ACCF HOME PAGE - AMERICAN COUNCIL FOR CAPITAL FORMATION
CONTACT US | SITE MAP
ABOUT ACCF | ACCF CENTER FOR POLICY RESEARCH | NEWS | NEWSLETTER | PROGRAMS | PUBLICATIONS

 

Click here to receive our newsletter via email.

Forward this page to a friend. Click here.

 

 

The Impact of Capital Gains Taxation
on U.S. Investment and Economic Growth

American Council for Capital Formation
February 15, 1995

Introduction

ACCF President Mark Bloomfield, accompanied by ACCF Chief Economist Dr. Margo Thorning, testified on February 15, 1995 as an invited witness at a Senate Finance Committee hearing that examined the impact of capital gains taxation on U.S. investment and economic growth. The executive summary and full text of the ACCF's testimony are presented here.

Executive Summary

1. Overview.
We commend the emphasis that Chairman Packwood places on the impact of capital gains taxation on the cost of capital, saving and investment, and economic growth. A capital gains tax cut will, if enacted, help reduce the burdens on capital formation imposed by current U.S. tax policy. It is clear that U.S. tax policies toward saving and investment must be revised if we are to increase real wages for U.S. workers and retain our leading role in world affairs.

2. Trends in U.S. Capital Formation. Recent U.S. saving rates and investment spending compare unfavorably with those of other nations as well as with our own past experience. The U.S. saving rate averaged 4.8 percent over the 1973-1991 period, compared to 19.1 percent in Japan and 10.7 percent in West Germany. During the same period, gross residential investment as a percent of GDP was lower for the United States than for any of our major competitors.

3. Tax Policy and Economic Growth. To those who favor a truly level playing field over time for individual and business decisions to save and invest, stimulate economic growth, and create new and better jobs, capital gains (and other forms of saving) should not be taxed at all. This view was held by top economists in the past and is held by many mainstream economists today. A number of studies show that U.S. economic growth would be enhanced if we relied more on consumption taxes or replaced the income tax with a fundamental tax restructuring plan such as those proposed by several prominent members of Congress.

4. Economic Case for Low Capital Gains Taxes. The economic case for low capital gains taxes rests on the beneficial impact of such a change on capital costs, capital mobility, entrepreneurship, and the ravages of inflation. One recent study shows that a substantial reduction in capital gains taxes, when macroeconomic "feedback" effects as well as "unlocking" of unrealized gains are included, would result in new and better jobs and lower capital costs, as well as increased capital formation, stronger economic growth, and higher federal tax revenues than under current law.

5. Taxation of Capital in Other Developed Countries. The United States taxes individual and corporate capital gains more harshly than most other industrialized countries. A survey of twelve industrialized countries showed that the U.S. capital gains tax rate on long-term gains on portfolio securities exceeded that of all countries except Australia and the United Kingdom and even these countries index the cost basis of an asset.

6. Capital Gains Rate Reductions and Tax Revenues. Capital gain revenue estimates involve three elements: the "static" revenue loss, the "unlocking" effect, and the "macroeconomic" impact. A National Bureau of Economic Research study in the late 1980s found the capital gains revenue maximizing rate to be in the range of 9 to 21 percent-the point at which there is sufficient "unlocking" because of a lower tax to compensate for the static revenue loss because of a lower tax rate. This maximizing rate does not account for the additional revenue stemming from the positive macroeconomic impact of increased investment, GDP, and employment that would result from a significant reduction in the capital gains tax. A new study by nationally renown economist Dr. Allen Sinai demonstrates this significant macroeconomic effect and the resulting additional federal revenues.

Finally, the historical experience of actual capital gains tax receipts during periods of low taxes (1978-1985) and high taxes (1987 to the present) strongly suggest that a reduction in the current capital gains tax would have a positive impact on federal revenues.

7. Conclusion. The hard fact is that we can no longer afford the luxury of government economic policies that reward consumption, discourage saving and investment, overregulate American business, and penalize economic growth. Enactment of capital gains tax reform would help move the United States toward a tax system that is more neutral toward saving and investment and pave the way for more fundamental tax restructuring.

ACCF Statement

Introduction


My name is Mark A. Bloomfield. I am president of the American Council for Capital Formation (ACCF). I am accompanied by Dr. Margo Thorning, our chief economist. The ACCF represents a broad cross section of the American business community, including the manufacturing and investment sectors, Fortune 500 companies and smaller firms, individuals, and associations. Our board of directors includes cabinet members of prior Republican and Democratic administrations, former members of Congress, prominent business leaders, and public finance experts. We appreciate this opportunity to present testimony on the impact of a capital gains tax reduction on U.S. investment and economic growth. In addition, I will share the results of an international comparison of capital gains tax rates and discuss the revenue maximizing capital gains tax rate.

To encourage a constructive debate on the taxation of capital gains this year, our affiliated public policy think tank, the ACCF Center for Policy Research, has prepared a special report for today's hearing, Questions and Answers on Capital Gains, which is attached to our testimony.

We commend the emphasis that Chairman Packwood places on the impact of capital gains taxation on the cost of capital, saving and investment, and economic growth. A capital gains tax cut will, if enacted, help reduce the burdens on capital formation imposed by current U.S. tax policy. It is clear that U.S. tax policies toward saving and investment must be revised if we are to increase real wages for U.S. workers and retain our leading role in world affairs.

Trends in U.S. Capital Formation, Productivity Increases, and Economic Growth

Investment spending in the United States in recent years compares unfavorably with that of other nations as well as our own past experience. From 1973 to 1991, gross nonresidential investment as a percent of Gross Domestic Product (GDP) was lower for the United States than for any of our major competitors (see Table 1). The U.S. saving rate averaged 4.8 percent over the 1973-1991 period, compared to 19.1 percent in Japan and 10.7 percent in West Germany. Even more disturbing is the fact that net annual business investment in this country has in recent years fallen to only half the level of the 1960s and 1970s. Net private domestic investment averaged 7.4 percent of GDP from 1960 to 1980; since 1991 it has averaged only 3.0 percent (see Table 2).

Reflecting the reduced share of GDP being invested each year, the U.S. capital stock has also grown more slowly. In the three decades prior to 1980, the total capital stock grew at 4.0 percent per year; in the 1980s and 1990s, the rate fell to 2.7 and 1.4 percent respectively (see Table 3). The stock of equipment, which many experts regard as critical for strong productivity growth, has increased only about half as fast since 1980 as in previous decades. Industrial equipment stocks, which grew at an average rate of 4.3 percent over the 1950-1979 period, increased by just 1.2 percent annually in the 1980s and 0.1 percent since 1990.

Link Between Investment, Productivity Increases, and Economic Growth

The importance of investment in plant and equipment for economic growth is emphasized in a new book by Harvard Professor Dale Jorgenson. Professor Jorgenson's book, Productivity: Postwar U.S. Economic Growth, analyzed economic growth between cyclical peaks in the business cycle over the 1948-1979 period. Allocating increases in output to three sources-growth in the capital stock labor, supply, and multifactor productivity-Professor Jorgenson found that increases in the capital stock contribute most to increases in output (see Figure 1).1

Investment's key role in advancing technological progress and productivity growth is also stressed in recent research by New York University Professor Edward N. Wolff.2 He argued that U.S. labor productivity growth rates are depressed by the recent slower growth in the capital-to-labor ratio-from a peak of 2.0 percent per year in the 1950s to 1.2 percent per year in the 1972-1992 period. He emphasized that the effects of the decline in U.S. capital-labor growth are perhaps even more pernicious than they appear at first glance.

Thus, a slowdown in capital formation may doubly hurt labor productivity growth-directly by slowing down the rate of capital deepening, and indirectly by slowing down the rate of technological advance. Professor Wolff's research also shows that U.S. labor productivity growth lags behind our competitors; OECD countries outstripped the United States during much of the 1950-1990 period. He noted that countries such as Japan and Germany, which experienced strong productivity growth in the 1970s and 1980s, showed significant gains in their capital-to-labor ratios. Our competitors' gains in capital-to-labor ratios are a direct result of their higher levels of investment.

Implications of Lagging Investment and Slow Growth in Labor Productivity for Current and Future Living Standards

Real family income in the United States has been nearly stagnant since the mid-1970s and in recent years has actually fallen. For example, real median household income was $39,869 in 1989; income has declined in each subsequent year, and in 1993 stood at $36,959. These trends have not only made it harder to maintain living standards but have also jeopardized our future economic health and our ability to remain the principal leader in international affairs.

In addition, looming in the future is the need to finance the retirement of the baby boom generation. Research by Stanford Professor B. Douglas Bernheim, commissioned by the ACCF Center for Policy Research, our public policy think tank, shows that current saving by members of the baby boom generation is seriously inadequate.3 The typical baby boom household saves at only one-third the rate required to finance a retirement standard of living comparable to that enjoyed before retirement.

Tax Policy and Economic Growth

To those who favor a truly level playing field over time for individual and business decisions to save and invest, stimulate economic growth, and create new and better jobs, capital gains (and other forms of saving) should not be taxed at all. This view was held by top economists in the past and is held by many mainstream economists today.

This is primarily because the income tax hits saving more than once-first when income is earned and again when interest and dividends on the investment financed by saving are received, or when capital gains from the investment are realized. The playing field is tilted because the individual or company that saves and invests pays more taxes over time than if all income were consumed and no saving took place. Taxes on income that is saved raise the capital cost of new productive investment for both individuals and corporations, thus dampening such investment. As a result, future growth in output and living standards is impaired.

A consumption-based tax system, under which all saving and investment would be exempt from tax, would be more favorable toward capital formation and economic growth than is our current income tax system, according to analyses by top public finance scholars over the past decade and a half. Studies by Stanford University's John Shoven and Lawrence Goulder; Harvard University's Dale Jorgenson; the University of Texas' Don Fullerton; and Joel Prakken of Laurence H. Meyer have used macroeconomic models that incorporate "feedback" and dynamic effects in simulating the effect of adopting a consumption tax as a full or partial replacement for the income tax. These studies, which use different types of general equilibrium models, all come to the conclusion that U.S. economic growth would be enhanced if we relied more on consumption taxes or replaced the income tax with a fundamental tax restructuring plan similar to those proposed by several prominent members of the U.S. Senate and House of Representatives.

The Economic Case for Low Capital Gains Taxes

Low capital gains taxes not only treat savers more fairly but also help hold down capital costs. Public finance economists refer to the tax on capital gains as a tax on retained income, which funds a large part of business investment. The higher the capital gains tax, the more difficult it is for management to retain earnings (rather than pay out dividends) for real investment in productive projects.

Although the economy is expanding, worries about the future appear to be multiplying. A cut in the capital gains tax to a top marginal rate of, say, 15 to 20 percent would by no means act as an economic panacea. However, it would surely give a boost to values of capital assets (e.g., real estate and the stock market), encourage investment by both mature and new businesses, and constitute fairer taxation of peoples' savings.

A new study by Allen Sinai, chief global economist at Lehman Brothers and highly respected economic forecaster, shows that a substantial reduction in capital gains taxes would, when macroeconomic "feedback" effects as well as unlocking of unrealized capital gains are included, result in new and better jobs, as well as increased capital formation, stronger economic growth, and federal tax revenues that are larger than under current law.

According to Dr. Sinai's study, a 50 percent capital gains exclusion, combined with prospective indexing for all taxpayers (individual and corporate) would, by the year 1999, increase real GNP by 2.3 percent, or about 0.5 percent per year compared to a baseline (see Table 4). In addition, the capital gains tax reduction would increase capital spending and capital formation, increase household net worth (household wealth), lower the cost of capital for business and increase business profits, increase employment and lower the unemployment rate, shift the financing of business activity away from debt to equity, and induce portfolio allocations by households toward equity to take account of changes in expected after-tax returns on stocks and bonds.

The economic case for a low capital gains rests on the beneficial impact on capital costs, capital mobility, the ravages of inflation, and entrepreneurship.

Capital Gains and Capital Costs

The user cost of capital is the pretax return on a new investment that is required to cover the purchase price of the asset, the market rate of interest, inflation, risk, economic depreciation, and taxes. This capital cost concept is often called the "hurdle rate" because it measures the return an investment must yield before a firm will be willing to start a new capital project.

Economists are in broad agreement that capital costs are affected by tax policy. For example, Stanford Dean John B. Shoven estimated that about one-third of the cost of capital is due to taxes. In other words, hurdle rates are 50 percent higher than they would otherwise be due to the tax liability on the income produced by the investment. Thus, the higher the tax on new investment, the less the investment that will take place. Although the Tax Reform Act of 1986 (TRA) substantially reduced corporate and individual income tax rates, the legislation's capital cost recovery provisions raised effective tax rates and capital costs for productive and pollution-control assets. Capital costs increased because of increases in capital gains taxes, the loss of the investment tax credit, lengthening of depreciable lives for many assets, and the creation of the corporate alternative minimum tax.

Low capital gains taxes help hold down capital costs. Research by public finance economists refer to the tax on capital gains as a tax on retained income, which funds a large part of business investment. The higher the capital gains tax, the more difficult it is for management to retain earnings (rather than pay out dividends) to fund real investment in productive projects. Research by Stanford Dean John Shoven, Ohio State Professor Patric Hendershott, and Dr. Allen Sinai, indicates that a cut in the capital gains tax rate to a range of 15 to 20 percent would reduce the cost of capital by 4 to 8 percent.

Capital Gains and Capital Mobility

High capital gains taxes reduce the mobility of capital-and thus economic efficiency-by keeping capital from flowing into its most productive uses. The "lock-in" effect was demonstrated in a recent Tax Notes article (December 26, 1994) by attorney Mark Greenstein.

Greenstein explains how high capital gains taxes "lock-in" investors through the example below:

Amy has a building that cost $50,000, is worth $1,000,000, and yields $70,000 in rent net of expenses. If Amy lives in a high-tax state, such as New York, more than one-third of the value of the building would disappear in taxes on disposition. To obtain the $70,000 she was receiving, Amy would have to obtain a yield of over 10 percent on the roughly $650,000 she would receive, net of taxes, if she sold the building. Because of this disincentive, it is likely that Amy will never sell her building. Instead, she will simply leave it to her heirs and, under current law, the gain will never be subject to an income tax.

Greenstein's example makes clear that if capital gains taxes were substantially reduced, Amy would be more likely to sell the building because any subsequent acquisition would not have to yield anything close to 10 percent to produce the $70,000 in income she earned on her original building. The example given above is also appropriate for investments in equities.

Capital Gains and Inflation

Opponents of capital gains tax reductions fail to recognize that capital gains investments are inherently high risk and that realized capital gains include purely inflationary gains that are not income. The willingness to invest is hindered by taxing capital gains, which are phantom earnings brought on by inflation. The combined effect of taxing inflationary gains and limiting the deductibility of capital losses leads to a severe over-taxation of many investments that will earn capital gains.

A study by National Bureau of Economic Research chairman Martin Feldstein and University of Michigan Professor Joel Slemrod documents the overtaxation of capital gains due to inflation. They found that in 1973 individuals paid capital gains taxes on more than $4.5 billion of nominal gains on corporate stock. Their finding provides evidence that capital gains taxation is distortionary and unfair. If the cost of these shares had been adjusted for inflation, the $4.5 billion nominal gain would be a real capital loss of nearly $1 billion. In other words, individuals paid a substantial capital gains tax even though, after inflation, they received less from their sale than they originally paid.

The distortion, it should be pointed out, was greatest for middle-class investors. That obviously makes little economic sense and is unfair.

Capital Gains and Entrepreneurship

Capital gains taxation has a particularly powerful impact on the entrepreneurial segment of the U.S. economy-a reality that econometric models do not capture-making possible new technological breakthroughs, new start-up companies, and new jobs.

A few words about the entire entrepreneurial process are pertinent. A number of factors are involved including entrepreneurs, informal investors, venture capital pools, and a healthy public market. All, I should stress, are sensitive to after-tax rates of return, which is why the level of capital gains taxation is important.

Foremost is the entrepreneur. By taxing his potential capital gains at a higher rate, either the pool of qualified entrepreneurs will decline or investors will have to accept a lower rate of return. In either case, the implications for the U.S. economy are clearly negative. To be successful, the entrepreneur, of course, needs capital. Fledgling start-ups depend heavily on equity finance from family, friends, and other informal sources. Professors John Freear and William Wetzel of the University of New Hampshire, in a survey of 284 new companies, found taxable individuals to be the major source of funds for those raising $500,000 or less at a time.4 The point to be stressed is that individuals providing start-up capital for these new companies pay capital gains taxes and are, therefore, sensitive to the capital gains tax rate.

The Taxation of Capital in Other Developed Countries

Capital Gains

The United States taxes capital gains much more harshly than does the rest of the world. In a survey of twelve industrialized countries undertaken by the American Council for Capital Formation, the U.S. capital gains tax rate on long-term gains on portfolio securities was found to exceed that of all countries except Australia and the United Kingdom, and even these two countries index the cost basis of an asset. Germany, Japan, and South Korea exempt or tax only lightly capital gains from portfolio stock (see Table 5). The U.S. corporate capital gains tax rate is at an historic high and it is not competitive with many other countries.

Do they know something we don't know? Perhaps, yes. They recognize the contribution a capital gains tax differential can make to lower capital costs, mitigate the distortions of inflation, increase capital mobility, nurture entrepreneurship, and stimulate new business creation.

While it is difficult to prove that low (or no) taxes on capital gains cause higher rates of saving and investment, the circumstantial evidence is compelling. Personal saving rates tend to be higher in countries with low or no tax on capital gains on portfolio securities (see Table 5), and investment is also higher (see Table 1). For example, Japan taxes long-term capital gains on securities lightly; Japan's personal saving rate as a percent of GDP averaged 11.9 percent over the 1973-1991 period and nonresidential fixed investment averaged 24.1 percent. The comparable saving rate for the United States is 5.9 percent; and U.S. investment averaged only 13.9 percent.

Other Capital Income and Investment

The U.S. tax on capital gains should also be evaluated in the broader context of the taxation of all capital income. On that score, the United States also fares poorly. Many industrialized countries (see Table 5) tax other capital income less harshly than does the United States. Most provide for the integration of corporate and individual taxes, and many tax interest income at lower rates than the United States.

In addition to having lower taxes on capital income than does the United States, many other industrialized countries tax new investment more favorably than we do. A new study by Harvard's Dale Jorgenson found that in 1990, the marginal effective corporate tax rate on investment in equipment was 18.5 percent in the United States compared 11.5 percent in Germany, 8.8 percent in Japan, and 8.0 percent in the United Kingdom.5 Thus, tax differentials may help explain why U.S. investment as a percent of GDP lags behind that of our competitors.

Consumption Taxes

Finally, an international comparison needs to address the more fundamental issue of the overall taxation of saving and investment. This is extremely important because the level of a country's saving and investment is a major factor in determining its economic growth. The evidence is clear that almost all of our international competitors rely to a much greater extent on consumption taxes to fund government expenditures than does the United States. On average, the OECD countries collect 30 percent of their tax revenues from consumption taxes such as the value-added tax, compared to only 15 percent from consumption taxes in the United States.

Capital Gains Rate Reductions and Tax Revenues

Capital gains revenue estimates involve three elements. First, there is "static" revenue loss stemming from taxing realizations at lower rates. Second, there is the "unlocking" effect peculiar to capital gains because it is a voluntary tax. Taxpayers tend to be locked in if the rate is too high and will unlock if the rate is lower, thereby generating tax revenues. Third, there is the macroeconomic effect of additional revenue generated by the impact of lower capital gains on capital costs, saving and investment, and economic growth. The challenge to this Committee is to evaluate all three dimensions and the net impact on total revenues to the U.S. Treasury.

Critics of a low capital gains tax argue that such cuts will result in significant federal revenue losses, and thus add to the budget deficit, absorb national saving, and raise interest rates and capital costs. Both economic analysis and experience effectively refute this view.

Is There a Revenue-Maximizing Rate?

In the late 1980s, experts at the prestigious National Bureau of Economic Research examined the question of the revenue maximizing capital gains tax rate or, at what point is there sufficient "unlocking" to compensate for the "static" revenue loss resulting from a reduction in rates. The study by former Harvard Professor Lawrence Lindsay (now a member of the Board of Governors of The Federal Reserve), which was based on academic models of the responsiveness of taxpayers to changes in the capital gains tax rates, found that the revenue maximizing rate ranged between 9 and 21 percent.6 The NBER study does not take into account the additional revenue stemming from the positive macro consequences of increased employment and growth, which result from a significant reduction in capital gains tax rates.

Subsequently, Professor Lindsay modeled the revenue impact of a 15 percent capital gains tax. He chose that rate because it fell in the middle of the revenue maximizing range of 9 to 21 percent. Professor Lindsay concluded that a 15 percent capital gains rate would have substantially increased capital gains revenues. Again, it needs to be emphasized that this analysis does not include the revenue impact of a stronger economy.

Historical Experience

Experience indicates that lower capital gains taxes have a positive impact on federal revenues. The most impressive evidence involves the period from 1978 to 1985. During those years the marginal federal tax rate on capital gains was cut from almost 50 percent to 20 percent-but total individual capital gains tax receipts increased from $9.1 billion to $26.5 billion. After surging to $326 billion in 1986 (the year before the rate increases in the Tax Reform Act of 1986 took effect), capital gains realizations have trended down and have remained at less than $130 billion per year in the 1990s. Taxes paid are averaging only $27 billion per year. Given the increases in the stock market, inflation, and growth in GDP since the late 1980s, realizations and taxes paid are almost certainly being depressed by the current high capital gains rates.

Unlocking and Macroeconomic Impact on Revenue

Scholars have researched two elements affecting capital gains tax revenues, the "unlocking" of unrealized gains and the macroeconomic impact of a low tax on capital gains. Estimates of unlocking are extremely sensitive to assumptions of the elasticity of taxpayer response. Very minor differences in assumptions can result in large differences in revenues. There is a wide range of credible assumptions about elasticity. The important point is that all the studies recognize a significant unlocking effect. For example, Princeton Professor David Bradford noted that the revenue estimates of President Bush's 30 percent capital gains exclusion resulted in a "static" loss over 1990-1995 of $100 billion, according to the Joint Committee on Taxation (JCT) and the Department of the Treasury. However, induced realizations-the "unlocking" effect-and depreciation recapture would have recouped almost 90 percent of the loss, according to the JCT, and 110 percent as estimated by the Treasury. This arithmetic accounts for only one behavioral response-the "unlocking" effect-and the Treasury recoups almost all of the revenue loss.

Government revenue estimates do not factor in the macroeconomic consequences of lower capital gains tax rates on U.S. capital costs, investment, economic growth, and overall tax revenues. However, Dr. Allen Sinai's new analysis (cited earlier) shows significant increases in GDP, employment, and investment as well as a positive impact on federal tax revenues as a result of substantial capital gains reductions.

Long-Run Strategies for Tax Reform

While capital gains taxes should be lowered or eliminated immediately to help encourage U.S. saving and investment, policymakers should have comprehensive tax reform as their long-term goal.

Restructuring the U.S. federal tax system to reduce the multiple taxation of saving and investment inherent in the income tax-and thus to promote productivity and higher living standards-should be high on the agenda of the 104th Congress. Several congressional tax reform proposals have been introduced or are close to being introduced as legislation.

Under the major taxation restructuring proposals before Congress, income in the form of capital gains is not taxed at all or only taxed if the proceeds are consumed. A common theme of the congressional tax reform proposals is that saving and investment are taxed more lightly and consumption more heavily than under current law.

Conclusion

The hard fact is that we can no longer afford the luxury of government economic policies that reward consumption, discourage saving and investment, overregulate American business, and penalize economic growth. Enactment of capital gains tax reform provisions would help move the United States toward a tax system that is more neutral toward saving and investment and pave the way for a more fundamental tax restructuring.

Notes

1. Jorgenson's analysis uses multifactor productivity, which relates output to inputs of both labor and capital. The traditional productivity measure commonly found in popular articles is labor productivity, which relates output to labor input alone.

2. Edward N. Wolff, "Capital Formation and Productivity Growth in the 1970s and 1980s: A Comparative Look at OECD Countries," in Tools for American Workers: The Role of Machinery and Equipment in Economic Growth (Washington, D.C.: ACCF Center for Policy Research, March 1993) pp. 46-76.

3. B. Douglas Bernheim, Do Households Appreciate Their Financial Vulnerabilities? An Analysis of Actions, Perceptions, and Public Policy. Special Report (Washington, D.C.: American Council for Capital Formation Center for Policy Research, August 1994).

4. John Freear and William E. Wetzel, Jr., "Equity Financing for New Technology-Based Firms," paper prepared for the Babson Entrepreneurship Research Conference, Calgary, Alberta. May 1988.

5. Dale Jorgenson, Productivity: Postwar U.S. Economic Growth (Cambridge, Mass.: MIT Press, 1995).

6. Lawrence B. Lindsey, "Capital Gains: Rates Realizations and Revenues," Working Paper No. 1893. April 1986 (Cambridge, Mass: National Bureau of Economic Research, Inc.)

ACCF
ACCF, 1750 K Street, NW, Suite 400, Washington, DC 20006 | Tel (202) 293-5811 | Fax (202) 785-8165 | info@ACCF.org