The United States Economy Performance and Issues [illustrated] 1557752311, 9781557752314

Citation preview

The United States Economy Performance and Issues By a Staff Team Headed by Yusuke Horiguchi Charles Adams Sharmini Coorey Krister Andersson Robert Corker A. Lans Bovenberg Liam P. Ebrill Bankim Chadha Owen Evans Daniel Citrin Steven M. Fries David T. Coe Lloyd Kenward Ellen M. Nedde

International Monetary Fund 1992

© 1992 International Monetary Fund

Library of Congress Cataloging-in-Publication Data

The United States economy: performance and issues I by a staff team headed by Yusuke Horiguchi. p. em. ISBN 1-55775-231-1 I. United States-Economic policy-1981- 2. United StatesEconomic conditions-1981- I. Horiguchi, Yusuke. II. International Monetary Fund. HC!06.8.U548 1992 338.973--dc20 92-16898 CIP This book's cover and its interior were designed by the IMF Graphics Section. Esha Ray of the IMF's External Relations Department edited the book for publication. Price: US$35.00

Please send orders to: International Monetary Fund, Publication Services 700 19th Street. N.W .• Washington, D.C. 20431, U.S.A. Telephone: (202) 623-7430 Telefax: (202) 623-7201

Preface This volume brings together technical papers prepared by the IMF staff for the Article IV consultations with the United States over the period 1988-90. The papers focused on key issues for U.S. economic policy in recent years and examined measures that might be expected to improve the prospects for growth over the medium term. The issues analyzed include macroeconomic questions related to fiscal and monetary policies, the current account, national saving, and exchange rates, as well as microeconomic topics such as the performance of the U.S. health care sector and the ramifications of U.S. tax and public expenditure structures. The North American Division of the IMF's Western Hemisphere Department was primarily responsible for the preparation of the papers but staff members of the Research and Fiscal Affairs Departments also made contributions. The authors are Liam Ebrill, Owen Evans, Charles Adams, Daniel Citrin, Lloyd Kenward, Robert Corker, Lans Bovenberg, Steven Fries, Sharmini Coorey, Krister Andersson, Bankim Chadha, and Ellen Nedde. The papers were prepared under the direction of Yusuke Horiguchi, at that time Assistant Director in the Western Hemisphere Department. The papers have benefited from discussions with U.S. officials and other staff members; however, the views expressed are those of the authors alone. Although time has elapsed since these papers were prepared, it was decided not to update them on the grounds that in their original form they would convey better the context in which they were written and the issues were discussed. S. T. Beza Counsellor and Director IMF Western Hemisphere Department

iii

Contents Preface

Page iii

Chapter 1. Overview

Yusuke Horiguchi, Liam P. Ebrill, and Charles Adams

1

PART I. NATIONAL SAVINGS AND PUBUC POUCY

Chapter 2. National and Personal Saving: Measurement and Analysis of Recent Trends

A. Lans Bovenberg and Owen Evans

19

Chapter 3. Tax Policy and National Saving: A Survey

A. Lans Bovenberg

52

Chapter 4. National Savings and Targets for the Federal Budget Balance

Owen Evans

77

Chapter 5. Social Security, Demographic Trends, and the Federal Budget

Liam P. Ebrill

116

Chapter 6. Some Microeconomics of Fiscal Deficit Reductions: The Case of Tax Expenditures 143

Liam P. Ebrill Chapter 7. Investment in Housing: A Portfolio Approach to Effects of Changes in Tax Policy

Krister Andersson

170

Chapter 8. Fiscal and Economic Impact of Federal Credit Programs Steven M. Fries

192

v

vi

•

Contents

PART II. OTHER DOMESTIC MACROECONOMIC ISSUES AND MEDIUM-TERM OUTWOK

Chapter 9. Tax Policy and Business Investment: Evidence from the 1980s

Robert Corker, Owen Evans, and Lloyd Kenward

221

Chapter 10. A System's Approach to Estimating the Natural Rate of Unemployment and Potential Output for the United States

Charles Adams and David T. Coe

252

Chapter 11. The Determinants of U.S. Real Interest Rates in the Long Run

Sharmini Coorey

314

PART Ill.MONETARY AND FINANCIAL ISSUES

Chapter 12. Broad Money Growth and Inflation

Liam P. Ebrill and Steven M. Fries

357

Chapter 13. An Expensive Thrift Industry

Steven M. Fries

373

PART IV.CURRENT ACCOUNT ADJUSTMENT AND THE EXCHANGE RATE

Chapter 14. A Dynamic Error-Correction Model of the U.S. Current Account

Ellen M. Nedde

399

Chapter 15. The Recent Behavior of U.S. Trade Prices

Daniel Citrin

445

Chapter 16. Structural Models of the Dollar

Charles Adams and Bankim Chadha

461

Contents

•

vii

PART V. STRUCTURAL ISSUES

Chapter 17. The U.S. Health Care Industry: Performance and Issues Liam P. Ebrill

499

Chapter 18. Implications of Integrating U.S. Corporate and Individual Income Taxes Krister Andersson

531

Chapter 19. Developments in Federal-State-Local Fiscal Relations: Implications for Infrastructure Investment and Education Liam P. Ebrill

574

In statistical matter throughout this book, dots(...) indicate that data are not available; a dash(-) indicates that the figure is zero or less than half the final digit shown, or that the item does not exist "billion" means a thousand million, and "trillion" means a thousand billion a short dash( ) is used between years or months(for example, 1989-90 or January­ -

October) to indicate a total of the years or months inclusive of the beginning and ending years or months a stroke(/) is used between years(for example, 1990/91) to indicate a fiscal year or a crop year a colon (:) is used between a year and the number indicating a quarter within that year (for example, 1991 :2) components of tables may not add to totals shown because of rounding.

Overview YUSUKE HORIGUCHI, LIAM P. EBRILL,

F

AND

CHARLES ADAMS

rom the fourth quarter of 1982 to the third quarter of 1990, the U.S. economy experienced its longest peacetime expansion. During this period, the growth of real GNP averaged 3. 6 percent a year. Employment in the nonfarm business sector increased by 22 million, bringing the unemployment rate down from a peak of 10¥4 percent in late 1982 to 5\14 percent during the first half of 1990. Inflation declined sharply from double digit rates in the early 1980s and remained in the 4-5 percent range throughout much of the expansion. Notwithstanding the relatively favorable performance, however, a number of developments in the U.S. economy during the 1980s have been of concern to U.S. policymakers and also to the international community. Most notably, the U.S. national saving rate has declined as a result of an unsatisfactory fiscal policy performance and a trend toward lower private saving. The weak national saving rate, in turn, has been accompanied by significant current account deficits and has contributed to an investment and productivity performance that has fallen short of that envisaged by the U.S. Administration. The exchange value of the dollar experienced large swings during the decade. Inflation, although falling sharply in the early 1980s, remains high by historical standards. In addition, a number of structural issues have arisen, concerning, inter alia, the allocative impact of the tax system, the safety and soundness of financial institutions, the provision of health care, and the prospects for a revival of productivity growth. The present volume brings together technical papers prepared by the staff of the IMF for its Article IV consultations with the United States over the period 1988-90 with a view to shedding light on these issues and contributing to debate on the policy measures that might enhance the 1

2 • Overview

prospects for sustained and strong growth over the medium term. For the convenience of the reader, an overview of these papers follows. (The year in which the papers were prepared is given in parentheses after the title of each chapter.) I. National Saving and Public Policy

The national saving rate in the United States dropped markedly in the 1980s. Gross national saving averaged l6Y2 percent of GNP in the period 1950--79, but fell to below 13 percent of GNP in the last three years (1988-90). In an accounting sense, the deterioration in the public sector balance was responsible for a little more than half of the fall in the national saving rate, with the remainder resulting from a decline in the private saving rate. The implications of the drop in the U.S. saving rate extend beyond the country's borders because of the sheer size of the U.S. economy and the role of the dollar in international trade and finance. A decline in U.S. saving puts upward pressure on both domestic and world interest rates with ramifications for investment and growth in partner countries, including those stemming from the effects of changes in the debt-service burden of the highly indebted countries. Furthermore, to the extent that the persistence of a lower U.S. saving rate manifests itself in a sizable current account deficit (and an accumulation of large external liabilities), the stability of exchange and financial markets may be affected and the world trading system could be exposed to the threat of increased protectionism. For some time, the U.S. Administration has emphasized the need to strengthen national saving performance if the U.S. economy is to achieve sustained growth over the medium term together with lower inflation and a reduced current account deficit. In recognition of the important role of fiscal policy in this regard, the Administration's January 1990 budget sought the achievement of approximate balance by fiscal year 1995 after excluding the cash-flow surpluses of the social security trust funds. The papers presented in Part I of this volume address a range of the empirical and public policy issues arising from the decline in U.S. national saving. Chapter 2, "National and Personal Saving: Measurement and Analysis of Recent Trends" (1988) by A. Lans Bovenberg and Owen Evans, documents recent trends in private saving, explores several issues regarding the measurement of saving, and considers how the principal theoretical models of saving behavior might be used to identify the factors behind the decline in the personal saving rate that has accounted for a

Yusuke Horiguchi, Uam P. Ebri/1, and Charles Adams •

3

major part of the deterioration in the private saving performance. As regards the appropriate measurement of personal saving, issues addressed are whether to include accrued capital gains and household purchases of nonresidential consumer durables; whether to correct for inflation; and whether to count social security contributions as private saving. Even with adjustments for these factors, however, the conclusion still holds that there has been a significant decline in the personal saving rate. It is suggested that improvements in wealth positions, in the living standards of the elderly, in social security pensions, and in private and public insurance mechanisms all contributed to the declining trend in personal saving. Demographic factors and increased efficiency of capital markets also seem to have played a role. Chapter 3, "Tax Policy and National Saving: A Survey" (1988) by A. Lans Bovenberg, explores how tax policy affects saving and examines which policy instruments might be most effective in raising the level and efficiency of national saving. The current income tax system in the United States discriminates between different types of saving and investment and largely exempts several forms of capital income from income taxation. The differential tax treatment-in particular the preferential treatment afforded owner-occupied housing and other durable consumer goodshas affected the allocation of saving between real and financial assets. Whereas the Tax Reform Act of 1986 and lower inflation rates have reduced some of these effects, the growing integration of world financial markets has tended to raise the adverse allocational effects of the differential tax treatment. On balance, the analysis of the tax treatment of private saving in the paper tends to support the view that, while its effect on the level of saving is relatively small and uncertain, its effect on the composition of saving and investment is significant. The paper concludes by observing that raising public saving (reducing the federal budget deficit) is the most direct and effective way to increase national saving in the United States. The appropriate target for the federal budget balance as a means for reversing the decline in the national saving rate is addressed by Owen Evans in Chapter 4, "National Savings and Targets for the Federal Budget Balance" (1989). Two alternative frameworks aimed at quantifying medium-term objectives for the federal fiscal balance are developed. The first is based on an explicit optimality criterion-namely, finding the neoclassical steady-state path on which the present value of per capita consumption is maximized. Illustrative calculations suggest that a net national saving rate of close to 10 percent of net national product (NNP), substantially higher than present levels, might be called for. The second

4 • Overview

approach calculates the net national saving rate consistent with a target growth for GNP. Assuming no reliance on foreign saving, an average annual output growth of 3Y4 percent over the medium term, as envisaged by the Administration at the time that the paper was written, would seem to require a net national saving rate of almost 10 percent of NNP. The two alternative frameworks analyzed therefore yield broadly similar results that imply that a federal budgetary surplus (on a unified budget basis) equivalent to around 3 percent of NNP would be an appropriate medium- or long-term objective for U.S. fiscal policy. The precise numerical results, of course, depend on the specific framework and the assumptions about certain parameters. Nevertheless, the paper's conclusions provide strong support for the view that the U.S. Federal Government should strive for the achievement of a substantial surplus on a unified budget basis over the medium to long term. One justification for calling on the Federal Government to aim for the achievement of budgetary surpluses is that some part of the decline in private saving may reflect the unintended consequences of public sector interventions (beyond those associated with the tax treatment of saving) for private market decisions. A potentially important example, in this regard, that is addressed in Chapter 5, "Social Security, Demographic Trends, and the Federal Budget" (1989) by Liam P. Ebrill, concerns the social security system. As background, the operations of the social security trust funds are currently running growing cash-flow surpluses for demographic reasons-notably, the passage of the "baby-boom" generation into its years of higher earnings capacity. These surpluses, however, will eventually give way to large and growing deficits as the baby-boom generation enters its retirement years. This paper analyzes how social security influences individual economic behavior and, on the basis of that analysis, examines the potential macroeconomic and policy implications of the interaction between the existing social security system and prospective demographic changes. The principal conclusion is that a strong case can be made for "saving" the social security surpluses, lending support to the Administration's medium-term objective-as set forth in the January 1990 budgetof achieving fiscal balance net of such surpluses. The considerations underlying this conclusion are twofold. First, if the social security surpluses are instead used to finance current government expenditures, then, irrespective of whether the social security system is viewed as a tax/transfer scheme or as a pension plan, there would in effect be a burden placed on the next generation when increased social security payments come due. That would be tantamount to reversing the inter-

Yusuke Horiguchi, Uam P. Ebri/1, and Charles Adams •

5

generational implications of the 1983 reform of social security that was in the direction of lightening the relative burden of the future generation. Second, if social security is analogous to a private pension plan, with individuals taking anticipated social security benefits into account when making their saving decisions, and if the social security cash-flow surpluses are a useful proxy for the impact of the prospective demographic changes on the social security system, then using those surpluses to defray current government expenditures would mean that there would be no commensurate act of investment to underwrite what individuals view as an increment to their savings. The question of how best to attain the goal of an improved federal fiscal balance raises difficult and complex issues-for example what is the appropriate mix between revenue increases and expenditure restraints? Focusing on the revenue side of the budget, Chapter 6, "Some Microeconomics of Fiscal Deficit Reductions: The Case of Tax Expenditures" (1988) by Liam P. Ebrill, suggests that the elimination of a number of tax expenditures could increase federal tax revenue while at the same time contributing possibly to improved economic efficiency. In view of the heterogeneity of tax expenditures, the paper individually evaluates the most important in terms of forgone revenue. On that basis, those to be considered either for reduction or elimination include the exclusion of employer contributions to medical insurance and health care; the deduction of mortgage interest on owner-occupied housing; the deductions for both property taxes and state and local income taxes; the step-up basis of capital gains at death; and the current exclusion of some social security benefits. An example of a tax expenditure which it might be argued could be retained is the exclusion from corporate income tax liability of employer pension-plan contributions and earnings because this exclusion may be viewed as contributing to an increase in national saving. While the reduction or elimination of tax expenditures such as those noted above is likely to yield gains in economic efficiency, it is also likely to involve capitalization effects that are large and potentially disruptive, particularly in those markets where investment decisions have been influenced by the existence of the specific tax expenditures under consideration. Chapter 7, "Investment in Housing: A Portfolio Approach to Effects of Changes in Tax Policy" (1989) by Krister Andersson, considers the possible effects of eliminating the tax deductibility of mortgage interest payments within the context of a portfolio model with three assets (bonds, housing, and stocks) calibrated to reflect actual portfolio shares in the U.S. economy. Since the stock of housing is fixed in the short run, the market's

6 • Overview reactions to the abolition of the tax expenditure would initially involve a decline in house prices as the effect of the change in the tax code is capitalized. This decline would tend to equalize the after-tax rates of return between housing and other assets. The long-run equilibrium would be characterized by a smaller stock of housing than otherwise, a diminished flow of saving into the housing sector, a relative price of housing that has recovered from its initial decline to a level determined by the long-run supply curve, and an increase in the before-tax rate of return on housing. Where the after-tax rate of return would end up would depend on the interaction between changes in the portfolio share of each asset and the corresponding changes in the risk characteristics of investors' portfolios. A key policy conclusion of the paper is that, while abolishing the mortgage interest deduction will result in long-term gains arising from a more efficient allocation of resources, some means of phasing in the reform might be considered with a view to mitigating the short-run capitalization effects that could be disruptive. A common theme shared by Chapters 5 through 7 is that the impact of the federal budget on economic activity in general, and on national saving in particular, cannot readily be summarized by a single statistic such as a budget deficit number, given the complex channels through which fiscal variables-such as taxes, expenditures, and borrowingaffect private decisions. Moreover, as suggested by Chapter 5 on the role of social security, care needs to be taken in selecting whatever measure of the fiscal deficit becomes the focus of policy debate. Chapter 8, "Fiscal and Economic Effects of Federal Credit Programs" (1990) by Steven M. Fries, further develops these points for the case of federal credit assistance programs. While the form of the programs in this area is varied, their long-run fiscal and economic (resource allocation) effects arise primarily from each program's subsidy component rather than annual cash flows. Under budgetary practices in effect at the time the paper was written, however, the cash flows of federal credit assistance programs were emphasized rather than their subsidy components. Because shortrun cash flows bear little or no relationship to credit program subsidies, this practice can render ineffective the Government's control of credit programs. Credit reform proposals by the Administration-proposals that are included in the Federal Credit Reform Act of 1990---would require the recording of the subsidy element of credit assistance programs in the budget and would make control of credit subsidies a focus of budgetary efforts. By subjecting these subsidies to annual appropriations, the Government would gain a more effective means to control the longrun fiscal effects of credit programs. Moreover, such reforms also would

Yusuke Horiguchi, Uam P. Ebri/1, and Charles Adams • 7

represent an important step in improving economic efficiency by eliminating unintended subsidies.

II. Other Domestic Macroeconomic Issues and the Medium-Term Outlook

Two key policy concerns in the domestic macroeconomic scene during the 1980s-which arose in part from the weak saving performance discussed above-were whether the U.S. economy could be expected to return over the medium term to the relatively rapid rates of growth recorded during the 1950s and 1960s and whether real interest rateswhich are generally viewed as having risen to relatively high levels during the 1980s-were likely to decline to their average levels experienced over much of the postwar period. The papers included in Part II of the volume are intended to shed light on these concerns by examining the performance of business fixed investment in the United States during the 1980s, the prospects for potential output and productivity over the medium term, and the factors determining the average level of U.S. interest rates. Chapter 9, "Tax Policy and Business Investment: Evidence from the 1980s" (1989) by Robert Corker, Owen Evans, and Lloyd Kenward, examines the behavior of business fixed investment in the United States, highlighting the role played by declining computer prices in its performance during the 1980s. The paper notes that the sharp declines in computer prices have complicated the assessment of investment during the 1980s. This index number problem arises because investment in high-technology equipment-whose relative price was falling-was growing very rapidly over the 1980s. Under these conditions, the earlier is the base period chosen to measure real investment performance the larger the weight given to a rapidly growing component of investment and the greater the measured increase in the overall index for investment. These difficulties are apparent when various measures of the ratio of investment to output in the United States are examined in order to determine the strength of investment during the 1980s. For example, the ratio of real gross business investment to real GNP measured in 1982 prices indicates a relatively strong performance while the corresponding ratio measured in 1987 prices is considerably weaker. In order to assess the prospects for investment over the medium term, the paper also presents estimates of investment equations for producers' durable equipment and nonresidential structures. The estimated equation for nonresidential structures appears to perform relatively well over

8 • Overview

the 1970s and 1980s but, as a result of the rapid growth of investment in office computing and accounting machinery, the equation for producers' durable equipment significantly underpredicts investment during the 1980s, suggesting that a different approach may be needed to predict the performance of this category of investment. The medium-term outlook for the U.S. economy depends importantly on the growth of potential output which is influenced by the underlying growth of factor inputs (labor and capital), the evolution of the natural rate of unemployment, and the rate at which total factor productivity is expected to increase. Chapter 10, "A System's Approach to Estimating the Natural Rate of Unemployment and Potential Output for the United States" (1989) by Charles Adams and David T. Coe, presents estimates of the natural rate of unemployment and potential output for the United States using an approach that systematically integrates wage and price data with real and structural variables that influence the behavior of factor and product markets. The estimation results indicate that the growth of potential output in the United States during the 1980s recovered somewhat from the low rates of increase during the previous decade, but remained below the rates of increase recorded during the 1960s. The natural rate of unemployment, after rising substantially during the late 1960s and much of the 1970s, was estimated to have declined in the 1980s, largely as a result of a change in the composition of the labor force toward groups with relatively low unemployment rates and declines in the minimum wage adjusted for the share of youth in the labor force. As regards prospects over the medium term, the paper's main conclusion is that, although an improvement in the growth of labor productivity is likely in part as a result of the aging of the population (and the greater maturity of the labor force it implies), this would be largely offset by a continuation in the slowdown of the growth of labor input. Compared with the 1980s, there is likely to be a modest pickup in the growth of potential real GNP to an average annual rate of slightly above 2Y2 percent in the period from 1990 to 1994. This estimate takes into account another result from the estimations, namely that the natural rate of unemployment over the same period is likely to decline further and to be down to about 5 percent by 1994. Another key question in assessing the U.S. economic outlook over the medium term is the prospective behavior of real interest rates. There appears to be broad agreement-measurement and other technical problems notwithstanding-that real interest rates during the 1980s were higher than during the 1960s and 1970s. However, there is no clear consensus on the possible factors behind the higher interest rates, with the

Yusuke Horiguchi, Uam P. Ebri/1, and Charles Adams • 9

lack of consensus primarily attributable to the inconclusive nature of available econometric evidence. The failure to identify the factors that explain the behavior of real interest rates is a handicap when seeking to determine systematically their evolution over the medium term. With this as background, Chapter 11, "The Determinants of U.S. Real Interest Rates in the Long Run" (1990) by Sharmini Coorey, attempts to shed light on this topic. The paper analyzes the time series processes followed by real interest rates and tests for comovements with other variables that, according to economic theory, would be expected to influence real interest rates. Economic theory suggests that real interest rates are determined by the balance of saving and investment and hence by the variables that influence these two aggregates. The empirical results of the paper indicate that real interest rates are cointegrated with fiscal deficits, government debt, private sector wealth, real money balances, demographic factors, and the productivity of capital. The estimated long-run influence of demographic and fiscal variables is particularly noteworthy because this result has usually been difficult to establish empirically. An important implication of the econometric evidence is that prospective demographic trends in the form of an aging in the population structure would tend to place upward pressure on real interest rates over the long run. In addition, while reductions in fiscal deficits would have an appreciable negative impact on real interest rates, the extent of their decline would be magnified or mitigated by whether or not such reductions were accompanied by a decline in the stock of government debt in relation to GNP. Given the upward pressure on real interest rates stemming from the aging of the population, fiscal action would have to be of a size large enough to make a significant dent in the debt-GNP ratio if it is to result in an appreciable reduction in real interest rates. Ill. Monetary and Financial Issues

Many of the policy issues that arose in the monetary and financial areas in the course of the last decade had a common thread in that they were related directly or indirectly to high and rising inflation in the latter part of the 1970s and early 1980s. The inflationary environment in this period led to an acceleration in the pace of financial innovation as providers of financial services sought to meet customers' needs to protect themselves against the harmful effects of inflation. Also, as inflation tended to magnify distortions caused by financial market regulations-such as those that mandated interest rate ceilings and that limited the range of

10 • Overview

activities depository institutions were allowed to undertake-the trend toward deregulation was set in train toward the end of the 1970s. Financial innovation and deregulation in turn had a pervasive influence on the way wealth owners managed their portfolios, in particular the way they managed monetary assets, leading to shifts in money-demand functions. These developments affected the way monetary policy came to be conducted. As doubts grew about the stability and predictability of money demand-the sine qua non of monetary targeting-the Federal Reserve began to rely less on the monetary aggregates and more on a broad range of financial and economic variables to guide the implementation of monetary policy. This trend culminated in the Federal Reserve's decision in 1987 to cease specifying an annual growth target range for Ml. As discussed in Chapter 12, "Broad Money Growth and Inflation" (1990) by Liam P. Ebrill and Steven M. Fries, however, recent research by the staff of the Federal Reserve pointing to an apparent long-run relationship between M2 and the price level, the so-called P* relationship, has revived interest in the possibility of using monetary aggregates as a guide to monetary policy. In this research, P* is defined as the price level that would prevail given the existing stock of M2, assuming that M2 velocity is at its long-run average level and the economy is operating at its estimated potential. The short-run deviations of the actual price level from P* can be decomposed into a velocity gap and an output gap. The Federal Reserve staff's analysis found that inflation showed a tendency to accelerate when the actual price level was below P* and that the velocity and output gaps were individually significant, with approximately equal coefficients, in explaining the short-term behavior of inflation. Chapter 12 examines the results of the Federal Reserve staff's research. The paper's orientation is technical in that a central issue addressed is whether the Federal Reserve staff's specification of the inflation equation in first difference form is statistically appropriate, but the conclusions it draws carry practical implications for the assessment of the usefulness of the P* relationship as a guide to monetary policy. The paper's empirical analysis indicates that an inflation equation in levels (rather than in first differences) may be more appropriate given the orders of integration of the explanatory variables used and reveals that the significance of the M2 velocity gap is not robust in this alternative specification. While not denying the existence of a long-run relationship between M2 and the price level, the paper suggests that the primary channel through which monetary policy influences inflation in the short run is the output gap. Inflation in the 1970s (and its aftermath) significantly affected the U.S. financial scene during much of the 1980s not only through its effects on the pace of financial innovation and deregulation but also through its

Yusuke Horiguchi, Uam P. Ebri/1, and Charles Adams • 11

disruptive effect on the performance of financial institutions, especially those that traditionally had engaged in substantial maturity transformation. Important examples of such institutions are savings and loan associations which, as a result of tax incentives and regulation then in place, held their assets primarily in the form of long-term, fixed-rate mortgages funded mostly by short-term deposits. Because of this maturity mismatch, savings and loan associations had to contend with considerable interest rate risk. Chapter 13, "An Expensive Thrift Industry" (1989) by Steven M. Fries, analyzes the plight of savings and loan associations and recent legislation aimed at reforming the industry. The savings and loan crisis that came under the political spotlight in the late 1980s can be traced back to the late 1970s and early 1980s, a period characterized by the acceleration of inflation and the sharp tightening of monetary conditions. The run-up in short-term interest rates, to peak levels of over 20percent in 1981-82, had such a devastating effect on the operating performance of savings and loan associations that, by the time greater stability of interest rates returned, the industry taken as a whole had been left with virtually no private capital. The Congress, state legislatures, and thrift regulators responded to the difficulties the industry faced by expanding the asset and liability powers of savings and loan associations while extending the scope of deposit insurance. This response, however, laid the basis for the re-emergence of widespread thrift losses in the later 1980s. Insolvent or significantly undercapitalized savings and loan associations that were permitted to continue their independent operations tended to assume greater risks, in some cases pursuing aggressively their expanded asset power to take full advantage of federal deposit insurance. In response to the rapidly worsening situation, in early 1989 the Administration proposed a comprehensive reform plan for the industry, which culminated in the enactment of the Financial Institution Reform, Recovery, and Enforcement Act of 1989 (FIRREA). The principal provisions of FIRREA included the establishment and initial funding of the Resolution Trust Corporation (RTC) to reorganize economically insolvent institutions; the reform of the regulatory structure of the savings and loan industry to parallel that of the banks, notably by raising minimum capital standards; the creation of the Savings Association Insurance Fund to replace the insolvent Federal Savings and Loan Insurance Corporation; and requiring savings and loan associations to hold an increased portion of their assets in mortgages. The paper concludes that on balance FIRREA represented a step toward the attainment of the Administration's goal of resolving the savings and loan crisis.

12 • Overview

However, there are questions whether the funding for insolvency resolutions would be adequate and whether the desire to preserve a separate housing finance system by increasing the proportion of assets to be held as mortgages would in effect weaken the industry, countering the positive impact of the other structural reforms.

IV. Current Account Adjustment and the Exchange Rate A key concern for policymakers during the 1980s was the emergence of large current account imbalances between the United States and its trading partners. Although the U.S. current account deficit, which peaked at 3Y4 percent of GNP in 1987, could be viewed as a manifestation of the U.S. national saving problem rather than a separate problem on its own, the unprecedented size and persistence of the imbalance was a source of tension in financial markets during much of the 1980s and may also have contributed to an upturn in protectionist pressure. At the same time, the initial failure of the U.S. current account deficit to narrow following the major decline in the dollar's external value beginning in 1985 gave rise to concerns about the efficiency of exchange rate changes in the international adjustment process. Some attributed the lack of current account adjustment to the unusual reaction of trade prices to exchange rate changes. More recently, as the U.S. current account deficit began to narrow substantially-and perhaps by more than predicted by conventional models-doubts have arisen whether these models capture adequately the supply responses to exchange rate changes. With this as background, the papers included in Part IV of the volume attempt to shed light on the nature of the current account adjustment process and the factors behind the movements in the dollar during the 1980s. Chapter 14, "A Dynamic Error Correction Model of the U.S. Current Account" (1990) by Ellen M. Nedde, outlines a quarterly model of the U.S. current account developed by the IMF staff in connection with its projections for the World Economic Outlook. The model is based on an error-correction methodology and relates trade flows and prices to variables such as current and lagged levels of economic activity at home and abroad and relative prices. This approach allows the data to identify more general lag structures than found in many conventional models and permits an evaluation of the roles of demand and supply factors in the current account adjustment process. In order to evaluate the model's ability to capture possible supply effects associated with exchange rate changes--effects that received a great deal of attention in the late 1980s-the paper includes several tests

Yusuke Horiguchi, Uam P. Ebri/1, and Charles Adams • 13

ofthe model's tracking ability. The main conclusion is that, notwithstanding the attention given to supply variables in attempts to explain the current account, standard variables found in conventional models-such as relative prices and economic activity-do a relatively good job in explaining the behavior ofthe U.S. current account after 1985, provided sufficient account is taken of lags in the adjustment process. Chapter 15, '''The Recent Behavior of U.S. Trade Prices" (1988) by Daniel Citrin, considers whether the increase in U.S. trade prices following the depreciation of the dollar after 1985 fell short of what would have been expected on the basis of historical relationships. There was-and perhaps still is-a view that the increase in U.S. trade prices after 1985 was surprisingly moderate given the significant decline of the dollar and that the lack of "pass-through" of exchange rate movements might have played a role in delaying current account adjustment. The paper notes, however, that much of the subdued behavior of prices as measured by the national accounts deflators for nonagricultural exports and non-oil imports after 1985 was attributable to a marked decline in computer prices and the growing importance of computer trade. Using fixed-weight price indices for exports and imports and manufactures-inclusive and exclusive of computer prices-the paper concludes that when allowance is made for the behavior of computer prices the adjustment of U.S. trade prices after 1985 was not out of line with experience, implying that there was no evidence that the lack of pass-through of exchange rate changes was delaying external adjustment. Chapter 16, "Structural Models of the Dollar" (1990) by Charles Adams and Bankim Chadha, explores various explanations for the U.S. dollar's movements during the 1980s in view of the rather poor performance of empirical exchange rate models. The paper analyzes the time series processes followed by dollar exchange rates and the extent to which any nonstationarity in exchange rates might be accounted for by the nonstationarity of economic variables included in empirical exchange rate models, such as money supplies, interest rates, fiscal and current account imbalances, as well as stocks of interest-earning assets denominated in different currencies. The main conclusion is that the time series processes followed by nominal exchange rates contain some predictable components and deviate from pure random walks in which exchange rates change unpredictably with new information. Nevertheless, given the amount of variability in exchange rates and the limited number of observations from the recent floating rate period, there is a basic difficulty in establishing the statistical significance of these deviations. When tests are carried out to determine whether and to what extent any nonstationarity in exchange rates can be explained by other economic variables, the

14 • Overview

paper finds a small number of cointegrating relationships but deviations from these long-term relationships play only a limited role in accounting for the short-term swings in the dollar during the 1980s. V. Structural Issues

While the U.S. economy is generally considered to be one of the most flexible among the major industrial countries, there are nonetheless a number of structural impediments to the efficient functioning of markets. For example, as noted earlier, several features of the current tax system distort the level and allocation of saving. Other examples include a range of imperfections affecting the market for the provision of health care that are causing efficiency losses and tending to raise the costs for providing health care; the differential tax treatment of debt and equity that heavily influences U.S. corporations' financing policy; government policies toward the farm sector that distort the production, consumption, and international trade in agricultural products; and the long-standing protection accorded to certain industries-such as steel and textiles-which have efficiency costs. At the same time that there are those concerns about undesirable government intervention in the market place, there also are growing concerns about the possible lack of government efforts in certain areas such as infrastructure and education. The volume contains three papers dealing with some of these structural issues. Chapter 17, "The U.S. Health Care Industry: Performance and Issues" (1989) by Liam P. Ebrill, notes that the share of health expenditures in the United States has risen from almost 6 percent of GNP in 1965 to 11 percent in 1986, and is projected-in the absence of reform of the health care sector-to rise to 15 percent by 2000. The paper argues that the health care sector has been influenced by a range of impediments to the functioning of the price mechanism that have tended to impart an upward bias to health care costs and have caused efficiency losses in the provision of medical services. These imperfections arise mainly from informational asymmetries, moral hazard problems associated with medical insurance, and a lack of competition in the market for health care. The main policy conclusion of the paper is that while reforming the health care sector would raise a number of difficult issues-including notably finding ways to resolve the dilemma created by the trade-off between the risk spreading afforded by medical insurance and establishing appropriate incentives to restrain expenditures-some features of government policy that are exacerbating the problem would need to be

Yusuke Horiguchi, Uam P. Ebri/1, and Charles Adams • 15

addressed promptly. These features include the current practice of allowing employer's contributions to health care costs to be excluded from individual taxable income even as those contributions are a deductible expense for the employer, thereby encouraging an excessive level of health insurance and implying significant revenue losses for the Government. The paper argues that other initiatives worthy of consideration include further reform of the malpractice and jury award system and addressing the reasons underlying the increasingly burdensome intensivecare expenditures in the United States. The separate tax treatment of corporate and personal incomes under current U.S. tax law is widely viewed as influencing a range of economic decisions, including the choice between debt and equity finance, the choice of legal form of business, the decision on whether or not to retain or distribute earnings, and the decision on how much to invest. Chapter 18, "Implications of Integrating U.S. Corporate and Individual Income Taxes" (1990) by Krister Andersson, considers the possible implications for economic efficiency and for federal revenues of integrating the corporate and personal income taxes in the United States. After briefly reviewing recent developments in the corporate sector-and the increased use of debt financing in particular-the paper evaluates the implications of alternative integration schemes. The paper concludes that the integration of corporate and individual income taxes could reduce substantially the effective tax rate on equity capital and lead to rather large gains in efficiency. In addition, while integration-and, particularly, the abolition of the double taxation of dividends--could result in revenue losses to the Government of the order perhaps of $35-40 billion a year, these losses could be reduced and possibly eliminated if some interest payments were made no longer deductible when calculating the corporate income tax. The paper notes, however, that even though the integration of corporate and personal income taxes would tend to raise economic efficiency, other considerations such as possible increase in administrative complexity would also need to be taken into account in any further reform of the tax system. In the decade of the 1980s there was a significant reduction in the real value of transfers from the Federal Government to state and local governments, and this development occurred against the background of growing concerns over the adequacy of the stock of infrastructure capital and the effectiveness of the education system, two areas that over time have been the shared responsibility of all levels of government. The final chapter of the volume, "Developments in Federal-State-Local Fiscal Relations: Implications for Infrastructure, Investment, and Education" (1990) by Liam P. Ebrill, reviews recent trends in these specific areas with a view

16 • Overview

to evaluating prospective needs and to determining the relative roles of the various levels of government in meeting those needs. As regards the adequacy of infrastructure, the paper concludes that in some cases increased investments would be needed just to maintain current standards. Less clear is the Federal Government's responsibility in this regard, particularly since much infrastructure appears to be in the nature of local public goods. Nevertheless, given a rapid deterioration in state finances and a secular decline in real grants-in-aid, some reassessment of the Federal Government's role might be appropriate. As regards education, while there is a strong case for arguing that the U.S. system is troubled, the paper questions whether a greatly enhanced federal financial role would necessarily be the appropriate response. The evidence indicates that greater financial resources in the aggregate might not achieve much and the theory of fiscal federalism would tend to suggest that most jurisdictional questions can be handled at the state-local level. Attention might need to focus more on the microeconomics of the industry, including ensuring that school districts face appropriate incentives.

PART I

National Savings and Public Policy

National and Personal Saving Measurement and Analysis of Recent Trends A. LANS BOVENBERG

AND

OWEN EVANS

T

he low level of national saving in the United States has generated widespread concern because it has reduced the prospects for future U.S. income growth. It is also seen as having contributed to a sharp widening of the U.S. trade deficit in the 1980s. Although the fall in public saving associated with a widening budget deficit has been widely publicized, the private sector accounted for about an equal share of the fall in national saving, with a declining trend in personal saving explaining most of the weakening of the private saving performance. This paper examines various issues relating to national and personal saving in the United States. Section I discusses how different components of national saving have evolved since 1950. Section II addresses several issues regarding the measurement of saving and concludes that the observed declines in national, private, and personal saving rates cannot be attributed to measurement problems. Section III examines factors that seem to have been behind the decline in U.S. personal saving. It suggests that improvements in wealth positions, in the living standards of the elderly, in social security pensions, and in private and public insurance mechanisms may all have contributed to the declining trend. Demographic factors and structural changes in capital markets also appear to have played a role. Section IV presents an empirical model of personal saving, in which demographic variables, the real after-tax interest rate, and the rate of This paper was published in Staff Papers, International Monetary Fund, Vol. 37 (September 1990), pp. 636-69. The authors are especially grateful to Yusuke Horiguchi and George Kopits for helpful comments.

19

20 • National and Personal Saving

inflation are important. According to simulation experiments, the changing demographic structure of the U.S. population appears to have played an important role in the declining personal saving rate in the 1980s. The rising ratio of wealth to disposable income and the decline in inflation were also significant, whereas the increase in the real interest rate mitigated the decline. Section V provides the conclusions. I. Recent Trends in Saving

This section discusses recent trends in the most commonly used saving measures, which are derived from the National Income and Product Accounts (NIPA). Table 1 presents net saving relative to net national product (NNP) for the decade averages over 1950-89 and also annually since 1980. Net national saving comprises net public and private saving, with the latter consisting of personal and corporate saving. The state and local government surplus and the federal government surplus add up to total public saving. The NIPA net national saving rate fell substantially in the 1980s to an average 3V4 percent from about 8V4 percent of NNP during 1950-79. Although rising dissaving by the public sector played an important role, falling private saving also contributed to the weakening national saving performance. The public and private sectors each accounted for about half of the decline in the NIPA net saving rate from 1950-79 to the 1980s. 1 The decline in the national saving rate was reflected in the widening imbalance between national investment and national saving, and in the increased reliance on foreign saving, which reached a peak of almost 4 percent of NNP in 1987. Moreover, whereas private saving traditionally provided sufficient funds to finance private investment in the post-1950 period, the private saving-investment surplus fell after 1982 and turned negative in 1987 and 1988. A declining trend in the personal saving rate explained most of the fall in the private saving rate. The net personal saving rate as a percentage of NNP declined from its historical average of about 5Yz percent in the beginning of the 1980s to about 3 percent in 1987-88, before increasing somewhat in 1989. Net corporate saving also declined in the 1980s compared with earlier decades. Gross private saving as a share of gross national product (GNP) is presented as a memorandum item in Table 1. In contrast to the share of net private saving in NNP, which started to fall in the mid-1970s, gross 1

The decade of the 1980s covers the years 1980-89.

A. Lans Bovenberg and Owen Evans • 21

private saving as a proportion of GNP has remained relatively stable within a range of 15\/z to 18 percent during the 1950-85 period. Nevertheless, after reaching a postwar high by the end of the 1970s, this saving measure has also declined recently, especially since 1984, averaging 15 percent of GNP in the period 1987-89-the lowest levels since the 1940s. Diverging trends in gross and net measures of saving point to important changes in depreciation rates. In fact, the share of depreciation in GNP rose significantly at the end of the 1970s and the early 1980s. This rising trend was due in part to a slowdown in productivity growth, which raised the capital-output ratio. More importantly, it reflected a shift in the pattern of nonresidential investment away from longer-lived assets, such as structures, toward shorter-lived assets, especially information processing equipment. 2 Although technological change may have contributed to the change in the composition of investment toward shorter-lived assets, taxation played a significant role as well. Several studies suggest that tax rules, especially when the rate of inflation was high, have typically favored equipment over structures. 3 Although net saving is the relevant concept for measuring the accumulation of wealth and the intertemporal allocation of consumption, gross saving is also an important measure. In particular, a higher rate of gross saving may lead to more rapid economic growth if technical change is embodied in new capital goods or if introducing new capital generates significant learning-by-doing effects. Moreover, gross saving measures tend to be more reliable than net saving measures, because, in contrast to measures for net saving, they do not rely on estimates for depreciation, which are subject to various statistical and conceptual problems. Nevertheless, the rest of this paper focuses on the net saving concept in view of its focus on the accumulation of wealth and intertemporal consumption decisions. II. Alternative Saving Measures

Aghevli and others (1990, Appendix I) deal with several issues regarding the measurement of national saving and its components in an inter2 Corker, Evans, and Kenward (1992; this volume) and Evans (1989) discuss in more detail recent movements in the composition of corporate investment, including the increase in the investment share of assets with relatively low service lives. See also de Leeuw (1989). 3 See, for example, Bosworth (1984), Gravelle (1983), and King and Fullerton (1984). Other studies have argued that additional uncertainty associated with higher and more variable inflation rates and anticipated larger fiscal deficits have raised long-term real interest rates relative to short-term real rates, thereby shifting the investment mix toward shorter-lived assets.

Saving and Investment

0.3 1.3

0.3

1.1

0.7

-

0.1

Net foreign investment

7.6 3.7 3.0 0.9

0.8

1.1

7.7 3.7 2.9

0.9 -1.9

-1.1

7.9 8.9 6.2 2.7

-0.1

8.2 3.4 3.8 1.0

Total net investment Plant and equipment Residential construction Inventory accumulation

-0.3

--

8.6 8.9 5.1 3.8 -0.3

Saving-investment surplus National saving-investment surplus Private saving-investment surplus

8.1 8.2 5.2 3.1 -0.1 -0.3 0.1

-1.8

1.1

-1.8

0.4

...

5.1

...

3.3 6.2 4.3 1.9 -2.9 -1.4 -4.3

0.4

1.3

1.7

0.5

0.2

-

3.9

-

7.4 2.7 2.7 2.0

5.8 2.9 2.6 0.3

5.3 2.0 3.1 0.2

5.3 1.6 2.9 0.6

5.4 2.0 2.7 0.6

0.6

4.7

0.3

0.6

0.2 -0.8 -0.2

0.7 -1.1 -2.7 -3.2 -3.6 -3.7 -2.7 -2.1

3.0

-1.3 -2.9 -3.1 -3.6 -3.5 -2.4 -1.5

6.2 2.3 3.5 3.6 2.4 1.5 1.7 0.8 2.2 0.9 -0.9 -0.2

0.3

5.5 3.7 2.2 -0.3

5.8 6.4 2.3 2.2 4.6 2.7 1.7 1.7 2.9 3.2 7.2 7.5 6.2 6.5 7.7 6.4 5.6 4.4 5.2 5.4 5.6 5.9 5.5 4.3 4.9 3.5 3.3 2.5 3.3 4.4 1.6 1.6 0.7 2.2 2.8 2.9 2.2 1.9 1.8 1.1 -1.4 -1.1 -4.0 -4.3 -3.1 -3.7 -3.8 -2.7 -2.2 -2.4 1.1 1.3 1.3 1.6 1.9 1.8 1.7 1.3 1.1 0.9 -2.5 -2.4 -5.2 -5.8 -5.1 -5.5 -5.5 -4.0 -3.3 -3.2

(In percent of NNP)

1950-59 1960-69 1970-79 1980-89 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989

Net national saving Net private saving Personal saving Corporate saving Net public saving State and local surplus Federal surplus

Measure

Table 1.

"tJ

lio share (right scale)

38

36 34

0

32 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987

market instruments, and pension fund reserves; and stocks are defined to include corporate equities and equity in noncorporate business. 21 The share of the portfolio invested in housing increased from about 30 percent in the late 1960s to 40 percent in the late 1970s. The rate of return on housing increased during this period (Figure 2). However, in the early 1980s, the rate of return on housing decreased significantly and in some years it was even negative. The share of household wealth in housing declined during this period but resumed its rising trend when the rate of return started to increase again. As expected, the share responds only slowly to changing rates of return. IV. Calibration of the Model

Assuming that the average rates of return and levels of risk obtained in the previous section are representative of what an investor may expect, it is possible (by minimizing the risk for different expected rates of return) 21 Another method of calculating portfolio shares would involve narrower definitions of each of these three asset categories and obtaining the shares by dividing each of those by a lower total wealth. However, the shares of assets would not have changed much if the alternative approach had been taken.

180 • Investment in Housing Table 3.

Shares in the Total Portfolio in 1987 and in the Model with a Required Rate of Return of 6.95 Percent

Share

Housing

Bonds

Stocks

1987 Model

36.5 40.3

37.2 36.8

26.3 22.9

to trace the frontier of efficient portfolios. The risk increases with the required rate of return. The portfolio with the least risk (which involves practically no risk at all) would consist of 94.5 percent bonds, Yz of 1 percent stocks, and 5 percent housing, and would have yielded a rate of return of 4. 78 percent. The actual portfolio held by the average U.S. investor in 1987 was very different from the least-risk portfolio (Table 3). As the investor requires a higher rate of return, the composition of the portfolio shifts in favor of housing and stocks. With increasing shares invested in stocks, an investor can achieve a very high rate of return but the risk associated with such a portfolio will be large compared with portfolios with a lower rate of return. Figure 3 shows how the allocation of the portfolio changes as the required rate of return increases. At a required rate of return of 6.95 perFigure 3. Share of Assets in the Portfolio-Base Case (In percent)

Share

7or-----------------------------------------------~

...

60 50

40 30 20

.... ···

Housing

.....................··· .........

...... ...... ......

.. ·.. ······ .. ·· ...

......

.. ··

•• •••••••

••• •••

-- _____ .-;,;;.

~ ~

··...

~

··~

~~

~~

··...•,

10 0~--~----~-----L----~----~~--~--~----~ 5.5 6 6.5 7.5 7 8 9 9.5 Rate of return

Krister Andersson • 181

cent, the composition of the observed portfolio and the risk-minimizing portfolio is rather similar. V. The Tax Deductibility of Mortgage Interest Payments

This section discusses how limiting the deductibility of mortgage interest payments could bring about a more efficient allocation of saving and analyzes the nature of the adjustments that might be involved. Such a tax change would initially result in capital losses for home owners, an adjustment that would be required to equalize the after-tax rates of return across alternative assets (adjusted for risk). The decline in the relative price of housing then would induce the response of the supply of resources to the housing sector. Investment in that sector would be discouraged and the stock of housing would be progressively reduced compared with what it otherwise would have been. The process would involve a recovery of the price of housing and would continue until the price returns to a level that covers resource costs. The forces driving the adjustment can be considered by analyzing how an individual investor would react to the change in the tax code, holding other factors constant. On the basis of equation (2) in the portfolio model at hand, the investors would then find that the rate of return on housing has decreased by 2.2 percentage points from 7 percent to 4.8 percent. 22 On the basis of this information, new shares in the portfolio with minimum risk can be found corresponding to each expected rate of return. In considering how investors might react to this changed environment, two polar cases can be distinguished. One is the case in which the investor requires the same rate of return as before the change despite the fact that one asset, housing, now yields a considerably lower rate of return. To get that rate of return, the investor must hold more of the risky asset, that is, stocks, and the risk of the overall portfolio increases. The other polar case is one in which the investor requires the same risk level as before but will accept a lower rate of return. However, the investor does not fall into either of these polar cases but he is willing to trade off rate of return for a lower risk level and find a risk-return combination that is superior to the two polar cases. 23 22 When simulating the change in tax policy, the variance-covariance matrix is assumed to be unchanged. However, re-estimating the variance-covariance matrix with the new rates of return on housing does not lead to any significant chc;tnges ofthe result (the change in the share of the housing is within 1 percentage pomt). 23 By assuming that the investor could have chosen any other portfolio, he reveals his preference for risk versus rate of return. Using this information for the relevant interval of rates of return, a superior combination between the two polar cases can be found.

182 • Investment in Housing

Figure 4. Demand for Housing1 (In percent)

1With

and without mortgage deductibility against federal income taxes.

The result from the model simulation shown in Figure 4 indicates that the trade-off between rate of return and risk leads to the new optimum portfolio that yields the rate of return of 6.7 percent (down from 6.95 percent before the change in the tax code but higher than a rate of return of 6.24 percent that the investor would have to accept if he wished to keep the risk level unchanged). The share of housing decreases from 40.3 percent to 26.5 percent, 24 while the share invested in stocks increases from 22.9 percent to 33.9 percent. The share of bonds also increases somewhat. The specific results that have been derived are based on a portfolio model in which the individual looks only at the rate of return and risk but has no other preferences for any particular asset. There are, however, reasons for believing that the share of housing would decrease by some24

An alternative approach, assuming that the tax effect is fully capitalized into the value of housing, results in a decrease of the existing housing stock from 36.4 percent to 25.5 percent, that is, by some 30 percent.

Krister Andersson • 183

what less because of a variety of general equilibrium effects and, in addition, because there are other-nonmonetary-returns connected with investment in housing: the benefits of living in a certain area, the correlation with the rental market, and so on. If the assets are imperfect substitutes, that is, individuals have preferences for certain assets beyond their risk/return characteristics, they would be willing to accept a lower rate of return or a higher risk level to have that asset in their portfolio. Section VII analyzes the implications of such preferences.

VI. Distributional Effects

On the basis of portfolio considerations, the share allocated to housing would be relatively small for an investor who requires a low rate of return, while it would be considerably larger for an investor demanding a relatively high rate of return (see Figure 3). Figure 5 shows what percent of the original housing share is maintained after mortgage interest deductibility is eliminated for investors with different required rates of return. The investor requiring a low rate of return would retain almost 80 percent of his initial housing share while the investor requiring a higher rate of return would only retain about 65 percent of his initial housing share. Figure 5. Relative Share of Housing to Its Previous Level (100=Unchanged level) Relative housing share (in percent)

80

75

70

65

60L---~----~-----L-----L-----L-----L----~----~

5.5

6

6.5

7

7.5

8

8.5

Required rate of return on total portfolio (in percent)

9

9.5

184 • Investment in Housing

Figure 6 shows how the shares would change for all three assets for an investor requiring a 6 percent or an 8 percent rate of return. The fact that different categories of investors will shift away from housing to different extents may mean that the elimination of the deductibility of mortgage interest payments would affect expensive and inexpensive housing units differently. When individuals demanding high rates of return (typically high-income earners) shift from housing to other assets, they are likely to demand less expensive housing units. Consequently, the demand for high-priced housing units is likely to fall and the price to drop, while the demand by these individuals for low- and medium-priced housing may increase relative to high-priced housing. According to the model, those investors already buying low- and medium-priced housing will not Figure 6. Shares of Assets in Total Portfolio at Different Required Rates of Return

Krister Andersson • 185

shift as much away from housing investment. Overall, the increased relative demand for low- to medium-priced housing may partly offset the decrease in the rate of return on such housing. Thus, the effect of the change in the tax rules on housing prices may be largest for expensive houses. VII. Imperfect Substitutability of Assets

This section introduces the possible role of asset characteristics other than risk and return. An individual's preferences for different attributes of assets will determine how willing he is to substitute one asset for another. Assuming for simplicity that all assets can be grouped into two categories, housing, H, and financial assets, FIN, a preference structure can be readily expressed. It may be noted that if there is no substitutability at all between assets, the preference structure is of the Leontief type. A less extreme case is constant expenditure shares, the Cobb-Douglas case. Another case is when the individual has no preferences between the two assets (separate from their risk/return characteristics), that is, they are perfect substitutes. In general terms the preference structure can be formalized as follows:

u = U(H,F/N) = [g. HP + (1 -g). FJNP] 11P, where g is a share parameter and p is an elasticity parameter. How much an individual changes his portfolio will depend not only on these parameters but also on the expected rates of return and risk characteristics of the assets. By assuming that investors are rational, the share parameters can be calibrated from actual portfolio shares (since the rates of return are already known). By assuming a value for p, it is possible to calculate how much the portfolio will shift when the rates of return are changed. 25 Two cases are considered. In the first, pis equal to 0 (an elasticity of substitution of 1), implying that the investor has only limited preference for housing. In this case, the housing share decreases from 40 percent to almost 31 percent (at a required rate of return of 6.7 percent) when the federal tax deductibility of mortgage interest is eliminated. In the other 25 The optimality condition shows that the shares of the assets depend on prices and share parameters as follows:

H =

IR~~NIT ·I (1 ~ gl· FIN,

where Tis equal to 1/(p - 1), g is a share parameter, RH is the rate of return on housing, and RFIN is the rate of return on financial assets.

186 • Investment in Housing

case, the individual has strong preferences for housing. 26 This translates to a p equal to -1 (an elasticity of substitution of 0.5). 27 The housing share in this case decreases from 40 percent to 36 percent when the deductibility of mortgage interest is removed. In the context of the model at hand, imposing a preference structure on the maximization problem results in a higher level of risk compared with the case in which the assets are perfect substitutes. However, for a relatively small increase in the risk level, a portfolio allocation with a larger housing share can be obtained. 28 The main effect on the other shares in the portfolio in this case, as it turns out, is almost a one-to-one trade-off of bonds for housing. Stocks are affected only to a limited extent because of the positive correlation between bonds and stocks in combination with a negative correlation between housing and stocks (Figure 7). The inclusion of imperfect substitutability of assets makes the portfolio model more realistic. An elasticity of substitution between 0.5 and 1 may be plausible, in which case eliminating the deductibility of mortgage interest when calculating the federal income tax would then result in a decrease in the housing share by some 4-9 percentage points. A shift to other assets, primarily corporate stocks, would take place. VIII. Alternative Policy Measures to Promote an Efficient Allocation of Resources

As observed earlier, mortgage interest is deductible in the United States even though the imputed income from housing is not taxed. The fact that interest on loans taken for business purposes is tax deductible, and likely to remain so in the future, could make it difficult to remove the deductibility of mortgage interest from an administrative point of view. Such a tax change would create incentives for tax arbitrage (to transfer loans from mortgages to business loans or to set up businesses whose only purpose would be to enable the home owners to deduct their interest payments). With this in mind, an alternative way to achieve a symmetric tax treatment would be by taxing imputed income from housing. Since most homes are assessed a value for property tax purposes, these values could 26 The individual in this case has a preference structure that is less elastic than the Cobb-Douglas case but more elastic than the Leontief case. 27 McGibany and Nourzad (1988) estimate that the elasticity of substitution between net private capital and long-term government bonds and notes is as low as 0.3. 28 The increase in risk will be larger, the more inelastic the preferences are.

Krister Andersson • 187

Figure 7. Demand for Housing1 (In oercentJ

I

With and without mortgage deductibility against federal income taxes. =Elasticity of substitution between housing and financial assets.

2e

provide a basis for calculating an imputed rental income for tax purposes. The main problem with this approach is that the value of property assessed for property taxes is in many cases far below the market value. 29 Another alternative would be to raise the effective tax rate for capital gains arising from houses. The existing provisions make the effective capital gains tax rate close to zero. 30 One argument for raising the capital gains tax rate on housing is that the preferential tax treatment has increased demand for housing and thereby induced some of the capital gains. Of course, only real, not nominal, capital gains should be subject to a higher effective capital gains tax, and inflationary gains should not be taxed. The general equilibrium effects of such a policy, including a possible increase in labor market immobility, would have to be carefully considered before such a tax change is enacted. Other potential negative effects could include increased lock-in effect of capital gains. 29 3

This is a problem when calculating property taxes as well.

°For a discussion of the capital gains tax treatment, see Andersson (1989).

188 • Investment in Housing

IX. Conclusion The asymmetric tax treatment of different assets, whether intended to achieve social goals or other objectives, is likely in general to hamper efficiency. In the specific case considered here, to the extent that the favorable tax treatment of housing diverts funds from other investment, it induces an inefficient allocation of saving and investment. 31 The results from the mean-variance portfolio model in this paper indicate that the elimination of the federal income tax deductibility of mortgage interest payments would cause individuals to shift their investment from housing to assets with higher returns, largely stocks. The extent of the portfolio adjustment would be greater for those who, other things being equal, are more willing to assume larger risks to achieve higher returns. If individuals have strong preferences for housing, and regard stocks only as an imperfect substitute for housing for investment purposes, the effect would be smaller than in the case of perfect substitutability. Under such an assumption, a decrease in the housing share in the portfolio resulting from the elimination of mortgage interest deductibility for federal income tax purposes is likely to be in the range of 4-9 percentage points. The general equilibrium effects of this policy change would be complicated as all asset prices in the economy would change to some extent. In the short run, the elimination of mortgage interest deductibility could entail some adjustment problems, including those related to capital losses for home owners. These losses would probably be greater for those who have invested in high-priced homes. In the long run, a more efficient resource allocation would be likely to lead to increased production capacity.

ANNEX A Description of Data Used The Rate of Return on Housing

The rate of return on housing is calculated in the following way. The value of houses is represented by the median sales price of existing single family homes, PROPV/2 and the mortgage interest rate is measured as the effective conven31 Although the research on income tax incentives and its effects on housing has taken numerous approaches, the studies have reached remarkably similar conclusions; tax preferences have favored investment in housing, raised the price of houses, and directed resources in favor of housing and away from other capital uses. For a summary of different approaches and studies on home ownership, see Smith and others (1988). 32 Compiled by the National Association of Realtors.

Krister Andersson • 189

tional mortgage rate for all homes. 33 The tax rate used is the marginal income tax rate, TOTMTAX, which is the combined federal and state and local income tax rate. The federal income tax rate is calculated as the average marginal tax rate for a family of four with twice the median income. 34 The state and local marginal income tax rate is calculated as the ratio of state and local personal tax receipts from income taxes to personal income. 35 The debt share, a, is calculated as the share of outstanding mortgages to the market value of the housing stock, PRO PV. 36 The property tax rate, PRO PTAX, is calculated as the ratio of property tax revenues to the market value of housing. The transactions cost connected with the purchase and sale of housing, TCOST, is estimated to have been in the range of 1 percent to 2 percent on an annual basis. The average holding period of a house is assumed to be seven years. 37 Taking into account that the estimated capital gains also include the increase in the value of the house due to improvements of the house, the total cost, including transactions costs, is estimated to be 2 percent on an annual basis. The Rate of Return on Bonds

The net of tax nominal rate of return on the least risky asset, bonds, is calculated as a weighted average of the rate of return on bank deposits, BDEP; the yield per annum on treasury securities at constant maturity of ten years, TSEC; 38 and the yield per annum on triple A corporate bonds, CORPBOND. 39 The rate of return on the composite "safe" asset can be written: (0.1· CORPBOND + 0.25 · TSEC + 0.65 · BDEP) · (1 - TOTMTAX). The Rate of Return on Stocks

The dividend price ratio, DIV, is obtained from Standard and Poor's 500 composite series, taking into account the value of the stock at the beginning of the period. 40 Capital gains, CAPGAIN, are calculated from the annual change of Standard and Poor's index. An accrued capital gains tax rate, CGTAX, is applied to capital gains, 41 while losses are assumed to be deductible against the 33

Compiled by the Federal Home Loan Bank Board. From the Office of Tax Analysis, U.S. Department of the Treasury. 35 From U.S. national accounts sources. 36 The share has varied from 41 percent in 1972 to 36.1 percent in 1978 and 1981 and up to a maximum of 43.9 percent in 1987. Source: balance sheet of the U.S. economy, Board of Governors of the Federal Reserve System. 37 Based on surveys by the National Association of Realtors. 38 Measured as the effective interest rate on passbook savings in the Federal Reserve's quarterly model of the U.S. economy. 39 Measured as the yield on Moody's triple A corporate bonds-Federal Reserve Board, Supplement to Banking and Monetary Statistics, Section 12, Statistical Release G .13. 40 Average of Wednesday's figures from Standard and Poor's "Current Statistics" and "Outlook." 41 Capital gains are taxed when realized. The accrued capital gains tax rate has been obtained from Fullerton and Karayannis (1987). For the period from 1972 34

190 • Investment in Housing statutory capital gains tax rate. 42 Therefore, the rate of return on stocks can be summarized as (1 - TOTMTAX) · DIV + (1 - CGTAX) · CAPGAIN.

REFERENCES Andersson, Krister, "Tax Reforms in Scandinavia During the 1980s," paper presented to the American Economic Association Annual Meetings in New York (1988). ---,"Implications of a Lower Capital Gains Tax Rate in the United States," IMF Working Paper No. 89/100 (Washington: International Monetary Fund, 1989). Bossons, J., "Housing Demand and Household Wealth: Evidence for Home Owners," in Urban Housing Markets: Recent Directions in Research and Policy, ed. by L. Bourne and J. Hitchcock (Toronto: University of Toronto Press, 1978). Ebrill, Liam P., "Some Microeconomics of Fiscal Deficit Reductions: The Case of Tax Expenditures," see Chapter 6 in this volume. ---,and Uri M. Possen, "Inflation and the Taxation of Equity in Corporations and Owner-Occupied Housing," Journal of Money, Credit, and Banking, Vol. 14 (February 1982), pp. 33-47. Frankel, Jeffrey, and Charles M. Engel, "Do Asset-Demand Functions Optimize over the Mean and Variance of Real Returns? A Six-Currency Test," Journal of International Economics, Vol. 17 (November 1984), pp. 309-23. Fullerton, Don, and Marios Karayannis, "The Taxation of Income from Capital in the United States 1980-1986," NBER Working Paper No. 2478 (Cambridge, Massachusetts: National Bureau of Economic Research, 1987). Hendershott, P., and J. Shilling, "The Economics of Tenure Choice, 1955-1979," in Research in Real Estate, Vol. 1, ed. by C.F. Sirmans (Greenwich, Connecticut: JAI Press, 1982). Ibbotson, R., and L. Siegel, "Real Estate Returns: A Comparison with Other Investments," Journal of the American Real Estate and Urban Economics Association, Vol. 12 (1984), pp. 219-42. Irwin, S., and D. Landa, "Real Estate, Futures, and Gold as Portfolio Assets," Journal of Portfolio Management (Fall 1987), pp. 29-34. Kendrick, D., and A. Meeraus, "General Algebraic Modeling System (GAMS)" (Washington: World Bank; and Austin, Texas: University of Austin, 1985). Lehmussaari, Olli-Pekka, "Exchange Rate Uncertainty and the Management of Official Reserves," D:64 (Helsinki: Bank of Finland, 1987). to 1979, the statutory capital gains tax rate was unchanged (28 percent) and the accrued capital gains tax rate of 1980 has been applied. 42 Capital losses may be used to offset capital gains and to some extent ordinary income (up to $3,000).

Krister Andersson • 191

Manchester, Joyce M., and James Poterba, "Second Mortgages and Household Saving," NBER Working Paper No. 2853 (Cambridge, Massachusetts: National Bureau of Economic Research, 1989). McGibany, James M., and Farrokh Nourzad, "The Substitutability of Real Capital and Financial Assets," Applied Economics, Vol. 20 (November 1988), pp. 1445-51. Meeraus, A., "An Algebraic Approach to Modeling," Journal of Economic Dynamics and Control, Vol. 5 (1983), pp. 81-108. Mills, E. S., "Dividing Up the Investment Pie: Have We Overinvested in Housing?" Business Review, Federal Reserve Bank of Philadelphia (March-April 1987), pp. 13-23. Neubig, Thomas S., and David Joulfaian, The Tax Expenditure Budget Before and After the Tax Reform Act of 1986, OTA Paper No. 60 (Washington: Office of Tax Analysis, U.S. Department of the Treasury, 1988). Organization for Economic Cooperation and Development, Urban Housing Finance (Paris, 1988). Rosen, Harvey S., and Kenneth T. Rosen, "Federal Taxes and Homeownership: Evidence from Time Series," Journal of Political Economy, Vol. 88 (February 1980), pp. 59-75. Slemrod, Joel, "Down-Payment Constraints: Tax Policy Effects in a Growing Economy with Rental and Owner-Occupied Housing," Public Finance Quarterly, Vol. 10 (April 1982), pp. 193-217. Smith, Lawrence B., Kenneth T. Rosen, and George Fallis, "Recent Developments in Economic Models of Housing Markets," Journal of Economic Literature, Vol. 26 (March 1988), pp. 29-64. Summers, Lawrence H., "Inflation, the Stock Market, and Owner-Occupied Housing," American Economic Review, Papers and Proceedings, Vol. 71 (May 1981), pp. 429-34. Zerbst, R.H., and B.R. Cambon, "Real Estate: Historical Returns and Risks," Journal of Portfolio Management (Spring 1984), pp. 5-20.

Fiscal and Economic Impact of Federal Credit Programs STEVEN M. FRIES

T

he U.S. Government's involvement in the allocation of credit to the private sector is extensive and in some forms is expanding rapidly. Through direct loans, loan guarantees, and activities of governmentsponsored enterprises-GSEs (see Section I), the Government aims to increase credit availability to certain target groups-such as home buyers, students, agricultural enterprises, small businesses, and exporters. However, in the wake of the savings and loan crisis and bailout of the Farm Credit System, as well as evidence of rising losses in federal mortgage insurance programs, all areas in which the Government may be exposed to significant financial risk have come under increased scrutiny. A priority in this regard is federal credit reform to improve control of program risks and to target resources more effectively. 1 While the form of federal credit programs varies, their fiscal and economic impact arise primarily from each program's subsidy component. For a direct loan, a subsidy exists to the extent that the credit terms offered by the Government are more favorable than those available from a private lender, provided that the Government's cost to originate and service the loans equals that of a private lender. A guaranteed loan conveys a subsidy when the Government charges a fee for its commitment to pay all or part of the loan principal and interest in the event of default that is less than the amount implied by actuarial considerations. A GSE receives a subsidy because its borrowing costs do not fully reflect the riskiness of its asset portfolio and contingent commitments. In each case, 1 Other areas in which the Government may be exposed to considerable financial risks are the deposit insurance system and the insurance of private pension plans. The former is examined in Fries (1992; this volume).

192

Steven M. Fries • 193

it is the subsidy component that determines the program's long-run fiscal impact in terms of the Government's net claim on private saving. Moreover, the program's impact on economic efficiency stems from the extent to which the subsidies alter the allocation of capital. Under current budgetary practices, however, the cash flows of federal credit programs, rather than their subsidy components, are emphasized. 2 Because short-run cash flows bear little or no relationship to credit program subsidies, this practice can render ineffective the Government's control of credit programs. For example, the Farm Credit System, a GSE that provides loans to agricultural enterprises, required an infusion of federal funds in 1988 to cover losses on its loan portfolio. 3 More recently, the Federal Housing Administration reported large losses in its Mutual Mortgage Insurance Fund, which provides mortgage insurance for lowand moderate-income home buyers. 4 These failures highlight the lack of effective budgetary control over credit programs and intensify concerns about their effect on the allocation of credit. The purpose of this paper is to examine recent credit reform proposals by the Administration and Senate. The essential ingredient in each proposal is to separate the estimated subsidy costs from the nonsubsidized cash flows of the credit programs. This change would make control of subsidies the primary focus of budgetary efforts. In addition, the Administration has proposed several reforms for particular credit programs that aim to improve their management of risk and targeting of resources. The remainder of the paper is organized as follows: Section I provides an overview of federal credit programs. The second section describes the current budgetary treatment of programs, while Section III explains various methods to estimate the subsidy components of existing credit programs. The fourth section considers the possible impact of current programs on the allocation of capital. Section V examines the proposed credit reforms. The sixth section offers some conclusions about the fiscal and economic effects of the proposed reforms. 1. An Overview of Federal Credit Programs

A substantial portion of funds that flow through the credit markets receives government assistance. In the fiscal years 1982 to 1989, federally 2 While GSEs are off-budget entities, their cash flows have been recorded in recent years in a supplement to the budget. 3 Department of the Treasury (1990), Appendix D, pp. 9-14. 4 General Accounting Office (1989b).

194 • Impact of Federal Credit Programs

assisted loans net of repayments averaged $110 billion a year, or 16 percent of total net lending in credit markets to the private sector. 5 At the end of fiscal year 1989, the outstanding amount of such loans totaled $1,558 billion. 6 In comparison, the largest U.S. commercial bank had assets totaling $231 billion at the end of 1989. 7 The form of government involvement in credit markets varies. For example, several federal agencies originate and service direct loans. These programs aim to redirect economic resources by providing credit on more favorable terms than would otherwise be available from private lenders. While the repayment of direct loans has exceeded disbursements in recent years, the outstanding amount of such loans totaled $207 billion at the end of fiscal year 1989 (Table 1). The Department of Agriculture and the Export-Import Bank operate the largest direct loan programs. Guaranteed loans, which grew steadily during the 1980s and totaled $588 billion at the end of fiscal year 1989 (Table 2), are privately held loans that the Government guarantees to pay in the event of default. The guarantee can cover all or part of the loan principal and interest, and thus transfer at least some of the default risk from the lenders to the Government. Guaranteed loans also include insured loans, where the Government pledges to use accumulated insurance premiums to secure lenders against default. The largest guaranteed loan programs are the mortgage insurance funds of the Department of Housing and Urban Development and the Department of Veterans Affairs. The troubled Mutual Mortgage Insurance Fund of the Federal Housing Administration alone covered $326 billion in loans at the end of fiscal year 1989. Guarantees of loans to students and small businesses also represent substantial contingent commitments of the Government. 8 GSEs are financial intermediaries that have been established and chartered by the Federal Government, but that are now, with one exception, privately owned. 9 Despite their private status, these enterprises maintain several links to the Government (Table 3). Because of these links, GSEs can borrow in private capital markets at interest rates just slightly above 0ffice of Management and Budget (1990), p. A-97. 0ffice of Management and Bud~et (1990), p. 229. Moody's Investors Service (1990). 8 In addition to the guarantee, the Government pays an interest subsidy on student loans. 9 This paper does not include as GSEs the Farm Credit System Financial Assistance Corporation, the Financing Corporation, or the Resolution Funding Corporation. These entities were created by the Government to finance through the Issue of nominally private bonds the resolution of insolvent farm credit and thrift institutions. 5

6

7

Steven M. Fries • 195

those on treasury securities of comparable maturity. While these obligations do not carry an explicit guarantee, investors perceive that GSE debt is implicitly backed by the Government, a perception reinforced by the Government's bailout in 1988 of the Farm Credit System. In addition to providing loans to agricultural enterprises, GSEs participate in the secondary markets for mortgages and student loans (as purchasers of loans for their own portfolios and issuers of pass-through securities), as well as provide loans to private financial intermediaries (primarily savings and loans ). 10 GSE loans outstanding, which include pass-through securities, grew rapidly during the 1980s and totaled $880 billion at the end of 1989 (Table 4). The largest GSEs are those that are active in the secondary mortgage market, the Federal National Mortgage Association and Federal Home Loan Mortgage Corporation. II. The Current Budgetary Treatment of Programs

The current budgetary measures for direct and guaranteed loan programs emphasize the amounts of cash actually paid or received in the fiscal year. While a cash-basis budget provides a comprehensive system for recording and controlling many receipts and outlays, it is an incomplete system when applied to federal credit programs because many economically significant aspects of the transactions are omitted. Most credit contracts and guarantees involve obligations for payments to or by the Government in future fiscal years. Thus, when the Government enters into credit commitments, many of the obligated amounts are excluded from that year's budget. This emphasis on cash flows in the current fiscal year tends to divert attention from the long-run costs of credit programs when budgetary priorities are being set. It is evident that cash-basis accounting for credit programs creates budgetary distortions. For example, the full amounts of direct loans are recorded as outlays when disbursed by federal agencies, whereas guaranteed loans have no positive outlays until defaults occur. Consequently, loan guarantees have been growing much faster than direct loans in recent years. Cash-basis accounting also results in treating equally some direct loans that differ in cost, since all loans of equal amounts have the same budget outlay in the disbursement period. No distinction is made between 10 Pass-through securities are claims to an underlying pool of loans. The GSEs that issue pass-through securities insure them against default risk from the underlying loans.

34,287 1,327

32,285 1,254

29,180 1,173

11,315 3,045

Department of Education Direct and guaranteed student loans College housing loans 11,816 3,025

9,269

8,801

8,307

12,439 2,676

9,759

16,007

12,484

25,927 27,201 7,333

24,385 26,022 7,013

11,866

11,840

11,762 24,207 24,368 6,556

24,037

1984

20,722

1983

17,358

1982

Department of Agriculture Agricultural Credit Insurance Fund Rural Housing Insurance Fund Rural Development Insurance Fund Commodity Credit Corporation commodity loans Public Law 480 long-term export credits Rural Electrification and Telephone Revolving Fund Rural Telephone Bank

Funds appropriated to the President International security assistance Agency for International Development

Federal Direct Loans Outstanding

13,336 2,300

14,418 2,229

35,941 1,434

10,389 1,194

34,323 1,447

11,219

10,622 10,046 35,636 1,383

18,577

21,608

15,105

11,706

32,570

1987

27,600 26,510 6,431

11,919

29,849

1986

28,698 29,295 7,957

28,563 28,860 7,708

11,855

26,532

1985

(In millions of U.S. dollars, end of fiscal year)

Table 1.

11,107 705

34,354 1,413

11,632

11,999

25,481 27,098 5,141

12,411

30,365

1988

11,983 679

35,129 1,485

12,204

8,904

22,547 27,867 3,798

11,853

22,422

1989

~

3(/)

Dl

~

:-1

Q.

:S.

~

::J Q.

Ill

(I)

~

):.

~

~

g Ill

27 4 • Unemployment and Potential Output

interaction with supply shocks, it is preferable to include, if possible, structural variables more directly related to labor market rigidities. Demographic Impacts on Unemployment

The impact of demographic factors on the natural rate of unemployment has often been emphasized (see Perry (1970) and Adams, Fenton, and Larsen (1987)). If different labor market groups have different natural rates of unemployment, perhaps because of different reservation wages or other characteristics, the changing demographic composition of the labor force may affect the natural rate of unemployment. This compositional effect (DEM) has been estimated as the difference between the aggregate unemployment rate and one constructed with constant labor market shares (Figure 1). In addition to the compositional effect of demographic changes, the relative minimum wage variable (RMW) has been multiplied by the labor-force share of youth aged 16 to 24 (SL ), the group most likely to be affected by the minimum wage. This implies that the impact on unemployment of a given change in the minimum wage will be greater the larger is the proportion of youth in the labor force. The total impact of both of these demographic effects turns out to be relatively small. Unemployment rate equation (4) in Table 4 tests for more direct effects on unemployment by also including the share of the labor force that is young (SL) and female (SF) (Figure 3). The proportion of youth, which has returned to about the same levels as in the mid-1960s, does not appear to have had a significant effect on unemployment over and above the two effects noted above. The estimated coefficient on the share of the labor force that is female, which has increased steadily from the mid-1960s, is significant and negativeY Given that unemployment rates for women are typically higher than for men, implying, if anything, a higher natural rate, it is difficult to rationalize a negative impact on unemployment rates from the increased proportion of women in the labor force. Structural Determinants of the Natural Rate

Many, perhaps most, of the structural features of labor markets are constant or evolve very slowly; and those that do change may be difficult to quantify. In the estimated equations in Table 4, the main structural 31 The estimated coefficients on both these labor-force groups are qualitatively similar to those reported in Table 4 when the variables are entered alone or in conjunction with the two supply-shock variables discussed above.

Charles Adams and David T. Coe • 275

variables affecting the natural rate of unemployment are union members as a percent of total private sector employment ( UNN), relative minimum wages (RMW), the average weekly unemployment insurance replacement ratio adjusted for the proportion of employees covered by the unemployment insurance system ( UIRR), and employers' contributions as a percent of total wages and salaries (NWLC) (see Figure 3). Average and marginal income tax rates were also entered in the equations, but the estimated coefficients were insignificant and/or incorrectly signed. The amount of time the unemployed search-an important determinant of the duration of unemployment-can be expected to depend, at least in part, on a comparison between the offered and reservation wage. The latter will be related to, among other factors, minimum wages, unemployment insurance benefits, and, perhaps, the degree of unionization. In the empirical analysis reported here, minimum wages and average weekly unemployment insurance benefits have been expressed relative to average wages of employees of private households, a relatively lowpaying service "industry" that is likely to be more relevant for a typical unemployed person and works better empirically than a measure of aggregate wages. Developments in relative minimum wages (adjusted for the share of youth) and unionization, particularly the decline in the 1980s (see Figure 3), are sufficiently similar so that it is not possible to get precise estimates of the contribution of each variable. 32 In view of this multicollinearity, which can be seen by comparing unemployment rate equations (1)-(3) presented in Table 4/3 the coefficients on these two variables are constrained to be identical in equation (1). The estimated coefficients on employers' contributions as a percent of total wages and salaries (NWLC) are statistically significant in each of the reported employment rate equations, and large relative to the other estimated structural coefficients. Except for equation (4) in Table 4, the size of the estimated parameters is also robust across the different specifications. In line with developments in most industrialized countries (see Chan-Lee, Coe, and Prywes (1987)), there were steady increases in 32 Neumark (1989) finds some evidence, based on a Phillips curve approach, that the degree of unionization affects the growth of union wages, but not the growth of aggregate wages. 33 When relative minimum wages and the degree of unionization are included separately in the equations, they are both significant, have similar coefficient estimates, and do equally good jobs of explaining unemployment rate developments (as in unemployment rate equations (2) and (3) in Table 4). When both variables are included in the same equation, however, neither is significant, although the goodness of fit of the equation is unchanged from those with either variable alone.

276 • Unemployment and Potential Output

Figure 3. Structural and Supply Factors Affecting Unemployment Percent

Percent

.------------------------:----,100 Unemployment insurance

60

Percent r~f employment covered (right scale)

90 55

..

50 _,

80

r , - _,.., ' " ' - '

Replclcemefl! ratio adjustedfor cm·erage (left scale)

/

45~~~~~~~~~~~~~~~~~ww~~~~uu~~ 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987

70

Current dollars

Percent

,-----------' 3.5

Minimum wages

I

20

I-

Minimum hourly wage (right scale)

3.0

2.5 15

I I

2.0

J

Re/atil·e minimum ~rage

~--------'

,-1

1

1.5

(left scale)

IO~uu~~~~~wwwwww~~~~~~uw~wwww~~~~ 1.0 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 Percent 25~------------------------,

Unionization

Union members as a percent of employment

20

15

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

Charles Adams and David T. Coe • 277

Figure 3 (concluded) Percent

21 18 15

Employers' l'Oilfrilmtions as a pen·n1t (~f' ~rages and .m/arie.\

12

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

Percent

50

Percent

26

Share of females and youth in the labor force

46

Youth. /6-2.J rear.\ ~ r (right scale I

42

I

...

I

~*-'-..~,,.,,

... ,

'

I

24

'_, '

22 lle/1 scale I

38

20 18

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

Percent

Percent

5

Supply shocks

45

,.

... ,_, ..,1"'1 t

filii

L

\'

Relati1·e export ) \ _,. 1/ prices· I (l·i~ht scale)

~,

30

''

15

'

-5

'

I

-15

0

1965 I 2 3

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

Minimum wage multiplied by the share of youth in the labor force. Export prices relative to import prices. Gasoline prices relative to the output deflator.

1987

5.85 14,100.00

4.80 7,800.00

4,800.00

20.6 10.2 10.4 5.3 3.4

19.4 9.3 10.1 4.3 4.0 1.8

6.70 35,700.00

6.13 25,900.00

1.7

1983

43,000.00

7.15

19.3 10.1 9.2 5.1 2.3 1.8

1987

Sources: U.S. Department of Commerce (July 1988, National Income and Product Accounts Tables 1.14 and 6.13) and United States Historical Abstract (various issues). 1 Includes group life insurance, workers' compensation, supplemental unemployment contributions, and directors' fees.

Memorandum items: Employers' social security contribution rate (percent) Maximum taxable earnings (dollars per year) 3.62

16.5 8.4 8.1 3.1 3.4 1.6

12.1 6.2 5.9 2.2 2.4 1.3

9.9 5.0 4.9 1.6 2.1 1.2

Total

Social security Other contributions Health insurance Pensions and profit-sharing Other 1

1975

1970

1965

Contribution

1980

3

I

s.

c::

.g

0

a ![

CD

~

-u

c.

::J

I ll

::J

CD

-

'


•

0

CD

(.,)

Steven M. Fries • 381

of savings and loans' assets for several reasons. Regulatory accounting standards and a cutback in budget resources allocated to thrift supervision permitted institutions to understate losses embedded in their portfolios, at least temporarily. When the magnitude of the problem became apparent, the relatively limited resources of the FSLIC further delayed regulatory action to take control of troubled savings and loans. Finally, some thrift institutions threatened with regulatory control were able to generate political support to delay or to stop such action. 18 In its initial response to the re-emergence of widespread thrift losses in the second half of the 1980s, the Congress sought to assist savings and loans and the FSLIC under the Competitive Equality Banking Act of 1987, which liberalized access to advances from the Federal Home Loan (FHL) Banks and contained a recapitalization plan for the insurance fund. To qualify for advances from a FHL Bank-a relatively low-cost source of funds for a thrift institution-the proportion of total assets that it must hold in the form of residential mortgages or related assets was set at 60 percent. To recapitalize the FSLIC, an off-budget governmentsponsored enterprise, the Financing Corporation (FICO), was established which was authorized to borrow $10.8 billion in private capital markets through fiscal year 1991. The proceeds of the FICO bond issues are placed with the FSLIC, an on-budget government agency, in exchange for nonvoting capital stock and capital certificates. 19 The interest payments on the FICO bonds are met from a special assessment on FSLIC-insured thrift institutions, which was levied in 1985 and which raised their deposit insurance premiums from 0.083 percent of insured deposits to 0.208 percent. However, this recapitalization plan proved inadequate for the task at hand. In 1988, the FSLIC resolved 205 insolvent thrift institutions at an estimated present-value cost of $31.7 billion (Table 2). Since this cost far exceeded the FSLIC's cash resources and net worth, most of the resolutions were in the form of assisted mergers. Typically, the FSLIC financed such a resolution by providing the acquirer with interest-bearing promissory notes, yield maintenance agreements, capital loss guarantees, and tax benefits. 20 In return, the acquirer of a troubled thrift institution generally invested additional capital, shared the tax benefits with the See, for example, Black (1989) and Jackson (1989). The effect of this complicated financing scheme is to remove from a measure of the budget deficit the FSLIC's outlays for insolvency resolutions. Bud~et accounting rules allow equity contributions to a government agency to offset Its outlays. 20 The additional liabilities of the FSLIC have value despite its negative net worth because of the implicit backing of the Federal Government. 18 19

382 • An Expensive Thrift Industry

FSLIC, and provided the FSLIC with some form of equity interest in the new institution. 21 Despite the record number of resolutions in 1988, there remained 364 GAAP insolvent savings and loans at the end of 1988. These institutions had $114 billion in assets and a tangible net worth of minus $16 billion. Another 392 institutions had GAAP capital-to-asset ratios of less than 3 percent with assets of $316 billion and only $1 billion in tangible net worth. Thus, at the year's end, 25 percent of all thrift institutions with 32 percent of the industry's assets were insolvent or significantly capital impaired. Structural Factors Underlying the Latest Crisis

An important cause of the marked deterioration of the capitalimpaired segment of the thrift industry in the second half of the 1980s is the "moral hazard" associated with federal deposit insurance coupled with an easing of capital adequacy standards and expansion of thrift asset powers. The moral hazard refers to the fact that the FSLIC cannot rely on the depositors of a thrift institution to protect its interest as the insurer of those deposits. Moreover, the value of the deposit insurance to the shareholders of a thrift institution-but not the insurance premium under the current system-increases with the riskiness of the institution's portfolio and decreases with the amount of capital at risk by its shareholders. 22 Thus, the FSLIC and thrift regulators restrict the permissible activities and stipulate capital adequacy standards of an insured savings and loan to limit the value of the deposit insurance liability. From this perspective, the easing of capital adequacy requirements in response to the industry's losses in the early 1980s and the expansion of asset powers increased the value of the FSLIC's contingent liabilities. A preliminary analysis of the 1988 insolvency resolutions tends to support the above hypothesis. 23 For over 90 percent of the thrift resolutions, the savings and loans had been insolvent for over one year, some for as long as ten years. During this time, the troubled institutions invested more heavily in direct investments, such as real estate and equity securities, as well as acquisition and development loans. To the extent that these investments increased overall portfolio risk, and ultimately the loan losses, this evidence substantiates the claim that the moral hazard 21 The FSLIC sought to share the tax benefits with the acquirer of a troubled thrift institution because of their uncertain value, while the FSLIC's equity participation limits a new institution's incentive to take excessive risks. 22 See Kane (1985). 23 Barth, Bartholomew, and Labich (1989).

Steven M. Fries •

383

associated with deposit insurance contributed to the insolvencies. Moreover, the troubled savings and loans tended to pay interest rates above the industry average by relying heavily on brokered deposits, permitting a more rapid expansion of their balance sheets and investment activities. Whether investments in particular assets are necessarily associated with increased portfolio risk is an unresolved issue, however. Expanded assets powers can facilitate diversification; or they can increase opportunities for risk taking. Two studies have attempted to assess these hypotheses, which are not mutually exclusive, by using the composition of portfolios, inter alia, to explain variations in risk among thrift institutions. 24 The studies generally concluded that the proportion of direct investments to total assets is insignificant in explaining the variation in risk among savings and loans. However, the proportion of capital to total assets is significantly and negatively related to risk, while the proportion of brokered deposits to total assets is significantly and positively related to risk. These results are consistent with the view that an insolvent or capitalimpaired thrift institution has an incentive to take excessive risks of all types and not just those permitted by expanded asset powers. 25 II. The Financial Institutions Reform, Recovery, and Enforcement Act of 1989

The FIRREA provides for a comprehensive reform of the savings and loan industry. The legislation largely conforms to the Administration's reform plan and contains five main elements. First, to resolve cases involving savings and loans that are currently insolvent or that become so over the next three years, a new government agency, the Resolution Trust Corporation (RTC), will be established. The RTC will operate under an oversight board composed of, among others, the Secretary of the Treasury, the Secretary of Housing and Urban Development, and the Chairman of the Board of Governors of the Federal Reserve System. The FDIC will manage the RTC unless removed by the oversight board. The RTC will be authorized to borrow $20 billion 24 Benston (1985) and Benston and Koehn (1989). In these studies, risk is measured as either the standard deviation of accounting rates of return on equity or of stock market returns; each is a proxy for the standard deviation of the return on assets. 25 A number of studies have found that the FSLIC's resolution costs are significantly and positively related to proportion of direct investments to total assets, including Barth, Brumbaugh, Sauerhaft, and Wang (1985), Barth, Brumbaugh, and Sauerhaft (1986), and Barth, Batholomew, and Labich (1989). However, these studies do not permit consideration of the role of expanded asset powers in reducing portfolio risk of an adequately capitalized thrift institution.

384 • An Expensive Thrift Industry

from the Treasury Department in fiscal year 1989 and a separate Resolution Funding Corporation (REFCORP) will be authorized to borrow a total of $30 billion in private capital markets in fiscal years 199~91. These proceeds will be transferred to the RTC for use in the insolvency resolutions. The second element of the legislation is to raise funds from the thrift industry to meet part of the insolvency resolution costs. Deposit insurance premiums for savings and loans will be raised from the current 0.208 percent of insured deposits to 0.23 percent in 1991, although they will be reduced to 0.18 percent in 1994 and to 0.15 percent in 1998. The FHL Banks, the 12 regional banks of the Federal Home Loan Bank System, will be required to pay the REFCORP from their accumulated retained earnings $1.2 billion toward resolution costs in fiscal year 1989 and additional amounts toward the purchase of the zero-coupon treasury securities that upon their maturity will repay the principal of the REFCORP borrowings. 26 In addition, the FHL Banks will contribute to REFCORP $300 million a year beginning in fiscal year 1992 to meet part of the interest payments on its borrowings. 27 The deposit insurance premiums for commercial banks also will be increased under the legislation. These premiums will be used to restore the reserves of the bank insurance fund to a more adequate level and will serve to lessen the net budget outlays associated with the legislation. 28 Third, to limit the thrift industry's influence over its federal regulators, the deposit insurance fund for savings and loans will be separated from the FHLBB. A single deposit insurance agency will be created by merging the FSLIC into the FDIC; however, separate insurance funds for thrift institutions and commercial banks will be maintained. 29 The FHLBB will be abolished. The Office of Thrift Supervision (OTS) will be established as an office in the Treasury Department and will be responsible for supervising all savings and loans as well as chartering federal institutions. This regulatory structure parallels that for national banks, with the OTS performing functions similar to those of the Office of the Comptroller of the Currency. The collateralization of a bond's principal with a zero-coupon treasury secuis often referred to as defeasance. 7 Each FHL Bank is owned by the member institutions of its district, as are each of the 12 regional Federal Reserve Banks. REFCORP's claim on the accumulated retained earnings of the FHL Banks and on part of their future earnings would reduce the value of savings and loans holdings of equities in the FHL Banks. 28 The reserves of the bank insurance fund fell below the statutory level of adequacy in fiscal year 1987. 29 The Savings Association Insurance Fund (SAIF) and Bank Insurance Fund (BIF), respectively. 26

rit~

Steven M. Fries •

385

The fourth element is to enhance the safety and soundness standards for the thrift industry. In particular, risk-based capital standards will be established that are no less stringent than those for national banks. Goodwill will be phased out from capital over five years, and bank holding companies will be permitted to acquire savings and loans to attract more capital to the industry. Moreover, investments in high-yield, noninvestment-grade bonds and equities will be prohibited or largely restricted. Fifth, to preserve a specialized housing finance system, thrift institutions will be required to invest an increased proportion of their assets in residential mortgages and related assets to qualify for advances from the FHL Banks. In addition, the system of 12 regional FHL Banks will be reorganized under a newly created Federal Housing Finance Board. The primary purpose of the FHL Banks will continue to be the provision of liquidity to their member institutions. As currently practiced, the system will borrow in private capital markets and advance the proceeds to member institutions for both short-term liquidity needs and long-term expansion of their balance sheets.

Ill. Estimates of the Present-Value Cost of Thrift Insolvency Resolutions

Estimates of the present-value cost of the insolvency resolutions, by both government agencies and private analysts of the thrift industry, have risen from a range of $1(}..30 billion at the end of 1986 to a range of $76-111 billion at present. 30 The significant upward revision of these estimates reflects the continued deterioration of the capital-impaired segment of the thrift industry. Also, the revisions and remaining differences among the estimates reveal the considerable difficulty in assessing the actual financial condition of savings and loans. Because of book value accounting standards, such as GAAP and RAP, which record assets at their historical cost, a decline in asset values is not reflected in financial statements until the losses are realized or the loans become nonperforming. To estimate the insolvency resolution costs, the balance sheets of troubled savings and loans must be converted to reflect market values, since these values provide a measure of the likely recoveries on assets in the event of a resolution. In practice, this conversion is generally undertaken for troubled thrifts that continue to operate independently by first 30

Wall (1989).

386 • An Expensive Thrift Industry

grouping these institutions according to their net worth as a percent of total assets (and in some cases taking into account their recent profitability). The loss rates on assets incurred in past resolutions of comparable institutions is then applied to obtain an estimate of future resolution costs. For insolvent savings and loans resolved prior to 1989, more detailed information on losses at each institution is available from the FSLIC. A rough assessment of the present-value cost of the insolvency resolutions can be constructed as follows: (1) the FSLIC reports a present-value cost of resolutions undertaken in 1987-88 of $39 billion; (2) the FDIC estimated that, for the 212 RAP-insolvent institutions placed under its supervision as part of the Administration's reform plan, the loss rate on their $73 billion in assets is 41 percent, with the losses amounting to $30 billion; 31 (3) to the remaining 139 GAAP-insolvent savings and loans with total assets of $33 billion a loss rate of 35 percent is applied, yielding additional losses of $12 billion; 32 and (4) finally, there were 111 GAAPsolvent savings and loans with $121 billion in assets but no tangible net worth and 279 marginally solvent institutions with $194 billion in assets, and these institutions would add a further $22 billion to the present-value resolution costs. 33 Overall, the calculations suggest a total presentvalue cost of $103 billion, which is at the upper end of the range of estimates available from various government agencies and private thrift industry analysts (Table 3). The above estimate of the present-value cost of the insolvency resolutions is subject to a number of qualifications. Foremost among them is that the RTC, with over $400 billion in assets at book value, will be managed and operated efficiently. Many of the RTC's assets will be concentrated in regions of the country with relatively weak economies and depressed real estate markets, and RTC will be under a legislative mandate to minimize the impact of its actions on local markets and to promote access to affordable housing by low-income households. Furthermore, the estimated present-value cost of the resolutions is based on several economic assumptions, including the absence of a recession, no change in the economy of the Southwest or in the local real estate markets, and broadly stable interest rates at around their current levels. 31

Federal Deposit Insurance Corporation (1989). This loss rate is based on the FSLIC's experience with its 1987 and 1988 resolution cases. See Federal Savings and Loan Insurance Corporation (1989). 33 The FDIC has experienced a loss rate on assets in the resolutions of similarly capitalized institutions of 12 percent and 4 percent, respectively. See Seidman (1989). 32

Steven M. Fries • 387

Table 3.

Estimates of the Present-Value Cost of Resolving the Thrift Insolvencies (In billions of U.S. dollars)

Cost of Pre-1989 Resolutions 42 39

Cost of New Resolutions 1 69 65

Total Cost of Resolutions 111 104

Brumbaugh, Carron, and Litan 2 Congressional Budget Office Federal Deposit Insurance Corporation3 41 61 102 Federal Home Loan Bank Board 38 38 76 General Accounting Office 39 39 78 Office of Management and Budget 39 50 89 Sources: Brumbaugh, Carron, and Litan (1989); Congressional Budget Office (1989); Seidman (1989); Wall (1989); Wolf (1989); Office of Management and Budget (1989). 1 Includes resolutions of primarily those thrift institutions that continued to operate but that were GAAP-insolvent or without any tangible net worth at the end of 1988. 2 Reported figures are the midpoints of ranges, with the estimate of total resolution costs ranging from $86 billion to $136 billion. 3 Reported figures are the midpoints of ranges, with the estimate of the total resolution costs ranging from $92 billion to $112 billion.

IV. A Cash-Flow Analysis of the Insolvency Resolutions

The uses of cash for insolvency resolutions under FIRREA will fall into three general categories: the actual resolution costs, capitalization of SAIF, and debt-service payments on borrowings. The sources of cash will be income from the thrift industry and borrowings by REFCORP, FICO, and the Treasury. The cash flows discussed in this section are based on the Administration's projections, adjusted to conform with the final provisions of the FIRREA and to include all debt-service payments. The total sources and uses of funds through fiscal year 1999 are projected to be $231 billion, although a sensitivity analysis suggests that the total could exceed $250 billion. Regarding the uses of cash, the insolvency resolution costs can be classified by when the resolutions have occurred or are expected to take place. First, because of the extensive use of FSLIC notes, yield maintenance agreements, and capital loss guarantees in past resolution cases, these cases will give rise to a series of cash outlays through fiscal year 1999 totaling $60 billion (Table 4). Second, as noted above, the RTC will

Sources of cash Income Thrift insurance premium Liquidation FHL Banks Other Borrowing REFCORP FICO bonds Treasury Total sources Sources: Office of Management

-

5.0 0.6 1.6 1.0 1.8 26.9

-

0.7

5.2 2.1 1.6 1.0 0.5 21.7 15.0 4.3 2.8 20.2 3.9 29.1 26.9 and Budget (1989) and

4.6 2.0 1.2 1.2 0.2 24.5

29.1

-

0.7

-

21.9 6.9 15.0

1990

28.4 8.4 20.0

1989

-

-

11.8 18.4

6.6 2.5 3.4 0.3 0.4 11.8

1.0 3.9 16.6

1.3 3.2 18.4

1.1 2.6 17.3 7.1 2.4 3.9 0.3 0.5 10.2

-

-

-

-

11.1 16.6

-

5.5 2.2 2.6 0.3 0.4 11.1

2.0 2.0 7.2 2.3

-

4.1 2.0 6.8 2.3

-

7.4 5.4

1994

9.6 5.5

1993

4.1 2.0 6.0 2.3

9.3 5.2

1992

10.2 5.7 27.2 17.3 Fund staff estimates.

-

6.5 2.3 3.4 0.3 0.5 20.7 15.0

6.9 1.7 2.0 1.1 2.1 27.2

-

20.3 5.3 15.0

1991

(In billions of U.S. dollars)

35.5 13.5 16.1 3.4 2.5 100.0 30.0 7.1 62.9 135.5

96.9 36.7 50.0 10.2 6.0 32.6 9.2 3.6 6.2 13.6 135.5

198994

52.8 26.1 18.3 4.9 3.5 178.3 30.0 7.1 141.2 231.1

134.7 60.5 50.0 24.2 16.0 80.4 20.7 3.6 11.7 44.4 231.1

198999

Projected Cash Flow Analysis of Savings and Loan Insolvency Resolutions, Fiscal Years 1989-99

Uses of cash Resolution costs Pre-1989 cases RTC cases Post-RTC cases Capitalization of SAIF Debt service REFCORP interest REFCORP defeasance FICO debt service Treasury interest Total uses

Table 4.

en

~

en

c

a.

:J

~

::::;

-l

;:::·

::I

X 1:J

m

::I

)>

•

(A) (X) (X)

Steven M. Fries • 389

resolve cases involving thrift institutions that are currently insolvent or that become so over the next three years. The FIRREA authorizes the RTC to borrow $20 billion from the Treasury and the REFCORP to borrow $30 billion in private capital markets and to advance the proceeds to the RTC for use in those insolvency resolutions. Third, the Administration anticipates a further $24 billion in resolution costs between fiscal year 1992 and fiscal year 1999, owing to a significant number of weakly capitalized thrift institutions that will not fall under the control of the RTC. These cases will be the responsibility of the SAl F. Altogether, the actual resolution costs in cash flow terms are projected to total $135 billion through fiscal year 1999. The capitalization of SAIF will begin in fiscal year 1992 at a rate of $2 billion in each fiscal year. The Treasury will make up any shortfall from this rate if net deposit insurance premiums from savings and loans prove inadequate. Under the Administration's assumptions, the SAIF's accumulated reserves will total 1 percent of insured deposits by fiscal year 1999. This accumulation of reserves is 0.25 percent below the designated target ratio for the deposit insurance funds under FIRREA. The debt-service payments on the borrowings to fund part of the resolution costs and capitalization of SAIF will fall into three categories. First, the REFCORP will require, in addition to $21 billion to meet its interest payments through fiscal year 1999, $3.6 billion to purchase the zero-coupon treasury securities that will back the principal of its borrowings. Second, the debt-service payments on the FICO's borrowings, authorized under the FSLIC's recapitalization plan of 1987, will total $12 billion through fiscal year 1999. Third, there will be substantial interest payments on treasury borrowings. These costs are estimated to total $44 billion through fiscal year 1999, using the interest rate assumptions contained in the Mid-Session Review of the Budget but adjusted to a fiscal year basis. 34 Taken together debt -service payments over the 11 fiscal years are projected to sum to $80 billion. About 23 percent of the $231 billion in total cash outlays through fiscal year 1999 will be met with income from the thrift industry. Savings and loan insurance premiums will amount to $26 billion during this period based on average rate of growth in deposits of 5.2 percent, while proceeds from the liquidation of assets from insolvent institutions will contribute $18 billion. The accumulated retained earnings of the FHL Banks and a portion of their future earnings will add a further $5 billion. In total, the thrift industry will provide $53 billion through fiscal year 1999. The remaining 77 percent of the total cash outlays over the 11 fiscal 34

Office of Management and Budget (1989).

390 • An Expensive Thrift Industry

years will be met through borrowing by government-sponsored enterprises and the Treasury. The REFCORP is authorized to borrow $30 billion from private capital markets. The FICO will borrow the $7 billion that remains under its borrowing authority. Finally, the Treasury will make up the shortfall, an estimated $141 billion. Total borrowings through fiscal year 1999 are projected to be $178 billion. The above-projected cash flows are sensitive to key assumptions about the present-value cost of the insolvency resolutions and future interest rates. If the present-value cost of the insolvency resolutions is about $100 billion (as indicated in the previous section) rather than about $90 billion (as assumed by the Administration), total cash outlays through fiscal year 1999 would be about $20 billion higher. Alternatively, if long-term interest rates decline to about 6.5 percent over the medium term rather than the 5.4 percent assumed by the Administration, total cash outlays would be $25 billion higher. V. Estimated Net Budget Outlays Under FIRREA

The funding of the insolvency resolutions under FIRREA was structured with a view to the Gramm-Rudman-Hollings (GRH) deficitreduction targets. Because of concern over a possible increase in the Government's borrowing costs if there were a waiver of the GRH targets, the Administration proposed that all RTC resolution cases be funded by the REFCORP, an off-budget, government-sponsored enterprise. 35 The proceeds of the REFCORP bond issues would be placed with the RTC, an on-budget government agency, in exchange for nonvoting capital stock. Since budget accounting rules allow equity contributions to a government agency to offset its outlays, net budget outlays associated with the RTC's resolution cases would be confined to the debt-service payments on the REFCORP borrowings. However, the REFCORP would have to pay a premium to borrow in private capital markets for want of an explicit government backing of its liabilities. Under FIRREA, the RTC's $20 billion in resolution costs in fiscal year 1989 will be funded by the Treasury to reduce the debt-service payments associated with the resolutions. No waiver of the GRH deficit-reduction target is required for the current fiscal year because it applies only to budget projections and not to the budget outcomes. The RTC's resolution costs in fiscal years 1990-91 will be funded by the REFCORP as proposed by the Administration, and thus will not be reflected in the budget deficits of those years. 35

Brady ( 1989b ).

Steven M. Fries • 391

To derive the net budget outlays associated with the FIRREA, two adjustments must be made to the treasury borrowings based on the above cash-flow analysis of the insolvency resolutions. First, the FSLIC funded part of its past resolution costs by issuing notes, and for budget purposes these costs were scored when the notes were issued even though there were no outlays of cash. Thus, a corresponding adjustment must be made when these notes are redeemed. Second, the additional bank deposit insurance premiums received by the FDIC under the legislation will serve to reduce the budget deficit. With these adjustments, over 70 percent of the net budget outlays associated with FIRREA through fiscal year 1999 will occur in fiscal year 1989 and fiscal years 1995-99 (Table 5).

VI. Structural Reforms for the Thrift Industry Developments in the thrift industry in the 1980s reveal several structural weaknesses, many of which are addressed under FIRREA. The key issues center around (1) capital adequacy standards; (2) movement toward market-value accounting; (3) the elimination of tax preferences for the acquisition of a troubled institution; and (4) separation of the deposit insurance agency, thrift regulators, and the FHL Banks that promote the long-term expansion of the industry. The FIRREA significantly strengthens the capital adequacy standards for the thrift industry and stipulates that they be no less stringent than capital standards for national banks. In the transition period, savings and loans will be required to maintain a tangible capital-to-assets ratio of 1.5 percent by February 1990 and a core capital-to-asset ratio of 3 percent by January 1991. 36 An institution not in compliance with the capital standards will be subject to a mandatory restriction on asset growth and a prohibition on obtaining funds through brokered deposits. To attract additional capital to that industry, bank holding companies will be permitted to acquire financially healthy savings and loans. However, a thrift institution or bank that is owned by a holding company will be liable for the losses to the FDIC from any affiliated institution. To enable the FDIC to limit the risk exposure of the SAIF, the FIRREA authorizes the agency to suspend temporarily the deposit insurance of a thrift institution if its tangible capital is fully depleted. The intent of the legislation is to provide the FDIC with the ability to control quickly an insolvent institution by threatening the institution with a run on its deposits. However, given the political consequences of imposing 36

Goodwill will be amortized over the period ending December 1994.

1991 5.7 -0.1

-0.4 -0.1 -0.9 -1.7 2.6 3.9 (1989) and Table 4.

1990 3.9 -0.4 -1.9 8.3

1992 10.2

1994 11.1 -1.6 -1.6 -2.1 7.4

1993 11.8 -1.0 -1.0 -2.0 8.8

198994 62.9 6.0 10.6 -4.6 -8.6 60.3

Estimated Net Budget Outlays from the Financial Institution Reform, Recovery, and Enforcement Act of 1989, Fiscal Years 1989-99

1989 Treasury borrowings 20.2 9.1 Adjustment for FSLIC notes Notes issued 10.6 Notes redeemed -1.5 Additional bank premiums Net budget outlays 29.3 Sources: Office of Management and Budget

Table 5.

198999 141.2 -9.4 10.6 -20.0 -21.7 110.1

~

en

c:

:r a.

~

:::!.

::r

-1

~·

en

::J

•

~ 1\)

Steven M. Fries •

393

losses on depositors, the FDIC's threat of using this authority may not be credible, since an institution with no tangible capital under GAAP often has a negative net worth in market-value terms. Movement toward a market-value accounting standard for the thrift industry is recommended for further study under FIRREA. Such a standard would strengthen the FDIC's ability to control a troubled savings and loan thereby limiting the losses it can impose on the deposit insurance fund. 37 However, a number of technical difficulties are associated with a market-value accounting standard. Many assets are infrequently traded and thus would be difficult to value in market terms. Off-balance-sheet assets and liabilities would also need to be converted to their market values. With these measurement problems, it would be difficult to specify a closure rule for an insolvent institution that prevented any losses to the deposit insurance funds, since the FDIC would be exposed to possible legal actions if it assumed control of a solvent institution. Nevertheless, movement toward market-value accounting, coupled with increased enforcement actions by thrift regulators, would serve to limit the resolution costs if an insolvency were to occur. 38 Market-value accounting also would promote greater accountability of the Congress, Administration, and thrift regulators for their intervention in the industry. 39 Such a standard would significantly curtail the ability to defer explicit recognition of deposit insurance losses. To the extent that resolution costs tend to increase while an insolvent institution is permitted to continue its independent operation, claims on the deposit insurance fund and potentially on the taxpayers would be reduced. A more complete accounting of insolvency resolution costs will be promoted by repeal of the tax incentive to an acquirer of a troubled savings and loan or bank under FIRREA. Their repeal also will eliminate a potential source of inefficiency in a regulator-assisted merger. The acquirer of a troubled savings and loan or bank will be the institution that can operate the troubled entity most efficiently rather than the one that can take maximum advantage of the tax concessions. The separation of the deposit-insured fund for savings and loans, thrift regulators, and the FHL Banks (which promote the long-term expansion of the industry) under FIRREA will help to reduce the industry's influence over its regulators. By placing the newly created OTS within the Treasury Department, the regulation and supervision of savings and loans may more fully reflect the interests of taxpayers. A similar effect See, for example, Benston and Kaufman {1988). For specific reform proposals, see Benston and Kaufman {1988) and Benston and others {1989). 39 See Kane (1989), pp. 158-79. 37

38

394 • An Expensive Thrift Industry

is anticipated by giving the FDIC administrative control over the SAIF. The FHL Banks, reorganized under the newly created Federal Housing Finance Board, will continue to provide advances to their member institutions for short-term liquidity and the long-term expansion of their balance sheets. While a number of structural reforms under FIRREA will strengthen the thrift industry, the legislation seeks to preserve a separate housing finance system by requiring savings and loans to hold a greater proportion of their assets than previously in the form of residential mortgages and related assets. To qualify for advances from a FHL Bank, a thrift institution must hold 70 percent of its assets in residential mortgages, up from 60 percent. The developments in the industry over the past decade indicate that this provision, combined with more restricted asset powers that prohibit direct investments and holdings of high-yield, noninvestment-grade bonds, is likely to weaken the thrift industry. 40

VII. Conclusion

The President's reform plan for the savings and loan industry had as its objectives the raising of funds to resolve the remaining insolvent savings and loans and the reform of the industry to prevent a crisis from recurring. The FIRREA represents a substantial step toward the attainment of these goals, although some problems remain. The FIRREA's provision for funding of the insolvency resolutions, while substantial, may prove to be inadequate. The estimated presentvalue cost of the insolvency resolutions developed in this paper is somewhat above that contained in the legislation. Both estimates assume that there will be no recession and that interest rates will remain broadly stable. In addition, based on the Administration's assumption, 77 percent of the estimated $231 billion in total cash outlays associated with FIRREA through fiscal year 1999 will be borne by the taxpayers. A sensitivity analysis suggests that total cash outlay could exceed $250 billion. The structural reforms promulgated under FIRREA will place the thrift industry on a sounder foundation. Capital adequacy standards will be raised to a level that is no less stringent than the capital standards for national banks. The separation of the deposit insurance agency, thrift regulators, and the FHL Banks will help to limit the influence of the thrift industry over its regulators. The legislation also recommends for further 40 The need for a separate housing finance system is unclear given the development of mortgage-backed securities and the range of institutions that originate mortgages.

Steven M. Fries • 395

study movement toward a market-value accounting standard for the industry. However, the measures to preserve a separate housing finance system by increasing the proportion of assets a thrift institution must hold in the form of mortgages and the restrictions on the asset powers of a thrift institution are likely to weaken the industry.

REFERENCES Barth, James R., R. Dan Brumbaugh, Jr., Daniel Sauerhaft, and George H. K. Wang, "Thrift Institution Failures: Causes and Policy Issues," in Proceedings of a Conference on Bank Structure and Competition (Chicago: Federal Reserve Bank of Chicago, 1985). Barth, James R., R. Dan Brumbaugh, Jr., and Daniel Sauerhaft, "Failure Costs of Government-Regulated Financial Firms: The Case of Thrift Institutions," FHLBB Research Paper No. 123 (Washington: Federal Home Loan Bank Board, October 1986). Barth, James R., and Michael G. Bradley, "Thrift Deregulation and Federal Deposit Insurance," FHLBB Research Paper No. 150 (Washington: Federal Home Loan Bank Board, November 1988). Barth, James R., Phillip F. Bartholomew, and Carol J. Labich, "Moral Hazard and the Thrift Crisis: An Analysis of the 1988 Resolutions," FHLBB Research Paper No. 160 (Washington: Federal Home Loan Bank Board, May 1989). Benston, George J., "An Analysis of the Causes of Savings and Loan Association Failures," Monograph Series in Finance and Economics, Monograph No. 1985 4/5 (New York: Salomon Brothers Center for the Study of Financial Institutions, New York University, 1985). - - - , R. Dan Brumbaugh, Jr., Jack M. Guttentag, Richard J. Herring, George G. Kaufman, Robert E. Litan, and Kenneth E. Scott, Blueprint for Restructuring America's Financial Institutions (Washington: Brookings Institution, 1989). Benston, George J., and George G. Kaufman, "Regulating Bank Safety and Performance," in Restructuring Banking and Financial Services in America, ed. by WilliamS. Haraf and Rose Marie Kushmeider (Washington: American Enterprise Institute, 1988). Benston, George J., and Michael F. Koehn, "Capital Dissipation, Deregulation, and the Insolvency of Thrifts," Shadow Financial Regulatory Committee Working Paper No. 89-02 (Chicago: Mid America Institute for Public Policy Research, January 1989). Black, William, "Memorandum to Edward Gray Regarding a Meeting with Senators Cranston, De Concini, Glenn, McCain, and Riegle," Wall Street Journal (New York), June 19, 1989, p. A20. Brady, Nicholas F. (1989a), Statement Before the U.S. Senate Committee on Banking, Housing, and Urban Affairs, Problems of the Federal Savings and Loan Insurance Corporation [FSLIC], lOlst Cong., 1st Sess., February 22, 1989 (Washington: U.S. Government Printing Office, 1989).

396 • An Expensive Thrift Industry

- - - (1989b ), "No Time to Waive Gramm-Rudman," Wall Street Journal (New York), July 11, 1989, p. A20. Brumbaugh, R. Dan, Jr., Thrifts Under Seige: Restoring Order to American Banking (Cambridge, Massachusetts: Ballinger, 1988). ---,and AndrewS. Carron, "Thrift Industry Crisis: Causes and Solutions," Brookings Papers on Economic Activity: 2 (1987), Brookings Institution (Washington), pp. 349-77. - - - , and Robert E. Litan, "Cleaning Up the Depository Institutions Mess," Brookings Papers on Economic Activity: I (1989), Brookings Institution (Washington), pp. 243--95. Carron, AndrewS., The Plight of the Thrift Institutions (Washington: Brookings Institution, 1982). Congressional Budget Office, "Cost Estimate for the Financial Institutions Reform, Recovery and Enforcement Act of 1989" (Washington, June 7, 1989). Executive Office of the President, "The President's Reform Plan for the Savings and Loan Industry" (Washington, February 6, 1989). Federal Deposit Insurance Corporation, "S&L Conservatorship Estimated Loss Profile" (Washington, May 17, 1989). Federal Savings and Loan Insurance Corporation, "Case Resolution Report" (Washington, June 30, 1989). Jackson, Brooks, "As the Thrift Industry's Troubles and Losses Mounted, Its PAC Donations to Key Congressmen Surged," Wall Street Journal (New York), February 7, 1989, p. A26. Kane, Edward J., The Gathering Crisis in Federal Deposit Insurance (Cambridge, Massachusetts: MIT Press, 1985). - - - , The S&L Insurance Mess: How Did It Happen? (Washington: Urban Institute Press, 1989). Office of Management and Budget, "Mid-Session Review of the Budget" (Washington, July 19, 1989). Seidman, L. William, Statement at Field Hearings Before the U.S. House Committee on Banking, Finance, and Urban Affairs, Financial Condition of the Federal Savings and Loan Insurance Corporation and Federal Deposit Insurance Corporation at Year End I988, 101st Cong., 1st Sess., March 11, 1989 (Washington: U.S. Government Printing Office, 1989). Wall, M. Danny, Statement Before the U.S. Senate Committee on Banking, Housing, and Urban Affairs, Problems of the Federal Savings and Loan Insurance Corporation [FSLIC], 101st Cong., 1st Sess., March 1, 1989 (Washington: U.S. Government Printing Office, 1989). Wolf, Frederick D., Statement at Field Hearings Before the U.S. House Committee on Banking, Finance, and Urban Affairs, Financial Condition of the Federal Savings and Loan Insurance Corporation and Federal Deposit Insurance Corporation at Year End I988, 101st Cong., 1st Sess., March 10, 1989 (Washington: U.S. Government Printing Office, 1989).

PART IV

Current Account Adjustment and the Exchange Rate

A Dynamic Error-Correction Model of the U.S. Current Account ELLEN M. NEDDE

C

onventional models of the U.S. current account, which express trade flows as functions of activity variables and relative prices, have recently been criticized on the grounds that they fail to reflect fully supply-side effects and, as a result, generate medium-term projections that may be overly pessimistic. These criticisms were partly fueled by the inability of many modelers to forecast accurately the turning point in U.S. trade and current account developments since 1987 and the degree of more recent improvements. 1 The inability of models to forecast accurately short-term movements in net flows should not be surprising. Net flows that are based on much larger underlying gross flows are inherently difficult to predict. A prediction error in the trade balance in 1989 equal to 3 percent of underlying gross flows (current dollar merchandise imports plus exports) would have implied an error of $25 billion, or 22 percent of the actual trade balance. While it is possible for short-run prediction errors to indicate more fundamental problems with a model's predictive power, one should test directly for structural breaks in the equations rather than examine the model's ability to track actual developments in net flows over a short horizon. This paper outlines a quarterly model of the U.S. current account that was recently estimated by the IMF's North American Division for use in projections associated with the World Economic Outlook. The model The author would like to thank Charles Adams, Liam Ebrill, and Yusuke Horiguchi for helpful comments and discussions and Fredesvinda Pham for research assistance. 1 See Hooper (1988) and Helkie and Hooper (1989).

399

400 • A Model of the U.S. Current Account contains 40 estimated equations which describe the behavior of volumes and prices for trade in goods and services, payments and receipts of investment income, and gross flows in the capital account. To the extent possible, foreign demand and price variables were taken from the "global economic environment" for the United States provided as part of the World Economic Outlook exercise. While some researchers have focused on additional variables that might capture missing supply effects, another possible source of misspecification is the assumed lag structure. The specification of the current model combines some conventional features of trade models with an error correction methodology. This methodology, combined with a general-to-specific testing-down procedure, permits the data to identify a more general lag structure and demand and supply specification than those contained in most existing models of the U.S. current account, and thus may be better able to capture longer-run supply effects. To evaluate the model's ability to capture supply effects, particularly since the decline of the dollar beginning in 1985, several diagnostic tests are performed to check for the presence of white noise residuals. In addition, a subsample of the data is reserved to test for structural breaks in the equations. Ex post simulation and forecast properties are then examined, revealing the model's overall ability to predict out of sample. Experiments with alternative assumptions for some key exogenous variables reveal further the model's simulation properties. The paper is organized as follows. Section I describes the general methodology of the paper, including the error correction methodology and general-to-specific testing-down procedure used and the degree of disaggregation of the current account for purposes of estimation. Section II contains descriptions of long-run equilibrium and short-run error correction equations for merchandise trade volumes and prices. Equations for trade in services are discussed in Section III, while Section IV contains equations for investment income and the capital account. The predictive performance of the model is examined in Section V of the paper. Section VI describes the response of the current account and its components to changes in assumptions regarding domestic and foreign growth and the value of the dollar. Concluding remarks are contained in Section VII. Estimated equations, associated test statistics, and tabulations of alternative estimates are presented in Annexes I and II. I. Methodology

For purposes of estimation, real trade flows are disaggregated into oil and non-oil imports and agricultural and nonagricultural exports. Implicit deflators for all but oil imports and fixed-weight prices for non-oil

Ellen M. Nedde • 401

imports and nonagricultural exports are also estimated. In the services account, equations are estimated for real payments and receipts of total services and their deflators. In the investment account, nominal direct investment income payments and receipts are estimated, as are implicit rates of return on private and government portfolio investments and stocks of direct investment and portfolio assets and liabilities. The stock of private portfolio liabilities is solved residually to ensure that external financing is consistent with the estimated current account balance. The model is estimated using the two-step procedure as developed by Granger and Engle. 2 This procedure involves the estimation of a long-run equilibrium relationship among levels of variables in the first stage. In the second stage, first-differences of variables and lagged residuals from the long-run equation (deviations from equilibrium) are used to estimate a short-run error correction equation which models short-run dynamics. In both the first and second stages of the estimation, a general-tospecific methodology was followed. In the first stage, alternative sets of variables (that were consistent with perfect and imperfect competition) were tested for cointegration with the dependent variable. In addition, volume equations were tested for cointegration with alternative activity variables, such as gross national product and total domestic demand, and for the importance of trend terms that might signal supply effects not fully reflected in prices. Prices were specified generally to allow the United States to be either a price setter or price taker in export and import markets. In the second stage, Granger causality tests were performed to test for exogeneity of possible independent variables. The lag structure specified in the error correction equations was a general one, with four lags in the dependent and independent variables included. Nested F tests were then used to test for a more parsimonious representation. All estimated equations, test statistics, and definitions of the tests performed are contained in Annex I; the results of these tests are not discussed in detail except for those cases in which they indicate a possible specification error. II. Merchandise Trade

As disaggregated here, the major components of U.S. merchandise trade are non-oil imports and nonagricultural exports. Demand for imports is modeled assuming that imported goods are imperfect substitutes for domestic goods; empirically, supply is best represented as a variable price markup over cost. While foreign exporters seem to exhibit some strategic pricing behavior, the U.S. market for exports is best represented 2 Granger and Engle (1985). See also Hendry (1986) and Granger (1986). This procedure is appropriate for series that are integrated of order one (/(1)) and are part of a linear combination of variables that yield /(0) residuals, implying cointegration.

402 • A Model of the U.S. Current Account

by perfect competition, with exports being viewed as imperfect substitutes for foreign goods on the demand side. Oil imports and agricultural exports are modeled separately because of the unique behavior of these components; both models contain elements of the perfect substitutes model under perfect competition. 3

Model of Non-Oil Imports and Nonagricultural Exports

Imports are modeled as imperfect substitutes for domestic goods so that the volume of non-oil imports demanded depends positively on a domestic scale variable and negatively on the price of imports relative to the domestic price level. On the supply side, the presence of product differentiation or other market imperfections is assumed to give foreign suppliers some degree of control over the price they charge. The price of imports is therefore not determined in the U.S. market, but is determined by a variable price markup relative to foreign costs, where suppliers adjust the markup in response to changes in competitive pressures in the U.S. market. 4 In level terms, the foreign export price (PX*) is equal to the markup, &, times foreign unit labor costs ( ULC*), where the markup is a function of U.S. export prices adjusted for exchange rates (PX ER) relative to foreign costs. The exchange rate is expressed in units of foreign currency per U.S. dollar. PX*

=

& ULC* = [PX ER I ULC*)b 1 ULC*.

(1)

Taking logs and assuming U.S. import prices are approximately equal to foreign export prices adjusted for exchange rates yields the following foreign export price and domestic import price equations. 5

px*

=

b1 (px + er) + (1 - b 1) ulc*

pmno = px* - er = b1 px + (1 - b 1) (ulc* - er).

(2) (3)

The price of non-oil imports is thus a weighted average of the domestic export price and foreign unit labor costs adjusted for exchange rates. Nonagricultural exports are modeled according to perfect competition, assuming that foreigners view exports as imperfect substitutes for their own domestic production. The volume of exports demanded (xnad) is 3

See Goldstein and Khan (1985) for a review of alternative trade equations. See Hooper and Mann (1989) and Woo (1984) for similar models. Hooper and Mann also allow for the effects of demand pressures as measured by capacity utilization in both the home and foreign markets. These effects were not significant in the present model. 5 All lowercase variables are in logs. 4

Ellen M. Nedde • 403

therefore a positive function of a foreign scale variable (y*) and a negative function of the price of exports (pxna) relative to foreign export prices adjusted for exchange rates (px$* = px* - er). The supply of exports is a positive function of the price of exports relative to domestic costs.

= c1 y* - c2 (pxna - px$*) xnas = dt (pxna - ulc ). xnad

(4) (5)

Setting export demand equal to export supply and solving for the equilibrium quantity yields a volume equation in which exports are a positive function of the foreign scale variable and a negative function of U.S. unit labor costs relative to the dollar price of foreign exports. c1 y* - c2 (lld 1 xna + ulc - px$*)

xna

=

xna

= C1 y* - C2 (ulc - px$*)

Ct

(6)

= Ct dtf(ct + dz), Cz = Cz dtl(ct + dz).

The price of exports then becomes a positive function of foreign demand and a weighted average of domestic unit labor costs and foreign export prices, adjusted for exchange rates.

lld 1 xna + ulc

lld 1 (C1 y* - C2 ulc + C2 px$*) + ulc

pxna

=

pxna

= D 1 y* + (1 - D 2) ulc + D 2 (px* - er) = ctf(ct + dz), Dz = Czl(ct + dz).

Dt

=

(7)

Finally, combining equations (2) and (7) allows U.S. and foreign export prices and the U.S. import price to be written as weighted averages of domestic and foreign unit labor costs. 6

px* = E 1 y* + (1- £ 2) (ulc + er) + £ 2 ulc* pxna

= F1 y* +

(1 - F2 ) ulc + F2 (ulc* - er)

E 1 = b 1 D/(1 - b 1 Dz), Ez Ft

pmno

= =

(8)

Dt/(1 - bt Dz), Fz

=

=

(9)

(1 - bt)/(1 - bt Dz) < (1 - bt)

Dz (1 - bt)/(1 - bt Dz) < Dz

E 1 y* + (1 - £ 2) ulc + E2 (ulc* + er).

(10)

6 Researchers are often interested in "pass-through" coefficients that describe the effects of exchange rates and foreign costs or prices on domestic trade prices. One should note that in the above model, for the same structural coefficients, the pass-through coefficients £ 2 and F2 for the effects of foreign costs on import and export prices are less than the coefficients (1 - b1) and D2 for the pass through of foreign export prices.

404 • A Model of the U.S. Current Account

Non-Oil Imports and Nonagricultural Exports in the Long Run

In the long run, the volume of non-oil imports was found to depend upon real U.S. total domestic demand and the relative price of imports adjusted by the average tariff rate. 7 The estimated demand elasticity is equal to 2.3 in the long run, a value that is in line with other estimates. 8 Others have included real GNP as the activity variable to account for the intermediate goods that the United States imports. An alternative longrun equation using real GNP was therefore tested; while the elasticity was little changed, the R 2 and cointegration statistics declined. A long-run price elasticity of -0.9 was estimated using a fixed-weight price index for non-oil imports; when the relative price term was constructed using the implicit deflator, this elasticity was slightly smaller in absolute value. Helkie and Hooper9 have attributed what is viewed as a high income (or demand) elasticity to the failure of trade equations to adequately capture longer-term supply effects. To capture these effects, they include the ratio of foreign to domestic capital stocks in their trade volume equations. One might expect U.S. exports to be a negative function and imports to be a positive function of this upward-trending variable. Various versions of the Helkie-Hooper model have produced mixed results on the importance of these effects, however; while the terms enter with the correct sign, they have not always entered significantly. When a trend term was added to the non-oil import volume equation in the present model, its estimated coefficient was negative and caused a significant increase in the estimated elasticity of demand. The fixed-weight price index for non-oil imports was found to be cointegrated with domestic unit labor costs and foreign unit labor costs, adjusted for exchange rates. The equation satisfies homogeneity since the sum of the coefficients on domestic costs and foreign costs adjusted for the exchange rate is approximately unity. 10 The equation also satisfies equal pass-through of foreign costs and exchange rates-a result that is expected from theory, but not always accepted by the data. In particular, one may not want to impose this equality in the short run since suppliers may respond differently to changes in exchange rates (which may be 7 In addition to these /(1) variables, the dock strike dummy developed by Isard (1975) was included. Being an /(0) variable, its inclusion should not influence cointegration tests, but does assist in estimatin~ unbiased long-run coefficients. 8 See the tabulation following equation (2) m Annex I for a comparison of alternative estimates. Sources are Dunaway (1988), Helkie and Hooper (1988), and Krugman and Baldwin (1987). 9 Helkie and Hooper (1988). 10 It is not possible to test this hypothesis statistically since the coefficients, while consistent, do not have an asymptotic normal distribution.

Ellen M. Nedde •

405

viewed as temporary) and changes in foreign costs (which may be viewed as more permanent). The importance of domestic unit labor costs in import prices (a weight of0.5) indicates that foreign suppliers pay considerable attention to U.S. cost conditions even in the long run.U If the United States were a price taker (setter) in the import market, one would expect a coefficient of unity (zero) on foreign costs adjusted for the exchange rate and a coefficient of zero (unity) on domestic costs. Others have noted the difficulties of estimating implicit deflators for non-oil imports (and nonagricultural exports). 12 Largely because prices of computers have been falling rapidly in recent years and because computers represent a growing share of U.S. trade, implicit deflators for U.S. exports and imports do not appear to reflect changing costs and exchange rates in a stable way. In particular, long-run cointegrating relationships for implicit trade deflators are difficult to find without including the price of computers explicitly and permitting somewhat large deviations from homogeneity. However, even if fixed-weight indices are used in the relative price terms of volume equations, implicit trade deflators need to be estimated in order to convert from volumes to values of merchandise trade. Implicit deflators were therefore estimated as functions of domestic costs, foreign export prices, the exchange rate, and the price of business machinery in the long run, even though homogeneity and equal pass-through of foreign prices and exchange rate changes would not be satisfied. 13 The long-run relationship estimated for the implicit deflator for non-oil imports includes the foreign export price, the exchange rate, the price of business machinery, and a trend term. 14 The volume of nonagricultural exports in the long run was found to depend positively on real foreign domestic demand and negatively on U.S. unit labor costs relative to foreign export prices adjusted for exchange rates. The addition of a deterministic trend was required to achieve cointegration. This term, entering with a negative estimated coefficient, is not inconsistent with the supply variable proposed by Helkie and Hooper, but given the mixed results associated with their relative capital stock term, it is not certain what the source of this unexplained time trend might be. The estimated long-run demand elasticity, at 2.1, is slightly less than the import demand elasticity, retaining Melick (1990) estimates similar long-run coefficients for the price of manufactured imports. 12 See Citrin (1992) and Meade (1990). 13 An equation linking changes in the implicit deflator to changes in the fixedwei¥,ht pnce index performed poorly in dynamic simulations. 1 See the tabulation following equation (6) in Annex I for a comparison of alternative estimates. 11

406 • A Model of the U.S. Current Account

the gap that is typically found between these elasticities. The price elasticity, at -0.5, is at the low end of other estimatesY The fixed-weight index for prices of nonagricultural exports was found to be determined primarily by domestic costs in the long run. The equation approximately satisfies homogeneity since the sum of the cost coefficients is close to unity. Demand did not enter the long-run equation, but as seen below, does influence price in the short run. The implicit deflator for nonagricultural exports was found to be cointegrated with domestic unit labor costs, foreign export prices, the exchange rate, and the price of business machinery. 16 Non-Oil Imports and Nonagricultural Exports in the Short Run

The short-run error correction equation for non-oil imports contains the independent variables of the long-run equation (but expressed in difference form) and the error correction term, which is equal to the residuals from the long-run equation lagged one period. The short-run demand elasticity of 1.4 is less than its long-run value, but the coefficient on the error correction term, at -0.5, implies relatively fast adjustment to equilibrium values. In contrast, the fixed-weight price of non-oil imports adjusts relatively slowly to its long-run equilibrium. All short-run coefficients are less than their long-run values, implying a gradual increase over time of the impact of changes in domestic and foreign costs and the exchange rate. The implicit deflator also adjusts slowly to equilibrium, with a coefficient on the error correction term of -0.2. Short-run movements in the fixedweight index improve the fit of the equation for the deflator significantly; these movements are the principal determinants of the implicit deflator in the short run, while foreign export prices, the exchange rate, and computer prices are its long-run determinants. As in the import volume equation, the short-run elasticities in the equation for nonagricultural exports are less than their long-run values. Changes in relative prices enter only after a two-period lag with an elasticity of -0.4. In the price equations, changes in the fixed-weight price index depend primarily on its own lagged values and lagged changes in demand. While differences in domestic and foreign costs enter the short-run equation, the exchange rate enters only through the error correction term. The implicit deflator depends on first differences of its long-run explanatory variables; it does not contain the demand effects that enter the fixed-weight price equation. 15

See the tabulation following equation (8) in Annex I. See the tabulation following equation (12) in Annex I for a comparison of alternative estimates. 16

Ellen M. Nedde • 407

Model of Oil Imports and Agricultural Exports

Oil imports are assumed to be perfect substitutes for domestic oil and are calculated as the residual from domestic consumption, production, exports, and stock changes. The latter three variables are not estimated. Consumption (in thousands of barrels a day) is modeled as a positive function of a domestic scale variable and a negative function of the price of oil imports relative to the domestic price level. The dollar a barrel price of imports is then used to translate to the dollar value of petroleum imports. The presence of subsidies and trade restrictions in the U.S. market for agricultural products makes agricultural exports difficult to estimate. The supply of agricultural exports in the United States seems to be best described by the perfect substitutes model, in which exports are the difference between supply and demand for agricultural goods. Domestic supply (as) is a positive function of the price of agricultural goods (pxa) relative to the domestic price level (or costs), while domestic demand for agricultural goods depends negatively on the relative price of agricultural goods and positively on a domestic scale variable. The supply of exports-which is equal to the difference between domestic supply and domestic demand-is therefore a negative function of the scale variable and a positive function of the relative price.

as=el(pxa-p) ad= / 1 y -

fz (pxa - p)

xas =as- ad= (e1 + fz) (pxa - P) -/1 Y ·

(11) (12) (13)

The demand for agricultural exports abroad is a positive function of a foreign scale variable and a negative function of the price of agricultural exports relative to foreign export prices, adjusted for exchange rates. Equating demand and supply yields agricultural exports as a positive function of the foreign scale variable and the relative price of agricultural goods domestically and a negative function of the domestic scale variable and the relative price of exports abroad. The export price depends positively on domestic and foreign scale variables and prices.

Oil Imports and Agricultural Exports in the Long Run

A long-run cointegrating relationship could not be found for the consumption of oil so that an error correction representation was not used for this equation. The short-run dynamic equation for oil and its long-run implications are discussed below.

408 • A Model of the U.S. Current Account

The volume of agricultural exports was found to be cointegrated with the expected demand and relative price terms, with the exception of domestic demand, which did not enter the equation significantly. While the model described above appears appropriate for the volume of trade, domestic and foreign costs and demand were poor explanatory variables for the export price and a long-run cointegrating relationship among these variables could not be found. The agricultural export price deflator moves closely with the commodity price index in the IMF's International Financial Statistics (IFS), but these two series also were not cointegrated. Because of the importance of the medium-term simulation properties of the model, a long-run relationship was imposed that implied proportionality between the agricultural export price and the IFS commodity price index, with a more general specification in the short run.

Oil Imports and Agricultural Exports in the Short Run

The short-run dynamic equation for oil consumption was estimated in both level and difference form. Coefficients were very similar and both equations performed satisfactorily in terms of diagnostic tests, but the equation in level terms performed better in dynamic simulations (the in-sample mean absolute error was reduced significantly). Oil consumption was found to depend on its own lagged values (up to five quarters) with the coefficients approximately summing to unity. Short-run demand and price elasticities for oil consumption were estimated at 0.8 and -0.1, respectively. The short-run dynamic equation for agricultural exports was estimated using a general specification that included differences (contemporaneous and lagged) of the relative price terms and demand terms, as well as lagged values of the dependent variable. However, sequential testing down reduced the equation to one with rather sparse dynamics, with only the dock strike dummy and the error correction term entering significantly, although none of the diagnostic tests rejected white noise residuals.17 In the short run, first differences in the agricultural export price are a function of its own lagged values and lagged differences of the IFS commodity price index. In the long run, the price is constrained to move with world commodity prices. The coefficient on the error correction terms was low ( -0.13), but significantly different from zero (t = -3. 9). 17 Hendry (1989) refers to this dynamic model as the dead-start model, since the dependent variable depends only on lagged values of the independent variables.

Ellen M. Nedde • 409

Ill. Trade in Services Model of Service Payments and Receipts

Trade in services includes travel, passenger fares, other transportation (freight and port services), royalties and license fees, and other private services such as education and business services. These services were aggregated and estimated in real terms using deflators for "other service" imports and exports from the national income and product accounts. 18 Real service payments are expressed as a function of a domestic scale variable, the price of service payments relative to the domestic price level, and the volume of merchandise imports and exports to capture the effects on freight and port services of growing international trade. Real service receipts are expressed analogously. Reflecting the prices used to deflate individual components of service payments and receipts in the national accounts, the deflator for service payments is expressed as a weighted average of the U.S. GNP deflator and the foreign consumer price index adjusted for exchange rates, and the deflator for services receipts is expressed as a weighted average ofthe U.S. GNP deflator and the U.S. consumer price index.

Service Payments and Receipts in the Long Run

Service payments and receipts were each found to be cointegrated with total domestic demand, relative prices, and merchandise trade volumes. Both long-run equations also contain several dummy variables to reflect recent revisions to the data. 19 The demand elasticity in the payments equation was estimated to be relatively low at 0.4, compared with an elasticity in the receipts equation of 1.0. This divergence offsets somewhat the effects of a higher elasticity of demand in the merchandise import equation relative to that of the export equation. However, since service flows are smaller than merchandise flows and because service payments respond more strongly to growth in merchandise trade than do receipts, the model still has the property that the current account grad18 The term "services" in this paper follows the new U.S. classification that excludes investment income. In contrast to Dunaway (1988), these aggregate equations were found to have smaller standard errors than individual equations for travel, transportation, and other services. 19 Trade in educational services was added beginning in 1981; a new survey for travel and passenger fares implies more complete coverage beginning in 1984; and other private services incorporate the results of a new benchmark survey beginning in 1986.

41 0 • A Model of the U.S. Current Account

ually deteriorates under assumptions of equal U.S. and foreign growth and unchanged real exchange rates. The deflator for service receipts was found to be cointegrated with the U.S. consumer price index and the GNP deflator as expected. However, the deflator for service payments was not cointegrated with the expected price series, although the estimated coefficients were consistent with theory and approximately satisfied homogeneity. Including the fixedweight price index for GNP rather than the implicit deflator had no effect on the estimated coefficients or the results of the tests for cointegration. However, as in the case of the agricultural export price, the medium-term simulation properties of the model were given precedence over its statistical properties, and the dynamic equation was estimated including this error correction term. 20 Service Payments and Receipts in the Short Run

Real service payments and receipts both respond strongly to deviations from equilibrium in the short run, with coefficients on the error correction terms of -0.7 and -0.5, respectively. Payments respond more strongly to changes in demand in the short run (0.7) than in the long run (0.4); in contrast, service receipts do not respond to short-run changes in demand except through the error correction term. The deflators for payments and receipts of services both depend on their own lagged values in the short run, as well as on lagged price levels and the error correction terms. Despite the poor performance in cointegration tests of the long-run equation for the payments deflator, the error correction term enters significantly (t = - 3.3), although with a relatively low coefficient (-0.1). IV. Investment Account Model of Investment Income

Investment income includes direct investment income, interest income on private portfolio assets, and interest income on government portfolio assets. Data on direct investment income payments and receipts are collected, while interest income is estimated by the Bureau of Commerce by applying an imputed interest rate to estimated stocks of assets and liabilities. The estimation method employed here mimics these tech20 As will be discussed in Section V, this short-run equation performed well in both in- and out-of-sample in dynamic simulations.

Ellen M. Nedde • 411

niques by calculating an implicit rate of return on private and government portfolio assets and liabilities and estimating the rate of return as a function of the U.S. Treasury bill rate. 21 For direct investment, income payments and receipts are estimated directly as functions of the relevant (book value) stocks, capacity utilization in the countries where the investments are located, and quarterly seasonal dummies. The receipts equation also includes the price of oil since a significant portion of U.S. direct investment abroad is in the petroleum industry. 22 Estimated asset and liability stocks are required to forecast investment income payments and receipts. The error correction methodology was not used for these equations, rather stocks were estimated as functions of their own past values, time trends, and quarterly dummy variables. Estimated Investment Income

The implicit rates of return on private portfolio liabilities and assets were found to be cointegrated with the U.S. Treasury bill rate (with long-run coefficients of 0. 7 and 0. 9, respectively). Long-run cointegrating relationships did not exist between the implicit rates of return on government portfolio assets and liabilities and the U.S. Treasury bill rate. In the short run, all four implicit returns depended on their own lagged values and contemporaneous and lagged values of the U.S. Treasury bill rate. 23 With the exception of the price of oil, which was found to influence direct investment income receipts only in the short run, cointegration tests confirmed the long-run relationships for direct investment income payments and receipts described above. Capacity utilization effects enter strongly in both the long run and the short run. The estimated coefficient on the asset stock was close to unity in the long run (1.1), which is 21 The U.S. Treasury bill rate is used as the independent variable in all four equations since most of U.S. portfolio assets and liabilities are dollar denominated. This treatment follows Helkie and Stekler (1987). 22 Capital gains and losses are excluded from these income series for purposes of estimation. Realized capital gains and losses due to differences between the sale price and book value of an asset are difficult to estimate without knowing the timing of particular transactions. Unrealized capital gains and losses associated with exchange rate changes were not estimated since they are currently being removed from direct investment income. These adjustments will appear instead as valuation adjustments in the stocks of foreign direct investment assets and liabilities. These revisions are not incorporated in the data used by the model. 23 The return on government portfolio assets has been quite volatile since 1987. A dummy in the first quarter of 1988 was included to account for the rescheduling ofinterest receipts from Egypt; however, the Jarque-Bera (1980) test still rejected normal residuals.

412 • A Model of the U.S. Current Account

somewhat surprising given the large discrepancy between this book value measure and the market value of foreign direct investment assets; the coefficient on the liability stock was estimated at 0. 7. As found by others, estimated standard errors are high for these equations. In the capital account, estimated equations for asset and liability stocks imply estimated gross capital flows, which are unlikely to match the estimated current account. Rather than place all discrepancies between estimates of the current account and capital flows in the statistical discrepancy, the inflow of private portfolio investment is taken as a residual to ensure that estimated net capital flows are consistent with the required financing of the estimated current account balance. In other words, the inflow of private portfolio investment (LPNET) is taken as the residual from private portfolio investment outflows (APNET), the balance on the current account (BCUR), net inflows of foreign direct investment (LDNET- ADNET) and government portfolio investment (LGNET - ADNET), the statistical discrepancy (SD), and the allocation of SDRs (SDRA ). (The statistical discrepancy and allocation of SDRs are taken as exogenous to the model.) LPNET

=

APNET - BCUR - (LDNET - ADNET) - (LGNET- AGNET) - SD - SDRA.

(14)

The stock of private portfolio liabilities (LP) is then calculated using cumulated gross flows. While only the net flows need satisfy this balance of payments identity, the distribution of the gross flows are important for forecasting investment income since the United States receives and pays a wide range of implicit rates of return on its assets and liabilities. Almost all assets and liabilities depend on a trend term in addition to their own lagged values. Diagnostic tests indicated that residuals from the equations for official assets and liabilities suffered from heteroscedasticity, possibly a result of increased intervention in the 1980s. V. Predictive Performance of the Model

This section discusses the performance of the model in static and dynamic simulations. 24 The equations were estimated over the period 1969:1-1987:2, retaining 1987:3-1989:2 for out-of-sample forecasts. 24 Static simulations use actual data as lagged values, while dynamic simulations use values generated by the model. Both kinds of simulations are deterministic, in the sense that the estimated coefficients and error terms in the model are taken to equal their expected values.

Ellen M. Nedde • 413 Table 1.

Root Mean Square Percent Errors 1

1971:4-1987:2

1987:3-1989:2

Static

Dynamic

Static

Dynamic

Non-oil import volume Fixed-weight price Implicit deflator

2.6 1.3

1.1

4.0 2.8 2.6

1.7 0.2 0.8

2.3 0.4 1.7

Oil import volume

5.8

19.7

4.7

4.2

Nonagricultural export volume Fixed-weight price Implicit deflator

2.5 0.9 0.8

3.7 2.4

6.1 0.5

1.3

3.2 0.5 0.6

Agricultural export volume Implicit deflator

7.0 1.9

10.7 4.4

6.6 2.2

8.2 7.5

Real service payments Implicit deflator

2.0 1.0

2.9 2.4

2.2 0.9

2.7 2.2

Real service receipts Implicit deflator

2.0

2.3 2.8

0.9

1.3

2.0 2.5

1.1

1.3

Trade balance 1.6 3.7 1.3 1.6 1.6 2.7 Net services 1.3 1.3 4.2 Net direct investment income 3.5 5.5 2.5 4.7 1.9 7.3 1.8 Net portfolio income Balance on services and investment income 2.6 1.4 1.5 1.9 1.0 0.9 Current account 1.4 2.9 1 Errors are expressed as a percentage of actual values for gross flows; for balances, errors are expressed as a percentage of actual underlying gross flows.

Root mean square percent errors for both periods for the major components of the current account are given in Table 1. The larger components of merchandise trade, non-oil imports and nonagricultural exports, tended to perform better than the smaller, more volatile, components of oil imports and agricultural exports both in and out of sample (Figures 1 and 2). The nonagricultural export equation began to underpredict post sample in the dynamic simulation, but otherwise the major components of trade volumes tracked the data closely. One marked difference among the four volume equations was the absence of an error correction term in the oil import equation and its tendency to miss some major turning points in the data. 25 The direction of the trade 25 Percent errors for oil imports appear large relative to the standard error of the estimated equation (2.5 percent) due to the translation from oil consumption to (the smaller volume of) oil imports. The root mean square percent errors for oil consumption are 2.3 and 2.0 for the static simulation and 8.3 and 1.8 for the dynamic simulation.

414 • A Model of the U.S. Current Account

Figure 1. United States: Actual and Predicted Import Volumes 100

Non-oil imports (In billions of 1982 U.S. dollars)

80

60

40

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

3or------------------,------------------------~--_,

Oil imports (In billions of 1982 U.S. dollars)

25

20

15

Dynamic

~

,•\,

':

10

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

Ellen M. Nedde • 415

Figure 2. United States: Actual and Predicted Export Volumes

90

Nonagricultural exports (In billions of 1982 U.S. dollars)

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

12.------------------------------------------,.---. Agricultural exports (In billions of 1982 U.S. dollars)

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

416 • A Model of the U.S. Current Account

volume errors post sample tends to be somewhat offsetting, with small underpredictions of the major components and small overpredictions of agricultural exports and oil imports. Figures 3 and 4 contain simulations of implicit deflators and fixedweight price indices for merchandise trade. Again, the smaller, more volatile components of trade produced larger percent errors than the two major components. As others have found, the import deflator increased by Jess than predicted during 1985-86, but these prediction errors were eliminated by mid-1987. In the post-sample period, the model began to overpredict the deflator for agricultural exports, but tracked closely both fixed-weight prices and implicit deflators for non-oil imports and nonagricultural exports. The portion of the model describing services performed relatively well in sample, with root mean square percent errors under 3 percent for both dynamic and static simulations (Figures 5 and 6). This ability to track the data closely continued in the post-sample period, with nearly all of the root mean square percent errors declining in value. Recall that tests for cointegration did not support the long-run equation for the service payments deflator, yet the dynamic equation for that deflator performs slightly better than the deflator for service receipts both in and out of sample. Figures 7-9 contain simulation results for the balances on portfolio income, total investment income (excluding capital gains and losses), the combined service and income account, the trade account, and the current account. Root mean square errors are calculated as percentages of the underlying gross flows relevant to the balance. 26 The predicted balance on portfolio income is partly influenced by errors in forecasting the current account, since these errors are accumulated in the stock of private portfolio liabilities. Because the current account balance was overpredicted in the dynamic simulation by small amounts over much of the in-sample period, the stock of liabilities was underestimated, causing the balance on portfolio income to be overestimated. During the out-of-sample period, both the balance on the current account and private portfolio income payments were forecast quite accurately; the overprediction of the balance on portfolio income was largely the result of an overprediction of government income receipts (arising from errors in both stocks and implicit returns). The combined balance on services and income was overpredicted slightly in sample because of the influence of portfolio balance errors. Out of sample, the balance was predicted more accurately because of offsetting errors. 26 For example, errors in the balance on portfolio income are expressed relative to the sum of private and government portfolio income receipts and payments.

Ellen M. Nedde • 417

Figure 3. United States: Actual and Predicted Trade Deflators

120 110

Non-oil import deflator (Index. 1982 = 100)

100 90 80 70 60

so 40 30 1969

110 100

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1977

1979

1981

1983

1985

1987

1989

1977

1979

1981

1983

1985

1987

1989

Nonagricultural export deflator (Index, 1982 = I 00)

90 80 70 60

so 40 30 1969

140 120

1971

1973

1975

Agricultural export deflator (Index, 1982 = I 00)

100 80 60 40 20 1969

1971

1973

1975

418 • A Model of the U.S. Current Account

Figure 4. United States: Actual and Predicted Fixed-Weight Trade Prices 140.-------------------------------~---------.---.

Fixed-weight non-oil import price (Index. 1982 = I 00)

120

100

so 60

40

20

1969

1971

1973

1975

1977

1979

1981

110 100

1983

1985

1987

1989

1985

1987

1989

Dynamic

Fixed-weight nonagricultural export price (Index. 1982 = I00)

90 80 70 60 50 40 1969

1971

1973

1975

1977

1979

1981

1983

Ellen M. Nedde • 419

Figure 5. United States: Actual and Predicted Real Services

16

Service payments (In billions of 1982 U.S. dollars)

14

12

10

8

6 1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1979

1981

1983

1985

1987

1989

22 Service receipts

20

(In billions of 1982 U.S. dollars)

18 16 14 12 10 8 6 4 1969

1971

1973

1975

1977

420 • A Model of the U.S. Current Account

Figure 6. United States: Actual and Predicted Service Deflators

140

Service payments deflator (Index. 1982 = I 00)

120

100

80

60

40

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1977

1979

1981

1983

1 0. The vector a is then called the cointegrating vector. 43 Several observations can be made about the concept of cointegration. Cointegration applies to the long-run relationships between variables, and describes a situation in which variables that may drift apart in the short run have a tendency to move together in the long run. As such it is well suited for determining the long-run link between the forcing variables in structural exchange rate models and exchange rates. Second, a necessary condition for exchange rates to be cointegrated with a set of variables is that all these variables are integrated of order one, given that the exchange rate is an J(l) variable. It is possible, however, to have variables that are integrated of a higher order, provided that there are cointegrating relationships among these variables that reduce their collective order of integration to that of the exchange rate. 44 Finally, if cointegrating relationships can be found, it is possible to specify an error-correction mechanism in which either the exchange rate or the forcing variables can be used to predict the rate at which the economy will return to long-run equilibrium. 45 The methodology of cointegration is applied in this section in three steps. First, the forcing variables identified in structural exchange rate models are individually tested to determine their order of integration. Second, tests are undertaken to determine whether exchange rates are cointegrated with these variables, yielding a number of cointegrating relationships. Finally, at the last stage, we test to see whether deviations of exchange rates from long-run values implied by the cointegrating vectors contain information for predicting future movements in exchange rates. Based on the review of structural exchange rate models, four sets of forcing variables are tested: money supplies and interest rate differentials; fiscal variables, including government spending, fiscal deficits, 43 For a simplified approach to cointegration, see Hendry (1986) and Granger (1986). 44 Hence, for example, individual variables such as the stock of foreign assets can be integrated of higher order than the exchange rate, provided that there are linear combinations of these and other variables that are integrated of the same order as the exchange rate. 45 As discussed in Section I, an ability to predict changes in the exchange rate is not necessarily inconsistent with market efficiency or the random walk property of exchange rates. A univariate random walk in exchange rates only implies that exchange rates cannot be predicted from their own history.

Charles Adams and Bankim Chadha • 483

and stocks of outstanding public sector debt; current account balances, in level terms and cumulated to give net foreign asset positions; and measures of economic activity or productivity, including real GNP/GDP, industrial production, labor productivity, and-as a proxy for capital productivity-real share prices. Given the large number of possible combinations of these variables, some simplification is obviously necessary. To make the number of tests manageable, all domestic variables are measured relative to their foreign counterparts. In the case of money supplies or interest rates, this is a reasonable approximation: the (logarithmic) difference between domestic and foreign money supplies appears in many structural models as does the differential between domestic and foreign interest rates. It is less apparent, however, how variables such as government expenditures, fiscal deficits, and public debt stocks should be treated. The approach adopted was to use the difference between domestic and foreign government expenditures as shares of GNP, and differences in the ratio to GNP of fiscal deficits and public debt across countries. 46 Details on how other variables are measured are provided in the accompanying tables. The results for the order of integration of forcing variables are presented in Table 8. For each variable, Dickey-Fuller and augmented Dickey-Fuller tests were used to test the null hypothesis that a given series was either integrated of order one or two. The results can be summarized as follows. Among the monetary variables, relative money supplies in all cases are integrated of order one, implying that they can potentially account for the nonstationarity in exchange rates. Interest differentials, both nominal and real, however, appear to be stationary and cannot therefore be cointegrated with exchange rates. The finding that real interest rate differentials are stationary is consistent with the findings of Meese and Rogoff (1986). 47 They interpreted this result to imply that the variance of the real exchange rate reflects changes in the (expected) long-run real exchange rate. 48 46 Combining domestic and foreign variables in this way also ensures that the forcing variables are all integrated of the same order as exchange rates. See below. 47 See also Meese and Singleton (1982). 48 This can be seen by writing the difference between the period t real exchange rate and the expected future real exchange rate in period t + k in terms of the real interest rate differential over k periods: E,q,+k- q, = k· [r,- ri]. If the nonstationarity in q, is not accounted for by the real interest differential, it must be accounted for by E,q, + k· Of course, if uncovered interest rate parity does not hold, the nonstationarity in the real exchange rate could be explained by nonstationarity in the risk premium.

484 • Structural Models of the Dollar Table 8. Estimated Order of Integration for Forcing Variables from Dickey-Fuller and Augmented Dickey-Fuller Tests

Variable Log(M1) 3-month differential 5-year differential 10-year differential 3-month real differential 5-year real differential 10-year real differential Log(GIY) DIY (JD)IY Log (Y') Log(RY) Log(RV) Log(XIL) Log(ULC)

UK

GR

CA

JA

Money and interest rates /(1) /(1) /(0) /(0) /(0) J(O) /(0) J(O) /(0) J(O) /(0) /(0) /(0) /(0)

/(1) /(0) /(0) /(0) /(0) /(0) /(0)

/(1)

Fiscal measures /(1) /(1) /(1) J(O) /(1) /(1)

/(1) /(1) /(1)

/(1)

Output and productivity /(1) /(1) /(1) /(1) /(1) /(1) /(1) /(1) /(0) /(1)

/(1) /(1) /(1) /(1) /(1)

/(1)

/(0)

J(O) /(0) /(0)

/(1)

/(1) /(1) /(1)

Current account and foreign asset stocks /(1) /(1) /(1) CABIY /(1) /(1) /(1) /(1) /(1) (JCAB)IY Note: All data are from International Monetary Fund, International Financial Statistics, various issues. All variables are measured as the value for the United States minus the value for the foreign country. Variable notation is as follows: G denotes nominal government consumption; Y denotes GNP at market prices; D denotes the central government balance; f is used to denote cumulative value of; Y' denotes industrial production; RY denotes real GNP; RV denotes real industrial share prices; (XI L) denotes output per man-hour in manufacturing; ULC denotes unit labor costs in manufacturing; and CAB denotes the domestic currency value of the multilateral current account balance. "l(n )" denotes that variable is integrated of order n; that is, needs to be differenced n times to obtain a stationary variable. The Dickey-Fuller and augmented Dickey-Fuller regressions include a constant and three seasonals. Real interest rates are computed using an AR(2) process for predicting inflation.

All the fiscal variables, with the exception of the (relative) U.K. fiscal deficit to GNP ratio, are integrated of order one and thus potential candidates for explaining the long-run variance of exchange rates. We interpret this to mean that differences in fiscal policy across countries hold the potential for explaining the long-run variance of exchange rates, independently of any short-run impact on interest rate differentials. The current account and cumulative net foreign asset positions of major

Charles Adams and Bankim Chadha • 485

industrial countries are also integrated of order one. All the measures of economic activity and productivity, with the exception of relative unit labor costs for the United Kingdom, are also characterized by the same order of integration as exchange rates. Explanations for the dollar's movements based on real factors such as productivity developments may therefore be able to explain long-run exchange rate movements. The tests for cointegration are based on the maximum-likelihood procedures developed by Johansen (1988) and Johansen and Juselius (1989). Unlike the Granger-Engle procedure, which presumes the (potential) presence of exactly one cointegrating vector among a set of variables, estimated by regressing one of the variables on the contemporaneous levels of the other variables, the Johansen procedure allows for as many cointegrating vectors as the number of variables. Before considering the relationships between exchange rates and other economic variables, it is interesting to determine whether the nonstationarity in dollar exchange rates derives from a single source, such as U.S. economic policies-that is, there is a "dollar phenomenon"-or reflects a more complex interaction with economic developments in particular countries. Accordingly, we tested whether the bilateral exchange rates of the dollar were themselves cointegrated and reflected a common source of nonstationarity. The results from these tests suggested that dollar exchange rates are not cointegrated, which means that long-run movements in the dollar are not dominated by developments in the United States and, hence, if there are explanations of long-run movements in dollar exchange rates, they must be of a (relative) country-specific nature. 49 The results from individual bivariate cointegration tests of each exchange rate and other economic variables are summarized in Table 9. In order to highlight the central results, the table does not show the values of individual test statistics and only indicates whether cointegrating relationships were found for particular variables. 50 The results suggest that little of the nonstationarity in exchange rates mirrors that of other economic variables in the long run. Nevertheless, a number of cointegrating vectors are identified that carry implications for understanding exchange rate behavior. 49 In order to conserve space, the results from these tests are not recorded in the tables. 50 As indicated earlier, the cointegration tests are based on the procedure developed by Johansen (1988). This procedure identifies a set of cointegrating relationships between the exchange rate and other economic variables. An alternative approach would be to use the cointegration procedures developed by Stock and Watson (1986).

yes

yes yes

yes

0

0

yes

-

CA JA

2

yes

yes

a. (I)

0

iii' -.

Q.

:T (I)

Q.

(if

s:: 0

CABIY (JCAB)IY

u c:

~

yes yes yes yes yes yes Note: All variables are measured as the value for the United States minus the value for the foreign country. Variable notation is as follows: G denotes nominal government consumption; Y denotes GNP at market prices; D denotes the central government balance; f is used to denote cumulative value of; Y' denotes industrial production; RY denotes real GNP; RV denotes real industrial share prices; (X!L) denotes output per man-hour in manufacturing; ULC denotes unit labor costs in manufacturing; and CAB denotes the domestic currency value of the multilateral current account balance. Table entries are as follows: "-" denotes no cointegrating vector found; "yes" denotes at least one cointegrating vector found at the 5 percent significance level; and " ... " denotes either data unavailable or that the variable was found to be stationary (see Table 8). A prior constant and three seasonals were first removed from the data.

Real UK GR

JA

0

log(ULC)

•

q}

Current Account and Foreign Asset Stocks

Results of Bivariate Cointegration Tests for Exchange Rates and Other Economic Variables

Output and Productivity Money Fiscal Variables Exchange Rate log(Ml) log(GY) DIY (JD)IY log(Y') log(RY) log(RV) log(XIL) Nominal yes UK yes GR yes yes CA yes yes

Table 9.

~ (X)

m

Charles Adams and Bankim Chadha • 487

Confirming results found with structural exchange rate models, monetary variables do not account for the long-run variance of nominal and real exchange rates. 51 The lack of cointegration between real exchange rates and monetary variables is to be expected on the basis of most structural models. The inability to find cointegration between nominal exchange rates and relative money supplies suggests that differences in monetary policy across major industrial countries have not been a major factor behind trends in nominal exchange rates over the recent floating rate period, or at least as measured through monetary aggregates. Fiscal variables explain some of the long-run variance in real and nominal exchange rates in all cases other than the U.S. dollar-yen exchange rate. In cases where cointegrating relationships are found, they either reflect the ratio of government spending to GNP, or fiscal deficit positions. No cointegrating relationships were found between public sector debt stocks and nominal or real exchange rates. Current account variables account for some of the long-run variance in exchange rates, particularly in the case of the exchange rate with the Canadian dollar. There is also some evidence of cointegrating relationships for the deutsche mark and Japanese yen exchange rates, but contrary to the predictions of portfolio models, current account variables do not systematically account for the longer-run variance of most nominal and real exchange rates. Among the variables used to measure economic activity and productivity, there are a small number of cointegrating vectors. The real exchange rates with Canada and Japan appear to be cointegrated with labor productivity; those with Germany and the United Kingdom are cointegrated with the proxy for capital productivity. These relationships are open to a number of interpretations, but one possibility is that they reflect the impact of productivity shocks on the real exchange rate. 52 In order to allow for the possibility that exchange rates could be cointegrated with a combination of the other economic variables, we also estimated a number of multivariate cointegrating vectors in which the nominal exchange rate was related to a complete set of monetary, fiscal, current account, and real productivity measures. For each bilateral nominal exchange rate at least two cointegrating vectors were found that were significant at the 1 percent level. One of these was picked and employed to measure the long-run level of the exchange rate as accounted for by 51 The failure to identify a role for monetary shocks over the recent floating rate period has led some researchers to conclude that the monetary approach to exchange rates has failed; see, for example, Boughton (1985). 52 1t is not clear, however, why capital productivity shocks matter for Europe, while labor productivity shocks matter for Canada and Japan. The result may reflect the sizable shifts in income distribution in Europe in the 1970s and 1980s, but it is unclear what implications these shifts would have for exchange rates.

488 • Structural Models of the Dollar

the forcing variables. 53 Deviations of the actual exchange rate from the long-run cointegrating vector can then be interpreted as measuring the equilibrium error in the system. Figure 3 plots the actual exchange rates and their long-run values as implied by the cointegrating vectors. As indicated in the figures, there are periodic substantial and long-lived deviations from the long-run values, which tend at times to display greater variability than exchange rates. There are two possible interpretations of the equilibrium errors shown in Figure 3. One is that they reflect bubbles in exchange markets, suggesting that exchange rates have at times deviated substantially from long-run economic fundamentals. An alternative explanation is that they reflect the systematic short-run movements in exchange rates identified in exchange rate models. In order to test this latter hypothesis, errorcorrection equations for exchange rates were estimated, in which the change in each bilateral nominal exchange rate was related to its own equilibrium error and the changes in the forcing variables. Consistent with the methodology of cointegration and error correction, these equations also included nominal interest rate differentials as a potential variable explaining short-run exchange rate movement. 54 k

Dsr

k

= L a;· Ur-; + L b; · R:~; i= I

k

+ i

L

=0

k"

C;·Rf'!; +

L

d;·DZr-i +

Er.

(9)

As indicated in equation (9), each error-correction equation relates the change in the exchange rate to an error-correction term that measures the deviation of the exchange rate from the long-run value implied by the forcing variables, U, short- and medium-run nominal interest differentials, R 5T and RLT, and lagged changes in the forcing variables, Z. The estimates of this equation for each bilateral nominal exchange rate are recorded in Table 10. Several features of the estimates are noteworthy. First, all coefficients on the equilibrium error are negative, so that the current change in the exchange rate is a negative function of the deviation of last period's exchange rate from its long-run equilibrium value. The estimated coefficient is, however, insignificant for the bilateral dollar rate with the pound sterling; for the deutschemark and the Cana53

There are no obvious criteria for choosing a cointegrating vector. We chose the one with the smallest variance for the equilibrium exchange rate error. 54 See Granger (1986). Given that nominal interest rate differentials are stationary and all variables integrated of order one are first differenced, the errorcorrection equation is not subject to the serious regression problem, and allows use of information on the levels of variables.

Charles Adams and Bankim Chadha • 489

Figure 3. (Log of) Actual and Long-Run Values of Nominal Exchange Rates Between the U.S. Dollar and Selected Currencies o.sr-----------------------------------------------~~

0.6

Nominal U.S. dollar-Japanese yen exchange rate

0.4

0.2 0~-----------.~T-~~~~~~-7~~~-.~----~

,

-0.2

-0.4 \ ... , Long-run ralue ~~___ 'J -0.6 ._____. •...___----J.__---1.____,____,___..._____.._____.___....____.____.___.._____, 1.5 1.0

•

Nominal U.S. dollar-deutschemark I\ exchange rate I \ I

0.5

Long-run ralut!

\ '\ \

"'

\

-0.5 -1.0.____,~__..

___...._____,____.___~--~--...____.____,~__..____..____,___~

0.1

-0.1

15 · 1.0

Nominal U.S. dollar-pound sterling exchange rate "

.. , ,

1\

I

0.~~---==--~~~~~,~~~~~='~-~~~~=~==~~~~~'~¥-,~'~i\~=,~-~,~~~:i\~~:~~ I

-0.5 -1.0

I\ I

I '

''

"T.,.,- '~

..,.

Actual

',I .,

'1

f LlmJ..

Krister Andersson • 545

The Tax Treatment of Domestic Investments in Selected Countries

The various methods of integration used in other countries increase the net of tax dividend income a shareholder receives. However, not only is the degree of integration important but also the overall level of corporate tax. The lower the corporate tax rate is, the less need there is to have any kind of integration. By looking at a gross income amount and comparing the income net of all taxes in different countries, we can calculate a measure of the overall level of taxation. 24 From Table 2, we can see that from a gross income of 100 units distributed as dividends, a U.S. investor gets to keep 39 units after corporate and individual taxes, whereas a U.K. investor would get to keep more than 61 units. Of the countries included in Table 2, only the Japanese investor retains less than a U.S. investor. The top marginal tax rate in Japan (applicable to incomes above approximately $135,000) is the principal reason for the low net of tax return for a Japanese investor. Both the tax treatment of dividend income and the tax treatment of capital gains are important when comparing the overall level of taxes in different countries. It is likely, for instance, that if a country imposes a relatively high tax burden on dividends, a large part of the return to equity will be in the form of retained earnings and capital gains. Although many factors influence dividend decisions, 25 one would expect a country such as Japan, with its relatively high tax rates on dividends and very low capital gains tax rate, to have a low dividend-earnings ratio, thereby allowing the Japanese investor to receive a large part of the return as capital gains. This is clear from Figure 6, where both tax rates for dividend income and capital gains together with dividend-earnings ratios are shown. The U.S. investor receives a much larger part of his return on equity as dividends than for instance a German investor. Germany does not impose any capital gains tax. 26 It is somewhat surprising to find that the United Kingdom, which has the highest capital gains tax rate, also has the highest dividend-earnings ratio. 27 24 It should be kept in mind that all international comparisons are subject to a large degree of arbitrariness and, in principle, entire tax systems rather than specific tax parameters should be analyzed at the same time. 25 The person with the highest marginal tax rate is not necessarily the "marginal investor" and even if he is the marginal investor, other more indirect ways of channeling investment funds could be used (probably with considerably lower tax liabilities). Any conclusions are, therefore, at best tentative and can only be interpreted as possible effects on incentives rather than on actual investment behavior. 26 Special rules apply to the taxation of capital gains from the sale of a significant holding (25 percent or more) in a business. 27 It must be kept in mind, however, that only gains accrued since 1982 are taxed

546 • Integrating Corporate and Individual Income Taxes

Given the tax structure of the United States with a rather high statutory capital gains tax rate and the lack of integration of corporate and individual taxes, a U.S. investor may very well face a higher tax burden than his counterparts in other countries. A higher tax burden may discourage U.S. investors from domestic investments, but the impact on saving in the United States will also depend on the tax treatment a U.S. investor receives if he invests abroad. The tax treatment of foreign investors in the United States, as well as the tax treatment of U.S. investors abroad, is central when evaluating whether or not U.S. investors are at a disadvantage vis-a-vis foreign investors. The Tax Treatment of Cross-Border Investments The Tax Treatment of Foreigners' Capital Income in the United States

The U.S. economy has become significantly more open to international financial capital in the last two decades, and the debate on the tax treatment of foreign-owned capital has intensified. The amount of foreign-owned assets in the United States grew more than 700 percent between 1975 and 1988 and more than threefold since 1980. 28 Since 1985, the amount of foreign-owned assets in the United States has exceeded the amount of U.S. assets abroad, although the comparison is of course heavily affected by the use of book value rather than market value accounting. 29 The United States exerts jurisdiction to tax all income, whether derived in the United States or not, of U.S. citizens, residents, and corporations. By contrast, the United States taxes nonresident aliens and foreign corporations only on income with sufficient nexus to the United States. Under the Internal Revenue Code, certain gross income of a foreign person is subject to a 30 percent U.S. withholding tax. Most U.S. income tax treaties with other countries reduce or eliminate the withholding tax, and business profits of an enterprise carried on by a resident of the treaty partner are not taxable by the United States unless the enterprise carries on a business through a permanent establishment in the United States. and gains since 1982 are indexed for inflation. Furthermore, the first £5,000 in carjtal gains for each individual are exempt. See Department of Commerce (1989). 29 In 1988, private, nondirect investments represented two thirds of foreign assets in the United States.

Krister Andersson •

547

Figure 6. Dividend-Earnings Ratio and Tax Rates In Selected Countries, 1990

Dividendearnings ratio

Top marginal tax rate (lnperce111)

Capital gains tax rate (In percent)

Source: Morgan Stanley ( 1990).

Foreign investors have been subject to this 30 percent withholding tax on dividends, rents, and royalties for a long time. However, in July 1984, a major component of the tax was removed with the elimination of withholding taxes on foreigners' interest income. The tax code now exempts from the 30 percent tax certain interest paid on portfolio obligations.30 The United States generally does not tax capital gains of a nonresident alien individual that are not related to U.S. real estate or U.S. trade or business. The U.S. Tax Treatment of U.S. Foreign Investments

The Administration discussed its rationale for U.S. tax policy toward international income in its 1985 tax reform proposal. The proposal states, "the general rule is that U.S. taxpayers are subject to U.S. tax on their worldwide income. A credit is allowed against U.S. tax for foreign income taxes paid in order to avoid double taxation .... The special 30

See sections 872(h) and 881(c) of the Internal Revenue Code.

548 • Integrating Corporate and Individual Income Taxes

measures include the deferral of U.S. tax on income earned by U.S.controlled foreign corporations until that income is remitted to U.S. shareholders. " 31 The most important goal for U.S. taxation of international income is to prevent distortion of the locational decisions of U.S. firms. This view is often referred to as capital export neutrality (CEN). The U.S. tax code does not however fully reflect CEN. The two major exceptions are the deferral of tax generally provided to active business income earned abroad and the limitation that no credit is provided for foreign taxes that, on average, exceed the U.S. tax rate. 32 Many countries take another view, the so-called territorial rule, and exempt profits of foreign subsidiaries and branch operations from domestic tax. 33 The Role of Tax Treaties

An important factor affecting U.S. tax policy regarding investment by foreign persons is the shift of the United States from net international creditor to debtor. Capital importing nations and capital exporting nations often have conflicting objectives concerning the country distribution of tax revenues. Capital importing countries tend to prefer that the tax be collected at the source, while capital exporting countries typically prefer the tax to be collected by the country of residence. 34 Through treaties with a number of countries, the United States has tried to avoid double taxation of income. The preferred U.S. tax treaty position has been expressed from time to time in model treaties and agreements. Nondiscrimination has been an important goal of U.S. tax treaty policy. 35 While withholding taxes on portfolio investments usually are creditable against income tax imposed by the country of residence, 31

United States (1985), p. 383. See Internal Revenue Code sections 903 and 904. The Administration's budget shows a revenue loss of $800 million from this feature. See Office of Management and Budget (1990), p. A-71. 33 Countries that fully or partially take this approach include Canada, France, Germany, and the Netherlands. See Frisch (1990). 34 According to the Joint Committee on Taxation (1990, p. 64), "United States tax policy, as expressed in tax treaty policy, for example, reflects its history as a capital exporter. Many argue that U.S. tax policy should respond to its new status by placing more emphasis on taxation at source." The report also mentions the problems connected with such a shift with increasing internationalization of business and financial markets and the possibility that the United States may once again become a net international creditor. 35 "Nationals of a Contracting State shall not be subjected in the other Contracting State to any taxation or any requirement connected therewith which is other or more burdensome than the taxation and connected requirements to which nationals of that other State in the same circumstances are or may be subjected." See the 1981 U.S. Model Income Tax Treaty, Article 24. 32

Krister Andersson • 549

no credit is generally available to portfolio investors with respect to denial of integration (imputation) benefits. 36 For direct investments, the integration benefits have generally not been extended to foreign direct investorsY The United States has consistently insisted on the extension of the integration benefit to the U.S. portfolio shareholders and the majority of the countries have granted these benefits. However, countries with integration generally do not extend integration benefits by treaty to U.S. corporate direct investors. 38 In their report on the taxation of foreign investment in the United States, 39 the Joint Committee mentions that a source country may have incentives to impose a tax if the foreign investment in that country is relatively insensitive to the tax levied. The larger industrial countries are those most likely to have sufficient market power to obtain an advantage from this kind of policy. A completely different strategy, typically used by small countries with little market power, is to provide tax exemption or tax holidays for foreign capital investment. The report rightly points out that a "beggar thy neighbor" policy would leave all nations worse off if generally adopted. 40 The introduction of some kind of integration (imputation) system in the United States could potentially imply a conflict with the previous nondiscriminatory line. 41 The U.S. Investor in an International Perspective

The combined effect of worldwide taxation of U.S. residents investing abroad while foreign investors are exempt from taxation on portfolio interest income and capital gains could put a U.S. investor at a disadvan36 In a treaty with Finland, ratified on June 14, 1990, Finland refused to grant any type of imputation credit or refund for U.S. shareholders of Finnish corporations. The Finns were willing to give credit only in situations in which their citizens receive reciprocal benefits and given the U.S. classical corporate tax system, no benefit for U.S. investors was given. The Treasury was however able to get a partial concession in the proposed U.S.-German income tax treaty. See Turro (1990). 37 One exception to this rule is the United Kingdom, which allows a refund to foreign direct investors equal to one half of the imputation benefit that would be available to a U.K. portfolio shareholder. 38 The Business and Industry Advisory Committee to the OECD (1990) urges the OECD to concentrate its efforts in the taxation area more toward the goal of eliminating international double taxation. One of their main concerns "arises from the fact that existing imputation systems, generally speaking, either do not deal at all with the problem of economic double taxation at the international level or may indeed add to the problem." Seep. 196. 39 See Joint Committee on Taxation (1990). 40 The report continues by indicating the potential desirability for some type of international coordination in tax policy. See Joint Committee on Taxation (1990), p. 62. 41 For a discussion, see Wrappe (1990).

550 • Integrating Corporate and Individual Income Taxes

tage. During the 1980s, the capital income taxes in the United States and Japan encouraged capital flows to the United States by favoring investment in the United States and by harming the country's relative saving performance. As regards the taxation of saving, for assets located in both countries, aU. S. saver faced a heavier tax burden than a Japanese saver. 42 While data are not available for investment combinations other than the U.S.-Japan case, it seems that the U.S. investor may face a heavier tax burden than many of his international competitors. There are several different ways to mitigate double taxation of equity capital in general and dividend income in particular. A lower capital gains tax rate would increase the net rate of return on the part that is retained in the corporation, while some kind of dividend relief would decrease the overall tax burden on dividends. The corporate tax rate has a major role to play in the taxation of capital income and the net subsidy it provides to debt financing. The decision whether to lower the capital gains tax and/or to mitigate the double taxation of dividends or even disallow deductibility of interest payments for the corporate tax should be based on efficiency, administrative, and equity considerations. The following section indicates some possible ways to integrate the individual and corporate tax systems. V. Integration of Corporate and Individual Taxes Some Possible Ways of Integrating Corporate and Individual Taxes

Given the existence of the corporate tax, the tax differential against the distribution of profits could be removed or partially mitigated in several ways. There are four possible ways of achieving a more equal tax burden on distributed and undistributed profits. One approach would be a splitrate system, under which there is one rate of corporation tax on undistributed profits and another and lower rate on distributed profits. Ger-

42 In 1987, for a debt-financed investment in the United States, a Japanese saver faced a net subsidy of 1. 7 percentage points in real terms, while aU .S. saver faced a subsidy of 0.6 percentage point. For an equity-financed investment in the United States, a Japanese saver faced a net real tax wedge of 2 percentage points, while a U.S. saver faced a tax wedge of 3 percentage points. For a debt-financed investment in Japan, a Japanese investor faced a net real subsidy of 0.3 percentage point, while a U.S. investor faced a net real tax of 0.8 percentage point. For an equity-financed investment the net tax was 6.5 percentage points for a Japanese investor and 7.8 percentage points for a U.S. investor. See Bovenberg and others (1990).

Krister Andersson • 551

many is among the countries using this method. 43 The split-rate system is similar in effect to a deduction for dividends paid (see below). A second method is a credit system under which all profits are first taxed at a single corporate tax rate and dividends are subsequently endowed with an imputed tax credit that can be set against the liability for direct personal tax. Italy and Canada are among the countries that apply this system. Between 1954 and 1963, U.S. taxpayers were allowed a credit of 4 percent of received dividends after a $50 exclusion ($100 for joint returns). 44 A similar system would be to allow corporations to deduct from their taxable income all or a portion of the dividends they pay out. This system was used in the United States in 1936 and 1937. One of the merits of this system is that it treats dividends like interest payments and all shareholders are treated equally. However, the method has been criticized for encouraging pay-out of earnings and for discouraging internal financing. A third method is a dividend exclusion system that permits individual income taxpayers to exclude all or a portion of their dividends from taxable income. In the United States, between 1964 and 1986 there was an exclusion of $100 ($200 for joint returns) for each taxpayer. 45 The fourth method is an "avoir fiscal" system, under which the recipient of cash dividends is once again endowed with an imputed credit (the avoir fiscal) but in which this tax credit is reckoned as a certain fraction of the corporation tax that has been levied on the profits used to pay the cash dividends. 46 The avoir fiscal system is used in France and Finland among other countries. For a summary of the different methods, see the annex. A very comprehensive study of tax integration was presented in Canada in 1966, by the Carter Commission. The Carter proposal includes a technique allowing for "voluntary allocation" by any corporation of earnings to shareholders even if earnings are retained by the corporation and such allocations (though retained) plus earnings actually distributed 43 In 1990, the tax rate on distributed profits was 36 percent and the tax rate on undistributed profits was 50 percent. 44 The form of credit granted the same relief on a dollar of dividends at all income levels but the value of the credit differed depending on the marginal individual income tax rate. For those subject to a zero rate, the credit was worthless, while the percentage of additional tax burden at the individual level removed by the dividend credit increased by income. This regressive pattern of relief led to its repeal. See Pechman (1987). 45 The exclusion was introduced as a compromise when the 4 percent dividend credit was repealed. 46 In the imputation system, the tax credit is reckoned as a given imputed rate of tax on the dividends.

552 • Integrating Corporate and Individual Income Taxes

would be "grossed up" and would carry a shareholder imputation credit for the corporate tax paid. Eugene Steuerle, former Deputy Assistant Secretary of the U.S. Treasury, has an alternative to other forms of integration (called simplified integrated tax (SIT)) that would assess tax once on income earned within a corporation. 47 The tax rate would equal the top rate applying to individuals or corporations, and low-income and nontaxable taxpayers would not benefit from lower rates. Taxes would be withheld at the corporate level and for most taxpayers withheld taxes would equal final tax liability. A number of other proposals have surfaced over the years but rather than going into the details of these, it may be more useful to assess the impact that different methods of integration would have on the cost of capital and revenue. Section VI develops a cost of capital methodology and reports tax wedge calculations for different integration schemes. Some Problems Connected with Integration

Integration of corporate and individual taxes raises a number of difficult issues. First, it would be inappropriate to assume that all corporations are subject to a 34 percent tax when calculating the credit for the individual taxpayer. All corporations are taxed at a reduced rate on the first $75,000 of their earnings. Furthermore, the effective tax rate on corporations is less than 34 percent since they are allowed accelerated depreciation for tax purposes and tax on foreign source income may be deferred. Second, as pointed out earlier, the tax treatment of foreign shareholders and tax-exempt institutions could be designed in several ways and is by no means noncontroversial. Third, withholding of taxes at the corporate level and credit at the individual level would entail increased administrative complexity. If not only dividends but also undistributed earnings are taxed at the individual level, the basis of the stock must be written up by the amount of taxed earnings, to prevent double taxation of any subsequent capital gains resulting from the retentions. This would make the tax system more complicated. 48 If the partnership approach is taken, the data processing requirements could be very substantial, since it would be necessary to report each person who held stock in a given corporation during any part of the year. Other areas of concern are intercorporate dividends and the overall progressivity of the tax system. One has to conclude, however, 47

48

See Steuerle (1989b), pp. 335-36. For a discussion, see McLure (1979), p. 148.

Krister Andersson • 553

that the problems of integrating the corporate and individual taxes depend to a large extent on the chosen method. Dividend relief at the corporate level for dividend payments when calculating the corporate tax would be relatively easy to administer. It would be possible only to give relief for dividend payments on new investments thereby avgiding a windfall gain to existing shareholders. VI. Cost of Capital Calculations of Some Integration Methods The User Cost of Capital

The present study uses the methodology of effective tax rates derived directly from the parameters of the tax system. 49 The essential concept used in the estimation of the tax rate on capital income is the tax wedge. The tax wedge provides a measure of the effective tax burden on new investments and can be explained by defining three rates of return: the required before-tax rate of return on investment, p; the market return (after corporate taxes), r; and the after-tax rate of return to the saver, s. All these returns are measured in real terms. In the case of debt finance, the market return corresponds to the real interest rate, and for equity financing, it amounts to the real return on equity (taking into account dividends and capital gains) before personal taxes. The total tax wedge, w, can therefore be thought of as consisting of two parts: a corporate tax wedge, we, which equals (p - r) and a tax wedge at the investor level, wi, equal to (r - s). When cross-border investments are considered, a more useful separation of the total tax wedge is a host country tax wedge and a home country tax wedge. The host country levies corporate taxes but often also withholding taxes on dividend and interest payments. The home country, in turn, taxes these returns, possibly subject to some form of double taxation relief. The corporate tax wedge is derived from the neoclassical theory of investment behavior, where firms carry out investments until the beforetax rate of return, p, equals the required real rate of return. 50 Solving for the before-tax rate of return, one obtains: p =

1 [(1- k- tc · z)(T + 8- 1r)]- 8, (1 - tc)

(1)

Earlier work in this field includes Fromm (1971 ), King (1977), and King and Fullerton (1984). 50 The expression for p is derived from the equality between the after-tax marginal benefit and the marginal cost of an mvestment project: (1 - tc) (p + 8) = (1- k- tc ·z) (T- 1T + 8). 49

554 • Integrating Corporate and Individual Income Taxes

where tc = statutory corporate tax rate

investment grant present value of depreciation allowances T = nominal discount rate 8 = economic rate of depreciation 1r = rate of inflation p = required before-tax real rate of return. k

z

= =

In the absence of taxes and grants, the nominal discount rate, T, is equal to the market rate of return on the financial asset. In the presence of corporate taxes, the difference between p and the real market rate of return on financial assets, the corporate tax wedge provides a measure of the burden (or subsidy} of the tax system on investment. The company's discount rate depends on the source of financing. If an investment is debt financed, debt-servicing costs are usually deductible when calculating the corporate tax liability, thereby reducing the company's financing costs and its discount rate. However, under the classical corporate tax system, no relief is given for investments financed by equity capital. 51 In general, the corporate tax system tends to favor debt financing while capital gains taxation at the investor level often leads to a favorable tax treatment of the part of an equity-financed investment that is retained in the company. The framework used here allows us to incorporate these effects (including the difference in discount rates for different types of financing) and compare tax wedges across modes of financing and for different integration schemes. Even at the investors' level, the taxation of dividend and interest income can differ. Hence, we need to define two rates of return: one for an equity-financed investment and one for a debt-financed investment. The real after-tax rate of return on a debt-financed investment, sd, can be expressed as sd = (1 - m) (r

+ 1r)

-

1r,

(2}

where ·m

=

marginal personal tax rate on capital income

= parameter representing relief for corporate taxes

r

1r

=

real interest rate

= inflation rate.

51 The discount rate will therefore be higher in this case, and the present value of depreciation allowances, z, will therefore be lower and the user cost of capital correspondingly higher.

Krister Andersson • 555

Table 3.

Tax Parameters and Some Economic Variables 1 (In percent)

1984

1987

Corporate tax rate 38.3" 49.5 Individual tax rate Dividend income 39.6 32.0 Interest income 25.8 22.4 Capital gains (accrual) 5.9 11.0 Assets life in years (machinery) 4.6 2 6.03 Dividend-payout ratio 56.7 78.2 Inflation rate 6.2 5.3 Interest rate (endogenous) 12.5 9.4 2.0 Net of tax real rate of return (exogenous) 2.0 1 The tax rates include an estimate of state and local taxes. 2 Assuming that 150 percent declining balance has been used for tax depreciation purposes and as first year convention, half a year's deduction. 3 Assuming double declining balance has been used for tax depreciation purposes and as first year convention, half a year's deduction.

For an equity-financed investment, the real after-tax rate of return, s., is equal to s,

= {a(l- m) + (1- a)(l- c)}I-L-

1r,

(3)

where a

= fraction of real earnings on equity paid as dividends = nominal return on equity before personal taxes c = tax rate on nominal capital gains.

1-L

By imposing an arbitrage condition at the investors' level, it is possible to calculate the tax wedges when the investor earns the same after-tax rate of return on a debt-financed investment and on an equity-financed investment.52 The present study imposes this arbitrage assumption, assuming that an investor requires a 2 percent real net of tax rate of return on both an equity-financed investment and a debt-financed investment. 53 Tax parameters broadly in line with the economic situation in 1987 have been used (Table 3). 52 Some studies include an exogenous risk premium on equity (see, for example, Feldstein (1986)). An alternative approach is to directfy estimate observed price-earnings ratios on shares (see, for example, Boadway and others (1987)). If the arbitrage condition is imposed at the corporate level, resulting in the same net cost for the firm regardless of the source of finance, the investor will typically receive a lower rate of return for an equity-financed investment than for a debt-financed investment. The user cost of capital is therefore not unaffected by the applied arbitrage assumption. 53 From the above formulation for the user cost of capital, it is obvious that the concept of effective tax rate is limited in several respects: it only considers explicit

556 • Integrating Corporate and Individual Income Taxes

Cost of Capital in the United States and the Tax Reform Act of 1986

By using the methodology described above, it is possible to make an overall assessment whether the Tax Reform Act of 1986 on average increased or decreased the incentives for a specific kind of financing. The U.S. corporate tax system favors debt financing over equity financing since interest payments are deductible for the corporate tax whereas dividend payments are not. In 1984, the total tax wedge for a debtfinanced investment was highly negative, that is, the tax system provided a subsidy to debt financing of more than 3 percentage points in real terms. An equity-financed investment faced a positive tax burden and the total tax wedge was almost 2.5 percentage points. The Tax Reform Act of 1986 increased the tax burden on both an equity-financed investment and a Figure 7. Total Tax Wedges in the United States (Percentage points in real terms)

I

Assuming unchanged economic environment from 1984.

taxes on capital income; it ignores quantitative restrictions and nontax policies; it is based on assumptions that tend to make the calculations of cost of capital static; it often does not take into account expected future changes in interest rates and tax rates; and, finally, it often abstracts from risks. In most countries, the effective tax rate depends on the type of investment or investor. Some agents are even tax exempt or are able to influence the effective tax rate that they will face by tax planning. The user cost of capital therefore only gives us a broad picture and all results should be interpreted with some caution.

Krister Andersson • 557

debt-financed investment. The subsidy to a debt-financed investment was reduced to less than 1 percentage point, while the total tax wedge on an equity-financed investment increased by more than 1 percentage point to 4.5 percentage points (Figure 7). The tax rates used in the calculations are average marginal tax rates and the effect of the tax reform may very well have been different for different investors. The general picture is, however, that the increase in the capital gains tax rate largely offset the lower tax rate on dividends and the lower corporate tax rate reduced the subsidy to debt financing. The tax reform does not seem to have reduced the incentives for debt finance and, if anything, it increased the relative incentive for debt financing. The inherent bias against equity financing therefore remains, but there are several possible options to mitigate the effect. These options are discussed in the next section. The Effect on Cost of Capital of Different Integration Schemes54

A Lower Capital Gains Tax Rate

As mentioned earlier, a number of possibilities exist for lowering the effective tax on equity capital and/or decreasing the bias in favor of debt financing in the present tax system. One proposal that has been widely debated in the last couple of years is a lower capital gains tax rate. In the 1991 U.S. budget, the argument in favor of a lower capital gains tax rate is that "Lowering the tax rate on capital gains would lower the cost of capital in vital areas of investment activity." 55 However, a lower capital gains tax would only influence the part of an equity-financed investment that is retained in the corporate sector. 56 No doubt a lower capital gains tax rate would influence dividend policies in the corporate sector, and a larger share of return on equity capital than today might well be received in the form of capital gains. A reduction in the capital gains tax rate to its 1984 level would reduce the total tax wedge on an equity-financed investment from 3.3 percentage points to 2.8 percentage points (Table 4). If we assume that a larger share will be received in the form of capital gains if the capital gains tax rate is reduced, the result is a further decrease 54 The comparisons described in this section are not revenue neutral, in the sense that increases (decreases) in some taxes are not compensated by corresponding decreases (increases) elsewhere. 55 See Office of Management and Budget (1990), p. 47. 56 A lower capital gains tax rate would have a number of other consequences for tax administration, tax arbitrage, and efficiency aspects as well. For a discussion of these aspects, see Andersson (1989).

2.00

17.83

3.33 1.77 1.56

-Equity

1987

2.00

-3.26 2.00

-0.53 17.30

-

14.04

2.80 1.66 1.14

Equity

-0.46 -2.57 2.11

Debt

No Integration but Reduced Capital Gains Tax; 1987 Payout Ratio

2.00

-2.98

-

14.04

-0.46 -2.57 2.11

Debt

2.00

-0.80 17.02

2.53 1.61 0.92

Equity

No Integration but Reduced Capital Gains Tax; 1984 Payout Ratio

2.00

-2.40

14.04

-0.46 -2.57 2.11

Debt

2.00

-1.39 16.44

1.94 1.49 0.45

Equity

Full Integration

2.00

-2.05

-

14.04

-0.46 -2.57 2.11

Debt

2.00

-1.74 16.09

1.59 1.42 0.17

Equity

Full Integration and Reduced Capital Gains Tax to 1984 Level

2.00

-1.66

-

14.04

-0.46 -2.57 2.11

Debt

2.00

-2.14 15.70

1.19 1.34 -0.15

Equity

Full Integration and No Capital Gains Tax

United States: The Effects of Different Tax Changes on the User Cost of Capital

Total tax wedge -0.46 Corporate tax wedge -2.57 Personal tax wedge 2.11 Change in total tax wedge from 1987 User cost of capital 14.04 Difference in user cost of capital between debt and equity -3.79 Return after all taxes (in real terms) 2.00

Debt

Table 4.

2.00

-1.07

1.33 15.37

0.87 1.28 -0.41

Debt

2.00

-1.39 16.44

1.94 1.49 0.45

Equity

No Deductibility of Interest Payments but Full Integration of Interest and Dividends

(11 (11

~ (/)