FRBSF Economic Letter
2000-37 | December 22, 2000
Over the last year the U.S. current account deficit has reached unprecedented levels. Figure 1 illustrates this by charting the falling trajectory of the current account balance, which essentially measures net exports of goods and services plus net income received from foreign investments. One implication of this deficit is that the U.S. overall must borrow from the rest of the world to pay for its excess of imports and to service its external debt.
Is the large current account deficit a problem for the U.S.? Economic theory offers some scenarios in which a current account deficit is a rational response to economic conditions or a response that may even enhance economic welfare. At the same time, recent research suggests that under certain circumstances, a large current account deficit may make the U.S. economy vulnerable to severe disruptions. This Economic Letter explores some recent theories and some data to understand how the current account deficit could be either an optimal outcome or a threatening one.
According to economic theory, the main cost of running a current account deficit is not the cost of actually paying off the accumulated debt but more precisely the cost of servicing the debt. Resources need to be set aside each year to pay the interest on the debt, and diverting economic resources away from alternative uses, like consumption, can lower a country’s standard of living.
Standard economic theory suggests at least two circumstances in which this cost may be worthwhile. First, if a country’s output is temporarily low, perhaps due to a recession, borrowing from abroad may be justified because buying imports softens the impact on consumption levels. Second, if a country has a sudden abundance of good investment opportunities, it may make sense to borrow from abroad when these opportunities cannot be financed out of domestic saving. After all, these investment projects may generate enhanced income in the future that can be used to finance the associated debt.
Theories of this type were developed in models that assumed optimizing economic agents trying to maximize their own welfare in a perfectly frictionless economic setting, that is, where there are no costs to transactions. One implication of these theories is that any current account deficit that arises exclusively from the actions of private agents in this context must be optimal in the sense that it is improving the welfare of people in the economy. So by definition, such a current account deficit is not something the government should worry about.
Recent developments in this theoretical literature have indicated that optimal current account deficits may be more elusive than the standard theory suggests. For example, Obstfeld and Rogoff (2000a) relax the assumption of a frictionless economic setting and augment the standard theoretical model by adding the costs of international trade, representing transportation costs or tariff barriers. They find that the presence of even modest trading frictions can strongly alter the model’s predictions; for example, small costs of importing goods can easily discourage optimizing private agents from purchasing imports. This finding suggests that it generally may not be optimal to run a current account deficit as a means to smooth consumption during recessions. In support of their model, the authors note that their prediction is consistent with the fact that national consumption levels tend to have a lower correlation across countries than do national output levels. After all, if countries tended to borrow from each other to smooth their consumption levels over periods of output fluctuations, their consumption levels should tend to move more closely together.
Similarly, the augmented model shows that transaction costs may make it expensive to import goods from abroad to undertake investment projects. As a result, optimizing agents in this environment tend to finance investment mainly out of domestic saving. The authors note that this prediction is consistent with the fact that the levels of saving and investment are highly correlated within most countries.
If the findings of Obstfeld and Rogoff are correct, then the costs of importing goods generally outweigh the potential benefits of running a current account deficit, whether it is to smooth consumption or to finance extra investment. These results, then, certainly make it more difficult to justify a large current account deficit as an optimal, welfare-improving phenomenon.
The empirical literature so far has not fully clarified how well these various theories apply to the real world. One strand of the literature tests these theories using a method called “present value tests.” These tests focus on the prediction that the optimal current account is a function of expected future changes in output net of investment. The empirical methodology then infers information about people’s expectations for future output using data on the current account itself.
These tests tend to reject the theoretical models quite strongly for most countries. The standard theory usually can explain why the current account goes into deficit when it does, typically in response to shocks to the level of output, but it usually cannot explain how big the current account deficit is. In other words, most countries tend to run larger current account deficits than can be justified in terms of optimizing behavior.
A recent contribution to this literature, Bergin and Sheffrin (2000), explores the possibility that shocks other than changes in output are driving the current account. In particular, the paper demonstrates with an extended present value test that large fluctuations in the current account in several countries can more easily be explained in terms of shocks to the interest rate or exchange rate. The idea is that fluctuations in such variables can induce households to “un-smooth” their consumption levels even if output levels are constant.
This finding also raises questions about considering a current account deficit an optimal outcome. The problem is that shocks to the interest rate or exchange rate may well arise from government action. If government action rather than purely private behavior is the underlying cause of the current account deficit, the argument stated previously for not worrying would no longer apply. For example, optimizing agents may generate a current account deficit simply as a way to make the best of a bad government policy. It is possible that corrective government action could improve welfare in this context. The implication is that, depending on the particular economic shock at work, there may be reason for governments to worry about a resulting current account deficit.
The theories also have mixed success when applied to the specific case of the U.S. By definition, the current account consists of saving (national saving plus government saving) minus investment. Figure 2 plots the U.S. current account and these components as shares of GNP. The last time the U.S. experienced a large current account deficit was in the mid-1980s. It seems that a major portion of the current account deficit at that time was low government saving, reflecting the large government budget deficit of the time. This relationship is often referred to as the “twin deficits” phenomenon. It is very hard to justify this current account deficit as an optimal response to the need to smooth consumption or to investment opportunities.
The current account deficit experienced in recent years seems to be different, however. Given that the government budget deficit has been resolved, clearly this is not a case of twin deficits. Instead, the figure suggests that the cause lies in a dramatic fall in private saving and a rise in investment. This explanation could be consistent with a theory of optimal current account deficits. For example, one feature of the recent economy has been the widespread implementation of new information technologies. If these represent profitable new investment opportunities, one might argue that it is optimal to run a current account deficit to finance them. Extending the argument further, these technologies and investments may have led consumers to anticipate a higher level of productivity and output in the future. One could argue this has helped spur the fall in private saving, as people try to smooth their consumption relative to the higher levels they anticipate for the future. One channel for these expectations to affect saving would be if the expected future productivity has spurred stock market gains and hence the wealth of consumers.
This argument merely points out that today’s large current account deficit could be explained as a rational and optimizing response to economic conditions. At a minimum, the present episode of current account deficits fits this type of explanation far better than the previous episode experienced in the 1980s.
Even if a large current account deficit can be justified from the perspective of macroeconomic theory, recent work by Obstfeld and Rogoff (2000b) offers a different reason for worry. They note that the debt accumulated by the U.S. to finance its deficits over the last decade has grown to about 20% of GDP. The authors acknowledge that this is smaller than has been accumulated by some other developed countries during the decade, and technically, it is possible to sustain and service a debt of this size. But it is unprecedented for such a large country to run a large deficit and debt of this type. The authors consider what might happen if the U.S. were forced to balance its current account suddenly because international investors decided to park their funds elsewhere. Such an event could result from a variety of unforeseen shocks, such as a slowdown in the U.S. growth rate relative to other countries or a sudden fall in the U.S. stock market.
Obstfeld and Rogoff argue that the effects of such a current account reversal would be especially potent for the U.S. economy, since the U.S. is less integrated into international markets than are other developed countries (perhaps because of the trading frictions the authors considered in the research discussed above). The result is that a modest shift in the quantity of the current account implies a large impact on the international relative price of U.S. goods and the real exchange rate. Using a simple model embodying their theoretical assumptions, the authors compute that a rapid reversal of the current account would generate a real exchange rate depreciation greater than 20% in magnitude.
The authors stop short of arguing that this real exchange rate depreciation would have large effects on the U.S. level of output or welfare. But it clearly has the potential to be severely disruptive. This suggests that even if a current account deficit may be justified as a reasonable response to shocks to output or investment opportunities, it nevertheless is also true that a large current account deficit creates a particular vulnerability to a variety of other shocks that might hit the economy down the road. This offers a warning that while the large U.S. current account deficit may well be rational and perhaps even welfare-improving, it nevertheless cannot be regarded cavalierly.
FRBSF Visiting Scholar
Assistant Professor of Economics, UC Davis
Bergin, Paul R., and Steven M. Sheffrin. 2000. “Interest Rates, Exchange Rates and Present Value Models of the Current Account.” The Economic Journal 110, pp. 535-558.
Obstfeld, Maurice, and Kenneth Rogoff. 2000a. “The Six Major Puzzles in International Macroeconomics: Is There a Common Cause?” NBER Working Paper 7777. http://papers.nber.org/papers/W7777 (accessed December 14, 2000).
Obstfeld, Maurice, and Kenneth Rogoff. 2000b. “Perspective on OECD Economic Integration: Implications for U.S. Current Account Adjustment.” Mimeo. Harvard University. http://www.economics.harvard.edu/~krogoff/jackson_hole.pdf (accessed December 14, 2000).
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120