Printer Friendly

The Adam Smith address: competing in the world economy: trade competition vs. growth competition.

AS I PREPARED my remarks for this meeting, I thought about the theme of this conference -- "Can the U.S. Prosper in a Competitive World?" -- and about what Adam Smith himself might have said if you could have had him as your speaker today.

Adam Smith taught us that a competitive market economy can serve the interest of consumers far better than government plans or government controls. But the topic of this year's NABE meeting is not about the virtues of that kind of competition. Rather it is about the competition between the U.S. economy and foreign economies.

Americans like to compete. We do it at work and we do it at play. We compete with others and with our own past performance. This kind of competition can bring out the best in us. And that is also true of international competition -- if we define it properly.


There are two fundamentally different senses in which we can think about competing in the world economy. The first relates to trade and the second relates to growth.

Much of the popular discussion about America's ability to compete deals with trade. Our trade deficit is taken as evidence of our lack of competitiveness. The goal of economic policy, according to this view, is to eliminate the trade deficit and to have a trade surplus instead. By this standard, we are doing very badly with our $100-plus billion annual trade deficit while the Japanese are doing very well with their $100-plus billion annual trade surplus.

If Adam Smith himself were here, he would recognize this focus on the trade balance as the same mercantilist approach he had argued against more than 200 years ago. He would remind us that the proper goal of economic policy is not measured by the size of the trade balance but by the level of national income. The wealth of a nation, to use Adam Smith's own language, is not its trade surplus or the resulting international reserves but the value of the goods and services that are produced. That is the proper domain for international competition now as it was when Adam Smith wrote The Wealth of Nations more than 200 years ago.

A positive trade balance does not give a nation a higher standard of living or a faster rate of economic growth. Although Brazil, for example, has a substantial trade surplus, real incomes are low, unemployment is extensive and economic growth is very slow. In the United States during the 1980s, real GDP grew at an above-trend rate and unemployment fell significantly even though we had an unprecedented series of massive trade deficits.

A trade deficit is not an indication that a nation has low productivity or low quality products. It is an indication that the domestic investment rate is high relative to the rate of saving. The fundamental relation that governs every nation's trade balance is the national income identity that the trade deficit equals the excess of investment over saving.

Unlike many things that we economists know, this relation is not based on an economic theory or an empirical regularity but is a basic accounting identity. We inevitably have a trade deficit if we as a nation spend more on investment than we save to finance that investment. The trade deficit brings with it the resources and the capital inflow that finance the excess of investment over saving.

A trade deficit can thus be either a reflection of a healthy vibrant economy with a high investment rate or an indication of an economy with a very low saving rate. Mexico's trade deficit of more than 6 percent of GDP is a sign of vigor and health. It reflects the strong rate of investment in the Mexican economy and the willingness of foreign investors to finance that investment because they know that it will give Mexico the ability to service its debt in the future.

Unfortunately, the U.S. trade deficit can be traced to a very low saving rate rather than to an unusually high rate of investment. In recent years, the U.S. net national saving rate has been an incredibly low 3 percent of GDP. Without a capital inflow, net investment would be limited to the same 3 percent of GDP, not enough to keep up with growth of the labor force and the population. Because of our trade deficit, we have had a capital inflow to supplement our domestic saving by amounts that varied from a high of 3.6 percent of GDP in 1987 to only 1.1 percent of GDP last year. That capital inflow permitted us to have a much higher level of investment and therefore a higher level of future productivity than would otherwise have been possible.

The gradual decline since 1987 in the size of this capital inflow (and in the current account deficit that it represents) illustrates a common process. International experience shows that industrial countries have a natural tendency to return to trade balance or near trade balance. While temporary changes in national saving rates are generally offset by a change in the international capital flow (and in the associated current account deficit), differences in saving rates among countries that persist for a decade or more cause corresponding changes in domestic investment.

This adjustment of domestic investment to domestic saving is something of a puzzle. Why don't the massive international capital flows that we see every day in financial markets permit national investment to be independent of national saving? The basic fact appears to be that these massive gross flows of funds are not matched by similarly large net flows of funds. On balance, the saving of a country tends to stay in the country where it originates.

This low net international mobility of capital may be a puzzle but it is clearly a fact. Japan has a gross saving rate of more than 30 percent of GDP while the corresponding figure for the U.S. is less than 15 percent. In spite of this 15-plus percentage point difference, Japan's capital outflow is only about 3 percent of its GDP and the U.S. capital inflow is less than 2 percent of our GDP.(1)

During the 1980s, the surge of the U.S. budget deficit led to a temporary surge in the trade deficit and a corresponding inflow of capital. The link between the two was christened the twin deficit problem. But by the end of the decade, in spite of the persistence of the budget deficit and a substantial decline in private saving, the U.S. current account deficit had fallen to less than 2 percent of GDP.

The real problem for the United States in the 1990s is not the possibility that the current trade deficits will grow but rather the risk that the combination of a continued low saving rate and a small trade deficit will keep investment exceedingly low. We have been able to cope with a low rate of investment in recent years because the overhang of real estate depresses the desire to build more offices and apartment buildings. But eventually that overhang will be absorbed and the demand for new construction will grow. If the saving rate and the trade deficit remain low, real interest rates will have to rise to restrain that demand and to crowd out other investment in order to balance the supply and demand for savings and investment.


That brings me to the second sense in which we can talk about international competition. I have argued that it is wrong to focus on competition with other countries for a larger trade surplus or a smaller trade deficit. It is not wrong though to think about competing in terms of productivity and growth. The rate of growth and the level of productivity are good standards by which to judge our own performance. Competing with other industrial nations to achieve higher economic growth and higher productivity may help to motivate public and private policies in the right direction.

Our performance by both of these standards does not measure up well. Over the past two decades, real GDP increased only about 60 percent in the United States. The percentage increase was nearly 40 percent larger in the rest of the OECD and more than twice as large in Japan. The different rates of increase of real GDP are primarily a reflection of differences in the rate of growth of productivity. According to OECD calculations, labor productivity in the business sector rose by less than 0.5 percent a year in the United States but by 1.4 percent a year in the OECD as a whole. Japan's labor productivity increased at nearly 3.0 percent a year.

It is of course much more difficult to compare the levels of productivity in different countries than it is to compare the rates of growth of productivity over time. Comparing productivity levels in the U.S. and Japan, for example, requires comparing output that is priced in dollars with output that is priced in yen. The conventional way of doing so is to convert the value of Japanese output to dollars, using a so-called purchasing power parity exchange rate. When that is done, manufacturing sector labor productivity appears to be about 50 percent higher in the United States than it is in Japan. Even in motor vehicle production, this calculation produces a result that U.S. productivity was 54 percent higher than Japanese productivity in 1989, up from 35 percent higher in 1980.|2~

I think such comparisons based on estimates of the purchasing power parity (PPP) exchange rates are very misleading. Although different studies use different values of the PPP exchange rate, a common estimate now is about 170 yen per dollar, far higher than the market exchange rate of about 105 yen per dollar. That PPP exchange rate is based on estimates by government statisticians of the relative prices in yen and dollars at which apparently "similar" products would be sold in the two countries. To be more concrete, the PPP estimate amounts to deciding what the relative values are for a particular model of a Toyota and a particular model of Chevrolet. The very high PPP value of the dollar amounts to saying that the Chevrolet is worth a good deal more relative to the Toyota than the actual prevailing market prices for the two cars in the United States. If the PPP exchange rate were a correct valuation of the products of both countries, the U.S. would not have a large trade deficit with Japan. Indeed, if the PPP exchange rate were really 170 while the actual exchange rate is 105, the yen would be terribly overvalued and that high value would make it very hard for Japanese firms to sell in the United States and to compete with American producers in other countries. In reality, however, American consumers see things in Japanese cars that the government statisticians apparently don't take into account. It may be particular features or a more favorable repair record. Whatever it is, the PPP approach as practiced overstates the value of U.S. cars relative to Japanese cars and therefore overstates the productivity of U.S. car producers relative to their Japanese counterparts. More generally, the fact that the U.S. has a large trade deficit with Japan at an exchange rate of 105 yen per dollar suggests that a PPP value over 150 is simply far too high. If the productivity comparison were made at an exchange rate that is likely to be more consistent with trade balance, say 100 yen per dollar, the Japanese productivity would be greater than U.S. productivity for most major industries.

We are of course a wealthy nation with a high standard of living. That reflects our level of education, our ability to use the best technologies in the world, and the capital stock that we have accumulated over past decades. Our real per capita income is likely to continue to grow in the future because of new technology even if there are no increases in the education of the work force or in the stock of capital. But while incomes will rise, they will rise more slowly in the United States than elsewhere. We will lose the international competition for higher growth and productivity that determine international differences in standards of living.

Improvements in education can play an important part in raising our standard of living over the next several decades. Experts are virtually unanimous in their opinion that our primary and secondary schools are not living up to their potential for creating a productive labor force. But improvements in education are not likely to add as much to output in the future as they did over the past generation when the fraction of the population twenty-five years old and over who graduated from high school doubled and the fraction that graduated from college tripled.

The lesson of international experience is that capital accumulation is the key to higher economic growth. Countries that invest more enjoy faster rates of economic growth both because the capital is inherently productive and because new capital embodies new technology.|3~ Remarkably, Adam Smith anticipated the link between capital accumulation and productivity when he observed that productivity depends on the division of labor and that in turn requires capital to finance inventories and the wages paid in advance of sales. Today we would place less emphasis on this working capital and more on the need for capital to finance purchases of machinery and equipment but we would agree with Smith's conclusion that capital is a key to higher productivity.


Because of the limited net international mobility of capital, increasing our rate of investment in the United States requires a higher national saving rate. Increasing national saving means reducing the budget deficit or raising private saving or both.

I have said and written so much about the importance of deficit reduction in the past that I will not add anything more now. Moreover, the need for deficit reduction is well accepted by policy officials, even if the recent budget legislation is far less than would be required to reduce the structural deficit significantly.|4~

In contrast, there has been no attention in the recent budget and tax discussions to the need to raise the private saving rate. Instead, several of the 1993 tax changes will have the effect of reducing private saving and depleting previously accumulated capital. For example, lowering the income limit on employers' contributions to qualified pension plans from $235,000 to $150,000 reduces substantially the incentive of higher income individuals to save. The increase in the marginal tax rate on the investment income of high income individuals from 31 percent to 39.6 percent has a similar adverse effect. With a 39.6 percent marginal tax rate, a ten-year Treasury bond now has a negative real after-tax yield!

The rise in the top tax rates on large estates also discourages accumulation by high income individuals. And the increase in the fraction of social security benefits included in the taxable income of upper income retirees penalizes those individuals who save enough during their working years to have investment income and pensions over about $30,000 a year in retirement. Taken together, the 1993 legislation is yet another blow to saving incentives in the United States. What could be done to encourage private saving in a tax-efficient way, i.e., in a way that raises private saving by more than it reduces tax revenue? One important possibility is to expand eligibility for IRAs to higher income taxpayers and to increase the amount of IRA contribution that can be made.

I am persuaded by a large volume of research that we have done at the National Bureau of Economic Research that individual retirement accounts do raise private saving in a tax-efficient way and therefore contribute to higher national saving. For example, a recent NBER study by Steven Venti and David Wise showed that raising the IRA limit by $1,000 would cause constrained families to increase personal saving by $834 (a rise in IRA saving of $856 of which only $22 would come from other saving) and to reduce taxes by only $269, so that consumption is reduced by $565.|5~ In other words, even though the higher IRA limit would cause some loss of tax revenue ($269), it would reduce consumption and raise national saving (by $565).

I believe that these very favorable estimates understate the true net positive effect of IRAs on national saving because they do not take into account the rise in corporate tax revenue that results from increased IRA saving. Some of the additional IRA saving finances new corporate equity capital. The return on that capital is subject to corporate tax as well as to tax at the individual level. The additional corporate tax revenue helps to offset the loss in personal tax revenue that comes from the IRA deduction and from the nontaxation of funds in the IRA.|6~

The corporate tax revenue is likely to be a significant factor in assessing the net tax effect. Indeed, with the so-called "back-ended" IRAs (in which initial contributions are not deducted from taxable income but funds can eventually be withdrawn tax free), the only loss of personal tax revenue is seen clearly to be the foregone tax on those funds that would otherwise have been saved in a taxable form. When the effect of the IRA on corporate tax revenue is taken into account, the back-ended IRA can actually be a source of higher tax revenue as well as increased personal saving.|7~

There are of course other tax-efficient ways of increasing national saving. Special saving accounts need not be tied to retirement. An account like the IRA from which funds could be removed after a relatively short period like five years might be more effective in attracting funds than a retirement account, particularly from younger individuals. Although the funds could in principle be withdrawn after five years, the ability to defer taxes by leaving the funds in the account might well result in relatively little withdrawal. Gradually raising the limits on such accounts would eventually convert our current income tax system to a cash-flow consumption tax to almost everyone. Reducing the relatively high capital gains tax rate would raise saving as well as improve the efficiency of the capital market by reducing the lock-in of assets with accrued gains. The effect on personal tax revenue of reducing the capital gains tax rate is so close to zero that the overall effect would be a tax-efficient increase in national saving.

These are the kinds of tax changes that our government should be pursuing if the United States is to compete effectively in the global race for higher productivity and growth. And that's the kind of world competition that will cause us to prosper as a nation and that will raise the living standard of the American people. It's also the kind of competition that Adam Smith would have welcomed and that we as economists should encourage.

1 See references at end of text.


1 More generally, an analysis of decade-average saving and investment rates in the OECD countries shows that a 1 percent of GDP rise in a nation's saving rate increases its investment rate by about 0.8 percent. This "savings retention coefficient" has been quite stable over time. See Martin Feldstein and Charles Horioka, "Domestic Saving and International Capital Flows," Economic Journal, 1980 for the basic evidence supporting this conclusion. Many authors have since reestimated this relation and found confirming evidence. For a good discussion of this subsequent work, see Jeffrey Frankel "Quantifying International Capital Mobility in the 1980s" in B.D. Bernheim and J. Shoven National Saving and Economic Performance, Chicago: University of Chicago Press, 1991. 2 These figures are from Keizai Koho Center Japan 1993: An International Comparison (1992) but similar figures are presented in studies by the OECD and by various private groups.

3 For a recent study that quantifies the importance of capital accumulation and of education in explaining international differences in economic growth, see N.G. Mankiw, D. Romer and D. Weil, "A Contribution to the Empirics of Economic Growth," Quarterly Journal of Economics, May 1992.

4 The Congressional Budget Office estimates that the structural deficit will decline from 3.3 percent of potential GDP in the current fiscal year to 2.3 percent of GDP by 1997. (Congressional Budget Office, The Economic and Budget Outlook: An Update, 1993) I believe that this is based on overly optimistic assumptions about (1) the tax revenue that will result from higher tax rates, (2) the reductions in government spending that will be achieved through better management systems and (3) the magnitude of the defense spending reductions. My reasons for this skepticism are discussed in two articles that I have written for the Wall Street Journal: "Clinton's Path to Wider Deficits" (February 23, 1993), and "Clinton's Revenue Mirage" (April 6, 1993).

5 See Steven Venti and David Wise, "The Saving Effect of Tax Deferred Retirement Accounts: Evidence from SIPP," in B.D. Bernheim and J. Shoven, National Saving and Economic Performance, Chicago: University of Chicago Press, 1991. A similar conclusion emerges from the work of Daniel Feenberg and Jonathan Skinner ("Sources of IRA Saving," Tax Policy and the Economy, 1989) based on tax panel data.

6 The method used by the Treasury and the Joint Tax Committee does not make any allowance for the effect of the IRA on corporate tax revenue. 7 I have discussed the effect of IRAs on corporate tax revenue in Martin Feldstein, "The Effect of Tax Based Saving Incentives on Government Revenue and National Saving," NBER Working Paper No. 4021, 1992.

Martin Feldstein is Professor of Economics at Harvard University and President of the National Bureau of Economic Research, Cambridge, MA.
COPYRIGHT 1994 The National Association for Business Economists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1994 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Feldstein, Martin S.
Publication:Business Economics
Article Type:Transcript
Date:Jan 1, 1994
Previous Article:Balancing Act: The New Medical Ethics of Medicine's New Economics.
Next Article:NABE presidential address: building improved official statistics for decisionmaking.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters