy Corinth, within the island’s other city states, and bring them over to Athens (Thuc. 3:86; 6:17). WHAT STALIN FEARED MOST: COLLUSION AMONG ENEMIES

Stalin demonized the former oppositionists partly by presenting them as in the pay of foreign powers.≥ Virtually all the victims of the major show trials held in Moscow between 1936 and 1938 were charged with acting in collaboration with or at the direction of foreign governments or intelligence services. The defendants implicated in the ‘‘Trotskyite-Zinovievite Terrorist Center’’ at the Kamenev-Zinov’ev trial (August 19–24, 1936) were alleged to have had contacts with the German Gestapo. The former oppositionists accused of involvement in the ‘‘Parallel Center’’ at the Piatakov trial (January 23–30, 1937) were accused of espionage and of trying to provoke a war in which Germany and Japan would defeat the Soviet Union. The Red Army commanders tried in June 1937 were charged specifically with treasonous dealing with Germany, as were many of their subordinates in subsequent processes. Finally, the defendants of the ‘‘Right-Trotskyist Center’’ at the Bukharin trial (March 2–13, 1938) were accused of spying for Britain since 1921 and planning to give away Soviet territory to the British Empire.∂ These were the public charges. Typically, the only evidence for them was the confessions of the accused, which had been beaten out of them. But Stalin did not instruct anyone to lie themselves or to extract a lie from others. Rather, he taught them how to construct evidence for the deeper ‘‘truth’’ that lay concealed beneath appearances, which he was the first to see clearly. Stalin’s belief in hidden enemies expressed the ‘‘dictator’s dilemma’’ described by Ronald Wintrobe (1998: 335–37): the more powerful the dictator, the less he can trust those around him when they claim to be loyal. The version of reality that he built on this belief would eventually appear in public court proceedings and in the columns of Pravda, but the first drafts of it are often found in the records of private conversations within the leadership (Fitzpatrick 1999: 21; Getty and

8

Mark Harrison

Naumov 1999: 15–24; Davies, Khlevniuk, et al. 2003: 12). In a private letter dated August 22, 1936, for example, Stalin explained to his Politburo colleagues how to respond to the foreign critics of the Kamenev-Zinov’ev trial, calling them ‘‘defenders of the gang of assassins and Gestapo agents’’ (Davies, Khlevniuk, et al. 2003: 335). Reacting on August 23 to testimony that Kamenev had sounded out the French ambassador concerning the French attitude to a future Soviet government of the opposition leaders, Stalin wrote to the Politburo: ‘‘I think Kamenev also sounded out the British, German, and American ambassadors. This means that Kamenev must have disclosed to those foreigners the plans for the conspiracy and assassinations of the [communist party] leaders. . . . This is an attempt by Kamenev and his friends to form an outright bloc with the bourgeois governments against the Soviet government’’ (Davies, Khlevniuk, et al. 2003: 338).∑ Perceptions like this were based on principles that Stalin followed consistently for many years. It is a hallmark of these principles that he adhered to them in private as well as in public. On June 4, 1932, for example, he wrote concerning rebellion in Mongolia to his deputy Lazar Kaganovich that the Mongolian government ‘‘should declare that the rebel chiefs are agents of the Chinese and especially the Japanese imperialists, who are seeking to strip Mongolia of its freedom and independence’’ (Davies, Khlevniuk, et al. 2003: 116, emphasis omitted). On July 2, he addressed the OGPU agents in Manchuria who, he believed, had overstepped the mark in organizing against the Japanese occupation, and ordered Kaganovich to ‘‘take Draconian measures against the criminals at the OGPU and the Intelligence Bureau (it is quite possible that these gentlemen are agents of our enemies in our midst)’’ (Davies, Khlevniuk, et al. 2003: 151). On August 7, 1932, Stalin mused on how to deal tactically with foreign specialists: ‘‘All foreign bourgeois specialists are or may be intelligence agents’’ (Davies, Khlevniuk, et al. 2003: 177; emphasis in original). In the midst of these matters Stalin was also dealing with the political crisis in the Ukraine resulting from the poor 1932 harvest and the excessive plan of procurements from it; on August 11 he warned Kaganovich, ‘‘Keep in mind, too, that the Ukrainian Communist Party (500,000 members, ha-ha) has quite a lot (yes, quite a lot!) of rotten elements, conscious and unconscious Petliura adherents [nationalists], and, finally, direct agents of [the Polish leader] Pilsudski’’ (Davies, Khlevniuk, et al. 2003: 180).∏ Stalin’s reference to ‘‘unconscious’’ nationalists expresses another dimension of his anxiety. He did not only fear hidden enemies, who had not revealed themselves yet; he also feared unconscious enemies—those who did not even know themselves yet, who believed themselves to be loyal. Sheila Fitzpatrick (1999: 194) conveys the consequences in the private thought of a diarist of the

The Dictator and Defense

9

time: ‘‘It was possible, evidently, to be a wrecker without meaning to be one or even knowing it. It was possible to wear a mask that deceived even oneself.’’ Stalin’s fears were heightened by the experience of the Spanish civil war. In October 1936 the Spanish Nationalist general Emilio Mola Vidal, asked which of his four army columns would take Madrid, declared that Madrid would be taken by the uprising of a ‘‘fifth column’’ of sympathizers within the city. Although this turned out to be more propaganda than fact, the idea of a fifth column hiding patiently at home until the opportunity presented itself to deliver the country to its external enemies exacerbated Stalin’s fears. It was his fear of a fifth column that sparked off the Great Terror of 1937 and 1938, when 1.6 million people from a working-age population of just under 100 million were arrested for counter-revolutionary violations; nearly half were executed (Kozlov et al. 2004, 1:609). Khlevniuk (1995: 167–68) concludes that the purges of those years aimed to achieve ‘‘the removal of all strata of the population, which in the opinion of the country’s leaders were hostile or potentially hostile’’ (my emphasis). Stalin’s anxiety pivoted on the expectation of war: the ‘‘potential’’ or ‘‘unconscious’’ enemies that he aimed to root out were those who appeared or believed themselves to be loyal in the normal, peaceful conditions of the time, whom he expected to betray their country at the first signs of foreign intervention. The framework for ‘‘mass operations’’ was set out in the NKVD operational order no. 00447, sent to the Politburo on July 30, 1937 (Getty and Naumov 1999: 473–80). Previous repressions had created wide swathes of the population that either remained under active suspicion or were seen as likely to have become embittered toward the regime. Although various small acts of reconciliation had been attempted, ‘‘the Stalinist leadership always considered terror as its main method of struggle with a potential ‘fifth column’ ’’ (Khlevniuk 1995: 169). Specifically designated for execution or imprisonment were those involved in the non-Bolshevik political parties, those who had fought for the losers in the civil war, the party oppositionists of left and right, those involved in religious activity, former kulaks, and those previously convicted of wrecking, spying, or other counter-revolutionary crimes; all qualified as actual or potential, conscious or unconscious enemies of the people. From the summer of 1937 the mass operations acquired a national tinge, with special actions directed against ethnic Germans (under an instruction of July 20, 1937), Poles (August 9), and Chinese (September 19); these were accompanied by orders to expel minorities from border regions—for example, Koreans from the Far East (August 21) (Khlevniuk 1995, 162–63). At the beginning of 1938 the transition to wholesale ‘‘national operations’’ began with a Politburo instruction of January 31 to the NKVD to complete, by

10

Mark Harrison

April 15, the destruction of the ‘‘counterrevolutionary nationalist contingent —Poles, Letts, Germans, Estonians, Finns, Greeks, Iranians, Harbintsy [former employees of the Chinese Eastern Railroad evacuated from Harbin, Manchuria], Chinese, and Romanians’’ and also Bulgarians and Macedonians (cited by Khlevniuk 1995: 164).π The fact is that Stalin was already personally secure in 1937 and his regime faced no immediate threat of any magnitude at home. This fact is sometimes used to criticize the idea that Stalin had a logically consistent rationale for the Great Terror. But the nonexistence of any serious domestic opposition at the time is beside the point. Stalin’s calculations were based primarily on his growing sense of the threat of a future war, and they were forward-looking; he was acting preemptively, not reacting. In this future war, Stalin feared, the elements in the domestic population that were presently no more than ‘‘potential’’ or ‘‘unconscious’’ enemies would reveal their true identities and rise up against him in concert with the foreign enemy. He had no reason to want to punish them, since they were doing nothing to merit punishment. He did not even want to warn or deter them; for deterrence, it would have been enough to select a few of them for heavily publicized show trials. He wanted simply to eliminate them beforehand, and this was best done without warning and in secret on a massive scale.∫ Authority for this interpretation is provided by Stalin’s prime minister Viacheslav Molotov. Looking back on these events after many years, Molotov observed that Stalin’s actions punished not just real enemies but also ‘‘many who vacillated, those who did not firmly follow the line and in whom there was no confidence that at a critical moment they would not desert and become, so to speak, part of the ‘fifth column.’ ’’ Molotov accepted this, but he did not regret it: ‘‘Stalin in my opinion, pursued an absolutely correct line: so what if one or two extra heads were chopped off, there would be no vacillation in the time of war and after the war’’ (cited by Khlevniuk 1995: 173). The ‘‘extra heads’’ that Stalin destroyed were innocent of any crime in the present. In Molotov’s eyes, Stalin had rightly decided to destroy them in advance—not to wait until they had had the opportunity to be guilty of something, when it might be too late.

Repression Is Flexible, Military Power Is Sticky To simplify, Stalin faced two threats with two instruments. The threats were his adversaries at home and abroad. His instruments were military force and repression. Military force was naturally adapted to meet the foreign threat, while repression could neutralize the threat at home.

The Dictator and Defense

11

Table 1.1. Soviet defense activity, 1926/27 and 1929 to 1940: Selected measures

∞Ω≤∏/≤π ∞Ω≤Ω ∞Ω≥≠ ∞Ω≥∞ ∞Ω≥≤ ∞Ω≥≥ ∞Ω≥∂ ∞Ω≥∑ ∞Ω≥∏ ∞Ω≥π ∞Ω≥∫ ∞Ω≥Ω ∞Ω∂≠

Regular personnel of the Red Army (thousands)

Munitions procured, unit index (% of ∞Ω≥π)

Defense outlays (% of imputed wage incomes)

∑∫∏ ... ... ∑∏≤ ∏≥∫ ∫∫∑ Ω∂≠ ∞,≠∏π ∞,≥≠≠ ∞,∂≥≥ ∞,∑∞≥ ... ∂,≤≠π

... ... ∞≤ ≤∂ ∑∞ π∏ ππ ∏∑ Ωπ ∞≠≠ ∞∑∏ ≤∂≠ ≤π≤

... ≤.∑ ≤.∑ ≥.≠ ∑.≥ ∑.∞ ∑.≠ ∑.π ∫.∏ ∫.π Ω.π ... ∞∫.∞

Source: Davies and Harrison (1997: 373, 375, and 395).

It is notable that Stalin did not try to use his armies to put down domestic enemies. OGPU decree no. 44/21 of February 2, 1930, for example, set out procedures for the ‘‘liquidation of the kulaks as a class’’ by arresting and imprisoning or deporting hundreds of thousands of people from the countryside. It warned the local secret police organizations ‘‘in no circumstances to involve units of the Red Army in the operation. To permit their employment only in extreme cases of the emergence of uprisings; by agreement with the local organizations, the Revolutionary Military Council, and the plenipotentiary OGPU representatives, where units of the OGPU troops are insufficient, covertly to organize military groups from reliable Red Army units that have been filtered by the special OGPU organs’’ (Kozlov et al. 2004: 96). The reason for this is clear: the Red Army’s rank and file consisted of conscripts representing all sections of society, not previously selected for loyalty, and could not be trusted if it came to a conflict with their own people. Military force and repression differed not only by the threats that they could be used to counter. They differed also in the speed at which they could be adapted to the threats as they changed. In brief, repression was relatively flexible and could be adjusted quickly. Military power, in contrast, was ‘‘sticky’’; it required considerable time to scale up and down. The main shifts in Soviet military spending and war preparations that we see

12

Mark Harrison

Figure 1.1. Soviet Defense Outlays, 1926/27–40: Selected Measures (percent of 1937) Key: A. The 1927 ‘‘war scare.’’ B. Japan invades Manchuria. C. Hitler takes power in Germany. D. Stalin recognizes Hitler as a threat. E. Japan invades China: World War II begins in the Far East. F. Germany invades Poland: World War II spreads to Europe. Source: Table 1.1; all series are normalized on 1937.

over the 1930s correspond in a rough way to the shifting external threats. To see this we must have measures, and quality effects make real military power and real outlays on it notoriously difficult to evaluate. Figure 1.1, based on Table 1.1, shows three alternative measures: the numbers of men and women in uniform, the volume of weapons purchased, and the burden of defense outlays relative to imputed wage incomes. The pattern they all show is one of rapid rearmament in the early 1930s followed by a hesitation, then acceleration in 1936 becoming more marked in 1938 and 1939. The rearmament of the early 1930s is often ascribed, reasonably, to fears over Japan, while the much greater rearmament at the end of the decade was clearly designed to counter the menace of the combined Axis powers (Davies 1996; Davies and Harrison 1997). In short, it appears that there was a general long-term intention to make the Soviet Union more secure externally. There were also fluctuations in the intensity with which the Stalinist leadership pursued this goal, and the fluctuations bear some relation to the intensity of external threats. The fluctuations were smoothed over time by long adjustment lags. These lags were forced by the time and effort required to build up military power. In the twentieth century it did not take much time to recruit soldiers, especially when the motherland was in danger. To train and equip them took longer,

The Dictator and Defense

13

Figure 1.2. Numbers Sentenced by the Security Agencies, 1921–53 Source: Table 1.2.

however, even when the equipment was ready. To build up the combat stocks of a mechanized army required specialized large-scale defense industries that took years to lay down. These war industries could not be maintained without comprehensive industrialization. Experience of the two world wars shows that industrial mobilization could be speeded up to a surprising degree when war broke out (Harrison 1998; Broadberry and Harrison 2005); in other words, the peacetime constraints on mobilization were not absolute and could be broken in time of national emergency. It follows that a dictator like Stalin, who did not respect social conventions even in peacetime, could probably have hurried things along a bit more during the 1930s. But the same experience also tells us that hurrying things along was ruinously expensive, so it is not really surprising if Stalin chose not to do more. Even as it was, a comparative view of European rearmament tells us that the Soviet military buildup shown in Figure 1.1 was more abrupt and determined than most. By contrast, Soviet experience shows that political repression could be turned on and off almost at the flick of a switch. Figure 1.2, based on Table 1.2, charts annual series for sentences for political offenses by tribunals, courts, the OGPU collegium, and NKVD special assemblies and troiki from the end of the civil war to Stalin’s death. From the point of view of establishing the cumulative total of victims of Stalinist repression, Michael Ellman (2002) has pointed out that these figures are incomplete. They interpret repression narrowly; they omit millions of victims imprisoned or forced into internal exile at various times for resisting specific policies such as the collectivization of agriculture, or violating

Table 1.2. Numbers sentenced by the security agencies, 1921–53

∞Ω≤∞ ∞Ω≤≤ ∞Ω≤≥ ∞Ω≤∂ ∞Ω≤∑ ∞Ω≤∏ ∞Ω≤π ∞Ω≤∫ ∞Ω≤Ω ∞Ω≥≠ ∞Ω≥∞ ∞Ω≥≤ ∞Ω≥≥ ∞Ω≥∂ ∞Ω≥∑ ∞Ω≥∏ ∞Ω≥π ∞Ω≥∫ ∞Ω≥Ω ∞Ω∂≠ ∞Ω∂∞ ∞Ω∂≤ ∞Ω∂≥ ∞Ω∂∂ ∞Ω∂∑ ∞Ω∂∏ ∞Ω∂π ∞Ω∂∫ ∞Ω∂Ω ∞Ω∑≠ ∞Ω∑∞ ∞Ω∑≤ ∞Ω∑≥ a a

Execution

Confinement in prisons and camps

Internal exile

Not specified

Total

Ω,π≠∞ ∞,Ω∏≤ ∂∞∂ ≤,∑∑≠ ≤,∂≥≥ ΩΩ≠ ≤,≥∏≥ ∫∏Ω ≤,∞≠Ω ≤≠,≤≠∞ ∞≠,∏∑∞ ≤,π≤∫ ≤,∞∑∂ ≤,≠∑∏ ∞,≤≤Ω ∞,∞∞∫ ≥∑≥,≠π∂ ≥≤∫,∏∞∫ ≤,∑∑≤ ∞,∏∂Ω ∫,≠≠∞ ≤≥,≤π∫ ≥,∑πΩ ≥,≠≤Ω ∂,≤∑≤ ≤,∫Ω∏ ∞,∞≠∑ ≠ ≠ ∂π∑ ∞,∏≠Ω ∞,∏∞≤ ∞Ω∫

≤∞,π≤∂ ≤,∏∑∏ ≤,≥≥∏ ∂,∞∑∞ ∏,∫∑∞ π,∑∂π ∞≤,≤∏π ∞∏,≤∞∞ ≤∑,∫∑≥ ∞∞∂,∂∂≥ ∞≠∑,∏∫≥ π≥,Ω∂∏ ∞≥∫,Ω≠≥ ∑Ω,∂∑∞ ∞∫∑,∫∂∏ ≤∞Ω,∂∞∫ ∂≤Ω,≥∞∞ ≤≠∑,∑≠Ω ∑∂,∏∏∏ ∏∑,π≤π ∏∑,≠≠≠ ∫∫,∫≠Ω ∏∫,∫∫π π≠,∏∞≠ ∞∞∏,∏∫∞ ∞∞π,Ω∫≥ π∏,∑∫∞ π≤,∑∑≤ ∏∂,∑≠∞ ∑∂,∂∏∏ ∂Ω,∞∂≤ ≤∑,∫≤∂ π,∫Ω∂

∞,∫∞π ∞∏∏ ≤,≠∂∂ ∑,π≤∂ ∏,≤π∂ ∫,∑π∞ ∞∞,≤≥∑ ∞∑,∏∂≠ ≤∂,∑∞π ∑∫,∫∞∏ ∏≥,≤∏Ω ≥∏,≠∞π ∑∂,≤∏≤ ∑,ΩΩ∂ ≥≥,∏≠∞ ≤≥,π∞Ω ∞,≥∏∏ ∞∏,∫∂≤ ≥,π∫≥ ≤,∞∂≤ ∞,≤≠≠ π,≠π≠ ∂,π∫π ∏∂Ω ∞,∏∂π ∞,∂Ω∫ ∏∏∏ ∂∞Ω ∞≠,≥∞∏ ∑,≤≤∑ ≥,∂≤∑ ππ≥ ≥∫

≤,∑∫π ∞,≤∞Ω ≠ ≠ ∂≥π ∏Ω∏ ∞π∞ ∞,≠≥π ≥,π∂∞ ∞∂,∏≠π ∞,≠Ω≥ ≤Ω,≤≤∫ ∂∂,≥∂∑ ∞∞,∂Ω∫ ∂∏,∂≠≠ ≥≠,∂∞∑ ∏,Ω∞∂ ≥,≤∫Ω ≤,∫∫∫ ≤,≤∫∫ ∞,≤∞≠ ∑,≤∂Ω ∞,∞∫∫ ∫≤∞ ∏∏∫ Ω∑π ∂∑∫ ≤Ω∫ ≥≠≠ ∂π∑ ∑ΩΩ ∑Ω∞ ≤π≥

≥∑,∫≤Ω ∏,≠≠≥ ∂,πΩ∂ ∞≤,∂≤∑ ∞∑,ΩΩ∑ ∞π,∫≠∂ ≤∏,≠≥∏ ≥≥,π∑π ∑∏,≤≤≠ ≤≠∫,≠∏Ω ∞∫≠,∏Ω∏ ∞∂∞,Ω∞Ω ≤≥Ω,∏∏∂ π∫,ΩΩΩ ≤∏π,≠π∏ ≤π∂,∏π≠ πΩ≠,∏∏∑ ∑∑∂,≤∑∫ ∏≥,∫∫Ω π∞,∫≠∏ π∑,∂∞∞ ∞≤∂,∂≠∏ π∫,∂∂∞ π∑,∞≠Ω ∞≤≥,≤∂∫ ∞≤≥,≤Ω∂ π∫,∫∞≠ π≥,≤∏Ω π∑,∞≤∑ ∏≠,∏∂∞ ∑∂,ππ∑ ≤∫,∫≠≠ ∫,∂≠≥

First six months.

Source: Popov (1992: 22); Kozlov et al. (2004, 1:608–9).

The Dictator and Defense

15

specific laws on, for example, labor discipline or the protection of state property. But a narrow focus is what we need here. The outstanding feature of the series is the sharp peak of repression in 1937 and 1938 that came about through the confluence of two processes. One was a public purge of Stalinist officialdom that carried away tens of thousands of victims between early 1936 and the end of 1938. The other was the mass operations secretly initiated at the end of July 1937, the victims of which are measured in hundreds of thousands. On a monthly or weekly breakdown, the series would probably show still greater volatility. Through the mass operations Stalin was able to ratchet up the rate of executions by several orders of magnitude in a few days; his brief instruction was sufficient to put an end to the mass operations in mid-November 1938. It is clear that repression was not perfectly flexible. The documents themselves show that preparation was required to both launch and terminate the mass operations. It took a few weeks to develop the rationale for the mass operations, gather the necessary information from the localities to fix their scope and scale, and disseminate the procedures for implementing them through NKVD order no. 00447. To scale them down took somewhat longer, ten months from the first signals in mid-January 1938 to their conclusion in midNovember. One factor in this delay was the need to conclude operations already in progress, but another was pressure from the NKVD in the localities to extend their scope (Khlevniuk 1995: 159–65). Eventually, however, the mass operations were abruptly closed down (Getty and Naumov 1999: 531–37). If repression was not perfectly flexible in the short term, longer-term rigidities also cannot be ruled out. The many smaller fluctuations in the level of repression before 1937 can be interpreted in two ways. One explanation is that they show the scope for flexibility and fine-tuning. Another is that there was high-level resistance. The resistance hypothesis implies that over many years Stalin developed a long-term plan to execute hundreds of thousands of people, but he was unable to implement it before 1937 because he first had to overcome resistance among his Politburo colleagues and at lower levels in the party. If this were right, it would follow that repression was not such a flexible instrument. Taken at face value, the resistance hypothesis is no longer particularly credible. Getty and Naumov (1999: 576–83) rule it out on two grounds: the archives of the years before 1937 have not produced any evidence that Stalin had a long-term plan of repression; on a surprising number of occasions other leaders in the Politburo and the regions were not an obstacle and were more eager for executions than Stalin himself. They conclude that Stalin did not have a long-term design, was often uncertain about how to optimize repression in the short run, and managed it from day to day in the light of results.

16

Mark Harrison

On a narrow interpretation the resistance hypothesis may now seem unconvincing; could it still have force in a broader sense? In a much earlier year such as 1932, Stalin was already a real dictator; for example, he had enough power that he could make decisions that precipitated a regional famine and killed millions of people unintentionally (Davies and Wheatcroft 2003) and survive without serious challenge. But he might not have gathered enough power yet to choose to kill hundreds of thousands of people deliberately, including some leaders who were not already marginalized or suspect but quite close to him. But my argument does not require that Stalin had unfettered arbitrary power. It requires only that he had the power to take the decisions he wanted when his perception of the present and future threats demanded it. In effect, I will argue that in 1932 the context did not make the execution of hundreds of thousands his best choice. If there would have been resistance to a Great Terror in 1932, it was because the context did not demand it of Stalin, as much as because Stalin did not demand it. To conclude, I do not maintain that repression was literally frictionless; however, the significant lags in scaling repression up and down were measured in days and weeks. These are momentary in comparison with the years required to build up spending on military equipment and defense supplies. Thus military spending was sticky; in contrast, Stalin adjusted repression almost at will.

Military Force and Repression in Theory A ruler must retain power in the face of real internal and external threats. Stalin’s problem began from the fact that the threat from each depends on the threat from the other. Intuitively, domestic opposition rises with the external threat because the latter raises the return to the dictator’s potential opponents from cutting a deal with the foreign enemy; by the same token, the danger of an external attack rises with discontent at home because the foreign enemy can expect to find more willing collaborators. Stalin’s own behavior and private attitudes show that he feared this synergy of domestic and foreign threats greatly and did everything he could to deter his own subjects from contemplating collusion with an external enemy. It follows from this that any instrument that opposes the foreign threat will also mitigate the threat from domestic agents. For example, discontented Soviet citizens sometimes expressed the private hope that an attack by foreign enemies would bring down the Soviet regime, and claimed that they would be ready to help. By rearming against Japan and Germany, Stalin lowered the expectations of his internal enemies that help might come from outside. Similarly, an instrument that represses domestic discontent will also lessen the

The Dictator and Defense

17

threat from abroad. Stalin may have thought that if he could visibly root out potential opposition on the home front, his enemies abroad would no longer expect Soviet traitors to help them in the event of war, and would have second thoughts about attacking Stalin in the first place. If rearmament and repression could both be used for the same objectives, how should a ruler combine them? This class of problem is well known in the economics of public policy; it was described and solved long ago by Jan Tinbergen (1952) and Robert A. Mundell (1962). It is normal for governments to have multiple objectives and multiple instruments of policy. It is also normal for each instrument to affect several objectives at the same time, making some easier and some harder to achieve. When this is the case, governments should match instruments against objectives in conformity with two rules. Tinbergen’s Rule requires that the government should have as many independent instruments as it has objectives, and in this case Stalin satisfied the rule by having two instruments, rearmament and repression, to neutralize two enemies, abroad and at home. Mundell’s Rule requires the government to assign each instrument to the objective for which it has a comparative advantage. In Stalin’s case, if a given increase in military force could diminish the external threat by more than it affected domestic opposition, and conversely for repression, then the ruler should mete out repression to his enemies at home and build military power against his adversaries outside the country. In short, suppose that the dictator wants to keep both domestic and foreign threats below some threshold level that will secure him from insurrection or military attack. Internally, he faces a ‘‘revolution constraint’’; looking abroad, he faces an ‘‘invasion constraint.’’ To stay in power, he must set the levels of military power and repression in such a way as exactly to neutralize his external and internal enemies at the same time and satisfy both constraints at once. Figure 1.3 shows some implications. The level of repression is measured along the horizontal axis, and military power is measured along the vertical axis. The steeper curve is the Revolution Constraint (RC): at all points along it, the domestic threat is neutralized. It is steep because a small reduction in repression must be compensated by a large increase in military power to neutralize a given domestic threat. The shallower curve is the Invasion Constraint (IC); at all points along it, the foreign enemy is held off. It is flatter because a small reduction in military power must be offset by a large increase in repression to keep the dictator safe from attack. Thus, repression and military power each contribute to both internal and external security; however, repression produces internal security directly and external security only indirectly, while the converse is true of military power; as a result, repression is comparatively

18

Mark Harrison

Figure 1.3. The Revolution and Invasion Constraints

more efficient in the production of internal security, while military power has a comparative advantage in producing external security. In this model repression and rearmament are inputs into the dictator’s power. I assume that the dictator has some other uses of power that he values and would like to maximize, and so it is in his private interest to hold power efficiently.Ω The intersection of the two constraints shows the efficient combination where the dictator sets military power equal to M≠ and repression equal to R≠. To the southwest of this point, the ruler is insecure both internally and externally. To the northwest he is safe from foreign enemies but his domestic enemies will overthrow him; to the southeast his domestic enemies are repressed but the foreign enemy will attack. In the shaded area to the northeast he has a safety margin on both fronts, but he must sacrifice other objectives unnecessarily to achieve this, unless he values safety for its own sake. This framework yields some simple comparative statics. For example, in the absence of any frictions, an increase in either threat should be matched by a proportional increase in the level of the instrument that is matched to it. Figure 1.4 shows the consequences of an increase in the external threat. The increase in the threat is shown by the upward shift of the invasion constraint to IC%: at any given level of repression, more military force is needed now to keep the dictator in power. However, the increase in the external threat will also raise the hopes of internal opposition and provoke more domestic discontent. RC, the revolution constraint, will move too, but by less. Taking both together, the dictator’s efficient choice is to maintain repression at R≠ and increase military power from M≠ to M∞.

The Dictator and Defense

19

Figure 1.4. An Increase in External Danger

In this case there is a vital complication: the time required for adjustment. We have seen evidence that military power was less speedily adaptable than repression. This means that the dictator’s choices over military power must be forward-looking. Because he cannot respond instantly, he must forecast foreign threats. What happened when Stalin made a mistake? On several occasions foreign enemies took Stalin by surprise. When faced with a sudden increase in external dangers Stalin would have liked to shift the country smoothly from M≠ to M∞, but rearmament would take time. In the meantime, his regime would be in peril because, if the external threat were not immediately countered, his internal enemies might be encouraged to rise up against him. What he should do in the long term is clear from Figure 1.4: he should rearm. But what should he do now? The solution is shown in Figure 1.5. In Figure 1.5 the dictator finds that the foreign danger level has risen, pushing the invasion constraint upward to IC% and the revolution constraint outward to RC%. He would like to rearm, but he cannot do this right away; he must buy time. In the figure, he buys time to rearm by raising repression to R&; this is the only option that will keep him inside his safety zone. In the short term, in other words, the dictator must rely on repression alone to counter both internal and external threats. The problem is that repression is the less efficient instrument for countering an external threat; the dictator must increase it, therefore, by a relatively large amount. Once the dictator has rearmed, he can scale repression back down so as to approach the efficient combination from the southeast, trading back along the invasion constraint IC% to M∞ and R≠ on the line shown by the

20

Mark Harrison

Figure 1.5. When Rearmament Takes Time

arrows. During the rearmament period, however, there is a temporary burden of excess repression, shown in the figure by the horizontal gap between R& and R%. The excess is required because the dictator cannot adjust the efficient instrument, military power, quickly enough, so in the interim he must use the inefficient instrument instead. R& is necessary to prevent invasion; it is not strictly necessary to prevent insurrection. R% would be enough on its own to prevent his domestic opponents from collaborating with each other to overthrow him. But in the circumstances repression does not just have to prevent insurrection; it also has to prevent internal enemies from colluding with likely external aggressors and ensure that the latter will get no help from any potential fifth column inside the gates. To summarize, a dictator who makes mistakes in forecasting external threats will be forced away from the balance that is efficient in the long term. The consequences of mistakes are not symmetrical, however. If he underestimates the required level of military power he must compensate with disproportionate repression. If he overestimates, however, society does not benefit by a disproportionate liberalization while he scales down military power; this is because although he can go outside the revolution constraint, he must not go inside it. This model yields two predictions. First, a ruler who correctly anticipates an increase in international tension should match it with a proportionate increase in military power. Second, however, faced with a rise in international tension

The Dictator and Defense

21

that is sudden and unanticipated, the same ruler should respond with an increase in political repression. The repression will be disproportionate to any internal threat, but transitory; it should be followed by a military buildup and a concurrent relaxation of the internal pressure on society.

Evidence on Military Power Versus Repression In the early 1930s Stalin’s dictatorship impoverished and embittered millions of his subjects. Society and politics were gagged and straitjacketed. Markets were suppressed and economic choices were greatly restricted. Heavy taxation enabled a huge diversion of resources into public projects. Peasant farmers were expropriated and suffered years of intense deprivation. Beyond the country’s borders, enemies gathered. The external dangers grew most markedly, early and late in the decade. Stalin responded with both repression and rearmament. Figure 1.6 charts the Soviet tradeoff between military power and repression during the 1930s. Repression is measured by annual sentences by the security organs in proportion to the population of working age; this indicator, while crude, captures well the large swings in Stalin’s propensity to victimize his less reliable subjects. Military power is indicated by the burden of defense outlays on imputed worker incomes; this measure is also imperfect, but has the merit of capturing the changing degree of consumer sacrifice that Stalin was willing to impose to realize his military policies. The vertical movement in the figure shows the progress of Soviet rearmament. Comparing the finishing point with the starting point, Soviet military power had multiplied several times while the level of political repression was roughly unchanged. This shows Stalin’s long-term adjustment to the growing external danger. While Soviet military power increased in the long term, the figure also shows several sharp increases in repression; these were often transitory, and were followed by substitution back toward rearmament. The general movement, and especially the repression ‘‘spike’’ of 1937, are readily interpreted in terms of Figure 1.5. In the first years of Stalin’s dictatorship his policies of forced industrialization, grain procurement, and collectivization evoked discontent and resistance; he easily increased repression in response. At this time the world was becoming noticeably more dangerous, particularly in the Far East, and at first this too demanded more repression; when he could, however, Stalin traded back to greater military power combined with less repression. But he could not do so at will, since it took time to increase the Soviet Union’s military capabilities. Given the inefficiency with which repression could address an unanticipated foreign threat, one might also expect heightened attention at such times to

22

Mark Harrison

Figure 1.6. Military Power Versus Repression, 1929–40 Sources: Military outlays in each year, expressed in percent of imputed wage incomes of the working population, are from Table 1.1; missing 1939 is imputed from data in the source. Numbers sentenced each year by the OGPU and NKVD special assemblies and troiki (as Figure 1.2), are calculated in percent of the population aged 15 to 64 within contemporary Soviet frontiers on 1 January each year from Andreev, Darskii, and Khar’kova (1990: 44–46).

nonmilitary instruments for influencing an adversary—for example, diplomacy. It is notable that Stalin gave special attention to diplomatic instruments at times of particularly rapid rearmament. For example, while the decisions were being taken that led to the first rearmament surge of 1931 to 1934, the Soviet government was taking an active part in the World Disarmament Conference at Geneva; to support these efforts the Politburo ordered the true increase in Soviet military outlays to be hidden for several years (Davies 1993). Later in the decade, when the military threat from the Axis became imminent, Stalin could not increase real defense outlays fast enough to protect his regime and restore stability by using the military instrument on its own. To cover himself during the transition, he resorted to a temporary repression that was hugely in excess of that required just to pacify domestic discontent. Then, having readjusted Soviet military power, as he thought, to the new threat, he could wind repression back down to its equilibrium level. Further illustration comes from the opening period of World War II. The

The Dictator and Defense

23

German invasion of June 1941 marked the collapse of Stalin’s hopes to deter Hitler, and the external threat to his regime became unexpectedly acute. The first six months of the war saw extraordinary efforts to recruit to the Red Army and reequip it, but these efforts were largely offset by early losses and the disruption of the economy. It was not until the second half of 1942 that the Red Army and war production were successfully stabilized at a much higher level (Harrison 1996). In the interim, the Soviet political society and its army were swept by ‘‘doubt and hesitation, dismay and fear’’ (Tochenov 2004: 42). In the countryside defeatism stimulated speculative talk about sharing out state grain stocks and collective livestock; in 1941 and 1942 there were widespread reports of collective farmers secretly agreeing to the redivision of the kolkhoz fields into private property in anticipation of the arrival of German troops (Barber and Harrison 1991: 104). Tochenov (2004) has vividly described unrest among the textile workers of Ivanovo: with production and pay collapsing and evacuation under way some debated, ‘‘Where better to live—under Hitler or under Soviet power? Who will win the war?’’ Others concluded, ‘‘It’s all the same to us to work for Hitler or Stalin.’’ One exclaimed to a party member: ‘‘God save us from the victory of Soviet power, they’ll hang all you communists.’’ A crowd shouted: ‘‘Down with Soviet power, long live little father Hitler.’’ An NKVD report blamed the unrest, characteristically, on ‘‘previously disguised spies and provocateurs—straightforward agents of German fascism and hostile people who have infiltrated the enterprises.’’ But the regional party committee noted that many party members had joined them. During this period Soviet society was held together by a wave of repressive measures (described by Barber and Harrison 1991: 63–67). Exceptional military mobilization was combined with a new wave of arrests for counterrevolutionary crimes; as Table 1.2 shows, 1942 also saw the highest number of political executions in any year outside the Great Terror. If it is right, this view of Stalin’s behavior can be tested. It is based on reasoning that should be valid for any ruler that has freedom of action under very general assumptions. According to this reasoning, when faced with an increase in some external threat, the ruler’s best response is to rearm—which is not surprising. The twist in the story comes when the change in the external threat is sudden and unanticipated, and for some reason military power cannot be adjusted quickly. In that case the ruler still rearms but, while he is doing so, he first kills those he fears, even though they have done nothing to deserve it. If this reasoning is right, then there should be evidence for it in other historical contexts. Although not plentiful, this evidence does exist. German historians (Gerlach 1998; Jersak 2003) have recently turned atten-

24

Mark Harrison

tion to the foreign-policy background of the Holocaust. They show another dictator responding to the sudden mounting of external threats, not by unifying the nation but by seeking to eliminate a potential ‘‘enemy within.’’ Hitler’s worst fear was of an international Jewish conspiracy. What did this mean to him, specifically? Christian Gerlach (1998: 786) suggests that Hitler viewed the Jews as coordinating a coalition of foreign enemies with their own potential to act on the territory of the Third Reich as ‘‘opponents, revolutionaries, saboteurs, spies, ‘partisans’ in his own backyard.’’ In this light, the steps that led to the Holocaust can be seen to follow a consistent pattern. In August 1941, although not yet at war, the United States administration made its first public commitment to the ‘‘final destruction of the Nazi tyranny,’’ when secretary of state Cordell Hull joined Winston Churchill in signing the Atlantic Charter. Hitler responded by secretly ordering the first deportations of Jews from Berlin (Jersak 2003). Japan’s surprise attack on Pearl Harbor at the beginning of December brought the United States into the war in Europe; at almost the same time, an unexpected Soviet counteroffensive began to drive German forces back from the gates of Moscow. Within days, Gerlach argues, Hitler had issued the secret orders that were intended to bring about the extermination of all the Jews in Europe. Another parallel is found in communist China. In 1962 the Soviet and Chinese communist parties came into open conflict over their interpretations of the Cuban missile crisis and relations with the United States. The former allies then took opposite sides when India and China clashed over a frontier dispute. In November 1964 Moscow rejected China’s terms for reconciliation. Meanwhile, the war in Vietnam was building up; in the spring of 1965 the United States launched an intense air campaign against one of China’s few remaining allies. In the Cultural Revolution, which Mao Zedong launched in 1966, we see again how a dictator responded to international isolation and weakness not by rallying the nation but by polarizing it in a campaign of far-reaching repression against potential—not actual—sources of domestic opposition. The short fifth century of civil war among the Greeks provides other parallels. Thucydides (4:80) describes how, in 424 b.c., Sparta was threatened unexpectedly by Athenian attacks on its territory. Its leaders, fearing that the slave population of helots would be inspired to revolt, selected two thousand of those who showed ‘‘most spirit’’ and slaughtered them in secret. Later in the war, in 405 b.c., a Spartan fleet appeared unexpectedly off the island of Samos, which was ruled by a democratic faction loyal to Athens. Faced with immediate blockade the Samian democrats responded by executing their aristocratic opponents, who would have favored installing an oligarchy and going over to Sparta (Kagan 2004: 478).

The Dictator and Defense

25

It must be said that clear parallels to the Great Terror are historically rare. Those that can be found tend to have arisen against a background of civil conflict. Perhaps there were unseen moral barriers that normally prevented a ruler from exploiting power so brutally; only the bitter experience of civil war could have broken moral inhibitions down so as to give Stalin the license to behave in the way that he did. We have asked why Stalin wanted a large army and defense industry. The short answer is that he had many enemies. He used Soviet military power to counter foreign threats. However, he also had enemies at home. Stalin’s decisions on military power and political repression were not independent. The interaction between foreign and domestic threats ensured that Stalin could not fix his response to one without considering its implications for the other. It follows that rearmament and repression are best seen as two elements of a package. Given time, military power could be efficiently adjusted to counter a foreign threat, while repression could efficiently neutralize domestic enemies. The two instruments differed, however, in that repression could be adjusted flexibly whereas rearmament plans took a considerable time to mature. As a result, it was difficult to respond quickly enough to foreign dangers that materialized unexpectedly. During the 1930s domestic tensions and foreign threats to the Soviet Union both accumulated. Successive actions by Japan and Germany ratcheted upward the fear of war. Stalin’s responses closely match those predicted by reasoning from first principles. There was a sequence of abrupt increases in repression followed by adjustments that raised military spending in a more sustained way while reducing repression. The underlying problem was that when foreign threats increased suddenly, the more efficient response was not immediately available. In the interim Stalin used repression to protect himself, although it was the less efficient response. The result was that he imposed a very large excess burden of repression on the Soviet society in some years. Notes I thank Stephen Broadberry, Keith Cowling, Simon Ertz, Arch Getty, Bishnupriya Gupta, James Harris, Oleg Khlevniuk, Andrei Markevich, Karl Maurer, Evan Mawdsley, William Mulligan, Eugenio Proto, and seminar participants at the London School of Economics, Royal Holloway University of London, and the Universities of Glasgow, Oxford, and Warwick for comments and advice. 1. This conclusion contradicts Roberta Manning’s (1993) suggestion that the Great Terror was a response to economic difficulties rather than political threats. Davies (2006) has shown that the timing is all wrong: Stalin launched the purge of the economic appara-

26

Mark Harrison

tus in early 1936 at a time when he had little cause to feel anything but satisfaction with the state of the economy, and those charged with economic sabotage in the show trials of 1936 and 1937 were not accused of causing the difficulties that began to materialize later in 1936. In short, the terror took place at a time of economic difficulties but the terror contributed to the difficulties, not the other way around. 2. RGASPI, 17/163/103 (January 3, 1927). 3. Stalin tended to demonize everyone. Davies (1989: 114) has noted that the public discourse of Stalinism relied heavily on the teaching ‘‘He that is not with me is against me’’ (Luke 11:23). For present purposes, what is of more importance is that this was not just a rhetorical pose for public consumption but accurately reflected Stalin’s worldview expressed, for example, in his private correspondence. Once he had overcome those openly opposed to him and had won unchallenged personal authority over the Soviet state and Bolshevik party, he continued to see the hand of enemies at work in all things. Commenting on mistakes made by the architects of the White Sea Canal on August 27, 1932, he called them ‘‘bunglers (or covert enemies)’’ (Davies, Khlevniuk, et al. 2003: 198); on September 24, 1933, he called those disputing a plan for tractor repairs ‘‘bunglers or outright enemies’’ (Davies, Khlevniuk, et al. 2003: 218). In an earlier incident Soviet trade officials in France and Finland were found to be corrupt; in private correspondence dated September 16, 1931, Stalin described this not as venality but as ‘‘betrayal’’ (Davies, Khlevniuk, et al. 2003: 91). Commenting on the circumstances of the famine of 1932/33 he wrote to the novelist Mikhail Sholokhov on May 6, 1933, that the peasants ‘‘were carrying out a ‘silent’ war against Soviet power. War by starvation’’ (Stalin to Sholokhov, cited by Davies 1996: 243; in fact it was predominantly the peasants who were starving). The previous September at a Central Committee plenum, he heard the speaker declare: ‘‘People who can look on cold-bloodedly while workers or their families don’t get bread for two or three days are degenerates.’’ Stalin interrupted: ‘‘They are enemies’’ (cited by Davies 1996: 257). 4. Conquest (1971: 148–76, 230–58, 497–573); Medvedev (1971: 152–239). Alleged links with foreign influences in these years were not confined to political troublemakers. Foreign connections were strongly featured in the Shakhty affair of 1928 (Carr and Davies 1969: 621–27), which inaugurated a new relationship between the Bolsheviks and professional workers. In a trial that lasted from May 19 to July 5, 1928, fifty-five mining engineers from the Shakhty district of the Donets coalfield, including three Germans, were accused of economic sabotage in the interests of former shareholders now in France; links were also alleged with German industrial interests and Polish intelligence. Four were acquitted, including two of the Germans, but the other German was imprisoned along with most of the rest; five were executed. In a speech of May 16, 1928, i.e., on the eve of the trial, Stalin (1954, 11:74) described the Shakhty affair as ‘‘the expression of a joint attack on the Soviet regime launched by international capital and the bourgeoisie in our country.’’ Similarly the leaders of the ‘‘Industrial Party,’’ on trial from November 25 to December 6, 1930, were accused of wrecking of the economy so as to create a context for French military intervention to establish a counter-revolutionary government (Davies 1989: 409 and 410n). An undercover reporter for an emigrant journal reported opinion in the Moscow streets as holding that the accused ‘‘sold themselves to the capitalists’’ and deserved execution. Trotskii, in exile abroad, at first believed that the accused were

The Dictator and Defense

27

indeed ‘‘a gang of agents of international imperialism.’’ At the Metro-Vickers trial of April 12–19, 1933, six British and twelve Soviet engineers were convicted of wrecking activities at power stations and spying for British intelligence (Davies 1996, 337). 5. In September 1935, during the preparation for the trial, Nikolai Ezhov told a conference of regional party secretaries: ‘‘Foreign intelligence officers, saboteurs, know that there is no better cover for their espionage and subversive operations than a party card. . . . We can assert firmly that Poles, Finns, Czechs, and Germans have been openly gambling on this. . . . They send people here and tell them: ‘Go and get yourself a party card’ ’’ (cited by Getty and Naumov 1999: 201). 6. An interesting case, secret until recently, followed in the summer of 1934, when a depressed gunnery officer was arrested for bringing a detachment of unarmed recruits to Moscow and exhorting them to overthrow the government. On August 8, 1934, Stalin wrote that the culprit, Nakhaev, ‘‘is, of course (of course!), not alone. He must be put up against the wall and forced to talk—to tell the whole truth and then severely punished. He must be a Polish-German (or Japanese) agent. The Chekists become laughable when they discuss his ‘political’ views with him (this is called an interrogation!). A venal mutt doesn’t have political views—otherwise he would not be an agent for an outside force’’ (Davies, Khlevniuk, et al. 2003: 248). At first Nakhaev did not cooperate, so Kaganovich reported to Stalin on August 12: ‘‘He is not showing his real roots yet. All of his behavior is confirmation that he is a foreign agent’’ (Davies, Khlevniuk, et al. 2003: 249). Some days later Nakhaev fell in with his allotted role, and on August 28 Kaganovich was able to tell Stalin: ‘‘As was to be expected Nakhaev confessed his connections with General Bykov, who . . . is an intelligence agent, according to what has been determined so far, for Estonia. We have to assume of course that it is not just Estonia’’ (Davies, Khlevniuk, et al. 2003: 264). From Stalin’s example, those around him learned to see the hand of foreign powers in quite minor matters. Thus on August 22, 1935, Kaganovich informed Stalin of the arrest of the Russian editor of a French-language newspaper published in Moscow: ‘‘The newspaper was conducted incorrectly and evidently Change of Landmarks [an emigrant group advocating reconciliation between the intelligentsia and the regime] elements, and perhaps spy elements as well, clustered around it’’ (Davies, Khlevniuk, et al. 2003: 298). Writing to Stalin about disruption on the railroads in Krasnoiarsk on January 26, 1936, Kaganovich noted: ‘‘The [repair] depot and the shop are contaminated with wrecker elements who are connected with the Japanese and Poles’’ (Davies, Khlevniuk, et al. 2003: 322). Reporting to a Central Committee plenum on December 4, 1936, concerning the newly uncovered ‘‘backup center’’ of the defendants in the Kamenev-Zinov’ev trial of the previous summer, Ezhov mentioned that it had improved its links with the country’s border regions. Lavrentii Beriia interjected: ‘‘And also in terms of its connections abroad’’ (Getty and Naumov 1999: 304). 7. Russians with contacts abroad were generally a target in the terror (Conquest 1971: 385; Solzhenitsyn 1974: 84), as was the contingent of foreign revolutionaries who had taken refuge from fascism in Moscow (Conquest 1971: 574–88; Medvedev 1971: 218–23). 8. Paul R. Gregory (2008) reaches the same conclusion, that the terror followed a strategy of elimination, from a study of the Soviet security police archives. I thank him for allowing me to read and cite this work while it was in press.

28

Mark Harrison

9. It is not necessary to the model to know what the value of power is to the ruler, but there is a wide literature that deals with such questions—for example, Olson (1993), Wintrobe (1998), Gregory (2004), and Açemoglu and Robinson (2006).

Published References Açemoglu, Daron, and James A. Robinson. 2006. Economic Origins of Dictatorship and Democracy. Cambridge: Cambridge University Press. Agursky, Mikhail, and Hannes Adomeit. 1978. ‘‘The Soviet Military-Industrial Complex and Its Internal Mechanism.’’ National Security Series no. 1/78. Queen’s University, Center for International Relations, Kingston, Ontario. Andreev, E., L. Darskii, and T. Khar’kova. 1990. ‘‘Opyt otsenki chislennosti naseleniia SSSR 1926–1941 gg. (kratkii rezul’taty issledovanii).’’ Vestnik statistiki 1990(7): 34–46. Aspaturian, Vernon V. 1973. ‘‘The Soviet Military-Industrial Complex: Does It Exist?’’ In Testing the Theory of the Military-Industrial Complex, ed. Steven Rosen, 103–33. Lexington, MA: Lexington Books. Barber, John, and Mark Harrison. 1991. The Soviet Home Front, 1941–1945: A Social and Economic History of the USSR in World War II. London: Longman. Broadberry, Stephen, and Mark Harrison, eds. 2005. The Economics of World War I. Cambridge: Cambridge University Press. Bystrova, Irina. 2000. Voenno-promyshlennyi kompleks SSSR v gody kholodnoi voiny. (Vtoraia polovina 40-kh-nachalo 60-kh godov). Moscow: Institut Rossiiskoi istorii Rossiiskoi Akademii nauk. Carr, E. H., and R. W. Davies. 1969. A History of Soviet Russia, vol. 3: Foundations of a Planned Economy, 1926–1929, Part 1. London: Macmillan. Conquest, Robert. 1971. The Great Terror. Revised edition. Harmondsworth: Pelican. Davies, R. W. 1989. The Industrialisation of Soviet Russia, vol. 3: The Soviet Economy in Turmoil, 1929–1930. Basingstoke (England): Macmillan. Davies, R. W. 1993. ‘‘Soviet Military Expenditure and the Armaments Industry, 1929– 33: A Reconsideration.’’ Europe-Asia Studies 45(4): 577–608. Davies, R. W. 1996. The Industrialisation of Soviet Russia, vol. 4: Crisis and Progress in the Soviet Economy, 1931–1933. Basingstoke (England): Macmillan. Davies, R. W. 2006. ‘‘The Soviet Economy and the Launching of the Great Terror.’’ In Stalin’s Terror Revisited, ed. Melanie Iliˇc, 11–37. London and New York: Palgrave. Davies, R. W., and Mark Harrison. 1997. ‘‘The Soviet Military-Economic Effort Under the Second Five-Year Plan (1933–1937).’’ Europe-Asia Studies 49(3): 369–406. Davies, R. W., Oleg Khlevniuk, E. A. Rees, Liudmila P. Kosheleva, and Larisa A. Rogovaia, eds. 2003. The Stalin-Kaganovich Correspondence, 1931–36. New Haven, CT: Yale University Press. Davies, R. W., and S. G. Wheatcroft. 2003. The Industrialisation of Soviet Russia, vol. 5: The Years of Hunger: Soviet Agriculture, 1931–1933. Basingstoke (England): Palgrave. Ellman, Michael. 2002. ‘‘Soviet Repression Statistics: Some Comments.’’ Europe-Asia Studies 54(7): 1151–72. Fitzpatrick, Sheila. 1999. Everyday Stalinism. Ordinary Life in Extraordinary Times: Soviet Russia in the 1930s. Oxford: Oxford University Press.

The Dictator and Defense

29

Gerlach, Christian. 1998. ‘‘The Wannsee Conference, the Fate of German Jews, and Hitler’s Decision in Principle to Exterminate All European Jews.’’ Journal of Modern History 70(4): 759–812. Getty, J. Arch. 1985. Origins of the Great Purges: The Soviet Communist Party Reconsidered, 1933–1938. Cambridge: Cambridge University Press. Getty, J. Arch, and Oleg V. Naumov. 1999. The Road to Terror: Stalin and the SelfDestruction of the Bolsheviks, 1932–1939. New Haven, CT: Yale University Press. Gorodetsky, Gabriel. 1999. Grand Delusion: Stalin and the German Invasion of Russia. New Haven, CT: Yale University Press. Gregory, Paul R. 1974. ‘‘Economic Growth, U.S. Defence Expenditures and the Soviet Defence Budget: A Suggested Model.’’ Soviet Studies 26(1): 72–80. Gregory, Paul R. 2004. The Political Economy of Stalinism. Cambridge: Cambridge University Press. Gregory, Paul R. 2008. Terror by Quota: State Security from Lenin to Stalin. New Haven, CT: Yale University Press. Harrison, Mark. 1996. Accounting for War: Soviet Production, Employment, and the Defence Burden, 1940–1945. Cambridge: Cambridge University Press. Harrison, Mark, ed. 1998. The Economics of World War II: Six Great Powers in International Comparison. Cambridge: Cambridge University Press. Harrison, Mark. 2003. ‘‘Soviet Industry and the Red Army Under Stalin: A MilitaryIndustrial Complex?’’ Les Cahiers du Monde russe 44(2–3): 323–42. Holloway, David. 1982. ‘‘Innovation in the Defence Sector.’’ In Industrial Innovation in the Soviet Union, ed. Ronald Amann and Julian Cooper, 276–367. New Haven, CT: Yale University Press. Holloway, David. 1983. The Soviet Union and the Arms Race. New Haven, CT: Yale University Press. Jersak, Tobias. 2003. ‘‘A Matter of Foreign Policy: ‘Final Solution’ and ‘Final Victory’ in Nazi Germany.’’ German History 21(3): 369–91. Kagan, Donald. 2004. The Peloponnesian War. New York: Penguin. Khlevniuk, Oleg. 1995. ‘‘The Objectives of the Great Terror, 1937–38.’’ In Soviet History, 1917–1953: Essays in Honour of R. W. Davies, ed. J. M. Cooper, Maureen Perrie, and E. A. Rees, 158–76. New York: St. Martin’s. Kozlov, V. P., et al. 2004. Istoriia Stalinskogo Gulaga. Konets 1920-kh-pervaia polovina 1950-kh godov. Sobranie dokumentov v semi tomakh. 7 vols. Moscow: Rosspen. Manning, Roberta T. 1993. ‘‘The Soviet Economic Crisis of 1936–1940 and the Great Purges.’’ In Stalinist Terror: New Perspectives, ed. J. Arch Getty and Roberta T. Manning, 116–41. Cambridge: Cambridge University Press. Mawdsley, Evan. 2003. ‘‘Crossing the Rubicon: Soviet Plans for Offensive War in 1940– 1941.’’ International History Review 25(4): 818–65. Mayer, Arno J. 2000. The Furies: Violence and Terror in the French and Russian Revolutions. Princeton, NJ: Princeton University Press. Medvedev, Roy A. 1971. Let History Judge: The Origins and Consequences of Stalinism. London: Macmillan. Mundell, Robert A. 1962. ‘‘Appropriate Use of Monetary and Fiscal Policy for Internal and External Stability.’’ IMF Staff Papers 9(1): 70–79.

30

Mark Harrison

Olson, Mancur. 1993. ‘‘Dictatorship, Democracy, and Development.’’ American Political Science Review 87(3): 567–76. Popov, V. P. 1992. ‘‘Gosudarstvennyi terror v sovetskoi Rossii 1923–1953 gg. (istochniki i ikh interpretatsii).’’ Otechestvennyi arkhiv, no. 2: 20–31. Raack, Richard C. 1995. Stalin’s Drive to the West, 1938–1941: The Origins of the Cold War. Stanford, CA: Stanford University Press. Rosen, Steven. 1973. ‘‘Testing the Theory of the Military-Industrial Complex.’’ In Testing the Theory of the Military-Industrial Complex, ed. Steven Rosen, 1–29. Lexington, MA: Lexington Books. Samuelson, Lennart. 2000. Plans for Stalin’s War Machine: Tukhachevskii and MilitaryEconomic Planning, 1925–41. London and Basingstoke: Macmillan. Simonov, Nikolai S. 1996. ‘‘ ‘Strengthen the Defence of the Land of the Soviets’: The 1927 War Alarm and Its Consequences.’’ Europe-Asia Studies 48(8): 1355–64. Solzhenitsyn, Alexander. 1974. The Gulag Archipelago, 1918–1956. London: Collins/ Fontana. Stalin, Joseph V. 1954. Works, vol. 11. Moscow: Foreign Languages Publishing House. Suvorov (Rezun), Viktor. 1990. Ice-Breaker: Who Started the Second World War? London: Hamish Hamilton. Tinbergen, Jan. 1952. On the Theory of Economic Policy. Amsterdam: North Holland. Tochenov, S. V. 2004. ‘‘Volneniia i zabastovki na tekstil’nykh predpriiatiiakh Ivanovskoi oblasti osen’iu 1941 goda.’’ Otechestvennaia istoriia 2004(3): 42–47. Uldricks, Teddy J. 1999. ‘‘The Icebreaker Controversy: Did Stalin Plan to Attack Hitler?’’ Slavic Review 58(3): 626–43. Weeks, Albert L. 2002. Stalin’s Other War: Soviet Grand Strategy, 1939–1941. Lanham, MD: Rowman and Littlefield. Wintrobe, Ronald. 1998. The Political Economy of Dictatorship. Cambridge: Cambridge University Press.

2

Before Stalinism: The Early 1920s andrei sokolov

In this chapter I will consider, first, the influence of historical continuity on Soviet military doctrine and military-industrial production. Specifically, what were the legacies of Tsarism, the events of World War I, the establishment of the Bolshevik dictatorship, and the civil war that led to Soviet Russia’s isolation in the international arena? Second, how did the defense industry adapt to market conditions under the New Economic Policy (NEP), and how did plan and market interact? This question has obvious importance for Russia today as it devises a military-industrial policy based on market relations between government, state-owned enterprises, and the private sector. Third, what was the real state of the Soviet defense industry and its management on the eve of Stalin’s emergence as dictator, and how did political and military leaders see its future at the time? Finally, to what extent did the immediate requirements of the defense industry contribute to the ‘‘Great Breakthrough’’ that ended NEP and established the Stalinist command system? Until recently, most Western textbook accounts described the economic context of this transition in purely civilian terms: the demands of capital construction, the grain crisis, and so on. Military-economic developments, when mentioned, were usually represented in terms of future needs, not present interests, although Carr and Davies (1969: 454–60) provided an honorable exception. With the blinkers of secrecy removed, more

31

32

Andrei Sokolov

recent studies by Simonov (1996a), Samuelson (2000), and Stone (2000) have begun to restore the balance. In fact, the interests of defense industry in the 1920s were a powerful influence that should not be ignored.

The Legacy of Tsarism and War Before 1914 Russia lagged behind the advanced countries in the production of modern weapons. This weakness became clear immediately after the outbreak of World War I in clashes with the better-equipped armed forces of imperial Germany. Allied deliveries compensated for these deficiencies only in part. Consequently the government had to resort to extraordinary mobilization measures in order to modernize and increase production of weapons and munitions. One reaction to wartime difficulties, which soon became a tradition, was the establishment of extraordinary agencies such as the Special Assembly for Defense of the State. The war demanded a degree of economic mobilization that was more intense in Russia than elsewhere, and it stimulated progress in aviation, tanks, vehicles, submarines, chemical weapons, and telecommunications. The new documents show considerable progress but with much delay and in quantities far less than needed. Documents from the Soviet period show that study of the experiences of World War I was important in forming the Soviet leadership’s thinking on military-industrial policy. A characteristic feature of defense industry in prerevolutionary Russia was the high proportion of state-owned enterprises, operating with backward technology and poor labor organization, that were engaged in supplying the armed service departments. Cash limits on spending and higher prices in the private sector hindered the government from switching orders to private and foreign establishments. In wartime sharp clashes developed between the state and the entrepreneurs and many private establishments working for defense underwent sequestration, opening the way for the subsequent Bolshevik nationalization of factories. Taking power in 1917, the Bolsheviks counted on concluding a peace treaty with Germany and her allies at Brest. Gambling on a lessening of military tensions, they began to demobilize industry and to curtail military programs. At this time some Bolsheviks debated whether war production was needed at all and, if so, whether for the world revolution or for national defense. However, the civil war in Russia quickly led to War Communism on the territory controlled by the Bolsheviks. What War Communism meant is well known; in its extreme centralization and suppression of the market, and especially in the enterprises nationalized by the Bolsheviks, it is easy to see features of the Russian state-owned factories before 1917.

Before Stalinism

33

War Communism created a vast number of extraordinary agencies and commissions dedicated to supplying the five-million-strong Red Army. These agencies were subordinated to the Extraordinary Plenipotentiary of the Committee of Defense (CHUSO, or Chusosnabarm), the antecedent of the State Defense Committee (GKO) that managed the Soviet defense effort during World War II. Military factories were subordinated in 1919 to the Council for Military Industry (SVP, Voenpromsovet) under CHUSO, and also to the Supreme Council for the National Economy (VSNKh), the ministry for stateowned industry. SVP itself became a special administration under the leadership of the Bolshevik P. A. Bogdanov, whose ideas helped to create the basic concepts of the Soviet defense industry complex.∞ War Communism bequeathed a lasting legacy to the defense industry. Some contemporary Russian authors (Prisiazhnyi 1994) fairly describe War Communism as an ‘‘economic plague’’ that released its ‘‘bacilli’’ into subsequent Soviet history. In fact the management of the Soviet defense industry in the 1920s reveals the strong influence of War Communism. The NEP reforms of privatization and the transfer of state-owned enterprises to independent trusts operating on commercial accounting principles had little effect on military factories. Most of them, sixty-two factories with one hundred thousand employees, remained under the management of the chief administration of military industry (GUVP) under the ministry for state industry as a whole. These became the nucleus of the Soviet defense industry complex; only a few were handed over to the trusts. After War Communism the Politburo was in charge of general economic and military strategy. It delegated decisions on the size and structure of the armed forces, the future of the defense industry, the scale of military equipment purchases, and the appointment of military leaders to the Revolutionary Military Council (RVS, Revvoensovet). Practical issues of defense construction were devolved to the Council of People’s Commissars (SNK, Sovnarkom), the defense ministry (NKVM, Narkomvoenmor), and the Red Army Staff. These were responsible for fixing requirements for particular types of armament; the Council of Labor and Defense (STO) operated under Sovnarkom as a specialist commission for practical issues concerning civilian and military production. Under STO Gosplan answered for current and perspective (fiveyear) plans, including plans for armament. The actual production of weapons was concentrated under the ministry for state industry (VSNKh), which, through its Committee for Military Orders (KVZ VSNKh), allocated equipment orders to enterprises including those subordinated to GUVP. To simplify, the market for weapons consisted of two chief players: a principal and an agent. The principal was the buyer and the agent was the supplier.

34

Andrei Sokolov

From this point forward we will call them the Army and Industry respectively, and this convention is also followed in the remaining chapters of the book. Of course these terms are a shorthand; they impose analytical simplicity on a messy historical reality. But they are a simplification that was already in use at the time. Our ‘‘Industry’’ translates the contemporary concept of Voenprom (voennaia promyshlennost’, military industry), which subsumed all the enterprises and coordinating bodies associated with the production of armament and direct provisioning of the Red Army. Similarly, our ‘‘Army’’ stands for the institutions and interests aligned with the contemporary Voenved (voennoe vedomstvo, the military department or departments) that acted as the purchaser or principal, including the army and navy staffs that compiled the production programs and plans.≤ Figure 2.1 shows these two chief players in relation to the other bodies whose role is discussed in this chapter. The conclusion of the civil war allowed for demobilization and reductions in defense production and in military reforms. In June 1921 Revvoensovet announced a cutback of military equipment orders. Red Army demobilization began; its strength fell to 1.6 million at the end of 1921, then 800,000 in 1922, and 610,000 in 1923. In 1924 a special commission of Revvoensovet set about a military reform. Some scholars (Simonov 1996a: 59) consider that this reform was based on the ideas of M. V. Frunze about a war of national defense; Frunze replaced L. D. Trotskii, who advocated the export of revolution, as defense minister in 1925. More likely, the Soviet state simply could not afford a large regular army, let alone the production of armaments. In accordance with the reform, military service was reorganized on the lines of a territorial militia. The regular strength of the Red Army, limited to 562,000, was to serve as the basis for training reserves who would remain in civilian employment. The domestic production of armaments was to be based on this much smaller military force. Military documents of the mid-1920s warn of pacifist and demobilizing sentiments with which the Army had to contend. This was pacifism not as an ideology but as an expression of extreme exhaustion after seven catastrophic years of war. Hence many officials were reluctant to engage with defense practicalities, despite their own rhetoric concerning the ‘‘looming imperialist war against the Soviet republic.’’ The unpleasant issue of preparing for a possible attack associated with border tensions was replaced by an energetic struggle for peace and international recognition. After the civil war Trotskii ceased to be involved in detailed military issues, being occupied with the power struggle. The central committee defense commission that he headed, the Trotskii commission, became inactive, and in 1924 Trotskii was replaced by A. I. Rykov, so it became the Rykov commission. With such sweeping

Before Stalinism

35

changes, practical concerns relating to the Army and its supply were left to the military leaders. Their demands were pushed down the queue; at times they were brushed off like irritating flies. The leaders of Industry continued to advocate the concentration of production. There were attempts to delegate this task to KVZ, the VSNKh committee for military equipment orders that allocated orders among the producers. GUVP chief Bogdanov wanted to assign this task to Gosplan or even establish a specialized interdepartmental agency under Sovnarkom or STO analogous to the old CHUSO. Most military leaders favored the idea of defense industry as an autarkic structure. Bogdanov based this on strategic and production considerations. Strategically the USSR would have to supply its forces solely from domestic resources; unlike Tsarist Russia it could not count on foreign assistance in the event of war.≥ Among the military it was increasingly believed that the country ought to be able not only to develop and deploy new types of weapons but also match their cost and quality to the levels achieved by other states. During the mid-1920s the leaders of Industry worked out what would become enduring features of the Soviet defense industry complex. Armaments were classified into three groups of varying technological proximity to civilian industry. The first group, most remote from civilian production, would be allocated to specialized military factories: small arms and ammunition, gunpowder, explosives and poison gas, mortar shells, and so on. The second group was closer to civilian industry; it comprised field and naval guns, shipbuilding, aviation, tanks, military optics, and radio communications. Here too it was considered desirable to close the production cycle as far as possible, so this group was also to be allocated to specialized military factories. The third group, comprising electrical equipment, transport and communications equipment, instruments, and personal kit and ration stores, was closest to civilian goods and could be allocated to civilian producers without detriment to defense interests.∂ These ideas subsequently led to a distinction between the military-industrial ‘‘cadre’’ factories or ‘‘numbered’’ factories, and the factories of the reserve; this corresponds with the Russian distinction between the regular (kadrovaia) army and the army of the reserve. In peacetime, the cadre factories would meet current equipment needs, engage in improvement and innovation, establish mass production, coordinate mobilization measures, and deploy those measures in the event of war. In peacetime the war production facilities of the reserve would be mothballed. The reserve factories would compile mobilization plans for procuring their own supplies in wartime from domestic, not foreign, sources.

Before Stalinism

37

World War I had shown that in a protracted war the foundation of defense was no longer the army but the sum of national resources. War required total mobilization, and this took time. Hence the decisive components of the national economy had to be developed beforehand, especially the branches supplying the defense industry, such as steel and machinery; these would become the vector of Stalinist industrialization. Mobilization plans were supposed to be based on the role of each enterprise in the wartime mobilization and division of labor, and to set out an order of transition to wartime work. Since military production is intensive in skill, the mobilization plan included a plan for advance training of civilian engineering and technological staff. The large-scale enterprises were obliged to develop close links with research and establish design bureaus (KB), while research institutes (NII) were to develop military technology. The ‘‘cadre’’ military factories were to be generously endowed with equipment for design and experimentation. Research institutes and design bureaus were to set up experimental production and establish direct links with Industry’s enterprises. Discipline, organization, and secrecy were fundamentals. Notable among the agencies established after the civil war to administer Industry was the VSNKh Committee for Demobilization and Mobilization (‘‘De- and Mobilization’’) of Industry (see Figure 2.1). At first it was charged with reconverting industry to peaceful production but it soon acquired further sweeping responsibilities for mobilization preparedness and mobilization planning throughout industry, the designation and security classification of defense-related projects, and the separate channeling of defense-related budgetary finance. Meanwhile, propaganda promoted the militarization of the civilian population. It claimed that the world bourgeoisie was planning to make war on the USSR. It sought to inspire popular trust in the Red Army and encouraged mass

Figure 2.1. The Army and Industry Key: The Army (voennoe vedomstvo), the service department(s) purchasing military goods: NKVM: People’s Commissariat for Military and Naval Affairs (the defense ministry). RVS: Revvoensovet, the Revolutionary Military Council (board of the defense ministry). SNK: Sovnarkom, the Council of People’s Commissars (government ministers). STO: Council of Labor and Defense. Industry (voennaia promyshlennost’), the industrial departments supplying military goods: GUVP: Chief Administration of Military Industry of VSNKh. KVZ: Committee for Military [Equipment] Orders of VSNKh. VSNKh: Supreme Council of National Economy (the ministry of state industry). Controllers: OGPU: Unified Chief Political Administration (the government agency responsible for state security). TsKK-RKI: the joint party Central Control Commission and state Workers’ and Peasants’ Inspection.

38

Andrei Sokolov

participation in a range of voluntary organizations to assist with both military and civil defense.

Defense Industry Versus the Market At the beginning of NEP the Soviet economy was in the deepest crisis. A huge number of tasks were pressing, and as economic priorities were reordered military affairs did not rank high. NEP reintroduced market relations and economic incentives as a matter of practical necessity, in order to nurse an ailing economy back to health. Civilian economic recovery was rapid, though not problem-free; its features are well known, so I turn immediately to the situation of the defense industry. The management reforms of 1921 to 1923 reorganized industry into incorporated trusts. In the state-owned sector, the largest and best-supplied enterprises were subordinated directly to the ministry of state industry (VSNKh). The rest were available for lease or franchise. The VSNKh enterprises were grouped and unified into trusts that operated on a commercial basis. Lossmaking or unprofitable enterprises were liable to closure or mothballing. In 1922 to 1924 a financial reform improved the public finances. Monetary management severely cut back the budget deficit. As a result the ruble was strengthened and could be used again for financial accounting. Enterprises that had come to a standstill, as in the case of the military factories, were assigned special budget credits and subsidies and were also allowed to fix premium (‘‘recovery’’) prices for finished output. The problem for defense industry was that, as Table 2.1 shows, state budget revenues in the mid-1920s remained significantly below the 1913 level. In preparing the financial reform the issue of how to protect defense outlays was considered more than once, but defense industry’s credits and subsidies were curtailed along with civilian producers. From 1921 to 1926 defense outlays fell behind other categories of spending and were reduced to 12.7 percent of budgetary outlays, well below the 1913 level. Industry was faced with the impossible task of supplying the Red Army under conditions of extreme fiscal austerity. In 1921 GUVP chief Bogdanov wrote to Lenin that the defense industry was on the brink of disaster. Reserves were exhausted. Some factories would require many years’ convalescence. Workshops were cluttered with useless equipment. Repair facilities had disintegrated. Buildings were dilapidated. Labor was badly organized; skills and discipline were low and absenteeism was prevalent. Few specialists were left. The quality of output had fallen. Liquidity and credit constraints had brought supplies and services to a standstill. It was stressed repeatedly that defense

Before Stalinism

39

Table 2.1. The Russian state budget of 1913 and the Soviet state budgets of 1922/23–26/27, at comparable prices Gold rubles (millions) ∞Ω∞≥ ∞Ω≤≤/≤≥ ∞Ω≤≥/≤∂ ∞Ω≤∂/≤∑ ∞Ω≤∑/≤∏ ∞Ω≤∏/≤π

∏,∫≥∏ ∞,∂∏≥ ≤,≤Ω∫ ≤,Ω∑∏ ≥,Ω∑Ω ∑,∞∑∑

Sources: Ob’iasnitel’nye zapiski (1927); Kuznetsov (1927).

industry was not like other branches; the restoration of good order would take a long time.∑ In short, Industry was being expected to raise efficiency and quality, cut costs, transfer to profit-and-loss accounting, and take on additional nonmilitary orders up to one-quarter of the volume of production, all in a period of financial austerity.∏ Under NEP the military factories were financed partly by selling output at fixed prices and partly from advances, credits, and accounts with the Army. ‘‘Guide’’ prices for military goods were based on actual unit costs, but without disclosing cost estimates to the purchaser. In cases of extreme necessity budget subsidies were available, and there was financial support for the ‘‘nursing’’ of enterprises. The central apparatus of Industry was maintained by the budget. Enterprises that worked only on military orders and experimental factories depended fully on the budget. Budget allocations to other defense producers were cut substantially. In factories where military orders accounted for threequarters of gross output, funding fell by one-half. In practice, however, financial accounting between the Army and Industry did not operate. Military factories were continually underfinanced and the financial support provided through advances was inadequate. Some goods for military use such as motors, light bulbs, and radio equipment bypassed defense industry entirely.π The government lacked the means to either mothball enterprises or nurse them back to health, let alone to rebuild production facilities. Reporting on the fulfillment of the ‘‘nursing’’ program for factories in 1923, GUVP indicated that ‘‘at present they are unable to supply armaments to a single army on the western front in the event of war in consequence of their economic and financial weakness. . . . Stocks of military equipment are

40

Andrei Sokolov

insignificant.’’∫ From 1921 to 1923 defense industry underwent every possible kind of crisis over fuels, materials, consumer goods, and finance.Ω Unlike light-industry enterprises, which had recovered their prewar output by 1923, the recovery of heavy industry, and particularly of military producers, was more difficult. Their lack of liquidity made it hard for them to pay suppliers. Industry officials fought to protect centralized allocation of supplies to the defense industry. The transfer of enterprises to corporate profit-and-loss accounting did not work well given that the defense industry was not operating in a competitive market, and GUVP turned repeatedly to the Politburo and Sovnarkom to amend the procedures for financing defense industry.∞≠ But there was little help there. Tasks were being set for defense industry as though it was like other branches while, as Bogdanov wrote, the state remained ‘‘deaf’’ to its needs; he emphasized that defense industry could not emerge from this crisis without help.∞∞ The simultaneous curtailment of military equipment orders and supplies forced enterprises, where possible, to turn to producing civilian commodities while selling off remaining material stocks and even equipment to gain a minimum level of working capital. Substantial indebtedness developed across Industry. The wage bill alone absorbed 88 percent of available funds in place of the planned 33 percent. Despite this, wage arrears were large. Yet wage increases were an indispensable condition for ‘‘nursing’’ factories back to health, given the collapse of real wages in the preceding years. Military factories paid workers in rations and money surrogates. Unlike other branches the defense industry could not pay in kind from its own products. As a rule the authorities tried to link wage increases with productivity gains and rising piece-rate norms, and productivity and wages remained lower in the defense industry than in other branches of industry. Increasing norms and wage arrears were the main source of growing worker discontent in the military factories. The volumes published so far in the documentary series Sovershenno sekretno (Top Secret) contain many reports of defense factory workers’ discontent (Sevost’ianov 2003– ). On this knife edge enterprises had to utilize every scrap of revenue for wage payments. Wage increases led in turn to rising unit costs of military goods, the prices of which approximately doubled in comparison to the prewar level.∞≤ Industry requested the government to set prices for military goods on a noncommercial basis free of indirect taxes. They justified this by the wide range of products in military equipment orders and the complexity of calculation, and they also claimed that heavy indirect taxes would weigh defense industry down.∞≥ The government, which wanted to extend the NEP reforms, was reluctant. At the end of 1924 STO adopted a resolution on standardizing

Before Stalinism

41

the pricing of military equipment on the same basis as for other industrial goods: the unit cost plus duties, excises, and a 3 percent profit tax.∞∂ In January 1925 a new resolution on trusts was adopted that again deprived defense industry of a further range of concessions and favors. Industry repeatedly sought to have prices, financing, and the collection of duties and taxes on military production regularized, and to have growing arrears excused or written off where possible.∞∑ The problems of defense industry continued into 1925/26. Output and employment continued to rise, but production fell short of the results that the Army hoped for so that the plan fell increasingly into arrears. Productivity lagged and labor discipline deteriorated. ‘‘Firm’’ prices turned out to be below costs, and the financial loss of the defense industry ran to millions of rubles. In short, the ‘‘soft budget constraint’’ that would characterize state-owned industry more generally under the command system, along with the continual renegotiation of prices, taxes, and subsidies, was already clearly visible in Soviet defense industry under NEP. The immediate causes of the shortfalls were said to be planning deficiencies, late delivery of supplies, and deficient technical documentation. But the fact is that the defense industry was not developing as the Soviet government wished. Moreover, it would now be required to move on from recovery to reconstruction. Industry was faced with substantial increases in production assignments that went together with funding for new projects.∞∏

Defense Industry and Military Needs The contraction of defense funding evoked protests from both the Army and Industry. In mid-1923 deputy GUVP chairman I. I. Smirnov reported to F. E. Dzerzhinskii in the latter’s capacity as a member of the Trotskii commission that the Red Army had been cut to the bone. It was pointless to expect foreign assistance in the event of war. On the contrary, he wrote, if there was a workers’ insurrection abroad the Soviet Union would be expected to supply weapons. The sixty-two factories at the disposal of Industry were clearly insufficient given Russia’s poor industrial mobilization potential and lack of mobilization plans; an investigation of mobilization plans in seventeen of the largest factories in Russia had exposed their lack of readiness.∞π As for the nursing of military factories, GUVP reported that 18 million rubles was not enough.∞∫ Reviewing the defense cutbacks, Gosplan claimed that if war broke out it would be possible to supply the army with rifles, but machine guns would be in short supply, and the supply of cartridges would be enough for no more than

42

Andrei Sokolov

two or three months.∞Ω And from the Army the military inspector S. S. Kamenev, a former army commander, declared that he was against further cutbacks not because of issues like war and world revolution, but just to conserve the army, its regular personnel, and its specialists.≤≠ The Army was among the first to compile current, long-term, and perspective plans for the Soviet economy. Revvoensovet first debated a five-year plan for military construction in May 1923. In terms of defense industry it set very modest objectives: to return the factories to their prewar level over five years. This would require 2,900 million rubles at current commodity prices over five years with continual year-on-year growth of investment. Beyond that, to raise the army just to the level of Russia’s neighbors would take tens of billions of rubles. The first year’s task was just to improve living conditions for service personnel, and raise pay and family support for the command staff. The second priority was to be aviation, although artillery was being supplied at only 60 percent of the required level and the cavalry at 75 percent. The plan would do nothing for the tank forces, and provide the navy only with money wages and salaries. Meanwhile defense industry could not be fully supplied from the weak domestic economy so some needs would have to be met from imports. Many items necessary for war production were virtually unobtainable, including nickel, tin, zinc, aluminum, lead, nitrogen, saltpeter, arsenic, and bromine.≤∞ Table 2.2 shows an estimate of the productive capacities of the specialized defense industry relative to requirements at the end of 1923. None were adequate; the situation was entirely unsatisfactory in relation to firearms, flares, and motors, and most threatening in relation to machine guns, rifles, cartridges, and motor vehicles. Just to undertake the first ‘‘nursing’’ measures it would be necessary to raise purchases abroad. Instead, however, the government curtailed foreign currency allocations to Industry on the pretext of the improving international situation.≤≤ The evidence of individual branches of the defense industry in the mid- 1920s confirms this overview. Soviet artillery lagged in range, rate of fire, and motorization.≤≥ Associated with this was the rudimentary state of the automobile industry. The tank industry was largely occupied with repairing vehicles captured in the civil war.≤∂ Aircraft production remained well below the level of 1916; the aircraft themselves were obsolete and relied on imported parts.≤∑ As for the navy, its shrunken fleet now relied on the Baltic shipyard, the only one that had succeeded in conserving its equipment.≤∏ In every case there were ambitious plans for industrial reconstruction and expanded military procurement, but the plans themselves carried little credibility. At times the resulting tensions led to infighting within the military as when, for example, the army raided the naval budget to strengthen the ground and air forces.≤π

Before Stalinism

43

Table 2.2. Productive capacities of GUVP factories compared with requirements, end of 1923 % of required Guns Gun matériel Aircraft Firearms, flares, and motors Machine guns Rifles Cartridges Motor vehicles

∏∑ ∑∑ ∂π ≤π ∞∏.∑ ∞∑ Ω.∑ ∞.∑

Note: The figures cover the factories of the GUVP (chief administration of military industry of the ministry for state industry). Source: RGVA, 4/1/42: 262–264.

In short, the claims of the Army on supplies of armament far exceeded what was possible under NEP. An intense struggle over priorities for the use of limited resources ensued. Concerns over the backwardness of defense industry and the other sectors supplying defense were more sharply voiced. The Soviet armed forces lagged behind Russia’s neighbors in both quantity and quality; if war broke out the neighbors’ armies were expected to be the first wave behind which the forces of economically more powerful countries would inevitably follow.

‘‘Who Is to Blame?’’ People began to ask who was to blame for the lamentable state of the defense industry. A natural first reaction was to blame those in charge of it. A barrage of criticism aimed at the leaders of Industry rose from year to year. The latter did not deny the problems and accepted that military supplies were not up to requirements. They rested their defense on the dislocation of military and civilian programs and the lack of means to maintain the defense industry; they sought the establishment of an agency with unified responsibility for its development and a framework of firm plans for the next three to five years.≤∫ But from the side of the Army, in 1925 a Revvoensovet resolution blamed Industry. It described the producers’ aims as unchanged since before the war. It complained of the low quality and high cost of military goods, continual underperformance, inadequate capacities, lack of communication and coordi-

44

Andrei Sokolov

nation, bad planning, and inadequate stockpiling of mobilization reserves.≤Ω In June 1925 Bogdanov asked Dzerzhinskii as minister of state industry to relieve him, stressing that he could not cope with the problems that Industry was required to solve.≥≠ The official propaganda of the 1920s claimed that the Red Army was acquiring new weapons, that artillery matériel was improved, and that aviation was advancing in quality as in quantity. It featured new means of chemical defense and military-medical equipment. It asserted that the Red Army was gradually converging on the level of the first-class armies. But the reality was different. The capacities of defense industry were not only less than in 1916, when Russia’s war production had attained its previous peak; they were also less than in 1913. Not even half the restoration of defense industry had been achieved, and this was especially depressing in the context of continuing advances in military technology in the West. Thus the exceptional secretiveness surrounding defense industry was motivated not only by a desire to conceal new developments, as was claimed, but also by a reluctance to disclose the weakness of defense to the population and to potential adversaries. In early 1927 Gosplan’s defense sector concluded from a review of war readiness that neither the Red Army nor the country was prepared.≥∞ A report by deputy defense minister Tukhachevskii concurred, stating that if war broke out their ‘‘meager combat mobilization stocks [would] scarcely last through the first period of the war.’’≥≤ As was usual in Soviet practice, a bad situation drew the attention of the political and control agencies. In December 1924 the OGPU special department reported that the GUVP apparatus was good for nothing; this had created a threatening situation for the country’s defensive capability. It was noted that the mobilization preparedness of civilian branches was in an especially bad state. Of the country’s 4,200 factories only 75 were included in higherlevel mobilization plans, and of these only 42 had their own factory-level mobilization plans. Investigation of the military factories had revealed losses, defective output, and low standards. Blame for this was laid on the old specialists who, it was said, were holding out one by one in the factories, cared only for their self-interest, and took no part in solving production problems. It was asserted that the scientific cadres in military enterprises had been contaminated by elements alien to the communists; a first priority had to be to restaff the apparatus with communists.≥≥ A further implication was the need for onthe-spot promotion of Bolshevik specialists who had gained leadership experience during the civil war. A state inspection commission also reviewed Industry. It reported that the factories’ civilian output amounted to 10 million rubles, much of which re-

Before Stalinism

45

mained unsold; it was expensive, of low quality, with many defective items. The losses were being covered out of funds assigned to military production. The same report listed irrational labor organization, huge numbers of whitecollar and auxiliary workers in the military factories, and high unit costs. The inspectors found the chief cause of these deficiencies in the fact that GUVP had maintained the same expensive apparatus unchanged since the civil war. But special emphasis was also laid on the ‘‘dominance of the spetsy,’’ meaning the pre-revolutionary specialists, and of the top brass of the old army.≥∂ These clearly show a tendency toward the ‘‘baiting’’ of the old specialists (spetseedstvo), a desire to heap blame on them and replace them with people who would measure up to the political tasks of the Soviet leadership rather than to professional standards; the prosecutions of specialists, including more than a few staff of Industry, were already looming. At first, however, the leadership tried to solve these problems by reorganization. In November 1925 a chief military-industrial administration (GVPU) was established under the ministry of state industry to lead the defense industry by bringing together all the structures of Industry; subsequently this became the production association of military industry (VPU, also known as ‘‘Voenprom’’). Thus the branch associations that would characterize industrialization under the ministries of the future were first observed in military production. The new Voenprom comprised four trusts: ‘‘Gun-Arsenal,’’ ‘‘Cartridge-Tube,’’ ‘‘Military-Chemical,’’ and ‘‘Rifle-Machine Gun.’’ But these trusts were not autonomous corporations; rather, each was really a ministerial subunit. At the same time the output of Voenprom was brought into a unified economic plan, and this was expressed in the control figures adopted the following year. On October 5, 1926, STO adopted a resolution that obliged the ministry of state industry to compile a unified plan for both military and civilian output with the agreement of the defense ministry, and to refer it to Gosplan for expert review, then to STO for confirmation.≥∑ Revvoensovet member I. S. Unshlikht turned to Stalin’s intimate colleague, the new defense minister and Revvoensovet chairman K. E. Voroshilov, with a call to establish new principles for interaction between the Army and Industry. The contracts that the army concluded with defense industry, he suggested, should not only guarantee fulfillment but also establish criminal and material responsibility for all shortfalls and violation. Revvoensovet should have the right to supervise and verify the progress of equipment orders, research and design work, stocks of materials, cost calculations, and real production capacities, in peacetime as in wartime. Reequipment, decommissioning, the transfer of production from one factory to another, and new factory building should be done only with the authority and agreement of Revvoensovet. The board of

46

Andrei Sokolov

the ministry of state industry should provide Revvoensovet with an exact count of all factories currently working on war-related orders under civilian industrial management; Revvoensovet should come to an agreement with the industry ministry concerning those that should be retained in the core of the defense industry. Unshlikht demanded that military employees should be appointed to reinforce the administrative and technological apparatus. The industry ministry should agree on nominations to both leading management positions and the specialized technological apparatus with Revvoensovet.≥∏ Unshlikht’s proposal fell. Although Stalin and his immediate supporters were increasingly in sympathy with the idea of going over to a command economy, they did not want soldiers in command. The politicians wanted to be in charge of the show themselves, and this required a defense industry under civilian control. The episode illustrates, nonetheless, the tendency to reinforce the role and influence of military considerations in the growth of the Soviet state. As Simonov (1996a,b) and Stone (2000) have shown in taking up the story of the later 1920s, this tendency would receive further impetus as a result of the ‘‘war scare’’ of 1927. The legacy of the past played a large role in the emergence of the Soviet defense industry complex. Whatever the new features introduced in the 1920s, the restoration of the defense industry and its steps out of the crisis followed a path that was already habitual for the country. Its management inherited many features of war communism which, in its turn, borrowed substantially from the traditions of pre-revolutionary Russia. The concept of the defense industry complex that was formed in the 1920s, its management systems and the principles on which it interacted with the civilian economy, were based on historical experience and created antecedents of the future Soviet administrative-command system. An overview of Soviet defense industry in the 1920s must conclude that in principle its problems could have been solved by more market reforms within NEP; however, neither the political leadership nor the military, holding to their revolutionary ideals and brought up in the spirit of war communism, were ready for this. Industry demanded more budgetary support, firm plans, centralized management, subordination of the civilian branches to its own requirements, concentration of production in the ‘‘cadre’’ military factories, and the rolling back of market relations. The Army proposed plan targets that far exceeded the possibilities of the economy. Both sides advocated forced industrialization to increase military power. Pressure from the military, kept within reasonable limits for a time, grew stronger. Thus, immediate militaryeconomic considerations contributed to the end of NEP and the transition to

Before Stalinism

47

directive planning. Their role was, if not decisive, then certainly more significant than has been thought. One might say that the military were increasingly influential advocates of industrialization as a way of solving the country’s economic problems, but this, although true, is not enough. More importantly, the military were a powerful voice on the side of solving problems through mobilization and repression. If supporters of a continuation of NEP were to be found in the civilian branches, the position of military leaders was surprisingly homogeneous and tended to converge on unanimous support for going over to a command system. In balancing priorities for the economy in general and military production in particular an arbiter became necessary to decide matters in favor of the military by an act of will, even when the means and resources were clearly insufficient. The establishment of Stalin’s dictatorship matched these tendencies. For a time the political leadership had kept aloof from the business of armaments. The ‘‘war scare’’ of 1927, however, would impel Stalin and his circle to become deeply absorbed in military production issues, and this in turn would presage further great changes.

Notes This chapter is a revision of the author’s article ‘‘Before Stalinism: The Defence Industry of Soviet Russia in the 1920s,’’ published in 2005 in Comparative Economic Studies 47(2):437–55. 1. Stone (2000: 28) gives priority to another defense manager of the time, V. A. Avanesov, whose importance was to think through the enforcement of political, not military, control of military industry. 2. This also matches the distinction between the supply departments and (armed) service departments used, for example, by Hancock and Gowing (1949) in their history of the British war economy. 3. RGAE, 2097/1/64: 8–24. 4. RGAE, 2097/1/64: 8–24. 5. APRF, 3/46/330: 7. 6. RGAE, 2907/5/181: 12–13. 7. RGAE, 7733/1/2205: 1–2. 8. GARF, 5446/55/348: 31. 9. RGAE, 2907/1/949. 10. GARF, 5446/55/348: 33. 11. APRF, 3/46/330: 7. 12. GARF, 374sch/28s/301: 95–98. 13. GARF, 5446/5à/549: 3–9. 14. GARF, 5446/5à/549: 69–69ob.

48

Andrei Sokolov

15. GARF, 574/5/969: 2–3. 16. RGASPI, 76/2/177: 55–64. 17. RGASPI, 76/2/17: 98–131; RGAE, 3429/10/144: 46–81. 18. RGVA, 33987/2/212: 309. 19. RGASPI, 325/1/524: 3–7. 20. RGASPI, 325/1/524: 11–16. 21. RGASPI, 76/2/17: 78–85. 22. RGASPI, 17/3/431: 16–19. 23. RGVA, 33988/2/651: 13–16; RGVA, 4/1/145: 3–8. 24. RGAE, 3429/10/113: 16–17. 25. RGAE, 3429/10/309: 118–121. 26. RGAVMF, R1483/2/7: 88–95. 27. RGAVMF, R-1483/2/17: 131–131ob. 28. GARF, 5446/55/747: 8–14. 29. GARF, 5446/55/747: 5–6. 30. GARF, 5446/55/747: 39–41. 31. RGAE, 4372/1/24: 1–3. 32. GARF, 8418/16/3: 335. 33. RGASPI, 76/3/377: 44, 50, and 51–54. 34. APRF, 3/46/330, 79–82: 85–93. 35. RGVA, 4/1/318: 1. 36. RGVA, 4/2/172: 227–228.

Published References Barber, John, and Mark Harrison, eds. 2000. The Soviet Defence-Industry Complex From Stalin to Khrushchev. London and Basingstoke: Macmillan. Bergson, Abram. 1961. The Real National Income of Soviet Russia Since 1928. Cambridge, MA: Harvard University Press. Bergson, Abram. 1964. The Economics of Soviet Planning. New Haven and London: Yale University Press. Bystrova, Irina. 2000. Voenno-promyshlennyk kompleks SSSR v gody kholodnoi voiny. (Vtoraia polovina 40-kh-nachalo 60-kh godov). Moscow: Institut Rossiiskoi istorii Rossiiskoi Akademii nauk. Carr, E. H., and R. W. Davies. 1969. Foundations of a Planned Economy, 1926–1929, vol. 1. London: Macmillan. Davies, R. W. 1993. ‘‘Soviet Military Expenditure and the Armaments Industry, 1929– 33: A Reconsideration.’’ Europe-Asia Studies 45(4): 577–608. Davies, R. W., and Mark Harrison. 1997. ‘‘The Soviet Military-Economic Effort Under the Second Five-Year Plan, 1933–1937.’’ Europe-Asia Studies 49(3): 369–406. Gregory, Paul R. 2003. ‘‘Soviet Defence Puzzles: Archives, Strategy and Underfulfillment.’’ Europe-Asia Studies 55(6): 923–38. Hancock, W. K., and M. M. Gowing. 1949. British War Economy. United Kingdom History of the Second World War, Civil Series. London: HMSO. Harrison, Mark. 1996. Accounting for War: Soviet Production, Employment, and the Defence Burden, 1940–1945. Cambridge: Cambridge University Press.

Before Stalinism

49

Harrison, Mark. 2003. ‘‘Soviet Industry and the Red Army Under Stalin: A MilitaryIndustrial Complex?’’ Les Cahiers du Monde Russe 44(2–3): 323–42. Istoriia OPK. 2003– . Istoriia sozdaniia i razvitiia oboronno-promyshlennogo kompleksa Rossii i SSSR. 1901–1963 gg. Dokumenty i materialy. Moscow: Rosspen. Kuznetsov, S. 1927. ‘‘Biudzhet za 10 let.’’ Vestnik finansov 1927(11). Moorsteen, Richard, and Raymond P. Powell. 1966. The Soviet Capital Stock, 1928– 1962. Homewood, IL: Irwin. Ob’iasnitel’nye zapiski. 1927. Ob’iasnitel’nye zapiski k biudhzetu 1925/26 i 1926/27 gg. Moscow. Prisiazhnyi, N. S. 1994. Ekonomicheskaia chuma: voennyi kommunizm v Rossii. Istoriko-ekonomicheskii analiz. 1918–21 gg. Rostov on Don: Rostovskii Gosudarstvennyi universitet. Samuelson, Lennart. 2000. Plans for Stalin’s War Machine: Tukhachevskii and MilitaryEconomic Planning, 1925–41. London and Basingstoke: Macmillan. Sevost’ianov, G. N., ed. 2003– . ‘‘Sovershenno sekretno’’: Lubianka-Stalinu o polozhenii v strane. 1922–1934. 10 vols. Moscow: Institut Rossiiskoi istorii Rossiiskoi Akademii nauk. Simonov, N. S. 1996a. Voenno-promyshlennyi kompleks SSSR v 1920–1950-e gody: tempy ekonomicheskogo rosta, struktura, organizatsiia proizvodstva i upravlenie. Moscow: Rosspen. Simonov, N. S. 1996b. ‘‘Strengthen the Defence of the Land of the Soviets: The 1927 War Alarm and Its Consequences.’’ Europe-Asia Studies 48(8): 1355–64. Stone, David R. 2000. Hammer and Rifle: The Militarization of the Soviet Union, 1926– 1933. Lawrence: University Press of Kansas.

3

Hierarchies and Markets: The Defense Industry Under Stalin mark harrison and andrei markevich

The defense industry was one of the most important components of Stalin’s military, economic, and political system. Its military significance is suggested by the fact that Soviet defense factories outproduced Germany in World War II and rivaled those of the United States for the next half century. Table 3.1 gives an idea of some bare numbers. These show, for example, that in the peacetime decade before World War II the Soviet aviation industry produced more than 4,000 aircraft per year. In wartime the rate went up to nearly 30,000 a year on average; it goes without saying that the metallic monoplanes that fought for the skies in the 1940s bore little comparison with the wood-and-canvas biplanes of the early 1930s. The defense industry was of great economic significance. The industry itself is best imagined as a relatively small hub of large-scale, specialized assembly plants and research units surrounded by a wide periphery of part-time subcontractors that provided the defense industry hub with the more generalpurpose materials, components, fuels and energy, and services that it needed while supporting the civilian economy as well. In 1938, for example, military orders accounted for just over 16 percent of the gross output of ‘‘civilian’’ engineering factories.∞ The 1939 plan for production of shell parts designated no less than eighteen different civilian ministries as subcontractors.≤ Specialized military factories naturally accounted for

50

Hierarchies and Markets

51

Table 3.1. Soviet war production, 1930–45 (units)

Aircraft Aircraft engines Armored combat vehicles Artillery units Mortars Machine guns (thousands) Machine pistols (thousands) Rifles (thousands) Shells (thousands) Bombs (thousands) Mortar shells (thousands) Rifle cartridges (millions) a

∞Ω≥≠–∂≠

∞Ω∂∞–∂∑

∂∑,≠∫∞ πΩ,≥∫≥a ≤∫,∂∫π π≤,∞≤Ω ∂∂,∫≤∑ ∑≠≥ ∞≠≥ ∏,≥∫∫ ∏≥,∑∑≤ ∞∂,≥∏∑ ≤∞,∏≤Ω ∞≠,∂∂≤

∞∂≤,π∂≠ ≤≠∫,∫π∑ ∞≠∏,∫∫∫ ∂Ωπ,∏∫∑ ≥∏∞,∏≠≠ ∞,∑∞≤ ≤,∞≥∑ ∞≥,≠∏≤ ≥∂∑,∂∑π ∂Ω,∫∑Ω ≤∏π,Ω∫≥ ∞≥,∫∫≠

1930–38 only.

Sources: 1930–40, Davies and Harrison (1997), except aircraft engines from Kostyrchenko (1992: 429); 1941–45, Harrison (1996).

the bulk of supply of all military items; in 1940, for example, the figure was 78 percent (Simonov 1996: 154). But they also contributed significantly to the supply of civilian products: as late as 1938 about one-fifth of their gross output still took the form of civilian items, and five years earlier the analogous figure had stood at more than one-third.≥ The interpenetration with the civilian economy makes the exact scope of the defense industry hard to define. The fuzziness was partly deliberate: the peripheral facilities were designed to swing over from civilian to war production in the event of war. The blurring of boundaries was also an inevitable product of rapid technological change: the pattern resulting when a radio wave rebounds from a solid object was a scientific curiosity in one decade and the object of a vast, secret, specialized industry in the next. As for orders of magnitude, a purely administrative definition of the industry gave the number of specialized defense factories in dozens in the 1920s, the low hundreds in the 1930s, and nearer to one thousand in the 1950s; the number of employees was more than a million before World War II, and many times that thereafter (Barber et al. 2000: 11). The political significance of the defense industry is not easy to weigh up. In the Stalin era the leaders of the defense industry tended to be men, never women, of the second or third rank. As a dictator Stalin did not tolerate rivals; no one could be as important as him. Nonetheless, he gave close attention to

52

Mark Harrison & Andrei Markevich

his military advisers and their demands on industry. After his death the defense industry leaders became more important; Dmitrii Ustinov, Stalin’s youthful minister of the armament industry in the 1940s, rose to become one of a handful of aging relicts that ruled Brezhnev’s Politburo in the 1970s. This book is about the Stalin era, though; Stalin himself is never far from the center of our attention and even when off the stage he cast a long shadow across it. Our subject is how the Soviet defense industry worked, and how well it worked. It is notable that in most countries the defense industry works badly. Defense contractors are notorious for cost overruns and kickbacks. Soldiers find that under combat conditions their products misfire, stall, or sink. Every army has its slang for equipment that is ‘‘u/s’’ (unserviceable), ‘‘duff,’’ or ‘‘kaput.’’ By some standards the Soviet defense industry worked quite well; it kept up with technology and supplied the armed forces with large numbers of weapons that were ruggedly adapted to unsophisticated soldiers fighting under unsophisticated conditions. The results are particularly impressive if we remember that the Soviet Union was, relatively speaking, a poor country with limited industrial resources. The best estimates available suggest that in the late 1920s, on the eve of forced industrialization, Soviet GDP per head of the population was about onequarter of that in the United Kingdom and one-fifth of the United States’ level; in the whole of Europe only a few neighboring countries to the south east were poorer (Maddison 2000). When Stalin said, in 1931, ‘‘we are fifty or a hundred years behind the advanced countries,’’ he was thinking in round numbers that were not intended to be precise; we can see with hindsight, however, that this estimate was quite accurate (Harrison 1994: 255–56). How did a relatively poor country achieve such results? This overview chapter presents a few basics of the organization of the defense industry in the context of Soviet political and economic institutions. Other chapters deal in more detail with the organization of particular activities including planning, mobilization, labor employment, and research. The institutional setting of the defense industry had a political dimension and a market dimension. Politically the Soviet defense industry operated in a context of dictatorship. Stalin and his immediate subordinates micro-managed the resources and priorities of the industry to a high degree by issuing commands that were often called ‘‘plans’’; for this reason the Soviet economy has been called both a ‘‘planned economy’’ and a ‘‘command system.’’ The plans had legal force; willful violation of them was a criminal offense. Thus a well-known legal text of the period (Pashukanis 1935/1980: 308) declared: ‘‘The plan is the law of the Soviet state. Fulfillment of the plan is the sacred obligation of every economic agency, of every manager, of every work-

Hierarchies and Markets

53

ing person. The obligatory nature of acts of socialist planning (plan discipline) is supported by various sanctions, in particular by the threat of criminal repression.’’ Although only a small proportion of such violations reached the courts, there is no doubt of the importance of the plan in the lives of citizens in every workplace and neighborhood. The defense industry also operated in a context of markets. At first sight it is not clear how markets could fit within a framework of dictatorship and a command economy. Markets are about decentralization and choice. Shall I take this job or that, and how hard should I work? Shall I spend my money on this item or that, or should I save it? A dictator’s agent, in contrast, must carry out orders; she ought not to make choices, other than between obeying and being punished for disobedience. In this book we will distinguish between ‘‘real’’ markets that existed independently of the dictator’s will and ‘‘internal markets’’ that the dictator deliberately created in order to solve his own allocation problems. Separate chapters of this book deal with internal markets for weapons and military inventions, and the real labor market where defense factories recruited their workers. The coexistence of the command system, real markets, and internal markets presents a fascinating problem in economics.

Command and Obedience AN ARMY-LIKE ECONOMY?

From 1918 onward the Soviet Union was a one-party state. The Bolshevik (later communist) party dominated the state through its structure, which paralleled the state in each department and at every level, and through its placemen, the nomenklatura, appointed to state posts from head of state and prime minister down through territorial and functional hierarchies to management personnel in enterprises. As a ruling party the communists carried on the conspiratorial traditions of their underground revolutionary past, in particular clandestine decision-making and the blind obedience of ordinary members to higher-level committees that were poorly controlled by infrequent elective conferences. Such rules were made for dictatorship, and by 1929 Stalin had emerged as the dominant leader of the ruling party; by 1932 his authority had become nearly absolute. Thus, the party ruled the state, but Stalin ruled the party. An army is a good analogy for Stalin’s command system in the obvious sense and also in less obvious ways. The core principle of military discipline is that of vertical hierarchy: a commanding officer at a higher level passes an order down to the rank and file, who obey and carry it out. The results depend on the

54

Mark Harrison & Andrei Markevich

quality of the orders and the quality of their execution. The quality of orders depends on good strategy, intelligence, and communications: an effective command must be purposeful, well informed about strategic requirements and resources available, and passed on without being distorted by the grapevine. In the context of the Soviet economy the strategy of the leaders, the quality of their information, and the difficulty of ensuring that the detailed plans received by enterprises corresponded with the grand overall perspectives for the economy as a whole are all well-known areas of research (for example, Zaleski 1971, 1980). The quality of compliance arises because in real armies a commander’s orders, however sensible or well founded, are not always obeyed in the spirit of the command. Soldiers may obey perfunctorily or ignore the command altogether, provided no one is watching them when they are shirking their duties. Unobserved, soldiers may take the initiative to sell off army property or go into business on their own account. Officers at intervening levels of command may also turn out to be brutal, venal, lazy, or incompetent to function satisfactorily in the eyes of their own high command. An army requires a code of discipline, a system that tracks supplies and money, and a military police. Finally, the quality of orders and the quality of compliance interact. Armies operate within a framework of international and domestic laws. In theory an order may be unlawful and in that case the soldier does not have to obey it and may even be obligated not to. This contradicts one of the chief requirements of an army: in a wide range of operational circumstances its success will depend on the instinctive or even blind obedience of the men; the chain of command cannot work well if those at lower levels are continually stopping to check their commanders’ orders against a rule book. The Soviet command system was like an army that stood above society in the sense that Stalin’s orders could override the law or were the law. By the same token Stalin did not allow his agents to shelter behind legal or constitutional restrictions when he demanded that they carry out his orders. One consequence was that agents at lower levels might have to break the law to comply with a superior’s command. For example, to fulfill a production quota, managers often had to go into illegal (but ‘‘real’’) markets or bribe suppliers to obtain the necessary supplies (Belova 2001). The inability of planners to meet all needs and account for all contingencies was an important reason why real markets persisted despite the command system. In turn this created a dilemma for superior officials: when they saw their subordinates violating the law, they might find it hard to tell whether the purpose of the lawbreaking was to support the plan or subvert it.

Hierarchies and Markets

55

THE ECONOMY’S GENERAL STAFF

How was the command system organized? One aspect was vertical: principals at higher levels received information from agents at lower levels and used this to formulate their orders. This is shown in Figure 3.1, Panel (A). Another aspect was horizontal: the economy was too complex to be organized in a single firm, so it was compartmentalized into many branches of production, distribution, transport, and so on. Goods and services then had to be exchanged among the various branches, and these exchanges were coordinated from above. This is shown in Panel (B). An important feature of the horizontal relations in the command economy was that agents should engage only in those that had vertical authorization; unauthorized transactions could be defined as disloyal or corrupt and criminal. Although agents did have a general authorization to buy or sell the ‘‘non-funded’’ commodities that fell outside the limits of the central plan at their own discretion, the fact is that virtually everything that entered or left the defense industry was ‘‘funded,’’ leaving defense industry managers with little or no theoretical freedom of action. As dictator, Stalin was the principal who stood above all others. He ruled through an interlocking series of party and state committees. Most important was the party Politburo, supposedly no more than the executive agent of the party’s large central committee but actually the handpicked group with which Stalin interacted most frequently, on whom he relied to keep him informed and to implement his decisions. The most famous names of the Stalinist era were there, including Viacheslav Molotov, his prime minister and later foreign minister, who sold western Poland to Germany in 1939 in return for the Baltic region and a temporary peace; Kliment Voroshilov, an old soldier of the civil war who became Stalin’s defense minister; Sergo Ordzhonikidze, who ran industry for Stalin until his suicide in 1937; and Lazar Kaganovich, Stalin’s deputy as party leader during much of the 1930s, whose brother Mikhail was an important official of the defense industry. Later additions included Andrei Zhdanov, who ran Leningrad and its big weapon factories; Lavrentii Beriia, chief of secret police and labor camps, who oversaw the development of new military technologies from radar to the atom bomb; Georgii Malenkov, who later took over the party apparatus and picked up an important role in the development of Soviet aerospace; Nikolai Voznesenskii, who ran the war economy; and Nikita Khrushchev, who would denounce Stalin after the latter’s death. As a principal who valued vertical obedience and truth-telling, and desired his agents to engage in only horizontal transactions that he had authorized,

56

Mark Harrison & Andrei Markevich

Figure 3.1. Vertical and Horizontal Relations

Stalin disliked it when his subordinates stood up for each other and favored those that were frank with him and did not cover for others. Thus he lost faith in Ordzhonikidze when the latter tried to protect his officials from the purges, and promoted Voznesenskii, whom the others disliked as the Boss’s pet. Later, however, when Malenkov and Beriia caused Stalin to doubt Voznesenskii’s loyalty, Stalin had him shot (Khlevniuk 1993; Gorlizki and Khlevnyuk 2004). In the field of defense economics the first formal link in the ‘‘transmission belt’’ from Stalin and the Politburo to the ministries was provided by a highlevel subcommittee of the party, the government, or sometimes both jointly. The succession is shown in Figure 3.2: in the 1920s a defense commission of the Politburo, then of the Council of Labor and Defense (STO); from 1930 to 1937 a joint party-government Defense Commission, and from 1937 to World War II a government Defense Committee; while Stalin himself tended to be an ‘‘ordinary’’ member of these subcommittees, they were always chaired by someone close to him such as Molotov, Lazar Kaganovich, or Voznesenskii. In World War II everything was done directly from Stalin’s war cabinet, the State Defense Committee. After the war things became less formal but the principle remained the same: a few people stood between Stalin and the government and processed his instructions in a small group before passing them on to the ministries. THE CHAIN OF COMMAND

Stalin’s orders went via the government, called the Council of People’s Commissars from 1917 and renamed the Council of Ministers in 1946, or directly to individual commissars. The People’s Commissars were government ministers who ran not only the usual departments for home and foreign affairs, defense, finance, education, and so on but also a wide range of ministries

Hierarchies and Markets

57

Figure 3.2. Stalin’s Defense Subcommittee

for production and distribution. The production and distribution ministries were the result of a state-owned economy; these ministries exercised the ownership rights over the means of production and distribution on behalf of the state. At this general level there were no fundamental differences in the way the defense industry was owned and managed from that applied to civilian branches. The general pattern of organization of a Soviet ministry is shown in Figure 3.3. There were three main levels: at the bottom were the direct producers organized in state-owned enterprises. The enterprises were grouped in together in larger associations. In the 1920s these tended to be called ‘‘trusts’’ in the spirit

58

Mark Harrison & Andrei Markevich

Figure 3.3. Minister, Chief Administration, and Enterprise

of capitalist holding companies, and the trusts had names that conveyed the sense of what they produced—for example, the ‘‘Gun and Arsenal Trust,’’ or the ‘‘Military Chemicals Trust.’’ Under Stalin the trusts lost even the fiction of operational independence that they had had in the 1920s and tended to be replaced by ministerial departments usually called ‘‘chief administrations.’’ Finally, the heads of chief administrations reported directly to the minister, with whom they belonged to the ministerial ‘‘collegium’’ or council; the more important heads of administrations had the status of deputy ministers. In practice the ministerial organization chart could be still more or less complex than that shown. Exceptionally large or important establishments could be kept under the personal supervision of the minister, bypassing the chief administration. At the other extreme particularly small-scale or otherwise negligible outfits could be lumped together into a ‘‘trust,’’ and such trusts could be agglomerated into ‘‘associations’’ that reported to a chief administration. In practice, therefore, the number of links in the transmission belt from the minister to the firm could range from one for the biggest factories to three or four for the least in size. But the general trend of the 1930s, as we shall see, was in the direction of simpler, shallower hierarchies with fewer levels. Table 3.2 provides a snapshot of the structure of the defense industry in 1936, when the industry became important enough to have its own ministry

Hierarchies and Markets

59

for the first time. At this time it comprised 274 separate establishments organized in eleven chief administrations and trusts; later the chief administrations were numbered rather than given names. The factories were highly diverse, ranging from large-scale mass production facilities for guns and ammunition to much smaller workshops in which artisans crafted aircraft out of wood and canvas; there were also research and experimental facilities, firing ranges, training colleges, and so forth. In the 1920s the ministerial structure was relatively simple: there was a single ministry for state-owned industry, grandly called the Supreme Council of the National Economy (VSNKh). Other ministries controlled foreign trade, transport, and the channeling of food from the countryside to the towns. With rapid industrialization under the five-year plans this simple structure soon became more complex. In 1930 the food industry was carved out of VSNKh; in 1932 what remained of VSNKh was split into separate ministries for the heavy and light industries. Further subdivisions came in the second half of the 1930s. In December 1936 a separate ministry was created for the defense industry; two years later, in January 1939, the defense industry was shared out among four new ministries for the aircraft, armament, ammunition, and shipbuilding industries. These ministries did not cover the whole of specialized war production, however. A ministry of engineering, hived off from heavy industry in 1937, continued to deal with things like armored vehicles and mortar armament. In September 1941, however, new ministries were set up to specialize in tank and mortar production for the duration of the war. After the war, completely new industries were created for the new atomic, missile, and radar technologies. Long-range ballistic missiles were absorbed by the ministry of armament, but atomic weapons and radar were managed by temporary high-level government commissions until after Stalin’s death, when they were handed over to new ministries of the radiotechnical industry and medium engineering. It was after World War II that the titles of some engineering ministries ceased to mean what they said: ‘‘medium engineering’’ meant uranium and plutonium processing, nuclear power, and nuclear bombs; ‘‘transport engineering’’ meant armored vehicles; ‘‘agricultural engineering’’ included short-range missiles as well as tractors and combine harvesters. The growing complexity of the ministerial structure in the 1930s and 1940s is illustrated in Figure 3.4. Why did it get so complicated? There are two main explanations, both plausible and consistent with the evidence, so at the moment we cannot easily identify how to share the burden of explanation between them. There is a simple economic argument: the ministerial structure became more complicated to match the growing complexity of the economy. There is a

Table 3.2. The ministry of defense industry, December 1936: 274 establishments

Chief administrations and other subunits

Number of establishments

∞. Chief Administration of the Aircraft Industry (GUAP): Aircraft and engine production, research, design, and training—e.g., Factory no. ∞ (the Aviakhim Factory), the Central Aero-Hydrodynamic Institute, the Rybinsk Aviation Technical School.

∏Ω

≤. Chief Administration of War Industry (GUVP): Small-arms, artillery, torpedo, optical, and instrument production and design—e.g., the ‘‘Bolshevik’’ Factory, OKB-≤ (the Shpital’nyi experimental design bureau).

∂∂

≥. Chief Administration of Ammunition (GUB): Shell, cartridge, explosive-tubing, and mine and bomb production, design, and testing—e.g., Factory no. ∏≤, the Central Design Bureau, the Sofrino firing range.

∑∑

∂. Chief Naval Administration (Glavmorprom): Shipbuilding and equipment production, research, design, and training—e.g., the Baltic Ordzhonikidze shipyard and TsKBS-≤ (submarine design bureau) in Leningrad, and the Shipbuilding Institute and Shipbuilding Training School in Nikolaev.

≤≥

∑. Chief Administration of the Organic Chemicals Industry (Glavorgkhimprom): Chemical agents production and experimentation—e.g., Factory no. ∑∞, Central Research Laboratory no. ≤.

∞≠

∏. Chief Administration of the Nitrate Industry (Glavazot): Explosives production and research—e.g., Factory no. ∂≠, the All-Union Chemical Research Institute.

∞≥

π. Chief Administration of the Low-Power Electrical Circuit Industry (Glavesprom): Telephone, radio, television, and electronic valve production, research, and training—e.g., the Red Dawn Telephone Factory, the All-Union Television Research Institute, the Radioelectronics Technical School.

≤∫

∫. All-Union Association of Factories of the Precision Industry (VOTI): Instrument production, research, design, and training—e.g., the Aviapribor Factory for aviation instruments, the Research Institute for Automatic Equipment and Gyroscopy, the

Hierarchies and Markets

61

Table 3.2. Continued

Chief administrations and other subunits

Number of establishments

Special Design Bureau, and the Mechanics Training School in Vladimir.

∞∏

Ω. All-Union Battery Trust: Battery production and experimentation—e.g., the Vossibelement Factory in Irkutsk, the Central Battery Laboratory.

∞≤

∞≠. All-Union Trust for Special-Purpose Steels (Spetsstal’): Special-purpose steel production—e.g., the Izhorsk factory.

≥

∞∞. Central Labour Institute (TsIT)

∞

Total

≤π∂

Source: Adapted from RGAE, 7297/38/91: 3–6 (December 21, 1936).

little more to it than this. Suppose the ministerial structure had stayed the same while the economy industrialized and diversified. With no change in the ministerial structure it would have been necessary to develop new subdivisions within the existing ministries. VSNKh, for example, was set up in January 1918 with just 14 internal divisions, called ‘‘chief administrations.’’ By 1920, including ‘‘centers and sections,’’ there were 74 of them (Carr 1966: 182n). A similar process can be seen at work a decade and a half later in the ministry of heavy industry. In April 1932, four months after its first formation, this ministry had 13 branch administrations; by 1938 these had grown to 34 in number, and this was despite the loss of a number of major branches to separate ministries.∂ The minister’s problem is illustrated in Figure 3.5. In Panel (A) he must coordinate two chief administrations responsible for production, say, one for guns and one for ammunition. In Panel (B) there is increased complexity of production so two new chief administrations have been added, say, one for airframes and one for aircraft engines; aircraft also need to be fitted with guns during manufacture and supplied with ammunition for operations. The number of chief administrations has grown to four. The scope of the ministry has doubled, but the problem of coordinating its component parts has quadrupled. In Panel (C) the minister has solved the coordination problem by devolving production to four new subordinate administrations and devolving coordination to the chief administrations; one chief administration coordi-

Figure 3.4. The Ministerial Structure of the Soviet Defense Industry, 1917–53 Source: Harrison (1985: 278–79); Crowfoot and Harrison (1990, unpublished Appendix A).

Hierarchies and Markets

63

Figure 3.5. A Coordination Problem

nates guns and ammunition and the other coordinates aircraft and engines, leaving the minister to coordinate between the two. Panel (C) looks like a possible solution. The main problem is that the chain of command from the minister to the producers has lengthened. More intervening links will blur his view of those below him and increase the scope for his subordinates to ignore or disobey orders. One solution to the chain-ofcommand problem is to break up the ministry; this is shown in Panel (D). Breaking up the ministry restores short chains of command. A consistent interpretation of the process of ministerial fragmentation shown in Figure 3.3, therefore, is that it accelerated when the authorities became particularly anxious about detailed control of subordinates. Supporting evidence is found in the fact that by the late 1930s ministerial hierarchies were typically much shallower than in the early 1930s, with no more than one or at most two levels intervening between the minister and the factory in place of three or even four in the earlier years. If ministerial subdivision was a solution, however, it creates a new problem, or more strictly it shifts the old problem to a higher level (Crowfoot and Harrison 1990): who will coordinate the ministers? Growing ministerial fragmentation often gave greater weight to the coordinating bodies that stood above ministers. One response to the rapid subdivisions of the late 1930s, for

64

Mark Harrison & Andrei Markevich

example, was the expansion of the apparatus of the Sovnarkom Economic Council (Figure 3.2) in 1940 into six mini-councils, one of them for the defense industry, each charged with oversight of one of the major branches of the economy. Thus, reducing bureaucracy in the ministries tended to result in more bureaucracy above them. There is also a political explanation of why the Soviet economy tended to see a growing number of smaller, more specialized ministries. This trend was not confined to the defense sector, although it may be argued that the defense sector saw it first. When the Soviet Union was constituted in 1923 there were 10 government ministers. By 1936 the number had risen to 18. Ministerial fragmentation pushed the number to 43 by 1941 and 59 during 1946 (Zaleski 1980: 20). The creation of new ministries had a multiplier effect on the number of middle-level posts, since every ministry required its complement of deputy ministers, heads of chief administrations, deputy heads, chief assistants, and so on. This was Stalin’s ‘‘other’’ job creation program: ‘‘I need not mention,’’ he reported in 1939, ‘‘that the division of organizations has made it possible to promote hundreds and thousands of new people to leading posts’’ (Stalin 1940: 650). On this interpretation Stalin bought the loyalty of his subordinates by offering promotion in return. A problem was that the number of people whose loyalty he needed tended to exceed the number of existing posts available or becoming vacant. He solved this further problem by frequent purges combined with ministerial subdivisions that created new leading positions in great numbers. But purges and reorganizations were also costly. Valery Lazarev (2005) suggests that after Stalin’s death the promotion mechanism tended to become unviable and threatened the command system with bankruptcy. Finally, Stalin may have had a more cynical motive. Perhaps a command system with a few powerful ministers ceased to suit him, to the extent that it enabled those at the next level below him to accumulate too much power. Perhaps he considered that it made it too easy for a rival to emerge as a potential successor. In the early 1930s, for example, one man ran most of Soviet industry: Sergo Ordzhonikidze. As industry grew, so did his influence. Increasing the number of chief administrations under him simply added to the layers of middle-ranking officials who owed their positions and allegiance directly to Ordzhonikidze. In contrast, increasing the number of other ministers diluted Ordzhonikidze’s influence, widened the circle of those who owed their loyalty to Stalin himself, and reduced the chances that another Ordzhonikidze would ever emerge. It took more than this to solve the problem of Ordzhonikidze himself, however: Stalin not only broke up his empire and destroyed those loyal to him, but also drove Ordzhonikidze to suicide (Khlevniuk 1993).

Hierarchies and Markets

65

The Internal Market for Military Goods Markets existed under Stalin’s dictatorship for good reason. The markets were, strictly, of two kinds: real markets and internal or quasi-markets. Quasi, from Latin, means, ‘‘as if it were’’; a quasi-market is market-like yet lacks some of the essential properties of real markets. The differences are summarized in Table 3.3. Real markets were formed by buyers and sellers who were independent of each other. When agents of the state or state-owned enterprises bought or sold in real markets, for example, the other party was usually a private agent or at least a private intermediary. They entered the market independently because it was in their own self-interest to do so. Prices were set by interpersonal negotiation, impersonal bidding, or preset in the presence of market power. The interaction of supply and demand led to one of several possible equilibria, depending on the institutional arrangements in the market (Morishima 1984: 13–31). Contract disputes were resolved by custom or law. The market steered resources in the general direction of their most profitable use. Internal markets, in contrast, were created by the state to allow its own agents to engage in decentralized transactions with each other. The agents entered the internal markets because they were told to. They were not supposed to behave in an independently self-interested way but to follow contingent rules. If they found themselves in dispute, the principal decided whether or not to intervene and whom to uphold. Prices and incentives in internal markets were formed by the principal’s decision, not by an equilibrating process. The principal usually calibrated incentives in such a way as to make his agents bring about a decentralized allocation of resources on lines already laid down in broad outline by centralized plans, only in more detail. The terms ‘‘quasi-market’’ and ‘‘internal market’’ are widely used in the modern economic analysis of resource allocation within large private and public-sector organizations. Our present usage of the term, however, owes more to Ludwig von Mises, who developed the idea of quasi-markets to describe what contemporary socialists like Oskar Lange wanted when they advocated market-like rules for bureaucratic socialist allocation. Mises regarded the idea of market socialism as inherently unworkable, likening it to a ‘‘triangular square.’’ He argued that because the agents would have no property of their own to lose, quasi-markets would be dominated by the ‘‘audacity, carelessness, and unreasonable optimism’’ of ‘‘the least scrupulous visionaries or scoundrels’’ (Mises 1949/1998: 705). We shall find that Mises had an element of prophetic truth on his side. At the same time the reality of internal markets in the Soviet defense sector was much more complex and interesting than this

66

Mark Harrison & Andrei Markevich

Table 3.3. Real markets and internal markets Real market

Internal market

Who forms the market?

Buyers and sellers.

Principal.

Who are buyer and seller?

Independent agents.

Both subordinate agents of same principal.

What motivates them?

Self-interest.

Principal’s command.

Who enforces exchange?

Custom or law.

Principal.

What forms prices?

Buyer and seller interact: supply and demand reach an equilibrium.

Principal’s decree tries to align agents’ self-interest with principal’s by trial and error.

What is role of market?

Allocate resources to most profitable use.

Implement principal’s allocation plan in detail.

would suggest, and the internal markets that Stalin created sometimes turned out to have a life of their own. If we ask how markets came to be nested within the command system as a matter of historical fact, the answer conforms fairly well to the real versus internal market distinction. Internal markets were created deliberately within the command system, whereas real markets persisted and were tolerated or eventually recognized in spite of it. The Soviet command system was formed in two episodes, the Russian revolution and civil war of 1917 to 1921 and Stalin’s ‘‘revolution from above’’ between 1929 and roughly 1934 (Davies 1994). The historical experience of these years shows us that the command system did not succeed in driving real markets out because it made too many mistakes and left too many needs unmet in the lower levels of society. Real markets persisted in the margins of the planned economy because, while carrying out their planned assignments, workers still had to go to the market to find bread and firewood, and managers still had to trade to get the fuel and materials they needed to fulfill the plan. It is less obvious, at first sight, why much of the allocation of defense goods was delegated to internal markets. It is easy to see why Stalin’s Politburo, meeting in the paneled offices of the Kremlin, should not know or care much about the provisioning of every household in every faraway provincial settlement. But the Red Army and the defense industry that supplied it were much more important to them than that. Stalin himself was obsessed with guns and

Hierarchies and Markets

67

Figure 3.6. The Defense Ministry, 1923–53 Source: Ivkin (1999: 170–72).

their technical detail. Despite this, much of the real story of defense allocation can be summed up as follows: each year Stalin gave the Army a bag of rubles and told it to buy the guns it needed from Industry in an internal market. That is why this book is called Guns and Rubles. In Chapter 2 Andrei Sokolov introduced the two sides of the internal market for weapons as the Army and Industry. He noted that each is shorthand for a somewhat complicated and imprecise empirical counterpart. The Army stands for the defense ministry, but the latter term itself embraces a complex sequence of official designations, shown in Figure 3.6, for the ministerial departments that governed the Red Army and Navy. In this book we shall refer to all these as the ‘‘defense ministry’’ unless the context demands greater preci-

68

Mark Harrison & Andrei Markevich

Figure 3.7. The Soviet Internal Market for Weapons

sion. Notably, Stalin himself occupied the post of defense minister from July 1941 to March 1947. The demand side of the internal market for weapons was constituted by the defense ministry’s purchasing staff, charged with fulfillment of the annual plan of orders for weapons and military equipment. These held the purse of rubles that Stalin had set aside for them. On the supply side, Industry represents the enterprises of the defense industry, organized under a ministry or, by the late 1930s, several ministries (shown in Figure 3.4). They also had a plan to fulfill, but this was a plan for the overall value of gross output. In theory these two plans were supposed to be coordinated, but perfect coordination was impossible in practice, and even limited reconciliation of military and economic plans proved to be costly and time consuming. The result was an internal market, shown in Figure 3.7. When guns met rubles the result was not a real market of the sort imagined by Léon Walras (1874/1954), who pioneered modern general equilibrium theory. In the Walrasian market masses of buyers and sellers interacted through an auctioneer who fine-tuned the prices of commodities to bring supply and demand into balance everywhere at once. In the Soviet internal market for weapons, hundreds of bilateral bargains took place in a chaotic sequence, one after another, as military officers visited factory managers and ministerial overseers to persuade Industry to produce what the Army wanted. In reaching a bargain each side had strengths and weaknesses. The main

Hierarchies and Markets

69

advantage of Industry was its market position. Most major military items were produced in a handful of industrial enterprises, and there was little substitutability among them. On a detailed product classification most producers were effectively monopolists. The context, moreover, was not one of market equilibrium but of a shortage economy of generalized excess demand. This gave industrial producers the whip hand in negotiating terms with most buyers. The strength of the Army was that it was not just any buyer. In the internal market for military goods it was almost entirely a monopsonist. It is true that in some periods the army and navy were organized under separate ministries with needs for armament and propulsive power that overlapped to some extent. It is also true that the NKVD, with its own internal security forces, was a third purchaser in the market, operating on a very small scale in comparison with the others. But there is no evidence in the documents that these three ever competed in the sense of bidding against each other for scarce goods or services. Thus, the relationship between buyer and seller in the internal market for military goods was almost always one of bilateral monopoly. The Soviet Union’s isolated position in the interwar global arena gave the Army very real strategic importance and ensured that Stalin would pay close attention to its needs. This was a source of both strength and weakness. Its strategic importance gave the Army the strength of high priority in the command economy; when funding or supplies were generally short, the Army would still get a large part of what it required. The weakness was that it made the Army a target for industrial suppliers who hoped to bargain away the Army’s resources in their own favor. For example, its decisive role in the country’s external security meant that the Army could not ultimately refuse to buy any weapons and end up without; it had to come to terms with Industry, and Industry worked to make these terms as advantageous as possible. The spirit of the resulting bargain can be illustrated with the help of Figure 3.8. The Army’s budget for weapons is a given purse of V rubles. Given perfect coordination of the plans of Army and Industry, the gross output plan of the defense industry measured in rubles would also be equal to V. The Army expects to purchase a quantity of weapons Q* at the planned average unit price P* that would just use up its budget. Industry’s problem is to take the Army’s purse as easily as possible. To produce Q* costs a certain level of effort; Industry would gain if it could induce the Army to settle for fewer weapons at a higher price, say Q% [ Q* and P% = V/Q%. So Industry is tempted to push up the prices of weapons relative to the effort cost of producing them. For existing weapons, it might be able to do this by reporting costs that are inflated relative to the true effort of producing them, or by introducing new weapons and reporting higher costs relative to effort. A further possibility is that Industry

70

Mark Harrison & Andrei Markevich

Figure 3.8. Supply and Demand for Weapons Note: For further reasoning underlying this figure see Gregory (2003b) and Harrison and Kim (2006).

could reduce effort by lowering the quality of output, since a quality reduction is equivalent to a price increase; Chapter 6 deals with this subject in detail. In other words, from Industry’s point of view the Army’s budget is a curve that happens to pass through the P* 0 Q* combination but, if Industry can report higher costs, and so move northwest along the curve to a more profitable combination than P* 0 Q*, it will do so. This vulnerability to exploitation by Industry left the Army in a weak position. There were two counterweights. First, the Army, like other consumers, found limited protection in the state arbitration courts that monitored the contracting process and tended to favor the buyer in their judgments (Kroll 1986, 1988). Second, the Army benefited from a unique system of direct monitoring of Industry that no other buyer was able to establish. It maintained a network of serving officers who acted as full-time representatives of the Army’s purchasing departments in the factories supplying defense. This limited the freedom of Industry to exploit the Army, but we will show that the defense suppliers had yet more cards up their sleeves in this asymmetric game. Why did Stalin put up with the rivalries and tensions that permeated the internal market? There seem to have been two reasons. The first reason is that however much he enjoyed the technical detail of weaponry, he could not carry out all the detailed allocation himself whether personally, or in his Politburo, or through his planners; the planners themselves did not want responsibility

Hierarchies and Markets

71

for this kind of work in case they ended up being blamed for the inevitable mistakes and failures (Belova and Gregory 2002). Even if he did not want to do it himself, however, Stalin could have made the Army do it. The Army knew what it wanted; why not make the Army responsible for the defense industry? In terms of modern economics, if it was costly to transmit the information required to balance supply and demand across the market place, why not replace the internal market with a single vertically integrated firm? Here the second reason comes into play: Stalin probably did not want to encourage the emergence of a powerful, integrated militaryindustrial complex that could challenge his power (Harrison 2003). The competitive rivalries and tensions between Army and Industry suited him; he preferred to divide and rule.

Third-Party Regulation Divide-and-rule is the stratagem of a ruler who mistrusts those around him, but it works by spreading mistrust widely through the institutions of governance. The functioning of the Soviet command economy reflected a lowtrust environment in which everyone had to be told what to do because those above them did not trust them to do it without being told. Moreover, having told their subordinates what to do, government officials were rarely confident that it had been done. In fact, one of the key problems of the Soviet governance system turned out to be that everyone was willing to put some effort into pretending to do as they were told when they were really doing something else. One result was that officials at higher levels were rarely satisfied with their own observance of those below them, and resorted to third-party regulation to supplement their own monitoring systems. The defense industry was subject to regulation by at least four types of external agency that reported directly to the government or the Politburo. These agencies dealt with planning and production accounting, contract arbitration, ‘‘control’’ or audit, and state security. In traditional accounts of the Soviet command system the role of the planners has been much misunderstood. Rather than being all-powerful goalsetters and distributors of resources, they largely did as they were told. It was the political leaders who set the goals and distributed the resources, while the planners’ task was to regulate and track the consequences. In doing this the planners, if anything, did as they were told more faithfully than most (Belova and Gregory 2002). One of their most important obligations was to tell the Politburo the truth about the state of the economy. To do this, side by side with their planning functions, they gathered information about the economy’s

72

Mark Harrison & Andrei Markevich

stocks and flows. It is true that this information came largely from the producers, who themselves could not be trusted to tell the truth. If we express this problem in the terms of Figure 3.8, it was part of the planners’ job to hold prices to P* so that the producers would have to fulfill their plan with Q* and not some lower level of output at a higher price. The planners did not succeed perfectly at this, but they did put limits on producers’ discretion to put up prices and reduce effort or divert resources. Centralized plans were fairly highly aggregated and did not prescribe the transactions between the Army and Industry in detail. The planners were not interested in regulating the countless petty disputes that arose over the detail itself. The detail of bilateral bargaining was handled through decentralized contracting procedures that were subject to oversight by the arbitration courts. According to Heidi Kroll (1986, 1988) the working of the decentralized contract system is best understood in terms of the need to economize on centralized planning costs, while the arbitration courts served to protect the buyer’s ‘‘right to be served’’ in the context of seller’s markets. Given the importance of the defense industry and the secrecy surrounding it, contract disputes with regard to military items were not handled in the local arbitration courts. In the early 1930s they were centralized in a special joint arbitration commission of the armed forces and the defense industry under the ministry of the ‘‘workers’ and peasants’ inspection,’’ a joint agency of the state and party. This ministry was dissolved in 1934 and the centralized resolution of contract disputes fell to the Soviet Union’s chief arbitrator. At first he was committed to handle all disputes ‘‘regardless of value’’; overwhelmed by a flood of trivial claims, he soon imposed a minimum value of 25,000 rubles on the claims that he would consider subsequently.∑ The institutions of party and state ‘‘control’’ or audit carried out detailed investigations of enterprise management and local government more generally in order to detect the flouting of instructions, corruption, and other abuses. In the 1930s the control agencies comprised the ministry of the workers’ and peasants’ inspection and subsequently, from 1934, separate commissions for party and state control, the latter becoming a ministry of state control in 1940 (Ivkin 1999: 181–83). Both agencies had special representatives in every region whose remit included looking into wrongdoing in industry; some appointed special investigators to their staff for defense matters.∏ At the end of 1937, however, the defense groups in both agencies were abolished, their powers passing to a new chief military inspectorate of the Defense Committee; the new chief inspector was granted an establishment of one secretary and nineteen subordinate inspectors for different branches of weaponry.π In September 1940 the state control commission became a ministry once

Hierarchies and Markets

73

more and absorbed the Defense Committee’s chief military inspectorate. Inspection of the defense industry became a much bigger and more specialized business. The new ministry organized inspection groups for each branch of the defense industry such as aviation, shipbuilding, and so forth.∫ The most coercive form of third-party regulation involved intervention by the state security police: in the 1920s and early 1930s the unified chief political administration (OGPU), from 1934 the chief administration of state security of the interior ministry (NKVD, from 1946 the MVD), and from 1943 the ministry of state security (NKGB, from 1946 the MGB) (Ivkin 1999: 176–77). Further oversight of the economy was also provided by the economic administration of the NKVD-MVD (Petrov and Skorkin 1999: 21, 23). The security police could investigate, arrest, and punish cases of wrongdoing that involved the suspicion of counter-revolutionary motivation or association. Such suspicions were easily aroused in anything involving military matters, specialized knowledge, or foreign technical collaboration from the time of the Shakhty affair in 1928. In Stalin’s Russia the security police acquired a close historical association with the high-technology branches of the defense industry. It seems likely that this was for several reasons. First, these were the branches in which the Soviet Union lagged furthest behind its rivals; catching up was most urgent and required the greatest leaps of understanding and mastery of technique. Second, these branches rested on advanced scientific foundations in which the relatively ill-educated political leaders were at the greatest disadvantage visà-vis the specialists. Third, because the science was relatively new, the Soviet Union had little choice but to select civilian specialists for their knowledge rather than their loyalty. The knowledge for which they were selected was crucially of scientific and technical developments abroad; this immediately made them potentially unreliable in an era when even ‘‘unconscious disloyalty’’ could be a crime (see Chapter 1). In short the progress of the defense industries was seen as vitally important, rested on science that politicians found difficult to evaluate, and relied heavily on civilian specialists selected on criteria that made their allegiance doubtful. These naturally combined to attract the attention of the security services. The tendency toward secret-police involvement in regulating militaryrelated research is symbolized by the emergence of a strange and unprecedented phenomenon: the sharaga, or prison design bureau. In the late 1920s the relationship between the industrial scientists and the regime underwent a crisis expressed in the leveling of wild accusations and criminal charges of economic sabotage and espionage against many engineers and economic specialists. First aircraft designers and then others were arrested and put to work

74

Mark Harrison & Andrei Markevich

on military projects as prisoners (Simonov 1996; Starkov 2000); the following information is drawn from work by Mikhail Morukov (2004). By the early 1930s the OGPU was employing more than four hundred imprisoned specialists in design bureaus located in Moscow, Leningrad, Khar’kov, and western Siberia. Of the total number on August 30, 1931, nearly half were working on defense projects ranging from new armored vehicles to submarines, artillery, explosives, and defenses against chemical agents. A second wave of arrests in 1937 and 1938 was associated with the Great Terror. Lavrentii Beriia, who took over the NKVD in the wake of the terror, set up a unified ‘‘special technical bureau’’ of the NKVD, answering directly to him, with 316 specialists organized in no less than eight subgroups for various aspects of aviation, aircraft engines, shipbuilding, propellants, explosives, armor steel, chemical agents, and chemical defense. During the war Beriia gained personal responsibility for the Soviet Union’s nuclear weapon project (Holloway 1994; Simonov 1996). Finally, in October 1946 the NKVD added thousands of deported German military designers to its roster of imprisoned scientific personnel (Harrison 2000). One reason that has been suggested for growing secret-police involvement in the defense industry is that the NKVD had a vested interest in monopolizing the development of military technologies. If so, this would have serious implications for our understanding of Stalin’s dictatorship; it would mean that the NKVD became independently powerful of Stalin, who was therefore not such an effective dictator after all. The evidence in favor of this idea is that the responsibilities of the NKVD grew remarkably through time. The evidence against it is that the NKVD seems to have been more interested in getting positive research results than in expanding its control of research assets. For example, engineers who achieved their assignments were often rewarded with early release. Although it does not resolve the issue, it may help the reader to understand that a similar debate has unfolded around the NKVD and its control of forced labor. The forced labor system grew under Stalin, and the NKVD controlled increasing numbers of forced laborers. Does this mean that the goal of the NKVD was to build an economic empire? Possibly. But there is also evidence that the NKVD planned to scale down forced labor more often than to expand it, that the arrests that recruited convicts for the labor camps were driven by Stalin more than by the NKVD itself, and that by the time of Stalin’s death the NKVD (by that time MVD) leadership was convinced that widespread forced labor was damaging the Soviet economic and political system and social order (Gregory 2003b). Although the jury is still out on this issue, the role that the OGPU-NKVD-

Hierarchies and Markets

75

MVD will play in this book is to a large extent that of Stalin’s loyal servant. The security police generally seems to have acted on Stalin’s behalf and in his interest rather than following its own private agenda. To conclude, in the institutional setting of the command economy Soviet defense factories had to satisfy many masters. They had to keep happy their own ministerial bosses, the defense ministry as their main purchaser, and an array of third-party regulators who reported directly to the Boss: to Stalin himself. This apparatus was costly for the country to maintain. Was it worth it? One aim of this book is to evaluate the resources and mechanisms of the Soviet defense industry from various angles in relation to its goals. It is for the reader to evaluate our success. The most important outcome of this chapter is to establish the concept of the internal market for weapons. When Industry produced guns for the Army, it did not do so by order, as the notion of a ‘‘command economy’’ might suggest. No matter how coherent and detailed was the procurement plan, it still had to be translated into contracts. The contracting process that exchanged guns for rubles was market-like. Being market-like is not the same as being a real market, and we address this through the concept of an internal or quasi-market. Well established in the regulation literature, this concept helps us to understand how a process that looks as if it should have been centralized to the extreme turned out to be dominated by informality and decentralization in its implementation. Notes 1. GARF, 8418/27/238: 27–36 (1938). 2. GARF, 8418/23/345: 145 (Kaganovich to the Defence Committee, January 10, 1939). 3. For the 1938 figure, covering the planned output of the ministry of the defense industry, see GARF, 8418/22/463: 5–7 (Defence Committee, February 20, 1938). In 1933 civilian items accounted for 36 percent of the output of the military factories under the ministry for heavy industry: GARF, 8418/8/9: 3–6 (Mezhlauk to Molotov, July 31, 1933). 4. 1934: RGAE, 7297/44/1: 116–117. 1938: RGAE, 7297/28/36: 1–2. 5. GARF, 5446/15a/1101: 1 (March 18, 1934); 5446/18a/893: 1 (December 16, 1936). 6. In 1937 the state control commission (KSK) special representatives had positions for twenty-three such defense investigators, but only seven were filled (GARF, 8418/12/402: 101). 7. GARF, 8418/12/402: 6, 13, 14 (December 7, 1937). 8. GARF, 8300/4/1: 1 (October 26, 1940).

76

Mark Harrison & Andrei Markevich Published References

Barber, John, Mark Harrison, Nikolai Simonov, and Boris Starkov. 2000. ‘‘The Structure and Development of the Soviet Defence-Industry Complex.’’ In The Soviet DefenceIndustry Complex from Stalin to Khrushchev, 3–32. Ed. John Barber and Mark Harrison. London and Basingstoke: Macmillan. Belova, Eugenia. 2001. ‘‘Economic Crime and Punishment.’’ In Behind the Façade of Stalin’s Command Economy: Evidence from the State and Party Archives, ed. Paul R. Gregory, 131–58. Stanford, CA: Hoover Institution. Belova, Eugenia, and Paul R. Gregory. 2002. ‘‘Dictator, Loyal and Opportunistic Agents: The Soviet Archives on Creating the Soviet Economic System.’’ Public Choice 113(3– 4): 265–86. Bystrova, Irina. 2000. Voenno-promyshlennyi kompleks SSSR v gody kholodnoi voiny. (Vtoraia polovina 40-kh-nachalo 60-kh godov). Moscow: Institut Rossiiskoi istorii Rossiiskoi Akademii nauk. Carr, E. H. 1966. A History of Soviet Russia, vol. 1: The Bolshevik Revolution, 1917– 1923, part 2. Harmondsworth: Pelican. Crowfoot, John, and Mark Harrison. 1990. ‘‘The USSR Council of Ministers Under Late Stalinism, 1945–1954: Its Production Branch Composition and the Requirements of National Economy and Policy.’’ Soviet Studies 42(1): 39–58. Davies, R. W. 1994. ‘‘Changing Economic Systems: An Overview.’’ In The Economic Transformation of the Soviet Union, 1913–1945, ed. R. W. Davies, Mark Harrison, and S. G. Wheatcroft, 1–23. Cambridge: Cambridge University Press. Davies, R. W., and Mark Harrison. 1997. ‘‘The Soviet Military-Economic Effort Under the Second Five-Year Plan (1933–1937).’’ Europe-Asia Studies 49(3): 369–406. Gorlizki, Yoram, and Oleg Khlevnyuk [Khlevniuk]. 2004. Cold Peace: Stalin and the Soviet Ruling Circle, 1945–1953. New York: Oxford University Press. Gregory, Paul R. 2003a. ‘‘An Introduction to the Economics of the Gulag.’’ In The Economics of Forced Labor: The Soviet Gulag, ed. Paul R. Gregory and Valery Lazarev, 1–21. Stanford, CA: Hoover Institution. Gregory, Paul R. 2003b. ‘‘Soviet Defence Puzzles: Archives, Strategy, and Underfulfillment.’’ Europe-Asia Studies 55(6): 923–38. Harrison, Mark. 1985. Soviet Planning in Peace and War, 1938–1945. Cambridge: Cambridge University Press. Harrison, Mark. 1994. ‘‘GDPs of the USSR and Eastern Europe: Towards an Interwar Comparison.’’ Europe-Asia Studies 46(2): 243–59. Harrison, Mark. 1996. Accounting for War: Soviet Production, Employment, and the Defence Burden, 1940–1945. Cambridge: Cambridge University Press. Harrison, Mark. 2000. ‘‘New Postwar Branches (1): Rocketry.’’ In The Soviet DefenceIndustry Complex from Stalin to Khrushchev, ed. John Barber and Mark Harrison, 118–49. Basingstoke (England): Macmillan. Harrison, Mark. 2003. ‘‘Soviet Industry and the Red Army Under Stalin: A MilitaryIndustrial Complex?’’ Les Cahiers du Monde russe 44(2–3): 323–42. Harrison, Mark, and Byung-Yeon Kim. 2006. ‘‘Plans, Prices, and Corruption: The Soviet Firm Under Partial Centralization, 1930 to 1990.’’ Journal of Economic History 66(1): 1–41.

Hierarchies and Markets

77

Holloway, David. 1994. Stalin and the Bomb: The Soviet Union and Atomic Energy, 1939–1956. New Haven, CT: Yale University Press. Ivkin, V. I., ed. 1999. Gosudarstvennaia vlast’ SSSR. Vysshie organy vlasti i upravleniia i ikh rukovoditeli. 1923–1991 gg. Istoriko-biograficheskii spravochnik. Moscow: Rosspen. Khlevniuk, Oleg V. 1993. Stalin i Ordzhonikidze: konflikty v Politburo v 30-e gody. Moscow: Rossiia molodaia. Kostyrchenko, G. V. 1992. ‘‘Organizatsiia aviatsionnogo krupnoseriinogo proizvodstva.’’ In Samoletostroenie v SSSR (1917–1945), ed. G. S. Biushgens, 413–36. Moscow: TsAGI. Kroll, Heidi. 1986. ‘‘Decentralization and Precontract Disputes in Soviet Industry.’’ Soviet Economy 2(1): 51–71. Kroll, Heidi. 1988. ‘‘The Role of Contracts in the Soviet Economy.’’ Soviet Studies 40(3): 349–66. Lazarev, Valery. 2005. ‘‘Economics of the One-Party State: Promotion Incentives and Support for the Soviet Regime.’’ Comparative Economic Studies 47(2): 346–63. Maddison, Angus. 2000. The World Economy: A Millennial Perspective. Paris: OECD. Markevich, Andrei. 2004. ‘‘Planning the Soviet Defense Industry: The Late 1920s and 1930s.’’ PERSA Working Paper no. 37. University of Warwick, Department of Economics. http://www.warwick.ac.uk/go/persa. Mises, Ludwig von. 1949/1998. Human Action: A Treatise on Economics. Auburn, AL: Ludwig von Mises Institute. Morishima, Michio. 1984. The Economics of Industrial Society. Cambridge: Cambridge University Press. Morukov, Mikhail Iu. 2004. ‘‘Deiatel’nost’ khoziaistvennykh podrazdelenii OGPUNKVD i oborona SSSR v 1929–1945 gg.’’ Candidate of Historical Sciences Dissertation. Moscow. Pashukanis, E. B. 1935/1980. ‘‘Course on Soviet Economic Law.’’ In Pashukanis: Selected Writings on Marxism and Law, ed. Piers Beirne, Peter B. Maggs (trans.), and Robert Sharlet, 302–45. London and New York: Academic Press. Petrov, N. V., and K. V. Skorkin. 1999. Kto rukovodil NKVD. 1934–1941. Spravochnik. Moscow: Zven’ia. Simonov, N. S. 1996. Voenno-promyshlennyi kompleks SSSR v 1920–1950-e gody: tempy ekonomicheskogo rosta, struktura, organizatsiia proizvodstva i upravlenie. Moscow: Rosspen. Stalin, J. V. (1940). Leninism. London: Lawrence and Wishart. Starkov, Boris. 2000. ‘‘The Security Organs and the Defence-Industry Complex.’’ In The Soviet Defence-Industry Complex from Stalin to Khrushchev, ed. John Barber and Mark Harrison, 246–68. Basingstoke (England): Macmillan. Walras, Léon. 1874/1954. Elements of Pure Economics, or the Theory of Social Wealth. London: Allen and Unwin. Zaleski, Eugène. 1971. Planning for Economic Growth in the Soviet Union, 1918–1932. Chapel Hill: University of North Carolina Press. Zaleski, Eugène. 1980. Stalinist Planning for Economic Growth, 1933–1952. Chapel Hill: University of North Carolina Press.

4

Planning the Supply of Weapons: The 1930s andrei markevich

In the twentieth century the waging of war rested on economic foundations. Armies in battle required weapons, ammunition, uniforms, and food rations. Planning for war, whether for attack or defense, was inconceivable without accounting for resources. This chapter reviews the Soviet way of planning the economic aspect of warfare that was laid down at the end of the 1920s and the 1930s following the abandonment of a mixed economy and the transition to forced industrialization. As a rule, scholars have tended to analyze the working arrangements of the Soviet civil and military agencies separately. The recent opening of the archives has made it possible to begin to see civilian-military interactions and transactions more clearly. Nikolai Simonov (1996a,b), David Stone (2000), Lennart Samuelson (2000a,b, 2001), and Oleg Ken (2002) have each looked at various ways in which Soviet military thinking influenced civilian preparations for war. Despite their efforts we still know relatively little about the day-to-day planning of the supply of defense with weapons and military equipment, general-purpose fuels and machinery, and transport and medical services, and about the planning of mobilization preparedness. In this chapter I first define the military input into economic planning. This came about in two ways: immediate military demands influenced economic plans for the distribution of current output, while long-term military require-

78

Planning the Supply of Weapons

79

ments were reflected in the distribution of investment resources. I also describe the institutional hierarchies through which military requirements were brought into contact with the command economy’s civilian allocation and management systems. Military requirements had a general impact on the economy as a whole, as well as specific implications for the defense industry. I go on to look at the problem of administrative reconciliation of planned supplies with planned demands. At the stage of implementation ministerial plans had to be translated in detailed contracts for specific enterprises to deliver specific goods to the military purchaser at agreed prices, and this process was decentralized to a considerable extent. In the course of implementation plans were vulnerable to amendment from above and below. This gave rise to wide scope for opportunistic behavior on both sides—for the military to put pressure on the industrial producers, and for the producers to escape from the pressure.

The Army and Economic Planning In the Soviet Union as elsewhere, the military prepared for war on the basis of alternative scenarios for the future evolution of events. Starting from the current political and international situation, the defense ministry and Red Army general staff assessed the most likely external threats and the military requirements that these posed. This informed their plans for military modernization, against which they measured the country’s military and economic potential. As Samuelson (2001: 77) and Andrei Sokolov (in Chapter 2) have shown, this led Tukhachevskii and other military leaders to press for greater military control over economic planning and the management of industry. In principle the economic issues that concerned the military may be divided into mobilization planning and operational planning. R. W. Davies examines mobilization planning in detail in the next chapter; here we will touch on it only to define it and establish its context. Mobilization plans prepared the economy for the contingency of a war that might or might not happen and set out the economic tasks that should be undertaken in that event but not otherwise. At this time most Soviet military experts imagined that the war for which they were planning would be protracted over several years; consequently they gave most attention to computing annual average resources and requirements. Thus, the activity of mobilization planning in the broadest sense went on at several levels, including computation of the military resources that would meet the ‘‘mobilization requirements for a year of waging war’’; economic plans for the ‘‘initial period of war,’’ usually covering three months; plans for the ‘‘initial year of war’’ when the country’s economy would not yet have been mobilized to the full; and also ‘‘mobilization plans’’ in the narrow sense of

80

Andrei Markevich

prescribing procedures for a transition to war production that utilized productive capacities as fully as possible. Operational planning, in contrast, was oriented on the immediate future. The most important economic components of military planning for the immediate future were proposals for national economic development that took into account the needs of defense, and plans of ‘‘current military orders’’ to procure weapons and military equipment, closely linked with economic plans for defense industry. The issues raised by operational and mobilization planning were clearly inseparable; some contemporaries regarded operational planning as no more than a component part of mobilization planning more generally.∞ Mobilization planning aimed to prepare the economy for a specific contingency, that of war, and to ensure a seamless transition to wartime production if that contingency was realized. Operational planning aimed to optimize the allocation of current resources for all contingencies, including that of war, and specifically the resources of the defense industry for the army’s peacetime needs. The resource constraints of a poor country posed cruel dilemmas. What was the cheapest way of supplying defense requirements? The Soviet authorities believed that they were faced with a choice between two extremes. At one extreme lay a peacetime defense industry of specialized ‘‘cadre’’ factories mostly or entirely devoted to military products. At the other lay what they saw as the ‘‘American’’ style of industrial mobilization based on the creation of dual-purpose capacities that could be switched from peacetime to wartime priorities at short notice. In the short run the first variant looked cheaper because specialized war production plants offered lower capital costs than factories with built-in flexibility. In the long term, however, the second variant looked cheaper because it would provide more civilian production as long as peace was maintained; meanwhile, specialized armament factories would be less fully utilized and risked rapid obsolescence. The military were more inclined to the first, and the civilian planners to the second. But it was the political leadership that had the decisive voice. The actual path that was chosen wandered between the two extremes, depending on the international context and fears about the planning horizon for war. As David Stone (2000) has shown, in the mid-1920s the Soviet leaders became extremely anxious about the country’s lack of preparation for another war. The ‘‘war scare’’ of 1927 showed them that war requirements were hugely in excess of the defense industry’s existing capabilities (Simonov 1996a). Under the first fiveyear plan the main stress fell on building specialized ‘‘cadre’’ factories so as to close the gap as quickly as possible. Then, during the 1930s, some of the emphasis switched to building up the potential of the strategic branches of

Planning the Supply of Weapons

81

civilian industry such as engineering, vehicles, aviation, and chemicals for wartime mobilization. Under the third five-year plan, however, mobilization requirements again rose greatly above the mobilization potential of civilian industry, and from 1938 to 1940 the Soviet leadership returned to the advantages of building up ‘‘cadre’’ factories (Samuel’son 2001: 144, 196, and 232). The links between mobilization and operational planning are illustrated in Figure 4.1. The military plan of war began from military thinking about the likely character of a future conflict. The main economic component of this plan was its mobilization requirement for resources in the event of war, which underpinned the mobilization plan. Analysis of the mobilization requirement relative to the resources that already existed showed up the likely bottlenecks in the specialized defense industry and the strategic civilian branches, and signaled how investment should be allocated now so as to eliminate these bottlenecks in the future. Thus mobilization planning fed into the investment aspect of the national economic development plan. This was a somewhat onesided interaction: military specialists expected to have significant influence over the civilians responsible for planning the defense industry and the economy as a whole, but were reluctant to grant civilians any access to military planning (Samuel’son 2001: 14). For present purposes the national economic development plan had two aspects: the investment plan and the plan for current output. Investment in defense industry and the other strategic branches was channeled through the specialized industrial ministries and chief administrations’ capital construction plans. But the current needs of the armed forces for combat readiness, operations, maintenance, and training were supplied from the economy’s current output. The latter came under the national economic plan in the form of targets for the gross value of output, disaggregated among the specialized ministries (for a new account of these arrangements in civilian branches, see Markevich 2004, 2005a). Some output was stockpiled in reserves for mobilization deployment in the event of war. The army’s current peacetime requirements were embodied annually in plans of ‘‘current military orders’’ for weapons and military equipment from industry. These procurements, financed by budgetary allocations, were fixed by the political leadership at the highest level. The plan of military orders provided the basis for contracts that were then drawn up between the defense ministry and individual suppliers of weapons and equipment, and these contracts were executed in turn through deliveries by one party and payments by the other. Figure 4.2 illustrates the relation between the Army’s procurement plans and the production plans of Industry. In short, the planning of output and investment in the defense industry

Figure 4.1. Planning for Defense

Figure 4.2. Planning the Defense Industry Note: I translate dogovor as ‘‘contract’’ (or ‘‘specific contract’’), tipovoi dogovor as ‘‘model contract,’’ and osnovnye usloviia postavki as ‘‘master delivery agreement.’’

84

Andrei Markevich

proceeded in two separate channels. One was a military channel that gave rise to mobilization requirements for investment and current orders for output. The other was the civilian channel for planning the economy as a whole, with the associated plans for investment and gross output of each sector including the defense industry.

Planning Institutions and Agents The defense planning agencies emerged at the end of the 1920s as part of the hierarchical system of the Soviet command economy. As Figure 4.2 shows, there were two parallel hierarchies, one civilian and the other military. The civilian agencies were primarily responsible for planning current production, while military agencies took the lead in mobilization planning. There were also centralized coordinating institutions that mediated between the two. Within this general picture the institutional details were continually redefined through the frequent reorganizations that proved to be a hallmark of the command economy. In this section I provide some of the detail. THE CIVILIAN AND MILITARY HIERARCHIES

The Plan of War. Forward thinking about the next war was the task of the Red Army general staff and the Revolutionary Military Council with other agencies subordinate to the defense ministry.≤ These worked out mobilization requirements and presented them to the government and Politburo so that defense needs could be incorporated into national economic plans. April 1928 saw a special decision to allow military involvement in all stages of compiling national economic plans; thus, defense ministry officials reviewed all the successive drafts of the first five-year plan. The Plan of Current Military Orders. The plans of military orders emerged from the purchasing administrations of the defense ministry, and were transmitted to the government for approval. Until 1929 the Red Army had just one administration for supply, which was then subdivided into specialized administrations for artillery, war chemicals, military equipment, and so forth. From 1929 to 1937 the Red Army appointed a specialized chief of armament to oversee the purchasing administrations; from 1930 he had the rank of a deputy of the defense minister. The Plan of Gross Output of Defense Industry. This was a component part of the civilian economic planning process. Defense industry current plans of gross value of output, finance, investment, labor, and so on, as for industry generally, primarily involved civilian agencies; the ministries, chief administrations, trusts, and enterprises producing military items included the same range of functional departments as other Soviet production-branch organizations.

Planning the Supply of Weapons

85

In addition to their current plans all Soviet economic agents had annually to work out a mobilization plan that set out what they would do in the event of war. In industry, for example, mobilization plans were compiled at all levels: for industry as a whole and then for every ministry, chief administration, and enterprise. The plans for industry and for each ministry followed a common layout with a plan for the supply of products by assortment, a plan of supporting measures, and a plan of ‘‘material-technical supply.’’ Ministerial plans included plans of measures to transfer subordinate enterprises to a war footing. Enterprise plans included a description of measures that would be implemented on the announcement of mobilization or other government instruction, a production plan for mobilization, and the preparatory measures that were necessary to implement it. Under rules promulgated in June 1927, most ministries established mobilization sections; from 1929 these began to appear in every state enterprise (Stone 2003). Defense sections in civilian agencies were staffed mainly by party members subject to security screening by the OGPU. Each had access, however, only to information directly relevant to his or her own responsibilities; ‘‘defense issues within the remit of the given department’’ could be made known only to the leadership and ‘‘a strictly limited number of responsible staff.’’≥ In the 1930s defense sections in civilian agencies, known variously as ‘‘military,’’ ‘‘special,’’ or ‘‘mobilization’’ departments, were involved not only in mobilization matters but also in the current planning of defense production to meet military orders. For this they had to work alongside the regular agencies and hierarchies for line management and the monitoring of accounts, prices, statistics, and so forth. The only exception was in the branches that supplied products solely for civilian use; there, naturally, mobilization departments had nothing else to do but plan for mobilization. In practice there was much variation in how mobilization planning and current planning were organized from one ministry to another. A summary of investigations carried out in 1938 concluded: ‘‘The mobilization agencies of ministries are designed on the most varied schemes’’; ‘‘There is no one organizational structure. [The ministry of the defense industry] has a mobilization administration subdivided into two departments and groups, [the ministry of heavy industry] has a military department subdivided into 8 branch groups, [the ministry of engineering] has a military department with 12 sectors, and [the ministry of local industry] has a special department with 3 sectors. . . . There is no approved statute for military departments.’’ The personnel of military departments at the ministerial level ranged from eleven in the ministry of state farms to seventy in the ministry of engineering. The military departments of chief administrations and enterprises showed similar variation.∂

86

Andrei Markevich

THE CENTRAL AGENCIES

While civilian and military agencies managed these matters from day to day, strategic decisions and coordination fell to the central agencies. Final responsibility, according to the Soviet constitution, lay with the Council of Ministers (Sovnarkom). Above the Sovnarkom, in practice, stood Stalin and the party Politburo. Stalin was extremely interested in the details of defense policy and reserved the right to intervene in them at any time; normally, however, he delegated management to a series of high-level defense subcommittees, the exact title and official subordination of which changed over the years (see Figure 3.2). A defense commission of the Politburo, then of the Council of Labor and Defense (STO), was succeeded in 1930 by a joint party-government Defense Commission, and from 1937 to World War II by a government Defense Committee. Finally, when war broke out in 1941 all these agencies collapsed into a single body endowed with absolutely untrammeled authority, the State Defense Committee. High-level coordination for current purposes involved a range of organizations. For the decade 1927 to 1937 the defense sector of Gosplan was chiefly responsible (Samuelson 2000a,b). The defense sector comprised both military and civilian staff and was supposed to unify the efforts of the defense ministry and general staff with those of Gosplan, the country’s lead organization for economic planning. The remit of the defense sector was to compile an economywide plan for the event of war, coordinate the mobilization plans of the economic ministries, reconcile the armed forces’ long-term plans with the five-year national economic development plans, and coordinate all defense issues that arose within Gosplan. The military strongly influenced the work of the sector and used it to lobby for the interests of the armed forces; in this sense Gosplan was not a monolithic agency. The proposals that emerged from the defense sector tended, however, to be moderate by comparison with the initial proposals for which the armed forces pressed; sometimes they were also more modest than the outcomes of decisions made at the highest levels (Samuel’son 2001: 92, 195). In the second half of the 1930s the whole system of defense industry administration was refashioned, and this shifted the locus of planning authority. At the end of 1936 a separate ministry of the defense industry was sliced out of heavy industry. April 1937 saw the final disbanding of the STO, while the joint Defense Commission was reorganized into the Defense Committee. In December 1937 the Gosplan defense sector was wound up and its powers transferred to a new mobilization sector; the Defense Committee staff were also fundamentally reorganized.∑ The Defense Committee assumed the predominant role in dealing with defense issues.

Planning the Supply of Weapons

87

The reorganized Defense Committee had four departments. The first of these regulated, coordinated, and approved mobilization plans and reserves; the spring of 1938 saw the short-lived establishment of a special militaryindustrial commission (VPK) of the Defense Committee to exercise leadership over mobilization planning; it lasted only until 1939. The second department dealt with military orders; it brought the defense ministry plan of military orders to the Defense Committee for review and monitored its implementation. The third department regulated and audited budget outlays on defense and reviewed ministerial requests for access to defense funds. The fourth department provided the Defense Committee with administrative support.∏ The personnel were picked from the Red Army command and leadership staff; this consolidated the privileged influence of the military in planning that they had previously exerted in the Gosplan defense sector. Civilian specialists, even those representing the ministries of defense industry, could obtain access to Defense Committee papers only if armed with special authorization ‘‘for this purpose by ministers personally.’’π Within the Defense Committee the first and second sections (and, in 1938/39 its VPK) had the dominant roles. This resulted from their responsibilities for mobilization planning and current planning, respectively. The binary division became still more clear after the reorganization of June 7, 1940, which reduced the Defense Committee structure to a ‘‘mobilization-planning department’’ and a secretariat. The department’s tasks were ‘‘work on the mobilization plans of industry, transport, and [air defense] on the basis of the requirements of [the ministries of defense and the navy] and the NKVD troops’’ and to prepare all matters concerned with the mobilization of industry and transport for review by the Defense Committee. The secretariat was to prepare current defense industry items for review by the Defense Committee including annual and quarterly plans for current military orders, military freight plans, military and naval construction, starting and stopping production of specific military items, and defense finance.∫

Defense in Economy-Wide Planning A main problem that arose in planning the needs of defense was to balance them against ‘‘the necessity of maintaining the forced pace of industrialization of the country.’’Ω Both military and civilian specialists recognized ‘‘that the perspective plan of preparation for defense should be organically incorporated in the overall plan of the national economy.’’∞≠ The basic difficulty was how to do this in practice. What should come first: the demands of the military or the needs of the economy? The prime responsibility for economy-wide plans fell to Gosplan. In confor-

88

Andrei Markevich

mity with a framework established by government decrees it issued directives comprising the chief indicators of the plan for the next period. Lengthy negotiations and deals with branch ministries subsequently converted preliminary guidelines into a finalized plan approved by the government. On the other hand the military was primarily responsible for coming up with proposals to strengthen the country’s defensive capabilities. More precisely, they had to work out the level of economic development that was theoretically required to support the mobilization requirement of a successful war effort at the end of the annual or quinquennial plan period. But it was not their responsibility to work out how to achieve this level. Thus Gosplan received proposals for national defense–related projects from the defense ministry, and for local defense projects from the staff of the territorial military districts; the latter had to conform, in turn, with the framework of instructions issued by the general staff to local planning commissions of the Union Republics, territories, and districts.∞∞ Because the military and Gosplan worked on these matters simultaneously and in parallel, it follows that Gosplan’s preliminary directives often failed to take into account the needs of defense as the Army saw it; it was on the basis of these preliminary directives, however, that all other economic agents calculated their own proposals. This gave rise to repeated supplementary directives concerned with planning the needs of defense, coupled with periodic complaints to the effect that defense requirements were being planned in isolation and then added mechanically to the overall plan for the economy. In sum, defense planning was a matter of ‘‘evaluation and correction of the overall economic plan from the standpoint of providing within it for the requirements of defense.’’∞≤ The compartmentalization of military-economic planning was also fostered by the regime of secrecy. Such plans appeared in two forms, either ‘‘covertly within overall [control figures]’’ or ‘‘in the form of a separate secret appendix’’ to more open documents.∞≥ Procedures for compiling overall plans and plans for defensive preparation were also separate and distinct. The joint government and STO decree of February 7, 1928, ‘‘On deadlines and procedures of compiling the five-year plan of the national economy,’’ for example, was followed by a supplementary decree of STO, ‘‘On the procedure of work on compiling the perspective plan of preparation of the national economy for defense.’’∞∂ Similar decrees were adopted in the case of annual plans.∞∑ To ensure that the needs of defense were ‘‘covertly’’ taken into account in the overall five-year plan it was envisaged ‘‘to receive the perspective plans of separate branches of defense (the five-year plan of construction of the armed forces, defense industry, strategic railroad construction, etc.) in the STO ex-

Planning the Supply of Weapons

89

ecutive session prior to final aggregation [do okonchatel’noi svodki] of the overall five-year plan.’’∞∏ The parallel planning of defense and the economy gave rise to frequent failures of process. The first five-year defense plan was never formally adopted, for example (Simonov 1996b: 89). Defense was included in the first five-year plan for the economy as a whole, but ‘‘covertly’’ (Samuel’son 2001: 93). During the first five-year plan period there was no perspective plan for rearmament, an omission that the planners themselves often lamented.∞π This story was repeated under the second five-year plan (see Chapter 5). Throughout the 1930s Gosplan did not assemble any annual plans for defense generally; ‘‘such plans were compiled only by the Gosplans of the union republics.’’∞∫ Officials at every level from ministries to enterprises often complained that instructions relating to defense needs were generally issued later than more general directives in the planning cycle; this tended to devalue the work done at earlier stages and required each ‘‘new assignment to be worked through in a great hurry.’’∞Ω The delay ‘‘makes it impossible to work attentively and thoughtfully with special purpose.’’ The result was ‘‘frequent revision and rebuilding of estimates for outlays on defensive readiness,’’ which put a brake on the ‘‘final scrutiny and approval of estimates for the next financial year’’ and on the signing and implementation of contracts.≤≠ This led agents, on one hand, to press for earlier circulation of directives relating to defense plans and, on the other, to delay the processing of overall plans until the intentions of the central agencies with respect to both civilian and military programs had been clarified. In 1929, for example, the ministries were supposed to supply Gosplan with papers relating to defense by August 1. As of August 20 ‘‘not one department and not one union republic has provided any papers to Gosplan with the exceptions of the OGPU troops, [the ministry of agriculture of the Russian republic], and very preliminary plans of [the ministries of transport and communications].’’≤∞ The STO executive session had no alternative but to extend the deadlines previously announced.≤≤ The reconciliation of defense plans and economy-wide plans between Gosplan and the military was further complicated by the intervention of a third party: the ministry of finance. Defense needs expressed in the mobilization requirement first and foremost influenced the allocation of investment. Defenserelated investment was financed mainly from the state budget. Ministries compiled special itemized lists of defense projects, and in their investment plans a separate column showed the financial limits for defense items.≤≥ At some point the itemized projects had to be reconciled with the funds available. When the 1929/30 annual plan was compiled, for example, it emerged that the initial financial limit for defense industry, 185 million rubles, ‘‘was coordinated nei-

90

Andrei Markevich

ther with the mobilization assignment under plan ‘B,’ based on the [defense ministry] triennial mobilization requirement, nor with the mobilization assignment under plan ‘P’ for October 1, 1930, that [the industry ministry] had established as a transitional step to achieving plan ‘B.’ ’’≤∂ In other words the plan indicators based on the military mobilization requirements, whether in the plan’s minimal or optimal variant, were inadequately financed. After this had become clear, the industry ministry asked the STO executive session either to supplement the funding or to curtail the program insofar as ‘‘the [industry ministry] presidium cannot take it upon itself to alter the mobilization assignment under ‘P’ nor can it decide which items of the [defense ministry] triennial requirement to put off beyond the October 1, 1931, deadline.’’≤∑ The finance ministry, intervening on behalf of the government in an expert capacity to reconcile declared goals with the funds available, typically took a hard line and tried not to concede anything, whether to the Army or to Industry.≤∏

Planning the Defense Industry A COORDINATION PROBLEM

Planning the defense industry was central to defense planning in the wider sense. In the 1920s and 1930s the Soviet Union differed from most other countries in the degree to which its defense industry formed a distinct, specialized branch. The ‘‘cadre’’ factories were brought together under their own trusts, chief administrations, and eventually ministries of defense industry. Most defense items came from the specialized ‘‘cadre’’ facilities, including nearly 80 percent of all military goods produced in the country in 1940 (Simonov 1996b: 154). There was relatively little participation by civilian factories. Insofar as the cadre factories were designed to meet the wartime needs of the armed forces, peacetime defense requirements could not utilize them fully and they were also employed in producing civilian goods; in 1933, for example, civilian items accounted for more than one-third of the gross output of the defense factories subordinate to the ministry of heavy industry.≤π In 1938, however, after a few more years of rearmament, orders for the ministry of defense and the NKVD troops had risen to 80 percent of the planned gross output of the factories of the ministry of defense.≤∫ The result was that the definition of the defense sector was intrinsically fuzzy, as Figure 4.3 illustrates: the planners dealt with the defense industry as an administrative bloc, but the defense industry also served civilian consumers while the armed forces procured their weapons and military equipment from a circle of producers that was actually wider than the defense industry as such. In the Soviet economy all agents were subject to higher-level plans, and

Planning the Supply of Weapons

91

Figure 4.3. Military and Civilian Producers and Products

defense enterprises and their ministerial fundholders were no exception. The defense industry was planned in the same general way as other branches: the government transmitted preliminary assignments down to ministries, chief administrations, and enterprises in which they were elaborated in more detail before finalized proposals were returned from below through the higher levels to the government.≤Ω There were some particular exceptions, however. For example, the plans of defense suppliers included separate columns for items for military and civilian use. The most important difference was that the ministries and enterprises supplying weapons and military equipment had to fix their plans not only within the guidelines set by the authorities above them and with their agreement, but also by agreement with the purchaser: the defense ministry. This procedure was instituted at the beginning of the command system. According to a resolution, ‘‘On the procedure for approving the plans of the military industry,’’ adopted by STO on October 5, 1927, the defense ministry was to supply the industry ministry with full ‘‘information necessary for drafting the five-year plan of recovery and further development of the military industry and of accumulation of materials of the mobilization reserve.’’ The industry ministry drafted this plan and presented it to STO for approval, but only after agreement with the defense ministry and Gosplan.≥≠ In parallel with plans for defense industry the military drafted both the five-

92

Andrei Markevich

year and annual plans for current military orders that expressed their demands on industry. There was virtually no export of arms before the war, and the state was the sole purchaser of military items.≥∞ In principle, therefore, the plan of current military orders (in Figure 4.3, A + C) should have been equivalent to the output plan of the defense industry (A + B) after subtracting the defense industry’s civilian products (B), and adding the civilian industry’s military products (C). Although the defense ministry drafted both the five-year and annual plans for current military orders, the annual plans never added up to the five-year totals. Actual military orders were based on the international situation, not on the five-year plan, which generally proved to be a poor forecast. As Eugène Zaleski (1980) showed in the case of the economy-wide five-year plans, therefore, the five-year plans for military orders had little to do with reality. The annual plans for military orders had more chance of being realized although they too were often modified in the course of their execution. As for the annual plans for military orders, the procedure for drafting them and linking them with the plans for defense industry failed to stabilize during the 1930s despite the fact that the circle of individuals involved showed little change. A document drafted in 1939 by staff of the Defense Committee states: ‘‘The defense ministry has no statute that sets out the procedure and deadlines for signing agreements with industry, the procedure for resolving pre-contract disputes, or the procedure for negotiating prices, etc.’’≥≤ The military themselves complained repeatedly about this. In early February 1940, for example, naval officials commented that the lack of clear rules for drafting plans of military orders ‘‘governing the normal drafting of the plan of orders through the coordination system up to the point of its approval and provision with financial credits places extreme burdens and delays on the work of the [navy ministry] in both compiling the plan and implementing it in industry.’’ In the light of the navy ministry’s experience with the 1938 and 1939 plans of orders, the authors stressed the need for the Defense Committee to issue a binding ‘‘Instruction on drafting the plan of orders’’ which ‘‘should include clear requirements and instructions of the [navy ministry] for the entire system of work on the [navy ministry’s] plan of orders and define at each stage the mutual relationships with the Defense Committee, the Economic Council, and the industrial ministries that arise in the process of this work.’’≥≥ This reached the Defense Committee a week after receipt of a similar letter from the Red Army general staff.≥∂ A result of the lack of rules was that the plan of defense industry and the plan of military orders emerged as outcomes of the disputes and deals that were typical of the Soviet economy. In the autumn preceding the planned year the defense ministry compiled its annual requirement for weapons and equipment and sent it to industry. The ministries supplying defense products held

Planning the Supply of Weapons

93

special meetings with military participation to review the military orders that were envisaged, determine which orders could be handled by their factories and which could not, assess their value, and so forth.≥∑ From these meetings emerged the ministerial responses to the planned military orders, including unresolved differences with the military. Disputes were referred to special commissions; the central authorities made the final decisions. As far as we can judge their motivations, the military generally wanted to extract the greatest possible volume of armament from industry. Industry, on the other hand, wanted first and foremost to obtain plans that it could fulfill. An example illustrates the scale of resulting disagreements. In drafting the plan of the defense industry ministry for 1938, the defense ministry asked for orders to the tune of 13.3 billion rubles, a sum 3.3 times larger than the value of orders that the defense industry had fulfilled the previous year. The defense industry ministry agreed to accept orders only in the sum of 9.5 billion rubles. Government leaders received a list of disputed items more than forty pages long.≥∏ Disputes on this scale were encouraged by the isolation of military planning from planning the economy. In connection with the 1938 plan, for example, the defense industry ministry commented that ‘‘to realize the [defense ministry] order in full will increase the defense industry’s requirement for machine tools and metals (in particular nonferrous) much more rapidly than the growth rates of engineering and metallurgy that are envisaged in the national economic plan for 1938.’’≥π Defense Committee staff noted the same thing. One document remarks, ‘‘Financial planning in the defense ministry proceeds in isolation from the planning of physical stocks. There is no one agency in the defense ministry that unifies issues of overall planning of financial and physical stocks. Physical planning is carried out without any account of the production possibilities of industry.’’ The result was that Industry could not fulfill the Army’s equipment orders, while the defense ministry could not spend the financial resources assigned to this purpose.≥∫ Not all disputes arose from Industry’s refusal to accept orders on the basis that they could not be filled. The defense industry sought to increase orders for those items that were already in production. The industry ministry requested a larger plan for 1929/30 for rifles, cartridges, and explosive powder, despite the fact that the military had no need of them. It was motivated, evidently, by a desire to be asked to produce items that its enterprises could easily produce and so fulfill the plan at a profit.≥Ω Because Industry refused to offer the items the Army wanted on the grounds of the ‘‘overburdening of capacity,’’ the defense ministry purchasing staff were forced ‘‘to place only the orders that would be accepted although at the expense of the composition of armament and military equipment.’’∂≠ Despite the high priority of defense supplies, therefore, the military could

94

Andrei Markevich

not negotiate complete satisfaction of their demands. There were several reasons. First, the defense ministry had to operate within a budget constraint. In planning military orders for 1940, for example, a key problem that was referred to a special commission established by the Defense Committee to manage disputes between the Army and Industry was ‘‘determination of the overall sizes of the budget and volume of military orders.’’∂∞ Second, there were too many disputes for the central agencies to go into each in detail. The central agencies could not cope with the flood of military complaints about the conduct of industry, as the following story shows. In 1930 the defense ministry asked the government to make industry take on a number of orders that it was refusing to accept. But the government responded that it could not review these disagreements and ‘‘not to burden the government with issues that can be solved by inter-ministerial negotiation’’; it recommended that ‘‘in future such issues, to the extent that they pile up, should be considered jointly by the USSR Revolutionary Military Council [the defense ministry leadership] and [the industry ministry]’’; it was permitted to refer such issues to the government only ‘‘in the event that they cannot be resolved by the efforts of the ministries.’’∂≤ The result was to differentiate issues by status: the more important the problem, the higher the committee to which it could be referred. Overall plan indicators and summary figures for military orders were reviewed in the Politburo and by Stalin personally.∂≥ Less important issues were dealt with in the Sovnarkom, STO, the Economic Council, or Defense Committee. Industry could freely refuse orders of lower priority, or negotiate their acceptance on advantageous terms. The orders that industry rejected were a permanent headache for the defense ministry. The military repeatedly asked the government to give the plan of military orders priority over the plans of industry, and remove the other ministries’ option ‘‘to refuse to fill orders.’’∂∂ A basic precondition for coordinating the output plans of defense industry with the Army’s plan of military orders was a procedure that aggregated the various items of plan information about defense production as a whole. In the early 1930s this did not yet exist. Special decrees that brought together all the indicators of the defense economy, for example, were not issued at this time; the practice emerged only after the establishment of a specialized defense industry ministry. In 1933, for example, Gosplan complained to the government, ‘‘In the situation that has been created, an overall, mutually coordinated plan for defense industry is not compiled at present. The plan of the ministry of heavy industry for the third quarter was not compiled and therefore was not reviewed by Gosplan.’’ The same prevailed in defense industry investment: ‘‘In practice construction goes ahead without itemized lists and plans for commissioning of production that have been reviewed by Gosplan and confirmed by

Planning the Supply of Weapons

95

the government. The trusts of the chief administration assemble and distribute itemized lists for implementation that have not been coordinated with each other and are not consistent in the spirit of unified objectives.’’ ‘‘The same situation has also emerged in the field of working out production programs, labor plans, quality indicators, and financial plans: these are all worked out by the trusts, and go essentially without the serious review or audit by higherlevel agencies that is necessary for coordination of the plan.’’∂∑ Gosplan requested the Sovnarkom to issue a decree obliging the ministry of heavy industry ‘‘to present overall annual and quarterly plans for the defense industry through Gosplan for confirmation by the Sovnarkom, with the [same] range of indicators prescribed for heavy industry as a whole and the same deadlines prescribed for presenting plans for the ministry as a whole.’’∂∏ The absence of overall plans for the defense industry in the first half of the 1930s resulted from the lack of a planning office that could act in this capacity. According to Gosplan, ‘‘The central planning sector of the [ministry of heavy industry] essentially stands aloof from planning the defense industry while GVMU [the ministry’s chief war-mobilization administration] brings together mobilization work in defense industry but not the current plan.’’ Secrecy made things worse. The same report noted, ‘‘Since the work of defense industry . . . is not open to press criticism, as a result we are faced with the great backwardness of defense industry in methods of work, standards of consumption of materials, cutting costs, and so forth.’’∂π It seems likely that this situation was one of the considerations behind the separation of the defense industry ministry from the ministry of heavy industry. The plan of defense industry and the plan of current military orders were finally reconciled by special government decrees. This, too, took some time to achieve. The output plan of the defense industry ministry for 1938, for example, was approved by a Defense Committee resolution of February 20, 1938. The resolution fixed the gross value of output of the industry, the volume of orders of the defense ministry and NKVD, the plan of output of the main military and civilian items in physical units, the average value of gross output per worker, the numbers to be employed and the wage bill, and a target for the percentage reduction in unit variable costs.∂∫ The procedure that confirmed the defense industry plan for 1939 was similar.∂Ω On the same procedure the Defense Committee approved plans for the defense industry following government approval of the annual plan for the whole economy, which already included assignments for the various ministries, including that of the defense industry, and of the state budget, which already included rows for outlays of the defense ministry, the NKVD, and the plan of military orders. This was a major difference from planning procedures in the civilian economy, where

96

Andrei Markevich

ministerial plans did not usually reach the stage of final approval (Markevich 2004). In short, the plans of the defense industry were somewhat more rigidly defined than those of the civilian branches, and this reflected their high priority. After the war, in contrast, the plans of the defense industry ministries were confirmed by ministerial decrees, like those of the civilian ministries.∑≠ Mark Harrison (2003) has proposed a model of Army-Industry relations in which Stalin exploited the lack of coordination of ex ante supply (Industry’s production plan) with demand (the Army’s plan of military orders) to foster their mutual rivalry and so divide and rule. The archives suggest that this model is oversimplified. In the strict sense it can apply only to the early 1930s. From the second half of the 1930s the Politburo did coordinate ex ante supplies and demands, at least in principle. I will show that this did not put an end to the jockeying between the Army and Industry for one-sided advantage in the internal marketplace for weapons. It appears, however, that this resulted from the failure of significant efforts to coordinate supply and demand rather than from a deliberate absence of effort to do so. EFFECTIVE COORDINATION AND ITS COSTS

From the second half of the 1930s the plans of the defense industry ministries began to be confirmed directly by the government. Other problems of Soviet planning that characterized the civilian branches persisted in the defense sector, however. First, deadlines for the approval of plans were rarely met; procedural difficulties in reconciling the plan of military orders with ministerial production plans were a major factor here. The annual defense industry plan for 1938, as noted above, was approved only on February 20. Confirmation of the navy ministry’s plan of orders for 1939 was delayed until March of that year, and with it the production plans of the branch ministries; the same plans for 1940 were not approved until February. ∑∞ The 1941 production plan for the ministry of ammunition was confirmed on February 14. In a memorandum to the party central committee dated April 8, 1941, the Gosplan mobilization department noted: ‘‘Large delays in the approval of the plan of defense industry by assortment in physical units are a persistent phenomenon in all the ministries, as are discrepancies between the physical assortment and the aggregate and quality indicators’’ (cited by Simonov 1996b: 131). Plans were passed down to chief administrations and enterprises with still longer delays. As Table 4.1 shows, the ministry of heavy industry distributed the finalized version of the plan of special orders for 1936 to its chief administrations only on February 23 of that year. In 1941, in the much smaller and more specialized ministry of ammunition, the same condition was reached only on March 11. As for the enterprise level, in 1935 the defense factories received their finalized plans only at the end of July; in 1936, during the second

Planning the Supply of Weapons

97

Table 4.1. Finalized annual plans: Ministerial distribution to chief administrations and enterprises Year

Date

Ministry

Distributed to:

∞. ∞Ω≥∑

End-July

Heavy Industry

Defense factories

≤. ∞Ω≥∏

February ≤≥

Heavy Industry (special orders)

Chief administrations

≥. ∞Ω≥∏

Second half; October/November in some cases

Heavy Industry

Defense factories

∂. ∞Ω≥π

March

Heavy Industry

Shipyards

∑. ∞Ω≥π

October ≤≥

Heavy Industry

Sarkombain aircraft factory

∏. ∞Ω≥Ω

June ≤∫

Armament

Molotov armament factory

π. ∞Ω∂∞

March ∞∞

Ammunition (military orders)

Chief administrations

Sources: Rows 1 and 3: 1935: RGAE, 1562/329/120: 30–33 (head of the statistical administration Kraval’, memorandum to Stalin, October 23, 1935); 1936, second half of the year: RGAE, 1562/329/120: 41–43 (acting head of the statistical administration special sector Tutenkov, explanatory memorandum on ‘‘Fulfillment of the production plan of military industry for nine months of 1936,’’ not dated). 1936, October or November: Hoover/RGASPI, 77/1/286: 53 (transcript of remarks of Anisimov [Voroshilov factory] at a meeting in the presence of comrade Zhdanov with the party committee secretaries of numbered factories, December 23, 1936). Row 2: GARF, 7297/38/247: 434 (ministry of heavy industry, decree no. 21s, February 23, 1936). Row 4: RGAE, 8183/1/146: 34ob (condensed transcript of remarks of Kharitonovich, planning official of the defense industry ministry second [shipbuilding] chief administration, at a meeting of the second chief administration party group of activists to discuss the outcomes of the February central committee plenum, April 11–13, 1937). Row 5: RGAE, 7515/1/156: 284–288 (head of the defense industry ministry third chief administration Vannikov, letter to director Novikov of the Sarkombain factory, October 23, 1937). Row 6: Hoover/RGANI, 6/2/250: 27 (KPK plenipotentiary for the Khabarovsk region, information to KPK chairman Andreev ‘‘On leadership and support of the Molotov factory from the ministry of armament and [its] chief administration,’’ November 21, 1939). Row 7: Simonov (1996b: 130).

half of the year and in some cases not until October or November. In 1937 the naval shipyards received their annual plans in March. In the same year its third-quarter plan reached the Sarkombain aircraft factory in Saratov on October 23, three weeks after the quarter had finished. In 1939 its annual plan reached the Molotov armament factory on June 28. A planning official in the defense industry ministry, Kharitonovich, re-

98

Andrei Markevich

marked: ‘‘It’s clear that if we get involved in the detailed reconciliation of the plan in all its components, we cannot issue it in time.’’ This was a problem in all branches of the economy: the need to trade the detailed accuracy of the plan against its timeliness. The solution that Kharitonovich envisaged was: ‘‘Even if we make mistakes in a few respects, we will gain from compiling the plan in good time. I would rather sacrifice the perfection of a few points in the plan and at least issue the plan in good time.’’∑≤ In summary we have the following picture. Factories always began the year or quarter with only the most general plan indicators, and these were preliminary and subject to further change, especially as regards production. The finalized, detailed version of the plan came much later and, in some cases, not at all.∑≥ The same Kharitonovich gives an example: At the end of 1936 the shipyards were given 1937 figures under the special heading of ‘‘control points.’’ These figures amounted to the yards’ portfolios of orders. The yards had to regard them as such, the more so insofar as at the beginning of the year the head of the chief administration called in the factory directors and chief engineers and told them that the points issued to them were not a plan but amounted to the portfolio of orders. The question was raised whether we would be expecting the factories to fulfill the control points in all aspects. [The reply was that] we would not, in particular because at the time when they were compiled there was no previously confirmed plan of defense ministry orders, which was being confirmed by a special procedure conditionally on the finances assigned to the defense ministry, and we had not yet received a firm plan, which was passed down to us only in March. We were unable to issue a firm plan to the factories at the end of 1936. To orient them in relation to the work in prospect we informed them of the substance of their portfolios of orders (the control points).∑∂

In the defense industry, meanwhile, as in the civilian branches, ‘‘dual planning’’ was widely practiced. Dual planning meant that the chief administration played safe by issuing a more ambitious plan to enterprises than had been confirmed by the government (as defense industry minister Moisei Rukhomovich recognized in May 1937 in remarks cited by Simonov [1996b: 109]). The planners themselves saw nothing wrong in this: What is it that amounts to so-called dual planning? Maybe it’s like this: for the second quarter the defense industry ministry approved a limit of 230 million to the chief administration, and we [the chief administration] gave out 250 million to the factories because the assortment plan that [the ministry] also approved for us doesn’t fit within 230 million. I think that a 10 percent difference is hardly a catastrophe. I wouldn’t try to cover up the shortcomings that we actually have: plans include elements that turn out to be unsupported

Planning the Supply of Weapons

99

in the factory; mistakes happen; we sometimes issue plans that are not always exactly in line with the directive instructions of the defense industry ministry. Here I recognize this and I consider that we should set ourselves the goal of eliminating these shortcomings in the future.∑∑

This style of work naturally reduced the quality of planning and promoted conflicts both between the plans of various economic agents and within each agent’s plan. Finally, the quality of defense planning was degraded by agents’ selfinterested manipulation of information, as in the economy more widely (Markevich 2004). Higher-level bodies approved the plans of defense agencies and enterprises at lower levels on the basis of the information that the latter had supplied. This information was largely taken on trust because the higher-level agencies had few means of verifying it. As a result, the agents responsible for executing plan assignments themselves had great influence over those who approved their assignments. Preferring to be given assignments that could be fulfilled without great effort, the agents tended to understate their possibilities and overstate their needs, and this in turn promoted the practice of understated planning. The central authorities understood this problem and recognized that ‘‘the industrial ministries typically do not get down to individual enterprises.’’∑∏ The quality of planning at lower levels ‘‘cannot be high, if the higher-level agencies do not review or cross-check their projections; large errors are an inevitable consequence of the lack of checking and internal consistency of the plan of defense industry as a whole.’’∑π Attempts to change this situation by means of periodic campaigns and audits had only short-term effects.

Decentralized Contracting The plans of the defense industry ministries and the plans of current military orders remained highly aggregated. Plans at lower levels were more detailed, but this means that they incorporated a mass of unsolved problems— for example, how to identify the specific source, user, and quality requirements of a specific line of output. These problems had to find their solutions in specific contracts between particular suppliers and purchasers. Contracts were a crucial element of the planned economy that would not have been able to function in their absence. No plan could anticipate the whole range of desirable links between producers and consumers; the transaction costs of compiling such a plan would have spiraled to infinity (Mises 1981; Hayek 1991). Unlike plans, the authority to make contracts was decentralized. Contracts specified a producer, a consumer, prices, quantities, and

100

Andrei Markevich

deadlines. Contracts were also necessary to smooth over planning errors. Contracts were operational by definition and so helped to solve such problems as how to work in the gap between the beginning of the plan period and the finalization of the plan, what to do when plan indicators were inconsistent or in conflict, and so forth. Contracts gave detail to plans. The processes of planning and contracting went in parallel: the preparation of contracts began shortly after the drafting of annual plans began, while the conclusion of contracts coincided roughly with the confirmation of plans. Often it happened that contracts were signed before plans were approved, but they could then be revised after plans were confirmed. A contract in this planned economy was different in principle from a contract in a market economy. It was not the outcome of unforced convergence of the parties on terms for mutual benefit. The context was set by the annual ‘‘contract campaign.’’ The official purpose of contracting was simply to fulfill the agents’ approved plans. In reality, the decentralized character of the contracting process left both sides room to maneuver. In some cases the dissatisfied party could reject the terms offered altogether. Centrally funded commodities tended to be those in shortest supply, so contracts for funded commodities tended to be relatively strictly tied to plans for their distribution. In negotiating for nonfunded items, both buyer and seller could exercise greater discretion. The contracting process differed in minor ways between the defense and non-defense branches. Generally, the ‘‘contract campaign’’ was initiated and managed from the center. The autumn of the year before marked its start with the publication of a raft of government resolutions and ministerial decrees. These legislative acts prescribed deadlines for the conclusion of contracts, listed the agents who should conclude them, and set out the procedure for resolving contractual disputes. The campaigns themselves proceeded at two levels. One level was the conclusion of ‘‘model contracts’’ and ‘‘master delivery agreements’’; the other was the conclusion of the specific contracts themselves. The campaigning unit was the ministry: each ministry had its own campaign, while the government fixed only the general rules of the game. Model contracts and master delivery agreements were concluded between the ministries producing and purchasing goods. They fixed the general procedures for delivery and acquisition of items, the requirements for quality and assortment, the rejection of substandard goods and associated penalties, and mechanisms for resolving disputes.∑∫ The list of commodities subject to master delivery agreements was revised every year by the government. In 1940, for example, the minister of armament had to sign thirty-nine master delivery agreements with the ministries supplying it.∑Ω Specific contracts were concluded at the enterprise level between the actual

Planning the Supply of Weapons

101

producer and consumer. This is where the detail of price, quantity, and assortment was thrashed out. The signature of specific contracts marked the closure of the contract campaign. In aggregate, specific contracts were supposed to fulfill overall plans in conformity with ‘‘master delivery agreements settled between supplier and consumer ministries’’ on one hand and model contracts on the other.∏≠ The side of the supplier bore primary responsibility for designing model contracts, master delivery agreements, and specific contracts. In the course of negotiation with purchasers, a mass of disputes and conflicts naturally arose. Conciliation could be pursued directly or through the state arbitration courts of Gosarbitrazh. The latter was organized hierarchically so that the level at which a review was held corresponded with the status of the contending parties. Thus a dispute among enterprises of the same ministry would be settled by intra-departmental arbitration. The arbitration system also reviewed the fulfillment of contractual terms and imposed penalties on violators (Belova 2005). The high degree of secrecy and priority in matters pertaining to defense industry affected procedures for resolving disputes between buyer and seller. An STO resolution of November 28, 1931, established a special financial and technical arbitration commission within the ministry of the workers’ and peasants’ inspection to examine disputes over military orders. In 1934 the inspection ministry was abolished and the arbitration commission was replaced by a new special procedure. Disputes were taken out of the hands of local arbitration agencies and centralized under the chief arbitration officer ‘‘regardless of the sum at stake.’’∏∞ Because the chief arbitration officer was immediately overwhelmed by the resulting caseload, in 1936 he fixed a minimum value for the cases that he would review, the requirement being initially 10,000 and later 25,000 rubles (see Chapter 3). Actions for smaller sums over military orders not classified as secret were relegated to the ordinary arbitration procedure.∏≤ In short, the process of contracting arose from the annual planning process, and most of the problems that arose in planning also cropped up in the contract campaigns. We will take the campaign of 1940 in the ministry of armament as an example. The preparatory stage began before the government had issued the framework resolutions. In mid-September 1939 the ministry of armament issued its own decree no. 264, ‘‘On preparation for the conclusion of contracts for 1940.’’∏≥ It ordered enterprise directors, ‘‘in consultation with operational staff for supply, sales, and finance, and legal advisers, to review the operating systems for contractual relations and ‘master delivery agreements’ ’’ and pre-

102

Andrei Markevich

sent proposals to the chief administrations and trusts not later than October 1, 1939. The trusts and chief administrations in turn were to draft the master delivery agreements and model contracts and present them to the ministry of armament arbitration staff by October 5 for transmission upward to the minister by October 15. This decree turned out to be premature given that the 1940 plan assignments for the ministry and its enterprises remained unknown; in consequence, its deadlines could not be realized. They were revised in subsequent decrees of the government and ministry. The Sovnarkom decree of November 8 ‘‘On the conclusion of contracts for 1940,’’ served to signal the real start of the contracts campaign.∏∂ Producer ministries were to provide master delivery agreements for an appended list of commodities by November 20. Deadlines were also given for the conclusion of specific contracts. Ministers now circulated the corresponding decrees in their own fiefs. Another such decree appeared in the armament ministry; the decree no. 264 that had preceded it was forgotten. Table 4.2 shows that the new ministerial decree required the master delivery agreements and model contracts to be drafted within more compressed deadlines than the center had envisaged, except for deliveries to ‘‘other’’ (therefore unimportant) purchasers. This was motivated partly by the minister’s desire to play safe and insure against the natural tendency of those below to ignore deadlines, and partly by a desire to give enterprises extra time for the negotiation of the contracts themselves. The ministry wanted its master delivery agreements confirmed by November 10, and by November 15 a model plan for ‘‘conclusion of agreements for 1940 in quantitative and summary terms.’’ This was at the same time that chief administrations were typically embarking on the first draft of enterprise production programs for the coming year (Markevich 2004). For conclusion of the specific contracts the ministry decree repeated the government deadlines. Responsibility for the timely conclusion and the appropriateness of contracts was laid personally on enterprise directors and heads of chief administrations and others at the equivalent levels. All were obliged to report progress to the ministry. A deputy minister was given personal charge of the contracts campaign as a whole. Its progress would be monitored by the ministerial arbitration and inspection officers.∏∑ Despite the theoretical coherence of the contracts campaign, its implementation was still a complex process. Of the thirty-nine master delivery agreements that the ministry of armament central administration of material supply should have signed with suppliers, none was ready by the November 20 deadline, and twenty-eight remained unsigned by the first of 1940.∏∏ Things were hardly better with the specific contracts and master delivery agreements that

Table 4.2. The 1940 contracts campaign in the ministry of armament: Government and ministerial deadlines Government deadlinea

Ministry of armament deadlineb

Nov. ≤≠

Nov. ∑

Central administration of material supply in consultation with chief administrations

Nov. ≤≠

Oct. ≤∑

Central administration of material supply

For items of mass consumption to be supplied by enterprises and agencies of the ministry of armament

Nov. ≤≠

Oct. ≤∑

Mass consumption trust

For items to be supplied for intermediate use by enterprises within the ministry of armament

Nov. ≤≠

Nov. ∞

Department of cooperation (i.e., intra-ministry subcontracting)

Nov. ≤≠

Nov. ∞

Ministry

Nov. ≤≠

Nov. ≤∑

Chief production administrations

Model contracts and master delivery agreements For items to be received by enterprises and agencies of the ministry of armament From other ministries

From Oboronpromsnabc

For items to be supplied by enterprises and agencies of the ministry of armament To ‘‘basic purchasers’’—i.e., the ministries of defense, the navy, and NKVD To ‘‘other ministerial purchasers’’ Specific contracts For mass consumption items

Agency responsible within the ministry of armament

Dec. ≤≠

Dec. ≤≠

...

For items of intermediate use and equipment

Jan. ≤≠

Jan. ≤≠

...

For freight transport

Jan. ≤≠

Feb. ≤≠

...

For capital construction in progress

...

Feb. ∞

...

For new capital construction

...

Mar. ∞

...

Sovnarkom decree no. 1648, ‘‘On the conclusion of contracts for 1940,’’ October 8, 1939. Ministry of Armament decree no. 297, ‘‘On the conclusion of contracts for 1940,’’ October 17, 1939. c Oboronpromsnab: the organization responsible for intermediate supply of defense industry enterprises. a

b

Source: RGAE, 8157/1/134: 25–26, 49–50.

104

Andrei Markevich

were the responsibility of other administrations. Only the ‘‘Master agreement for cooperation and sales of mass consumption goods’’ was ready on time. The ‘‘Master agreement for delivery of goods by Oboronpromsnab’’ (see the note to Table 4.2) was approved ten days late, with a seventeen-day delay for the model contract for supply of principal products of the ministry. An audit showed that the heads of chief administrations and the central administration for material supply ‘‘have not engaged directly with this work, delegating it to second-rank staff of the chief administrations.’’ In the same way the task of concluding specific contracts proceeded at ‘‘unsuitable [lack of ] speeds.’’ By mid-December 1939 the conclusion of contracts had only just started.∏π At the end of December the difficulties with the campaign were specifically raised in a meeting of the ministry of armament collegium, which obligated the heads of chief administrations and enterprise directors ‘‘personally to take all measures to complete the conclusion of contracts for mass consumption goods in the coming days and finish the conclusion of the remaining contracts within the deadlines approved by the Sovnarkom resolution of October 8, 1939.’’ It required them immediately to compile plans of the conclusion of contracts for 1940 in summary terms. The chief of the central administration for material supply was to complete the negotiation of master agreements with the supply ministries without delay and circulate them to the armament enterprises.∏∫ There were two underlying causes of delay. First, both the contracting parties tended to hold back from agreement until they had been informed of their future plan obligations so as not to conclude agreements unnecessarily and end up having to supply more than the plan would have required.∏Ω Second, one of the parties could often gain a one-sided advantage by using delay as a bargaining tactic in fixing contract terms. Reporting to the ministry on progress with contracts at the end of 1939, the heads of chief administrations noted that the signing of contracts ‘‘is being delayed by a number of disagreements over quantity, quality terms, and prices’’; ‘‘Contracts with [the navy ministry] have not yet been concluded because the navy ministry considers it possible to proceed to signing contracts only after confirmation of the program of orders for 1940 by the government.’’π≠ Delays in signing specific contracts and disputes over detailed contract terms culminated not infrequently in a collapse of negotiations; thus, the theoretically straightforward relationship between plan and contract sometimes broke down. In such cases, examples of which are scattered through the archives, the industrial supplier and military purchaser failed to sign any contract for items that had been planned. Called ‘‘precontract disputes,’’ these were subject to automatic review by the arbitration agency. The outcome of arbitration was formally binding on both sides. The state arbitration agencies

Planning the Supply of Weapons

105

annually adjudicated on tens of thousands of cases of precontract disputes, amounting to between 5 percent and 15 percent of their overall workload (Belova 2005). Even this authority was not always effective in compelling the parties to agree. The enterprises’ contract disputes with the military were in large part an extension of the conflicts that arose in reconciling the defense industry’s production program with the defense ministry’s plan of military orders. By refusing to come to specific contract terms, the parties tried to re-engage in battles already lost, as they had thought, at the stage of reconciling the production plan with the plan of military orders. Another factor in the military rejection of orders was changes in their perception of future wartime requirements. It is not possible to say which side bore the chief responsibility for breakdowns, but the frequency of breakdowns is hard to overstate. At the end of 1938, for example, the ministry of defense industry requested the government to trim its ministerial gross output plan by 535.2 million rubles, or 4.4 percent, appending a detailed list of the assortment and value of items that it wished to drop. In each case the justification offered took one of two forms.π∞ Either the purchaser had refused to conclude a contract to accept the item concerned, or a contract had been signed for a lesser value than that initially planned. These reasons accounted for more than half of the value of the items that it was proposed to drop.π≤ Thus Industry tried to present the situation as precipitated by the actions of the Army. Clearly, however, the responsibility should have been shared because the Army’s refusal was based on the terms that Industry offered: it takes two to disagree. If the enterprise walked away from negotiations, then it typically blamed the technological difficulties of satisfying the military’s requirements: the lack of production facilities, the mismatch between the product and the available technologies, and so on. Underlying this lay the producers’ desire to limit their efforts to ‘‘easy’’ items that did not require new investments or special efforts. This limited the range of goods for which they would willingly accept orders. If it was hard to refuse top-priority orders, those of secondary status provided Industry with much more room to maneuver, particularly because the center could not monitor everything at once. The Army employed several stratagems to limit Industry’s opportunistic behavior. First, they could use the official route of referring disputes to Gosarbitrazh. Second, they could work more closely with enterprises at the earlier stage of reconciling the plan of military orders with the production plan. This route was taken in 1940, for example, by the command staff of the first and second Far Eastern Red Banner armies, when they sought to win direct influence over the production program of the Molotov no. 106 factory, which

106

Andrei Markevich

supplied artillery and small arms production and repairs to these armies.π≥ Third, one of the tasks of the military agents in industry was to lobby for the conclusion of contracts (Markevich 2005b). They had to report all cases of refusal to sign contracts to the central organs of the defense ministry so that steps could be taken. A timely signal enabled the defense ministry to step in and force acceptance of the contract concerned. In 1938, for example, defense industry factory no. 145, the sole supplier of two-headed lubricators, refused a delivery contract on the grounds that the workshop was under reconstruction. The military agent reported to the Red Army artillery administration, ‘‘The factory . . . has reequipped the workshop . . . it has sold off some of the equipment for manufacture of lubricators. The factory is also selling off the lubricators in stock that it has not declared.’’π∂ This enabled the artillery administration to intervene and force the restoration of production before it was permanently stopped. The army did not always succeed in forcing the acceptance of the contracts that it needed. The result, in the words of defense industry minister Rukhimovich in 1937, was that ‘‘contracts with the defense ministry have not matched either annual or quarterly plans for output in terms of either quantities or deadlines’’ (cited by Simonov 1996b: 109).

Plan Revisions and Contracts The process of planning the defense industry did not finish with the approval of plans and conclusion of contracts. As in other branches, plan assignments and contracts for defense items were liable to frequent interventions and revisions. The high priority of defense did not enable greater plan stability; on the contrary, it may have increased the frequency of plan amendments. Many revisions resulted from high-level intervention. Changes in the international environment and the government’s foreign policy orientation, for example, were speedily reflected in reviews of military requirements and defense industry assignments. Long-term plans were strongly affected by such reviews. In the spring of 1929, for example, the government adopted a series of measures that dealt with rearmament and revisions to the five-year plan for military orders.π∑ A new ‘‘five-year plan of construction of the Red Army’’ was adopted and this led to further review of the armed forces’ mobilization requirement and plan of military orders. On the basis of the defense ministry’s new five-year requirement, Industry was required ‘‘to work out operational plans (capital investment, amounts and sources of finance, etc.) of the amount and provision of the Red Army’s construction and requirements during the five-year period.’’π∏

Planning the Supply of Weapons

107

Annual plans and contracts were less vulnerable to high-level intervention; in the course of a year, however, such interventions could still be large enough to require review of the plans of particular agents and even of the entire defense budget. Measures to increase the delivery of military items that were adopted by the Defense Committee in 1932, for example, led to an increase in the military budget for 1933. Forced into a corner, defense minister Voroshilov turned to heavy industry minister Ordzhonikidze for advice: ‘‘What is to be done?’’ (cited by Samuel’son 2001: 195). The quarterly plans and annual plan of the defense ministry third chief administration for 1937 had to be revised several times in the light of new government assignments.ππ The final variant of the annual plan was confirmed on December 22, 1937, with nine days to go to the end of the plan period.π∫ The Molotov armament factory’s investment plan for 1939 was amended five times.πΩ Most revisions that were initiated from below resulted from mismatches between plans and mistakes within them, and were triggered by the difficulties that arose in attempting to implement them, particularly problems of supply. When the third chief administration of the armament ministry sought changes in the fourth-quarter plan and contracts of factory no. 2 for 1937, it justified two changes on supply shortfalls, two more on the lack of familiarization with particular items of production, and one on poor quality, presumably of supplies.∫≠ When supplies were missing the factory might substitute other products for which the necessary stocks existed. ‘‘Take this case. The plan says series 9. During the year there is a complication with the diesels. Is this not something that ought to be reflected in the plan? Instead of [the original] series we’ll put another one in the plan, but within the same quota that was [already] given.’’∫∞ Although subordinates could adopt variations within the constraints of higher-level plans freely and more or less with impunity, plan reductions required higher-level approval. In the words of an economic official, ‘‘The chief administration has the right to reduce the plan, but no right to vary government decisions.’’∫≤ To summarize, interventions from above and below led to the situation clearly depicted by one military agent: ‘‘The plan is compiled and assignments are handed out to factories. The factories set the plan in motion but in a short period of time the whole thing goes to nothing and begins again from the beginning.’’∫≥ The defense industry’s planners and managers regarded this as entirely normal; they just adopted the view of Stalin (1937: 413), who had declared that planning only begins with compiling the plan. They affirmed, ‘‘For the plan to be vitalized requires [us] to introduce corrections while it is being implemented; flexibility is necessary and the factories must show versatility such as to allow them to switch to other tasks.’’∫∂ Before reporting on

108

Andrei Markevich

plan fulfillment at the end of each period it was necessary to work out revised plans that took into account all the interventions that had accumulated during the year. Thus defense planning did not escape the general problem of Soviet planning that in principle the plans were intended to overcome the supply constraints, but in practice the constraints tended to take precedence and the plans had to adapt to the constraints. The dominant influence of current supply became particularly apparent during World War II; the example of the tank industry shows that in such rapidly changing circumstances, factory plans were often drafted either retrospectively or not at all.∫∑ The high priority of defense made one difference: the constraints on its supply attracted the special attention of the top leadership. According to armament minister Boris Vannikov, ‘‘Stalin studied the summaries of aircraft and aeroengine production every day, and demanded explanations and adopted measures in every instance of a deviation from the target. . . . One can say the same about his involvement in reviewing issues of the tank industry and naval shipbuilding’’ (cited by Simonov 1996b: 128). A wide range of central agencies was drawn into meeting the dictator’s requirements, among them the Defense Committee staff who complained: ‘‘Practice shows that the existing establishment of the Defense Committee secretariat . . . in addition to its core work on compiling annual and quarterly plans of current military orders, has been burdened mainly by issues of the material and technical provision of current military orders, their frequent alteration, and issues of capital construction of the defense industry ministries.’’∫∏ Intensified regulation had the effect that plan alterations and interventions to assist the defense industry were shifted up the hierarchy to the highest level. Another consequence of the high priority of defense was the effect on its contribution to civilian supplies: when difficulties arose with the delivery of military items, the first reaction was often to cut back on the plan of civilian goods. In August 1932, for example, the planned output of instrumentation was almost halved (from 18.9 to 10 million rubles) and a special resolution of STO authorized the instrumentation association of the heavy industry ministry ‘‘to review contracts for the delivery of its products with all parties other than organizations of [the defense ministry].’’∫π Similarly, the civilian shipbuilding plan was cut back in 1935 ‘‘in the interests of naval shipbuilding.’’∫∫ All this did not go far enough, however, to satisfy the ambitions of the leadership for the defense industry. A memorandum prepared by Defense Committee staff for Molotov about defense industry plan revisions in 1938 declared, ‘‘The Defense Committee decisions on curtailment of the defense industry ministry’s program for 1938 were effectively brought about by the underfulfillment of the defense industry ministry’s program for 1938.’’∫Ω In

Planning the Supply of Weapons

109

mid-December 1938 defense industry minister Mikhail Kaganovich requested the Defense Committee to incorporate all the reductions into a finalized version of the 1938 plan.Ω≠ In other words, Industry sought to fulfill its production plan by having the target lowered to the achieved level of production rather than by raising production to hit the target. The armament ministry made a similar request on similar grounds in 1939.Ω∞ In a broad sense the planning of defense generally, and of the defense industry in particular, and the contract campaigns in the defense sector followed the same patterns as planning in the economy as a whole. The main differences followed from the exceptional priority of defense and included the participation of the military in plan procedures and the development of a complex methodology for negotiating military requirements with civilian planning agencies. The military contribution to planning current military orders had an important influence on defense industry plans; the military mobilization requirement likewise powerfully affected the allocation of investment and perspective plans for the economy as a whole. Other differences followed. The central agencies were preoccupied with issues of the country’s war readiness. This raised the status of all agencies with special responsibility for the supply of defense, even those that dealt with only minor aspects. The detailed drafting and negotiation of plans between the Army and Industry was important enough to be handled in the secret central core of the command system, by the defense sector of Gosplan and the Defense Committee staff. It was so important that formal procedures for finalizing plans were given effect in the defense industry long before they were extended to civilian branches, which continued to operate almost entirely under ‘‘preliminary’’ or draft plans at this time. High priority had some adverse effects on planning. The parallel activities of military and civilian planners were often poorly reconciled; the attempts at reconciliation prolonged the drafting process and made deadlines frequently unrealistic. Effective procedures for taking defense requirements into account did not emerge in economy-wide planning as it was practiced in the 1930s to match those followed in defense industry. Final decisions were the result of prolonged disputes that subsided only with the signing of specific contracts. The high priority of defense may have increased the likelihood of interventions and plan revisions in the course of their implementation. In the final analysis, military requirements were more important than the aspirations of industry; in this sense the plans of current military orders carried more weight than the defense industry’s supply plans. Despite this, industry had many means with which to defend its interests. It faced multiple re-

110

Andrei Markevich

quirements. As a result the government could not attach identical priority to all of them. From this, in practice, emerged a hierarchy of tasks. One result was that the aggregate of specific contracts for the delivery of military goods tended to add up neither to the plan of military orders nor to the output plan of the defense industry. The execution of these contracts depended greatly on the supply situation of industry, which ultimately dominated the outcomes of both plans of military orders and industrial output plans. Notes Thanks to Paul Gregory and Mark Harrison for advice and comments; I am responsible for remaining errors. 1. GARF, 8418/8/9: 3–6 (Gosplan deputy chairman Mezhlauk, letter to prime minister Molotov, ‘‘On the procedure for confirming annual and quarterly plans in defense industry,’’ dated July 31, 1933). 2. My use of the term ‘‘defense ministry’’ embraces a complex sequence of official designations for the ministerial departments that governed the Red Army and Navy. From 1923 to 1934 there was a single unified defense department, the people’s commissariat for military and naval affairs (NKVM). In 1934 it became the people’s commissariat of defense (NKO). From 1937 to 1946 it was subdivided into separate departments for defense (NKO) and the navy (NKVMF). In 1946 all the people’s commissariats were renamed ministries. The Revolutionary Military Council was the title of the defense ministry ‘‘collegium’’ (i.e., governing body) from 1918 to 1934. 3. Hoover/RGASPI, 85/27/405: 1–6 (STO executive session, ‘‘Provisional statute of the procedure for staffing the defense agencies of civilian departments, conditions of employees’ service, and procedure for processing defense issues,’’ June 25, 1927). 4. GARF, 8418/22/79: 5 (report prepared by staff of the Defense Committee, May 9, 1938); GARF, 8418/26/250: 32–33 (report on working arrangements in military departments in the listed ministries, June 21, 1938), 34–39 (report on the condition of mobilization work in the ministry of heavy industry, July 23, 1938), 39–40 (report on investigation of mobilization work of the special department of the ministry of local industry, July 22, 1938), 41–44 (report on the condition of the military department of the ministry of engineering, July 22, 1938), and 53–56 (report of investigation of the fifteenth administration of the ministry of defense industry, July 20, 1938). 5. GARF, 8418/12/402: 6 (Defense Committee resolution no. 190s, ‘‘On the apparatus of the Defense Committee of the USSR Sovnarkom,’’ December 7, 1937). 6. GARF, 8418/12/402: 10 (Defense Committee resolution no. 190s, December 7, 1937: Annex 2, ‘‘Statute on the apparatus of the Defense Committee of the Sovnarkom’’). 7. GARF, 8418/12/402: 9 (Defense Committee resolution no. 190s, December 7, 1937: Annex 1, ‘‘Establishment of the apparatus of the Defense Committee of the Sovnarkom’’). 8. GARF, 8418/24/226: 1–2 (Sovnarkom resolution no. 983/372ss, ‘‘On the reorganization of the Defense Committee apparatus,’’ June 7, 1940). 9. GARF, 8418/1/113: 16 (Gosplan, letter to the STO executive session, September 17, 1927).

Planning the Supply of Weapons

111

10. GARF, 8418/2/28: 1 (excerpt from minute no. 15 of the STO session, ‘‘On the procedure for work on compiling the perspective plan of preparation of the national economy for defense,’’ February 20, 1928). 11. GARF, 8418/22/631: 2–4 (chief of Red Army general staff Shaposhnikov and military commissar of the general staff Zaporozhets, letter to Defense Committee secretary Basilevich, April 21, 1938). 12. GARF, 8418/3/93: 2–3 (STO executive session resolution, ‘‘On the procedure and deadlines of compiling defense c[ontrol] f[igures] for 1929/30,’’ July 28, 1929); emphasis added. 13. GARF, 8418/1/61: 4 (STO executive session resolution, ‘‘Directives on accounting for the needs of defense in working out c[ontrol] f[igures] for 1927/28,’’ June 25, 1927). 14. GARF, 8418/2/28: 1, 11 (excerpt from minute no. 15 of the STO session, February 20, 1928). 15. For example, GARF, 8418/1/61: 2–4 (STO executive session resolution, ‘‘Directives on accounting for the needs of defense in working out c[ontrol] f[igures] for 1927/ 28,’’ June 25, 1927); GARF, 8418/3/93: 2–3 (STO executive session resolution, ‘‘On the procedure and deadlines of compiling defense c[ontrol] f[igures] for 1929/30,’’ June 28, 1929). 16. GARF, 8418/2/28: 1 (STO resolution, ‘‘On the procedure for compiling the perspective plan of preparation of the national economy for defense,’’ February 20, 1928). 17. For example, GARF, 8418/1/61: 5–6 (Gosplan, letter accompanying draft, ‘‘Directive on accounting for the needs of defense in working out c[ontrol] f[igures] for 1927/ 28’’); GARF, 8418/3/150: 8–46 (Gosplan report, ‘‘Military industry and the military products of civilian industry in the c[ontrol] f[igures] for 1929/30 (capital construction and development of capacities),’’ October 1, 1929). 18. GARF, 8418/22/631: 2–4 (chief of Red Army general staff Shaposhnikov and military commissar of the general staff Zaporozhets, letter to Defense Committee secretary Basilevich, April 21, 1938). 19. GARF, 8418/3/150: 8–46 (Gosplan report, ‘‘Military industry and the military products of civilian industry in the c[ontrol] f[igures] for 1929/30 (capital construction and development of capacities),’’ October 1, 1929). 20. GARF, 8418/3/93: 12 (deputy minister of transport, letter to the secretary of the STO executive session, June 21, 1929). 21. GARF, 8418/3/93: 20 (Gosplan, letter to the STO executive session, August 21, 1929). 22. GARF, 8418/3/93: 1 (secretary of the STO executive session Appoga, letter to Gosplan, the defense ministry, VNSKh, and other departments, August 23, 1929). 23. GARF, 8418/9/198: 1–2 (STO resolution no. K-11s, ‘‘On the procedure of financing and material provision of defense measures in 1934,’’ January 15, 1934). 24. GARF, 8418/3/150: 8–46 (Gosplan report, ‘‘Military industry and the military products of civilian industry in the c[ontrol] f[igures] for 1929/30 (capital construction and development of capacities),’’ October 1, 1929). 25. GARF, 8418/3/150: 5 (Presidium of VNSKh, letter to the STO executive session, ‘‘On the c[ontrol] f[igures] for defense work,’’ August 9, 1929). 26. See the process of assigning funds to the defense ministry for military orders in 1927/28: GARF, 8418/1/113: 21–22 (ministry of finance, letter to the STO executive session, October 1, 1927).

112

Andrei Markevich

27. GARF, 8418/8/9: 3–6 (deputy Gosplan chairman Mezhlauk, letter to prime minister Molotov, ‘‘On the procedure for confirming annual and quarterly plans for military industry,’’ July 31, 1933). 28. GARF, 8418/22/463: 5–7 (Defense Committee resolution no. 29ss, ‘‘On the plan of production of the ministry of defense industry in 1938,’’ February 20, 1938). 29. For example, RGAE, 7515/1/156: 200–332 (ministry of defense industry sixth chief administration, correspondence with factories and with the ministry leadership concerning the compilation of plans of the sixth chief administration and its enterprises for the third and fourth quarters of 1937); RGAE, 8157/1/389: 1–236 (ministry of armament, correspondence with Gosplan concerning the compilation of plans of the ministry of armament, its chief administrations, and enterprises for the second, third, and fourth quarters of 1939 and the annual plan for 1939). 30. GARF, 8418/1/4: 1 (STO resolution, ‘‘On the procedure of planning the military industry,’’ October 5, 1927). 31. Strictly speaking the army and navy were not the only purchasers of weapons; the NKVD also bought military equipment for its internal security troops. NKVD purchases were relatively unimportant, however: its planned orders for 1938 amounted to 128 million rubles compared with 9.9 billion rubles for the defense ministry. See GARF, 8418/22/463: 5–7 (Defense Committee resolution no. 29ss, ‘‘On the plan of production of the ministry of the defense industry for 1938,’’ February 20, 1938). 32. GARF, 8418/25/544: 121 (Defense Committee staff, note, ‘‘On the financial department,’’ not dated or signed, but 1939). 33. GARF, 8418/24/2: 7–9 (deputy head of the chief naval staff Ivanov and military commissar of the chief naval staff Nikolaev, letter to Defense Committee secretary Safonov, February 4, 1940). 34. GARF, 8418/24/2: 44–45 (chief of the Red Army general staff Shaposhnikov, acting military commissar of the general staff Gusev, and chief of the armament department of the general staff Ermolin, letter to Defense Committee secretary Safonov, January 26, 1940). 35. For example, GARF, 8418/23/91: 83–89 (minutes of meeting in the ministry of transport central military department, October 27, 1939). 36. GARF, 8418/22/463: 134–174 (table of disagreements in defense ministry orders for 1938). 37. GARF, 8418/22/463: 73–76 (defense industry minister Kaganovich, letter to deputy prime minister Chubar’, February 2, 1938). 38. GARF, 8418/25/544: 125 (Defense Committee staff, note, ‘‘On the financial department,’’ not dated or signed, but 1939). 39. GARF, 8418/3/111: 2 (VSNKh, letter to the STO executive session, ‘‘On disagreements with the defense ministry on the bulletin of compulsory orders for 1929/30,’’ June 11, 1929). 40. GARF, 8418/24/2: 44 (chief of the Red Army general staff Shaposhnikov, acting military commissar of the general staff Gusev, and chief of the armament department of the general staff Ermolin, letter to Defense Committee secretary Safonov, January 26, 1940). 41. GARF, 8418/23/732: 145 (Defense Committee staff official Ivanov, aide-mémoire, ‘‘Procedure for working out the plan of current military orders for 1940,’’ November 2, 1939).

Planning the Supply of Weapons

113

42. GARF, 8418/4/39: 1 (prime minister and STO chairman Rykov, letter to defense minister Voroshilov and Kuibyshev [for VSNKh], February 26, 1930). 43. Having approved the budgetary control figures of the defense ministry for 1927/ 28, for example, the STO executive session resolved to refer them upward ‘‘for confirmation by the instantsiia’’—i.e., by the Politburo (GARF, 8418/1/113: 1 [the STO executive session, excerpt from minute no. 10, October 8, 1927]). In February 1938 prime minister Molotov requested the Politburo to confirm the defense, navy, and NKVD order for armament and military equipment (GARF, 8418/22/463: 122–126 [Molotov, draft memorandum to the Politburo, no date]). 44. GARF, 8418/24/2: 7–9 (deputy head of the chief naval staff Ivanov and military commissar of the chief naval staff Nikolaev, letter to Defense Committee secretary Safonov, February 4, 1940). 45. GARF, 8418/8/9: 4–5 (Gosplan deputy chairman Mezhlauk to prime minister Molotov, ‘‘On the procedure for confirming annual and quarterly plans in defense industry,’’ letter dated July 31, 1933). 46. GARF, 8418/8/9: 7 (Sovnarkom draft decree, ‘‘On the procedure for confirmation of annual and quarterly plans for military industry’’). 47. GARF, 8418/8/9: 4 (Gosplan deputy chairman Mezhlauk, letter to prime minister Molotov, ‘‘On the procedure for confirming annual and quarterly plans in defense industry,’’ July 31, 1933). 48. GARF, 8418/22/463: 5–7 (Defense Committee resolution no. 29ss, ‘‘On the plan of production of the ministry of defense industry in 1938,’’ February 20, 1938). 49. GARF, 8418/23/345: 145 (defense industry minister Kaganovich, letter to the Defense Committee requesting approval of the defense industry ministry plan for 1939, January 10, 1939). 50. For example, RGAE, 8157/1/1258: 3–98; 8157/1/1259: 5–119; 8157/1/1260: 4– 121; 8157/1/1261: 4–121 (minister of armament, decrees approving various plans of the ministry of armament for 1949). On the confirmation of postwar plans of civilian ministries, see Markevich (2004). 51. GARF, 8418/24/2: 7–9 (deputy head of the chief naval staff Ivanov and military commissar of the chief naval staff Nikolaev, letter to Defense Committee secretary Safonov, February 4, 1940). 52. RGAE, 8183/1/146: 33ob–34 (condensed transcript of remarks of Kharitonovich, planning official of the defense industry ministry second [shipbuilding] chief administration, at a meeting of the second chief administration party group of activists to discuss the outcomes of the February central committee plenum, April 11—13, 1937). 53. The Molotov armament factory never received its 1939 first-quarter plan for labor (Hoover/RGANI, 6/2/250: 31 [KPK plenipotentiary for the Khabarovsk region, information to KPK chairman Andreev, ‘‘On leadership and support of the Molotov factory from the ministry of armament and [its] chief administration,’’ November 21, 1939]). 54. RGAE, 8183/1/146: 34ob (condensed transcript of remarks of Kharitonovich, planning official of the defense industry ministry second [shipbuilding] chief administration, at a meeting of the second chief administration party group of activists to discuss the outcomes of the February central committee plenum, April 11–13, 1937). 55. RGAE, 8183/1/146: 35 (condensed transcript of remarks of Kharitonovich, planning official of the defense industry ministry second [shipbuilding] chief administration,

114

Andrei Markevich

at a meeting of the second chief administration party group of activists to discuss the outcomes of the February central committee plenum, April 11–13, 1937). 56. Hoover/RGASPI, 77/3/22: 1 (central committee, draft closed letter, ‘‘On defects in the city and district party committees’ leadership of work in industry and measures to root out these defects,’’ no date, but 1940). 57. GARF, 8418/8/9: 5 (Gosplan deputy chairman Mezhlauk, letter to prime minister Molotov, ‘‘On the procedure for confirming annual and quarterly plans in defense industry,’’ July 31, 1933). 58. For example, RGAE, 8157/1/134: 44–47 (model contract between the defense ministry and the ministry of armament for 1940). 59. RGAE, 8157/1/134: 67 (head of the chief inspectorate Svistunov to minister Vannikov, ‘‘On the conclusion of contracts for 1940 by [the central administration for material supply],’’ December 26, 1939). 60. The words cited are from RGAE, 8157/1/134: 48 (Sovnarkom resolution no. 1989, ‘‘On the conclusion of contracts for 1940,’’ December 5, 1939). 61. GARF, 5446/15à/1101: 1 (Sovnarkom decree no. 561/93s, March 18, 1934). 62. GARF, 5446/18à/893: 1 (Sovnarkom, letter to Gosarbitrazh, December 16, 1936). 63. RGAE, 8157/1/134: 24 (Ministry of Armament decree no. 264, ‘‘On preparation for the conclusion of contracts for 1940,’’ September 16, 1939). 64. RGAE, 8157/1/134: 49–50 (Sovnarkom resolution no. 1648, ‘‘On the conclusion of contracts for 1940,’’ October 8, 1939). 65. RGAE, 8157/1/134: 25–26 (Ministry of Armament decree no. 297, ‘‘On the conclusion of contracts for 1940,’’ October 17, 1939). 66. RGAE, 8157/1/134: 67–70 (head of the chief inspectorate Svistunov to minister Vannikov, ‘‘On the conclusion of contracts for 1940 by [the central administration for material supply],’’ December 26, 1939). 67. RGAE, 8157/1/134: 21–23 (ministry of armament chief arbiter Beshtau, memorandum on the course of fulfillment of Sovnarkom resolution no. 1648, ‘‘On the conclusion of contracts for 1940’’ [October 8, 1939], to the ministry of armament collegium, December 25, 1939). 68. RGAE, 8157/1/134: 3–4 (ministry of armament collegium, resolution, ‘‘On the course of fulfillment of Sovnarkom resolution no. 1648 of October 8, 1939, and ministry of armament decree no. 297 of October 17, 1939, ‘On the conclusion of contracts for 1940’ ’’). 69. At a meeting that he convened to discuss the tank industry on March 23, 1936, Leningrad party boss Zhdanov remarked: ‘‘A contract is a good thing, but comrade Ruda [a Leningrad tank factory director] is saying: Until I get an assignment from the chief administration, I won’t sign a contract’’ (Hoover/RGASPI, 77/1/826: 11–12). 70. RGAE, 8157/1/134: 51 (deputy head of the ministry of armament first chief administration Zak to minister of armament Vannikov, December 27, 1939) and 60–61 (head of the ministry of armament sixth chief administration Eremin to minister of armament Vannikov, December 22, 1939). 71. GARF, 8418/22/463: 46–63 (minister of defense industry Kaganovich, letter to Defense Committee chairman Molotov asking for a reduction in the defense industry plan of gross output for 1938 with a list of items to be removed, December 15, 1938).

Planning the Supply of Weapons

115

72. GARF, 8418/22/463: 45 (Defense Committee staff, note to Molotov on the request of the ministry of defense industry to reduce the defense industry plan of gross output for 1938, December 20, 1938). 73. Hoover/RGANI, 6/2/250: 24 (KPK plenipotentiary for the Khabarovsk region, note by cable, ‘‘On account of the requirements of the first and second Far Eastern RedBanner armies for repair of artillery and products of the Red Army vehicle and armor administration by the Molotov no. 106 armament factory,’’ to KPK chairman Andreev). 74. RGAE, 7515/1/404: 46–53 (acting head of the Red Army artillery administration Savchenko, letter to defense industry minister Kaganovich, March 23, 1938). 75. GARF, 8418/3/48: 1 (excerpt from minute no. 32 of the STO executive session, ‘‘On strengthening the Red Army artillery and amendment of the five-year plan of orders for artillery,’’ April 16, 1929; GARF, 8418/3/76: 1–2 (STO executive session resolution, ‘‘On Red Army construction in the second period of the five-year plan and the five-year plan of special preparation of the country for defense,’’ May 27, 1929). 76. GARF, 8418/3/76: 1–2 (STO executive session resolution, ‘‘On Red Army construction in the second period of the five-year plan and the five-year plan of special preparation of the country for defense,’’ May 27, 1929). 77. RGAE, 7515/1/156: 196–199, 200, 301 (head of the defense industry ministry third chief administration Vannikov and head of the sixth department Sharanin, correspondence with the head of the defense industry ministry planning administration, July 26 and 27 and November 15, 1937). 78. RGAE, 7515/1/156: 332 (head of the defense industry ministry planning administration, letter to head of the defense industry ministry third chief administration Vannikov, December 22, 1937). 79. Hoover/RGANI, 6/2/250: 31 (KPK plenipotentiary for the Khabarovsk region, information to KPK chairman Andreev, ‘‘On leadership and support of the Molotov factory from the ministry of armament and [its] chief administration,’’ November 21, 1939). 80. RGAE, 7515/1/156: 220 (head of the defense industry ministry third chief administration Vannikov and head of the sixth department Sharanin, letter to the defense industry ministry planning administration on the causes of alteration of contracts for factory no. 2, August 14, 1937). 81. RGAE, 8183/1/146: 35 (condensed transcript of remarks of Kharitonovich, planning official of the defense industry ministry second [shipbuilding] chief administration, at a meeting of the second chief administration party group of activists to discuss the outcomes of the February central committee plenum, April 11–13, 1937). 82. Hoover/RGASPI, 77/1/826: 64 (remarks of Rubenov at Zhdanov’s meeting with party secretaries of numbered factories, December 23, 1936). 83. RGAE, 8183/1/146: 79ob (condensed transcript of remarks of Kudakov, official of the navy ministry, at a meeting of the second [shipbuilding] chief administration party group of activists to discuss the outcomes of the February central committee plenum, April 11–13, 1937). 84. RGAE, 8183/1/146: 34 (condensed transcript of remarks of Kharitonovich, planning official of the defense industry ministry second [shipbuilding] chief administration, at a meeting of the second chief administration party group of activists to discuss the outcomes of the February central committee plenum, April 11–13, 1937).

116

Andrei Markevich

85. RGAE, 8752/1/248: 1–40 (meeting of chief accountants of the tank industry ministry at the Uralmash factory, Sverdlovsk, October 26–30, 1942). 86. GARF, 8418/24/226: 11 (memorandum of Defense Committee official Aleksandrov to Molotov, May 29, 1940). 87. GARF, 5446/12à/16: 1 (STO resolution no. 1219, September 27, 1932). 88. GARF, 5446/16à/84: 1. 89. GARF, 8418/22/463: 45 (Defense Committee staff, note to Molotov on the request of the ministry of defense industry to reduce the defense industry plan of gross output for 1938, December 20, 1938). 90. GARF, 8418/22/463: 46 (defense industry minister Kaganovich, letter to Defense Committee chairman Molotov, December 15, 1938). 91. GARF, 8418/23/732: 86 (Defense Committee staff, note to Voznesenskii on alteration of the ministry of armament plan for 1939, December 2, 1939).

Published References Belova, Eugenia. 2005. ‘‘Legal Contract Enforcement in the Soviet Economy.’’ Comparative Economic Studies 47(2): 387–401. Harrison, Mark. 2003. ‘‘Soviet Industry and the Red Army Under Stalin: A MilitaryIndustrial Complex?’’ Les Cahiers du Monde russe 44(2–3): 323–42. Hayek, F. A. 1991. The Fatal Conceit: The Errors of Socialism. Chicago: University of Chicago Press. Ken, Oleg N. 2002. Mobilizatsionnoe planirovanie i politicheskie resheniia. Konets 1920-seredina 1930-kh gg. St. Petersburg: Evropeiskii Universitet v Sankt-Peterburge. Markevich, Andrei M. 2004. ‘‘Byla li sovetskaia ekonomika planovoi? Planirovanie v narkomatakh v 1930-e gg.’’ In Ekonomicheskaia istoriia. Ezhegodnik 2003, ed. Leonid I. Borodkin, 20–54. Moscow: Rosspen. Markevich, Andrei M. 2005a. ‘‘Soviet Planning Archives: The Files That Bergson Could Not See.’’ Comparative Economic Studies 47(2): 364–86. Markevich, Andrei M. 2005b. ‘‘Sovetskoe znachit nadezhnoe: voenpredy i problema kachestva v sovetskoi oboronnoi promyshlennosti.’’ In Ekonomicheskaia istoriia. Ezhegodnik 2005, ed. Leonid I. Borodkin, 364–406. Moscow: Rosspen. Mises, Ludwig von. 1981. Socialism: An Economic and Sociological Analysis. Indianapolis, IN: Liberty Fund. Samuelson, Lennart. 2000a. ‘‘The Red Army and Economic Planning.’’ In The Soviet Defence-Industry Complex from Stalin to Khrushchev, ed. John Barber and Mark Harrison, 47–69. Basingstoke (England): Macmillan. Samuelson, Lennart. 2000b. Plans for Stalin’s War Machine: Tukhachevskii and MilitaryEconomic Planning, 1925–41. London and Basingstoke: Macmillan. Samuel’son [Samuelson], Lennart. 2001. Krasnyi koloss. Stanovlenie sovetskogo voennopromyshlennogo kompleksa. 1921–1941. Moscow: AIRO-XX. Simonov, Nikolai S. 1996a. ‘‘Strengthen the Defence of the Land of the Soviets: The 1927 War Alarm and Its Consequences.’’ Europe-Asia Studies 48(8): 1355–64. Simonov, Nikolai S. 1996b. Voenno-promyshlennyi kompleks SSSR v 1920–1950-e gody: tempy ekonomicheskogo rosta, struktura, organizatsiia proizvodstva i upravlenie. Moscow: Rosspen.

Planning the Supply of Weapons

117

Sokolov, Andrei K. 2004. ‘‘NEP i voennaia promyshlennost’ Sovetskoi Rossii.’’ In Ekonomicheskaia istoriia. Ezhegodnik 2003, ed. Leonid I. Borodkin, 95–117. Moscow: Rosspen. Stalin, I. V. 1937. Voprosy leninizma, 10th ed. Moscow: Gospolitizdat. Stone, David R. 2000. Hammer and Rifle: The Militarization of the Soviet Union, 1926– 1933. Lawrence: University Press of Kansas. Stone, David R. 2003. ‘‘Mobilisation and the Red Army’s Move into Civil Administration, 1925–31.’’ Kritika: Explorations in Russian and Eurasian History 4(2): 343–67. Zaleski, Eugène. 1980. Stalinist Planning for Economic Growth, 1933–1952. London and Basingstoke: Macmillan.

5

Planning for Mobilization: The 1930s r. w. davies

Mobilization planning involved the preparation of the economy for the event of a future war that was widely anticipated but might or might not be launched from any particular direction at any particular moment. The first part of this chapter describes the emergence of the mobilization system in the inter-war years: Future mobilization needs had to struggle for priority over the current needs of defense and industrialization. The second part examines the results of the mobilization plans in practice as they affected major aspects of the economy: the armaments industries, the civilian sector, the localities, and the workforce.

Policies, Plans, Procedures During the first two five-year plans (1928 to 1937) the proportion of the Soviet national income devoted to investment approximately doubled. Investment was concentrated on the capital goods industries, with strong emphasis on fuel and power, iron and steel, and machine building. The pattern and speed of industrialization were dictated to a considerable extent by the strongly held Soviet view that war between Soviet socialism and the capitalist world was inevitable—and the emergence of a modern army and armaments industry capable of resisting the economically more advanced capitalist powers depended on the establishment of a large, modern capital goods industry. 118

Planning for Mobilization

119

In the short term the defense sector competed for resources with civilian investment and consumption. How far an immediate increase in defense expenditure was required depended on an assessment of the time available before the outbreak of war. If war was likely to come sooner, more investment should be allocated directly to investment in the armaments industries, to the production of armaments, and to the maintenance of the armed forces. The general assumption by the Soviet leaders that sooner or later the capitalist powers would attack was dramatically justified by the Nazi-German invasion of June 1941. But in the shorter term their analysis of the international position of the Soviet Union suffered from two major defects. First, it tended (except, ironically, in 1941) to overestimate the immediate military threat. Thus the Soviet Union was militarily weak and internationally isolated in 1927, following the defeat of the revolution in both Europe and Asia. But the claim by both the majority of the Politburo and the Left Opposition that there was an immediate danger of war was not justified. Second, both the political and the military leaders overestimated the strength of the enemy. In August 1940, for example, the general staff claimed that Germany had 13,900 aircraft, of which 12,000 were directed against the Soviet western frontiers (Naumov 1998, 1:181–82; the document is signed by marshals S. K. Timoshenko and B. M. Shaposhnikov). In fact, there were only some 4,700 German aircraft in the Soviet theater of war on June 22, 1941 (Meliia 2004: 226). These arrangements considerably overestimated the peacetime strength of the potential enemy. But during the war Germany, supported by the industry of most of the rest of Europe, was able to produce far more armaments than either the German or the Soviet military planners had assumed. Germany did not confront the Soviet Union with 12,000 aircraft in 1941, but in 1944 it produced as many as 32,000 aircraft. In practice, Soviet assumptions about the military strength that had to be mobilized to overcome the enemy proved not to be unrealistic. Nevertheless, the current production of some weapons, particularly tanks, in the 1930s was greater than was required by the immediate military threat, particularly given the rapidity of technical advance. By 1941, much of the large Soviet stock of armaments was already out of date. This ‘‘overproduction’’ has led some historians to assume that the Soviet economy was already ‘‘militarized’’ in the mid-1930s. David R. Stone (2000: 216) wrote in his otherwise excellent book on the first five-year plan that Stalin ‘‘saw no need to slow the growth of the Red Army for the sake of the overall economy’’ and that ‘‘military planners never truly saw the need to decide between devoting resources to expanding production capacity and increasing current production.’’ This is a mistaken view. In practice, far fewer resources were allocated to defense than would have been required if war was imminent. After 1927,

120

R. W. Davies

only a moderate increase in defense expenditure took place. The share of defense in investment and production increased substantially in 1931 and 1932, following the increased threat from Japan, and still further from 1936 onward, partly in response to the belief that the danger of war had increased, and also because a technological revolution in the armaments production of all the major powers made defense more expensive. But it was not until the last couple of years before the German invasion that the economy began to be transferred to a war footing. During the 1930s, the conflict between the short-term needs of defense and the goal of rapid industrialization was partly resolved by establishing facilities that were not immediately used for defense purposes but could be rapidly transferred to defense when mobilization was declared. Specialized armaments factories were developed with spare capacity temporarily used for civilian production, or kept in reserve. Civilian factories were designed or adapted so that they could be transferred to the production of armaments or their components. In the inter-war years, the Soviet Union devoted thought and resources to preparation for rapid conversion to wartime conditions to a greater extent than any other country, with the possible exception of Germany on the eve of the war. From 1927 onward, the military prepared mobilization claims (zaiavki) to the government that set out the military units and armaments production that should be available when war broke out, usually expressed in terms of requirements for ‘‘one year of war’’ or ‘‘the first year of war.’’ The government then approved national mobilization requirements (zadaniia), usually substantially reduced as compared with the military claim. On this basis every unit of the economy adopted mobilization plans (mobplany). To operate this system a network of mobilization departments was established, under conditions of extreme secrecy, at every level of the economic hierarchy, including major civilian factories. The mobplan compiled in 1927 was replaced by ‘‘MV-10’’ at the end of 1930. MV-10 was in turn revised within a few months, and then gave way to ‘‘MV-12’’ in 1931 (Samuelson 2000: 121–23 and 126–34; see Table 5.1). In 1933 the Red Army prepared a new ‘‘mobilization claim,’’ M-18, for 1938. The corresponding plans for industry and the other civilian ministries were reduced by the Defense Commission, a high-level agency attached to both the Politburo and Sovnarkom, and were then known as ‘‘M-3’’ or ‘‘MP-3.’’ At first each mobplan was more ambitious than its predecessor (see Table 5.1, cols. 1 to 4). These capacities for production stipulated in the mobilization claims and requirements, and in the mobplans, were vastly in excess of actual peacetime current production. According to the defense ministry claim of June 1933, for

Table 5.1. Production of major weapons: Planned capacity for one year of war, 1927–1938 (units)

Artillery Shells (millions) Rifles (thousands) Machine guns (thousands) Rifle cartridges (millions) Aircraft Aircraft engines Tanks Ammunition powder propellants (thousands of tons) Explosives (thousands of tons) Poison gas (thousands of tons)

MP-≥≥ (for Jan. ∞, ∞Ω≥∂)

Defense Comm. variant, ∞Ω≥≥ (for Jan. ∞, ∞Ω≥∫)

Gosplan variant, ∞Ω≥≥ (for Jan. ∞, ∞Ω≥∫)

Government variant, ∞Ω≥π (for Jan. ∞, ∞Ω≥∫)

(∂) ≥≤,∂∑Ω ∞∏∑ ≥,≠≠≠ ≤≤≤ ∞≥,≠≠≠ ≥∞,≤≠≠ ∫≤,∑≠≠ ∂≤,∑≠≠

(∑) ≤Ω,π∏∏ Ω≤.∫ ≤,∂≠≠ ≤∞∞ Ω,≠≠≠ ∞∫,≤≠≠ ∂π,∑≠≠ ∂≤,≠≠≠

(∏) ≥Ω,∞∫≠

Government variant, ∞Ω≥π (for Jan. ∞, ∞Ω∂≤)

MP-∞ of July ≤Ω, ∞Ω≥Ω (for ∞Ω≥Ω)

Peak production in wartime

(∫) ∑∞,∫∞∫ ≤≥≥ ≤,π∂∞

≤∂,ΩΩ≠e

(Ω) ∞≥≠,≠≠≠ ∞∫∂ ∂,≠∂Ω ∂∑Ω π,∂≠∏ ≥≥,≤≠≠e ∑≤,ππ∏ ≤Ω,≠≠≠

∞Ω≤π

∞Ω≥≠ MV-∞≠

(∞) ≥,π∏≥ ≥π.Ω ∞,≠≠≠ ∂≥.∑ ... ≤,Ω≠∑

(≤) ∞≤,∏∞≠ ∂≠ ∞∑π∑ Ω≠.∏ ... π,≠Ω∫

∞∑≠

≤≠,≠≠≠

...

...

...

≤≠∞

∞≤∑

...

...

≤∫∑

∞≤π

...

...

...

∞∫≤

∞≥∑

...

...

∏∞∏

...

≤π.≤

π∑.∑

...

≤≤∑

∞∂π

...

...

≤≤∫

...

(≥) ≤∂,∂∞≠ ∏≠ ≤,≠≠≠ ∞≤≠ ≥,≠≠≠ ∞∑,∫∏≠c ∞∑,∞∑≠ ≥≠,∫≠≠d

≤,≥≠≠ ≤∑≠ ≤≠,∏≠≠ ∂≤,≥≠≠ ≥∑,∂≠≠

(π) ∞∞Ω,∏≠≠ ∞∑≠b ∑,≠≠≠ ∂∑≠ ∞≥,≠≠≠b ≥∂,∞≠≠a ππ,≠≠≠a ∏≠,ππ∑

∞∏,∏∂≠ ≤π,≤∏≠

January 1, 1940. January 1, 1939. c Includes 8,730 trainers. d Includes 15,000 pocket tanks and armored cars. e Includes 5,700 armored cars. a

b

Sources: Cols. 1 and 2: Samuelson (2000: 122). Col. 3: Simonov (1996: 117–18), citing the archives (adopted by the ministry of heavy industry on the basis of a STO decision, November 1933). Cols. 4 and 5: RGAE, 4372/91/1448: 181–180 (Botner and Kolesinskii of Gosplan defense sector to Kuibyshev and Mezhlauk), dated April 4, 1933. Cols. 6 and 7: RGAE, 4372/91/3002: 139–132 (Smirnov, Gosplan, to Molotov), June 5, 1937. Col. 8: Simonov (1996: 120), citing the archives. Col. 9: Table 5.4.

122

R. W. Davies

Table 5.2. Planned capacities for one year of war, 1931–1933 (units)

Aircraft Aircraft engines Tanks Shells (millions)

Capacity

Date to be achieved

≥∑,≠≠≠ ∫≥,∑≠≠ ∂≠,≠≠≠ ∞∞π

January ∞, ∞Ω≥∑ January ∞, ∞Ω≥∑ By end ∞Ω≥≥ (sic) By end ∞Ω≥∂

Source: RGAE, 4372/91/1448: 181–179 (report to Kuibyshev and Mezhlauk, then head and deputy head of Gosplan, from Botner and Kolesinskii). The tank figure was reduced during 1933 to 38,400, and the shell figure to 76 million, to be achieved by 1935 (Naumov 1998, 1:181–82).

example, peacetime production in 1937 would amount to only 6 percent of capacity in the case of tanks, and 4 percent in the case of shells.∞ The mobplans also provided for the development of military facilities apart from the capacity to produce armaments. Much attention was devoted to the transport network, which had to be capable of moving supplies, armaments, and troops in war conditions. Plans were also drawn up for the greatly increased quantities of food and clothing that would be supplied to the military in the event of war. Mobilization planning also extended in principle, and irregularly in practice, to the local soviets. The mobplans also provided for the establishment of stocks of weapons, metals, grain, and other foods, and clothing and consumer goods under the Committee of Reserves, ready for use in the event of war (Davies 1996: 115). For all these purposes investment allocations were provided by the state budget. During the second five-year plan the revision of the mobplan was largely an ad hoc affair. In 1933 ferocious discussions took place about the military aspects of the plan, but no five-year plan for defense was adopted by Sovnarkom or the defense ministry, apart from general figures for investment, armaments orders, and maintenance.≤ And no five-year plan was approved for the armaments industries as a whole or for individual armaments such as aircraft and chemicals.≥ Instead, the armaments industries worked within the framework for M-3, supplemented by specific plans approved by the Defense Commission. The plans stipulated both current production and capacity for the first year of war. Table 5.2 shows the plans for capacity, approved as early as 1931, that were still notionally operative in 1933. Between 1933 and 1937 M-3 underwent many revisions, at first to cope with the failure to achieve the original targets on time, and then to incorporate the substantial expansion required by the looming danger of war. A report prepared by A. I. Egorov, chief of the general staff, on behalf of the Red Army

Planning for Mobilization

123

Table 5.3. The Red Army requirement for one year of war, 1935 (units) In ∞Ω≥∑

Tanks Aircraft Shells (millions) a

In ∞Ω≥Ω

Red Army claim

Approved by government

Red Army claim

Approved by government

∂≠,≠≠≠ ∞∂,∑≠≠ ∞≤∂.≤

≥∑,∫≠≠ ∞∞,π≠≠ π∏

∑≠,≠≠≠ ∂∑,≠≠≠ ≤≤∑

... ≤∏,∏∑≠ ∞≠≠a

To be achieved in 1936.

Source: Ken (2002: 290–91) (report dated April 14, 1935).

in April 1935, presented the position shown in Table 5.3. The revisions normally appeared in the form of specific plans prepared by the Defense Commission and approved by STO. Thus, presumably following Egorov’s claim, the Commission resolved on April 25, 1935, that by 1938 capacity should reach 27,000 for aircraft and 79,000 for engines. These figures for 1938 were lower than those approved in 1931 to be achieved in 1935.∂ Two months later, on June 17, 1935, the Commission decided that shell capacity should amount to 150 million by January 1, 1939, as compared with the 34 million capacity expected on January 1, 1936, and capacity for propellant powders should amount to 220,000 tons by January 1, 1939. The situation was extremely anomalous in the case of artillery pieces, rifles, and tanks: the plans had not been revised since 1933, but the date by which the targets should have been reached had passed.∑ Further revisions in the capacity plans were made by May–June 1937 (see cols. 6 and 7 of Table 5.1). These revisions did not greatly change the targets already adopted by 1935. However, as they were expressed in crude physical units, they did not reveal the large technical changes that were taking place in the aircraft, tank, and other industries. The tank of 1938 was a more complex weapon than the tank of 1932, and required the use of more complex materials and components. In value terms, the new weapons were at first much more costly than their predecessors. Thus in 1936 a T-26 tank cost 71,500 rubles, but the T-46 that was intended to replace it cost 200,000 rubles.∏ However, experience and economies of scale reduced the initial costs of new weapons. The price of the more powerful T-34, 270,000 rubles in 1941, was reduced to 135,000 rubles in 1943 (Harrison 1996: 181). The extent to which the mobplans disaggregated weapons capacity into specific weapons varied considerably from plan to plan. The military complained on numerous occasions about the absence of an

124

R. W. Davies

up-to-date mobilization plan for industry. On November 25, 1935, marshal M.N. Tukhachevskii sent a firm memorandum to Stalin: ‘‘The question of preparation for the mobilization of industry has not yet been discussed at a sitting of the Defense Commission. However, the state of industrial mobilization preparedness is extremely threatening.’’ He pointed out, for example, that the M-3 mobplan for the ministry of heavy industry stipulated a capacity of only 65 million shells in 1935, while the 1933 decree of the Defense Commission authorized a capacity of 76 million. Moreover, actual production of shells in 1935 amounted so far to only about one million, as compared with the military order of 6.3 million, and this was less than 10 percent of the required mobilization capacity.π According to Tukhachevskii, mobilization readiness for the production of artillery pieces and other weapons was equally unsatisfactory. There was no proper coordination between the manufacture of artillery and explosives, which were located under different chief administrations of the ministry of heavy industry: ‘‘If the present situation drags on, in the event of war the army will eventually meet with very serious disruptions.’’∫ Following this demarche, in December 1935 the defense ministry presented a report and draft decree to the Defense Commission on the system for elaborating mobplans—but no decision was adopted.Ω In August 1936, Egorov wrote to the Defense Commission insisting that new mobplans should be adopted simultaneously for the whole economy.∞≠ In the following month, a further memorandum from the general staff pointed out that the army had now itself adopted ‘‘MP-4’’ (Mobilization Plan No. 4), and insisted that it should come into force in the civilian ministries by October 1.∞∞ At the end of the year the issue was taken up by Voroshilov in a trenchant memorandum to Molotov. He complained that as a result of the failure of the ministries of heavy and light industry to prepare MP-4, ‘‘the existing position is a very serious threat to the mobilization readiness of the army and of the country as a whole.’’∞≤ While no new mobplan was adopted for heavy industry, where most armaments were produced, other ministries had already begun to prepare new plans. On May 25, 1935, STO instructed ministers to prepare MP-4, covering the years 1935 and 1936. Most ministries obliged in the course of the next few months.∞≥ In May 1937 a Gosplan report to Molotov acknowledged that the ministries, apart from those responsible for the armaments industries, had prepared MP-4s, but complained that a number of sections of the plans, including supplies, labor, fuel, and power, had not been completed. Moreover, the plans were now already out of date, but Sovnarkom had made no decision about a fresh mobplan. In its absence the army had sent out new mobilization requirements to the ministries so that they could work out new mobplans for

Planning for Mobilization

125

1937 and 1938.∞∂ At this time the mobilization department of an armaments factory wrote to Gosplan complaining of the anomaly that the armaments industry was still using ‘‘revised M-3,’’ while light industry and part of heavy industry were using MP-4, and the army had moved on to ‘‘revised MP-4.’’∞∑ In 1937 criticism of the effectiveness of the mobplans was increasingly entangled with charges of wrecking. In August G. I. Smirnov, appointed head of Gosplan in February 1937, wrote to Molotov complaining, ‘‘The preparation of the economy for defense is one of the spheres of activity which has suffered from wrecking activity.’’ Both M-3 and M-4 were out of date, and lacked a plan for the supply of metal and other materials in the event of war. Smirnov demanded that the ministry of heavy industry should immediately prepare a report and decree for Sovnarkom on a mobplan for 1938.∞∏ In December 1936 most of the defense industry had been split away from heavy industry. Rukhimovich, the first minister of the defense industry, was arrested on October 16, 1937, and Smirnov himself was arrested on the following day. In February 1938 Rukhimovich, in his testimony to the NKVD, repeated the familiar story that M-3 was out of date: the rapid mechanization of the army, and its growth in size, meant that present requirements were more than double those set out in M-3. He now improbably attributed this situation to the deliberate wrecking activities of Piatakov, Smirnov, and himself.∞π By this time Rukhimovich’s testimony was itself out of date. In November 1937 Voroshilov sent a long report to Stalin on the reorganization and development of the Red Army in 1938 to 1942, including a decree for the Defense Committee ‘‘On the mobilization of the Red Army for 1938–1939.’’ The decree stated that the new army mobplan should be known as ‘‘Mobplan 22,’’ and the plan for the supply ministries should be known as ‘‘Mobplan no. 8.’’ Voroshilov’s report indicated the number of artillery pieces, tanks, aircraft, and vehicles that should be available for immediate mobilization in 1938, and also in the first year of war, and compared these figures with the existing stock of weapons.∞∫ The arrangements for mobilization planning always suffered from the weakness that they were not in practice controlled by any central authority. Until the end of 1936 practical decisions about the production of weapons, and the capacity for them, rested with the military and mobilization administration of the ministry of heavy industry, which managed the trusts concerned with the different branches of the arms industry.∞Ω More general decisions, and the coordination of the various defense activities (insofar as it existed) rested in practice with the defense sector of Gosplan and the Defense Commission. The defense sector was a well-informed and influential agency, staffed by experienced planners and military men, but like the rest of Gosplan

126

R. W. Davies

it lacked executive power. The Defense Commission certainly carried the authority of the Politburo and Sovnarkom—and of Molotov and Stalin. But at least until 1937 mobilization planning was never high on its list of priorities. The secret resolutions of the Politburo, Sovnarkom, and the Defense Commission itself were rarely concerned with the problems of mobilization. The lack of a controlling hand in mobilization planning was a frequent subject of complaint from those engaged in this activity. At the end of 1932 Efimov, responsible for mobilization planning in the general staff, complained in a memorandum to Molotov that Gosplan had failed to issue directives on the national economy in war conditions. The ministries did prepare directives, but ‘‘only the most insignificant part has been pushed through with great difficulty.’’ Efimov claimed that I.P. Pavlunovskii, head of the military and mobilization administration of the ministry of heavy industry, who had the ear of Ordzhonikidze, had resisted the establishment of a high-level commission to examine the problem because he was ‘‘afraid that the commission . . . [would] compel him to work on the mobilization of industry,’’ and appealed to Molotov for support.≤≠ Four years later, in December 1936, Buzinkov, an army officer concerned with mobilization planning in the artillery administration of the Red Army, made similar complaints in a memorandum to Efimov, who was now head of the artillery administration, entitled ‘‘Considerations on the principles of organization of the preparation of industry for serving the needs of the Red Army in time of war in connection with the formation of the people’s commissariat of the defense industry.’’ According to Buzinkov, ‘‘During the whole 8 or 9 years of the work of preparing industry the absence has been felt of a unified methodology and of an agency with ‘the leading role’ in all this work.’’ As a result, ‘‘interdepartmental struggle’’ had prevented normal development: ‘‘In practice serious achievements in the preparation of the mobilization of industry have been completely lacking. This situation cannot continue.’’ Buzinkov called for the establishment of a special agency attached to the defense ministry with the authority to resolve inter-ministerial disputes. Three months later Buzinkov, who had been transferred to the mobilization administration of the new ministry of the defense industry, complained in a further memorandum that since the reorganization ‘‘mobilization work has remained in a scattered state as before, and is perhaps getting worse.’’ He now called for the establishment of something like ‘‘an economic general staff, leading and influencing work in the economy as a whole, both in the army and in industry.’’ At present ‘‘various decisive issues are not worked through and no one is responsible for their content.’’ The proposal for a military-economic general staff with overriding powers revived the doctrines unsuccessfully proposed by the leading military thinker Svechin a decade before.≤∞

Planning for Mobilization

127

After another three months had elapsed, the persistent Buzinkov sent yet another memorandum, this time to Andreev in the central committee, complaining that his previous memorandum had been ignored both by the army and by the defense industry: ‘‘They evidently did not read it as the original does not have any marks on it, and I found it was with comrade Levin almost 2 months after it was sent.’’≤≤ By this time several factors made it essential to establish firm arrangements for the preparation and enforcement of the mobplan. The defense industry broadly defined had been divided between ministries for heavy industry and the defense industry in December 1936, and in August 1937 a new ministry for machine building was split off from heavy industry (see Figure 3.4). The upheaval of the purges removed the key people concerned with defense planning and disrupted administrative arrangements generally. Simultaneously, both the armaments industries and the size of the armed forces increased rapidly. In April 1937 STO and the Defense Commission were replaced by an equally high-level Defense Committee, nominally attached to Sovnarkom, and chaired by Molotov (see Figure 3.2).≤≥ During 1937 Gosplan’s defense sector continued to bear the main responsibility for coordinating the mobilization plans. But on December 7, 1937, it was abolished, and its functions were transferred to the Defense Committee (Khlevniuk et al. 1995: 34); Gosplan retained a small mobilization sector. The first department of the committee was the ‘‘mobilization and planning department,’’ divided into four sectors responsible for mobilization capacity and its prospects, and for the mobilization plans of the different branches of the defense industry and of the civilian economy.≤∂ It had a small staff of nineteen.≤∑ Following the abolition of the Gosplan sector of defense, the Defense Committee reviewed the state of mobilization planning in the ministries. In February 1938, a memorandum evidently written by G.D. Bazilevich, the secretary of the committee, reported to Molotov, ‘‘The present mobilization plan is in an extremely bad state and threatens the danger of the disruption of industrial mobilization. In most ministries, although mobplans exist on paper, they are unrealistic, contain wrecking proposals, are out of date, and not backed up by resources. In the ministries of the defense, machine-building, and heavy industries, following the division of the ministry of heavy industry, there are no mobplans which include the whole system of industry of these ministries.’’ The memorandum insisted that partial modifications of the mobplan had merely ‘‘increased the confusion.’’ A new mobplan must be prepared, but the plan ceilings or limits (limity) had not yet been approved by the government. These would in effect have to be based on the capital investment available in 1938. The memorandum warned that ‘‘the full cycle of work on the new

128

R. W. Davies

variant of the mobplan will take one year’’; even if the plan limits were approved immediately, the plan could not be completed until October 1 at the earliest.≤∏ Meanwhile, in the autumn of 1937, Molotov, in his capacity as chair of the Defense Committee, had instructed the army general staff to prepare a draft decree ‘‘On the system for elaborating mobilization plans.’’ The draft proposed that the defense ministry should prepare a general ‘‘mobilization claim,’’ the main plan limits of which would be approved by the Defense Committee. On the basis of these plan limits, the defense ministry should prepare a revised claim, disaggregated to the various ministries. The claim for all weapons would be sent to the ministry of defense industry, which would coordinate the work between its own factories, and, via the ministries concerned, with the factories of other ministries. This plan would be approved by the Defense Committee, as would the mobplans of other ministries for food, clothing, transport, and so on. The draft decree was approved by Voroshilov and by Mikhail Kaganovich, at this time minister of the defense industry. However, Lazar, Mikhail’s brother, who was far more senior in the political hierarchy, was at this time minister for heavy industry, responsible for the supply of fuel, metals, and chemicals to the armaments industries. He strongly objected to the excessive powers that the decree afforded to the ministries of defense and the defense industry, and demanded that each ministry should receive its mobilization plan direct from the Defense Committee. His proposal was rejected on the grounds that it would require a huge expansion in the staff of the mobilization departments of the two ministries, and would also exclude preliminary negotiation between the ministry of defense industry and the other ministries about what the latter would supply to the weapons program. The decree approved by the Defense Committee on April 8, 1938, retained the arrangement that the plan limits, and the lists of weapons and other items to be produced in one year of war, should be approved by the Committee, and that the ministry of defense industry would ‘‘prepare lists of enterprises of other collaborating ministries for approval by the Committee.’’ Similarly, the ministry of heavy industry would be responsible for metal, fuel, and power, and the ministries for the food industry and agricultural procurements would be jointly responsible for food and fodder.≤π A further decree, dated April 29, 1938, set out the new mobilization requirements for mobplan MP-10.≤∫ This was a time of great organizational uncertainty. A further decree on June 1, 1938, transferred mobilization planning to a separate ‘‘military industrial commission’’ attached to the Defense Committee, headed by Chubar’. Its importance was indicated by the decision that Chubar’ should be freed from

Planning for Mobilization

129

all his responsibilities as a deputy chair of Sovnarkom.≤Ω However, this was also the climax of the purges, and two weeks after his appointment, on June 16, 1938, Chubar’ was dismissed from his post and removed from the Politburo. He was replaced as head of the Military Industrial Commission by M. M. Kaganovich.≥≠ On the day of his dismissal Chubar’ sent a memorandum to Stalin, Molotov, and Voroshilov, with the poignant comment, ‘‘I will no doubt be unable to communicate with you personally.’’ Referring to the ‘‘mobilization requirement’’ approved by the Defense Committee for the supply of munitions in the year July 1, 1938, to July 1, 1939, in the event of mobilization, he described it as a ‘‘colossal task for industry,’’ which had met with ‘‘considerable resistance’’ in the supply ministries, whose own proposals did not meet the requirements: ‘‘There is no militant alarm yet in the ministries, but unstated thoughts that the mobilization requirement is extreme are obviously circulating behind the back of the leadership. It is impossible to explain the facts that even the ministries cannot find the strength in themselves to guarantee the mobilization requirements, except by lack of belief in the obligation to fulfill them.’’ He criticized industry for its ‘‘stubborn unwillingness to reveal and mobilize all the possibilities of our economy,’’ but also pointed out that the requirement would cost more than 100,000 million rubles in 1938 prices, even without shipbuilding—ten times as large as the current production plan for the armaments industries. This task could be achieved only by establishing an experienced staff in the army to deal with mobilization, and appointing full-time competent heads of the military departments of the ministries, training special engineers and skilled workers for factories that were particularly important for the mobilization requirement, and establishing mobilization commissions in the regional and republican soviets. All concerned must be equipped with the necessary textbooks and instructions: ‘‘There is a lot of talk in the ministries about wrecking in the preparation for mobilization, but no real elimination of the consequences of wrecking has occurred.’’≥∞ An undated and unsigned document in the files of the Defense Committee confirmed the figures cited by Chubar’. It compared the current military order for 1938, 10,579 million rubles, with the cost of the mobilization requirement for the year 1938/39, 95,195 million rubles (without shipbuilding). It pointed out that the mobilization plan called for the production in 1938/39 in the event of war of 6.8 times as many tanks, 2.29 times as many aircraft, 11.4 times as many shells, and 3.5 times as many artillery pieces as in the current order for 1938. Following the appointment of M. M. Kaganovich the military industrial commission was allocated a staff of seventy to eighty persons, far larger than

130

R. W. Davies

the staff of the previous mobilization department in the Defense Committee. The decree authorizing this establishment stated that the commission’s functions included ‘‘coordinating the mobilization-industrial plan with the national economic plan (jointly with the Mobsektor of Gosplan).’’≥≤ The new commission failed to complete the work of preparing mobplan MP-10, supposed to cover July 1, 1938, to June 30, 1939, by August 1938 as planned. This was primarily a matter not of administrative delays, but of the availability of resources to cope with the demands of the military. Memoranda prepared during 1938 strongly implied that the military mobilization requirement for 1938/39 could not be achieved. According to a later report, the Defense Committee, following representations from the commission, reduced the mobilization requirement of MP-10 by 25 to 30 percent. Simultaneous estimates of the availability of materials showed that only 70 percent of the aircraft requirement and 50 percent of the tank requirement were feasible.≥≥ A document prepared in October 1938 calculated the cost of the 1938/39 mobplan at 56,443 million rubles in 1938 prices as compared with the 1938 order of 10,826 million, and the previous estimate of 100,000 million rubles.≥∂ It is not clear whether the reduction is due to a recalculation, or to the reduction of the demands presented by the defense ministry earlier in the year. The reorganization of mobilization planning was not yet at an end. In January 1939 the ministry of defense industry was divided into four new ministries for the aviation industry, weapons, ammunition, and shipbuilding; the ministry for machine building was divided into three, and the ministry of heavy industry into six (see Figure 3.4). The unitary ministry of heavy industry of 1936 was thus divided into thirteen new ministries, each represented on Sovnarkom. This administrative upheaval solved the issue raised by L. Kaganovich a year previously. As the ministry of defense industry had been broken up, the coordination of arms production had to be undertaken either by the four new ministries separately or by the Defense Committee and its military industrial commission. Control was still further centralized on June 19, 1939, when the military industrial commission was again absorbed into the Defense Committee. Revision of the mobplans continued until the eve of the war. The military industrial commission and the Defense Committee approved a further mobplan, known as MP-1, in a series of decrees on April 29, September 12, and December 18, 1938, and July 29, 1939.≥∑ MP-1 was declared obsolete by the Defense Committee in June 1940. On July 25, 1940, a Sovnarkom decree affirmed that MP-1 was no longer in force, and called for the preparation of a mobplan for 1941 in time to enable the industrial enterprises to complete the preparation of their own mobplans by December 1, 1940 (Naumov 1998, 1:128–29; Gerasimov 1999: 5).≥∏ But on September 10, 1940, the Defense

Planning for Mobilization

131

Committee approved an interim ‘‘mobilization requirement’’ for the fourth quarter of 1940, which was known as MP-2. This listed all the main weapons and their components in considerable detail. It was dispatched to thirty-two ministries and other departments, which were required to allocate these requirements among their factories and ‘‘prepare the factories by October 1, 1940, for the delivery of arms in full.’’≥π Following this partial mobilization, which is discussed further below, in February 1941 the defense ministry and general staff jointly presented a ‘‘plan’’ or ‘‘scheme’’ for the mobilization of the Red Army in 1941 and 1941/42 to Stalin and Molotov, replacing the plan of 1938 to 1939 (Naumov 1998, 1:607–40 and 640–50). This was followed by the preparation of elaborate plans for 1941/42 in all the armament ministries. The mobplan for munitions was approved in June 1941 (see below), but no mobplan for armaments was accepted by the government before the German invasion on June 22 (Gerasimov 1999: 5). These plans, which were more specific and coherent than earlier mobplans, are discussed further below.

Mobilization Plans in Practice The documents prepared in Gosplan in the course of 1933 to 1937 implied that for most major armaments the capacity planned for 1938 in the Gosplan variant of the second five-year plan had been achieved. However, this was far short of the original requirement of the Defense Commission (compare cols. 4 and 5 of Table 5.1 with col. 6 of Table 5.4). Insufficient information is available to enable us to check how far these statements about the capacity achieved were realistic. The many complaints that little attention had been given to the mobplans by industry are strong evidence of the need for skepticism. However, in the case of some armaments—for example, guns and shells—the archives provide quite detailed accounts of the location and type of weapon to be produced in wartime.≥∫ Perhaps the most reliable Soviet estimates of available capacity were made in the course of 1940. Table 5.5 (cols. 1 and 2, estimated at the beginning of 1940) shows that the estimated capacity lagged considerably behind the requirements of MP-1 for the first year of war, except in the case of aircraft and rifles. Column 3 of the table shows the output planned for immediate mobilization in the autumn of 1940 for MP-2. This was a plan for immediate operation in the last quarter of 1940, and the estimate was even more modest than that made at the beginning of 1940. These figures are closer to the capacity estimated for January 1, 1936 (see Table 5.4). It is also relevant to note that at its peak actual armaments production, in spite of the occupation by Nazi Germany of the territory on which most

Table 5.4. Production of major weapons: Claimed actual capacity for one year of war, 1930–1938 (units)

Date of estimate:

Artillery Shells (millions) Torpedoes Rifles (thousands) Machine guns (thousands) Rifle cartridges (millions) Aircraft Aircraft engines Tanks Ammunition powder propellants (thousands of tons) Explosives (thousands of tons) Poison gas (thousands of tons) Optics (millions of rubles)

∞Ω≥≠ (S-≥≠) Jan. ∞∏, ∞Ω≥∑

Jan. ∞, ∞Ω≥≥ April π, ∞Ω≥≥

Jan. ∞, ∞Ω≥≥ June ∑, ∞Ω≥π

(∞) ≥,≤∏∂ ≤≥.≥ ... Ω≠≠ ∑∞ ≥,≠≠≠ ≤,π∑∫ ≤,∫≠∂ ∞,≠≠≠

(≤) ∞∞,≤≥∂ ∂∑.∑ ... ∞,∂≠≥ ∞∞∂ ∑,∑≠≠ π,∑≠≠ ∞∑,∂≠≠ ∞≠,∑≠≠

(≥) ∞≥,π∫≠ ... ... ∞,≠≠≠ ∞∞∑.π ... π,∞∑≠ ∞≥,∏≠≠ ∞≠,≠≠≠

≤π.∫ ∂≥.∑ ∞π.∑ ...

∂≥ ∑∂.≤ ∂∫.π ...

... ?a ... ...

Jan. ∞, ∞Ω≥∑ Jan. ∞∏, ∞Ω≥∑ (∂) ≤≥,≠≠≠ ∏∑b ... ≤,≤≠≠ ∞≥≠ ∂,≠≠≠ ∫,≠≠≠ ≤∑,≠≠≠ ≥≠,≥≠≠ ∏∏.∫ ... ∏≠.≠ ...

Jan. ∞, ∞Ω≥∏, preliminary Nov. ∞∂, ∞Ω≥∑ (∑) ≤≥,∏≠≠ π∏ ≤,≠≠≠ ≤,≠≠≠ ∞∑≤ ∏,≠≠≠ ∞∞,≠≠≠ ∞∏,≠≠≠ ≥≠,≠≠≠ ∏∂.≥ ∏Ω ∑∫ ∞Ω≤

Jan. ∞, ∞Ω≥∫, plan May ≤∏, ∞Ω≥π (∏) ≥Ω,∞∫≠ ∞≠∞ ≥,≠≠≠ ≤,≥≠≠c ≤∑≠ π,∑≠≠ ∞∏,Ω≠≠d ∂≤,≥≠≠ ≥∑,∂≠≠ ∞≥∑ ≤∫≠ ∞≤≤.≤ ∂∑≠

Peak wartime production ... (π) ∞≥≠,≠≠≠ ∞∫∂ ... ∂,≠∂Ω ∂∑Ω π,∂≠∏ ≥≥,≤≠≠e ∑≤,ππ∏ ≤Ω,≠≠≠f ∞≤π ... ... ...

Given as 198,000 tons: something peculiar here! Includes shells produced by civilian industry (other columns are not specific). c Given as 2,420 in RGAE, 4372/91/3217: 125–123. d 20,600 including training aircraft. e Combat aircraft. f Simonov (1996: 177) gives 85,515 in 1942, including self-propelled guns (SAU). Sources: Col. 1: RGAE, 4372/91/2112: 34 (dated January 20, 1935). Col. 2: RGAE, 4372/91/1445: 181–180 (Botner and Kolesinskii of Gosplan defense sector to Kuibyshev and Mezhlauk). Col. 3: RGAE, 4372/91/3002: 139–132 (Smirnov, Gosplan, to Molotov), except aircraft and engines for January 1, 1933: RGAE, 4372/91/2761: 131, dated November 14, 1935 (memorandum from Mezhlauk), and gunpowder and optics for January 1, 1938: RGAE, 4372/91/3217: 125–123. Col. 4: as col. 1. Col. 5: RGAE, 4372/91/2761: 131, dated November 14, 1935 (memorandum from Mezhlauk). Col. 6: as col. 3. Col. 7: See Table 5.7, Harrison (1996: 180), Simonov (1996: 172–73), and (for propellants) Vernidub (2003: 41). a

b

Planning for Mobilization

133

Table 5.5. Planned and actual capacities, 1938–1940 (units)

Artillery pieces Rifles (thousands) Rifle cartridges (millions) Aircraft Tanks

Capacity for year of war planned in ∞Ω≥∫ (MP-∞)

Actual capacity achieved (early ∞Ω∂≠ estimate)

Mobilization capacity planned (autumn ∞Ω∂≠)b

(∞) ∑∞,∫∞∫ ≤,π∂∫a ∞∏,∏∂≠ ≤π,≤∏≠ ∞Ω,≤Ω≠

(≤) ≥∞,Ω∂≠ ≤,≥∑≠ ∞≠,∫π≥ ≤∑,ππ∑ π,∫∑≠

(≥) ≤π,ππ∏ ≤,∞≤≠ ∏,∂≤∂ ... π,≠∫∂c

Misprinted in the source as 274.8. I have multiplied the quarterly plans by four to obtain annual figures. c Includes armored cars; tanks alone were only 5,140. a

b

Sources: Col. 1 and 2: See Simonov (1996: 120, 170), citing archives. I have not obtained access to the documents from which Simonov obtained these figures. Col. 3: GARF, 8418/24/327: 2–51.

prewar arms production was located, was higher than the estimate made in October 1938.≥Ω In physical terms, the wartime production of major weapons was also much higher than the wartime capacity claimed for 1938 (see last column of Table 5.1). It was also higher than the estimate made at the beginning of 1940 shown in the above table, except for cartridges. The minister for armament, Vannikov, claims that the ready availability of spare capacity is shown by the rapid expansion of armaments production in the first months after the German invasion, before much Soviet territory was occupied. According to Vannikov (1969: 129–30), armament production in July 1941 was 30 percent greater than in June, and increased by a further 8 percent in August.∂≠ This claim is confirmed by Harrison’s (1996: 190) estimate that military production, 71 percent above the 1940 level in April–June 1941, before the invasion, had risen to 143 percent above the 1940 level in July–September, and declined only in October–December, when a large part of Soviet territory had already been occupied by the enemy.∂∞ Vannikov attributed this increase to the rapid transfer of materials and personnel from civilian to armaments industry immediately after the outbreak of war. The huge gap between current production and wartime potential capacity was covered in three ways. First, spare capacity was set aside in the armaments factories that formed part of the defense industry, ready for use in the case of mobilization. At a meeting of the council of the ministry of heavy industry in May 1935, the following exchange took place between Pavlunovskii and other senior officials:

134

R. W. Davies [m. m.] kaganovich. You have (u tebia) 500 or 600 machine tools tied up with wire. pavlunovskii. It’s not my fault that there is no war. (Laughter.) It is better that these machine tools should remain tied up. They are mobilization resources, special equipment, and we must base our current estimates on equipment which is actually working. piatakov. Of course. ordzhonikidze. If we look at the Zlatoust factory, we see that there are huge halls there, with a huge number of machine tools, but this is not the issue.∂≤

The armaments factories, as well as keeping machines in reserve, in peacetime used much of their capacity not immediately required for arms production to produce civilian goods. With the rapid increase in armaments from the mid-1930s onward, the nonmilitary production of the armaments industries declined as a proportion of total production (Table 5.6). Nevertheless, during the second five-year plan it increased substantially in absolute terms. Data for five branches of the industry show that their non-military production increased by as much as 78 percent between 1932 and 1936 (derived from the source cited for Table 5.6). In the preparation of the third five-year plan, the authorities continued to presume that new armaments factories would produce a wide range of non-military goods. Thus a document about the planned new artillery and ammunition factories proposed that the artillery factories should manufacture refrigerators, compressors, and road construction equipment, shell factories should make machine tools and ball bearings, explosives factories should make bicycles, gramophones, and metal goods, and gunpowder factories should make plastics and artificial leather.∂≥ Second, civilian factories, many of which in peacetime were already producing some armaments, or their components, were prepared for a switch to armaments production in the event of mobilization. Major civilian factories, in all the industrial ministries, included ‘‘special departments’’ (spetsial’nye otdely) responsible for the application of the mobplan in their enterprise; these departments were subject to a particularly severe regime of secrecy. As early as 1931, the mobplan included 150 major civilian enterprises (see Davies 1996: 116). The extent of the potential involvement of civilian industry varied from armament to armament. Aircraft frames and engines were to be assembled entirely, or almost entirely, in specialized factories (though many of their components were produced outside the industry). In contrast, according to a Gosplan estimate in April 1933, at the end of the second five-year plan (January 1, 1938) over 60 percent of tanks, between 50 and 64 percent of shells, and all poison gas would be manufactured in civilian industry.∂∂ Third, production at the armaments factories would be more intensive in

Planning for Mobilization

135

Table 5.6. Civilian output of the defense industry: The value of nonmilitary items by branch of industry in 1932, 1935, and 1936 (percentage of gross output)

Artillery Ammunition Tanks Optics Precision industry Batteries Average of the above Average, not including the ammunition and precision industries

∞Ω≥≤

∞Ω≥∑

∞Ω≥∏

∑∞.∂ ∑≤.∫ ≥≥.≠ π∑.≥ π∫.≠ ∫≠.≠ ∑∞.∑

∑∞.≥ ... ≥∑.∂ ∑∞.∏ ... πΩ.∞ ...

≥Ω.∏ ∂π.∫ ≥∏.≥ ∑∞.∞ ∂≥.≠ π∞.Ω ∂≥.∑

∑≠.∑

∂∫.∞

∂∞.∞

Note: Data for nonmilitary production in the aircraft, naval, chemical, and other defense industries were not available for the years before 1935. These figures refer to industries subordinate to the ministry of defense industry in 1937. Source: Calculated from data in RGAE, 4372/91/3217: 118–116 (dated May 20, 1937).

wartime: both the machines and the personnel should be employed for more hours a week. Thus the mobplan for light industry in the Russian republic in 1938 assumed that factories would work for 300 to 359 days a year in three shifts of 7.5 hours.∂∑ If achieved, this would more than double the capacity in use. It has not been possible to determine how far the huge expansion in armaments that took place during the war was due to the preparation for mobilization undertaken before the war, and how far to wartime emergency measures, the possibility of which had been underestimated in pre-war planning. ARMAMENTS INDUSTRIES

I illustrate the progress of mobilization in the armaments industries from the examples of the small arms, artillery, and ammunition industries, already well established before World War I, and the tank industry, which was very largely a product of the 1930s. Small Arms, Artillery, and Ammunition. A mobilization plan for these industries was prepared by the military in 1912.∂∏ The stocks of weapons it provided for were largely achieved by the outbreak of war in July 1914, and the annual production of weapons and ammunition proposed in the plan was greatly exceeded during the war. At the peak in 1916 rifle production was nearly double the wartime capacity as estimated in 1912, and the production of cartridges was more than treble (Gatrell 1979: 317–24).

136

R. W. Davies

In the peacetime Soviet Union the 1916 production of artillery pieces was not exceeded until 1937 and the 1916 level of rifle production was not reached until 1939. By the beginning of 1941, however, the stock of these traditional items—in the case of artillery technically far more sophisticated than in World War I—considerably exceeded the 1914 level.∂π And the wartime capacity planned in all the successive mobplans from 1931 onward was much higher than the actual production achieved during World War I. Although a substantial part of the capacity established by 1917 was not yet fully utilized in the first half of the 1930s, new factories and extensions to existing factories were already being constructed in anticipation of wartime needs. In January 1935 a report prepared by the Gosplan defense sector for the VII Congress of Soviets listed four new factories and one reconstructed factory in the artillery and ammunition industries, built since the previous congress of soviets in March 1931 at a cost of 173 million rubles.∂∫ The expansion of the industry continued during the second five-year plan, but attention was now particularly directed toward the production of ammunition and the supply of explosives. It was common ground between the military and the defense industry that investment in ammunition was inadequate. In April 1933, the Gosplan defense sector, in a memorandum to Kuibyshev and Mezhlauk, head and deputy head of Gosplan, complained that the investments allocated to ammunition and explosives in the draft plan, although greater than the investment in small arms and artillery, meant that ‘‘artillery fire, which is a bottleneck in the mobilization program at present, will continue to be a bottleneck during the second five-year plan. In 1937 our capacity will be 92 million shells, i.e. less than the Germans had in 1917, the propellant powder capacity will be 125,000 tons, while America at present had about 250,000 tons, and France about 215,000 tons.’’∂Ω From 1934 onward investment in the ammunition industry, which had been lower than investment in artillery in 1932 and 1933, now exceeded artillery investment by an increasing margin.∑≠ Nevertheless, complaints about its inadequacy continued. In May 1934 Voroshilov, in a trenchant memorandum to Stalin, pointed out that in the case of both rifle cartridges and explosives there was ‘‘almost no accumulation for the conduct of war.’’∑∞ Then in December of the same year the mobilization department of the general staff criticized the defects in the small arms, artillery, and ammunition programs, particularly emphasizing that the ministry of heavy industry had failed to involve civilian factories in these industries: at present the plans included 145 factories, less than during World War I and far less than in wartime Germany.∑≤ By this time the Defense Commission had already decreed that the wartime capacity for shells, cartridges, and explosives should be greatly increased by 1939, and had

Planning for Mobilization

137

Table 5.7. Peak wartime production of weapons by the major powers (thousands)

Artillery pieces Rifles and carbines Machine guns Combat aircraft Tanks and self-propelled guns

Germany

Japan

USSR

USA

U.K.

∞∂∫a ≤,∫∑∏a ∑≠Ωa ≥∂.∞a ∞∫.≥a

∫∂a ∫∫∑a ∞∑∏a ≤∞.≠a ∞.≤b

∞≥≠c ∂,≠∂Ωb ∂∑Ωc ≥≥.≤a ≤Ω.≠a

≤≤∞c ∑,∏∫≥c ∫≥≠c π∂.∞a ≥∫.∑c

∞∞∫c Ω∞≠c ≤∫∂b ≤≤.πa ∫.∏b

1944. 1942. c 1943. a

b

Source: Harrison (1998: 15–16).

allocated 4,000 million rubles for investment in these sectors, together with small arms and optics, in 1936 to 1938.∑≥ In 1937 the Gosplan defense sector, now preparing the third five-year plan due to start on January 1, 1938, complained, ‘‘In 1937 our capacity in heavy artillery and fire power lags behind Germany and Japan,’’ and estimated that German wartime annual artillery capacity was 47,000.∑∂ This proved to be a considerable underestimate (see Table 5.7). In spite of all the difficulties, the estimated capacity for one year of war of the main items of arms and ammunition, by January 1, 1938 (as estimated by Gosplan in May 1937), exceeded the Gosplan estimate for 1938 made in 1933, except in the case of cartridges (see Table 5.1, col. 5, and Table 5.4, col. 6). With the end of the purges and the division of the ministry of defense industry into four more specialized ministries at the beginning of 1939, the new ministries considered the extent to which their industry was ready to fulfill MP-1, adopted during 1938. A Gosplan report dated November 10, 1939, firmly stated, ‘‘The basic fault of MP-1 was that it had no material backing, i.e. the requirement of the ministries for raw and other materials, and equipment, and tooling, and for manual and technical labor, was not estimated.’’ As labor requirements had not been established, key workers in a number of key supply ministries, including iron and steel and non-ferrous metals, had been called up for the Red Army.∑∑ These strictures were confirmed at a meeting of the collegium of the ministry of ammunition on December 20, 1939. The minister reported that on receiving MP-1, ‘‘enterprises proved not to be able to carry it out. The main reason for this was that the managers did not know all the possibilities of the factories for working at full capacity. Capacity was decided without a profound study

138

R. W. Davies

of the real production possibilities, without the perspective of revealing and freeing all the possibilities concealed in production.’’ Although labor was available, it was not put to work in many cases because materials were lacking as a result of the plan not being tied in with enterprises in other branches.∑∏ Over a year later, in April 1941, the ministry of ammunition presented its proposals for the achievement of the mobplan for July 1941–June 1942 to Sovnarkom. These included detailed schedules for the contribution that twenty-five supply ministries and other departments could make to ammunition production, and also for the supply of machine tools and other equipment, iron and steel, nonferrous metals, chemicals, and paints by its own enterprises and those of other ministries. It requested the allocation of 85 million rubles in foreign currency for the purchase of equipment abroad (mainly from Germany . . . ). A draft Sovnarkom decree proposed to reduce civilian production and use the capacity released for the ammunition industry: the supply of copper for cables should be cut by 40 percent, and copper should no longer be used for the manufacture of consumer goods; the production of electrical items should be reduced by 30 to 40 percent.∑π Simultaneously, the ministry of ammunition presented an estimate to the Defense Committee of the capital investment required in 1941 so that the capacities of the industry could be increased to the level required by the mobplan for 1941/42. An accompanying note (zamechanie) from minister P. N. Goremykin stated that investment had amounted to 1,000 million rubles in 1940, and the 1941 allocation had been increased to 1,878 million rubles. In response to the mobplan it would need to be further increased to 4,030 million rubles in 1941, and a further 200,000 workers would need to be employed in capital construction. Supplies of materials, equipment, and transport for construction would need to be more than double the present claim for 1941—but this claim was being met only to the extent of 50 to 60 percent. The note concluded laconically and dramatically: ‘‘I consider that such a mobilization plan will not be realistic for carrying out in the remaining 8 months of 1941.’’∑∫ The elaborate mobplan for 1941/42 was approved on June 6, 1941. Two weeks later Germany invaded the Soviet Union, and on June 23, the day after the invasion began, the government agreed to put the plan into effect (Pospelov 1974, 1:21). Wartime production did not reach the level required by the mobplan, but it amounted to twelve times the 1940 level in the case of shells, and nearly treble that level in the case of cartridges. The lack of realism and precision in the estimated capacities is strikingly illustrated by the case of shells. On April 29, 1938, the Defense Committee decided that the mobilization capacity to be achieved for 1939 should amount to 219 million shells. This requirement was allocated to as many as 800 enter-

Planning for Mobilization

139

prises in sixteen ministries. But the available capacity proved far from sufficient, amounting to 108 million shell cases, propellants (including stocks) for 136 million shells, and explosives and fuses for a mere 72 million shells. In consequence, the military-industrial commission and the Defense Commission approved a requirement of 112 million, distributed among 300 factories. This figure was not adopted as part of MP-1 until August 3, 1939, and in consequence a mobplan of 56 million shells was adopted only for the six months from July to December 1939. The report to Gosplan in November 1939 stated that MP-1 suffered from the following deficiencies: a. the effort to maximize shell production led to the allocation of orders to machine-tool factories, which would have led to the disruption of machinetool production, an unambiguously negative fact; b. the production plan for shell bodies was allocated to vehicle, tram, and trolleybus factories without considering their stock of machine tools. This would have led to the disorganization of the repair of vehicles and would not have resulted in the production of shells, as there were no machine tools; c. even large factories, such as Uralmash, were required to produce shell bodies . . . which would have disrupted the production of major capital equipment. In the following June, in the new mobilization plan for July 1940 to June 1941, the defense ministry proposed the huge figure of 351 million shells, but industry claimed that in a year of war it would be able to produce only 77 million, less than the estimate for 1938 made in May 1937 (see Table 5.4).∑Ω It should be noted, however, that 83 million shells were in fact produced in 1941, as many as 56 million of these in July–December, and that production in 1944 reached 184 million (Harrison 1996: 180). Information about the mobplans for propellants (ammunition powder), and the capacities actually achieved, is sufficiently detailed to enable us to assess how far in this case the mobplans were realistic. In the last three and a half years before the war, with the switch of the economy to a war footing, the gap between estimated actual capacity and the capacity stipulated in the mobplans substantially narrowed: estimated actual capacity was 24 percent of the mobplan in 1938, and increased to 75 percent in 1941. At the same time current production increased from 15.6 percent of the mobplan and 65.7 percent of available capacity in 1938 to 55.1 percent of the mobplan and 72 percent of actual capacity in 1941 (see Table 5.8). On the eve of the war, the production and available capacity for ammunition still lagged behind the availability of artillery: the Soviet Union had more artillery pieces but less ammunition than Germany. During the war itself,

140

R. W. Davies

Table 5.8. Propellants: Plans and outcomes, 1938–1941 (thousand tons)

Mobilization plan Estimated actual capacity Realized output

∞Ω≥∫

∞Ω≥Ω

∞Ω∂≠

∞Ω∂∞a

∞∏∫.≠ ∂π.≠ ≤∏.≤

∞≤∑.≠ π≥.≠ ∑≠.∏

∞Ω∑.∞ ∫≥.Ω ∑Ω.∂

∞∑∏.∏ ∞∞∫.≤ ∫∏.≠

The figures for 1941 are very rough. The mobilization plan and realized output were obtained by doubling the figures for July–December 1941; the estimated capacity is at July 1, 1941.

a

Source: Vernidub (2003: 37, 41).

production of propellants was at first drastically reduced as a result of the seizure by the enemy of the territory on which much of the industry was located. Two factories in areas that were not occupied doubled their capacity during the war (Table 5.9). However, two major factories ceased production altogether and others were evacuated, and as a result the total capacity of factories already operating in 1941 expanded only slightly during the war. Additional capacity was supplied partly by factories that were slowly built during the 1930s, and started production only in 1941 or after, and partly by factories constructed from scratch during the war. Thus by 1944 factories already operating in the 1930s provided 42 percent, factories built during the 1930s 45 percent, and factories built during the war 14 percent of capacity (Table 5.10). The substantial role of factories built during the 1930s, which had not been available at that time for immediate production, shows that in this case the peacetime development of capacity for wartime production played an important part in Soviet wartime military strength. Wartime Soviet production of propellants provided 64 percent of the total amount used at the front during the war. This was supplemented by the use of the reserve stocks of the army artillery administration, which provided 12 percent; and by imports from the United States, Britain, and Canada, which provided a further 24 percent. In addition, substantial amounts of ethyl alcohol and other products were imported and used in the Soviet production of propellants (Vernidub 2003: 56–57). Tanks. The 1931/32 programs assumed that in wartime the vast majority of tanks would be produced by the tractor and automobile industry. The Stalingrad factory would produce 12,000 T-26s a year, Kharkov would produce 2,000 BTs, and the Gorkii factory 12,000 pocket tanks. The Voroshilov works, which was part of the armaments industry, would not produce more than 1,500 T-26 tanks.∏≠ Recognizing that these objectives could not be achieved by the

Planning for Mobilization

141

Table 5.9. Propellants: Two factories, 1941 and 1944 (thousand tons) July ∞, ∞Ω∂∞

Jan. ∞, ∞Ω∂∂

≤≥.≠ ∞∑.∑ ≥∫.∏

≥≠.≠ ∂≥.≤ π≥.∫

No. ∂≠ (long-established, Kazan’) No. ≥Ω≤ (began production ∞Ω∂∞) Total for these two factories Source: Vernidub (2003: 40–41).

Table 5.10. Propellant capacity by age of factory, 1941–1944 (thousand tons)

Factories operating before ∞Ω∂∞ Factories built by ∞Ω∂∞ but not yet operating Factories built during the war Total capacity Actual productiona

July ∞, ∞Ω∂∞

Jan. ∞, ∞Ω∂≤

Jan. ∞, ∞Ω∂∂

∫π.≤ ≥∞.≠

∂∑.∏ ≥∞.≤

π∂.∂ ∫≠.∂

... ∞∞∫.≤ ∫∏.≠b

≠.∏ ππ.∂ ∏π.πc

≤∂.≤ ∞πΩ.≠ ∞≤∏.Ωd

Total output was less than total estimated capacity, but we are unable to allocate the shortfall among different ages of factory. b Output from July to December 1941 at annual rate. c Annual output for 1942. d Annual output for 1944. a

Source: Estimated from data in Vernidub (2003: 36–41).

beginning of 1933 as planned, the Politburo resolved that the Stalingrad factory should be able to reach its wartime capacity of 12,000 by the summer of 1933 (see Davies 1996: 171, n. 149). At this time no major developments appear to have been undertaken at the other factories scheduled for tank production. The conversion of the Stalingrad factory proved very expensive, and took much longer than planned. The cost of conversion, including the construction of a second assembly shop, escalated by August 1932 from an initial 11 or 12 million to 90 million rubles.∏∞ Nevertheless, by January 1, 1935, 99 million had already been spent.∏≤ But an investigation by Khalepskii, head of the ‘‘administration for motorizing and mechanizing the Red Army,’’ reported that ‘‘the factory [was] not ready,’’ and that no decision had been taken on who would supply the armor plating, semis, ball bearings, and electrical items. Reporting this situation to the Defense Commission in December 1934, Voroshilov complained that the situation was ‘‘intolerable.’’∏≥ Ordzhonikidze did

142

R. W. Davies

not reply until April 1935, when he claimed that in 1935 the factory would be ready ‘‘in the main’’ for the wartime production of 6,500 tanks, and that the long-awaited capacity of 12,000 tanks would be reached by 1936.∏∂ After further delays, in the spring of 1937 a report from the Gosplan defense sector claimed that the conversion was more or less complete, and the full capacity would be reached by January 1, 1938. However, the factory was now going over to the T-46, and its capacity would reach 18,000, plus spare parts, by 1942 (this plan proved abortive). Gosplan also stated that the Gorkii factory was now capable of producing 10,000 pocket tanks a year in wartime, to be increased to 18,000 by 1942. In all, civilian factories would be responsible from January 1, 1938, for 22,300 of the total wartime production of 35,400 tanks.∏∑ Some progress had certainly been made by this time. By March 1937, for example, the Izhora, Mariupol’, and Kulebaki-Podol’sk works had a wartime capacity of 14,000 armor-plated bodies out of the 35,000 required. The capacity for the armor plating itself was on a similar scale. In February 1937 the ministry of defense industry urgently requested Sovnarkom to allocate an additional 34 million rubles to the ministry of heavy industry so that Krasnyi Oktiabr’ in Leningrad, the Andreev factory in Taganrog, and the Krasnoarmeiskaia shipyard in Stalingrad could supply armor-plated bodies to the Stalingrad factory.∏∏ The stock of spare parts for tanks was extremely small. The defense sector of Gosplan pointed out to Sovnarkom that the value of the spare parts held by the Committee of Reserves amounted to only 24 million rubles; this must be built up to 179 million rubles.∏π In May 1937 Gosplan anticipated that an annual capacity of 35,400 tanks, including 16,000 pocket tanks, would be reached by January 1, 1938. The majority of these, 22,300 tanks in all, including 10,000 pocket tanks, would be produced by factories managed by the ministry not of defense industry but of heavy industry.∏∫ However, this figure was rendered obsolete by the drastic shift to modern armor-plated tanks. The estimated capacity in 1940 was far lower than that required by MP-1. But eventually, annual wartime production, consisting almost entirely of medium and heavy tanks, reached 29,000 (see Table 5.1). MOBILIZATION OF THE CIVILIAN SECTOR

The role of civilian industry in mobilization planning is illustrated by the examples of transport and communications, and the food industry. Transport and Communications. Investment in transport for defense purposes was partly directed toward facilitating defense activities in peacetime, but its primary purpose was to prepare the system for the event of war. Trans-

Planning for Mobilization

143

port took the lion’s share of defense investment in the civilian sector. During the second five-year plan, 1933 to 1937, investment in transport and communications for so-called narrow defense purposes amounted to 1,970 million rubles as compared with the 5,937 million rubles invested in the armaments industries.∏Ω Of this total, 41 percent was invested in the railroads, 40 percent on the road network, and the rest on waterways and communications. The main outcome was the construction of 1,201 kilometers of new railroad lines, out of a total constructed in this period of 4,800 kilometers, and 6,667 kilometers of strategic roads.π≠ The length of strategic roads completed annually greatly increased from 1935 to 1937 after the adoption of a revised plan. In addition to these developments categorized as defense investment, a number of lines and roads listed by the Gosplan defense sector as of military significance were built as part of normal investment. These included a line from Volochaevka to Komsomolsk-na-Amure and the main roads from Moscow to Minsk and Moscow to Kiev. The double-tracking of the Trans-Siberian line was also accomplished primarily during the second five-year plan, together with the strengthening of the line to enable it to carry heavy military loads. From 1933 to 1937 substantial investments were also undertaken on the accumulation of mobilization stocks earmarked for the use of transport and communications in the event of war, and stored by the Committee of Reserves. Their cost amounted to 1,959 million rubles, equal to the amount spent on defense investment in the sector.π∞ The weakness of the transport system in the Far East played a major part in the inability of the Soviet Union to take a strong stand against Japan after the invasion of Manchuria in 1931 (Davies 1996: 164–67 and 171). Accordingly, the investment in both new railroads and mobilization stocks for transport was directed primarily toward Siberia and the Far East. The vehicle industry was a new development in the 1930s, and only a limited number of road trucks was available for army transport in the event of war. In 1937 a Gosplan report estimated that the Red Army would have 43 vehicles and tractors available per 1,000 soldiers. The equivalent German figure even before German rearmament got seriously under way was 33 per 1,000, but a German truck had twice the capacity of a Soviet truck.π≤ The army and the Gosplan defense sector were therefore concerned to ensure that adequate horse-drawn carts would be available for the military. An anxious memorandum from the defense sector to the head of Gosplan dated May 20, 1937, pointed out that the total number of carts in the USSR had declined from 16.3 million in 1926 to 9.6 million in 1933. Since 1933 no census had been carried out, but the number of carts had probably fallen to 6.5 million, of

144

R. W. Davies

which over one million had been produced from 1934 to 1937, mainly after 1934. To cope with the shortage of carts in the event of war, high-level decisions in 1935 ordered every collective farm to prepare between one and three carts, depending on its size, together with their equipment, for transfer to the military in the event of mobilization.π≥ These decisions, if carried out in full, would make available some 600,000 to 700,000 carts for the army. In shipping, the pre-war mobilization plan proposed that sixty-five civilian ships should be armed and mobilized, and transferred to the Northern Fleet, within one month of mobilization. In fact, 122 ships were equipped and transferred in forty-eight days.π∂ The Food Industry. Throughout the 1930s the mobilization departments of all the civilian industries prepared for the vast expansion of the army that would follow the outbreak of war. The most important industries were of course the food, drink, and tobacco industries, and light industry responsible for clothing and footwear, as well as for textiles for parachutes and other purposes. The food industry is considered here as an example. The patchy evidence indicates, as would be expected, that investment for defense purposes increased rapidly throughout the decade. Investment increased from 15 million rubles in 1932 to 48 million in 1933 and a planned figure of 77 million rubles in 1937.π∑ By 1936 twelve new factories had largely been completed, producing meat, fish, salt, soap, and other commodities. Nearly all these factories were built in Siberia and the Far East.π∏ Following the usual pattern, the mobplans for the food industry were based on estimates of requirements during the first year of war. They were concerned primarily with the army itself, but also dealt more cursorily with the needs of the civilian population. The requirements steadily expanded during the 1930s, as Table 5.11 illustrates. Plans were also made for the production of longlasting rusks (sukhari), which were not manufactured in peacetime.ππ The extent of the army requirements as a proportion of total production varied considerably from product to product. A table comprising twenty major food, drink, and tobacco commodities, prepared in 1936, showed wartime requirements from the ministry of food industry ranging from 2.3 percent of the production planned for that year in the case of matches, to 40 to 60 percent for meat and dairy products.π∫ The most important requirement was of course grain. The amount of bread, groats, and fodder needed by the army in one year of war was estimated in 1936 at 6,077,000 tons compared with the 938,000 tons actually allocated in the agricultural year 1935/36. The increased demand for fodder was particularly large owing to the planned conscription of a large number of horses into the army.πΩ Throughout the 1930s the authorities struggled, eventually successfully, to build up an ‘‘untouchable fund’’ and a

Planning for Mobilization

145

Table 5.11. Food products: Army requirements and available supplies, 1930–1936 Annual army requirement

Hard biscuits (thousands of tons) Vegetable oil (thousands of tons) Condensed milk (millions of standard cans) a

∞Ω≥≠

∞Ω≥∂

∞Ω≥∏

Total output for all purposes, ∞Ω≥∏

(∞) ≤π.∞ ∑≠.∫

(≤) π∏.∂ ∑∂.∂

(≥) ∞≠∏.≥ ∑≥.∂

(∂) ∞≠∂a ∂≠∏

...

π.≤

∞Ω.∏

≤∫a

Planned output.

Sources: Cols. 1 and 2: RGAE, 4372/91/3025: 3–2 (dated January 10, 1934). Col. 3: RGAE, 4372/91/2691: 35. Col. 4: as col. 3, except vegetable oil from RGAE, 8453/2/26: 355ob.

‘‘mobilization fund’’ (also known as a ‘‘state fund’’) of grain, partly to guard against bad harvests, partly for the army and the civilian population in the event of war. To store the grain, extensive warehouses were constructed, particularly in Siberia and the Far East; this work was transferred to the NKVD at the end of 1935.∫≠ The ministry of food industry also prepared wartime plans for the many socalled non-food commodities produced by its factories, such as soap and glycerin, showing the capacity required and the allocation of the commodities among ministries. The work was hampered in the mid-1930s by the absence of an up-to-date national mobplan. For example, on July 7, 1937, the mobilization department of the industry summoned a meeting attended by representatives of the major ministries, including defense, to establish the requirements for glycerin, a trihydric alcohol with many uses in foodstuffs and chemicals; most of the defense industry’s allocation was consumed in the preparation of explosives, but it was also used in airframes, parachutes, army clothing, and industrial fabrics. Two of the five ministries present at the meeting had no idea of their wartime requirement. The ministry of light industry reported that it ‘‘has no new mobilization targets for 1937 and cannot say anything,’’ and the ministry of heavy industry stated that ‘‘a centralized mobilization claim cannot be compiled, and none can be presented.’’ Defense, the defense industry, and the timber industry were able to state how much they required, but the amounts were much higher than previous claims and exceeded the existing production capacity in the food industry. The meeting had to refer the whole matter back to the ministries and to the Gosplan defense sector.∫∞ By 1937 the food industry was engaged in compiling ‘‘MP-4-1937,’’ to

146

R. W. Davies

show requirements in the event of war in 1937. On April 28, 1937, it asked all government departments to fill in forms listing their wartime needs, and submit them by June 15.∫≤ As we have seen, in the case of glycerin these figures were still not available a month after this date. But by October the ministries had evidently already complied, and the ministry of food industry issued a further set of instructions on the procedure for allocating each minister’s total allocations between their subdepartments. By this time the mobplan was described as covering ‘‘1937 and 1938 war years.’’∫≥ With the growth of the armaments’ industries, civilian industry sought to retain the authority in the event of war to utilize as much as possible of its peacetime capacity and its peacetime allocation of materials. On June 4, 1936, STO ruled that the annual wartime capacity to produce gas masks should be increased to 60 million by January 1, 1939. As the planned wartime capacity of the twelve factories scheduled to produce gas masks was only 50 million, it was decided that the deficit should be met by the Odessa factory of the food industry. Minister of the food industry, A. I. Mikoian, protested, and on May 28, 1937, the ministry of defense industry conceded that the existing plants could achieve the extra capacity, provided that Gosplan prepared a plan for this during 1938.∫∂ A striking example of the problems of coordination between the armaments and the civilian industries is provided by the tinplate (belaia zhest’) used for canning food. The existing mobplan for the ministry of defense industry required the use of a large amount of tinplate for the manufacture of grenades, gas masks, and smoke pots (shashki). One April 15, 1937, the mobilization department of the ministry of food industry sent a memorandum to the mobilization administration of the defense industry ministry, pointing out that the proposed allocation of tinplate could be met ‘‘only by reducing the production of tinned food, particularly meat and fish, of which the army is a consumer.’’ The ministry of food industry therefore proposed that substitutes for tinplate should be found; sheet steel (chernaia zhest’) had already been used for gas masks, and plastics or glass was another alternative.∫∑ LOCAL MOBILIZATION

In every regional, town, and district soviet, units were established for mobilization planning; and the local arrangements for mobilization were a frequent matter of concern to the military districts, each of which usually covered several regions. By 1934 Leningrad, for example, had prepared an elaborate plan for evacuating the city in three phases. The third phase involved the evacuation of 118 defense factories. During the next four years, the plan was changed locally without coordination among the various local agencies,

Planning for Mobilization

147

and without any input from central government. The senior military official responsible for mobilization in Leningrad described this activity as ‘‘amateurish’’ and castigated the local officials concerned for ‘‘illiteracy over mobilization.’’ In 1937 and 1938 the Leningrad executive committee appealed in vain to the higher authorities for a revised plan. Some progress was made in spite of the absence of a revised plan. In May 1937 the NKVD checked mobilization preparedness at eight defense factories; as a result, a party resolution criticized the state of the mobilization reserves and established mobilization officers at the armaments plants (a staff of eight persons was scheduled for the Kirov works) (Losik and Shcherba 2000: 181). Then in 1939 on the basis of a memorandum from the Leningrad military commissar, a quite elaborate plan was prepared for the evacuation of children’s homes. But activities remained piecemeal and inadequately resourced until a few months before the German invasion (Cherepenina 2005: 22–23). The failure of the higher agencies to bring the mobplans up to date alarmed the local authorities. In 1937 the regional party secretary in the Kalinin region complained that the ministry of heavy industry had not revised the mobplan for its wagon factory since 1933, so that the factory was still instructed to produce horse-drawn pontoons that were no longer used by the army: ‘‘On the first day of war the factory will be drowned in materials, but it will not receive the right ones.’’ A shoe factory in the region was scheduled to make 315,000 pairs of boots with rubber soles, but a satisfactory process for sticking on the soles had not been devised. Factories required to produce army clothing had not been allocated supplies of materials. Regional stocks of army clothing had been accumulated as long ago as 1927 to 1929, and 30 to 40 percent of the stocks were already useless. In general, the staff of the mobilization departments (‘‘special sectors’’) of trusts, enterprises, and establishments were poorly qualified, and the managers of factories and other establishments ‘‘do not engage in serious work on mobilization.’’∫∏ The serious defects in the national mobplans for the regions did not mean that local authorities undertook no preparation for war. A report on mobilization preparations in West Siberia in 1936 noted that resources were obtained from the central and local budget in roughly equal amounts for implementing ‘‘special control figures of defense expenditure.’’ These resources were used for preparing hospitals and schools for wartime use, converting road trucks so that they could be used to transport the wounded, constructing barns for military stores of hay, and accumulating stocks for wartime use. In addition, a larger sum was used for what were described as ‘‘measures of defense significance.’’∫π In the Omsk region reserve stocks included soap, macaroni, tea, cigarette paper, and preserves. (However, the military complained that these

148

R. W. Davies

stocks were inadequate, and also that the footwear factory in the region was not ready for wartime production.)∫∫ In the Voronezh region, the ‘‘narrow presidium’’ of the regional executive committee, with six members, undertook a quite modest series of measures. As in West Siberia, these included the adaptation of trucks and buses to carry the wounded. They also included the preparation of schools to be used as hospitals, supplying schools with rifles and appropriate storage, and preparing the capacity to produce iron barrels for the army. The local authority was also responsible for checking the lists of persons liable for military service and in the reserve, and for checking the number and quality of horses and gear that would be sequestered for the army in the event of war.∫Ω THE MOBILIZATION OF LABOR

The mobplans assumed that labor would be used more intensively in the event of war, but made surprisingly modest assumptions about the extent of the intensification that the invasion of the Soviet Union would require. In April 1937, E. I. Kviring, in a memorandum to Molotov on behalf of Gosplan, pointed out that the 1933 mobplan for labor now needed to be substantially revised and explained that the ministry of defense industry, in preparing a new mobplan, had assumed that the working day would last 7.5 hours with two days off a month in metalworking and similar shops (so about 2,558 hours a year), eight hours with three days off a month in industries where continuous production prevailed (that is, about 2,682 hours a year), and 144 or 180 hours a month (so 1,728 or 2,160 hours a year) in the chemical industry. Gosplan complained that the ministry of heavy industry and others had adopted different working regimes, and urged that the government should approve a national standard. However, although it pointed out that ‘‘in a year of war we will have insufficient labor, particularly skilled labor, mainly in the defense and heavy industries,’’ it did not suggest that the working day should be further increased.Ω≠ In fact, hours worked in wartime were far higher than those proposed in 1937, reaching a peak in 1943 of 2,908 hours in the machine-building and metalworking industry (which during the war was devoted very largely to armaments) and 2,708 in industry as a whole (Harrison 1996: 259). In the civilian mobilization departments a major preoccupation was the retention of labor in the event of war. On December 15, 1937, a deputy minister of the food industry sent a memorandum, ‘‘On the system for preparing plans to secure labor in wartime,’’ to the constituent departments of the ministry, and to the republican food industry ministries. This instructed them to prepare information on the labor shortage that would occur in the first month of war as a result of the call-up of army reserves, and to propose how

Planning for Mobilization

149

the deficit would be met, suggesting that more women could be employed, and also men who were not of army age. The memorandum also called for the adoption of measures to increase production per worker by raising output norms, increasing the number of shifts and the length of the working day, cancelling days off, and transferring labor from lower-priority food enterprises. Compulsory overtime up to fifty hours a month could be introduced, but ‘‘only in exceptional cases . . . when all other reserves of the enterprise . . . have been exhausted.’’Ω∞ The evidence available is not sufficient to enable a quantitative estimate of the resources allocated to mobilization. This is partly because of the extreme degree of secrecy applied to mobilization matters. But the defense budgets and the plans of the armaments industries do not seem to have divided expenditure between current and mobilization purposes, and the issue of the relative resources to be devoted to the two purposes was not discussed in the 1930s in the sources I have seen. It is certain that far less attention was devoted to the preparation of capacity for wartime use than to current production planning, and to investment for the immediate production of armaments. Until 1938, the Politburo, Sovnarkom, and the Defense Commission issued a rather small number of decrees and instructions about mobilization planning, and mobilization received far less attention in the ministries than current planning. The available documents indicate that concern about mobilization planning particularly declined in the mid-1930s, with the concentration on efforts to increase immediate military strength and to improve the quality and sophistication of weapons. Mobilization activity was particularly hindered in the civilian ministries by secrecy; as Mark Harrison will illustrate in Chapter 9, even in the armaments industries, the representatives of the army were prevented from inspecting the progress of preparations for mobilization. Nevertheless, the Soviet Union seems to have been more advanced than its rivals in mobilization planning. Throughout the 1930s senior officials and party leaders, both in the army and in Gosplan and the ministries, frequently urged Stalin, Molotov, and the leading members of the Politburo to devote more attention to mobilization, and to establish arrangements for coordinating it on a national scale. A network of staff concerned with mobilization existed throughout the 1930s at every level from the ministries to the major factories and the administrative districts. With the evidence available, it has not been possible to establish at all reliably the capacity ready for use in war conditions. From 1938 onward, mobilization planning received much more attention at central and local levels, and something like a national coordinating mechanism seems to have emerged.

150

R. W. Davies Notes

This paper has drawn on the pioneering research of Lennart Samuelson and Nikolai Simonov. I am grateful to Julian Cooper, Mark Harrison, Andrei Markevich, and Lennart Samuelson for comments, and to Markevich, Samuelson, Christopher Joyce, and the archivists of GARF, RGAE, RGASPI, and RGVA for the supply of material. Samuelson is undertaking further work on mobilization planning on the eve of World War II. This chapter was prepared with the financial support of Grants Nos. R000239543 and R000221443 from the U.K. Economic and Social Research Council. 1. Tank production in 1937 was planned at 2,850 units, shell production at 7 million units (Samuelson 2000: 177.) 2. RGAE, 4372/91/3217: 10 (report from Gosplan defense sector, dated May 11, 1937). 3. RGAE, 4372/91/3217: 89 (for aircraft), and 63 (for chemicals). 4. RGAE, 4372/91/2761: 132–131 (report prepared by Mezhlauk, November 14, 1935). The report noted that aircraft capacity on January 1, 1936, would be 11,000, and engine capacity 16,000. 5. RGAE, 4372/91/2761: 131. 6. Estimated from data in RGVA, 4/14/1814: 174 (memorandum from Voroshilov to Molotov dated December 26, 1936). 7. The actual number of shells supplied to the armed forces in 1935 reached 2.4 million (Samuelson 2000: 182). 8. RGVA, 33987/3/400: 257–261. 9. RGVA, 40438/1/2218: 7. 10. RGVA, 40438/1/2218: 7 and 7ob. 11. RGVA, 40438/1/2218: 298 (dated September 27, 1936). 12. RGVA, 40438/1/2218: 300 (dated December 9, 1936). 13. The memorandum from Egorov states that the ministries of transport and water transport had prepared MP-4 for 1935, and the ministries of food industry, internal trade, and RSFSR (Russian republic) local industry had prepared MP-4 for both 1935 and 1936 (RGVA, 40438/1/2218: 7). MP-4 of the ministry of light industry had not appeared by the end of 1936 (RGAE, 4372/91/2968: 74–69); by the time it was completed the period it covered had already expired. 14. RGAE, 4372/91/2968: 101–100 (memorandum from Kviring, deputy head of Gosplan, dated May 20, 1937). 15. RGAE, 4372/91/2968: 74–69 (memorandum from the acting head of the mobilization department of factory No. 204 to Gosplan defense sector, dated March 23, 1937). 16. RGAE, 4372/91/2968: 103–102 (dated August 16). 17. Khaustov et al. (2004: 479) (NKVD report to Stalin dated February 8, 1938). Stalin sent the report to other members of the Politburo with the comment ‘‘Rukhimovich’s testimony deserves attention, as it gives the possibility of lightening the task of eliminating wrecking in the defense industry’’ (Khaustov et al. 2004: 484). 18. Naumov (1998, vol. 2: 532–57). Both report and decree were marked ‘‘In favor. Stalin. Molotov.’’ The decree was dated November 29, 1937. 19. Chemical warfare, including the provision of gunpowder and other explosives, and

Planning for Mobilization

151

the aircraft industry, however, came under separate chief administrations. At the end of 1935 the military and mobilization administration was subdivided (see Mukhin 2004: 142). 20. GARF, 8418/6/2519: 116–118. 21. RGAE, 4372/91/2968: 143–140 (dated March 13, 1937). 22. RGAE, 4372/91/2968: 144 (dated July 2, 1937). In this time of denunciations, Buzinkov sent the memorandum over the head of Levin, who was in charge of the mobilization department. Buzinkov had evidently looked at his original memorandum in Levin’s office. 23. RGASPI, 17/3/987: 18 (dated April 27). Its members were Stalin, Kaganovich, Voroshilov, Chubar’, Rukhimovich, and Mezhlauk, with Gamarnik, Mikoian, Zhdanov, and Ezhov as candidate members. 24. Decree dated December 31. 25. Decree 190s of the Defense Committee, dated December 7, 1937. 26. The memorandum was based on a report sent to Bazilevich in January 1938 by a certain Major Ivanov, assistant to the chief of the first department of the Defense Committee, in January 1938. 27. On the same day a top-level ‘‘military-technical bureau’’ with a staff of fourteen was attached to the Defense Committee; its function was to handle the receipt, distribution, and assessment of intelligence material. Chaired by Molotov, its members were Stalin, Voroshilov, both Kaganovich brothers, and Ezhov (decrees 49 and 50ss). 28. GARF, 8418/23/92: 12, refers to this decree. 29. GARF, 8418/26/179: 1 (article 107). Its functions were set out in a decree of July 23 (GARF, 8418/28/35: 35–38). 30. Chubar’ was arrested on July 4, 1938, and sentenced to death on February 26, 1939. 31. RGVA, 33387/3/1075: 76–63. 32. Decree 161ss of the Defense Committee, dated July 23, 1938. 33. GARF, 8418/23/92: 12–14 (memorandum from Aleksandrov to Voznesenskii, dated November 10, 1939). 34. In addition, ships to be made available in 1938/39 in the event of war would cost 1,476 million rubles as compared with the 1938 order of 856 million. 35. These decrees are referred to in GARF, 8418/23/92: 12–14 (memorandum of November 10, 1939 by Aleksandrov). I have not been able to obtain access to these decrees. 36. I have not obtained access to the decree of July 29, 1939, referred to in the decree of July 25, 1940. 37. GARF, 8418/24/327: 2–51 (the decree was signed by Voroshilov as chairman and M. Pugaev as secretary of the Defense Committee). 38. For guns and shells, see RGAE, 4372/91/3217: 48 (dated May 13, 1937). 39. The value of the finished output of munitions (excluding naval munitions) in 1944 has been estimated at 65,800 million rubles at 1937 prices (Harrison 1996: 187) as compared with the 1938 estimate of 56,400 million at 1938 prices (see above). 40. Vannikov, an industrial worker before the revolution, graduated from the Bauman institute in 1926, was appointed director of the Tula small-arms factory in 1933, and by

152

R. W. Davies

January 1939 rose to be minister for armaments; he was under arrest from June to August 1941 and had to advise Stalin from his prison cell. 41. These figures exclude naval construction. See also the data in physical terms (Harrison 1996: 180). 42. RGAE, 7927/38/177: 113. 43. RGAE, 4372/91/3222: 197, 192, and 190 (dated May 27, 1937). 44. RGAE, 4372/91/1648: 176 (dated April 7, 1933, signed by Botner and Kolesinskii, head and deputy head of the defense sector of Gosplan). 45. GARF, 8418/26/296: 9 (instruction prepared in June 1938). 46. For background information on these industries, see Cooper (1976). 47. Stock of weapons:

Rifles (thousands) Machine guns (thousands) Artillery pieces

July 1914

Jan. 1, 1941

4,652 4.2 c. 7,500

6,781 338 40,812

Sources: 1914, from Gatrell (1979: 317–24); 1941, from army mobplan of February (Naumov 1998, 1: 607–50). 48. RGAE, 4372/91/2112: 29 (dated January 13). 49. RGAE, 4372/91/1448: 181–178 (dated April 7). These comments refer to the lower second variant of the plan; the higher variant was regarded by the authorities as not feasible. 50. See data in RGAE, 4372/91/3217: 115 (dated May 20, 1937). 51. RGVA, 22987/8/533: 265 (dated May 14). 52. RGVA, 40438/1/2260: 63–71 (dated December 27). 53. RGAE, 4372/91/2761: 132–122 (memorandum by Mezhlauk, dated November 14, 1935, citing decree of June 27, 1935). 54. RGAE, 4372/91/3002: 139–138 (dated May 24). German wartime shell capacity was estimated at 216 million. 55. GARF, 8418/23/92: 12–14 (memorandum from Aleksandrov). Voznesenskii did not accept even this scathing memorandum, and wrote on it: ‘‘The memorandum covers up the bureaucratic nature of a number of aspects of MP-1. Evidently comrade Aleksandrov does not want to disclose these bureaucratic aspects.’’ 56. RGAE, 7516/1/272: 1–3. 57. RGAE, 7516/1/790: 1–41 (dated April 29, 1941). 58. RGAE, 7516/1/789: 1–44 (dated April 19, 1941). 59. The information in this paragraph is pieced together from memoranda to Voznesenskii from Matveev, dated July 31, 1938 (GARF, 8418/23/92: 53–54), and by Aleksandrov, dated November 10, 1939 (GARF, 8418/23/92: 12–18), and from data available from http://www.soldat.ru/doc/mobilization/mob/chapter1 — 4.html. 60. RGAE, 4372/91/1475: 11–9 (unsigned report to the Gosplan defense sector dated March 4, 1932); 4372/91/1050: 134–129 (report at meeting chaired by Unshlikht, August 10, 1932).

Planning for Mobilization

153

61. RGAE, 4372/91/1050: 133–132 (statements by Botner and Goldberg, August 10, 1932). 62. GARF, 8418/10/127: 2, 2ob, and 3. 63. GARF, 8418/10/127: 2 and 2ob (memorandum to Molotov dated December 7). 64. GARF, 8418/10/127: 5–6 (dated April 1). 65. RGAE, 4372/91/3217: 47–42 (unsigned report dated May 13, 1937). This figure included three hundred heavy tanks to be made in the Kirov works. 66. RGAE, 4372/91/2888: 85–80, 77, and 75–74 (correspondence between Rukhimovich, Molotov, Gurevich, and G. Smirnov, February 16–March 22, 1937). 67. RGAE, 4372/91/2888: 35–32 (Botner to Mezhlauk, January 27, 1937). 68. RGAE, 4372/91/3222: 176 (dated May 26, 1937). 69. For transport and communications investment, see RGAE, 4372/91/3217: 26 (dated May 18, 1937); for armaments industry investment, see Harrison and Davies (1997: 384), and (for 1937) Samuelson (2002: 193). These figures cover industry subordinate to the ministry of the defense industry in 1937. The figure for transport is probably too high, because it includes a preliminary figure for 1937, in which year the investment plan was considerably underfulfilled. These figures exclude the Baikal-Amur railroad. 70. RGAE, 4372/91/3217: 68–66 (Kononenko, defense sector, to Kviring, deputy head of Gosplan). In addition, 103 kilometers of secondary lines were constructed for military use. Although the main track constructed almost equaled the amount in the plan for 1933 to 1937, these secondary lines were a mere 7.5 percent of the plan. 71. RGAE, 4372/91/3217: 69 (this figure seems wildly improbable). 72. RGAE, 4372/91/3217: 129–128. 73. RGAE, 4372/91/3222: 54–53 (Kononenko to Smirnov). The Sovnarkom central committee decision of August 9 and the TsIK-Sovnarkom decision of October 7, 1935, instructed collective farms to have available one cart if they included fewer than 50 households, two carts if they included 50 to 100 households, and three carts if they included more than 100 households. 74. See http://www.warwick.ac.uk/go/aviaprom/ver4/1941.doc. 75. RGAE, 4372/91/2025: 3–2 (dated January 10, 1934). The figures for 1932 and 1933 refer to the ministry of supply, the figure for 1937 to the ministry of food industry. 76. For a brief account of each of these factories, see report of the food industry mobilization department in RGAE, 4372/91/2473: 52–51 (the report is dated December 22, 1935, but handwritten notes indicate that the factories would be completed in 1936). For a proposal to construct a bakery in the Far East under the consumer cooperatives, see RGAE, 4372/91/2804: 17–16 (dated May 31, 1935). 77. RGAE, 4372/91/2025: 3 (dated January 1, 1934). Rusks stored since 1913 were consumed by the soldiers defending Moscow in 1941 and 1942 (interview in program Oboznaia voina, Moscow TV Pervyi Kanal, 2003). 78. RGAE, 4372/91/2691: 55. 79. For the demand, see RGAE, 4372/91/2691: 55, for the grain allocated in 1935/36, see RGAE, 1562/12/2118: 129–132. 80. See the Politburo decision of October 7, 1935, that in 1936 the NKVD should construct grain warehouses with a capacity of 300 million puds (4.9 million tons) (RGASPI, 17/162/18: 192).

154

R. W. Davies

81. RGAE, 4372/91/2691: 43–42 (minutes dated July 22, signed by A. Loos, head of the mobilization department of the ministry of food industry). 82. RGAE, 4372/91/2728: 82–81, signed by a deputy minister and circulated in eighty-eight copies. 83. RGAE, 4372/91/2968: 132–131 (dated October 26, 1937); in this year of purges, the document was signed by a different deputy minister, as the previous signatory had been dismissed. 84. RGAE, 4372/91/2728: 94–92 (memorandum from Kott, head of mobilization administration of the ministry of defense industry). 85. ‘‘On Economizing Tin in Wartime’’ (RGAE,4372/91/2728: 89). The outcome of this proposal is not known. 86. RGVA, 33387/3/987: 16–20. The report is not dated, but was written after the ‘‘Stalin constitution’’ of December 1936. It is colored by the atmosphere of the purges, claiming that in the mobilization staff ‘‘wrecker-Trotskyites have been exposed.’’ 87. The first category was allocated 1.8 million rubles, the second 2.5 million. See report of West Siberian executive committee dated September 13, 1936: RGVA, 40438/1/2217: 8– 15. In the North Caucasus military okrug in 1937 0.75 million rubles were spent in the North Caucasus region, and 2.3 million in the Azov-Black Sea region (presumably on the first category) (RGVA, 40438/1/2217: 247–241, dated March 15, 1937). 88. RGVA, 40438/1/2217: 23–5 (dated January 17, 1937). 89. RGVA, 40438/1/2217: 82–95 (minutes dated February 15, 1937; the meeting lasted only 1 hour 20 minutes, so the plans were evidently prepared beforehand). The planned expenditure in 1937 was only 630,000 rubles; 87 percent was to come from the local budget, and nearly all the remainder from the RSFSR budget. 90. RGAE, 4372/91/2968: 106–105 (dated April 29). 91. RGAE, 4372/91/2968: 161–155, issued in 60 copies. Enterprises were divided into ‘‘mobilized,’’ ‘‘equivalent to mobilized,’’ and ‘‘not mobilized,’’ with different arrangements applying for each category.

Published References Cherepenina, N. 2005. ‘‘The Demographic Situation and Health Care on the Eve of the War.’’ In Life and Death in Besieged Leningrad, 1941–1944, ed. John Barber and Andrei Dzeniskevich, 13–27. Basingstoke (England) and New York: Palgrave Macmillan. Cooper, Julian. 1976. ‘‘Defence Production and the Soviet Economy, 1926–1941.’’ Soviet Industrialisation Project Series no. 3. University of Birmingham, Centre for Russian and East European Studies. Davies, R. W. 1996. The Industrialisation of Soviet Russia, vol. 4: Crisis and Progress in the Soviet Economy, 1931–1933. Basingstoke (England): Macmillan. Davies, R. W., and Mark Harrison. 1997. ‘‘The Soviet Military-Economic Effort Under the Second Five-Year Plan, 1933–1937.’’ Europe-Asia Studies 49(3): 369–406. Gatrell, Peter. 1979. ‘‘Russian Heavy Industry and State Defence, 1908–1918: Prewar Expansion and Wartime Mobilization.’’ Ph.D. dissertation, University of Cambridge. Gerasimov, G. I. 1999. ‘‘ ‘Mobilizatsiia est’ voina . . .’ ’’ Voenno-istoricheskii zhurnal, no. 3: 2–13.

Planning for Mobilization

155

Harrison, Mark. 1996. Accounting for War: Soviet Production, Employment, and the Defence Burden, 1940–1945. Cambridge: Cambridge University Press. Harrison, Mark. 1998. ‘‘The Economics of World War II: An Overview.’’ In The Economics of World War II: Six Great Powers in International Comparison, ed. Mark Harrison, 1–42. Cambridge: Cambridge University Press. Ken, Oleg N. 2002. Mobilizatsionnoe planirovanie i politicheskie resheniia. Konets 1920-seredina 1930-kh gg. St. Petersburg: Evropeiskii universitet v Sankt-Peterburge. Khaustov, V. N., V. P. Naumov, and N. S. Plotnikova, eds. 2004. Lubianka. Stalin i Glavnoe upravlenie gosbezopasnosti NKVD. 1937–1938. Moscow: Mezhdunarodnyi fond ‘‘Demokratiia’’ Khlevniuk, Oleg V., A. V. Kvashonkin, L. P. Kosheleva, and L. A. Rogovaia, eds. 1995. Stalinskoe Politbiuro v 30-e gody. Sbornik dokumentov. Moscow: AIRO-XX. Losik, Aleksandr, and Aleksandr Shcherba. 2000. ‘‘The Defence-Industry Complex in Leningrad (1): The Interwar Period.’’ In The Soviet Defence Industry Complex from Stalin to Khrushchev, ed. John Barber and Mark Harrison, 173–83. London and Basingstoke: Macmillan. Meliia, A. 2004. Mobilizatsionnaia podgotovka narodnogo khoziastvo SSSR, 1921– 1941 gg. Moscow: Al’pina Biznes Buks. Mukhin, M. Iu. 2004. ‘‘Reformy upravleniia voennoi promyshlennost’iu SSSR i tempy rosta proizvodstva v 1936–1941 gg.’’ In Ekonomicheskaia istoriia. Ezhegodnik. 2004, ed. L. I. Borodkin and Iu. A. Petrov, 141–59. Moscow: Rosspen. Naumov, V. P., ed. 1998. 1941 god. 2 vols. Moscow: Mezhdunarodnyi fond ‘‘Demokratiia.’’ Pospelov, P. N., ed. 1974. Sovetskii tyl v Velikoi Otechestvennoi Voine. 2 vols. Moscow: Mysl’. Samuelson, Lennart. 2000. Plans for Stalin’s War Machine: Tukhachevskii and MilitaryEconomic Planning, 1925–41. London and Basingstoke: Macmillan. Simonov, N. S. 1996. Voenno-promyshlennyi kompleks SSSR v 1920–1950-e gody: tempy ekonomicheskogo rosta, struktura, organizatsiia proizvodstva i upravlenie. Moscow: Rosspen. Stone, David R. 2000. Hammer and Rifle: The Militarization of the Soviet Union, 1926– 1933. Lawrence: University Press of Kansas. Vannikov, B. L. 1969. ‘‘Oboronnaia promyshlennost’ SSSR nakanune voiny (iz zapisok narkoma).’’ Voprosy istorii 1969(1): 122–35. Vernidub, I. I. 2003. Boepripasy—frontu. Ocherki. Moscow: Sergiev Posad.

6

The Soviet Market for Weapons mark harrison and andrei markevich

We introduced Chapter 3 by noting that, in all countries, markets for military goods work poorly. This is to a large extent independent of the constitution of the state and the social and economic system. In all countries, whether ownership is private or collective, and whether rulers are democratic or authoritarian, the agents on each side of the defense market are powerful and well connected. On one side a senior minister manages a government monopsony: there is only one significant customer for such items as heavy artillery, aircraft, and battleships. On the other side is a charmed circle of big defense contractors. A few large-scale corporations supply such weapons; their ability to squeeze money out of government is augmented by the fact that they are too important for production, employment, and national security for the government to let them fail. As a direct result, defense markets everywhere are notorious for cost overruns, delayed deliveries, quality shortfalls, subsidies, and kickbacks. It would be a mistake, however, to conclude that defense markets everywhere are uniformly the same. Just among the market economies, national arrangements have been shown to vary significantly in the degree of competition, public accountability, rent seeking, and ‘‘softness’’ of budget constraints on defense suppliers (Eloranta 2002, 2008). The Soviet market for military goods also shows several unique and fascinating features; despite the fact that both buyer and sellers were state owned, so that it was only an internal market

156

The Soviet Market for Weapons

157

in the sense defined in Chapter 3, it supplied an army that won World War II and threatened the West for the next half century. Thus it is fully worthy of detailed study. In writing about the internal market for weapons we do not mean that there was a market relationship between the Army and Industry as units. At this level there was no market exchange but a political relationship between the defense minister and a few industrial ministers. The market tended to emerge at lower levels (see Figure 3.7) in which individual military purchasing administrations had to bargain in detail with individual defense factories. We suggest that the market had less scope to develop for products where models were established and were in serial production year after year so that each year’s contracts could be planned in advance on the basis of the previous year’s experience. But for many lines including aircraft, ships, tanks, and engines the Army was continually trying to place contracts for new or unique items. Innovation in military machine technologies seems to have been particularly rapid in the mid-1930s and this accelerated the year-to-year turnover of products (Davies and Harrison 1997). In such periods even the crudest version of directive planning was impossible because it was never clear beforehand who would produce what products and how many, to what quality standards, or at what price. This greatly extended the scope for market-oriented behavior. The most important problem in the Soviet military market was the quality of weapons. By ‘‘quality’’ we mean the observable characteristics of fabricated goods such as their reliability or performance. Both quantity and quality can be observed. But they differ in the ease with which each can be verified, or proved to a third party. Quantity is more easily verified, whereas verifying quality takes relatively much more time and effort.∞ The Soviet economy had a general problem with quality because, sheltered from competition and guaranteed economic survival by state plans, factory managers faced strong temptations to seek a quiet life for themselves and their employees by fulfilling the plan for least effort (Granick 1954; Berliner 1957). The authorities assigned plans in rubles of gross output subject to fixed plan prices and quality specifications (tekhnicheskie usloviia). Quality, however, was costly to the producer. As we now know, virtually everything in the Soviet command system that appeared fixed was negotiable in practice, including plans and prices. Once plans and prices had been written down, however, the main scope for the factory to economize on effort lay in finding ways to reduce quality that were hard to verify. Immediately upon transition to the command system, a rapid deterioration in product quality began that was eventually halted and reversed only with great difficulty (Davies 1989: 88–89, 313–14, and 384–85; 1996: 108, 394–95, 404, and 484).

158

Mark Harrison & Andrei Markevich

In the hope of limiting such producer opportunism the authorities relied first on industrial self-regulation. Thus, every factory had its own quality department, or OTK (otdel tekhnicheskogo kontrolia), responsible for ensuring that its products came up to standard. Not surprisingly, this was largely ineffective: managers had little incentive to make self-regulation stick, and the staff employed to carry out quality assurance typically saw themselves as lowstatus employees paid to provide a fig-leaf to cover up for management when things went wrong; when they tried to work professionally to external benchmarks, managers slapped them down.≤ Above the factory level, the ministers in charge of the supply of military goods had to account for their quality to Stalin and this forced them to care about quality; periodically, at least, they said that they did. When they spoke up for quality, they often made inspirational speeches and issued decrees about the enforcement of standards and benchmarks that were accompanied by fearsome threats of punishment for violation. In practice, however, the ministry had its own plan to fulfill; conscientious adherence to quality standards could threaten not only the incomes of workers and managers but also the authority and prestige of the minister. If the minister was for quality before the event, then after the event quantity became the important thing and quality was allowed to slide unnoticed. What means were available for the Soviet buyer to bring independent pressure to bear upon a poor-quality supplier? Under Soviet legislation of 1929, strengthened in December 1933 and July 1940, factory managers became criminally liable for negligence in relation to product quality. The problem lay not in the law but in its enforcement; in 1939, for example, the decree of December 1933 was already a dead letter (Solomon 1996: 144–47). The buyer could also claim a refund and seek damages through the civil arbitration courts. The buyer’s expected gain was limited, however, by two factors: the procedure was time consuming, and it opened up the buyer to retaliation by the seller in the future. This limited the buyer’s expected gain from appealing to higher authority. By focusing on the problem of quality we do not mean to imply that the Red Army’s military equipment was not good enough to fight wars and win battles. The quality of weapons has both economic and military aspects that are conceptually distinct, although practically related. The economic aspect of quality decides whether the equipment creates producer and consumer surpluses sufficient for both buyer and seller to be willing to agree the terms of an exchange beforehand and remain satisfied with the results afterward. The military aspect decides whether the buyer can use the weapons to beat the enemy. In World War II, Soviet weapons such as the T-34 tank, BM-13 ‘‘Katiusha’’

The Soviet Market for Weapons

159

rocket mortar, and Il-2 assault aircraft won a reputation for rugged serviceability and firepower. Militarily, they were good enough. This does not mean that they always performed according to contract. This chapter is about the economic aspect of quality: on what terms was Industry willing to provide it, and did the Army get what it paid for? The economic aspect of quality is important because, even if the weapons were militarily ‘‘good enough,’’ what price had to be paid to get them? This chapter is organized as follows. In the first part we set out a framework for analyzing the problem of quality in military markets, in which Industry ‘‘holds up’’ the Army for gain. The second part describes the main solution that the Army adopted, that of deploying its own supply enforcement agents throughout Industry. The third part describes these agents’ daily work, which brought them into frequent conflict with the industrial suppliers. The fourth part describes the result, which was a compromise over both the quantity and quality of goods accepted. The fifth part concludes.

The Hold-Up Problem in Defense Industry The ‘‘hold-up’’ problem provides a way of understanding quality issues in the Soviet market for weapons. A hold-up can arise wherever one partner must invest in an exclusive relationship with another in order to realize the benefits of a potential exchange.≥ In a market in which buyer and seller have an exclusive relationship, the hold-up can arise on either side or both sides at once, but in our case it will generally be one sided: the buyer cannot identify and select the best-matched seller without first undertaking a costly search, exchange of information, and negotiation. Suppose the buyer’s selection cost is S, which is also the cost of switching from one seller to another. Once formed, the relationship with the seller is then worth at least S to the buyer and this is what the buyer stands to lose if the initial relationship breaks down. S becomes part of the profit created by the relationship, but it need not accrue to the buyer. Who actually receives it will depend on post-contract renegotiation. The seller can hold up the buyer: by threatening to withdraw from the relationship, the seller can face the buyer with a potential loss at least equal to S so the buyer should be willing to pay the seller up to S to avoid this loss. The extent of the transfer will depend on the relative bargaining strengths of the two sides; the party with more to lose is more likely to lose it. The risk presented by the hold-up problem is that in order not to be held up and so make a loss, agents will avoid investing in the relationship-specific assets that make them vulnerable; as a result, society as a whole will lose the

160

Mark Harrison & Andrei Markevich

gains from trade. The hold-up problem is not without standard solutions, however, that should bring the incentives of the buyer and supplier back into alignment (Schmitz 2001). One is vertical integration, which brings the parties together under a single authority and completely replaces their market relationship by hierarchy. There are also intermediate solutions that retain the market relationship but regulate it by long-term contracts with some combination of joint financing of initial joint costs and contingent rules for distributing the subsequent benefits. Defense markets are generally thought to have a potential for hold-up problems (Rogerson 1994). First, there is an exclusive relationship, with only one buyer and little room, perhaps, for more than one supplier. Second, the relationship requires both sides to invest in it before gains can be realized. The Army must invest in selecting its suppliers and securing their goodwill; if a relationship breaks down it must start again, so its investment will be lost. Similarly, Industry’s firms must acquire the specialized capital assets required to produce the particular items that the Army alone wants; if the relationship breaks down these specialized assets will be less valuable in their best alternative use. Thus, both sides have something to gain and something to lose, and the result is that each can be held up by the other. In the Soviet case the hold-up problem was one sided. First, Soviet firms generally did not pay for capital goods that were free of charge to the user, the cost to society being met by grants from the state budget.∂ If we consider only the financial aspect, then the price of the firm’s specialized capital assets was zero.∑ This would weaken the hand of the Army and eliminate the scope for it to gain by holding up Industry. Industry could still hold up the Army, however, as long as the Army faced a positive switching cost S. This was likely less of an issue for established products that did not change from one year to the next. Once suppliers and their capacities were known from experience they could be written into plans, and this limited their bargaining power. More important for us is the case in which the Army needed a fresh source for a new product and had to expend resources on selecting the supplier and negotiating a deal. In Chapter 4 Andrei Markevich described how the Army was forced to wage a frustrating ‘‘contracts campaign’’ in order to place new orders with industrial suppliers each year. In an earlier study Harrison and Simonov (2000: 231) identified major obstacles as ‘‘the difficulty of finding willing suppliers of new defence products, and the desire of industry to secure a relatively homogenous assortment plan which would allow concentration on long runs of main products without a lot of attention to spare parts and auxiliary components, no matter how essential to the customer’’; the resistance of Industry could go so far as to leave significant

The Soviet Market for Weapons

161

orders completely unfilled. We conclude that switching costs would have left the Army vulnerable to a hold-up. Given this, what form of hold-up should we expect? Under Soviet arrangements, once higher-level plans had been issued and contracts agreed, the main opportunity for Industry lay in undershooting on quality, knowing that the Army could not take its business away. We illustrate this with an example that has three stages: a contract, the hold-up, and readjustment. Suppose the Army has a fixed budget of 100,000 rubles that it is willing to exchange with Industry for a particular gun. This gun is available in look-alike versions of two different qualities, ‘‘low’’ and ‘‘high.’’ Industry reports to the Army that in the low-quality version the gun will cost 500 rubles, but the high-quality one will cost a thousand. Officially Industry just needs to cover its costs, so if the Army pays 100,000, Industry will offer to deliver any mix of high and low quality that satisfies the condition (in thousands) 100 = 1 0 H = 0.5 0 L, using H and L for the numbers of high and low quality, respectively. In Figure 6.1 these relative costs are reflected in the gradient of the bold line labeled C, which gives the maximum amounts of either quality that Industry will write into a contract worth 100 thousand rubles; the C line has a downward slope of 1/2 because for every extra H in the contract Industry will offer two units less of L. The Army is willing to pay a thousand rubles each for high-quality guns but would pay only 250 for low-quality ones; it will promise its cash for any combination that meets the condition (in thousands) 100 = 1 0 H + 0.25 0 L. The fact that it values low-quality items below their production cost makes the Army willing to pay 100,000 provided that it receives only high-quality items. The thin line labeled V shows the combinations that the Army would accept as worth 100,000 and so embodies the Army’s relative evaluation of guns of different qualities; it has a downward slope of 1/4 because the Army would give up four L for every extra H in the contract. The V line touches Industry’s offer line at the vertical axis, and this makes both parties just willing to trade; they will exchange a contract for one hundred items of exclusively high quality at a thousand rubles each. Figure 6.2 shows the hold-up. After the event, Industry violates its contract; while sticking to the contract terms in quantity, it defaults on quality by following the arrow pointing southeast to x. The arrow has a downward slope of exactly forty-five degrees, meaning that Industry is substituting low-quality for high-quality guns, one for one. How far Industry can go is limited by the cost the Army would incur to select another producer. Suppose the Army’s switching cost is 45,000 rubles. Then Industry can cut the value of its delivery to the Army by up to this amount before the Army will tempted to break the contract. In this case, at x Industry mixes up to sixty low-quality guns with at least forty

162

Mark Harrison & Andrei Markevich

Figure 6.1. Before the Hold-Up: Contract

of high quality, keeping the sum of units at 100 as in the contract, and so fulfilling the plan in quantity if not in quality. The Army must now pay 100,000 for forty high-quality items worth 40,000 plus sixty low-quality items that it values at only 15,000, making its procurement worth only 55,000, so it has lost 45,000. The Army knows it has been cheated because quality is observable, but it can do nothing about it because quality is not verifiable in the dictator’s court; if it broke the contract, the Army would have to accept the equal or greater loss of having to find another supplier at short notice. Once the Army learns to anticipate such losses, what can it do? The standard solutions listed above involve market regulation by long-term contracts or market suppression through vertical integration. In the Soviet context we see that the standard solutions could not apply. Stalin ruled out vertical integration of the Army and Industry because he did not want to encourage the formation of a powerful military-industrial complex. The historical record shows that military interests advocated integration with the defense industry, but Stalin opposed it and quickly ruled it out. In 1927 army commanders Tukhachevskii, chief of the general staff, and Unshlikht, a member of the Revolutionary Military Council, sought powers for the Red Army over appointments to the defense industry, plans and reports of defense producers, and plans for capital investment in the industry (Samuelson 2000: 42–47; see

The Soviet Market for Weapons

163

Figure 6.2. After the Contract: Hold-Up

also Chapter 2). These proposals were rejected (Harrison and Simonov 2000: 230). Tukhachevskii’s subsequent resignation as chief of staff was most likely prompted by the failure of his ambition to control the defense industry (Samuelson 2000: 55–59). As for Stalin’s motivations, divide-and-rule was a basic mechanism on which he built his power, and this included keeping soldiers and industrialists at odds (Harrison 2003). An intermediate solution to the hold-up problem is long-term contracting. But again, this could not be applied in the Soviet context; the reason is that, as Andrei Markevich showed in Chapter 4, under Soviet rules all contracts were rewritten every year; no long-term contract was worth more than the paper it was printed on. More formally, the dictator could not credibly promise to uphold long-term agreements between the Army and Industry for sharing the gains from trade because he clearly had the power to break any contract and could not bind himself. In the absence of other solutions, Figure 6.3 shows what may happen next. The important thing is that at x, the Army’s loss exceeds Industry’s gain. This is because of the production of low-quality guns: it costs Industry 500 rubles

164

Mark Harrison & Andrei Markevich

Figure 6.3. After the Hold-Up: Readjustment

to produce every low-quality item, but this is twice what these items are worth to the Army. Thus the Army values the sixty low-quality guns it has received at only 15,000 rubles, but they cost Industry 30,000 rubles to produce, and the 15,000-ruble difference is a deadweight loss that benefits no one. As a result both parties could gain by raising quality. If they could agree to trade back along V% to the vertical axis, for example, Industry would cut its costs by another 15,000 without further loss to the Army, which would now receive only fifty-five guns, but all would be of high quality. Alternatively, trading back along C% would give the Army seventy guns, all of high quality, and cut the Army’s loss to 30,000 rubles, without detriment to Industry, whose costs would not change. Or they could agree to split the gain; the shaded triangle in the figure shows the scope for compromise, and the arrow pointing northwest shows the direction in which it lies. There is an obstacle, however. For Industry, the point of maintaining the combined amounts of H + L at 100 was to avoid a verifiable contract violation. Anywhere else in the shaded area than at x, Industry will underfulfill the plan in quantity, and the violation will be obvious: total output will be up to forty-five items short. The Army may wish to forgo some of these items in order to get higher quality overall. The danger for Industry is that the Army can afterward

The Soviet Market for Weapons

165

denounce it to the dictator for breaking the contract, putting Industry’s gain from the hold-up at risk. To eliminate the deadweight loss at x, Industry must bind the Army not to denounce it afterward, and the Army must be willing to be bound. There has to be a mechanism for collusion: the Army must join a conspiracy that hides not only the original hold-up operation (the shift to x) but also the subsequent readjustment that restores quality at the expense of quantity. Otherwise, both sides must accept the deadweight loss at x. In this chapter we will see how the whole thing worked in reality. The Army tried to reach across the internal market for weapons by deploying thousands of military engineers to the factories of the defense industry. These agents had a dual role. Their first duty was to prevent the Army from being held up and to enforce its contracts. They monitored the process of contract fulfillment with special regard to quality, and aimed to reject items for purchase when their quality fell below some threshold level. The work of the military agents made the quality of military goods more verifiable. When Industry sought to cut the supply of high-quality items, the Army sought to prevent their replacement by low-quality items, and this opened Industry up to penalties for defaulting on quantity. In practice, however, the military agents and their superiors tended not to make trouble for Industry over quantitative shortfalls. This suggests that Industry’s cooperation was available at a price: the Army had to accept shortfalls on quantity and help conceal them from the dictator’s prying eyes.

The Military Agents As the mixed economy of NEP gave way to the command system the Army had to face up to its adverse consequences. In 1930 a radical reform set the aim of achieving ‘‘a breakthrough in the work of industrial enterprises in fulfilling military equipment orders.’’ The reform entitled the Army to appoint special military agents (voennye predstaviteli, voenpredy) to regulate procurement from Industry.∏ The 1930 statute also defined the rights and obligations of Industry and the Army in relation to product quality. These were left largely unchanged in subsequent versions enacted in 1933/34 and 1939.π The statute of 1939 charged the military agents in industry with ‘‘observance of the process of manufacture of military products . . . , the technical acceptance of finished items, and monitoring the enterprises’ mobilization readiness.’’∫ Their responsibilities included checking that production adhered to technological standards and that enterprises fulfilled their plans; they were obliged ‘‘to report to the Red Army chief of armament through the chief of the appropriate equipment [purchasing] administration’’ concerning all shortfalls in suppliers’ fulfillment of military equipment orders: the use of substandard

166

Mark Harrison & Andrei Markevich

Figure 6.4. Principals and Agents in the Military Marketplace

materials, shortages of raw materials and semi-manufactures for the enterprise, departures from approved processes and blueprints, poor work by the factory OTK, missed deadlines for military orders, and so on.Ω To fulfill these obligations the military agents were endowed with rights of free access to the entire factory site at any time, day or night, and to all documentation relating to technology, production, and mobilization. The management was obliged to support the military agents with necessary accommodation and equipment. Faced with substandard products the military agents could halt acquisition and, if necessary, production, but they were prohibited from doing so if the purpose was to exert pressure on the management. Managers had no right to interfere directly in the work of the military agents, but they could appeal over their heads to higher authority. To protect their independence from management the military agents were paid only by the defense ministry and were prohibited from accepting rewards or benefits from the side of Industry. Figure 6.4 illustrates the structure of agency that resulted. During the 1930s the numbers of military agents appear to have risen dramatically, and their qualifications also improved. Within the defense ministry, separate chief administrations for artillery, the air force, chemical weapons,

The Soviet Market for Weapons

167

and so forth dealt with the purchase of specialized equipment (see Figure 6.3). Each maintained its own military agents at suppliers. Two factors swelled their numbers. First, the agents themselves were serving officers, but the defense ministry also engaged civilian employees to support them. Second, an enterprise that supplied more than one purchasing administration of the defense ministry had to accommodate agents from each of them, and this also added to numbers. At 16 factories in Iaroslavl’ in 1943, for example, a total of 144 agents worked on military acceptance, including 19 senior command staff, 30 middle ranking officers, and 89 hired employees. Some factories accommodated agents of up to five separate Army and Navy purchasing administrations.∞≠ The growth of numbers employed as military agents is hard to judge because we lack global figures for the early period. At the beginning of 1930 one of the Red Army’s purchasing administrations, that for military maintenance (voenno-khoziaistvennoe upravlenie), accounted for just 263 local procurement agents.∞∞ Numbers appear to have grown rapidly thereafter; by 1938 the total of military agents and their employees had reached two to three thousand, and more than 20,000 by 1940.∞≤ This growth probably reflected supply and demand. On the demand side the economy and especially its defense sector were expanding with exceptional rapidity (Davies and Harrison 1997). At first, demand outstripped supply; at the beginning of the decade skilled engineers were so scarce that recruiting standards had to be lowered to fill vacancies for military agents.∞≥ In 1933 the government admitted, ‘‘The defense ministry acceptance staff do not measure up to their job descriptions.’’∞∂ Frequent military complaints about the shortage of agents and the amount of overtime they had to work, leading to poor control of quality and deadlines, persisted through the mid-1930s.∞∑ Two factors eventually overcame this shortage. One was the expansion of Soviet higher education, which greatly augmented the supply of professionally qualified personnel. The other is that more privileged terms of employment were established to make up requirements by recruiting skilled civilian personnel.∞∏ In 1938 military agents’ pay was raised up to and subsequently beyond the level of Industry’s own quality staff; as numbers increased, their workload was also cut back.∞π The turnaround in relative pay and conditions evoked notable resentment among those employed in industrial self-regulation of quality. In October 1947 a meeting was held in the ministry of armament for factory OTK officials. According to one speaker, ‘‘A leading military employee [responsible] for a single product gets 1,400 to 1,500 rubles [monthly]. An OTK deputy [chief ] for metallurgy in charge of 17 workshops gets 1,350 rubles and an OTK head of workshop gets 900 rubles. This pay gap ensures they get people

168

Mark Harrison & Andrei Markevich

with more skills, higher discipline, and better training since these are all associated with high pay.’’∞∫ Another gave the average monthly pay of OTK staff at his factory as 400 rubles including bonuses, while hired employees of the military agents got 600 rubles and the officers up to 2,000 rubles.∞Ω A third compared wages in the OTK unfavorably not only with the earnings of the military agents but also with production workers’ pay. The basic pay for OTK workers equaled that of production workers, but the latter could expect large piece-rate bonuses whereas OTK staff got nothing for additional effort.≤≠ Industrial quality workers also complained about the military agents’ easy life. ‘‘Our team from the chief artillery administration comprised a lieutenantcolonel, a captain, and three hired staff. They needed 40 minutes to take ‘decisions’ and the rest of the time they could catch flies, sing songs, and undertake staff development.’’≤∞ This was not an urban myth; the Army considered it normal that ‘‘the workload of military product acceptance on military agents and their staff does not exceed 50 percent.’’≤≤ The generally privileged position of the military agents would appear to have been important in limiting their corruptibility; as we report elsewhere (Markevich and Harrison 2006), we have found little evidence of bribery, and some evidence that such cases were exceptional.

Into Battle with Industry In this section we analyze how the Army’s agents worked to enforce quality on Industry. Officially, the Army and Industry had common interests; disputes arose only because of ‘‘misunderstanding,’’ which could be overcome through procedures to identify and manage disagreements such as joint meetings.≤≥ In reality, however, ‘‘mutual relations of the factories with ministry of defense and [navy] representatives are unbearable.’’≤∂ Underlying this lay the hold-up problem between the Army and Industry, which gave rise to persistent antagonisms focused on the role of the military agent. The mutual attitudes of managers and military agents can be illustrated from both sides. A defense industry manager spoke up for Industry in 1928: ‘‘Less regulation. It is our misfortune that they regulate us so much.’’≤∑ Nearly a decade later, a shipyard worker told party activists: ‘‘The handover of vessels must be simplified. We are losing a lot of time doing unnecessary trials.’’ A military agent replied for the Army: ‘‘[The previous speaker] said that the trials are implemented in too much detail. But I say that detailed trials are essential. . . . We have to eliminate all defects from the key items through exhaustive trials.’’≤∏ Another military agent put it bluntly: ‘‘Don’t argue with us, just do what we say because we’re not making it up.’’≤π

The Soviet Market for Weapons

169

Those who spoke for Industry typically accused military agents of incompetence and having a lack of realism. ‘‘There are good acceptance agents but there are also agents who don’t understand the things they are supposed to accept. How can someone be a good acceptance agent if they tell him to deal with soap today, hay tomorrow, and belts the day after?’’≤∫ ‘‘If the [naval agency] is staffed with weak employees then they will set requirements wrongly. Often a ship isn’t handed over because there is more squabbling going on than work.’’≤Ω In a development once predicted by the political scientist David Holloway (1982: 325n.) they considered the agents to be useful only to exert pressure on their own subcontractors.≥≠ The agents themselves realized that Industry regarded them with contempt, as ‘‘blunderers who . . . give us nothing useful,’’ or ‘‘formalists who . . . shove spokes in our wheels’’ and so on.≥∞ This hostility arose because the military officers acted as the Army’s loyal agents. The chief instrument at their disposal for enforcing quality, and perhaps the only one that was effective, was their right to refuse to accept goods that were not up to standard. By rejecting the goods that Industry offered they threatened the ability of Industry to show compliance with supply plans and contracts. This was a powerful threat, but not as potent as might appear at first sight. In theory plan and contract violations could carry direct administrative and legal penalties. In practice, however, military agents rarely looked to higher authority to impose punishments for low quality, and when they did they were typically unsuccessful. In 1933, for example, a military agent tried to use the party committee of aircraft factory no. 24 to bring to account those responsible for ‘‘malicious toleration of defective parts,’’ but without success.≥≤ We have found only one case, that of naval armament factory no. 347, in which a military agent took the managers to court on criminal charges of supplying substandard goods; the court cast doubt on the accusations and the file was returned for further enquiries. A review by KPK, the ruling party’s ‘‘control’’ or audit commission described in Chapter 3, found that the judicial route was inappropriate and substituted dismissal for the criminal charges.≥≥ Financial penalties mattered more. When plans failed workers, managers, and ministerial officials lost bonuses; contract failures deprived the enterprise and ministry of revenue. Although it did not have the same significance as in a market economy, money did matter. Just as important, plan and contract violation attracted complaints and was a signal for investigation. For those to whom a quiet life mattered more than money, to underfulfill a plan or agreement usually led to unpleasantness and disruption. Other classic investigations confirm how important it was for Industry to avoid this by fulfilling the plan (Berliner 1957).

170

Mark Harrison & Andrei Markevich

The frequency with which Industry failed to fulfill the Army’s contracts is one measure of the military agents’ activism. At armament factories nos. 74 and 286 in 1946/47, for example, the share of output that the military agents rejected rose above 40 percent.≥∂ Military agents could reject the entire monthly output of a given factory—for example, that of defense industry factory no. 205 for March 1938 ‘‘in view of the totally unsatisfactory installation of electric plugs in all articles supplied.’’≥∑ Enquiries into the failure of defense orders by KPK often laid the fault at the military agents’ door. According to KPK records, in January and February 1934 the Tula gun factory produced 3,000 carbines and 106 ShKAS machine guns, but only 800 rifles were accepted for the defense ministry and no machine guns at all. The 3,000 carbines ‘‘were presented for acceptance 23,000 times, almost 8 times per carbine on average.’’≥∏ KPK auditors concluded that ‘‘discord between management and representatives of military acceptance on the score of product quality’’ lay behind persistent plan breakdowns.≥π In 1944 the KPK official for the Khabarovsk region reported that ‘‘vexatious litigation,’’ with the managers on one side and the OTK and military agents on the other, had taken hold of aircraft factory no. 126 on the issue of parts and components that did not conform to the blueprints. ‘‘These disputes . . . sometimes drag on for weeks . . . while business stands still.’’ In the first quarter of 1940 rejected goods amounted to 375,000 rubles.≥∫ The military agents’ screening could outdo OTK control by an order of magnitude. Among the aircraft that the OTK of factory no. 126 passed in 1940, the military agent found up to eighty defects.≥Ω In the first nine months of 1940, of 6.6 million shell cases produced at munitions factory no. 184 the OTK scrapped 2.74 percent; after that, the military agent scrapped a further 10.5 percent.∂≠ Not all military agents refused to compromise on quality issues or demanded unconditional adherence to agreed standards; in 1937, for example, naval officers warned against the common practice of accepting vessels without the necessary technical documentation.∂∞ KPK factory audits of the period report other failures of a similar type. At a naval armament factory the military agent was reported to have accepted substandard mines.∂≤ At aircraft factory no. 39 in 1939, it was said, ‘‘[The] senior military agent . . . and regional military engineer . . . have impermissibly weakened control over the quality of accepted goods, established the practice of accepting unfinished aircraft subject to written factory guarantees, and left aircraft armament unchecked.’’ Aircraft with unserviceable machine guns, and bombers with engines that suffered overcooling when cruising in level flight, were accepted and put into service. Iron replaced chrome-molybdenum for rivets with the silent

The Soviet Market for Weapons

171

consent of the military acceptance officers, and so forth. Significantly, chief of the air force purchasing administration Efimov was accused of colluding with these malpractices: ‘‘Not only did [he] not take measures to restore order but [he] even suppressed criticism of the defects, describing the communists who raised the criticisms as ‘cry-babies’ and threatening them with dismissal.’’∂≥ Efimov was one of the top supply officials in the defense ministry; if this was his attitude, the case of factory no. 39 cannot have been unique. In the years of rapid prewar expansion, equipment supplied to military units often turned out to be unfit for service even though the military agents had previously passed them as acceptable. In March 1938, for example, the air force complained to defense industry minister Mikhail Kaganovich about numerous defects in I-16 fighters and UTI-4 trainers, and requested that the factories themselves dispatch special repair brigades to military units.∂∂ Military agents’ standards appear to have slipped markedly with the outbreak of war. The records of tank factory no. 183 show that in every year of the war more than half the tanks taken into military service were registered with one or more defects at the point of acceptance. The worst year was 1942, when only 7 percent were reported free of defects. The high rate of defects at this stage of the war was attributed to the fact that factory no. 183 was newly assembled out of plants evacuated from five locations in the war zones. The frequency of defects fell back in subsequent years, however, as wartime output expanded and experience accumulated.∂∑ The situation was no better elsewhere. For example, of the T-34 tanks that factory no. 174 presented to the military agent in August 1943, only 4.5 percent were free of defects and more than half had three defects or more. From April to August 1943 roughly a tenth of vehicles were in such a bad state that they were returned to the factory for remedial work before retesting.∂∏ The same happened to more than 20 percent of tanks supplied by the Kirov factory in Cheliabinsk.∂π Subject to repeated testing, however, military agents eventually accepted virtually all tanks produced; across the industry, in July 1943, tanks accepted ran at 99 percent of those supplied.∂∫ It seems that in wartime, at least, only totally unserviceable goods were rejected; most equipment was taken for the army following retesting, defects and all. The result was a steady flow of complaints by military units. In April and May 1943 the Army made seventy-seven complaints to Industry about cracks in tank bodies.∂Ω A recent study of the tank industry by the young Russian historian Arsenii Ermolov (2004) provides further detail. During the war 12 percent of all tank losses were ascribed to technical faults; this proportion was higher in 1942 and 1943. In the summer of 1942 the military agent at tank factory no. 183 found that every tenth new vehicle sent to the front was

172

Mark Harrison & Andrei Markevich

being reported as needing repair. In his view this understated the true position: only one-quarter of actual defects was being reported; military units were either tolerating the remainder or fixing them at their own expense. A senior officer responsible for armored equipment recalled, ‘‘In one particular engagement on the Stalingrad front, when our tank numbers were evenly matched with the Germans, only one quarter of our tanks actually took part—say, 100 out of 400 tanks.’’ The standards that military agents applied to armament were probably more stringent than those for personal kit and transport stores. Although the gap is inherently difficult to measure, KPK documents give the impression that military agents allowed more defects in soldiers’ clothing and footwear and that their superiors in the central supply staff of the defense ministry agreed with this. A KPK audit of 1937 found, ‘‘The army is supplied with footwear made out of leather of completely unsatisfactory quality.’’ ‘‘Neither the ministry for light industry and its plant managers, nor the Red Army administration for supply of troops is giving the necessary attention to the quality of military footwear.’’ ‘‘[Each] military agent in the localities has to service four to six or more production establishments and cannot systematically check up on the footwear plants.’’ At some factories up to half the footwear that the military agents had accepted was substandard. ‘‘The [supply administration] has systematically tolerated a lowering of requirements in the footwear supplied, with regard to both soles and materials.’’ In this case the mutual rights and responsibilities of buyer, seller, and military agents were undefined because the draft regulations had been under consideration by the ministry for light industry for two years.∑≠ The situation persisted for three more years: in 1940 a KPK report found, ‘‘Defense ministry acceptance agents in factories and plants [of the light and textile industries] are tolerating substandard items on a massive scale.’’∑∞ Why did military agents not rigorously enforce defense ministry guidelines on substandard equipment? The main reason is that out of loyalty to the Army, they could not reject everything that Industry offered them. One of the OTK chiefs at the armament ministry meeting held in October 1947 let the truth slip: ‘‘I don’t agree that we cannot come to terms with the military acceptance staff. . . . They are state officials the same [as us] and they are responsible for equipment orders to the same extent [as us].’’∑≤ The same logic also led their chiefs on the supply staff to collude with them and not punish them for lowering standards. If agents demanded inflexible adherence to standards, they laid themselves open to criticism for excessive zeal or caution. For example, a KPK factory report of 1940 condemned the OTK and military agent at aircraft factory no.

The Soviet Market for Weapons

173

126 for ‘‘a tendency to over-insurance.’’∑≥ Surveying the work of military agents in 1943 the KPK demanded, ‘‘The military agent should in most cases rule on the acceptability of one or another deviation [from standards] so as not to delay products for the front.’’∑∂ Thus, while military agents may have tried not to accept goods that were clearly unserviceable, there was pressure on them to tolerate some level of defects. It may be asked why, through repeated exchanges, Industry and the Army did not learn each others’ preferences and resources so as to converge on a mutually beneficial equilibrium in which the Army obtained goods of the quality it required and Industry was able to fulfill its plans without the need for costly rejections and plan failures. One reason may be that the annual process of plan and contract revision prevented the hold-up problem from being solved by long-term contracting. Instead, the planning process focused each side on extracting the maximum short-term advantage from the other, year after year. Another reason was that learning was inhibited by very rapid change in the product assortment: in the 1930s, for example, one year’s procurement of aircraft rarely replicated the purchases of the year before to any significant extent. In this context the Army viewed the results of its procurement apparatus as thoroughly worthwhile. Even in wartime when the front line desperately needed career officers, the defense ministry refused to cut numbers of military agents by merging its specialized purchasing administrations into one.∑∑ There were at least three wartime proposals to do this, one in 1941 and two in 1943; the ministry rejected them all on the grounds that ‘‘creating a unified apparatus for regulation and acceptance of military production, independent of the chief administrations, would lead to a loss of accountability in regulating the production of armament and munitions, and to a reduction in their quality.’’∑∏

Bargaining for Quality In setting out the hold-up problem in the market for weapons, we made two predictions. First we suggested that when held up by Industry, the Army would find it more important to uphold quality than quantity. This was shown in Figure 6.3, in which the Army’s valuation of quality, given Industry’s postcontract offer at point x, exceeded that of Industry. Second, we suggested that Industry and the Army would be jointly interested in collusion to conceal the resulting shortfall on quantity. Consistent with these expectations, we find that military agents typically took a harder line over quality than quantity. They were ready to offer some leeway to Industry over quantitative fulfillment as the price for maintaining quality. The outcome was a bargain that fell short of the

174

Mark Harrison & Andrei Markevich

initial contract but was more efficient than Industry’s initial post-contract offer. The Army was still held up, but less inefficiently than would appear at first sight, and in return allowed its agents to help conceal Industry’s otherwise verifiable shortcomings. Procurement delays were often concealed. The KPK archive contains many cases of reports falsified by both civilian and defense enterprises. The usual form was to exaggerate output over the accounting period by including pripiski, goods that did not exist yet but would be produced in the next period. Pripiski allowed the enterprise to claim fulfillment of the plan and entitlement to a bonus by ‘‘borrowing’’ future output. This practice involved criminal deception. A single enterprise could not undertake it successfully in isolation; therefore, ministerial superiors had to know about it and the customer had to go along with it in silence. The wider the circle became, the greater were the risks of disclosure. Despite such risks, however, in the seller’s market for civilian goods the power of suppliers was often enough to win the cooperation of both superiors and purchasers (Berliner 1957). Pripiski were widespread, also, in the Soviet defense industry. A KPK report of 1946, for example, claimed that a tank factory director was ‘‘systematically engaging in the pripiska of goods that have not finished production’’ and that his chief administration, although aware of this, ‘‘has not only not prevented but has even rewarded it.’’∑π Similarly, the KPK found that in 1944 the relevant administration of the armament ministry told a factory director ‘‘to report inflated information to the ministry.’’∑∫ In September 1944 the KPK acknowledged that pripiski were widespread: in 1943 and 1944 an armament factory had ‘‘continually reported falsely inflated information about the fulfillment of the factory’s program, typically using from 5 to 20 days of the following month to complete production’’; an aircraft factory had reported ‘‘incorrectly inflated information about plan fulfillment’’ in 1943 and for the months of January, February, and March 1944; the managers of a tank factory ‘‘have also been deceiving the government and ministries by reporting false information on the fulfillment of the production program.’’∑Ω There were even pripiski in a vehicle repair factory of the defense ministry itself; the ministry’s vehicles administration, though ‘‘aware of all the factory’s shortfalls and lack of management, took no measures to overcome them.’’∏≠ Widespread pripiski indicate a systematic tendency for Industry to ignore delivery deadlines: goods were regularly delivered to the Army a month or more late. The military agents could never have been unaware of this. In the Western literature there have been divergent views on whether Soviet military agents would have colluded with pripiski for the sake of maintaining the

The Soviet Market for Weapons

175

producer’s goodwill. Arthur J. Alexander (1978: 59n) thought this likely, while the Soviet émigré Mikhail Agursky, writing with Hannes Adomeit (1978: 23), judged it improbable. In fact Alexander was right: military agents virtually never took action to enforce deadlines. Of all the cases of pripiski that the KPK uncovered, only two were reported by military agents. In September 1941 a military engineer reported an unacceptable delay in an order for gas protection equipment placed with the ministry of general engineering.∏∞ Intervention by the KPK secured a new deadline for the order, but no penalty for the delay. In 1943 a military agent and his senior technician reported on ‘‘deception and irregularities’’ at an electrical factory; this led to a special audit commission that confirmed the various violations.∏≤ External KPK auditors themselves uncovered other pripiski. When they did so, they found that the military agents had colluded tacitly or openly in the deception. In 1944, for example, the military agent had joined the director of an armament factory in signing a cable reporting 101.5 percent fulfillment of the April program when both knew this to be false because it took part of the May program into account. Significantly, higher officials representing both Industry and Army had approved the pripiska by April 30.∏≥ They justified this on the basis of precedent; the defense official noted that he had approved similar arrangements in other cases ‘‘to avoid a breakdown of the plan and provision for the needs of the troops.’’∏∂ It was the same in the tank factories. In 1942 the KPK officer for the Sverdlovsk district found evidence of largescale pripiski for September, October, and November at the Uralmash factory, not just ‘‘with the ministry’s knowledge’’ but ‘‘on the instruction’’ of the minister and deputy minister, most of which the military agent went along with.∏∑ In short, deadlines for the supply of armament seem to have caused little anxiety to military agents, and even their superiors were ready to approve a degree of delay. They did have to look as if they supported firm deadlines. This led them to collude with enterprise managers in falsifying reports of plan fulfillment. In return, they gained cooperation over quality. Military marketplaces display obvious inefficiencies under most institutional arrangements, but the internal defense market of the Soviet Union was characterized by monopoly and exclusivity to an unusual degree. This presents a particular problem in the scope for one side to hold up the other. We have shown that in the Soviet defense market it was the seller, Industry, that was best placed to hold up the buyer, the Army. The form that the hold-up typically took was for Industry to default on quality. This hold-up problem could not be resolved by the conventional means recommended by economic theory: vertical integration was not in Stalin’s

176

Mark Harrison & Andrei Markevich

political interest, and long-term contracting was ruled out by the discretionary logic of command planning under a dictator. Instead, the Army sought to solve the problem by deploying agents through Industry to verify quality and reject substandard goods, threatening Industry with an easily verifiable shortfall on quantity. The Army was prepared to pay tens of thousands of agents and pay them well for their loyalty, even in the midst of a total war. These agents, however loyal, still had to reach a compromise with Industry. In the typical bargain the military agents agreed to overlook quantity violations in return for greater cooperation on issues of quality. Notes The authors thank Paul R. Gregory, Michael Waterson, and participants in the University of York Centre for Historical Economics and Related Research seminar, in particular Keith Hartley, for comments and advice. 1. In this chapter we look at the problem that arises when quality is observed before purchase but cannot be verified: the buyer is aware of quality defects, for example, but cannot prove them to a third party. Markevich and Harrison (2006) have looked at the further problem that arises when quality is also costly to observe so that the buyer may not know the quality of what is bought until after purchase; this problem is also common in defense markets. 2. Harrison and Simonov (2000) and Markevich and Harrison (2006) discuss industrial quality self-regulation at the factory and ministerial levels in more detail, including the interests of and positions adopted by the industrial ministers themselves. 3. Goldberg (1976: 439) provides the original formulation; see also Williamson (1985: 61–63). 4. At the end of the 1980s the U.S. Department of Defense was doing the same for a substantial proportion of private-sector defense-related investment and R&D expenditures in order to overcome defense contractors’ fears of being held up by the government, according to Rogerson (1994: 67–68). During World War II, for the same reason, the U.S. Defense Plant Corporation and other federal agencies provided and afterward wrote off capital facilities for war production to the private sector that Robert J. Gordon (1969) valued at $45 billion (at 1958 prices). 5. Their true cost to the firm may have been greater than this implies because, as Berliner (1957) showed, the firm had to negotiate capital grants with higher authority, and the negotiation took time and effort and required the expenditure of goodwill. In recent research Gregory and Lazarev (2002) have demonstrated how Soviet firms had to bargain for a specific class of capital goods: motor vehicles. 6. RGVA, 33991/1/65: 7–8 (1930). Harrison and Simonov (2000: 229) have described how this arrangement emerged from the pre-revolutionary procurement system. 7. GARF, 8418/8/175, ff 10–14 (decree of the Council of Labor and Defense, November 28, 1933; decree of the ministries of defense and heavy industry, August 4, 1934); 8418/23/314: 1–5 (decree of the Defense Committee, July 15, 1939). 8. GARF, 8418/23/314: 2 (July 15, 1939).

The Soviet Market for Weapons

177

9. RGVA, 33991/1/65: 11 (March 1930). 10. Hoover/RGANI, 6/2/49: 8 (July 7, 1943). 11. RGVA, 47/5/207: 1 (1930). 12. 1938: a document dated April 16 of that year (GARF, 8418/22/508: 6) gives the number of locally hired employees of military and naval agents as 1,695; the serving officers can hardly have exceeded this number. 1940: Harrison and Simonov (2000: 229). 13. RGVA, 33991/1/65: 1 (February 27, 1930). 14. GARF, 8418/8/175: 10–12 (November 28, 1933). 15. GARF, 8418/22/508: 8 (May 29, 1938). 16. GARF, 8418/8/175: 3 (August 4, 1934). 17. GARF, 8418/22/508: 1 (June 5, 1938). 18. RGAE, 8157/1/4105: 102 (Zvonarev, October 21, 1947). 19. RGAE, 8157/1/4105: 140 (Dovzhenko, October 21, 1947). 20. RGAE, 8157/1/4105: 110, 112 (Koloskov, October 21, 1947). 21. RGAE, 8157/1/4105: 203 (Dul’chevskii, October 21, 1947). 22. Hoover/RGANI, 6/2/49: 8 (July 7, 1943). 23. ‘‘Common interests,’’ RGVA, 47/9/83: 102 (Budnevich, 1928); RGAE, 8183/1/146: 81 (Kudak, April 13, 1937). ‘‘Mutual misunderstanding’’ to be overcome through ‘‘joint meetings,’’ RGAE, 7515/1/403: 180 (Kulik to M. Kaganovich, February 7, 1938). Advocating ‘‘joint meetings’’ ten years previously, RGVA, 47/9/83: 96 (Dybenko, 1928). 24. RGAE, 8183/1/146: 80 (Kudak, April 13, 1937). 25. RGVA, 47/9/83: 30 (Penin, 1928). 26. RGAE, 8183/1/146: 53–53ob (shipyard worker Serdiuk versus naval agent Aliakrinskii, April 13, 1937). 27. RGAE, 8183/1/146: 39 (Blagoveshchenskii, April 13, 1937). 28. RGVA, 47/9/83: 23 (Bobrov, 1928). 29. RGAE, 8183/1/146: 48 (Serdiuk, April 13, 1937). 30. For example, defense industry minister Kaganovich wrote to chief of the Red Army artillery administration Kulik asking him to tighten up the work of military agents at engineering factories that were supplying defective shell casings to defense factory no. 12 (RGAE, 7515/1/404: 247, June 20, 1938). 31. RGAE, 8183/1/146: 80 and 39 (April 13, 1937: ‘‘Blunderers,’’ Kudak; ‘‘formalists,’’ Blagoveshchenskii). 32. Hoover/RGANI, 6/1/91: 10 (March 17, 1934). 33. Hoover/RGANI, 6/6/1616: 128 (May 13, 1941). 34. RGAE, 8157/1/4105: 213 (Mandich, October 21, 1947). 35. RGAE, 7515/1/404: 158 (Savchenko to M. Kaganovich, 1938). 36. Hoover/RGANI, 6/1/22: 34 (March 7, 1934); emphasis in the original omitted. 37. Hoover/RGANI, 6/1/22: 36 (March 7, 1934). 38. Hoover/RGANI, 6/2/27: 108–109 (July 29, 1940). 39. Hoover/RGANI, 6/2/27: 108 (July 29, 1940). 40. Hoover/RGANI, 6/2/34: 158–159 (December 27, 1940). 41. RGAE, 8183/1/146: 38 (Blagoveshchenskii, April 11–13, 1937). 42. Hoover/RGANI, 6/6/1616: 127 (May 13, 1941). 43. Hoover/RGANI, 6/2/17: 47 (KPK bureau decree, December 3, 1939).

178

Mark Harrison & Andrei Markevich

44. RGAE, 7515/1/404: 4–6 (March 29, 1938). 45. RGAE, 8798/4/17: 231–232 (‘‘History of Tank Factory no. 183,’’ manuscript). 46. RGAE, 8752à/4/293: 180, 182 (August 11, 1943). 47. RGAE, 8752à/4/293: 188 182 (August 11, 1943). 48. RGAE, 8752/4/293: 66 (August 11, 1943). 49. RGAE, 8752/4/293: 114 (August 11, 1943). 50. Hoover/RGANI, 6/1/72: 77, 82–84 (June 10, 1937). 51. Hoover/RGANI, 6/2/250: 41–42 (May 14, 1940). 52. RGAE, 8157/1/4105: 136 (Dovichenko, October 21, 1947); emphasis added. 53. Hoover/RGANI, 6/2/27: 109 (June 29, 1940). 54. Hoover/RGANI, 6/2/49: 9 (July 7, 1943). 55. Hoover/RGANI, 6/2/49: 8–10 (July 7, 1943). 56. Hoover/RGANI, 6/2/49: 9 (July 7, 1943). 57. Hoover/RGANI, 6/2/98: 81, 85 (August 2, 1946). 58. Hoover/RGANI, 6/2/67: 11 (1944). 59. Hoover/RGANI, 6/6/1583: 10–13 (July 15, 1944). 60. Hoover/RGANI, 6/6/1583: 31 (October 26, 1948). 61. Hoover/RGANI, 6/6/47: 18 (September 29, 1941). 62. Hoover/RGANI, 6/2/55: 1–2 (KPK bureau decree, October 28, 1943). 63. Hoover/RGANI, 6/2/63: 159–160 (June 5, 1944). 64. Hoover/RGANI, 6/2/63: 21 (July 8, 1944). 65. RGAE, 8752/4/108: 151–151ob (December 7, 1942).

Published References Agursky, Mikhail, and Hannes Adomeit. 1978. ‘‘The Soviet Military Industrial Complex and Its Internal Mechanism.’’ National Security Series no. 1/78. Queen’s University, Center for International Relations, Kingston, Ontario. Agursky, Mikhail. 1976. ‘‘The Research Institute of Machine Building Technology.’’ Soviet Institution Series no. 8. Hebrew University of Jerusalem. Alexander, Arthur J. 1978. Decision Making in Soviet Weapons Procurement. Adelphi Paper no. 147 8. London: International Institute for Strategic Studies. Berliner, Joseph S. 1957. Factory and Manager in the USSR. Cambridge, MA: Harvard University Press. Davies, R. W. 1989. The Industrialisation of Soviet Russia, vol. 3: The Soviet Economy in Turmoil, 1929–1930. Basingstoke (England): Macmillan. Davies, R. W. 1996. The Industrialisation of Soviet Russia, vol. 4: Crisis and Progress in the Soviet Economy, 1931–1933. Basingstoke (England): Macmillan. Davies, R. W., and Mark Harrison. 1997. ‘‘The Soviet Military-Economic Effort Under the Second Five-Year Plan (1933–1937).’’ Europe-Asia Studies 49(3): 369–406. Eloranta, Jari. 2002. ‘‘The Demand for External Security by Domestic Choices: Military Spending as an Impure Public Good Among Eleven European States, 1920–1938.’’ Ph.D. dissertation, European University Institute, San Domenico di Fiesole. Eloranta, Jari. 2008. ‘‘Rent Seeking and Collusion in the Military Allocation Decisions of Finland, Sweden, and the UK, 1920–1938.’’ Economic History Review, forthcoming.

The Soviet Market for Weapons

179

Ermolov, Arsenii. 2004. ‘‘Narodnyi komissariat tankovoi promyshlennosti SSSR v gody Velikoi Otechestvennoi voiny. Struktura i deiatel’nost.’ 1941–1945 gg.’’ Moscow State University: Candidate of Historical Sciences Dissertation. Moscow. Goldberg, Victor P. 1976. ‘‘Regulation and Administered Contracts.’’ Bell Journal of Economics 7(2): 426–52. Gordon, Robert J. 1969. ‘‘$45 Billion of U.S. Private Investment Has Been Mislaid.’’ American Economic Review 59(3): 221–38. Granick, David. 1954. Management of the Industrial Firm in the USSR. New York: Columbia University Press. Gregory, Paul R. 1990. Restructuring the Soviet Economic Bureaucracy. New York: Cambridge University Press. Gregory, Paul R., and Valery Lazarev. 2002, ‘‘The Wheels of a Command Economy: Allocating Soviet Vehicles.’’ Economic History Review 55(2): 324–48. Harrison, Mark. 2003. ‘‘Soviet Industry and the Red Army Under Stalin: A MilitaryIndustrial Complex?’’ Les Cahiers du Monde russe 44(2–3): 323–42. Harrison, Mark, and Nikolai Simonov. 2000. ‘‘Voenpriemka: Prices, Costs, and Quality Assurance in Interwar Defence Industry.’’ In The Soviet Defence-Industry Complex from Stalin to Khrushchev, ed. John Barber and Mark Harrison, 223–45. Basingstoke (England): MacMillan. Holloway, David. 1982. ‘‘Innovation in the Defence Sector.’’ In Industrial Innovation in the Soviet Union, ed. Ronald Amann and Julian Cooper, 276–367. New Haven: Yale University Press. Markevich, Andrei, and Mark Harrison. 2006. ‘‘Quality, Experience, and Monopoly: The Soviet Market for Weapons Under Stalin.’’ Economic History Review 59(1): 113– 42. Rogerson, William P. 1994. ‘‘Economic Incentives and the Defense Procurement Process.’’ Journal of Economic Perspectives 8(4): 65–90. Samuelson, Lennart. 2000. Plans for Stalin’s War Machine: Tukhachevskii and MilitaryEconomic Planning, 1925–41. London and Basingstoke: Macmillan. Schmitz, Patrick W. 2001. ‘‘The Hold-Up Problem and Incomplete Contracts: A Survey of Recent Topics in Contract Theory.’’ Bulletin of Economic Research 53(1): 1–17. Solomon, Peter H. 1996. Soviet Criminal Justice Under Stalin. Cambridge: Cambridge University Press. Williamson, Oliver E. 1985. The Economic Institutions of Capitalism. New York: Free Press.

7

The Market for Labor in the 1930s: The Aircraft Industry mikhail mukhin

In the 1920s aircraft manufacture was everywhere a cottage industry dominated by artisan methods of small-scale manufacture. The Soviet industry had an annual capacity to produce aircraft and engines in the low hundreds. In the next decade this industry was the subject of two sharply opposing pressures. Both emanated from the leaders of the government and the industry, who wanted two things at once. On one side they demanded the enhancement of technology, rising levels of skills and qualifications, the growth of productivity and efficiency in the use of resources, and improved adherence to enhanced standards. In the mid-thirties the industry’s profile moved rapidly away from the plywood and canvas biplanes of the Great War to the highpowered metallic monoplanes of the next war. On the other side the authorities pushed insistently for rapidly rising quantities of output. As Figure 7.1 shows, the raw quantities of aircraft and engines climbed at a dizzying rate. There was a setback in 1935, not only in aviation but throughout the defense industry, disrupted by sweeping modernization of military equipment and production technologies (Davies and Harrison 1997). After that, growth was resumed at a high rate, and by 1939 output exceeded 10,000 airplanes a year, all of which were required for military use. The industry struggled, however, to deliver rising numbers and rising standards at the same time.

180

The Market for Labor in the 1930s

181

Figure 7.1. Soviet Aircraft and Engines, 1930–40 (units) Note: In principle the gap between production and defense procurement should be explained by civilian purchases (including purchases for the internal security forces of the interior ministry), the surely negligible balance of foreign purchases, and changes in inventories. The sources are independent, however, and accounting discrepancies may explain the apparent small excess of defense procurement over production in 1939 and 1940. Sources: Production of aircraft and engines: Kostyrchenko (1992: 329 and 432–33). Aircraft procurement: GARF, 8418/25/14: 2–3.

This chapter recounts that struggle from the point of view of the workforce and its conditions of pay and employment. It is said that it is not weapons that fight, but people. By the same token it is not factories that build aircraft, but the people in them. My research relies on a broad range of evidence. Although some is already in the public domain, including official histories, memoirs, and the research of other scholars, especially that carried out since the veil of secrecy was lifted at the end of the 1980s, its core is primary documentation from the central state and party archives, the central archives of the defense and aircraft industries, and local archives of the aircraft factories and their party organizations. This documentation is introduced into scholarly circulation for the first time. Part 1 of the chapter sets out the problem of ‘‘technological culture’’ in the eyes of the industry’s leaders. Part 2 describes the main trends in pay and productivity of the aviation workforce through the mid-1930s. Part 3 adds to this the results of the vast spurt of effort in the last years before the war. Part 4 summarizes the conditions found in a single factory at the end of the decade, taken as a microcosm of the industry. Part 5 concludes.

182

Mikhail Mukhin

The Technological Culture Technological culture, by which I mean the degree of conformity to worldwide best practices, is fundamental to the aviation industry. Within a dozen years Soviet aeroengineering changed from a cottage industry based around a few artisans to a large-scale operation with factories and production lines for serial manufacture. This involved much more than just establishing a number of enterprises and hiring hundreds of craft workers so as to build a few hundred aircraft a year. It demanded a shift to a much higher level of technological culture. BARANOV’S DIAGNOSIS

The starting point was not high. It was difficult for the industry to conform to high standards when the standards themselves were soft. The frequency of alterations to designs and processes was an important influence. Even factory no. 1, the standard bearer of Soviet aircraft manufacture, suffered continually from design variations that often had to be corrected for previous ‘‘corrections.’’ After the factory had started production of the R-1 reconnaissance aircraft, for example, it received a directive to replace the wooden undercarriage with a metallic assembly. An instruction to overturn the previous directive and revert to wooden undercarriage naturally arrived only after the factory had converted to the metallic one. The drawings for the I-2b fighter reached the factory nine months late; five hundred errors had been identified by the time production started. There were two sides to the process. The purchaser was often a source of delay: the factory produced a batch of 152 I-3 fighters based on a design approved by the air force; then the defense ministry required three hundred design modifications before it would accept the aircraft. Not surprisingly, a considerable proportion of the batch ended up in the scrapyard.∞ This hardly encouraged the workers to take responsibility for outcomes: ‘‘Why bother? We’ll have to do it all over again next month.’’ It also hindered the stabilization of production processes for a given model. But the producer was also often at fault. The rapid expansion of aircraft production in the late 1920s and early 1930s inevitably diluted the stratum of skilled workers. Because they were in short supply, their employment share tended to shrink. In 1929 highly skilled workers accounted for 8.5 percent of the industry workforce and semi-skilled workers for 48 percent. By 1931 these shares had shrunk to 2 percent and 28 percent, respectively (Talanova 1999: 88). The skill shortage encouraged factories to adapt designs to their own technological capacities, usually to the detriment of design quality.

The Market for Labor in the 1930s

183

Air force chief P. I. Baranov offered a penetrating diagnosis of the problem.≤ In 1921 the Soviet aircraft industry had been rebuilt from nothing around copied Western designs, the R-1 and U-1 aircraft and M-2, M-5, and M-6 engines; it has taken the industry eight years to get them into production and even now there are hold-ups everywhere and continual demands from industry to reduce quality standards. . . . From a design standpoint they [the aircraft and engines] are extremely complex and are not fully understood; in some aspects of their operation they yield only to approximate calculation . . . it takes three to four years for the sort of aircraft the air force needs to get from the commissioning of a design to a prototype and industrial production. In the context of developing Soviet aviation, this lag is completely unacceptable. At this rate an aircraft with design features that would be advantageous in combat would be obsolete by the time it left the factory and suitable at best for training.

The process of developing a new model, according to Baranov, actually went like this: so long as the test results showed some approximate correspondence with the required ‘‘tactical-technical characteristics’’ the air force officials gave the nod for it go into serial production, hoping to narrow the gaps, as one would say today, ‘‘interactively.’’ But this often did not work out, leading to procurement refusals, demands for remedial manufacture, and general disruption. In contrast, the models presented for testing had typically been produced by the handiwork of master craftsmen. When adapting them to the requirements of serial production the factory tended to make simplifications that inevitably degraded their tactical-technical characteristics. The use of substitutes for materials in short supply also undermined quality. In Baranov’s view, however, ‘‘the basic fault in the work of Aviatrest [was] working without blueprints, from unauthorized outline sketches that [had] been crossed out and corrected innumerable times.’’≥ Management attempts to ‘‘simplify’’ designs and improvise adaptations were a daily occurrence. In July 1932 the aircraft industry administration abandoned the struggle to make factories build aircraft ‘‘as laid down in the blueprints sent from the design bureau’’ and officially permitted factories to introduce design modifications ‘‘to suit the factory’’ (Talanova 1999: 69). Subsequently the industry’s leaders made despairing efforts to break this tendency by circulating threatening decrees on the enforcement of responsibility for technological discipline two or three times a year. There was no discernable improvement, however, until the minister issued decree no. 518 on October 2, 1940. This directive categorically prohibited alteration of designs and blueprints for serial production without preliminary testing and the permis-

184

Mikhail Mukhin

sion of the chief designer and the minister. It also banned any variation of the approved production process without permission of the factory chief technician, chief engineer, and chief designer, countersigned by the head of the chief administration superior to the factory (Talanova 1999: 111–12). This limited the factory’s discretion over the adjustment of designs to manufacturing capabilities to some degree. TREATMENT AND RESULTS

Workers and managers bore some responsibility for not regulating labor utilization effectively. Visiting the United States in 1930, the designer A. N. Tupolev (cited by Kerber 1999: 50) was astonished not just by the American factories’ technological level but also by the American workers’ attitudes to working time. ‘‘Their factories [give] a surprising impression—you walk from one workshop to another and you never see a single idle person. I gave Petr Ionovich [Baranov] a nudge: I bet you’re waiting like me to find some workers standing behind this door having a smoke. But there weren’t—we saw nobody smoking or chatting!’’ This is not what Tupolev was used to in the Soviet Union. Tupolev thought discipline could be best improved not by penalizing cigarette breaks more severely but by enhancing working conditions. He inspected factories regularly and not infrequently made scandalous discoveries: once he found a fishbone in a glass of fruit juice; another time he marched a factory director to the workers’ toilet and lectured him on the relation between filth and product quality (Kerber 1999: 53). The general level of working culture was not particularly high, however, even in the leading factories. A report on factory no. 1 notes: ‘‘Labor discipline low. The workers drink, a great deal at times, and turn up drunk, especially after being paid. . . . The factory is severely lacking in higher and middle technical personnel and skilled workers.’’∂ A serious bottleneck was the low proportion of working time that was used productively. In a lecture on ‘‘Modern Aircraft Abroad’’ (cited by Ivanov 1995: 150–51) the designer N.N. Polikarpov lamented: Reading our newspapers we see that in a number of branches we find the phenomenon of so-called storming (shturmovshchina) when 50 to 60 percent of the production program is completed in the last week of the month, but [only] 2 or 3 percent in the first week. . . . I am always amazed by the planning and discipline that exist abroad. You arrive at the factory and you see that no one is in a rush, everyone is working calmly. If they see you, even if they take a break, it doesn’t break into their work.

The Market for Labor in the 1930s

185

You think: working like that, how can they get first-class results? It turns out that of the eight or nine hours that they spend at work each day they are actually working eight or nine hours; meanwhile we are rushing around, doing something or other, but actually working two or three hours a day and we have no idea how we spend the rest of the time without making an input.

Even if he exaggerated for effect, for two to three hours of real work per shift is surely too little, Polikarpov’s overall judgment is supported by other evidence. The party central committee’s industrial department, for example, investigated aircraft factories no. 24 and no. 29 in connection with their successful completion of the 1936 annual plan. The audit established that in the first ten months of the year only 57 percent of the aircraft plan had been completed, and 76 percent of the aircraft engine plan.∑ The lag of reality behind the plan that gave rise to the need for storming had accumulated not only within each month but also across the months of the year. In short, Polikarpov’s bitter words were hardly news for the industry’s leaders. The investigation defined the industry’s management problems in the mid-1930s precisely as follows: Managers rely on storming instead of production improvement. In the first nine months of 1936 in just seven factories (no. 1, 21, 22, 23, 24, 26, and 29) there were 3,156,000 hours of downtime and 5,676,000 hours of overtime. Substandard goods made up 31.4 percent of the value of commodity output of factory no. 29 and 15 percent for factory no. 24. The current Sovnarkom decree on factories’ material accountability for substandard goods is not normally enforced. Chief Brandt of no. 2 machinery workshop, factory no. 24, claims: ‘‘We are afraid to fine shoddy workers since even without that we are experiencing a labor shortage.’’ A very large proportion of substandard work is the consequence of unpunished violations of production processes and weak labor and technological discipline. . . . Those who violate labor discipline are often leading factory employees. As a rule the workshop chiefs at factory no. 22 go home at midnight or 1 a.m.; at least five times a month the director holds meetings with the workshop chiefs that begin not earlier than 11 p.m. and finish at 3 or 4 a.m. Because of this the workshop chiefs are usually late to work. . . . When the [overall] program is not completed the great majority of workers [still] overfulfill their [individual] work norms. At factory no. 1 only 4 percent of the workers have not fulfilled their norms. Factory no. 22 has fulfilled [only] 40 percent of the annual program, but less than 7 percent of workers have not fulfilled their norms. Factory no. 31 has fulfilled 18 percent of the program, but only 28 workers have not fulfilled their norms. A huge quantity of work time is spent on substandard work, repairs, and remedial work. There is colossal labor turnover, 25 percent in total across the aircraft industry.∏

186

Mikhail Mukhin

In some respects the aircraft industry came out ahead of other high-technology sectors. It led, for example, in its approach to pay and incentives. In 1930, for example, 65 percent of the employees of factory no. 1 were on piece rates compared with only 42 percent at another technologically advanced enterprise, the Moscow Elektrozavod.π But this lead was not maintained. In 1938 and 1939 fewer than 62 percent of the total working time in the aircraft industry was paid by piece rates, and this was less than the 63 percent recorded in 1932.∫ One explanation may be that factory no. 1 was atypical; it was one of the oldest plants not just in the industry but in the country. It was beyond the capacity of the newer factories, often scattered far from the established industrial centers and plagued by recruitment difficulties, to catch up in this regard. The Stakhanov movement did not pass by the aircraft industry. In 1936 no fewer than 1,758 of the 8,452 employees of factory no. 19 were registered Stakhanovites.Ω This certainly gave some striking results. Before Stakhanovism it took two workers per shift to produce seven or eight rear engine covers. Using Stakhanovite methods two workers, Fedineev and Tot, raised the rate to 18. Their earnings per shift also rose from 10 or 11 rubles to 25 to 30 rubles. This was a real achievement. But it could not be generalized to all lines of work or all factories (Mukhin 2003).

The Aviation Workforce: Some Trends PRODUCTIVITY AND PAY

We will now compare the position of workers in the aircraft industry with that in other defense branches. The aircraft industry was to some extent gathering weight at this time, its employment share in the ministry of defense industry rising from 17 percent in 1933 to 19 percent in 1936.∞≠ Figure 7.2 shows that for most years of the second five-year plan period, ruble productivity in the aircraft industry was close to or fractionally below that for the entire defense industry; wages, in contrast, were higher. The aircraft industry’s officials rated their own capability to regulate wages as low. Deputy minister Mikhail Kaganovich reported in 1936: In 1934 and 1935 issues of wage organization in the aircraft factories were set aside. No one from the chief administration or the factories really engaged with these issues. There was a labor department in the chief administration, weak both in numbers, 5 persons, and in their make-up. It was headed by an enemy of the people. He had no authority in the factories. At the factories labor issues were scattered under different headings . . . wages, payment

The Market for Labor in the 1930s

187

Figure 7.2. Productivity and Pay: The Aircraft Industry Compared with Defense Industry as a Whole, 1933–36 (rubles) Note: The ‘‘gross surplus’’ is gross value of output at prevailing prices, less average earnings per manual worker. Source: RGAE, 8044/1/ 2737: 3–4.

accounts, and incentive systems were dealt with in factory planning departments, while technological departments dealt with norm setting and piece rates. At the largest works [only] one to three people dealt with wage issues. There was one at factory no. 39 and two at factory no. 18. In factory plans wage issues were regarded as subsidiary and did not get enough attention so that wage-leveling, large unjustified side-payments, and distorted norms and piece-rates ruled on the shop floor.∞∞

This was not self-criticism, however; Kaganovich was criticizing his predecessors and had a clear interest in portraying the state of affairs that he had inherited as one of gloom and desolation so as to puff up his own achievements. We noted above that by 1932 the aircraft industry already had a high proportion of its workers on piece rates. In other respects, however, the wages policy of the early 1930s does not look so good. Many different pay scales were in effect at different enterprises: in 1935 there were no less than 50. On the grade I scale there were 28 different points for time workers and 28 for piece workers. Characteristically, the ratio of grade VIII to grade I pay varied between 2.58 and 2.83 to one.∞≤ Thus there was little incentive for workers to acquire new skills and win promotion from grade to grade. In an effort to break with this situation the industry leadership attempted to review work norms and restructure pay scales in 1936. The number of points on the grade I

188

Mikhail Mukhin

Figure 7.3. Enterprises of the First Chief Administration of the Defense Industry: Numbers Employed, 1932–36 Source: RGAE, 8044/1/2737: 37.

scale was cut back from 56 to 20. In 1938 it was intended to cut back the number of points to five for the Moscow region and 10 for the whole country. The grade VIII to I ratio was increased to 3.5 to one in order to increase incentives. This degree of wage inequality was relatively uncommon in the Soviet economy at that time; for example, at the Moscow Elektrozavod it varied between 2.8 and 3 to one.∞≥ Piece workers were paid one-third more than time workers as a rule. As shown above, the share of intellectual work in the defense branches was above the average for industry as a whole. Figure 7.3 shows two things: the rapid growth of total employment in the aircraft industry, which rose by onehalf in the five years leading up to 1936; at the same time the share of engineering and technical specialists rose somewhat, from less than 12 percent to over 14 percent. Also noticeable is the shrinkage in apprenticeships, which declined absolutely. The latter trend was reversed in 1936, however, because of the appearance of a new category: ‘‘adult apprentices,’’ or workers who were being assimilated to new occupations.∞∂ Evidently this stood for the induction of new masses of unskilled workers into the industry. The basic proportions of manual and specialist workers, however, persisted through 1937 and 1938.∞∑ Table 7.1, showing the distribution of manual workers across pay grades, suggests that their skills were average for the Soviet economy at the time. As Figure 7.4 indicates, the nominal pay of the main groups of workers in the aircraft industry approximately doubled between 1932 and 1936. The persistent labor shortage of the time might be expected to have compressed the

The Market for Labor in the 1930s

189

Table 7.1. Manual workers in the aircraft industry: Numbers by pay grade, 1937 (percent of total) Grade I II III IV

Number

Grade

Number

∑.≥ ≤≤.∑ ≤∑.∞ ∞∫.∑

V VI VII VIII Total

∞∂.≠ ∫.∏ ∂.∞ ∞.Ω ∞≠≠.≠

Source: GARF, 8418/22/34: 18.

structure of earnings somewhat. If so, then Figure 7.4 would show a negative association between earnings growth and initial earnings. There is some evidence of for this: highly paid engineers lagged behind the manual workforce, and low-paid youth workers benefited from the highest earnings growth of all. But low-paid apprentices lagged behind, and office staff did worst of all. Because the engineering and technical workforce could not be put on piece rates much ingenuity went into working out alternative incentive schemes. They were rewarded mainly by the degree of fulfillment of the production plan and cost reductions and penalized for the supply of substandard goods. Those who contributed to the results of a unit below the level of a workshop were rewarded on the basis of results of both unit and workshop; those who contributed directly to the results of the workshop as a whole were rewarded by the results of both workshop and factory.∞∏ The relative pay of engineers and technicians had an important influence on production. Engineers were paid according to rates set out in a handbook for the industry approved in 1932. By the second half of the 1930s it was out of date. Some posts now arising had not been foreseen and the pay for others was falling behind that of manual workers. The same was happening to office workers. The aircraft industry leadership complained, ‘‘Valuable employees are transferring to manual employment and getting a 25 to 30 percent pay increase, so that production is being deprived of literate ledger clerks, accountants, and time keepers.’’∞π The year 1935 was one of rapid progress. Until 1935, as Figure 7.5 shows, average productivity and losses from shoddy work were rising together. In 1936 and 1937, however, there was a significant breakthrough in productivity while losses fell.∞∫ The fact that engineers’ pay kept pace with that of the manual workers must have been largely due to the higher average seniority of the specialists com-

190

Mikhail Mukhin

Figure 7.4. Earnings Growth in the Aircraft Industry by Employment Status, 1932–36 Source: Calculated from figures in RGAE, 8044/1/2737: 37.

Figure 7.5. Productivity and the Quality of Work in the Aircraft Industry, 1932–37 Source: RGAE, 8044/1/2750a: 1.

pared with that of the workers, among whom there was much higher turnover and who were therefore newly recruited to a much higher degree. Figure 7.6 shows that most manual workers in the aircraft industry had less than two years’ seniority while fewer than one-third had more than three years’. In contrast, almost one-half of the specialist workers had three years’ seniority. This also helped them to maintain their relative pay.

The Market for Labor in the 1930s

191

Figure 7.6. Manual Workers and Engineering and Technical Workers in the Aircraft Industry: Seniority on November 1, 1939 Source: RGAE, 8044/1/2748: 2ob.

When earnings in the aircraft industry are compared with those available in other defense branches, it transpires (Figure 7.7) that the aircraft workers were obtaining a small premium of around 5 percent, which was fairly stable and was present in all years except 1935, when other branches temporarily caught up; the following year the aircraft industry differential was restored. Only shipyard employees were consistently better paid in all years. It is important to bear in mind that the position of different aircraft factories varied substantially. Those in Asia suffered the worst labor shortages. In the spring of 1937, for example, factory no. 126 (Komsomol’sk-on-Amur) had 2,860 employees compared with an established strength of 3,600; 200 of these were ‘‘former’’ people who had ‘‘socially alien’’ origins. The factory had no access to white bread. Its need for black bread was put at 25 tons per day, but only 16 to 18 tons were being baked, and long queues formed in consequence. The list of goods for July of which the employees had only fond memories is startling: pasta, none since March 1; fresh fish, none since June 1; sugar, none since June 10 ‘‘and it is not known when there will be.’’∞Ω There was no flour or milk on sale, but since what date is not stated. Worse than these, the poor accommodation, lack of children’s foodstuffs, and absence of medical care in the factory settlement were leading to epidemics and high mortality among the children.≤≠ In Moscow, it goes without saying, factory no. 1 was immune to this kind of deprivation. Even so, the factories on European territory had their own prob-

192

Mikhail Mukhin

Figure 7.7. Annual Average Earnings of Manual Workers in the Aircraft Industry, 1933–38 (percent of other branches of the defense industry) Source: Calculated from RGAE, 8044/1/2750à: 80. Average earnings of manual workers in other branches of the defense industry are computed as unweighted means for shipbuilding, artillery and shells, munitions, radio equipment, chemical weapons, armored vehicles, tanks, optical equipment, instruments, gunpowder, and batteries.

lems. The workers of the Rybinsk factory no. 26, for example, complained that they had received no smoked foodstuffs such as sausage for months, and no oil or fats for weeks. Much of the foods allocated centrally from Moscow were ‘‘lost’’ by the local supplier in Ivanovo.≤∞ Factory no. 22, one of the oldest and largest aircraft plants, suffered from a persistent housing shortage. For 19,236 persons dependent on factory housing in April 1928 there were 72,833 square meters of space available, or 3.8 square meters per person compared with a sanitary norm of 8 square meters. Nearly half of this population, 8,874 people, lived in ‘‘piled-up wooden shacks of a temporary kind that have been in use for 10 to 12 years.’’≤≤ A particularly acute problem was the 2,100 workers who had requested to be classified as in acute housing need. It was not possible to meet their requests because from 1934 to 1938 the factory had not built a single housing unit.≤≥ For a while the factory management had managed the problem by hiring only family members of existing employees. By 1936, however, when the number of those not seeking additional housing space was down to 3,500, this source of recruitment was exhausted; moreover, the factory faced the need to hire 2,000 more workers in order to fulfill the plan while the issue of how to accommodate them had not been considered.≤∂ In 1938 the factory’s production program was doubled, requiring the further recruitment of 7,000 employees. There was

The Market for Labor in the 1930s

193

no possibility of meeting this target without more housing.≤∑ Only at this point did the factory succeed in ‘‘carving out’’ twenty-five standard housing blocks and sites under construction ‘‘on the account of other ministries,’’ or by robbing Peter to pay Paul.≤∏ The same mechanism of exploiting the urgency of high-priority government assignments to solve morbid issues arising from everyday life was used by various factory managers. In 1938, for example, when it became urgent to bring the new I-153 fighter into serial production, the industry leadership succeeded in getting the Defense Commission to order the Moscow city soviet to build three multi-story apartment blocks for factory no. 1 to be ready not later than the third quarter of 1939.≤π DOWNTIMES AND EMERGENCIES

The irregular pace of production of the aviation industry eventually led workers not only to adapt to their environment but to gain from maintaining the status quo. Here the leaders of factory no. 22 describe how the plan was fulfilled in 1938: ‘‘Until the fifteenth of the month the assembly line stood still. In 15 [more] days we [had to] achieve the program; that is why there was no final assembly going on, that is why everybody was on downtime. But in this situation someone had ‘caught a fish in muddy water’ because when the twentieth came and it was clear that we would not make the program the factory director came onto the shop floor and talked to the foremen, and then when he saw that nothing was happening he said: ‘Here’s so much money for you, now get stuck in.’ And they got stuck in.’’≤∫ On one hand the factory managers had to bow to the judgment of the skilled craftsmen who were the experts on how to ‘‘get stuck in.’’ On the other hand it is clear that the same experts were the ones who profited most from the situation. In addition to the respect they won from management, they also gained in overtime rates of pay. Corresponding to this was the spread of overtime that is seen in Table 7.2. Aircraft industry leaders were especially worried by the rapid growth of overtime toward the end of this period (Table 7.3). The efforts to accelerate production seem to have disrupted working patterns in that overtime and downtime began to rise simultaneously. Overtime payments were not a large proportion of pay, however. Of manual workers’ earnings in the first nine months of 1939, only 3.5 percent came from overtime; almost two-thirds of earnings were made up by piece rates and bonuses based on piece rate norms.≤Ω It was hard for managers to restrain earnings from piece rates, which tended to grow rapidly. In 1939 the ministry was increasingly convinced of the necessity for an urgent review of work norms. Many workers were fulfilling out-of-date

194

Mikhail Mukhin

Table 7.2. Downtime and overtime in the aircraft industry, 1938 and 1939 (hours and percent)

Time worked (thousands of hours) Of which: Overtime Downtime % of total: Overtime Downtime

∞Ω≥∫

∞Ω≥Ω

≥∫∑.∂

∂∏∑.∑

∞∏.≠ ∫.∂

≤π.≥ ∫.∫

∂.≤ ≤.≤

∑.Ω ∞.Ω

Source: RGAE, 8044/1/2748: 2ob.

Table 7.3. Downtime and overtime per worker in the aircraft industry, January– October 1939 (monthly average and percent of total time worked)

First quarter Second quarter Third quarter October

Overtime

Downtime

∑.∫ π.∞ π.∫ ∞∂.∂

≥.∏ ≤.∑ ≤.∂ ≥.≤

Source: RGAE, 8044/1/2750: 17.

norms with ease and claiming large payments on this basis. At the same time the ministry was struggling to fulfill its production program (Table 7.4). The revision of work norms is discussed further below. Long-service employees, the ‘‘labor aristocrats,’’ cohabited with a much larger number of migratory workers who contributed to very high turnover. At factory no. 45, for example, with a workforce of 1,500, more than 1,000 employees left and were taken on in just eleven months of 1939.≥≠ Turnover, moreover, was on an upward trend (Table 7.5). Although the high rate of hirings can be explained by the industry’s rapid expansion, the high rate of departures is a less favorable indicator. Sometimes, turnover was attacked by nontraditional means. Thus on September 23, 1938, decree no. 266ss/ov transferred factory no. 22 to a state of mobilization, meaning that the employees liable to conscription were counted as being on active service. They carried on as members of the trade union but could no longer take part in arbitration or grievance procedures that con-

The Market for Labor in the 1930s

195

Table 7.4. The fulfillment of production plans and work norms in aircraft factories, January–October 1939 (percent)

Factory

Production plan fulfillment

Work norm fulfillment

No. ∞∫ No. ≤∞ No. ≤≤ No. ∞∑≥

∫∏.≠ π≥.∞ Ωπ.≤ π≥.Ω

≤≠∫ ∞π≤ ∞∑Ω ∞≥Ω

Source: RGAE, 8044/1/2750: 10–11.

Table 7.5. Turnover of employees in aircraft factories, 1938 and 1939 (thousands and percent)

Thousands: Hired Left % of employees: Hired Left

∞Ω≥∫

∞Ω≥Ω

∞≠∫.π ππ.∏

∞∑∞.∞ ∞∞≥.∂

∑≠.Ω ≥∏.≥

∏∞.∞ ∂∑.Ω

Source: RGAE, 8044/1/2748: 2ob.

cerned labor discipline, departures, or transfers to other work. Those mobilized came instead under the jurisdiction of military tribunals and the Red Army code of discipline.≥∞ This resolved the problem of turnover. The method was soon extended to recruitment. On November 23, 1938, the defense ministry called up 2,500 skilled workers and transferred them to factory no. 22.≥≤ By March 21, 1939, 11,000 of the factory employees had the status of Red Army servicemen.≥≥ There was a spoonful of tar in this honeypot, however. The mobilized workers could not leave, nor could they be sacked. The result was a rapid decline in discipline.≥∂ The mobilization also gave rise to an array of administrative follies. The local police would not grant residence permits to workers without passports, but the passports of mobilized workers were held by the defense ministry. This not only impeded the reception of newly recruited workers into the factory hostels but even threatened the expulsion of 202 factory employees who were already living there. Although this could be sorted out eventually,

196

Mikhail Mukhin

Table 7.6. Pieceworker’s basic monthly wage and average actual monthly earnings based on 175 hours worked, May 1939 (rubles) Grade II III IV V VI

Wage rate

Actual earnings

≤≤≠ ≤∏∏ ≥≤≤ ≥∫∏ ∂∑Ω

≥≤≤ ∂≠∑ ∂Ω∑ ∏≤≠ π∑≠

Source: GARF, 8418/23/952: 4.

things went worse with the penalization of absentees and shoddy workers who wore a uniform. As servicemen they could not be punished under the civilian labor code. At the same time the military prosecution service refused to get involved on the grounds that it had no legal basis to act.≥∑ Despite this the factory management considered that the advantage of putting a stop to voluntary turnover more than made up for the other defects of the new system. Defense Commission decree no. 108ss of May 7, 1939, empowered the factory management to fix overtime, subject to the minister’s agreement but not that of the trade union; the only concession to the mobilized workers was a percentage markup on the basic wage based on the duration of unbroken employment at the factory.≥∏ As early as July 1939 the factory no. 22 director was requesting more contingents of mobilized workers; this action suggests that, at the very least, the management was adjusting to the new system.≥π At the same time, however, the general tone of its reports indicates that worker discontent was rising and the management was still looking for ways to improve worker incentives. Evidently, even mobilized workers still needed something positive to work for. Thus a Defense Commission decree of August 1939 authorized the seniority premium created by the July decree to be calculated on workers’ actual earnings rather than their basic wage; this was of some importance because earnings usually exceeded wage rates by a substantial margin (Table 7.6). These provisions limited turnover but did not prevent it completely; in the aircraft industry it fell from 45 percent of average employment in 1939 to 34 percent in 1940.≥∫ Meanwhile, war preparations were tending to militarize the management of the whole economy. In June 1940 the Supreme Soviet presidium adopted its well-known ‘‘Decree on Transition to an 8-Hour Working Day and 7-Day Working Week and Prohibition of Voluntary Departure of Manual and Office Workers from Enterprises and Establishments’’ (IVMV

The Market for Labor in the 1930s

197

1974, vol. 3: 374). Serfdom and indentured labor had returned, although not only to the Soviet Union: Germany had enacted similar measures in April 1939. In this context, on July 4, 1940, the experiment with mobilization of the aviation employees was terminated and all the workers were returned to civilian status. By this time, what was meant by ‘‘civilian status’’ had changed and was now somewhat more ‘‘military’’ than before. Nonetheless, the outright militarization of labor had evidently been judged ineffective.

The Prewar Spurt In the course of 1940 the pace of work was intensified in the aircraft factories (Mukhin 2006); managers gained the right to introduce overtime on their own authority and to introduce two or three shifts working in workshops and entire factories. Plans for the last quarter of 1940 and the first of 1941 envisaged average overtime of two to three hours per worker.≥Ω Increased overtime was often associated, however, with deterioration in the coordination of work. At factory no. 18, for example, downtime in the first quarter of 1941 stood at 5 percent of total working time while overtime stood at 7.8 percent.∂≠ If the downtime could have been eliminated, the overtime could have been cut by two-thirds. Although the overall capacity of the aircraft industry grew extensively in the last prewar years, the forcing of the pace had some adverse results. First was that as new factories sprang up, the thin layer of skilled personnel was further diluted and labor productivity fell. This was the starting point for a whole series of further ramifications. Downtime per worker jumped from 37 hours in 1939 to 61.4 hours in 1940, leaving more than 3 percent of working time idle.∂∞ The general picture of management and skill shortages is conveyed by a sequence of anecdotes. On December 4, 1937, the Defense Committee ordered ‘‘factory no. 24 director Mar’iamov to be removed from the position of director as not coping with business. To appoint I. T. Borisov director.’’∂≤ On January 1, 1939, a decree of the eighteenth chief administration of the ministry of defense industry ordered ‘‘deputy director [of factory no. 24] Borisov P. A. to be relieved of duties for repeated crude violation of budgetary and financial discipline and to be employed in less responsible work. To name factory no. 24 director Borisov for toleration of violation of financial and budgetary discipline.’’∂≥ On January 13 the ministry of the defense industry resolved in decree no. 16/k ‘‘1. To remove comrade Borisov I. T. from the work of director of factory no. 24; 2. to appoint comrade Sokolov D. M. as director of factory no. 24, relieving him of work as director of factory no. 20.’’∂∂ Finally, on January 20, factory no. 20 was given a new director: I. T. Borisov.

198

Mikhail Mukhin

In short, the outcome of this series of revelations and reprimands was that a man found to be incompetent in charge of one factory was put in charge of another. Clearly the number of suitable candidates fell far short of the number of vacant posts. The crisis extended to the workforce as a whole. From 1939 there was massive compulsory recruitment of skilled but untrained workers into the aviation enterprises; these workers were to learn on the job, while engaged in production on the shop floor. Many of them had earned seven to eleven years’ seniority in their previous posts; they tended to resent their compulsory reclassification because it threatened loss of status and downgrading of their pay. To compensate for the compulsion involved these new workers were allowed to keep the seniority earned in their previous employments. The older ‘‘cadre’’ factories were swamped by the new workers. In 1940 the industry as a whole took in 30,000 skilled but untrained workers (Kostyrchenko 1994: 200). Some work groups doubled in size. When workshop chiefs refused to take untrained recruits the director of factory no. 22 gloomily replied: ‘‘No one is going to give us trained assembly workers, riveters or mechanics.’’∂∑ There were none to be had. In some workshops the level of substandard products rose to 50 percent. The situation was complicated by the fact that factories were supposed to train the new workers and raise output at the same time. The 1939 production plan for factory no. 22 was twice that of the year before.∂∏ Shoddy work in the factory in 1939 and 1940 ran at nearly twice the level of 1938, the last prewar year.∂π There was some improvement the following year, however. Of 439,000 trained graduates of the new system of State Labor Reserves, the government sent 50,000 to the aircraft, armament, and ammunition industries and to defense industry construction sites (TsSU 1972: 355). The industry struggled to supply inexperienced recruits with instruments and equipment. Managers were at a loss: ‘‘In two days I have taken on 48 people, just fewer than I have existing workers. . . . I went [to the instrument department] and I asked what they could give me—10 hammers, no drills or pneumatic hammers, 5 electric hammers’’; in short, three workers to a hammer.∂∫ Inevitably, while employment rose average productivity fell. An example is provided by a time-and-motion study of milling-machine operator Milekhin at factory no. 22 in February 1941.∂Ω In one and a half shifts Milekhin worked 262 minutes out of 660. Of the remaining time he spent 35 minutes setting up his machine, 35 minutes obtaining instruments, and 30 minutes idle, and he began to tidy up 27 minutes before the end of his shift. The uses of the remaining time are not specified. Milekhin lay at one

The Market for Labor in the 1930s

199

extreme but of six workers studied in this way none worked more than 494 minutes and the typical downtime was 200 to 300 minutes. There were also general observations on the deterioration of labor discipline. Twenty or thirty minutes before the dinner break queues formed in the canteens.∑≠ Foremen were reluctant to notify timekeepers when workers were even half an hour late; to fire a trained colleague would hurt the work group more than the one dismissed.∑∞ Established employees complained that the new workers had caused the work culture to decline because they regarded their new duties as unwanted and temporary. Factory no. 22 party committee members claimed: ‘‘We made the rounds of three quarters of the work positions . . . at any machine you open the desk and there’s some bread and some dirty rags and so on. You tell [the worker], surely he can’t eat the bread but he says ‘it’s all right, it’ll do.’ At the machines there’s wires and shards lying about, like with pigs. . . . Some machines are broken because they treat them outrageously. . . . Just about every bench is broken, the locks are broken, every chair has a chain to stop it being stolen, the vises are broken, there’s dust, filth, you can’t get past it.’’∑≤ There were frequent cases when tasks were specifically delayed to the evening so as to count as overtime; this practice also extended to the engineering and technical workers. Basic duties began to be paid as contract work, enabling technicians on a basic wage of 800 rubles to earn 1,000- to 2,400-ruble bonuses for assembling equipment.∑≥ The urgency with which the industry’s and the country’s leaders wanted to bring new aircraft models into serial production had a decisive influence on the formation of incentives and side payments. In 1939, for example, the Defense Committee promised bonuses to leading specialists for successfully starting up new production lines as follows: 25,000 rubles each to the factory director, chief designer, and chief engineer; up to 15,000 rubles to the production chief, the serial production office chief, and the chief of production preparation; and up to 10,000 rubles to the chief dispatcher, the quality section chief, and the supply section chief.∑∂ Subsequently the ministry leadership continued to pay close attention to management incentives. In February 1941 minister A. I. Shakhurin told Stalin that the current monthly pay of directors and chief engineers was 2,000 rubles in aircraft and assembly factories and 3,000 rubles in aeroengine factories. Because workshop chiefs were able to take home 2,000 to 2,500 rubles including bonuses, he asked to fix the pay of a long list of directors and chief engineers at 2,500 to 3,000 rubles.∑∑ Subsequently attention turned to rewards at lower levels. In January 1941, for example, in connection with phasing out the SB bomber and starting up production of the Ar-2 and Pe-2 at factory no. 22,

200

Mikhail Mukhin

Table 7.7. Average hourly pay of a grade I manual worker in the aircraft industry, 1941: Selected factories (kopecks) Factory

Location

Pieceworker

Timeworker

No. ≥∫≥ No. ∂∂Ω No. ∞∑∑

Moscow Saratov Khar’kov

π∫.π∑ π≠.≠≠ π∫.∫≠

π≠.≠≠ ∏∞.≤∑ ∏∑.∏≠

Source: RGAE, 8044/1/647: 20–23.

deputy minister V. P. Balandin authorized the factory to offer foremen bonuses for plan fulfillment up to 50 percent of their basic pay.∑∏ Taking an overview, it appears that Soviet economic managers had lost faith in the power of repression to motivate effort, and were returning to cash as a stimulus. It was said above that in 1939 the ministerial leaders of the aircraft industry put revision of work norms on the agenda. In 1940 they went from words to deeds. In January and early February 86 percent of norms were reviewed; on average norms were raised by 38.7 percent, implying a cut in piece rates by 25.8 percent. As a result the proportion of norms that the workers overfulfilled fell from 91.8 to 43.9 percent. Meanwhile, the share of piece work in overall working time fell marginally from 62 to 61 percent.∑π The pay structure for manual workers in the aircraft industry in 1941 had eight grades with a 3.6:1 ratio between the top and bottom grades, more or less the same as in 1938.∑∫ Grade I was particularly important because it provided the base for the higher grades, and the grade I scale varied considerably across factories (Table 7.7). Such discrepancies were bound to influence turnover, and managers exploited them to their own advantage. An extreme case is provided by factory no. 464, established in Riga on the basis of an existing facility but virtually without existing personnel. Appointed to head the factory, the designer O. K. Antonov advertised locally for recruits, luring them with promises of high pay which he spread using mailshots and word of mouth. He was so successful that the Riga Arsenal complained to high authority about the aircraft builders’ poaching its best workers.∑Ω The situation was particularly bad in the aircraft works of the eastern regions. In an attempt to consolidate employment and curb the rising turnover on November 23 the Economic Council authorized the industry to boost the pay of specialists and office workers in these distant regions by the correction coefficients shown in Table 7.8; the coefficient rose from west to east.

The Market for Labor in the 1930s

201

Table 7.8. Average pay of engineering, technical, and office workers of aircraft factories in remote regions: Correction coefficients introduced on November 23, 1940 Factory

Location

No. ∞∞∏ No. ∞≤∏ No. ∫≥ No. ΩΩ No. ∞≤∑ No. ∞∑≥

Semenovka, Primorskii krai Komsomol’sk-on-Amur Khabarovsk Ulan-Ude Irkutsk Novosibirsk

Correction coefficient ∞.∫ ∞.π ∞.∏ ∞.∑ ∞.≥∑ ∞.≤

Sources: RGAE, 8044/1/408: 38 and, for locations, Dexter and Rodionov (2004).

While rewards were increased in the more distant factories, the ministry sought to restrain wage expenditures in those closer to the center. In March 1941 there was a regular review of work norms at a number of factories that led to a general cut in piece rates.∏≠ This increased the pressure on the workers; at factory no. 16 in the first quarter of 1941, for example, 18.6 percent of piece workers failed to fulfill their norms.∏∞ Labor discipline not only did not improve as a result of such innovations but also remained unsatisfactory in the first months of the war despite new wartime legislation. On September 29, 1941, the factory no. 22 party committee reported that August had seen 42 instances of lateness and 66 of absenteeism compared with 29 and 13 respectively in July.∏≤ In addition the report noted ‘‘aimless wandering around in working time, failure to start and finish work on time, sleeping in working time, and failure to fulfill management instructions.’’∏≥

A Factory’s Problems These problems can be viewed in the microcosm of a single plant, the May the First Factory no. 7 in Moscow, on the basis of documents of the factory party committee in 1939 and 1940. This was a newly converted factory but it was affiliated to one of the oldest aircraft works, factory no. 22. Problem no. 1. The party committee defined the first problem as ‘‘storming,’’ which was ‘‘consuming the factory.’’∏∂ Figures for the summer and early autumn of 1939, illustrated in Figure 7.8, suggest that it was normal for virtually no output to be recorded in the early days of the month. The ministry shared this view. In his memoirs Shakhurin (1985: 94) wrote: ‘‘I became minister in January 1940, but in February the factories were still supplying the goods accepted in December 1939. Working like this the first

202

Mikhail Mukhin

Figure 7.8. Storming in Factory no. 7, July–September 1939 Source: Calculated from figures of monthly plan fulfillment for gross value of output given in TsAODM, 217/1/156: 51.

two weeks of each month were spent clearing up what had been left undone in the previous month, and in the last ten days would begin a storm to fulfill the plan somehow or other. Roughly they would supply half of the total output in the last ten days—at best. In the worst case it would take still longer.’’ At the May the First Factory the workshop chiefs defined the cause of storming as arising from the lack of reserve stocks of finished goods carried over from one period to the next: ‘‘Storming happens because [we have no reserves of ] finished goods. There is no carryover as such and so the workshops have to make their start each month from the first day and are working on the carryover up to the last day.’’∏∑ In other words, because production always started from nothing it was excessively vulnerable to disruption. The factory management, however, refused to take responsibility for it and blamed the workshop chiefs: ‘‘The workshop chiefs are to blame for storming. Who is it, one may ask, that doesn’t begin work on the first of the month, who is it that prevents additions to the carryover?’’∏∏ At first glance the position of the workshop chiefs looks stronger: a workshop is an integral subunit of the factory and to operate it without interruptions needs coordination above. If materials were not supplied at the right time the workshop could scarcely ‘‘begin work on the first of the month.’’ The warehouse would respond to an order from the factory director, not the workshop chief. In short, coordinating supplies to ensure uninterrupted operations was a management responsibility; if the director failed, then storming was chiefly his fault. The party committee leadership offered its own version of the root of the

The Market for Labor in the 1930s

203

problem. ‘‘Departments [such as] production preparation and supply do not get worked up in advance about provisioning work in the first 20 days [of the month]. Business contacts don’t happen between workshop and department chiefs, and one workshop chief doesn’t care about the fate of another—this makes storming inevitable in the workshops where the production plan is finished off.’’∏π Interestingly, the party committee did not raise the issue of the lack of higher management coordination; rather, the workshop chiefs were blamed for not freely coming to a collective agreement. The same document notes: ‘‘The machine workshop has never yet provided a full assortment to the assembly shop. The supply department is remote from production and does not live by its interests.’’∏∫ But if the individual workshops and departments should have been able to come together and coordinate their actions in a decentralized way, what was management for? Who, if not the director, was supposed to make each take account of the requirements of the others? This leads us to the next item. Problem no. 2. In the 1930s the term ‘‘workshop sectionalism’’ (tsekhovshchina) was applied to cases in which the workshop leaders gave priority to the interests of their own section rather than to those of the factory as a whole. Where it took hold, the workshop leaders seceded from the rule of the factory management and began to act as independent proprietors. It is hard to know what was typical, but the experience of factory no. 7 is certainly of interest: ‘‘in the rifled cylinders workshop, for example, there is a leak in the roof; instead of putting a bucket under the leak, they have made a hole in the floor and let the water through into the store.’’∏Ω Problem no. 3. This was the disinclination to work for basic pay. Overtime and contract work were another matter. ‘‘In the department of the chief mechanic, particularly, it is possible to complete work within normal hours but they try by all means to drag it out under a [sideline] contract or outside normal hours.’’π≠ Indirect evidence of workers’ reluctance to rely on basic pay is the growth in the number of Stakhanovite workers, which, in factory no. 7, proceeded from below among the workers rather than from the party committee above.π∞ Dispirited by this the party committee suggested: ‘‘The growth of Stakhanovite numbers is going ahead haphazardly and spontaneously because of the workers’ clear material interest under piece rates.π≤ In the departments, where they work for basic pay, the workers limit themselves to minimal effort; naturally therefore there is a wide divergence between workshops and departments.’’π≥ Table 7.9 illustrates the difference. Problem no. 4. This was turnover. There were several causes. First, the management expected a high proportion of new recruits to fall out quickly and made little fuss over them. ‘‘We have high turnover at the factory, and the

204

Mikhail Mukhin

Table 7.9. Factory no. 7: The number of Stakhanovites in 1939, by month

July August September a

In production

In departmentsa

∞Ω∫ ≤≥∏ ≤∂≤

∞≥ ≥ ≤∞

From the context, in the offices where workers relied on basic pay.

Source: TsAODM, 217/1/156: 51.

new recruits have no craft organization, the management does nothing for them, and they have no comradely connections at first. We need to gather the new recruits at meetings from time to time, explain the factory’s tasks, and surround them with comradely concern; then turnover will fall off’’; or so it was suggested at a party meeting.π∂ Evidently there was a need for ‘‘comradely concern’’ to be material as well as moral; it seems that the most lucrative overtime and sideline contract payments were monopolized by the established workers, and the new recruits were kept away from them. Party meetings perhaps alluded to this in veiled terms: ‘‘When directed to the workshops they do not meet with proper attention on the part of the workshop leaders. The bureaucratic reception of new recruits, the time they spend idle thanks to incompetence and disorganization of the workplace, and the ungratifying information about our workshop procedures push people into going absent or quitting voluntarily.’’π∑ The factory party leadership offered other explanations for high turnover. Party secretary Likhonin observed: ‘‘They often say it is ‘lack of housing’ but I say: we take people without carefully checking up on them. . . . 1. Anisimov I. P. Hostel superintendent. Joined August 14 [1939], not registered for residence. On October 4 the police required his dismissal. He did not even settle his [wage] account. 2. Bozhko V. V. Milling-machine operator. Joined September 13, dismissed October 10 for lack of work. Asked why he was hired, not known. 3. Dolgova I. G., joined August 28 as storekeeper in the welding workshop, dismissed September 21 for lack of work. 4. Antonova M. D., joined as accounts inspector on September 16, dismissed October 14 for lack of accommodation.’’π∏ But it seems unlikely that more careful checks could have changed the situation. These people were hired but the factory did not sort out their housing or right of residence, or left them idle on basic pay without the possibility of earning bonuses or supplements. Their voluntary or involuntary departure cannot have been a surprise.

The Market for Labor in the 1930s

205

Problem no. 5. Poor discipline was blended from time to time with drunkenness. A June 1940 party meeting noted: Cases of drunkenness among individual communists are becoming more frequent. They have even begun to drink the alcohol spirit in the factory store. The day before he was due to get his party candidate membership card, Lakov got drunk, came to work, and caused a minor fracas. On November 13 on Zaretskii’s shift a lathe operator cut through the thread of a spindle. Zaretskii not only did not take steps against the guilty party but also did not inform the workshop chief. On November 11 foreman Alabin of the iron foundry workshop [let a casting go cold and solidify in the furnace] under pressure and did not take it up with the guilty party and even tried to make light of the facts. On November 10 at 4 a.m. Alabin was sleeping on his shift with those under him. Naturally the duty janitor left his post and went to sleep too. On November 8 the duty worker . . . Zhukov turned up for duty completely drunk and was allowed onto the shop floor. The communists are giving a bad example. Men’shov and Kosarev came to work with such a powerful smell of wine that it was impossible to go close to them.ππ

It is difficult to tell what was cause here and what was effect. Whether alcoholism, a low level of work culture, or poor management lay at the root, the fact is that excessive drinking was a serious workplace problem. Problem no. 6. Everything was compounded by poor accounting for materials and work in progress. In the summer of 1940 it was said: Despite the large accounting staff there is no progress. Warrants are issued without [stating] quantities and as a result materials are used at will, parts are given out without [checking] quantities, shoddy work is concealed, and all because there is no accounting. . . . We draw materials from the warehouse without analysis and the result is shoddy work.π∫ . . . There is none of the accounting we need in our workshops. There is no system in the accounting for the movement of products. The [planning and forecasting department] doesn’t know about the movement of parts, even the workshop that is allocated the parts doesn’t know either. . . . The dispatch of materials, you can say, is done by eye and isn’t checked on receipt even by the workshops.πΩ

This situation was perfectly consistent with workshop sectionalism. If management took no responsibility for coordination the mutual relations among workshops would soon naturally degenerate to anarchy. We began this study of employment and ‘‘technical culture’’ in the Soviet aircraft industry by distinguishing between the pressures for improvement and expansion, which pulled in opposite directions. On one side the leaders of the

206

Mikhail Mukhin

industry would have liked, we suppose, to have transformed it into the isolated havens of science and rationality that they saw abroad, with sober and disciplined workers accurately translating blueprints from the cutting edge of technological progress into gleaming ranks of new aircraft. The requirements of rapid growth, however, deprived the aircraft industry of the chance to select its workforce and enhance its skills and conditions in the way that its leaders would clearly have preferred. Instead, massive recruitment and stretched budgets held back the development of skills and the improvement of conditions. The industry had no chance to become a privileged enclave; continually hungry for fresh human resources, it could not detach itself from the society that supplied them. Its staff and employees, therefore, continued to reflect the trends that were general in the civilian economy. As a result, employment in the aircraft industry at this time brought little sense of privilege or stability. It is true that in order to attract and retain workers in the numbers required, pay levels in the industry were somewhat better than elsewhere and rose continually. But after nearly a decade of exceptional growth most workers had less than two years’ seniority. Only the thinnest layer of tenured ‘‘labor aristocrats’’ was beginning to form. The Soviet authorities were more than willing, moreover, to replace financial incentives with coercive measures so as to force productive effort when necessary in their eyes. Any privilege was temporary and could be withdrawn at any time. On the eve of war the aircraft industry underwent explosive growth, and this further intensified its personnel problems. Again the layer of skilled workers was spread thinner. Mass transfers of skilled workers from other branches did not help; a skilled joiner could not take the place of a skilled riveter. Turnover and ‘‘storming’’ increased; discipline declined and with it the level of technological culture. These efforts were not for nothing. They laid the basis for the exceptional expansion of production in the first days of the war and, later, in wartime evacuation. If the aircraft industry’s workforce continued to mirror society, it carried the virtues as well as the demerits of the Soviet working class as a whole. The victory in 1945 was due, not least, to these men and women. Notes I thank Andrei Sokolov for guidance and for not letting me be satisfied by past achievement. 1. TsAODM, 3/11/722: 22. 2. TsAODM, 3/11/722: 38. Petr Ionovich Baranov (1892–1936) is sometimes called ‘‘father of the Soviet aircraft industry.’’ Having joined the Bolsheviks in 1922 he became a political assistant to the Red Army air force chief in the following year and by the end of

The Market for Labor in the 1930s

207

1924 had been appointed to head the Red Army air force. In 1931 he stepped sideways, simultaneously taking charge of civil aviation within the defense ministry and the aircraft industry within the industry ministry; he retained the latter post in the new ministry of heavy industry formed in 1932. He was killed in an air crash on September 5, 1933. 3. TsAODM, 3/11/722: 40. 4. TsAODM, 3/11/726: 47–50. 5. GARF, 8418/11/103: 43. 6. GARF, 8418/11/103: 47–49. 7. Factory no. 1: TsAODM, 3/11/726: 52. Elektrozavod: TsMAM, 2090/1/581: 4. 8. 1932: Kostyrchenko 1992: 420. 1938 and 1939: RGAE, 8044/1/ 2748: 2ob. 9. GARF, 8418/11/100: 16. 10. RGAE, 8044/1/ 2737: 2. 11. GARF, 8418/22/34: 12. 12. GARF, 8418/22/34: 17. 13. TsMAM, 2090/1/579: 3ob; TsMAM, 2090/1/1686: 16. 14. RGAE, 8044/1/2737à. ff. 6–7. 15. RGAE, 8044/1/2750à. ff. 95–97. 16. GARF, 8418/22/34: 22. 17. GARF, 8418/22/34: 28. 18. It is interesting to compare Figures 7.1, 7.3, and 7.5. Figure 7.1 shows that numbers of aircraft produced declined absolutely in 1935. Figure 7.3 shows that employment increased smoothly, so aircraft per worker must have fallen too. Figure 7.5 values output per worker at plan prices, however; these tended to drift upward as new products were introduced; the setback of 1935 is visible only as a short-lived deceleration. 19. GARF, 8418/12/143: 164–168. 20. GARF, 8418/12/143: 164. 21. GARF, 8418/19/9: 42. 22. GARF, 8418/12/180: 15. 23. GARF, 8418/12/180: 1. 24. GARF, 8418/12/180: 21. 25. GARF, 8418/22/271: 11. 26. GARF, 8418/22/271: 6. 27. GARF, 8418/22/266: 5. 28. TsAODM, 217/1/98: 92. 29. RGAE, 8044/1/2750: 9–10. 30. RGAE, 8044/1/401: 8. 31. GARF, 8418/23/386: 1; GARF, 8418/23/952: 1. 32. GARF, 8418/23/386: 4. 33. GARF, 8418/23/386: 9. 34. GARF, 8418/23/386: 20–21. 35. GARF, 8418/23/386: 24–25. 36. GARF, 8418/23/386: 1. 37. GARF, 8418/23/386: 4. 38. RGAE, 8044/1/2775: 20. 39. RGAE, 8044/1/674: 44.

208

Mikhail Mukhin

40. RGAE, 8044/1/674: 65. 41. RGAE, 8044/1/2775: 20–21. 42. GARF, 8418/12/142: 54. 43. TsMAM, 690/1/135: 5; there is no evidence of kinship between the two Borisovs but it is hard to imagine that the director was out of the picture when the deputy was engaged in ‘‘crude violation.’’ 44. TsMAM, 690/1/135: 2, 8, 24. 45. TsAODM, 217/1/49: 14–15. 46. TsAODM, 217/1/49: 3–4. 47. TsAODM, 217/1/96: 196; TsAODM, 217/1/96: 18. 48. TsMAM, 690/1/135: 25. 49. TsMAM, 690/1/98: 6. 50. TsMAM, 690/1/135: 9. 51. TsMAM, 690/1/135: 23. 52. TsMAM, 690/1/135: 25–99. 53. TsAODM, 217/1/96: 164. 54. RGAE, 8044/1/133: 175. 55. RGAE, 8044/1/698: 122. 56. RGAE, 8044/1/684: 17. 57. RGAE, 8044/1/2775: 21. 58. RGAE, 8044/1/647: 19. 59. RGAE, 8044/1/651: 130. 60. RGAE, 8044/1/674: 44. 61. RGAE, 8044/1/674: 71. 62. TsAODM, 217/1/196: 52. 63. TsAODM, 217/1/196: 52. 64. TsAODM, 217/1/156: 49ob. 65. TsAODM, 217/1/156: 2. 66. TsAODM, 217/1/156: 3. 67. TsAODM, 217/1/156: 50. 68. TsAODM, 217/1/156: 50ob. 69. TsAODM, 217/1/156: 28. 70. TsAODM, 217/1/156: 27ob. 71. The diffusion of the Stakhanov movement from above, through management encouragement and patronage, was widespread in large Soviet enterprises such as the Moscow Elektrozavod (Zhuravlev and Mukhin 2004: 120–24). 72. At this point a hand has inserted ‘‘and the [illegible] consciousness of the workers’’ into the typescript. Perhaps it was feared that without this phrase the interpretation offered might be found politically unacceptable. 73. TsAODM, 217/1/156: 51. 74. TsAODM, 217/1/156: 28. 75. TsAODM, 217/1/156: 51ob. 76. TsAODM, 217/1/156: 51ob. 77. TsAODM, 217/1/156: 48.

The Market for Labor in the 1930s

209

78. The same warehouse with the hole that another workshop had made in its ceiling. As the saying goes: ‘‘Don’t spit in the well.’’ 79. TsAODM, 217/1/156: 61.

Published References Davies, R. W., and Mark Harrison. 1997. ‘‘The Soviet Military-Economic Effort Under the Second Five-Year Plan, 1933–1937.’’ Europe-Asia Studies 49(3): 369–406. Dexter, Keith, and Ivan Rodionov. 2004. The Numbered Factories and Other Establishments of the Soviet Defence Industry, 1928 to 1967: A Guide, Part I. Factories and Shipyards: Version 5.0. University of Warwick, Department of Economics. http://www .warwick.ac.uk/go/vpk. Ivanov, V. P. 1995. Aviakonstruktor N. N. Polikarpov. St. Petersburg: Politekhnika. IVMV. 1974. Istoriia Vtoroi Mirovoi voiny 1939–1945, vol. 3. Moscow: Voenizdat. Kerber, L. L. 1999. Tupolev. St. Petersburg: Politekhnika. Kostyrchenko, G. V. 1992. ‘‘Organizatsiia aviatsionnogo krupnoseriinogo proizvodstva.’’ In Samoletostroenie v SSSR (1917–1945), vol. 1, ed. G. S. Biushgens, 413–36. Moscow: TsAGI. Kostyrchenko, G. V. 1994. ‘‘Aviatsionnoe promyshlennost’ nakanune i v gody Velikoi Otechestvennoi voiny (1939–1945 gg.).’’ In Samoletostroenie v SSSR (1917–1945), vol. 2, ed. G. S. Biushgens, 197–238. Moscow: TsAGI. Mukhin, M. Iu. 2003. ‘‘ ‘U sovetskikh—sobstvennaia gordost.’’ Spetsificheskie metody trudovoi stimuliatsii v SSSR 30-kh godov.’’ In Ezhegodnik istoriko-antropologicheskikh issledovanii. Moscow. Mukhin, M. Iu. 2006. Aviaprom. Aviatsionaia promyshlennost’ SSSR v 1921–1941 gg. Moscow: Nauka. Shakhurin, A. I. 1985. Kryl’ia pobedy. 2nd edition. Moscow: Politizdat. Talanova, L. E. 1999. ‘‘Sovetkaia voennaia aviapromyshlennost’ v 1929–1945 gg. na primere zavoda no. 21.’’ Candidate of Historical Sciences Dissertation. Nizhnii Novgorod. TsSU. 1972. Tsentral’noe Statisticheskoe Upravlenie SSSR. Narodnoe khoziaistvo SSSR. 1922–1972. Iubileinyi statisticheskii sbornik. Moscow: Statistika. Zhuravlev, S. V., and M. Iu. Mukhin. 2004. ‘‘Krepost’ sotsializma’’: Povsednevnost’ i motivatsiia truda na sovetskom predpriiatii, 1928–1938 gg. Moscow: Rosspen.

8

The Market for Inventions: Experimental Aircraft Engines mark harrison

The second quarter of the twentieth century witnessed an astonishing revolution in military technology. World War I saw the beginnings of motorized warfare; aircraft were used in combat and the first tanks appeared. By the end of World War II they were the primary armament of continental warfare. The interwar period also saw the scientific breakthroughs that would eventually lead to radar, guided missiles, and atomic weapons. The Soviet economy was large but poor, and it was particularly poor in the scientific and information infrastructure that made the other powers rich. Despite this, the Soviet defense industry was invariably close to the forefront of global developments. Behind the scenes there was a good deal of emulation of foreign progress and outright copying; copying, however, was rarely as simple as might appear at first sight because what could not be copied was the ‘‘how-to’’ knowledge of materials and working them that gave rise to precision and reliability. Technologically, therefore, we are looking at a success story. In writing the economic history of the underlying processes, hindsight is very useful because it tells us that the successes actually happened. To understand how they happened, however, how they were organized and how the decisions were made that brought them about, there comes a point where we must throw hindsight away; we must try to see the process through the eyes of

210

The Market for Inventions

211

contemporaries who did not know where their actions would lead. When we do this, we find that managing technological change is quite similar to managing mobilization planning (Chapter 5) in two respects. First, those responsible faced huge uncertainty about what they were planning for. Second, everyone had an axe to grind. The uncertainty was over the most promising directions of technological development. There were many possibilities, of which only a few would prove fruitful; as the economic historian Joel Mokyr (1990: 176–77) has pointed out, the many inventions that failed were part of the cost of success. The specialists themselves did not appear to suffer from this uncertainty, however: they were usually certain about the right way to go forward, or behaved as if they were certain. The problem was that they did not agree among themselves, and could not all have been right at the same time. In aircraft propulsion in the 1930s, for example, there appear to have been at least four groups. One group believed that the future lay with rockets, so they designed rocket aircraft. Another group believed, equally strongly, that the rocket engineers were fools, and that the future lay with the jet engine. A third group, about which history has almost forgotten, believed that the technology of steam turbines, already tried and tested in naval propulsion and power generation, could be applied to aviation. A fourth group thought that the first three groups were all dreamers, and the practical way forward was to concentrate on improving the existing reciprocating engines. Stalin, who took a keen personal interest in the modernization of the Red Army’s weaponry, warned the aircraft designer Aleksandr Iakovlev (2000: 501): A designer is a creative worker. Like the painter of a picture or the writer of a literary work, the product of a designer’s or scholar’s creativity can be successful or unsuccessful. The only difference is that from a picture or verse you can tell the author’s talent right away. . . . With a designer it’s more complicated: his design can look very attractive on paper, but final success or failure is determined much later as a result of the work of a numerous collective and after the expenditure of substantial material means. . . . Most designers get carried away with themselves and are convinced of their own and no one else’s righteousness; on the basis of an overdeveloped self-regard and the mistrust that is characteristic of every author they tend to attribute their own failures to prejudice against themselves and their creations.

With hindsight we can see that from the point of view of fighting the coming war, the fourth group of conventional improvers was right, because none of the others contributed anything practical to the Soviet war effort. We also see that the third group, the steam turbine engineers, was marching confidently

212

Mark Harrison

into a dead end. Using a longer time horizon, one that extends beyond 1945, the first two groups will each be found to have made a valid contribution, but the rocketeers’ was valid only by accident: the rocket was not the answer for aviation, and would turn out to solve the problem of strategic bombardment only because of the atomic bomb, which had not yet been invented. The jet engine, in contrast, would dominate military and commercial aviation into the next century. Its designers expected this and said so at the time; as it happens, therefore, they were the ones who were proved right in the long term. Those responsible for directing and funding research and development (R&D), in contrast, knew nothing of this. Their first handicap was that they did not have a clear idea of the appropriate time horizon over which to demand results; they did not know whether there would be a war or how long it might last. Second, they were also less well qualified than the specialists themselves to judge competing claims. What were they supposed to do? Given limited ruble funding of military research and many competing claims on it, they tended to ration it out across a wide range of projects, giving nearly everyone a little and no one as much they wanted. The axes being ground were the special interests of the rival parties. Each declared that their own motivation was unselfish and based on the interests of the collective: they had no private interest, only the interests of the party and the government at heart. It was those who stood in their way that were selfishly motivated. Economic reasoning suggests, however, that each had a clear private interest at stake, and that this would have important consequences for the efficiency of resource allocation. For illustration, suppose that people generally differ in the amount of talent they have, and this decides how productive they are; for simplicity, they can be talented or untalented. In most occupations what kind they are would quickly become obvious, and as a result two salary levels would be established, high (for talented people) and low (for the untalented). But not all work is like that. Military invention provides cases of projects that lasted many years before coming to fruition or being closed down. In long-term projects it may not become apparent how talented or productive are the employees until the project is finished; in the interim, everyone looks alike. Meanwhile, all must be paid the same, regardless of true talent. What are the likely consequences? For aircraft propulsion between the wars, the Soviet authorities had a budget from which they could choose to employ many or few designers. Other things being equal, the more designers they employed, the greater their chance of final success. But other things would not be equal. The number employed would influence the average salary paid: the greater the number, the lower the salary.

The Market for Inventions

213

Suppose the funding authority paid the average salary in the economy: the average would be above the low rate paid to untalented people, and these would be attracted into military research. Effectively, military R&D would become a field in which the untalented could gain a temporary salary premium while hiding their lack of talent. But the average would also be below the rate payable to talented people elsewhere, and many would leave military projects in response. Some would stay: those with sufficient inner motivation to compensate for the financial sacrifice. But talented people, even if some do not care about themselves, are also likely have families and dependents, and this means that few can ignore the pay disadvantage completely. If the authorities responded by cutting back employment and raising salaries above the rate for talent, they would win more talented designers back into the field but the premium on the untalented would now be exceptionally high and these people would crowd into employment in even greater numbers. Simultaneously, those seeking work would rise while those obtaining it would fall; the talented minority would be squeezed by the untalented crowd, and this would risk more damage to the general chances of final success. This is the problem of adverse selection: there were two types of project, good and bad; because the authorities could not tell them apart, bad projects would tend to drive out good ones.∞ Finally, although technically ignorant, the authorities were not economically stupid and rationally mistrusted the designers that stayed in the market for this very reason. Even in the context of a highly centralized command economy, it is impossible to describe this process without using the terminology of markets. The first part of this chapter describes the ‘‘market for inventions’’ in more detail: who the actors were, what the market covered, and how decisions were made in it. The second part looks at the designers and their active role in promoting research and securing funding. In the third part we look at the market from the point of view of the funders, who faced the difficult problem of deciding when to discontinue funding for unsuccessful research. The fourth part concludes. This chapter is about the market for experimental aircraft engines. The market was opened in 1932 and closed down in 1946. I do not try to narrate the story of what happened in the market in any detail.≤ The early and mid- 1930s saw a wide-ranging exploration of alternatives. Soviet designers worked on lines parallel to progress in other countries, but in isolation from it and lagging behind by a margin that was often narrow. As war approached, the authorities’ enthusiasm for radically new long-term developments waned and their interest became narrowly focused on quick results. Research on steam turbines was recognized to be going nowhere and came to an end. Interest in jet and rockets revived during the war, magnified by the appearance of German V2 rockets and

214

Mark Harrison

German and British jet aircraft and, at the war’s end, the Allied forces’ seizure of technological trophies in occupied Germany. These developments drove away the uncertainties, forced the resolution of differences, and led to the closing down of the market. By 1946 Soviet specialists were working to Stalin’s order on German rockets and British and German jet engines. They quickly brought them into production and also began to improve them. Their ability to do so was an achievement in its own right; it would hardly have been possible without their background of independent endeavor.

The Market for Inventions THE PLAYERS

There were four main sets of players in the Soviet market for inventions: a Dictator, the Army, Industry, and the Designers. Stalin presided over the market process through the committees and commissions responsible for defense matters described in Figure 3.2; he also took an active personal interest. Strategic decisions were taken at this level that framed the market for inventions by defining government funding priorities and authorizing major organizations to enter or leave the market; for example, on July 4, 1932, the Defense Commission issued the decree that first approved initiatives in steam and gas turbines and rocketry; subsequent decrees taken at a similar level through the 1930s and 1940s created new research institutes and gave priority to specific projects (Danilov 1981: 71). The result was an internal market, not a real market, as we defined them in Chapter 3. But, like the market for weapons, the internal market for inventions had a life of its own. The market process was played out by actors representing the Army and Industry; these are illustrated in Figure 8.1. The Army’s chief role, fulfilled by the defense ministry as consumer, was to disburse much of the funding for military research. Most commonly the Army contracted out inventive activity to the research institutes and design bureaus of Industry. The most important, and interesting, direction of the flow of funding therefore followed the solid arrows from the Dictator through the Army to Industry and Industry’s Designers. As we shall see, however, there were also subsidiary flows, shown in the figure as dashed arrows: the Army carried out some research and design activity in-house, and Industry also funded some R&D on its own account. Normally, the Army and Industry independently formulated operational plans for research and experimentation that were then coordinated through a contracting process. The most important planning horizon was annual. The

The Market for Inventions

215

Figure 8.1. Dictator, Army, Industry, and Design Organizations: Funding Flows

Red Army had an annual plan for the development of military inventions, most of which it contracted out to industrial organizations through the internal market for inventions. Industrial ministries also had their own R&D plans —for example, the annual plan for experimental aircraft engines to be carried out by the institutes and bureaus of the aviation industry, part of which was made up by contracts accepted from the Red Army. These arrangements imposed the following structure on competition among designers. There were many designers and many design organizations. Because projects had to have a home, the industrial design bureau was the main vehicle for this competition. Designers competed for funding from a few potential sources. In principle the defense industry monopolized the market, but in practice its monopoly was sometimes threatened by other parties: other industries with sideline interests that led them to seek diversification, military men interested in the scope for vertical integration, and the Dictator who could revoke the delegation of his powers to the market at any time and impose direct control under the NKVD. As the principal funder the Red Army made various attempts to bypass the internal market for inventions and substitute itself for Industry. Military research institutes and design bureaus carried out some in-house research. During his time in charge of the procurement of Red Army equipment, marshal

216

Mark Harrison

Mikhail Tukhachevskii, executed in 1937, was an eager modernizer and an enthusiast for rocket armament and aviation; he was particularly keen on keeping it under direct military control (Holloway 1982: 321). The NKVD also intervened from time to time by seizing the designers and managing them on a prison basis (Albrecht 1993: 133–35; Starkov 2000: 255–60; Mukhin 2004a). The owner of factory no. 16 in Kazan’, for example, was the ministry for the aircraft industry, but its engine design bureau was a prison bureau staffed by prisoners and run by the NKVD fourth special department. SCALE AND SCOPE

Between 1932 and 1946 there were, in total, roughly thirty major projects in new kinds of aviation propulsion. It is a rough count based on the activities of major designers, institutes, and bureaus recorded in the plans, reports, and memoranda of the ministries of defense, internal affairs, and the heavy, defense, ammunition, and aircraft industries.≥ The count is likely to be incomplete to the extent that records were incomplete and some activity was unplanned, but it has the great advantage of being based on records that were compiled without hindsight, the point being that hindsight tends to lose sight of the false starts and failures that were an essential part of the invention process. The breakdown of these projects may surprise the reader. There were seven major projects in rocket research; these projects include the rocket engines and experimental aircraft of the future Soviet chief missile designers Vladimir Glushko and Sergei Korolev, whose work remained heavily focused on aviation until after the war. Because of these projects’ future importance for the postwar missile and space races, Western analysts and historians have given them close attention.∂ But this was actually the smallest group of projects. More was spent elsewhere. The next group was the ten major projects in steam turbine development. It may be unexpected to find that so much serious effort went into the unsuccessful attempt to build steam aircraft. But it is only surprising with hindsight. As long as the problem of a reliable, efficient, long-range, high-altitude successor to the reciprocating engine remained unsolved, it was sensible to go on looking at all potential answers; the 1930s saw similar projects in Britain, France, Germany, and the United States, a fact of which the Soviet authorities were well aware.∑ It is likely, however, that more was spent on this line of research in the Soviet Union than in the rest of the world put together. The largest group was the thirteen jet engine projects. These projects were highly varied and only two would result in workable engines by the end of our period: the first Soviet turbojet and turboprop engines, designed by Arkhip

The Market for Inventions

217

Figure 8.2. Experimental Soviet Rockets and Gas and Steam Turbine Aircraft Engines, 1932– 46: Cumulative Investment (major project-years) Source: Steam turbines from Harrison (2003); jets and rocket motors from Harrison (2005).

Liul’ka and Vladimir Uvarov, respectively. Most of the other projects involved either simple ramjets that could boost a conventional aircraft only when it was already flying at speed, or ‘‘hybrid’’ attempts to work around the most daunting technological challenges of the turbo jet. The usual workaround was to use a conventional reciprocating engine rather than a gas turbine to supercharge the jet engine’s air intake; the Italians famously flew a hybrid jet aircraft of this type, the Caproni-Campini N1, from Rome to Milan in 1940. Figure 8.2 shows how the investments in R&D may have mounted up. The unit of measurement is a ‘‘major project-year’’; the measure is arbitrary, but perhaps no less arbitrary than a monetary scale. The figure shows, on this measure, that knowledge and experience built up most rapidly for steam turbines until 1939 when, at sixteen project-years, the authorities determined that this path was a dead end and they did not need to find out any more. Jet engine research did not catch up until 1942 but then accelerated, reaching forty-four project-years by 1946. Even by 1946, rocket research had barely caught up with the prewar effort in steam turbines. There was considerable project turnover, and this is of interest because we would like to know more about how projects came to be seen as promising enough to win startup funding, and how they came to be terminated as failures. Figure 8.3 illustrates the turnover for experimental engine projects in the aggregate. In most years there were a couple of startups, sometimes one, sometimes three. Terminations were less frequent and were more bunched. A few events lie

218

Mark Harrison

Figure 8.3. Experimental Soviet Rockets and Gas and Steam Turbine Aircraft Engines, 1932– 46: Major Projects, Total Source: Steam turbines from Harrison (2003); jets and rocket motors from Harrison (2005).

behind the irregularities seen in the figure. The Great Terror of 1937/38 led to the arrest of some famous rocket designers, including Glushko and Korolev, and a complete break in rocket research, but because projects in progress were few there was little influence on the figures in the chart. In 1939 there was a major shift in emphasis from steam to gas turbines; this shift is reflected in both the high casualty rates of 1938 and 1939 that brought steam turbine research to an end, and the large number of fresh startups in 1939. The outbreak of the war in June 1941 led to a temporary suspension of jet research from which no quick payoff was expected, and a turn back to rocket aviation, which was thought at the time to be nearer to a practicable high-speed interceptor. As a fraction of national resources, the sums involved were tiny. The 1937 annual plan for the most important research organization in rocketry envisaged 476 staff, including 118 engineers, with a value of work of 4.5 million rubles.∏ But three-fifths of its work was unrelated to aviation.π In contrast the overall value of equipment orders for the army and navy in the same year was 5.7 billion rubles (Davies and Harrison 1997). EFFORT AND REWARD

Those who worked in the market for inventions could expect to be modestly better off than others. The average monthly pay of specialist (‘‘engineering and technical’’) workers at research institute no. 3 (NII-3) for ammunition in the first quarter of 1941 was 818 rubles, roughly two and a half

The Market for Inventions

219

times the average industrial wage of 1940.∫ More detail is available for the 250 ‘‘management and administrative’’ staff of Liul’ka’s experimental aircraft factory no. 165 in August 1946; this category included everyone from the chief designer (6,000 rubles a month) to the floor sweepers in the labs (200 rubles). The median monthly wage was 875 rubles, compared with 626 rubles for the average industrial wage in December of that year (Filtzer 2002: 235). The pay was enough to staff the factory fairly fully; there were only eight vacant posts, of which seven offered less than 400 rubles.Ω Basic pay was just the start; there were plentiful supplements and incentive payments. The director of research institute no. 3, for example, was said to have received 19,250 rubles on top of his salary in 1939 and the first half of 1940.∞≠ The evidence is piecemeal and we have no clear picture of how such sums were fixed or allocated. Documents suggest that officials tended to seek approval for making side payments to mark significant reorganizations, achievements, and anniversaries. In November 1933, for example, the chief of the Red Army administration for military inventions asked for 2,500 rubles for bonuses to mark the first Soviet liquid-fueled rocket and the establishment of the new jet propulsion research institute (RNII).∞∞ In July 1940 the director of research institute no. 3 asked for his most outstanding staff, not named, to be given medals to mark the successful exploitation of its rocket shells in combat against Japan and Finland.∞≤ In July 1947, armament minister Dmitrii Ustinov and NKVD chief Sergei Kruglov wrote to Stalin for permission to award commemorative decorations to the former ‘‘enemies of the people’’ working in the prison design bureau no. 172 (OKB-172) to mark its tenth anniversary.∞≥ What is striking is the lack of rules and guidelines; everything was argued ad hoc. To put it another way, any excuse probably would have done. The lack of rules was a source of danger to those making or approving such requests, which could easily be made to look corrupt. Prewar investigations, for example, redefined many instances of side payments and awards as unjustified after the event. A finance ministry audit of defense industry research establishments in 1938 found that the central institute for aeroengineering (TsIAM) was running no less than nineteen separate incentive schemes on which it had spent 1.2 million rubles in 1937 along with another 200,000 rubles on rest cures and sickness benefits.∞∂ An audit of two years for research institute no. 3 identified a loot chain through which the director not only made ‘‘unjustified’’ side payments to himself but also used incentive schemes to pay off his colleagues and bosses.∞∑ The point is not whether such payments were truly unjustified, but that the lack of rules at the time makes it impossible for us to judge after the event whether or not the payments were justified.

220

Mark Harrison

Finally, of significance equal to or greater than cash in a shortage economy was the privileged consumer provisioning available to those whose jobs gave them the right to a Moscow residence permit. In the 1930s the aircraft industry opened a number of new factories in the provinces, some near Moscow, some thousands of kilometers to the east. Mukhin (2004b) has shown that Moscow-based aviation specialists could be persuaded to leave the capital only with great difficulty, and then only when the ministry gave a written guarantee of their future right of return. Their reluctance was the same, regardless of whether ‘‘the provinces’’ were on the Pacific coast or only just outside the Moscow city limits.

Who Drove the Market? Although it looked as if planned from above, the internal market for inventions was actually driven by the designers. The stereotype of a command system might lead one to expect the dictator to have controlled the market by simultaneously issuing funds to the Army and orders to Industry, leaving only a small role for the subsequent negotiation and exchange of contracts for the work to be carried out. As Andrei Markevich has shown in Chapter 4, something like this was the intention in the planning of the defense industry’s current production, although it was not and could not have been implemented perfectly. As far as the planning of inventions is concerned, this model was not an option at all. The difficulty was that given the uncertainty surrounding the future of technology, Stalin did not and could not know what orders to give in the first place. The designers knew better, and the dictator had to adapt to this reality. This applied even to the highest-level strategic decisions, which seem to have been strongly influenced by lobbying from below. Not surprisingly, successful designers had to be active lobbyists; they were what Donald Mackenzie (1996: 13) has called ‘‘heterogeneous engineers,’’ capable of building networks as well as machines, and reshaping organizational as well as technological constraints. Everyone played this game. As the minister of the aircraft industry later explained in February 1941, ‘‘Work on the creation of jet propulsion engines at home in the USSR . . . began on the initiative of a few engineers taking the form of inventors’ proposals.’’∞∏ The rocket engineers found their sponsor in the soldier Tukhachevskii, whose personal plan seems to have been to build exclusive links with them and try to monopolize new technologies in artillery and aviation for the Army (Siddiqi 2000: 4–7). Some of them paid a heavy price for this private connection in 1937, when Tukhachevskii was arrested and executed as a spy. The steam engineers eventually found their patron in Industry, under the aircraft industry

The Market for Inventions

221

leader Mikhail Kaganovich, who later mentioned how it came about: ‘‘Three years ago comrade Tsvetkov came to me and proposed making such a turbine, I went to the boss, the people’s commissar signed a decree to the effect that, in urgent order, under personal responsibility, [inaudible] to build a turbine.’’∞π Liul’ka, father of the Soviet turbo jet, started his work in 1936 on the back of a steam turbine project in Khar’kov, but managed to attract independent funding only after a three-year struggle with the authorities including a personal encounter in Moscow with minister Kaganovich and a written petition to prime minister Molotov (Berne and Perov 1998: 78–81). Kaganovich was a second-rate patron; despite his much more famous brother Lazar, one of Stalin’s closest associates, Mikhail lacked clout and his ministerial career ended in failure in 1940. After Tukhachevskii’s execution in 1937, therefore, the cause of aviation jet propulsion lacked a powerful sponsor until Stalin’s deputy Georgii Malenkov began to take an interest in 1943.∞∫ This was a timely move because both German and British inventions were about to materialize in the skies above Europe. One consequence of the designers’ initiative was that projects tended to proliferate beyond intended limits, in an uncontrolled way. Designers worked to secure ministerial approval and the funding that followed. If refused at one level, they appealed to the next. If necessary they began work without waiting for authorization; they illegally diverted resources of their own design organizations that had been allocated to other uses and then used the preliminary results to support subsequent attempts to gain official backing. The director of research institute no. 3 outlined the consequences at an internal meeting held in May 1942, when the wartime strain on resources was at its most acute: As an example of how we are forced to diffuse the attention of our cadres I will take the first research department. There are 26 [research] topics for 10 engineers. Some of these topics are incidental to our institute and do not match its profile or specialization. These topics arose because there were people to put them forward and instead of passing them on to those organizations for whom such topics were more appropriate we engaged in them ourselves. . . . It’s characteristic of such topics that working on them involves unnecessary investigations since [we have] no corresponding experience. Often what is done is done many times, and all because we took on what was not our business, because we have neither experience nor cadres to work on items that don’t match the profile of our institute.∞Ω

Finally, designers monitored the foreign military and technical press. When they found evidence of aerospace experimentation in rival states, they worked it up so as to demonstrate the advances being made elsewhere, contrast it unfavorably with the state of affairs at home, and promote their own bids for

222

Mark Harrison

funding.≤≠ Conversely, when foreign press information became sparse, they used its disappearance to call urgently for increased funding on the grounds that foreign powers were evidently forging ahead of the Soviet Union in secret.≤∞ There is no indication in the aircraft engine designers’ files that they gained any information about progress in jet engines or rocketry in other countries that was not freely available in the press. If Soviet spies did acquire such information, it did not reach the designers.

Not Knowing When to Stop To summarize: the authorities had no difficulty eliciting proposals to start up research. Because there were many projects to fund, they funded them in installments. At the start of a project the officials usually had little more than an engineer with some qualifications and letters of personal recommendation who could talk at them knowledgeably and with enthusiasm about matters beyond their experience. By limiting startup funding and making further installments conditional upon progress reports, they could increase their information about the quality of projects beyond that which was available initially.≤≤ The time for approving the next installment became an opportunity to review each project, evaluate its results so far, and decide whether to continue or discontinue funding. The evidence suggests, however, that the authorities did not normally have a good idea of when to stop. Rather, funding decisions were reactive; the fact that a project had been previously approved so that initial funds had been disbursed and work begun was a sufficient reason, other things being equal, for funding to be continued. There are two possible reasons for this, one political and the other economic. The political mechanism would work as follows. In every political system there are channels that allow resources to be exchanged for loyalty. Military research was possibly one such channel in the Soviet system. The rocket scientists that Tukhachevskii patronized, for example, clearly developed loyal feelings toward him; did he fund their work because he wanted their inventions, or because he wanted their loyalty, now or in the future? If the latter, then in this case the market for inventions was primarily a political market in which loyalty was the desired final product and inventions were an incidental by-product. The alternative is that the funding authorities actually wanted technological results and were relatively uninterested in the personal loyalties of the technologists. Why then did they appear ready to roll funding over from year to year, even when progress was slow or nonexistent? Possibly because they were rationally unable to enforce the strict success criteria that they demanded

The Market for Inventions

223

initially. The reasoning of Mathias Dewatripont and Eric Maskin (1995) suggests the following example; to keep it simple, the discount rate is zero. Year 0. A project is expected to cost 100,000 rubles a year and last two years, so its total cost should be 200,000 rubles. Its result will be worth one million rubles if it succeeds, and the funding authority estimates its chance of success at 25 percent, giving an expected value of 250,000 rubles and a surplus over costs of 50,000. The project is approved for two years. If the project’s expected duration were three years, however, implying a total cost of 300,000, a loss would be expected and the idea would be rejected. Year 2. Two years have passed and the 200,000 has been spent with no result. Should the funder close the project down or let it continue? The designers ask for another year and another 100,000; they say they can still obtain the desired result. The funder, now more skeptical, revises the project’s chance of success down to 20 percent, say, so its expected value has fallen to 200,000. But this expected value of 200,000 is now available for an outlay of only 100,000, leaving a surplus equal to 100,000, which is more than before! The funder will approve the project for another year, or lose the expected surplus. It is true that after the extra year the total outlay of 300,000 will have exceeded the expected value of the project; if the funder had known this beforehand, the project would never have been allowed to start. But the first 200,000 is a ‘‘sunk cost’’; it is gone beyond recall. Because it is gone, it should no longer enter into the funder’s calculation; two years having gone by, only the marginal cost of the third year remains relevant. In short, once the first installment has been paid and has become a sunk cost, the payment of the next installment becomes more likely. Projects can win continued funding even after it is known that they should never have been started. The evidence from the Soviet market for military inventions, although not entirely straightforward, allows us to rule out the political mechanism in most cases. If the main motive for funding projects was to win the agent’s loyalty, it is necessary that the principal should have signaled this, because there had to be a way of letting the agent know that the funding was more than what any objective social planner would have provided; otherwise, why should the agent offer loyalty in exchange? We can imagine the sort of ceremonial that would have welcomed success and excused failure alike: ministers would have made speeches emphasizing the common goals, the difficulties of the task, the comradeship of the struggle, the great efforts made, the experience gained, the foundations laid, the possibilities of future progress, and the valuable spin-offs generated along the way. The evidence does not match this at all. When faced with a lack of results, ministers did not excuse failure and praise effort but became impatient and

224

Mark Harrison

bullying. Kaganovich accused the steam turbine engineers of losing touch with reality, and mocked them for confusing aviation with railroads: ‘‘You could put a F[eliks] D[zerzhinskii] locomotive in an aircraft, but then the aircraft would weigh 2,000 tons. This is comrade B[inaudible]’s fantasy, he’s got 245meter wings and a 45-meter fuselage. . . . We’re not talking about a boiler on a Tsvetkov locomotive [Tsvetkov was a turbine specialist]. . . . If we put an airscrew on a locomotive it’ll turn, but we need to put it in an aircraft at altitude. . . . I can’t sit for three years and see no results.’’≤≥ As the evidence of failure mounted the engineers pleaded the value of potential spin-offs from their work on aviation for naval and locomotive engineering, to no avail (Rodionov 2005, under January 15, 1938). It is possible that Stalin had favorites and this gave them some limited protection. Gennadii Serov (1997: 4) has suggested that Stalin favored the rocket designer Andrei Kostikov with disproportionate funding. In November 1942 Stalin authorized an unproven Kostikov design, the 302 rocket fighter, for development at a time when other new initiatives were being ruthlessly subordinated to the mass production of existing lines. When the 302 proved a failure Kostikov was sacked, then arrested, but released a year later, and he was allowed to retain his military rank and decorations. Korolev’s biographer Iaroslav Golovanov (1994: 511) claims, ‘‘Stalin needed Kostikov, since [the latter] was one of the standard bearers of the Stalinist world order.’’ Stalin was habitually suspicious of those he favored, however. In 1950, for example, he suddenly accused his favorite aircraft designer Aleksandr Iakovlev (2000: 395) of diverting state funds into excessive salary and bonus payments: ‘‘Do you know what they say about you behind your back? They tell me you’re a thief.’’ What saved Iakovlev was the support of his boss, minister of the aircraft industry Mikhail Khrunichev, who proved to Stalin that Iakovlev’s design team and production workers were fewer in number, lower paid, and less well equipped than those of the other designers. In short, the funding of research proceeded generally on the basis that the authorities wanted results, not loyalty. In the presence of sunk costs, however, there were no clear rules for defining failure or terminating failed projects. In the absence of rules the authorities tended to roll funding over from year to year until their patience suddenly ran out; then, the money stopped and heads rolled. In the case of rocket research at the jet propulsion research institute, for example, frustration boiled over in the context of the Great Terror.≤∂ Tukhachevskii was arrested in May 1937. The purge of the rocket scientists began in October with arrests including the director and the rocket motor specialist Vladimir Glushko. In June 1938 work on the Korolev-Glushko rocket aircraft

The Market for Inventions

225

project was suspended on the grounds that it was diverting funds away from work of more immediate military utility such as rocket artillery; Korolev was arrested, charged with being a Trotskyist saboteur, and sentenced to ten years’ forced labor. Another rocket establishment was closed down in the following way. Tukhachevskii had set up military design bureau no. 7 (KB-7) in 1935 to promote rocket aviation. In January 1938, with Tukhachevskii gone, the design bureau was transferred from the Army to Industry, handed over to the ministry of the defense industry’s chief administration for ammunition. But design bureau no. 7 never produced any results; its annual report for 1938 says, for example: ‘‘for armament in 1938 nothing supplied, in view of the long-term [ perspektivnyi] character of work.’’≤∑ Then, in early 1939 the Red Army and Navy decided unilaterally to stop funding 47.5 million rubles’ worth of research and experimentation out of 77.5 million previously agreed with the ammunition industry; this left design bureau no. 7 entirely with funding.≤∏ The bureau was closed down before the end of the year; according to Siddiqi (2003) it imploded with the staff denouncing each other and the director, who was arrested and imprisoned. To summarize: the funding authorities were vulnerable to a failure of commitment: before the event, they intended to fund only good projects and were ready to discontinue the funding of bad ones, but after the event it was harder than expected to carry out their intention, and low- or no-return projects continued to squeeze the funding of those that would eventually bear fruit. In retrospect the surprise is that the Soviet market for inventions worked as well as it did. Research and development is, first and foremost, a forwardlooking activity. No one would undertake it that did not have a long time horizon or was unwilling to wait patiently for results. Of next importance, R&D contracts normally involve a complicated sequence of immediate advances by the funder in the expectation of future results from the designer; normally, they also allow for sharing risks of success and failure proportionally between the two sides. One might expect, therefore, that R&D would work best within stable rules so that each side could make clear long-term commitments, confident in the expectation that the commitments made by the other side would be fulfilled. Yet this is not what we see. The Soviet market for inventions was chaotic. It was driven by the designers, who did not form a quiet orderly line for funding but joined an unruly mob that pestered officials without mercy and ruthlessly jostled each other out of the competition. Once in the market many designers obtained some financial returns to their efforts, but the rewards seem to have borne little relation-

226

Mark Harrison

ship to results. The designers also incurred nontrivial personal risks, dressed up as punishment for failure, but the criteria for penalization, including the time horizon over which they were judged to have failed, were fixed arbitrarily and revised without warning. The lack of rules, not particularly surprising in itself, was the product of a harsh dictatorship that ruled by decree in all areas of economic life. Despite this, the Soviet market for inventions succeeded in the sense that it gave rise to a stream of excellent designs that kept the Red Army close to the world military-technical frontier. In short, chaos did not frustrate intentions. It is possible that the lack of rules helped the authorities to mitigate the problem of adverse selection. Adverse selection arose when there were two types of project, good and bad. The funding authorities, unable to tell them apart, had to offer the same funding to both types. This would have made research and design a safe occupation for the misguided, the talentless, and the time-servers, while squeezing the funding available for the truly talented. Hence the risk that bad projects would drive out the good ones. The authorities’ interventions changed the incentives by turning research and design into a dangerous activity. It was not exactly that they punished failure; they clearly tried to, but failure was something for which they had no clear definition. As a result those who were swept up in purges, arrested, and jailed, included some like Korolev and Andrei Tupolev who eventually succeeded and became iconic figures of postwar Soviet technology. In other words, Stalinist repressions made R&D more dangerous for everyone, talented and untalented, alike. The result may have been to deter some of the untalented but risk-averse designers from promoting bad projects. At the same time the truly talented designers were not deterred because they were more driven from within; they were more willing to undertake risks, or were more confident of achieving final success, or both. What made the great designers great is that they just wanted to build aircraft and jets and rockets and they would attempt to overcome any obstacle to do so. They did not want a quiet life. This motivation could survive years of frustration and poor conditions including, for some, moral and physical abuse, imprisonment, and forced labor. Thus Stalin’s terror tried their mettle and proved their comparative worth in ways that more orthodox tests could not.

Notes I thank the Leverhulme Trust, the British Academy, and the University of Warwick Research and Innovations Fund for financial support of my research on ‘‘Invention, Imitation, and the Birth of Soviet Aerospace’’; the University of Warwick for study leave;

The Market for Inventions

227

and the staff of the State Archive of the Russian Federation (GARF), the Russian State Economic Archive (RGAE), and the Russian State Military Archive (RGVA) for access to documents. 1. George Akerlof (1970) argued that if sellers offer secondhand automobiles that are either of bad quality (‘‘lemons’’) or good, and sellers know the quality of what they offer but buyers do not know the quality of what they buy, buyers will offer an average price that will be attractive only to the sellers of bad automobiles; it will be too low for sellers of good automobiles to wish to sell, and bad automobiles will drive out good ones. This is one of the propositions for which he shared the 2001 Nobel Prize for Economics. 2. See Harrison (2000, 2003, and 2005). For a standard account written before the archives, see Holloway (1982). 3. These and other figures that follow are found by adding together the numbers of research and design projects in steam turbines from Harrison (2003) with those in jet and rocket propulsion from Harrison (2005). 4. See, for example, Ordway and Sharpe (1979), Holloway (1982), Albrecht (1993), and Siddiqi (2000 and 2003). 5. The projects outside the Soviet Union were surveyed for German readers by Knörnschild (1941) and for the British in wartime by Smith (no date: 36–40). A memorandum of February 28, 1937 (RGAE, 8328/1/919, 77) indicates the extent of Soviet prewar knowledge of these foreign projects, which was fairly complete and mostly correct but lacked technical detail. 6. RGAE, 8162/1/16: 4 (no date but about February 1937). This was the jet propulsion research institute (RNII). 7. RGAE, 8159/1/6: 74 (December 1936). 8. RGAE, 8162/1/449, 87 (April 10, 1941). 9. RGAE, 8044/1/3079, 82–91 (August 27, 1946). 10. RGAE, 7516/1/692, 3 (November 21, 1940). 11. RGVA, 34272/1/146: 145 (Terent’ev to Tukhachevskii, November 16, 1933). 12. RGAE, 8162/1/306: 186–187 (Slonimer to Sergeev, July 22, 1940). 13. GARF, 9401/2/170: 213–228 (July 13, 1947). 14. RGAE, 7515/1/379: 134–137 (April 19, 1938). 15. RGAE, 7516/1/692: 1–7 (November 21, 1940). 16. RGAE, 8044/1/460: 59 (minister Shakhurin to deputy prime minister Voznesenskii, February 5, 1941). 17. RGAE, 8328/1/824: 35 (August 22, 1936). The very first projects in steam propulsion, however, were sponsored by military design organizations, including one under Tukhachevskii’s direct control at the time. 18. For Malenkov’s briefing by aircraft industry minister Shakhurin, see RGAE, 8044/1/984: 264–275 (October 22, 1943). 19. RGAE, 8162/1/574: 101 (Kostikov, director of research institute no. 3 of the ammunition industry, formerly the jet propulsion research institute, May 7, 1942). 20. For examples, RGVA, 34272/1/105: 91–94ob (May 20, 1931) and 118–120 (May 1931); RGAE, 7516/1/324, 1–4 (April 9, 1939); RGAE, 8162/1/305, 30 (April 16, 1940). 21. RGAE, 8159/1/149: 220 (July 26, 1936), 219 (September 29, 1936), and 218 (October 13, 1936); RGAE, 7516/1/324: 10 (no date but 1939).

228

Mark Harrison

22. For annual reports of the jet propulsion research institute and its successor organization for 1936, 1939, and 1940, for example, see RGAE, 8159/1/137: 2–28 (no date but 1937), 8162/1/240: 9–63 (January 9, 1940), and 8162/1/449: 2–61 (January 14, 1941), respectively. The annual reports for 1937 and 1938 were removed from the archive in the 1960s, apparently on the instruction of the USSR Academy of Sciences, and have not been traced. 23. RGAE, 8328/1/824: 12, 15, and 52 (August 22, 1936). 24. For recent studies of the rocket scientists’ purge, see Harrison (2000: 128–30) and Siddiqi (2000: 10–11). The aircraft industry yields a number of design organizations that were closed because of a lack of results. Bartini, Grigorovich, Miasyshchev, Petliakov, Polikarpov, Sukhoi, and Tupolev are all cases of chief designers imprisoned along with their teams; Kalinin was executed (Albrecht 1993: 133–36 and 214–15). The same happened in other lines of work; those charged with designing the Soviet Union’s first atomic bomb, for example, expected medals if it worked, and imprisonment or execution if it didn’t (Holloway 1994: 215). 25. RGAE, 8162/1/89, 125 (no date but 1939). 26. RGAE, 8162/1/299: 36–54 (reports to Sergeev, forwarded by him to Kulik, Frinovskii, and Molotov, March to April 1939).

Published References Akerlof, George A. 1970. ‘‘The Market for ‘Lemons’: Quality, Uncertainty, and the Market Mechanism.’’ Quarterly Journal of Economics 84(3): 488–500. Albrecht, Ulrich. 1993. The Soviet Armaments Industry. Chur (Switzerland): Harwood Academic Publishers. Berne, L. P., and V. I. Perov. 1998. ‘‘Istoriia sozdaniia pervogo otechestvennogo turboreaktivnogo dvigatelia (K 90-letiiu so dnia rozhdeniia A. M. Liul’ki).’’ In Iz istoriia aviatsii i kosmonavtiki, vol. 72: 77–94. Moscow: Institut istorii estestvoznanii i tekhniki RAN. Danilov, B. 1981. ‘‘Iz istorii sozdaniia reaktivnoi aviatsii.’’ Voenno-istoricheskii zhurnal, 1981(3): 70–75. Davies, R. W., and Mark Harrison. 1997. ‘‘The Soviet Military-Economic Effort Under the Second Five-Year Plan, 1933–1937.’’ Europe-Asia Studies 49(3): 369–406. Dewatripont, Mathias, and Eric Maskin. 1995. ‘‘Credit and Efficiency in Centralized and Decentralized Economies.’’ Review of Economic Studies 62(4): 541–55. Filtzer, Don. 2002. Soviet Workers and Late Stalinism: Labour and the Restoration of the Stalinist System After World War II. Cambridge: Cambridge University Press. Golovanov, Iaroslav. 1994. Korolev. Fakty i mify. Moscow: Nauka. Harrison, Mark. 2000. ‘‘New Postwar Branches (1): Rocketry.’’ In The Soviet DefenceIndustry Complex from Stalin to Khrushchev, ed. John Barber and Mark Harrison, 118–49. Basingstoke (England): Macmillan. Harrison, Mark. 2003. ‘‘The Political Economy of a Soviet Military R&D Failure: Steam Power for Aviation, 1932 to 1939.’’ Journal of Economic History 63(1): 178–212. Harrison, Mark. 2005. ‘‘A Soviet Quasi-Market for Inventions: Jet Propulsion, 1932 to 1946.’’ Research in Economic History 23: 1–59.

The Market for Inventions

229

Holloway, David. 1982. ‘‘Innovation in the Defence Sector’’ and ‘‘Innovation in the Defence Sector: Battle Tanks and ICBMs.’’ In Industrial Innovation in the Soviet Union, ed. Ronald Amann and Julian Cooper, 276–414. New Haven: Yale University Press. Holloway, David. 1994. Stalin and the Bomb: The Soviet Union and Atomic Energy, 1939–1956. New Haven: Yale University Press. Iakovlev, A. S. 2000. Tsel’ zhizni. Zapiski aviakonstruktora, 6th ed. Moscow: Respublika. Knörnschild, E. 1941. ‘‘Dampftriebwerke für Flugzeuge.’’ Luftwissen 8(12): 366–73. MacKenzie, Donald. 1996. Knowing Machines. Cambridge, MA: MIT Press. Mokyr, Joel. 1990. The Lever of Riches: Technological Creativity and Economic Progress. Oxford: Oxford University Press. Mukhin, Mikhail. 2004a. ‘‘Aviaprom. Sovetskaia aviapromyshlennost’ v 1921–1941 gg.’’ MS in preparation. Mukhin, Mikhail. 2004b. ‘‘Employment in the Soviet Aircraft Industry, 1918 to 1940: Work Culture, Organization, and Incentives.’’ PERSA Working Paper no. 36. University of Warwick, Department of Economics. http://www.warwick.ac.uk/go/persa. Ordway, Frederick I., and Mitchell R. Sharpe. 1979. The Rocket Team. London: Heinemann. Rodionov, Ivan. 2005. The Aviation and Aircraft Industry of the Soviet Union, 1916 to 1946, Version 5. University of Warwick, Department of Economics, available at http:// www.warwick.ac.uk/go/aviaprom. Serov, Gennadii. 1997. ‘‘V nachale reaktivnoi ery.’’ Samolety mira (3–4): 2–7. Siddiqi, Asif A. 2000. Challenge to Apollo: The Soviet Union and the Space Race, 1945– 1974. Washington, DC: NASA History Division (NASA SP-2000–4408). Siddiqi, Asif A. 2003. ‘‘The Rockets’ Red Glare: Technology, Conflict, and Terror in the Soviet Union.’’ Technology and Culture 44(3): 470–501. Smith, G. Geoffrey. No date. Gas Turbines and Jet Propulsion for Aircraft. London: Flight Publishing Co. Starkov, Boris. 2000. ‘‘The Security Organs and the Defence-Industry Complex.’’ In The Soviet Defence-Industry Complex from Stalin to Khrushchev ed. John Barber and Mark Harrison, 246–68. Basingstoke (England): Macmillan.

9

Secrecy mark harrison

All governments have secrets but some are more secretive than others. In modern democracies public debate often takes the merits of transparent government and an open society for granted. But even in societies where transparency and freedom of information are officially the norm, there is always a core of government where information is gathered and decisions are taken in secret. There are also states where most things are secret. The Soviet state was of the latter type, and was among the most secretive states that have ever existed. Many things were kept secret that in most other societies would be regarded as information open to all. It was not just a matter of keeping this information from the public in the press, on the streets, and in the factory canteens. Just as significant were the many secrets that relatively high-ranking officials kept close to themselves and preserved zealously from each other. Evidently, Soviet leaders were willing to pay a substantial price to hold power in secret. Secrecy was costly. Directly, it was costly to enforce; indirectly, it also damaged the efficiency of the economy. Enforcement required procedural rules for the creation, numbering, distribution, tracking, conservation, and filing or destruction of secret documents. It also required an apparatus to monitor and investigate of cases of disclosure. The punishment of disclosure resulted in the loss of human capital previously invested in agents who turned out to be disloyal. Efficiency costs arose because secrecy created bar-

230

Secrecy

231

riers to the sharing of information that was required to allocate resources efficiently; secrecy meant, for example, that principals decided the general allocation of resources in ignorance of specific facts, while agents determined allocations in detail although they were kept in the dark as regards the broader implications of their decisions. The likely costs of Soviet secrecy suggest that we should inquire closely into its fundamental purposes. Why secrets? Who benefited from the regime of secrecy, and how? Connected with this is the further issue of ‘‘excessive’’ secretiveness. What is the optimal level of secrecy for a government? Did Stalin’s regime take secrecy too far? Why was the Soviet system so extremely secretive, and did the costs of secrecy contribute to its eventual collapse? One of the obstacles to research on secrecy is that the regime of secrecy was itself a secret. The issues that it raises could hardly be investigated while this regime persisted. The chance to study it empirically emerged only after the Soviet Union ceased to exist. Even then, as the historian Jonathan Bone (1999: 66) has pointed out, the attention of historians was diverted from it by the transfer of interest away from the ‘‘top end’’ of the Soviet system to Soviet society and, in particular, to the ‘‘subtleties of the state-society interface’’; and by the greater interest in ‘‘exhuming the Stalinist past rather than in the relatively prosaic work of analyzing it in full forensic detail.’’ Indeed, it is not at all obvious how to investigate secrecy. The evidence of Soviet secrecy is preserved everywhere in the archives, in millions of documents each stamped with its security classification and numbered for limited circulation. But the evidence of its consequences, of how people adapted to it, and how they sought to exploit it, is not held anywhere in particular. It is scattered randomly through the archives. Most of the evidence in this chapter was found accidentally in the course of other research, and most of the latter was found by others who kindly drew it to my attention because they were aware that I would be interested. In this chapter I will describe some aspects of the official practice of secrecy in the Soviet Union under Stalin, using the evidence primarily of the defense industry. This description is necessarily incomplete, because it is not balanced by evidence of the degree of secrecy either in civilian branches, or in the prerevolutionary market economy. Then, I will look at the enforcement of secrecy. I will make a simple point: it is easy to see why governments and states should value secrecy collectively, but secrecy will not be upheld unless each individual agent of the state upholds it. This makes the operation of secrecy a problem of individual motivation. Finally, I will look at the reasons why it might have suited principals and agents in the defense industry to withhold or trade information.

232

Mark Harrison

The Practice of Secrecy Soviet secrecy seems to have been tightened under Stalin in a series of discrete steps; this claim is a first approximation and there is no doubt that future research will refine or modify it. The steps are set out in Figure 9.1, but the figure is just an illustration and the reader should not take it too seriously for a number of reasons: (1) we have no way of measuring units of secretiveness so as to calibrate the vertical scale; (2) the implied ‘‘zero’’ of secretiveness where the horizontal and vertical axes intersect is completely arbitrary; (3) there is no reason to suppose that secrecy increased by equal increments at each step; (4) we have not had the opportunity to make any special study of secrecy before the Revolution so as to provide a baseline for ‘‘normal Russian’’ secrecy; (5) the idea of ‘‘normal Russian’’ and ‘‘normal Soviet’’ secrecy is a conjecture, not a fact. As well as rising secrecy the figure also measures falling standards of disclosure under Stalin’s rule, and this does have a quantitative index, proposed by Abram Bergson (1953), in the length of successive fiveyear plan documents on publication. The Bolshevik Revolution of 1917 brought to power an underground revolutionary party organized on conspiratorial lines with secret lines of communication and secret decision-making processes. Its habits became the habits of government, which began to rule by conspiracy; one of its first acts created political censorship (Goriaeva 2002). There were many secrets already in the 1920s, with much political and military information classified in continuity with the pre-revolutionary practices of the imperial government. Jonathan Bone has shown that the whole system for classifying and handling information was codified by the OGPU in 1927, following a scandal involving the careless handling of secret papers. The documents he has cited (1999: 70–74) make clear that at this stage the whole of defense industry fell within the bounds of secrecy. Classified ‘‘top secret’’ as ‘‘matters of a military nature’’ were all orders, plans, correspondence, financial allocations, and other information relating to mobilization plans for industry and the economy generally, both in detail and ‘‘in any way revealing the general mobilization system.’’ Also top secret were the ‘‘condition and production plans of military and aviation factories’’ and their new construction. Merely ‘‘secret’’ were information concerning naval shipbuilding and repair, correspondence relating to the procurement of imported military equipment, information about the location of individual defense factories, information concerning their supply with equipment ‘‘giving the possibility of drawing conclusions about factory capacity,’’ and photographs or plans of civilian factories supplying military goods. It was at this time that the core factories of the defense

Secrecy

233

Figure 9.1. The Rise and Decline of Soviet Secrecy, 1917–56 Key: S≠: ‘‘normal Russian’’ secrecy. S∞: ‘‘normal Soviet’’ secrecy. 1917: the Bolshevik Revolution. 1927: the OGPU codifies ‘‘normal’’ secrecy. 1937: secrecy ends the publication of current statistics. 1947: Stalin orders still tighter secrecy within the state. 1956: post-Stalin return to ‘‘normal’’ secrecy.

industry lost their names and addresses for public purposes and became known, if at all, by factory number and mailbox number (Cooper 1999). Notable also, under ‘‘material of a general nature,’’ is the ‘‘secret’’ classification of ‘‘the organization of secret document handling; procedure for conducting and archiving secret correspondence’’ (Bone 1999: 74); this ensured that the conduct of secrecy was kept hidden. The transition to the command system that followed almost immediately in 1929/30 further reduced society’s access to information. This came about naturally, without any new measures, because the state expanded at the expense of society and so automatically monopolized a much larger share of the information being produced. The fog thickened further in 1937. The publication of previously available economic statistics was suspended, and this condition persisted for two decades. This made little difference to the defense industry, which was entirely secret already. Many other interesting aspects of Soviet life disappeared from the public record at this time, however. Anne Applebaum (2003: 109–11) has described how, in the first years after 1930 when the Gulag was established,

234

Mark Harrison

the reeducation of criminals through labor was seen as fit for public discussion. By 1937 most of those associated with the propaganda of ‘‘corrective labor’’ had been arrested and their writings had disappeared from the public sphere. The procedures for complete concealment of the Gulag were in place. By 1940 the internal correspondence of the NKVD gave rise to 25 million secret courier items annually. The approach of war in 1941 made little extra difference because just about everything was secret already. It was after the war, in 1947, that secrecy was heightened even further. In response to unauthorized disclosure in the west of some potentially valuable Soviet medical research Stalin introduced a new law that further widened the scope of state secrets and committed further resources to enforcing them. At this time secrecy was at a level that was difficult to enforce: ministers complained that the new law required thousands of officials to act to preserve state secrets, but many of these same officials could not be informed of their new obligations because they were not cleared to receive information at a level of classification corresponding with that of the decree (Gorlizki 2002: 721). But this may not have been a new state of affairs. Speaking with the party secretaries from Leningrad’s secret defense factories in the mid-1930s, the local party chief Andrei Zhdanov referred to the ‘‘game when secrecy is such that the people who are accountable for implementing programs and introducing new production lines don’t know what the programs are, and on the other side enemy sources acquire information earlier than our party people.’’∞ Stalin’s death, and Khrushchev’s distancing from his former patron, brought some relief from this suffocating condition. Addressing the twentieth party congress in February 1956, Stalin’s former deputy and trade minister Anastas Mikoian (1956) remarked drily: ‘‘Without the most careful examination of all the statistical data that which we possess in far larger measure than at any other time and in any other country, without organising these data, without analysing them and generalising from them, no scholarly economic work is possible. It is a source of regret that the statistical data are still classified secret in the central statistical administration in comrade Starovskii’s safes.’’ (Vladimir Starovskii was the Soviet Union’s chief statistician, holding office continuously from October 1940 to August 1975.) In the same year the statistical authorities were allowed to resume the annual publication of an economic yearbook. But they retained a strict monopoly of statistical publication; as late as 1987 the appearance of the first unofficial estimates of Soviet economic growth in a literary journal was a publishing sensation (Khanin and Seliunin 1987). Under Khrushchev and his successors, right up to the end of the Soviet state, the defense industry remained top secret. In his memoirs Mikhail Gorbachev (1996: 136, 215) has written that until he took office as general secretary in

Secrecy

235

1985, ‘‘All statistics concerning the military-industrial complex were top secret, inaccessible even to members of the Politburo’’; ‘‘only two or three people had access to data on the military-industrial complex.’’ He recalls that the long-serving defense minister Dmitrii Ustinov ‘‘essentially had monopoly control’’ over defense information; it was a serious breach of protocol for outsiders, including other Politburo members, even to question him. According to military sources of the same period (cited by Firth and Noren 1998: 260n), the true scale of military funding was known to ‘‘only four men . . . the General Secretary, the Council of Ministers Chairman, the Minister of Defence, and its Chief of the General Staff.’’ Iurii Masliukov, a leader of the Council of Ministers military-industrial commission under Gorbachev, has confirmed, ‘‘Until 1988 summary figures concerning the defense of the country were considered to be a secret of exceptional state importance; a limited circle of people (the leadership of USSR Gosplan and not even all Politburo members) were familiar with them. It was forbidden to copy such figures in the typing pools, and they were circulated in documents by authorized individuals from hand to hand’’ (Masliukov and Glubokov 1999: 105). To summarize: secrecy pervaded the command system of the Stalin era. Secrecy rules built elaborate firewalls that impeded information flows not only from state to society, or from the Soviet state to other states, but also within the state itself. Even within the privileged official sphere information was shared on the basis of need rather than right to know, and the need to know was defined within limits that appear to have been extraordinarily narrow. Secrecy rose and fell; at the moment we do not have a good explanation for this, other than to say that perhaps a dictatorship needs a lot of secrecy to flourish, and increasing secrecy was one of the first steps of the Bolshevik regime. There can also be too much secrecy and it seems as though the secrecy of the late 1940s was too much even for a secretive dictatorship.

Enforcing Secrecy Soviet law penalized the disclosure of official secrets by various means. In the extreme, Article 58(6) of the Russian republic’s criminal code punished ‘‘espionage, i.e. the transmission, theft, or collection, with a view to transmission to foreign States, counter-revolutionary organizations, and private individuals, of information accounted by reason of its contents an especially guarded State secret’’ (Conquest 1971: 743–44). In principle the same law distinguished cases leading to ‘‘especially grievous consequences to the interests of the USSR,’’ punished by execution, from less serious cases for which a term in the Gulag was prescribed. The practical distinction between large and small secrets was probably arbi-

236

Mark Harrison

trary and time-varying. The limited official propaganda suggested that petty revelations were as damaging as more serious ones. As a journalist wrote in 1953: ‘‘In questions of the conservation of party and state secrets, in information that is not intended for disclosure there are no such things as trifles. Sometimes information that is insignificant at first glance can be of great value to a spy.’’≤ The testimony gathered by Alexander Solzhenitsyn (1974: 63–64) suggests that the violations that led to terms of forced labor were often trivial or fabricated. The burden of proof in such cases was evidently low, possibly because it was the kind of thing almost anyone could have done, and therefore in fact probably had done regardless of the evidence that was actually available. Soviet penal practices do not seem to have placed much emphasis on guilty intent or premeditation, as opposed to the accidental or negligent disclosure of state secrets. The sheer scope of secrecy made careless disclosure a serious problem. John Barber et al. (2000: 21) recount that in the spring of 1937 the heavy industry commissariat published figures for the gross output of its civilian products alone, while Gosplan simultaneously published the overall gross output of heavy industry, permitting anyone to compute the value of defense output as the residual. An alarmed reaction from within Gosplan demanded strict punishment of the responsible officials in industry. A clampdown on statistical publication began at about this time and lasted until the post-Stalin thaw. Extensive secretiveness and the ease of careless disclosure made for an environment in which it was virtually impossible for everyone to keep the right side of secrecy regulations at all times. Anyone could let slip a ‘‘trifle’’ at any time and, even if they did not, could readily be accused of having done so. This forged the law into a powerful instrument of repression. Cases of espionage under Article 58(6) on its own or in conjunction with other articles made up 15 percent of the roughly 8,000 executions carried out by the NKVD in the Leningrad district in August, September, and October 1937 (Iliè 2000: 1529). The same proportion applied to the national figure of 681,692 executions by the NKVD in the course of 1937/38 would suggest up to 100,000 cases of espionage that received capital sentences across the country in the years of the Great Terror. The underlying national figure was probably less than this, because the proportion of cases in Leningrad may have been raised by the high concentration of military and defense-related facilities in that locality. At the same time, recall that execution was reserved for more serious cases, and some cases were presumably judged to be trivial. For this reason, executions for espionage were probably the tip of an iceberg comprising all the cases where espionage was alleged. Stalin and Beriia used the fact that the laws on secrecy were difficult to uphold conscientiously in the postwar ‘‘Gosplan affair.’’ Stalin had lost faith in

Secrecy

237

his former favorite, Gosplan chief N. A. Voznesenskii. Investigation then revealed a history of poor handling of secret documents in Gosplan, with many papers that were missing or, if not missing, should have been destroyed. This established the charge on which Voznesenskii and many of his subordinates were executed (Khlevniuk et al. 2002: 274–307).

Collective and Private Motivations What was secrecy for, officially? Throughout this period it was seldom acknowledged in public that secrets existed, and then only in the abstract. A limited-circulation newspaper of the early 1950s provides a rare rationalization of official secrecy and illustrates it at the same time. The Noril’sk combine was the major Soviet producer of nickel and a number of other nonferrous metals of which the defense industry was a major consumer. On August 18, 1953, its daily newspaper published a leading article, ‘‘Strictly Observe Party and State Secrets’’; this was after Stalin’s death and the arrest of Lavrentii Beriia but before the Thaw.≥ The unnamed writer reminded readers that the party’s constitution obliged its members ‘‘to observe party and state secrets, to show political vigilance, remembering that the vigilance of communists is necessary in every organization and every circumstance. Revelation of party and state secrets is a crime before the party and incompatible with membership of its ranks.’’ The article continued: The imperialists are assigning hundreds of millions of dollars to disruptive work against the camp of socialism and democracy. The capitalist encirclement is dispatching its agents to our country and is looking for persons ready to betray the interests of the Motherland and fulfill the assignments of the intelligence agencies of the bourgeois states to undermine Soviet society. Lacking social support in the Soviet land, despairing at its unforeseen moral and political unity, they try to exploit the dregs of society in the persons of diverse renegades and degenerate elements. . . . In the party midst there are still to be encountered individual chatterboxes and scatterbrains. They are not averse to bragging of their inside knowledge among friends and acquaintances, in the circle of their families, by telephone and in personal correspondence and so forth. . . . [We must] explain to the Komsomols [party youth members] that questions considered at closed party meetings cannot be the subject of public scrutiny, that the contents of secret party and ministerial documents are not a subject for conversation even with the most intimate persons.

The rationalization of secrecy in this quotation was, in other words, as follows: hostile states threaten us from outside; for this reason we must conceal our arrangements and capabilities. In this context, secrecy is a collective good.

238

Mark Harrison

The illustration of secrecy is that the Noril’sk combine was an MVD forced labor camp, the existence of which was itself a state secret: reflecting this, it managed to publish a daily newspaper that avoided printing any information that could reveal which combine, who owned it and whom it employed, where it was located, or what it produced. What is the problem for which secrecy was a solution? Western social and historical science has given rise to two views of secrecy. In the tradition of Max Weber (1922/1968) secrecy is seen as providing a private benefit to bureaucrats, who naturally incline to secrecy because it protects them from criticism and the need to account for their actions. Those who adhere to this tradition recognize that the effects of secrecy may be costly but the costs are borne by society, not the bureaucrat. Some costs are political: secrecy tends to undermine democratic values by stifling debate and weakening accountability (Colby 1976; Moynihan 1997). Secrecy may also harm economic efficiency; for example, technological secrecy may be associated with a high level of duplicated inventions (Medvedev 1977) and military secrecy may encourage scientific fraud (Park 2000). Because the bureaucrats do not pay these costs, however, they will always choose secrecy, which, from this perspective, is simply the natural condition of government. A Weberian approach to Soviet secrecy (Tarschys 1985) therefore emphasizes secretiveness as a shared historical feature of European bureaucracies, and as a legacy that Russian autocracy passed on to Stalinist dictatorship. It does not explain why ‘‘normal Soviet’’ secrecy was so much more intense than ‘‘normal Russian’’ secrecy in the absence of Soviet institutions, nor does it explain why Stalin also took secrecy so far beyond any concept of the normal. An alternative approach to secrecy stems from the rational-actor tradition in the theory of international relations (Schelling 1963). The rational actor is the state as a whole, not the individual official. Secrecy is seen to provide collective benefits to the state, but again it is costly. This approach is not concerned with the social costs; rather, the state itself bears opportunity costs of implementing and enforcing secrecy. Abram Bergson (1953: 14), for example, understood Soviet secrecy in terms of two motivations, national security and ‘‘effective propaganda to create favorable impressions,’’ but he also alluded to the costs of secrecy to the state measured by the conflicting ‘‘need to release data for the operation of a nationwide planning system, including the training of personnel.’’ As long as the state is behaving rationally by counting the costs as well as the benefits, the rational-actor tradition suggests that we should see secrecy as a choice, not a natural condition; such a state would choose to cut secrecy back, for example, if its marginal costs began to exceed its marginal benefits.

Secrecy

239

Given the existence of a voluminous literature on the economics of information, economists might be expected also to have shown an interest in secrecy. Much has been written about information that happens to be costly to observe or verify, but official secrecy is of greatest interest where it concerns the state’s need to erect artificial barriers to the transmission of information that would otherwise be easily observed and shared—for example, the location and business of a factory. Similarly, there is a great deal of work on the sort of information that people have a natural self-interest in keeping to themselves—for example, their private medical history. Again, official secrecy is of interest precisely because those who are entrusted with secrets do not necessarily have a direct personal stake in keeping them; secrecy benefits the state as a whole, rather than any individual official. In short, there is no ready-made theoretical framework for understanding the political economy of official secrecy. My starting point is Bergson’s observation that the Soviet state had a natural collective incentive to conceal information for the sake of national security and the reputation of the state. From this point of view it is not hard to understand why the defense industry should have been kept secret. I go beyond this, however, in claiming that collective motivations alone cannot explain the development of secrecy within a state. Secrecy may be in the collective interest of governments and other organizations, but that is not enough. History tells us that secrecy is effective only when it also corresponds with the private interests of individual decision makers; otherwise, they will have private incentives to ignore rules and reveal information or permit others to reveal it. Dilbert illustrates this perfectly in Figure 9.2. Every time we read in the press that some unattributed source inside a company or close to the administration has leaked damaging information, we see that private and collective incentives have gotten out of alignment and the private one has prevailed. Effective secrecy requires some mechanism to bring the private incentives back into line with the interests of the organization. Otherwise, it will fail. The Soviet context provides evidence that this was not just a theoretical possibility but an everyday problem. On one side Stalin regarded secrecy as vital to national security. On the other, the officials responsible for it often did not care enough about it themselves to stick conscientiously to rules. Officials of the interior ministry and the labor camp system regularly put secret documents at risk by failing to store or transport them securely (Harrison 2004 gives examples). In the defense industry itself, the party central committee issued a decree in May 1934 on the system of security passes and guard duties at defense industry establishments. A subsequent investigation by KPK, the central committee’s audit commission, revealed an apathetic response and

240

Mark Harrison

Figure 9.2. Dilbert on the Value of Information Source: United Media. Dilbert: ∫ Scott Adams/Dist. by United Feature Syndicate, Inc.

widespread violations. Reprimands and recommendations had little effect over the following weeks. As follow-up investigations and measures showed, only widespread punitive measures elicited change in the desired direction.∂ The economic historian Avner Greif (2000) has proposed that any institution that persists must be understood as an equilibrium at the level of all the individuals taking part in it: each person participates because it is in his or her interest to do so, conditional upon the expectation that others have made the same calculation. If this is not the case, the regime will fall apart. The inference that I draw is that to explain the extreme secretiveness of the Soviet state, we must look at who benefited from it and how the benefits were shared.

Market Information Information comes in many shapes and sizes, not all of them valuable. In this chapter we will think about two kinds of valuable information that I will call market information and strategic information. We can think of bureaucratic principals as having clear incentives to limit the sharing of both strategic

Secrecy

241

and market information, and we will think briefly about the reasons for doing so and the obstacles they faced. First, market information: this is the information that buyer and seller must exchange in order to make a deal. Without this information, exchanges would be of lower value, or would not take place at all. The information is valuable because sharing it makes both sides better off. An example would be market research. Buyers carry out research to learn the reputation of sellers and the characteristics of their products. Sellers carry out research to identify buyers and customize products to their preferences. As a result, everyone can make more efficient choices. We can stylize the process of planning in the Soviet economy as follows (this account is based on Harrison 2005). A principal issued an order to an agent at a lower level. The order was a demand for output, but the plan also included the advance of some resources for the agent to use in order to carry out the order: for example, a wage and some equipment and materials. This plan was also a legal claim to monopolize the agent’s activities. The plan stated what the agent was legally required to do with the principal’s advance, but it went beyond this: anything that was not in the plan was prohibited and the agent was legally obliged not to do it. But the principal could not observe what the agent did next, and only got to see the end results in some later period. In between, the agent had a choice: be obedient and try to fulfill the plan, or set out to cheat the principal. Of the many kinds of cheating the one we will focus on here is the idea that the agent could use the principal’s advance to go into business and trade on her own account. If that paid better than fulfilling the plan, then cheating was the agent’s best choice. Cheating was bad for the principal, however, not only in the obvious sense that it might leave the principal materially worse off. Even if the agent could find a way of gaining while leaving the principal no worse off and none the wiser, cheating could still damage the principal by shifting the distribution of power and authority in the economy gradually in favor of the agent. Taken to its logical conclusion, it clearly had the potential to subvert the command system altogether and lead to its replacement by a market economy. To illustrate market information, consider Figure 9.3. Agents 1 and 2 report to their principal, say, the minister for atomic weapons. Agent 1 supplies fissile materials such as uranium and Agent 2 is the bomb builder. The basic structure of the command system that we outlined in Chapter 3 requires that the minister orders the materials supplier (1) to advance uranium to the bomb builder (2), who returns the finished device to the minister. In theory everything goes according to plan. Market information is exchanged only vertically via the principal. The principal sets the requirements and funding of Agent 1,

242

Mark Harrison

Figure 9.3. Principal, Agents, and an Outsider

and knows the capabilities and costs of Agent 2. There is no need for Agents 1 and 2 to have any horizontal contact with each other, apart from the moment when the materials change hands. The market information adds value to the transaction but only the principal, who holds all the information, can extract the added value. One of the standard criticisms of the planned economy is that this is a very costly and imperfect way of using market information; moreover, the agents have no incentive to disclose it because they do not gain from sharing it. Because they do not share their information, there will be mistakes and gaps in the inside principal’s knowledge of each of his own agents taken separately, and he is likely to make more mistakes in his allocations than if his agents were left to agree between themselves on a voluntary market transaction. These mistakes will reduce the total of value that is added. If the principal were altruistic, the efficiency loss should be sufficient to induce him to give up his coordination role and leave coordination to market interaction among the agents. But, although this would lead to fewer mistakes and more value, the principal would not only be unable to extract any of the gain, which would accrue to the agents, but would also lose the ability to extract any surplus at all. In other words, the power of a selfish principal depended on preventing a market from arising among the agents below him. It is one more step to suppose that this power rested, in part, on stopping market information being exchanged at lower levels. Again, consider the agents in Figure 9.3, the supplier of fissile materials and the bomb producer. There is scope for one or both of these to do a little market research and start trading on their own account, either with each other or with an outsider. Plenty of people are potentially interested in privately acquiring

Secrecy

243

fissile materials. What’s to stop Agent 2 delivering a little uranium on the side to a research physicist short of materials for experimentation, the agent of a foreign power interested in atomic blackmail, or a criminal gang with plans for organizing the illicit trade? There are many more mundane examples, too, from gun-running to the theft of valuable materials for relatively trivial uses. From Soviet times, the author recalls seeing hand-crafted bottle openers on sale in a Moscow market, made out of titanium diverted from a weapons factory. For serious business, the agent had to be ready to exchange market information with outsiders. She needed to undertake some market research and also advertise her goods. The better her outside market knowledge, and the better the market’s knowledge of her, the more likely she was to find a deal that improved on the inside wage offered by her principal. She also had to build trust with outsiders. As Avner Greif (2000) has emphasized, market transactions that depend on a sequence in which one side has to move first must overcome the fear of being cheated. Where the law is ineffective, and crime is unorganized, illegal transactions rely on trust, or ‘‘honor among thieves.’’ Thieves build mutual trust partly by the prior exchange of verifiable information: ‘‘How do I know you have the goods?’’ ‘‘Let me see the money.’’ Just as she is proposing to break trust with the principal so as to steal from him, the agent has to trust the outsiders in the illegal market not to steal from her. And they have to trust her not to betray them to her principal. This is where official secrecy helped the Soviet principal: it reduced the agent’s expected gain from betraying the principal and going into private business by a discounting effect and a threat effect. The discounting effect arose because secrecy made information unverifiable and so put a discount on its value; this is bad for horizontal trust. To continue as an unofficial trader the agent must work harder to offset the discounting effect and take more risks to establish her credibility: ‘‘Look, I could get ten years for this too.’’ The threat effect arose because when the sharing of secret information was detected, both sender and receiver risked punishment. The threat was directly costly to the agent. Indirectly, it also damaged the outside network in a way that would reduce profit opportunities. Outside opportunities were likely to vary directly with the scope of private networks in the same way as economic efficiency generally increases with market size. But as more people were involved the risk of detection increased. Thus criminalizing the sharing of information was likely to reduce the scope of criminal organization: ‘‘We don’t need them; they don’t need to know.’’ In turn, narrower networks implied lower outside profits. Evidently, secrecy was not a sufficient condition for eliminating illicit trade

244

Mark Harrison

and corruption. In the defense industry in the late 1930s some suppliers of non-combat equipment sold it more profitably on the side than to the Red Army (Harrison and Simonov 2000: 236). A maker of gun accessories was caught selling off capital equipment and undeclared stocks of finished goods rather than continuing to supply the military (Markevich and Harrison 2004). But it is hard to know whether this was typical.

Strategic Information SPYING AND BARGAINING

In contrast to market information, strategic information is valuable because sharing it makes one side better off at the expense of the other. Market information creates a two-sided gain: when it is shared, both sides make a profit. In the case of strategic information there is a one-sided advantage that changes hands with the information; as a result, one side wants to get it and the other side wants to keep it secret. Strategic information is valuable in wars, races, blackmail, and bargaining. Examples include military plans and deployments; technological secrets such as the recipe for Coca-Cola or how to make the atomic bomb; information that can damage reputation like love letters or evidence of the use of slave labor; and financial information that a labor union can exploit to undermine the negotiating position of an employer. The Soviet Union concealed the true level of military outlays, for example, from 1930 to 1933 because the truth would have undermined its negotiating position at the Geneva disarmament talks (Davies 1993). The pretense continued in 1934 and 1935 because, although the original motivation had disappeared, to close the gap between truth and reality too quickly would reveal the lie and undermine the future credibility of all such Soviet claims (Davies and Harrison 1997). Strategic information could be traded in two distinct contexts, one obvious, the other less so. The obvious one is espionage. In terms of Figure 9.3, the outsider is a foreign power or its intermediary who is willing to reward Agent 2 in return for information about the principal’s capabilities or intentions; from the point of view of the principal, Agent 2 knows ‘‘too much.’’ What is worse, Agent 2 can share this information without any overt sign of disobedience, because nothing is physically lost from the principal’s advance or what the agent does with it. Here official secrecy is not only the obvious solution but the only solution. Again it helps the principal through threat and discounting effects. The discounting effect lowers the credibility of the information that the agent has to offer. For example, most observers regarded Soviet budgetary series for mili-

Secrecy

245

tary spending from the mid-1960s through the end of the 1980s as a blatant lie, and a number of alternative estimates were prepared including some that relied on intelligence sources, but because the truth could never be verified and the alternative estimates were often wildly divergent doubt was cast on all figures, including those based on intelligence that could easily be interpreted as disinformation or deliberate lies designed to mislead (Harrison 2003). The discounting effect could not solve the problem on its own, however, for two reasons. A secret was likely to retain some value even when it could not be verified. More worryingly, the very fact that information was made secret could give it a cachet of strategic value. We could think of this as establishing a rationale for the principal to do some fairly obvious things. One was to impose secrecy indiscriminately and classify a lot of non-strategic or trivial information as well, to mask the few bits of information that really mattered, while claiming in public that ‘‘there are no such things as trifles.’’ Another was to engage in the spreading of lies, or disinformation; both actions would damage the market for secondhand information that was true and valuable.∑ Secrecy and lies would damage the market by driving down the customer’s willingness to pay for information that had not been and probably could not be verified, but as long as there was some market value to a secret these stratagems alone would not prevent the trade altogether. It was important, therefore, to create legal penalties as an additional barrier to entry. Thus, the risk of punishment remained essential to the enforcement of secrecy. There is another way in which strategic information could be valuable, less obvious than espionage but probably much more commonplace. Now suppose that the outsider is the authorized customer of Agent 2 (see Figure 9.3). In terms of the subject matter of Chapter 6, we could think of the principal and his agents as Industry, with the Army as the outsider. The Army and Industry had an adversarial relationship insofar as Industry wished to extract the Army’s budget for the procurement of weapons with as little effort as possible. Industry could shift the ratio of effort to rubles to its own advantage by pushing up the prices of its products relative to their quality and the effort embodied in them. In principle weapons were priced on the basis of unit variable costs plus a markup to cover capital charges. Costs reflected both the quality of materials and the amount of labor effort used in production. In this context, production costs became strategic information, and official secrecy promoted the cheating of the buyer through covering up cost inflation and price gouging. More generally, we find self-interested agents throughout Industry exploiting the cloak of secrecy to deny strategic information to a variety of outsiders including the Army and the party; the latter, representing the encompassing interests of the dictator, usually sided with the Army.

246

Mark Harrison INDUSTRY AND THE ARMY

The defense industry provides a number of cases in which managers and officials exploited secrecy in their own self-interest. Industrial managers enthusiastically classified cost statistics as important military secrets so as to withhold them from the defense ministry’s purchasing agents. The Army was deprived of verifiable information about costs while Industry was freed from adherence to cost-plus pricing rules for defense products, or won time to inflate costs so as to justify the price set. We have several examples from the 1930s. A KPK audit commission report of September 1935 slated defense enterprises for excluding military agents from information about production costs and exploiting secrecy to engage in wholesale ‘‘deception’’; similar behavior caught the attention of the finance minister two years later (Harrison and Simonov 2000: 235). Let’s be clear, in case the reader finds it too bizarre: the position of Industry was that the Army’s representatives could not be allowed to know the production costs of the weapons they were buying on the grounds that this information was an important military secret. What is more, in the context of its time and place this apparently absurd stratagem was effective. In 1938 the defense ministry succeeded in getting the ministry for the defense industry to agree that its factories would disclose pricing calculations to the military agents. Locally, however, this agreement was subject to widespread sabotage on the side of the factories. In a letter of March 29, 1938, to defense industry minister Kaganovich, officials of the Red Army artillery administration reported that ‘‘the obstacles to normal calculation and the proper estimation of actual costs of artillery administration orders’’ had not been overcome. They complained: Despite frequent appeals to the planning and finance administrations of the defense industry ministry nothing has been put into effect up to now. The finance administration of the defense industry has not implemented the direct instruction of your deputy B. L. Vannikov to provide the artillery administration with the calculations. Locally the practice is continuing of the factories’ holding back the calculation work of the military agents. Just in the last few days the military agent at factory no. 12 has informed us that the factory is refusing to supply calculations ex post, referring to your decree no. 54 of February 9 this year. Such a refusal is a direct violation of the government decision no. 108ss of September 3, 1937, by which the defense ministry is entitled to receive annual calculations ex post.

It was necessary, they continued: to make arrangements to regularize mutual relations with the artillery administration in the sense of providing it with full opportunity to do calculation

Secrecy

247

work and in all cases to obligate factories to provide the artillery administration with calculations ex post for orders covering 1937 in fulfillment of the government’s decision. All this is especially necessary taking into account that the defense industry ministry and defense ministry will shortly be working together to set prices for 1938. Correct decisions will only be reached under conditions of the artillery administration’s most detailed familiarization with the production costs of the goods to be ordered and joint business preparation.∏

In other words, the Army’s agents needed to know about Industry’s costs in order to carry out their statutory responsibilities, while Industry was able to keep this information back by means of nothing more complicated than simple foot-dragging. The military agents encountered similar problems in attempting to carry out their obligation ‘‘to oversee the condition of mobilization planning’’; the following account is based on work by Markevich and Harrison (2004). In 1937 the government Defense Committee issued a special resolution that included a stipulation of military agents’ right of access to enterprise mobilization plans. The defense industry ministry, however, ‘‘forgot’’ to include this in the decree that it issued to implement the Defense Committee resolution. The decree limited the prerogatives of military agents to ‘‘the right to participate in working out and auditing the provisioning . . . of enterprises, the right to check the factual correspondence of technological processes with working drawings and technical specifications, and the provision of technology with equipment . . . and so forth.’’ Enterprises then cited the decree in refusing the military agents’ access to mobilization planning. The Red Army artillery administration appealed to the defense industry ministry on this issue three times: in February, March, and April 1938. The defense industry ministry based its refusal on the need to ensure the secrecy of mobilization assignments; in August, however, it agreed grudgingly to open up enterprise mobilization planning to the military agents subject to a rigmarole of special procedures and permissions. The underlying situation was that the defense industry ministry was exploiting secrecy to cover up the lamentable state of mobilization planning at the enterprise level. Thus, on receiving the April 1938 demand from the Red Army artillery administration to give their military agents the right to see the defense factories’ mobilization plans, a ministry official, evidently the minister or one of his deputies, wrote on the letter: ‘‘After approval of the mobilization plans’’ (emphasis added). As in production, so in research and development, military designers were able to exploit secrecy to keep their ideas to themselves. In 1938, for example, specialists from a naval research institute in Leningrad wanted to evaluate the work that the designer Vladimir Uvarov was doing on gas turbines in Moscow

248

Mark Harrison

(for an overview of this work and its military and economic significance, see Chapter 8). They wrote to the defense industry ministry complaining of obstruction: Uvarov would not let them into his bureau. His obstruction, they suggested, was motivated by personal animosity ‘‘and in addition ostensibly special instructions about the secrecy of the work.’’ On the side of Industry, however, the ministry officials stood up for their ‘‘own’’ designer and refused curtly to intervene.π The defense industry also tried to keep the plan and budget authorities continually starved of information about its activities on the grounds that they constituted important military secrets. The central government took repeated steps to counter this tendency and enforce the supply of information. A Politburo resolution of January 1932, for example, required that defense industry production should be included in the calculated totals for industry as a whole. In January 1935 deputy minister for heavy industry Iurii Piatakov proposed to prime minister Molotov that defense industry should no longer have to report its progress to the finance ministry or Gosplan’s statistical administration on grounds of national security. In March, following a counter-claim from the statistical office, the government made limited concessions to Piatakov but still required defense industry to report both real outcomes and ruble aggregates to Gosplan in Moscow, real outcomes for civilian products only to local statistical agencies, and ruble aggregates to the ministry of finance (Simonov 1996: 1362, 1364n; Markevich 2000). INDUSTRY, THE PARTY, AND ‘‘PUBLIC’’ OPINION

Above both Army and Industry stood the dictator Stalin, who used his party to enforce his interests. In this context it was Industry, not the Army, that was behaving badly and needed to be brought under control. In Chapter 4 Andrei Markevich quoted the opinion of Gosplan chief Mezhlauk in July 1933 that secrecy explained the ‘‘backwardness of defense industry in methods of work, standards of consumption of materials, cutting costs, and so forth’’ because secrecy freed the industry from press criticism. In fact this problem emerged simultaneously with the creation of the command system. In March 1930, less than one year after adoption of the first five-year plan, defense minister Voroshilov wrote: ‘‘In its most recent two decrees on defense industry (no. 3 of July 15, 1929, and no. 3 of February 25, 1930) the Politburo has indicated the excessive reclassification (zasekrechivanie) of production at military and aviation factories as secret, in consequence of which non-party and party activists have been effectively excluded from taking part in the organization and rationalization of production.’’ He called on the government audit commission to follow the matter up urgently so as to put a stop to

Secrecy

249

excessive secrecy ‘‘under the sign of which bunglers and others skilled in bad work conceal their ‘artwork.’ ’’∫ At a lower level of the apparatus the idea that managers and workers could collude to exploit secrecy in an opportunistic way also occurred to Andrei Zhdanov in Leningrad. During 1936 the latter met several times with party activists from the local defense factories to discuss plan discipline. He linked poor performance in the tank industry ‘‘with the fact that the factories prefer to pursue their own interests . . . considering that civilian production may be reported in the press, but they’ll get around to military production sometime later.’’Ω At a subsequent meeting he remarked acidly of failures in the artillery program: ‘‘Our misfortune is that we can’t drag you outside; if they wrote about these customs in the press the public opinion of our country would eat you alive, to put it mildly. . . . We can’t do this unfortunately and we appeal to your [sense of ] party duty.’’ A few minutes later Zhdanov noted that public opinion can reward as well as punish. ‘‘It’s easy,’’ he remarked, ‘‘to get a medal for civilian production and make a stir in the papers, that is, it’s easy to be good. But in the military industry, since there’s no public regulation [obshchestvennogo kontrolia net] and they publish nothing in the papers, this is a business that can be shelved indefinitely.’’∞≠ Later that year, Zhdanov again commented on the inferior fulfillment of plans for the defense industry compared with those of the civilian economy, despite the former’s supposedly higher priority. Once more he ascribed this to secrecy: ‘‘They never notice you in the press on account of secrecy, do they? And this contributes to a certain degree of freedom from accountability [beskontrol’nost’] and this demands an effective way of working that compensates for the absence of public criticism with robust and very effective organization from below.’’∞∞ Before we leave Zhdanov it is worth considering what he tells us about the role of the press and ‘‘public’’ opinion in the Stalinist system of power. The press was controlled by the party, and public opinion was no more than the opinion of the party expressed in public; what could the party media do in public to discipline industry that the party itself could not achieve in private? Zhdanov looked to the press, evidently, to deliver public abuse and humiliation in return for failure, and social honor in return for success, and he saw this as a uniquely powerful mechanism for aligning private incentives with the interests of the party. Secrecy neutralized this mechanism, however, leaving only appeals to sense of duty backed up by vague threats. Possibly, this tells us as much about Zhdanov’s intellectual formation as it tells us about the empirical behavior of Soviet defense industry managers. It is of interest, therefore, to find his interpretation, and that of Voroshilov, fully

250

Mark Harrison

confirmed by evidence from other sources a full half century later. Citing press reports from the spring of 1988, the time of Gorbachev and perestroika, Julian Cooper (1990: 188) remarked, Critics of poor-quality civilian goods manufactures at defense-industry enterprises now openly express their frustration at the way in which secrecy is used to obstruct the investigation and exposure of shortcomings. One author, discussing the fire hazard presented by low-quality televisions, notes that some of the producers ‘‘have hidden themselves in zones literally closed to criticism. State ‘secrets’ are invoked. Even people’s control . . . has difficulty breaking through into the ‘boxes.’ ’’ Another author, in a remarkably outspoken article entitled ‘‘On ‘boxes,’ open secrets, and departmental interests,’’ berates the aviation industry for its use of secrecy to protect its own interests and shows the absurdity of some of the security measures adopted.

(In this quotation ‘‘people’s control’’ refers to the state and party auditors, and the ‘‘boxes’’ are the mailbox numbers that continued to conceal the locations of the secret defense factories.) We have looked at the secretiveness of the Soviet state from various angles, using the defense industry as an example. We have tried to explain blanket secrecy, elaborate internal firewalls, fierce punishments, and a curious mix of laxity and negligence in some respects combined with overzealousness in others. We have made only a first pass at the subject, and much more is still to be learned. A comparison with secrecy in purely civilian branches of the economy could perhaps throw even more light on this murky topic. It seems clear, nonetheless, that secrecy had more than one dimension. As regards the secrecy of market information the interests of principal and agent were generally opposed: the principal used secrecy to limit the agent’s options, whereas the agent was indifferent to secrecy or preferred openness. In the case of strategic information the principal used secrecy to limit an outsider’s options, and this created a motivation for the agent to sell it without authority. Such cases encouraged blanket secrecy combined with powerful threats to induce the agent to stick to the rules of secrecy. This may help explain the severity of punishments for violating Soviet state secrets, which continued long after Stalin’s death. There was another dimension to secrecy, and this contributed to the sturdiness of the many firewalls that prevented information from spreading from one bureaucracy to another. It involved situations in which the agent could exploit secrecy to conceal strategic information and so extract something from an authorized outsider. For example, Industry’s agents, the managers of defense plants, exploited secrecy to conceal and exaggerate their true costs and

Secrecy

251

extract a surplus from the Army. In such cases the Army could object and the party could disapprove, but their right to know was fatally weakened because the agent’s superiors were unwilling to enforce it, colluded with the concealment in practice, and can be thought of as having gained from it. This proposition is supported by the evidence. In the cases that have come to light in which the agent used secrecy in this opportunistic way it was never the principal that blew the whistle. Some outside agency always exposed the facts: usually the customer, the planner, the auditor, or, in later years, the press. It is easy to work out why. As long as the agent exploited secrecy at someone else’s expense, her principal did not object. The cloak of secrecy enabled the agent, as seller, to extract something from an outsider, keep some of it as reduced effort, and pass the rest upward to her boss as a show of increased compliance with plans and contracts for fewer resources used. In return, the boss defended his agent. Vertical loyalty was thus a two-way street. The 64,000,000-ruble question remains: why, exactly, was the Soviet state of the twentieth century so excessively secretive, so much more extremely secretive than its Russian predecessor, or its successor, or the states of most competing market economies? This chapter has shown that we do not need to attribute the excesses and extremes to individual folly or miscalculation. The behavior of the great Stalin and all the little Stalins under him is largely understandable in terms of the rational principle: I am entitled to privacy, You are defensive, He or she has something to hide.

What was different about Soviet bureaucrats, and what made their secretiveness take on such fantastical forms and dimensions, was the lack of limits that would have been imposed on their behavior by political or market competition. They monopolized both political and economic power. This deprived them of the incentives that drive others to limit secrecy and prefer varying degrees of openness. Rivalry, whether competition between firms for market shares, or rivalry among the political parties contending for government, limits secrecy for two reasons. First, when information is valuable the ability of competing rivals to create incentives for disclosure is correspondingly large, so that important secrets are quickly spilled into the public arena; also, political and business leaders have difficulty in then imposing large penalties on their own employees or agents for disclosing secrets because public opinion will punish them if they do. Second, those who allow the costs of upholding a regime of secrecy to get out of hand will suffer a competitive disadvantage because of their higher

252

Mark Harrison

costs, and this is likely to end in driving them from the market; self-interest makes them ready to limit the uses of secrecy. These limits to secrecy form, in turn, one of the most important advantages of a society based on economic and political competition.

Notes I thank Paul Gregory for guidance; Leonid Borodkin, Andrei Markevich, and Andrei Sokolov for assistance; Stephen Broadberry, Michael Ellman, Melanie Iliè, Peter Law, Oleg Khlevniuk, and Wolfram Schrettl for conversations and comments; and the staff of the Hoover Institution Archive, Stanford, California, for access to documents. 1. Hoover/RGASPI, 77/1/826, 67 (December 12, 1936). 2. Hoover/GARF, R414/4/193, 296. This passage is from an article titled ‘‘Strictly Observe Party and State Secrets’’ in the August 18, 1953, issue of the daily newspaper of the Noril’sk combine, discussed further below. 3. Hoover/GARF, R414/4/193, 296. 4. For the KPK investigation, Hoover/RGANI, 6/1/35: 10, 15, 51, 52. For the followup measures, Hoover/RGANI, 6/1/41: 71–72. Harsh penalties sometimes led to overcompensation; the director of the military chemical industry’s factory no. 6, for example, had his juniors kidnapped to test their courage and see how they would behave ‘‘in the event of arrest by adventurists’’ (Hoover/RGANI, 6/1/91, 16, 19). On the treatment of documents in the interior ministry, see Harrison (2004). 5. The market for information is damaged in a similar way as Akerlof (1970) shows the offer of ‘‘lemons’’ (low-quality used cars) to damage the market for secondhand automobiles. See Chapter 8, note 2. There are some differences between secondhand automobiles and strategic information, of course. One difference is that information may be copied and shared relatively freely in contrast to automobiles, which have to be produced using capital, labor, and materials. The main cost of trading strategic information was fear of the consequences of being found out. 6. RGAE, 7515/1/403: 303 (March 29, 1938). 7. RGAE, 8328/1/995: 5–6 (May 3 and 15, 1938). 8. Hoover/RGASPI, 85/27/86: 1–2 (March 3, 1930). 9. Hoover/RGASPI, 77/1/826: 12 (March 23, 1936). 10. Hoover/RGASPI, 77/1/826: 19–20 (May 5, 1936). 11. Hoover/RGASPI, 77/1/826: 28 (December 23, 1936).

Published References Akerlof, George A. 1970. ‘‘The Market for ‘Lemons’: Quality, Uncertainty, and the Market Mechanism.’’ Quarterly Journal of Economics 84(3): 488–500. Applebaum, Anne. 2003. Gulag: A History of the Soviet Concentration Camps. London: Allen Lane. Barber, John, Mark Harrison, Nikolai Simonov, and Boris Starkov. 2000. ‘‘The Structure

Secrecy

253

and Development of the Soviet Defence-Industry Complex.’’ In The Soviet DefenceIndustry Complex from Stalin to Khrushchev, ed. John Barber and Mark Harrison, 3– 32. London and Basingstoke: Macmillan. Bergson, Abram. 1953. ‘‘Reliability and Usability of Soviet Statistics: a Summary Appraisal.’’ American Statistician 7(3): 13–16. Bone, Jonathan. 1999. ‘‘Soviet Controls on the Circulation of Information in the 1920s and 1930s.’’ Cahiers du Monde russe 40(1–2): 65–90. Colby, William E. 1976. ‘‘Intelligence Secrecy and Security in a Free Society.’’ International Security 1(2): 3–14. Conquest, Robert. 1971. The Great Terror. Revised ed. Harmondsworth: Pelican. Cooper, Julian. 1990. ‘‘The Defense Industry and Civil-Military Relations.’’ In Soldiers and the Soviet State: Civil-Military Relations from Brezhnev to Gorbachev, ed. Timothy J. Colton and Thane Gustafson, 164–91. Princeton, NJ: Princeton University Press. Cooper, Julian. 1999. ‘‘Introduction.’’ In The Numbered Factories and Other Establishments of the Soviet Defence Industry, 1928 to 1967: A Guide, Part I. Factories and Shipyards: iv–viii. Compiled by Julian Cooper, Keith Dexter, and Mark Harrison. University of Birmingham, Centre for Russian and East European Studies, Soviet Industrialisation Project Series. Occasional Paper no. 2. Available from http://www.war wick.ac.uk/go/vpk. Davies, R. W. 1993. ‘‘Soviet Military Expenditure and the Armaments Industry, 1929– 33: A Reconsideration.’’ Europe-Asia Studies 45(4): 577–608. Davies, R. W. 2001. ‘‘The Making of Economic Policy.’’ In Behind the Façade of Stalin’s Command Economy, ed. Paul R. Gregory, 61–80. Stanford, CA: Hoover Institution. Davies, R. W., and Mark Harrison. 1997. ‘‘The Soviet Military-Economic Effort Under the Second Five-Year Plan, 1933–1937.’’ Europe-Asia Studies 49(3): 369–406. Firth, Noel E., and James H. Noren. 1998. Soviet Defense Spending: A History of CIA Estimates, 1950–1990. College Station: Texas A&M University Press. Gorbachev, Mikhail. 1996. Memoirs. London: Doubleday. Goriaeva, T. M. 2002. Politicheskaia tsenzura v SSSR. 1917–1991. Moscow: Rosspen. Gorlizki, Yoram. 2002. ‘‘Ordinary Stalinism: The Council of Ministers and the Soviet Neopatrimonial State, 1946–1953.’’ Journal of Modern History 74(4): 699–736. Greif, Avner. 2000. ‘‘The Fundamental Problem of Exchange: A Research Agenda in Historical Institutional Analysis.’’ European Review of Economic History 4: 251–84. Harrison, Mark. 2003. ‘‘How Much Did the Soviets Really Spend on Defence? New Evidence From the Close of the Brezhnev Era.’’ PERSA Working Paper no. 24. University of Warwick, Department of Economics. Available from http://www.warwick.ac .uk/go/persa. Harrison, Mark. 2004. ‘‘Why Secrets? The Uses of Secrecy in Stalin’s Command Economy.’’ PERSA Working Paper no. 34. University of Warwick, Department of Economics. Available from http://www.warwick.ac.uk/go/persa. Harrison, Mark. 2005. ‘‘The Fundamental Problem of Command: Plan and Compliance in a Partially Centralised Economy.’’ Comparative Economic Studies 47(2): 296–314. Harrison, Mark, and Nikolai Simonov. 2000. ‘‘Voenpriemka: Prices, Costs, and Quality in Defence Industry.’’ In The Soviet Defence Industry Complex from Stalin to Khrushchev, ed. Mark Harrison and John Barber, 223–45. London and Basingstoke: Macmillan.

254

Mark Harrison

Khanin, G. I., and V. Seliunin. 1987. ‘‘Lukavaia tsifra.’’ Novyi mir, no. 2: 181–201. Iliˇc, Melanie. 2000. ‘‘The Great Terror in Leningrad: A Quantitative Analysis.’’ EuropeAsia Studies 52(8): 1515–34. Khlevniuk, O. V., I. [Yoram] Gorlitskii [Gorlizki], L. P. Kosheleva, A. I. Miniuk, M. Iu. Prozumenshchikov, L. A. Rogovaia, and S. V. Somonova, eds. 2002. Politbiuro TsK VPK(b) i Sovet Ministrov SSSR. 1945–1953. Moscow: Rosspen. Markevich, Andrei. 2000. ‘‘Otraslevye narkomaty i glavki v sovetskoi ekonomike 30-ykh gg. (na primere NKTP i GUMPa).’’ Moscow: Russian Academy of Sciences Institute of Russian History. Markevich, Andrei, and Mark Harrison. 2004. ‘‘Quality, Experience, and Monopoly: Regulating the Soviet Seller’s Market for Military Goods.’’ PERSA Working Paper no. 35, University of Warwick, Department of Economics. Available from http://www.war wick.ac.uk/go/persa. Masliukov, Iu. D., and E. S. Glubokov. 1999. ‘‘Planirovanie i finansirovanie voennoi promyshlennosti v SSSR.’’ In Sovetskaia voennaia moshch’ ot Stalina do Gorbacheva, ed. A. V. Minaev, 82–129. Moscow: Voennyi Parad. Medvedev, Zhores A. 1978. Soviet Science. New York: W. W. Norton. Mikoian, A. I. 1956. Speech by A. I. Mikoyan at 20th Congress of the Communist Party of the Soviet Union, February 16, 1956. London: Soviet News. Moynihan, Daniel Patrick. 1997. ‘‘Secrecy as Government Regulation.’’ Political Science and Politics 30(2): 160–65. Park, Robert L. 2000. Voodoo Science: The Road from Foolishness to Fraud. Oxford: Oxford University Press. Schelling, Thomas C. 1963. The Strategy of Conflict. Oxford: Oxford University Press. Seliunin, Vasilii, and Grigorii Khanin. 1987. Lukavaia tsifra. Novyi mir 1987(2): 181– 201. Simonov, N. S. 1996. ‘‘ ‘Strengthen the Defence of the Land of the Soviets’: The 1927 ‘War Alarm’ and Its Consequences.’’ Europe-Asia Studies 48(8): 1355–64. Solzhenitsyn, Alexander. 1974. The Gulag Archipelago, 1918–1956, vol. 1. London: Collins/Fontana. Tarschys, Daniel. 1985. ‘‘Secret Institutions in Russian Government: A Note on Rosenfeldt’s Knowledge and Power.’’ Soviet Studies 37(4): 524–34. Weber, Max. 1922/1968. Economy and Society: An Outline of Interpretative Sociology, ed. Guenther Roth and Claus Wittich. 3 vols. New York: Bedminster.

Afterword mark harrison

In this book we have looked at some issues of Soviet defense and the defense industry under Stalin, using simple tools of historical economics and political economy. What have we learned? Defense issues were at the core of Stalin’s dictatorship (Chapters 1 and 2). We cannot understand how Stalin’s dictatorship came into being, and how he ruled, unless we take into account the active role played by military and military-industrial interests as well as the importance of the armed forces in Stalin’s system of political power. Stalin wanted military power to defend himself against external threats. At the same time, the Soviet military authorities supported his program of forced industrialization and transition to a command system. Once he had become dictator, however, Stalin did not want, and came to fear, strong military and industrial leaders with the independence to stand up to him or the potential to join against him (Chapter 3). In Stalin’s time the last of these were the heavy industry minister Sergo Ordzhonikidze, whom Stalin drove to suicide in 1936, and marshal of the Red Army Mikhail Tukhachevskii, executed in 1937. World War II threw up new leaders of the military and economic fronts; afterward, Stalin humiliated, imprisoned, or executed a number of them. Stalin was able to rule, therefore, without having to concede anything to the claims of a powerful, self-interested military-industrial lobby. 255

256

Afterword

The Soviet defense sector, although organized, was not monolithic (Chapters 3, 4, 5, and 6). In theory everyone in the defense sector might have gained by cooperating to share the proceeds of privilege. In practice there were vertical and horizontal fault lines everywhere that kept them at odds with each other and limited cooperation. Probably this was not an accident. It was Stalin’s system of rule to keep the Army and Industry divided from each other, and to structure the incentives facing them so that they gained more from mutual rivalry, including shirking mutual obligations and cheating each other, than from setting out to cooperate. They had to cooperate at some point, of course; when cooperation was achieved, it usually emerged from conflict rather than arising spontaneously. We cannot understand how the defense sector was managed without taking this into account. The defense sector was privileged but not self-contained (Chapters 3, 4, 5, and 7). Its rapid growth and continually, unpredictably changing requirements prevented it from detaching itself from the Soviet economy. Qualitative change in weaponry required the continual militarization of formerly civilian technologies. Quantitative expansion continually sucked equipment and workers from the civilian sector into military industry. As a result there were not two Soviet economies, one military and the other civilian. The people in the defense sector were rooted in Soviet society and remained part of it. The strengths and weaknesses of this society were also the strengths and weaknesses of the defense sector. The context of the defense industry was a command system, but in the command system disobedience and disloyalty were normal and were controlled rather than prevented (Chapters 1, 2, and 3). Everyone tried to influence, reinterpret, or even ignore the commands that they received in their own interests. This has an important implication: we can transfer the tools of modern economic analysis, developed to analyze market behavior, across to the Soviet institutional setting. We have studied several kinds of outwardly dysfunctional institutions and irrational behaviors in the Soviet defense industry. These may seem at first glance to have been specific to Russian culture, Soviet institutions, or Stalinist ideology. On closer examination, borrowing from the economics of contracts and information, we can understand them more simply as the calculated actions of rational, self-interested agents in the situations in which they found themselves (Chapters 6, 8, and 9). Taking our study as a whole, what comes across most strongly is that Stalin’s management of the defense economy was both single-minded and remarkably effective. It is true that the experiences of the others involved, ministers, soldiers, and workers, were chaotic, fraught with tensions, and often corrupting or demoralizing. But the outcomes measured by the weapons pro-

Afterword

257

duced and the military divisions equipped and ready for battle were sufficient to beat off external threats and maintain his power. The extent to which Stalin was able to secure these outcomes is one measure of his efficiency as a dictator. The detailed results of our book can also throw light on wider issues that we have not addressed directly; further research is required to explore them more fully. One wider issue is whether the Soviet defense sector used its resources more effectively than other branches of the Soviet economy. At the present it is still not clear whether the defense industry was more efficient than other Soviet industries, or merely better supplied. Stalin’s agents were rarely motivated to cut costs, but as yet we cannot say whether they were less or more careless of resources in the defense industry than in other branches. Another wider issue is whether the Soviet defense industry used its resources more effectively than the defense sectors of the market economies of comparable size and military strength in Germany, Britain, or the United States, for example. We have noted that most countries face significant problems in organizing military production because defense markets tend to be highly concentrated and it is all too easy for business interests to use political action to create market opportunities, push up prices, and cheat the taxpayer. The Soviet defense industry did not have to work very well in any absolute sense to do comparatively well. But whether it did better than the defense industries of other countries has not been determined yet. Finally, we may ask whether Stalin deserves some credit for the way he managed Soviet defense issues and the defense industry. This may seem an odd question to ask. Stalin was a bloody dictator who directly ordered the deaths of around one million of his own subjects; his internal policies probably led directly or indirectly, sometimes predictably, but not intentionally, and therefore with a lower degree of culpability, to the deaths of around 10 million more.∞ These terrifying figures do not include the many lives that were additionally lost because Stalin entered World War II without the officers he had slaughtered in 1937, or because of his own disastrous military and diplomatic miscalculations and mismanagement on the eve of the war and during it, although it should be remembered that Stalin was not the only one to make mistakes; democratic leaders like Prime Minister Churchill and President Roosevelt could also make costly strategic errors. World War II is rightly at the center of our focus when we look back on Stalin’s effectiveness as a defense manager and military leader. It encapsulates all the contradictions of his time. Stalin’s sole claim to our gratitude arises because it was his good fortune to rule when he did. He was cruel enough, but the twentieth century threw up someone worse: measured by the number of

258

Afterword

his deliberately intended victims, Hitler was even more vicious than Stalin and so had the effect of making Stalin look good by comparison. Stephen Wheatcroft (1994), for example, has compared the numbers of victims of Stalin’s repressive policies with those of Hitler’s. The figures indicate that Hitler and Stalin were each responsible for many millions of deaths, but their level of culpability differs. Most of Hitler’s victims were intended and he deliberately murdered at least 5 million people. In contrast, most of Stalin’s victims died from neglect, indifference, or the unintended consequences of his policies; the number of his premeditated murders is just a million or so. In most societies morality and law condemn both murder and killing without premeditation, but treat murder as the worse of the two. They also treat two murders as worse than one. Are 5 million murders worse than one million? It is a difficult question; on a logarithmic basis 5 million exceeds one million by less than one whole order of magnitude. If Hitler was truly worse than Stalin, we might evaluate some of Stalin’s measures more positively than we would have done in Hitler’s absence. Briefly, a German victory over the Soviet Union in World War II would have given Hitler domination over the whole of continental Europe; it would most likely have forced Britain out of the war and prevented America from coming to the aid of European civilization. Hitler should have had an easy victory over Stalin. The Soviet Union was a relatively poor country, and the history of two world wars shows clearly what happens when rich countries meet poor ones in total war: the poor country is defeated either immediately or gradually and, in the latter case, its economy collapses. At least, this was the case for market economies in both world wars. Stalin’s command economy is the sole exception to this rule (Harrison 1998a, 2005; Broadberry and Harrison 2005). The Soviet economy proved unexpectedly resilient; it was able to absorb the shock of massive invasion and devastating territorial loss, followed by mobilization on a scale similar to or greater than that of the much more highly developed market economies of Germany, Britain, and the United States (Barber and Harrison 1991, 2006; Harrison 1998b). This resilience owes much to Stalin’s military-economic policies, which left the Soviet defense industry sufficiently well organized and supplied to equip the Red Army and help destroy Hitler’s forces in the east. (Surprisingly, it also owes something to his policy of farm collectivization in the early 1930s; this policy was catastrophic from the perspective of peacetime economic welfare, but in wartime it stopped the Soviet economy from disintegrating by ensuring the priority of Red Army soldiers and war workers over peasants in the wartime allocation of food.) As a result of these policies, the Soviet economy did not collapse as Russia’s had done in World War I.

Afterword

259

In the long term the Soviet wartime resistance to Hitler probably saved tens of millions of Russian, American, and European lives. Those alive today as a result have no need to feel grateful to Stalin for organizing Soviet resistance because Stalin did it self-interestedly, not as a favor to others. But it would also be hypocritical to condemn the resistance, to which Stalin made an indispensable contribution. To conclude, this book has investigated a core aspect of Stalin’s legacy. Stalin’s legacy was more lasting than Hitler’s, and morally more complex. It seems reasonably certain, however, that, if Hitler had somehow failed to come to power in Germany or had been prevented from launching his drive to the east, no one would wish to remember Stalin today as anything more than a murderous dictator who squandered his people’s rubles on guns. Notes 1. For the one million, see figures under ‘‘execution’’ in Table 1.2. The 10 million is a round number with an error margin of perhaps 2.5 million on either side, made up as follows: 2 to 2.5 million deaths arising from deportations and detention in prisons and labor camps, but not counting judicially or administratively prescribed executions (Ellman 2003); plus 4.6 to 8.5 (but probably around 5.7) million premature deaths in the famine of 1932 and 1933 (Davies and Wheatcroft 2003: 415); plus 1 to 1.5 million more premature deaths in the famine of 1947 (Ellman 2000). Higher figures are in circulation, for example the 15 million suggested in the Black Book of Communism (Courtois 1999). A figure of 15 million, however, lies above the upper limit of the error margin on the figures used here by about 2.5 million.

Published References Barber, John, and Mark Harrison. 1991. The Soviet Home Front, 1941–1945: A Social and Economic History of the USSR in World War II. London: Longman. Barber, John, and Mark Harrison. 2006. ‘‘Patriotic War, 1941 to 1945.’’ The Cambridge History of Russia, vol. 3: The Twentieth Century, ed. Ronald Grigor Suny, 217–42. Cambridge: Cambridge University Press Broadberry, Stephen, and Mark Harrison. 2005. ‘‘The Economics of World War I: An Overview.’’ In The Economics of World War I, ed. Stephen Broadberry and Mark Harrison, 3–40. Cambridge: Cambridge University Press. Courtois, Stephane. 1999. ‘‘Introduction: The Crimes of Communism.’’ In The Black Book of Communism, ed. Stephane Courtois, Mark Kramer (translator), Jonathan Murphy (translator), Jean-Louis Panne, Andrzej Paczkowski, Karel Bartosek, and Jean-Louis Margolin, 1–31. Cambridge, MA: Harvard University Press. Davies, R. W., and Stephen Wheatcroft. 2003. The Industrialisation of Soviet Russia, vol. 5: The Years of Hunger: Soviet Agriculture, 1931–1933. Basingstoke: Macmillan. Ellman, Michael. 2000. ‘‘The 1947 Soviet Famine and the Entitlement Approach to Famines.’’ Cambridge Journal of Economics 24(5): 603–30.

260

Afterword

Ellman, Michael. 2003. ‘‘Soviet Repression Statistics: Some Comments.’’ Europe-Asia Studies 54(7): 1151–72. Harrison, Mark. 1998a. ‘‘Economic Mobilization for World War II: An Overview.’’ In The Economics of World War II: Six Great Powers in International Comparison, ed. Mark Harrison, 1–42. Cambridge: Cambridge University Press. Harrison, Mark. 1998b. ‘‘The Soviet Union: The Defeated Victor.’’ In The Economics of World War II: Six Great Powers in International Comparison, ed. Mark Harrison, 268–301. Cambridge: Cambridge University Press. Harrison, Mark. 2005. ‘‘The USSR and Total War: Why Didn’t the Soviet Economy Collapse in 1942?’’ In A World at Total War: Global Conflict and the Politics of Destruction, 1939–1945, ed. Roger Chickering, Stig Förster, and Bernd Greiner, 137– 56. Cambridge: Cambridge University Press. Wheatcroft, Stephen. 1994. ‘‘The Scale and Nature of German and Soviet Repression and Mass Killings, 1930–45.’’ Europe-Asia Studies 48(8): 1319–53.

Contributors

r. w. davies is Emeritus Professor and Senior Fellow of the Centre for Russian and East European Studies, University of Birmingham. paul r. gregory is Professor of Economics at the University of Houston and Research Fellow of the Hoover Institution on War, Revolution, and Peace, Stanford University. mark harrison is Professor of Economics at the University of Warwick; Distinguished Visiting Fellow of the Hoover Institution on War, Revolution, and Peace, Stanford University; and Senior Research Fellow of the Centre for Russian and East European Studies, University of Birmingham. andrei markevich is Marie Curie Research Fellow of the New Economic School, Moscow; Associate Fellow of the Department of Economics, University of Warwick; and Research Fellow of the Interdisciplinary Center for Studies in History, Economy, and Society, Moscow. mikhail mukhin is Research Fellow of the Institute of Russian History of the Russian Academy of Sciences. andrei sokolov is Professor of the Institute of Russian History of the Russian Academy of Sciences.

261

This page intentionally left blank

Index

Page numbers in italic indicate tables and figures. Adomeit, Hannes, 2, 175 adverse selection problem, 213, 226 agents. See military agents; principals and agents Agursky, Mikhail, 2, 175 aircraft industry workforce, 180–209; discipline problems of, 184–185, 198– 199, 201, 205; and downtime and overtime, 193, 193–194, 194, 203; excessive drinking by, 205; growth of, 180–181, 181, 206; and incentives, 196, 199–201, 201; management and skill shortages in, 197–199; materials accounting by, 205; and mobilization, 197–201; numbers and types of workers, 188, 188, 189, 191; pay rates and productivity of, 186–191, 187, 190, 192, 192, 196, 200, 200–201, 201; problems at factory no. 7, 201–

205; and quality, 170–171, 182–184; and shortages of worker housing and food, 191–193; Stakhanovite workers, 186, 203, 204; storming by, 184–185, 201–203, 202; turnover rates, 194– 197, 203–204; work norms, 193–194, 195, 200; workshop sectionalism by, 203 aircraft propulsion research and development, 210–229; basic pay and incentives, 218–220; improvement of existing engines, 211–212; jet engines, 211–212, 216–217, 217, 218, 220, 221; major projects overview, 216– 217, 217; and the market for experimental engines, 213–214; politics and funding decisions in, 222–225; project startup and termination in, 217–218, 218, 220–222, 222–225; rocketry, 211–212, 216, 217, 217, 218, 218, 220, 224–225; secrecy and, 247–248; special interests in, 212–213; steam

263

264

Index

aircraft propulsion (continued) turbines, 211–212, 216, 217, 217, 218, 218, 220–221, 224; turbo jets, 221; uncertainty in, 211–212 Akerlof, George, 227n1, 252n5 Albrecht, Ulrich, 216 Alexander, Arthur J., 175 ammunition industry, 35, 59, 135–140, 137–141 Antonov, O. K., 200 Applebaum, Anne, 233–234 arbitration, xxii, 70, 72, 101–102, 104– 105, 158 armament ministry contract campaign (1940), 100–104 Army (ministry of defense, Red Army, Navy), xxi; as advocate of industrialization, 46–47; Army/Industry rivalries, tensions, and disputes, 70–71, 93, 104–106, 166, 246–248; and contracting campaigns, 100–106; coordination of defense plans with Gosplan, 87–90; coordination of defense plans with Industry plans, 90–91; and defense plan revisions, 106–108; and demobilization and pacifism of 1920s, 34–35; inspection of Industry by, 72– 73, 165–173; and mobilization (see mobilization plans (mobplans)); needs of, vs. productive capacities under NEP, 41–46, 43; operational planning, 78–81; organization of, 1920s, 33–35, 36; and Plan of War, 81, 82, 84; as principal, 33–35, 36, 67–68, 68, 68– 71, 156–157; priority of, 109–110; and quality (see quality in military markets); role in the market for inventions, 214–216, 215, 220, 222–225; and secrecy, 245, 246–248; strengths and weaknesses of, in internal market for weapons, 68–71; use of term, xxi, 34, 36, 67–68, 110n2 Article 58(6) of the RSFSR criminal code, 235–236 artillery industry, 136, 249

Aspaturian, Vernon V., 2 audit (control), 71, 72. See also KPK Balandin, V. P., 200 Baranov, P. I., 183 Barber, John, 23, 51, 236, 258 Bazilevich, G. D., 127–128 Belova, Eugenia, 54, 71, 101, 105 Bergson, Abram, 232, 238, 239 Beriia, Lavrentii, 55, 56, 74, 237 Berliner, Joseph S., 157, 169, 174 Berne, L. P., 221 blanket secrecy, 250 Bogdanov, P. A., 33, 35, 38, 40, 44 Bolsheviks, 4, 26n4, 32–33, 53, 232, 235 Bone, Jonathan, 231, 232, 233 Broadberry, Stephen, 13, 258 Buzinkov (army officer in mobilization planning), 126–127 Bystrova, Irina, 2, 3 cadre factories, 35, 37, 80–81, 90 Carr, E. H., 26, 31, 61 Cheka, xxiii chemical weapons and defense, 32, 44, 60, 61, 74, 121, 132, 134, 146, 150n19, 175 Cherepenina, N., 147 chief administration of military industry. See GUVP chief administrations (glavnoe upravlenie, GU), xxii, 58, 58–59, 60– 61, 61, 64, 96, 97, 102 China-Soviet relations, 24 Chubar’, Vlas, 128–129 CHUSO, 33 civilian factories (producers of defense goods), 35, 120, 142–144 clothing and footwear industries, 144, 147, 172 Colby, William E., 238 command system: army as analogy of, 53–54; and contracts, 99–101; disobedience and disloyalty in, 256; existence of markets in, 53, 65–71, 220; organi-

Index zation and structure of, 55–56, 56, 56–64, 57; and quality problems, 157– 158; and regulation and inspection, 71–75; and secrecy, 233–235; Stalin’s role in, 54, 55–56, 64; success of, 258– 259; use of market information in, 240–244. See also plans and planning commisariats/commisars, xxi, xxii, 57– 58. See also ministries Committee for Demobilization and Mobilization of Industry, 37 Committee for Military Orders (KVZ VSNKh), 33, 35, 36 communications, 142–144 Communist Party, 53 Conquest, Robert, 3, 235 contracts/contract campaigns: decentralization of, 99–100; disputes and resolution, 72, 101, 104–105; hold-up problems and, 160–165; ministry of armament campaign of 1940, 100– 104; and plan of current military orders, 81; process of, 100–101 control. See audit (control); secret and security police Cooper, Julian, 233, 250 corrective labor. See Gulag (chief administration of labor camps) Council for Military Industry (SVP, Voenpromsovet), 33 Council of Labor and Defense (STO). See STO Council of Ministers. See Sovnarkom (Council of People’s Commissars, SNK) Crimean War, 5 Crowfoot, John, 63 Cuban missile crisis, 24 Danilov, B., 214 Davies, R. W., 3, 5, 6, 8, 12, 16, 22, 26, 27, 31, 66, 79, 122, 134, 141, 143, 157, 167, 180, 218, 244 ‘‘Decree on Transition to an 8-Hour Working Day and ... Departure of

265

Manual and Office Workers from Enterprises and Establishments’’ (Supreme Soviet presidium), 196– 197 Defense Commission (1920s), 34, 36, 56, 57, 86 Defense Commission (1930–37): and labor, 196, 214; and mobilization planning, 123, 124, 125–126, 127, 136–137, 149; succession of, 56, 57, 86 Defense Committee: and defense planning, 92, 95, 108–109; and mobilization planning, 125, 127, 128, 129, 130, 138; organization of, 86–87; succession of, 33, 56, 57, 72–73, 83 defense industry. See Industry (military industry, ministries and factories) defense markets. See market for weapons defense ministry. See Army (ministry of defense, Red Army, Navy) defense planning, 78–110; agencies and institutions, 83, 84–87; contract campaigns, 99–106; coordination of Army and Industry plans, 69–70, 70, 90–96; current output aspect of, 81–82; and defense industry regulation, 71–72; delays in plan approval and use of dual planning, 96–99, 97; and external threats, 1–2, 4–5, 11, 11–13, 12; and internal opposition, 5–10; investment aspect of, 81–82; ministry of finance and, 89–90; operational planning, 78– 84; plan revisions, 99–106; priority of defense and, 108–110; vs. economywide planning, 87–90. See also mobilization plans (mobplans) Defense Subcommittee, 83 design bureaus (KB), 37, 215, 215–216, 225 designers, 211, 215, 215–216, 220–222, 225–226 Dewatripont, Mathias, 223 dictators/dictatorships, 7–9, 17–21, 235 Dilbert, 239, 240

266

Index

domestic threats (domestic opposition). See internal threats drunkenness, 205 economic development plan, 81, 82, 93 economic planning. See Gosplan economy: significance of defense industry to, 50–51, 51. See also command system; plans and planning Efimov (in artillery administration), 126, 171 Egorov, A. I., 122–123, 124 Ellman, Michael, 13 Eloranta, Jari, 156 Ermolov, Arsenii, 171 espionage, 7, 235–237, 244–245, 245– 246 external threats: and defense planning, 1–2, 4–5, 11, 11–13, 12; and mobilization planning, 118–120; secrecy and, 237–238; Stalin’s fears of, 1–2, 27n6, 255; and use of military force and repression, 17–21, 18, 19, 20, 21– 25 fifth column, 9 Finland, 4 Firth, Noel E., 235 Fitzpatrick, Sheila, 6, 7, 8–9 food industry, 122, 128, 144–146, 145, 148–149 forced labor, 74–75, 237–238 Gatrell, Peter, 135 Gerasimov, G. I., 130, 131 Gerlach, Christian, 23–24 Germany, 4, 5, 23, 138 Getty, J. Arch., 3, 7, 9, 15 GKO (wartime State Defense Committee), 33, 56, 57, 86 glavnoe upravlenie (GU). See chief administrations (glavnoe upravlenie, GU) Glushko, Vladimir, 216, 218, 224–225 glycerin supplies, 145–146

Golovanov, Iaroslav, 224 Gorbachev, Mikhail, 234–235 Goriaeva, T. M., 232 Gorlizki, Yoram, 56, 234 Gorodetsky, Gabriel, 2 Gosarbitrazh (state arbitration service), xxii Gosplan, xxii; on defense cutbacks, 41– 42; and defense industry plan, 94–95, 96; and mobilization, 129, 148; responsibilities of, 33, 35, 87–88; and secrecy, 235, 236–237, 248; on war readiness, 44 Gosplan affair, 236–237 Gosplan defense sector, 44, 86, 109, 125–126, 127, 136, 137, 142 Granick, David, 157 Great Terror (1937–38), 3, 9–10, 15, 25, 74, 224–225, 236 Gregory, Paul R., 1, 71, 74 Greif, Avner, 240, 243 GU. See chief administrations (glavnoe upravlenie, GU) Gulag (chief administration of labor camps), x, xxii, 233–234, 235–236 GUVP, 33, 36, 38, 39–40, 41, 43, 44–45, 60 GVPU, 45 Harrison, Mark, 3, 5, 12, 13, 23, 52, 63, 71, 74, 96, 123, 133, 139, 148, 149, 157, 160, 163, 167, 168, 180, 218, 239, 241, 244, 245, 246, 247, 258 Hayek, F. A., 99 heavy industry ministry (NKTP), xxi, 61. See also Industry (military industry, ministries and factories) Hitler, Adolf, 5, 23, 24, 258 hold-up problem, 159–165 Holloway, David, 2, 3, 74, 169, 216 Holocaust, 24 horse-drawn carts, 143–144 Iakovlev, Aleksandr, 211, 224 incentives, 187, 219

Index Industry (military industry, ministries and factories), xxi; as agent, 33–35, 36, 67–68, 68, 68–71, 156–157; Army/Industry rivalries, tensions, and disputes, 70–71, 96, 104–106, 166, 246–248; and civilian producers and products, 90–91, 91; classification of factories (cadre factories, reserve factories, civilian factories), 35; complexity and fragmentation of, 59, 61–64; contracting process, 100–102; economic significance of, 50–51, 51; expansion and investment in artillery and ammunition, 135–137; financial problems under NEP, 38–41; GUVP, 33, 36, 38, 39–40, 41, 43, 44–45, 60; illicit trade and corruption in, 243–244; inability to meet Army needs under NEP, 41– 46, 43; ministries of, organization and structure, 45, 58–59, 60–61, 130; and mobilization (see mobilization plans (mobplans)); organization of (1920s), 33, 36, 37; peak wartime production (major powers comparison), 137; and planning (see defense planning; mobilization plans (mobplans)); Plan of Gross Output of Defense Industry, 83, 84; political significance of, 51–52; priority of military requirements and, 108–109; and quality (see quality in military markets); regulation and inspection of, 71–75; role in market for inventions, 214–216, 215, 220– 221; and secrecy, 232–235, 233, 237– 238, 239–240, 245, 246–251; and Stalin’s fears of military-industrial collusion, 2–3, 70–71, 162–163, 256– 257; strengths and weaknesses in internal market for weapons, 68–71; use of term, xxi, 34, 36, 67–68; War Communism agencies and commissions, 32–33; and World War I, 32. See also VSNKh (Supreme Council of the National Economy, industry ministry); specific industries

267

industry ministry. See VSNKh (Supreme Council of the National Economy, industry ministry) information markets. See secrecy internal markets, 65–71; and real markets, 65–66, 66. See also market for inventions; market for weapons internal threats, 17–21, 18, 19, 20, 21– 25 Invasion Constraint, 17–21, 18, 19, 20 Ivanov, V. P., 184, 192 Ivkin, V. I., 72, 73 Japan, 5, 12 Jersak, Tobias, 23–24 jet engines, 211–212, 216–217, 217, 218, 220, 221 Kagan, Donald, 24 Kaganovich, Lazar, 8, 55, 56, 130, 246 Kaganovich, Mikhail, 55, 109, 128, 129– 130, 134, 171, 186–187, 221, 224 Kalinin, Mikhail, 6 Kalinin region mobilization plan, 147 Kamenev, S. S., 7, 8, 42 Kamenev-Zinov’ev trial, 7, 8 Ken, Oleg, 78 Kerber, L. L., 184 Khalepskii (Red Army administrator), 141 Khanin, G. I., 234 Kharitonovich (planning official), 97–99 Khlevniuk, Oleg, 3, 5–6, 8, 9, 10, 15, 56, 64, 127, 237 Khrunichev, Mikhail, 224 Khrushchev, Nikita, 55, 234 Korolev, Sergei, 216, 218, 224–225, 226 Kostikov, Andrei, 224 Kostyrchenko, G. V., 198 Kozlov, V. P., 3, 9, 11 KPK, xxii, 169, 170, 172–173, 174, 239–240, 246 Kroll, Heidi, 70, 72 Kruglov, Sergei, 219 kulaks, 6, 11

268

Index

Kviring, E. I., 148 KVZ VSNKh, 33, 35, 36 labor: forced labor, 74–75, 237–238; mobilization planning for, 148–149; Supreme Soviet presidium decree (June, 1940), 196–197. See also aircraft industry workforce Lange, Oskar, 65 Lazerev, Valery, 64 lemons, market for, 227n1, 252n5 Leningrad mobilization plan, 146–147 Likhonon (party secretary), 204 Liul’ka, Arkhip, 217, 219, 221 loyalty, 2, 64, 73, 176, 222, 223–224 M-3 (MP-3), 120, 122–123, 124, 125 M-18, 120 Mackenzie, Donald, 220 Maddison, Angus, 52 Malenkov, Georgii, 55, 56, 221 Mao Zedong, 24 market for inventions: organization of, 214–216, 215; role of designers in, 220–222; success of, 225–226. See also aircraft propulsion research and development market for labor. See aircraft industry workforce market for lemons, 227n1, 252n5 market for weapons: Army and Industry as principal and agent, 33–35, 36, 67– 68, 68, 68–71, 156–157; as internal market, xiii–xiv, 75, 156–157. See also quality in military markets market information, 240–244 markets. See internal markets Markevich, Andrei, 78, 81, 96, 99, 102, 106, 160, 168, 220, 244, 247, 248 Maskin, Eric, 223 Masliukov, Iurii, 235 master delivery agreements, 100 Mawdsley, Evan, 2 Mayer, Arno, 6 Medvedev, Roy A., 3, 26, 27, 238

Mezhlauk (Gosplan chief), 248 Mikoian, A. I., 146, 234 military agents, 165–173, 173–175, 247 military force vs. repression, 3, 4, 17–21, 18, 19, 20, 21–25 military industrial commission, 128–130 military-industrial complex, x, 2–3, 70– 71, 162–163, 256–257 military markets. See market for weapons ministries, xxi, 56–64; complexity and fragmentation of, 59–64; Council of Ministers, 56–57; Defense Committee as link to, 56, 57; mobilization sections of, 85; organization of, 56–59 ministry of armament contract campaign (1940), 100–104 ministry of defense. See Army (ministry of defense, Red Army, Navy) ministry of finance, 89–90 Mises, Ludwig von, 65–66, 99 mixed economy. See NEP (New Economic Policy) mobilization: and aircraft industry labor, 194–197, 195; of armaments factories, 134–140; and use of repression, 13, 15–16 mobilization plans (mobplans), 79–80, 118–155, 137–141; agencies and organizations responsible for (1920s), 37; for civilian sector, 142–146; criticism of, 123–125; estimated vs. real capacities (1938), 137–141; external threat estimation and, 118–120; and food industry, 144–146, 145; government reorganization and, 127–128; for labor, 148–149; lack of central control of, 125–127; under military industrial commission, 128–130; ministries and, 85; regional and district, 146–148; resources allocated to, 149; revision of, 120, 121, 121–123, 130–131; secrecy and, 247; for tank industry, 140–142; vs. actual capacities, 131, 132, 133, 133–135; and wrecking activity, 125

Index mobplans. See mobilization plans (mobplans) model contracts, 100–101 Mokyr, Joel, 211 Molotov, Viacheslav, 10, 55, 56, 128, 148, 221, 248 Morishima, Michio, 65 Morukov, Mikhail, 74 Moynihan, Daniel Patrick, 238 MP-1, 130, 131, 137 MP-4, 124, 125 MP-10, 128 Mukhin, Mikhail, 216, 220 Mundell, Robert A., 17 MV-10, 120, 121 MVD (interior ministry), xxii, xxiii, 73, 74–75. See also NKVD (interior ministry); OGPU; secret and security police

269

operational planning, 79, 80,81 Ordzhonikidze, Sergo, 55, 56, 64, 107, 134, 141–142, 255 OTK, 158, 167–168, 170, 172–173

Naumov, Oleg V., 8, 9, 15, 27, 119, 130, 131 NEP (New Economic Policy), xxii, 31, 33, 38 NKOP (defense industry ministry), xxi NKTP (heavy industry ministry), xxi. See also Industry (military industry, ministries and factories) NKVD (interior ministry): defined, xxii; and forced-labor, 74–75; involvement in high-technology defense industry, 73, 74; in the market for weapons, 69; and mobilization planning, 145, 147; and prison design bureaus, 216; and repression, 9–10, 13, 15, 236 (see also MVD (interior ministry); OGPU); and secrecy, 234 NKVM (defense ministry), 33 Noren, James H., 235 Noril’sk combine, 237–238

Parallel Center, 7 Park, Robert, 238 Pashukanis, E. B., 52 Pavlunovskii, I. P., 126, 133–134 pay rates. See wages Peloponnesian wars, 6–7, 24 people’s commissariats, xxi. See also ministries Perov, V. I., 221 Petrov, N. V., 73 Piatakov, Iurii, 7, 125, 134, 248 Plan of Current Military Orders, 80, 81, 84, 91, 92, 95. See also defense planning Plan of Gross Output of Defense Industry, 84. See also defense planning Plan of War, 81, 82, 84 plans and planning: legal force of, 52–53; national economic development plan, 81, 82, 93; role of planners, 71–73. See also command system; defense planning; Gosplan; mobilization plans Poland, 4 Polikarpov, N. N., 184–185 Politburo (Political Bureau), xxii–xxiii, 33, 55–56, 86, 96, 235, 248 press and secrecy, 248–251 principals and agents, 33–35, 36, 55–56, 56, 65–71, 66, 68, 156–157, 250–252 pripiski (pripiska), 174–175 Prisiazhnyi, N. S., 33 prison design bureaus (sharaga), 73–74 propellants, 139–140, 141 public opinion and secrecy, 248–251

OGPU, 11, 13, 44–45, 73, 74, 75, 85, 232. See also MVD (interior ministry); NKVD (interior ministry) Omsk region mobilization plan, 147– 148

quality in military markets, 156–179; acceptance of substandard equipment, 170–171, 172–173; Army-Industry collusion to conceal shortfalls and delays, 173–175; Army-Industry

270

Index

quality in military markets (continued) market-like relationship, 157, 166; command system and, 157–158; criminal prosecution, 169; design changes and, 182–184; economic and military aspects of, 158–159; at expense of quantity, 164–165, 173–175; factory managers and military agents antagonism, 168–169; financial penalties, 169; Industry hold-up and Army response, 159–165, 175–176; market for lemons, 227n1, 252n5; military agents and, 165–173; problems as inherent in the command system, 156– 157; quality defined, 157; rejection of goods, 169, 170, 171–172; secrecy and, 249–251; self-regulation at factory and ministry level, 158, 167–168, 170 quasi-markets. See internal markets Raack, Richard C., 2 railroads, 143 rational-actor tradition, 238, 256 Red Army. See Army (ministry of defense, Red Army, Navy) repression: and domestic opposition, 9– 10; mobilization of, 13, 14, 15–16; numbers sentenced, 14, 22; rearmament vs. repression model, 16–21; as response to external threats, 4, 21–25; secrecy and, 236; ‘‘unintentional’’ explanation of, 3–4 research and development. See aircraft propulsion research and development research institutes (NII), 37, 215, 215– 216 reserve factories, 35 Revolutionary Military Council (RVS, Revvoensovet), 33, 42, 45–46 Revolution Constraint (RC), 17–21, 18, 19, 20 Revvoensovet, 33, 42, 45–46 Right-Trotskyist Center, 7 roads, 143

rocketry, 211–212, 216, 217, 217, 218, 218, 220, 224–225 Rodionov, Ivan, 224 Rogerson, William P., 160 Rosen, Steven, 2 Rukhomovich, Moisei, 98–99, 125 Rykov, A. I., 34 Samuelson, Lennart, 1, 32, 78, 79, 81, 86, 89, 107, 162, 163 Schelling, Thomas C., 238 Schmitz, Patrick W., 160 science and technology, 73–74, 210– 213. See also aircraft propulsion research and development; market for inventions secrecy, 230–254; collective motivations for, 237–239; costs of, 230–231; in defense planning, 87–88; and discounting effect, 243, 244–245; and espionage, 7, 235–237, 244–245, 245–246; Industry exploitation of, 246–251; laws regarding, 235–237; limits to, through rivalry and competition, 251–252; and market information, 240–244; private benefits of, 238, 239–240; purposes of, 231; rise and decline of, 232–235, 233; and role of press and public opinion, 248–251; secrecy of, 231; and strategic information, 244–245 secret and security police, xxiii, 13, 13– 16, 14. See also MVD (interior ministry); NKVD (interior ministry); OGPU Seliunin, V., 234 Serov, Gennadii, 224 Shakhty affair, 26n3 Shakhurin, A. I., 199, 201–202 Shaposhnikov, B. M., 119 sharaga (prison design bureaus), 73–74 ships and shipping, 144 Siddiqi, Asif, 220, 225 Simonov, N. S., 2, 5, 6, 32, 34, 46, 51, 74, 78, 80, 89, 98, 108, 160, 163, 244, 246, 248

Index Skorkin, K. V., 73 Smirnov, G. I., 125 Smirnov, I. I., 41 soft budget constraint, 41 Solomon, Peter H., 158 Solzhenitsyn, Alexander, 236 Sovnarkom (Council of People’s Commissars, SNK), xxiii, 33, 56, 86, 138. See also Gosplan Sovnarkom Economic Council, 64 Spanish civil war (1936), 9 specialists, 44–45, 218–219 specific contracts, 100–101 Stakhanov movement (Stakhanovites), 186 Stalin, Josef, 16–17, 96; and command system, 54, 55–56, 64; deaths attributed to, 257–258, 259n1; and defense planning, 86, 256–257; fears of collusion among enemies, 7–10, 26n3, 26n4, 27n6; fears of external threats, 1–2, 27n6, 255; fears of hidden and unconscious enemies, 7–9; fears of military-industrial collusion, 2–3, 70– 71, 162–163, 256–257; plans to conquer western Europe, 2; power of, 53; priority on defense, 108; role in market for inventions, 211, 214, 220, 224, 226; and secrecy, 236–237; and use of military force, 3, 4, 10–11, 16–17, 21– 25; and use of repression, 3–4, 7, 9– 10, 10–11, 13, 13–16, 14, 21–25; worldview of, 7–9 Starkov, Boris, 74, 216 Starovskii, Vladimir, 234 State Defense Committee. See Defense Committee steam turbines, 211–212, 216, 217, 217, 218, 218, 220–221, 224 STO (Council of Labor and Defense), xxiii, 33, 40–41, 45, 83, 86, 87–88, 88–89, 91, 101, 108, 127 Stone, David, 32, 46, 78, 80, 119 storming, 184–185, 201–203, 202 strategic information, 244–245

271

Supreme Council of the National Economy, 33, 37, 38, 59, 61. See also GUVP; VSNKh Supreme Soviet presidium decree (June, 1940), 196–197 Suvorov, Viktor, 2 SVP. See Council for Military Industry (SVP, Voenpromsovet) switching costs, 160, 161–162 Talanova, L. E., 182, 183, 184 tank industry, 140–142, 171–172, 249 Tarschys, Daniel, 238 technological culture, 181, 182–186, 206 technology. See science and technology threat effect, 243 Thucydides, 7, 24 Timoshenko, S. K., 119 Tinbergen, Jan, 17 tinplate supplies, 146 Tochenov, S. V., 23 transport and communications, 142–144 treason, 7 Trotskii, L. D., 34 trusts, 57–58 Tukhachevskii, Mikhail, 5, 44, 79, 124, 162–163, 216, 220, 222, 224, 225, 255 Tupolev, A. N., 184, 226 Uldricks, Teddy J., 2 Unshlikht, I. S., 45, 46, 162 Ustinov, Dmitrii, 219, 235 Uvarov, Vladimir, 217, 247–248 Vannikov, Boris, 108, 133, 246 vehicle industry, 143–144 Vernidub, I. I., 140, 141 Voenprom (military industry), xiii, 34, 45. See also Industry (military industry, ministries and factories) Voenpromsovet. See Council for Military Industry (SVP, Voenpromsovet) Voenved, 34. See also Army (ministry of defense, Red Army, Navy)

272

Index

Voronezh region mobilization plan, 147– 148 Voroshilov, K. E., 45, 55, 107, 124, 125, 136, 141, 248 Voznesenskii, Nikolai, 55, 56, 237 VPK (Defense Committee special military-industrial commission), 87 VPU. See Voenprom (military industry) VSNKh (Supreme Council of the National Economy, industry ministry), xxi, 33, 37, 38, 59, 61. See also GUVP wages, 40, 186–191, 196, 199–201, 218–220 Walras, Léon (Walrasian market), 68 War Communism agencies and commissions, 32–33

war scare (1927), 4–5, 6 Weber, Max, 238 Weeks, Albert L., 2 West Siberia mobilization plan, 147 Wheatcroft, Stephen, 6, 16, 258 Wintrobe, Ronald, 7 workshop sectionalism, 203 World Disarmament Conference (Geneva), 22 World War I, 32, 210, 258 World War II, 22–23, 56, 108, 157, 158– 159, 176n4, 255, 257–259 wrecking activity, 26n4, 73, 125, 127, 129, 150n17 Zaleski, Eugène, 54, 64, 92 Zhdanov, Andrei, 55, 234, 249–250