Information about Western Hemisphere Hemisferio Occidental

18 Does the Current Account Matter?

Jacob Frenkel, and Morris Goldstein
Published Date:
September 1991
  • ShareShare
Information about Western Hemisphere Hemisferio Occidental
Show Summary Details
W. Max Corden

In this paper I shall contrast the “old view” and the “new view” of the current account. I subscribe to the new view, but I shall devote much of this paper to exploring possible qualifications to it. Given that the two “views” are by now well known, the contribution of this paper is the pursuit of these qualifications. The old view is that the current account certainly does matter and that, in general, so-called current account imbalances are undesirable and require some policy action, especially if they are unlikely to be “sustainable.” The new view is that the current account does not matter from a policy point of view at all, even though various elements that determine it are certainly relevant for policy.

The old view will be described in Section I, the new view in Section II, and possible qualifications to the new view in Section III. Section IV looks at the implications of the Barro-Ricardo debt neutrality theorem for the issue under discussion here. Section V sums up. I shall limit the scope of the paper in one particular way: it will deal only with countries that have current account deficits. While the new view is applicable also to surplus countries, as is also the discussion of qualifications, there are some special features applying to the surplus cases that will not be dealt with here.

Jacques Polak did not directly write on this subject.1 But during his long and fruitful reign, an important paper came out of the Research Department of the IMF—namely, Salop and Spitaller (1980)—which was, in fact, one of the earliest statements of the new view.2 I recall Jacques citing this paper to me with approval several years later. So the new view is hardly new now, no more than the New Economics was new when Keynesian economics finally reached Washington in 1961. But it is new in the sense that only lately has it crept into the International Monetary Fund’s World Economic Outlook, its Staff Studies for the World Economic Outlook, and the Economic Outlookof the Organization for Economic Cooperation and Development (OECD), as well becoming widely accepted in the academic world as “obvious.”3 Yet, the old view still lives, at least implicitly, with continuing concerns about “imbalances” or expressions of delight that these imbalances have been declining and may decline further.

Before launching into details, let me make two general points. First, this issue can be compared with the free trade-protection debate. The usual way in which economists approach the free trade-protection issue is to begin with the principle of comparative advantage, a principle not widely understood outside our profession, and endlessly expounded and defended by economists worldwide. This principle is, in my view, comparable with the new view. It is where one must begin, even though qualifications have to be considered. But, as we well know, given specified conditions, sound arguments for protection or intervention of some kind can be made. They do not negate the principle of comparative advantage but qualify its practical conclusions in particular cases. The issue is whether the conditions for particular arguments for protection exist, and specifically whether various arguments would actually outweigh the basic principle significantly so as to destroy the presumption in favor of free trade. The qualifications to the new view that I will explore are comparable to the explorations of arguments for protection. They may qualify the new view without destroying it.

The second point concerns the attitude one should take to the perception of “problems.” The world is full of problems, and it is certainly a role of economists to study them, to draw attention to problems that may not be widely perceived, and to urge policy changes that will avoid problems in the future. But it is surely important not only that problems be well defined but also that we do not divert our attention to imaginary problems. For example, I would argue that when German capital was moving into Spain on a large scale in recent years, that was hardly a problem in itself. Rather, it was a natural movement of funds from a high savings economy to an economy with emerging investment opportunities. Of course, conceivably it could lead to problems, as any movement of factors or goods could, but to suggest that current account imbalances within Western Europe were a problem per se is to manufacture artificial worries in a world that has enough real problems to worry about.

I. The Old View

Consider the following simple case. A country is committed to a fixed exchange rate regime. For whatever reason, there is little or no private international capital mobility. If it does exist potentially, the private sector has little or no borrowing capacity abroad. Finally, the government or central bank cannot borrow abroad, other than short term for emergency purposes. Apart from such short-term emergency borrowing, notably through the IMF, a current account deficit must thus be financed by use of the country’s own foreign exchange reserves. This broadly describes the situation that did exist for all countries other than the United States in the early years of the Bretton Woods system.

In this situation, a country can run a current account deficit for a limited period. But no positive deficit is sustainable indefinitely. It must come to an end. If it keeps going, well before the foreign exchange reserves run out, there is likely to be a foreign exchange crisis generated by expectations of devaluation. Perhaps there is a steady and predictable inflow of direct investment, as there was into Western Europe in the 1960s. This makes a prolonged current account deficit possible, but no greater than the capital inflow. Hence, the sustainable deficit may not actually be zero.

The practical conclusion is that the current account must be carefully watched so as to avoid a sudden crisis. An increase in a current account deficit, or a shift from surplus or balance into deficit, is a matter of concern, even when the deficit is seen to be temporary and hence self-correcting. On the other hand, its causes—for example, whether an increase in it is caused by a rise in private investment, a fall in private savings, or an increase in the budget deficit—are irrelevant. Any current account deficit that is expected to be lasting for a longer period in the absence of a policy change, and that exceeds predictable direct investment inflows, is “worrisome” or “a matter for concern,” to use some popular official phrases.

This is the simplest way of making sense of the old view. The trouble is that times have changed. There is private capital mobility, and in addition governments that wish to maintain their exchange rate can and do borrow on the international capital market directly to finance deficits. For all developed countries, as well as some developing ones, that is surely a correct description of their situation since the early 1970s and, to some extent, even earlier. Yet the old view has survived, at least until very recently.

It is obvious that a current account figure is a flow figure and what matters, even for the old view, is the movement to a stock situation—that is, the running down of foreign exchange reserves in the example I have given to a level that would lead to crisis. Thus, it would be a cheap criticism of the old view to say that it makes no sense to have flow targets. The only target can be a stock, such as a target foreign exchange level, combined with a view about the optimal time profile of current account flows to attain the stock. Thus, I do not distinguish the old view from the new on this ground. It seems reasonable to presume that the believers in the old view know the difference between stocks and flows. When they show concern about a current account situation or prospective path of the current account, they are concerned that it is not falling fast enough to attain the desired stock level (or ratio of the stock to, say, imports or gross national product).

The old view has been implicit in a mass of writing in this field coming from the Fund, the OECD, the Bank for International Settlements, many governments, and independent commentators, including some authors from the Institute for International Economics. The problem for me here is that any clear theoretical foundation has not been provided. The view has, indeed, only been implicit. It is reflected in extended discussions of current account imbalances and their prospects and brief remarks suggesting that reductions are welcome and represent “progress” toward an objective.

It is important not to be unfair here. It is possible that concern about current account imbalances has been based not on the logic that has just been advanced—that is, on a model where an international capital market is not available to a country—but rather on a judgment, often implicit, that one or more of the qualifications to the new view discussed in Section III below are empirically so important that they outweigh the simple logic of the new view itself. This would suggest that the new view is not new even in the minds of those who have apparently put forward the “old view.”4

The empirical evidence that the old view has influenced actual government policies is perhaps more important than what is written down. It appears that governments have deliberately sought to avoid or moderate current account imbalances, though this has been less so in the 1980s than earlier. This conclusion emerges from the search by Bayoumi (1990) and Artis and Bayoumi (1990) for explanations of the correlations between national savings and national investment, originally noted by Feldstein and Horioka (1980) and attributed by the latter to low capital mobility.5

II. The New View

In its most general form, the new view may be put as follows. The current account is the net result of savings and investment, private and public. An increase in a current account deficit can be caused by an increase in investment or a fall in savings, or any combination of these, again distinguishing private and public investment and savings. Indeed, there are many kinds of investment, and many different agents who may save, so it is not just a matter of four variables. If, just for exposition, we simplify by aggregation and suppose that there are just two agents, the private sector and the public sector, and just two decisions for each, the investment decision and the savings decision, one can conceive of an optimal outcome when all four decisions are optimal.

There are problems about this approach because the four decisions are not necessarily independent, and I shall come back to this point in Section IV. But, to proceed, the optimal outcome will carry with it a particular current account level at any point in time. Yet, this cannot really be described as the “optimal current account” at which policy should aim, because it could result from various combinations of savings and investment, public and private, and not just from the optimal combination. The current account that would result from the optimal levels of saving and investment could also be a by-product, for example, of a situation where the budget deficit was too high and private investment too low. Looking at the current account figure on its own tells us nothing. One must look at the separate decisions.

An increase in a current account deficit might have been caused primarily by an increase in the budget deficit, as in the case of the United States since 1982. Alternatively, it might have been caused by an increase in private investment combined with a decrease in private savings, as in the case of the United Kingdom and Australia recently. These are very different situations and require very different policy changes, if indeed they require policy changes at all. For example, if investment increased because the perceived productivity of investment had increased, and there is reason to believe that it is equal to or greater than the rate of interest on borrowed funds abroad, no policy change at all would be required. There are numerous factors determining optimal savings and investment, and also divergences between actual levels of these and the optimal levels, and it is these that are relevant for policy consideration.

A simple parallel from trade theory can be drawn here. I shall abstract from terms of trade effects for the moment, and assume that we are concerned with a small country facing given terms of trade and a given world interest rate. The trade volumes and trade patterns resulting from free trade are then optimal for a country if there are no “domestic distortions,” that is, if underlying production and consumption decisions are not distorted by taxes, subsidies or controls, lack of information, and so on. If there are distortions, the first-best policy is to remove them. Looking at what happens to the volume of trade tells us absolutely nothing about the distortions, and it is impossible to determine what the optimal trade volumes and trade patterns are, other than by seeing what happens, or speculating what might happen, if the distortions were removed. In a decentralized system, it is necessary to get the traffic signs and signals right—which includes providing appropriate taxes and subsidies or other incentives to deal with externalities—and then the optimal traffic flow will result.

There are now many theoretical papers that either imply or clearly state the new view and that provide intertemporal analyses of external balance. They usually assume that private (or national) savings and investment are determined independently and optimally, so that a time path for the optimal current amount emerges endogenously.6 This recent literature represents a considerable advance on earlier models that, rather misleadingly, identified current account imbalances with net national wealth decumulation or accumulation, hence assuming away domestic investment.

I put the new view in its strongest form in Corden (1977, pp. 50-51). One should assume that private savings and investment decisions are optimal unless there are particular reasons to believe to the contrary. There is no reason to presume that governments or outside observers know better how much private agents should invest and save than these agents themselves. These private decisions should not be a matter for public policy concern, other than to ensure that there are no government-imposed distortions. “Various divergencies between social and private costs and benefits should be corrected, so that private decisions are made on the basis of price signals that indicate social and not just private costs and returns. But there is no need for concern with particular quantitative outcomes, and certainly not for any public-policy quantitative targets.” On the other hand, public sector behavior—that is, the budget balance—is a matter for public policy concern and the focus should be on this. This does not mean that budget balances should be zero (however calculated), but only that a public policy issue of the appropriate fiscal policy does arise.7

It follows that an increase in a current account deficit that results from a shift in private sector behavior—a rise in investment or a fall in savings—should not be a matter of concern at all. In the United States, there is widespread confusion as to whether the problem (if any) is the budget deficit or the current account deficit, since they have tended to move together since 1982. Sometimes it has hardly seemed necessary to make the distinction. But, by contrast, recent developments in the United Kingdom and Australia have highlighted the issue raised in this paper. In both countries the central budgets have moved into rough balance or even surplus, while current account deficits have increased owing to private consumption and investment booms. Thus, these countries do not have twin deficits, and one can clearly distinguish the current account issue, which is the focus of this paper, from the fiscal policy issue which is, in principle, quite distinct.

The official position in Australia has been that the current account deficit is certainly a problem, and this view has support in the financial community and from others. But John Pitchford—in Pitchford (1990) and elsewhere—has strongly put forth the argument suggested here that there is indeed no problem, and in Australia the new view is known as the “Pitchford line.” When the United Kingdom went into current account deficit in 1988, the “Lawson doctrine” propounded the same view, though it was much disputed. As noted above, it was put forth earlier in Congdon (1982).

There was a similar situation in Chile from 1979 to 1981.8 The budget had been brought into balance, but there was massive private sector borrowing abroad, generating a large current account deficit. This is possibly the only example of this kind of situation since 1973 in a developing country. The official position was that, for the reasons given here, the current account deficit was not a problem. The unfortunate outcome in that case was that, when the debt crisis came, the government was compelled by foreign creditors to take over the private sector debts. It is a kind of situation to which I shall return below when discussing qualifications.

III. Qualifications and Complications

Let us now assume that a country’s current account deteriorates, perhaps quite suddenly and sharply, owing to a private spending boom, that is, a rise in private investment or a fall in savings. The interest rate will tend to rise, which draws in capital from abroad, and the real exchange will appreciate as a result of some combination of rise in domestic prices and nominal appreciation. If the authorities wish to avoid temporary inflation—leading possibly to the generation of inflationary expectations and hence longer-term problems, these developments will be associated with some monetary contraction, which may avoid the rise in domestic prices and which will appreciate the nominal exchange rate further. In any case, there will be a real appreciation, with familiar redistributive effects affecting tradable goods producers adversely, and probably raising real wages, at least in the nontradable sectors. We assume that there is no change in fiscal policy, though the budget balance could, of course, be affected by these developments. Naturally foreign debt will build up as a result. The question now is, again, whether the current account deficit and buildup of foreign debt are matters for public policy concern. One can think of several channels through which, conceivably, they might be.

Unsound Private Borrowing and the Signaling Role of the Current Account

Perhaps the spending boom is unsoundly based. This could be because the government or central bank is understood to have provided implicit or explicit guarantees for domestic borrowers (a moral hazard problem) or because, for whatever reason, the country is just experiencing the usual unsound boom destined to end in a bust. Even without guarantees involving the possibility of rescue operations, private losses are bound to be spread to the public sector, if only through reduced tax collections.

Basically I am assuming now either that there is a divergence between private and social interest (a domestic distortion) or that myopic private agents subject to a boom mentality do not really know what is best for themselves, and in doing themselves and their shareholders and creditors harm, they are also harming the society as a whole. One recognizes the Chilean case—where the government ended up taking over private debts—and also negative views of private sector decision making that are frequently expressed, for example, in Australia recently. Anyone familiar with the recent activities of various Australian entrepreneurs is likely to have some sympathy for such views.

There may then be a case for policies that either moderate the boom through monetary contraction, or that offset its effects through fiscal contraction. The move into sharp current account deficit can then be regarded as a warning signal of a problem. The deficit matters in the sense that it needs to be watched.

This argument does not mean that a current account deficit must always be moderated or avoided, but only that a spending boom that caused it possibly should be.9 It is still necessary to look in detail at the sources of the deficit, as taught by the new view. Not all booms are “unsound,” or even partially so, and justify second-guessing by the authorities. Furthermore, there would also be a problem if the boom manifested itself not in a current account deficit, but in higher interest rates, leading to crowding-out of “sound” investment. The new view is not really dented by this consideration.

Contamination Effect

The more private agents or governments borrow, the larger the share of their equities or the debt instruments they issue in international portfolios. The larger this share, the greater the risk factor will be, and thus the interest rates borrowers have to pay, or the lower the prices of their equities.10 This is obviously not an argument for keeping current account deficits to zero, since there cannot be any presumption that the existing stock of the nation’s financial assets held abroad happens to be optimal to start with. Furthermore, the risk factor surely depends also on the state and prospects of the economy, which depend, among other things, on the way in which the borrowed funds have been used. The expected time paths of the debt-GDP and debt-export ratios may shed some light on the risk factor, though that is the most that can be said about these favored and much cited statistics.

All this is not really relevant for the central issue of whether the current account—as distinct from the buildup of publicdebt held at home or abroad—should be a matter of public policy concern. The real issue is whether changes in the risk factor resulting from increased borrowing are wholly internalized for the various agents, private and public, or whether there is some externality or “contamination” effect. If they were wholly internalized—so that increased borrowing by one agent would not raise the risk factor facing another agent—there would be no public policy issue resulting from current account deficits. This is the key point. There has to be an externality. If there is not, then there is no need to make elaborate calculations, for example, of the likely growth in the share of U.S.-issued assets in the world’s financial portfolios, or of changes in the U.S. debt-GDP ratio. Individual agents will incorporate expected adverse effects of increased borrowing in their own borrowing—and hence savings and investment—decisions. In the absence of externalities, the new view stands completely.

Yet there is surely a possibility that borrowers docontaminate each other, so that there is an externality. Markets are concerned with country risk and do look at a country’s total debt ratios. It is not unreasonable to presume that the more some private agents of a country borrow, the higher have to be the interest rates that are paid both by the government of the country and by other private agents. Perhaps this is not always so, and if markets are sensible, they will also look at the details—how much went into consumption, how much into investment, and so on. Yet, such a concern of markets with country risk is not wholly irrational and can be explained by considerations already discussed.

Governments may have to rescue private agents in trouble, and so may encounter more financial difficulties themselves as a result of excessive private borrowing (hence increasing the need for rescheduling their own debts, for example), and, conversely, private agents may have to rescue governments in trouble through paying higher taxes to finance mounting sovereign debt-service obligations. Furthermore, if some private agents get into trouble, governments may have to raise taxes or reduce government services to finance their rescue, and this would then create financial problems for other private agents.

We have thus a genuine qualification to the new view resulting from the existence of country risk: from this aspect, at least, the current account as a whole, and not just the sources of its changes, is relevant, and increases in the deficit are “worrisome.” Decentralized decision making could lead to excessive borrowing from a national point of view. But this qualification will only become relevant once debt ratios and current account deficits exceed certain levels, and particularly when increased borrowing is for consumption rather than for investment.

There is an additional factor in the case of a very large borrower in world terms—that is, the United States. The more such an economy borrows, for whatever purpose, the more it raises world interest rates for a given risk factor and not just interest rates facing the United States. If it is a net debtor, this has an adverse effect on the economy as a whole (like any deterioration in the terms of trade), an adverse effect which will not be internalized. This kind of effect provides a standard argument from a strictly national point of view for taxing capital inflow.11 Of course, it provides no basis for targeting the current account, but only for an optimal tax (with the tax rate changing over time) that would have the net effect of reducing the stream of current account deficits to some extent. More relevant in present conditions are the international effects of the rise in interest rates brought about by growing U.S. net indebtedness. Debtors lose and creditors gain as a result; it could be argued that the loss to developing country debtors is an undesirable international redistribution effect that U.S. policies should take into account.

The Present and Future Real Exchange Rate

Much of the concern with current account deficits and their sustainability appears to be connected with exchange rate considerations and with changes in the absolute and relative positions of the tradable and nontradable sectors of the economy, as well as real wages. In other words, it is not concerned with optimal borrowing or lending considerations at all, but with relative price and distributional effects of changes in current accounts.

If a current account goes into deficit in the way described earlier, the problem can be of two kinds. First, the actual distributional effects of the change may be thought undesirable, and will certainly be perceived as such by losers. Thus, the move into deficit is likely to be associated with a real appreciation (whether through domestic prices rising as a result of the boom, or nominal appreciation resulting from capital inflow, or both), and this will adversely affect the tradable goods producers. Second, the current account deficit may be predicted to be temporary, being based perhaps on an inevitably temporary investment boom, or it may be “unsustainable” because of a growing debt-GDP ratio or growing ratio of the country’s debt instruments in international portfolios, leading to an increasing risk factor and an inevitable adjustment to a lower deficit. The temporary nature of the deficit, and hence of the real appreciation and national output patterns that go with it, are often thought undesirable because, presumably, they are sources of instability.

These two concerns—the immediate distributional effect and the dislike of a temporary shock—are somewhat contradictory. If distributional effects are temporary, they are, presumably, less sectorally damaging than if they were permanent. And if they are known to be temporary, they should not lead to costly reallocations of resources that will later be regretted as the real exchange rate depreciates again when the boom and the capital inflow come to an end.

Indeed, there seems to be some confusion about the desirability or otherwise of an “unsustainable” deficit. Is a development necessarily undesirable because it is not likely to, or indeed cannot, go on forever? Must a private consumption or investment boom, and the associated current account deficit, necessarily be stopped because eventually it will stop in any case? An investment boom may be perfectly “sound”—the expected marginal productivity of capital exceeding the foreign rate of interest—yet it will inevitably be temporary. It can be interpreted as a sound stock adjustment—a switch in the nation’s asset portfolio out of financial assets into real assets because the expected profitability of real assets has improved. Similarly, an apparent consumption boom is often really a boom in the acquisition by households of consumer durables and here, again, it can be regarded as an adjustment of asset portfolios. The policy literature shows an extraordinary concern with “sustainability.” But we might bear in mind that the rate of growth of a teenager is also not sustainable. Does that make it nonoptimal?

Nevertheless, it can be conceded that stability of the real exchange rate, and hence of the size and profitability of the tradable relative to the nontradable sectors, could be desirable, other things being equal; it could be one of a number of arguments in a plausible social welfare function. Indeed, only by introducing real exchange stability as a separate concern, independent of optimal borrowing and lending, can one rationalize a great deal of policy advocacy in this field. 12

This suggests that some public policy action designed to modify or offset, at least in part, the current account and real exchange rate effects of a private sector boom could be justified. Intervention in the foreign exchange market designed to prevent appreciation would be inflationary and hence fail to avoid real appreciation, unless there were a simultaneous fiscal contraction. With high capital mobility, monetary contraction would have little effect in reducing private spending. Hence there would have to be offsetting action in the form of fiscal contraction. This, of course, may lead to a nonoptimal budget balance—government spending becoming too low or taxes too high—an adverse effect that would have to be traded off against the objective of real exchange rate stability. But such stabilization policy aimed at the real exchange rate is subject to all the modern critiques of short-term demand management and stabilization policies.

At this point, one might also reflect on the “hard landing” scenario put forth frequently in the mid-1980s, and apparently providing an argument in favor of reducing the U.S. current account deficit because of its probable exchange rate effects.13 The argument was that a depreciation of the dollar, implicitly not just nominal but also real, was inevitable, and at some stage this would lead to a sudden shock-depreciation that would have further adverse repercussions, apparently both inflation and recession.

It has never been clear to me why a depreciation that was so widely expected, and did seem to follow logically from many forecasts, should have to happen suddenly and in a crisis atmosphere. As we know now, it has not done so. One can concede that a depreciation could have adverse effects (more inflation for given overall employment), and that reduction of the current account deficit would be associated with a real depreciation. But why would it be desirable to bring this about earlier rather than later? Possibly the argument was really that the budget deficit was excessive, and that it needed to be—and would eventually have to be—reduced. The current account and exchange rate effects would be adverse by-products of the inevitable adjustment. If this was the position, it would fit in with the new view, since the policy issue was really concerned with the budget deficit.

Current Account Deficits Generate Protectionism

The following line of thought is well known. In the United States the current account deficit is said to give rise to protectionist pressures, this presumably being a response to the adverse effects of the deficit, or of the real appreciation, on import-competing producers. Thus the argument is not that the distributional effects themselves are bad but that the pressures that result from the distributional effects have an adverse effect on the national economy.

It is worth noting that, given a floating or flexible exchange rate system, there is no presumption that increased protection would actually improve the current account—that is, increase savings or reduce investment (Corden (1990)). Furthermore, protection would just yield gains to the protected import-competing producers at the expense of the nonprotected (or less protected) ones; thus it would reshuffle, not reduce overall, the losses suffered in the tradable sector as a whole as a result of the real appreciation.

It can hardly be disputed that better ways to resist such pressures should be sought if there are no other reasons for wishing to reduce the deficit. But, if it is true that a deficit or increased deficit does generally generate such pressures and these become effective in producing more protection, there is possibly some argument here for moderating a current account deficit, even though this would be a second-best approach. But is it true? One has no difficulty thinking of some surplus countries that are by no means free traders, nor countries (such as the United Kingdom and Australia) where a shift into deficit has not led to increased protection. But the relationship between current account balances and protection is an empirical question suitable for a Ph.D. thesis.

Thinking about surplus countries, one might go on to assume that the effect could be symmetrical, deficits leading to increases and surpluses to decreases in protection. One would then expect that the effect of the Japanese and German surpluses—which have been, more or less, the mirror images of the U.S. deficit—would have been to reduce protection in those countries. This appears to have been true for Japan, but not Germany. If there were actually symmetry, the basic argument would really be destroyed. World current account imbalances would not necessarily bring about increases in world protection as a whole, since higher protection by the deficit countries (or by countries that have shifted into higher deficit) would be offset by reduced protection by the surplus countries. A reduction in the U.S. deficit associated with real depreciation of the dollar might be expected to lead to increased protection (or greater reluctance to liberalize) by other countries. Thus, the argument actually assumes an asymmetrical reaction.

Two More Qualifications?

In my determined search for qualifications to the new view, I shall try to give precise content to two other popular arguments.

(1)The first argument applies to the United States, which borrows in domestic currency. It is said that the larger the stock of dollars held abroad as a result of U.S. current account deficits, the greater is exchange rate volatility—a dependence of the exchange rate on the fickle views of foreigners. This argument should be distinguished from the one discussed above, which concerned real exchange rate instability in the medium run.

There appears to be a flaw in the argument, insofar as it focuses on current account deficits as the source of the problem. Given international capital mobility, the larger the holdings of liquid dollar-denominated assets by foreigners or by U.S. residents, the more likely are big swings in the exchange rate owing to asset market effects—that is, owing to short-term changes in expectations. Possible capital flight presents just as much of a problem in a country like Brazil where the local-currency-denominated assets (whether or not described as “money”) are held wholly by domestic residents. When the pool of liquid assets in a particular currency builds up, there can indeed be a problem, but this also comes about when local savers rather than foreigners hold large amounts of domestic money or short-term government debt.

The problem is, again, the budget deficit and not the current account. In the case of the United States, one can surely assume that the vast holdings of liquid dollar assets by U.S. residents are more than enough to ensure that the foreign exchange market will respond rapidly and possibly with large movements to changes in “news.”

(2)Next I come to the argument that continued current account deficits lead to increased and undesirable foreign ownership and control of the domestic economy. This view is at present popular in the United States, especially with regard to Japanese investment. There is a vast literature on this supposed foreign ownership problem, much of it stimulated by the postwar flow of U.S. investment abroad.14 Let us grant for the purpose of the discussion that there is a public good (or “public bad”) factor here: at the margin, at least, domestic control of capital is better than foreign control, and this consideration does not enter the private decision-making calculus. The first-best policy is then to tax income from direct investment by foreigners above the rates of tax appropriate for other reasons.15 Taxing income from all foreign investments, including income from portfolio capital and debt instruments (where no foreign control is involved), would be second-best or worse.

The outcome of a tax, whether on income from all foreign capital inflow or only on income from direct investment alone, would be to raise the domestic interest rate and the return on equities, and this would possibly increase domestic savings and certainly reduce domestic investment, the latter, at the margin, being foreign financed. Thus the outcome would indeed be a reduction in the current account deficit below the level it would have reached otherwise. But the first-best approach is surely to impose an (estimated) optimal tax on direct investment—if one really believed that reducing foreign control is a public good—and then allow the market to determine the savings, investment, and hence current account outcomes. The alternative old view approach of aiming directly at a current account target would nonoptimally discourage also foreign capital inflow that does not involve control, so that the attempt to deal with one distortion would yield a new, by-product, distortion. In addition, and most importantly, a current account target would disregard numerous factors that would normally influence and change savings and investment and that a free market modified by given rates of taxation would automatically take into account.

IV. The New View and the Barro-Ricardo Theorem

In view of the immense attention that the Barro-Ricardo debt neutrality theorem has received in the fiscal policy literature, it can hardly be ignored here.16 The basic idea is that government deficits or surpluses affect future expected taxes, and these expectations would, in turn, affect private sector savings. For given government spending, a reduction in current taxes would raise the budget deficit, and this would be offset, at least to some extent, by a rise in private savings designed to provide for increases in taxes in the future. In the extreme case there would be a full offset: a shift from tax finance to debt finance would not change total national savings.

The evidence seems to suggest that there is sometimes a tendency in that direction but certainly not a full offset. This tendency can be found in the recent British and Australian episodes referred to earlier, though the shift of the Australian current account into large deficit appears to be explained primarily by an investment boom, and I am not convinced that the Barro-Ricardo story is the right explanation for the consumption booms in the two countries. These booms are more likely to have been caused by financial liberalization that made borrowing easier. The United States since 1982 has been a laboratory experiment for the proposition, and it is obvious that private savings in the United States have not responded in the way suggested.17

The implication for our present discussion is that government and private savings or dissavings are possibly related, and certainly should be related, if individual private agents are to behave optimally from their own point of view, or from the point of view of themselves and their heirs and successors combined. This leads to an issue concerned with optimal savings. For any given level and pattern of private and government investment, given expected productivity improvements and expected terms of trade, there is, in principle, an optimal path of national savings (public and private combined). If there were an increase in the expected productivity of investment or an expected future improvement in the terms of trade, optimal national savings would fall, since some of the expected future gains should be passed back to the present, allowing present consumption to rise. In other circumstances, national savings might need to increase. If these and similar circumstances did not change (and there were no change in social time preference), national savings should stay constant.

The next step is to disregard distinctions among citizens, and to assume that the average taxpayer at any point in time is also the average saver. If the private sector were to behave optimally from a national point of view (future generations being included in “the nation”), it should then offset any changes in public sector savings. If the Barro-Ricardo theorem applied in its extreme form, the private sector would indeed do so. A shift from tax finance of government spending to debt finance would not alter national savings and hence the current account. Of course, the current account balance could still change because of changes in public and private investment levels and, in addition, optimal national savings may change for various reasons, and this would change the current account through appropriate changes in private sector savings.

It follows that ifone made the assumption that private agents were able to “see through” the implications of switches between taxing and debt finance, and, in addition, behaved optimally from a national (including future generations) point of view, private savings decisions would ensure the optimality of national savings, and any change in the current account resulting apparently from changes in savings should not be a matter of concern. One might also like to assume that private investment is always optimal, given available information, in which case only changes in the current account resulting from changes in government investment would be matters of concern. The moral is then that one should look directly at government investment, since most sources of changes in the current account would result from optimal decisions.

It has already been observed that, even ignoring the Barro-Ricardo theorem, it is hardly reasonable to assume that private savings and investment decisions are always made on the basis of undistorted signals, nor that private agents always behave optimally even from their own point of view. But now we have an additional complication or problem. The evidence seems to be pretty clear that private agents do not fully (or at all) “see through” the future tax implications of government debt finance or, even if they see through it, do not respond in their savings behavior in the extreme way that the simplest version of the Barro-Ricardo theorem suggests. There is then an additional source of nonoptimality. In fact, public debt finance gives rise to an additional “domestic distortion,” leading to a divergence between private sector savings behavior and the optimal savings from a national intergenerational point of view. Public debt finance is likely to lead to national savings that are too low, and the greater the debt-financed deficit, the greater the discrepancies between actual national savings and optimal savings (unless there are other distortions pushing in the opposite direction).

All these complications do not alter the main message of the new view. It is pointless only to look at changes in the current account. The distinction between changes originating in the private sector and the public sector, and between changes in investment and changes in savings, must still be made.

Yet, having said that, I shall continue my relentless search for qualifications to the new view. The Barro-Ricardo idea may give one a lead to understanding the common intuitive reaction that a budget deficit financed by domestic savings is likely to be less of a problem for the future than a budget deficit financed through a current account deficit.

Let us compare two countries—call them the United States and Italy—both of which have bond-financed budget deficits of similar size (relative to GDP), with the deficits having similar implications for future taxpayers.18 Let us also assume that private investment in both countries is optimal in the sense that the expected marginal social return is equal to the expected world interest rate over the relevant periods. But the United States has low private savings, just sufficient, say, to finance private investment, so that the U.S. current account deficit matches the inflow of foreign funds to finance the budget deficit. On the other hand, Italy has high private savings sufficient to finance both private investment and the budget deficit. Thus, Italy does not have a current account deficit. Actually, for the argument it is necessary only that Italian savings are higher (relative to GDP) than in the United States, but the case I describe is a very clear-cut one.

What this example does say is that the higher current account deficit of the United States reflects its lower private savings ratio. The current account deficit of the United States is then a problem if the U.S. savings ratio is too low. In practice, the current account deficit could also reflect a higher investment ratio resulting from greater investment opportunities, but we assume that this is not so in this case.

Now the question is whether Italy is saving too much or the United States too little. The intuitive view is that it is the United States and not Italy that has a problem in this case. The implicit theory could be the following. Italian taxpayers, present and future, are Barro-clever and can see through the veil of debt finance, and, furthermore, they show foresight and a concern for their heirs, and save more as a result. Thus, they ensure that there will be no problem for them in the future because of the budget deficit. By contrast, Americans have not pierced the debt veil or—more likely—do not show foresight or a concern for future generations—so that they do not save much, if at all, just because of the budget deficit. The farsighted Barro-Ricardo-thinking policy adviser or policymaker sees the U.S. problem, which manifests itself as a problem of foreign debt accumulation. The central point is that the low savings of the United States could conceivably be attributed to a failure of private savers to take into account the dissavings of the government. But it still remains true that just looking at the current account deficit is not enough, since this deficit also depends on investment.

V. Conclusion

In conclusion, one might ask whether the various qualifications and complications that have been discussed here seriously dent the new view of the current account. I do not think that they do.

First, one can concede that the current account (as distinct from the fiscal balance) is worth watching as a signal of problems in private saving and investment decisions: spending booms may be “unsound” or subject to distortions. But the new view is not even dented because one must still look at private investment and saving separately and directly, and search for distortions or signs of foolish decision making.

Second, if there are contamination effects owing to country risk, the nation’sdebt (possibly relative to GDP, or relative to exports) becomes relevant, and not just separate agents’ debts. In this case, the actual and expected path of the current account deficit must indeed be watched. But this consideration only becomes significant once debt ratios are fairly high; one may guess that it is not significant for the United States or the United Kingdom but could be significant now for Australia.

Third, there is the concern with real exchange rate stability. Here, of course, it is the real exchange rate, and not actually the current account, that matters, but the two are closely related. The question really is whether there ought to be as much emphasis on real exchange rate stability as one finds in the literature. It certainly has to be set against other objectives, notably that saving and investment, private and public, be as close to optimality as possible. Is it really sensible to forgo an investment boom because it leads to instability of the real exchange rate?

Fourth, there is the argument common in the United States that the current account deficit generates protectionist pressures that become effective and impose familiar losses on the economy. This is essentially an empirical question where the evidence (on casual empiricism) seems to be somewhat ambiguous. But perhaps this argument could justify policies—essentially second-best policies—designed in some circumstances to reduce a deficit below where it would be otherwise. The weakest link in the argument is that it assumes a lack of symmetry.

Fifth, there is the concern with foreign ownership. Even if one grants this (and I hesitate to do so), the new view is not dented at all in this case since the relationship between the current account and the increase in foreign control is quite indirect; taxes directed specifically to this concern would be first-best.

Finally, national savings may be below the optimum when there is a debt-financed fiscal deficit because of the failure of private agents to compensate fully by increasing their own savings. The evidence suggests that they are unlikely to behave like perceptive Barro-optimizers. But this consideration does not dent the new view; in fact, it fully supports it because it means that the source of the problem is specifically the budget deficit. As stressed in the new view, the budget deficit is that determinant of the current account balance which one should watch most closely to see whether there is a policy problem.


    Artis,Michael, andTamimBayoumi,“Saving, Investment, Financial Integration, and the Balance of Payments,”in Staff Studies for the World Economic Outlook (Washington: International Monetary Fund, 1990).

    • Search Google Scholar
    • Export Citation

    Barro,RobertJ.,“The Ricardian Approach to Budget Deficits,”Journal of Economic Perspectives (Nashville, Tennessee), Vol. 3 (Spring1989), pp. 37–54.

    • Search Google Scholar
    • Export Citation

    Bayoumi,Tamim,“Saving-Investment Correlations: Immobile Capital, Government Policy, or Endogenous Behavior?”Staff Papers,International Monetary Fund (Washington), Vol. 37 (June1990), pp. 360–87.

    • Search Google Scholar
    • Export Citation

    Bruce,Neil, andDouglasD. Purvis,“The Specification and Influence of Goods and Factor Markets in Open-Economy Macroeconomic Models,”in Handbook of International Economics,Vol. II, ed. by RonaldW. Jones andPeterB. Kenen(Amsterdam: North-Holland, 1985).

    • Search Google Scholar
    • Export Citation

    Congdon,Tim,“A New Approach to the Balance of Payments,”Lloyds Bank Review (London), No. 146 (October1982), pp. 1–14.

    Cooper,RichardN.,“Comments on Sachs,”Brookings Papers on Economic Activity: 1 (1981), Brookings Institution (Washington), pp. 269–73.

    • Search Google Scholar
    • Export Citation

    Corbo,Vittorio,“Reforms and Macroeconomic Adjustments in Chile During 1974-84,”World Development (Oxford), Vol. 13 (August1985), pp. 893–916.

    • Search Google Scholar
    • Export Citation

    Corden,W.Max,Trade Policy and Economic Welfare (Oxford: Clarendon Press, 1974).

    Corden,W.Max,Inflation, Exchange Rates and the World Economy(1st ed., Chicago: University of Chicago Press, and Oxford: Oxford University Press, 1977; 3rd ed., Oxford: Clarendon Press, 1985).

    • Search Google Scholar
    • Export Citation

    Corden,W.Max,“Trade Policy and the Current Account,”in The Direction of Trade Policy,ed. by CharlesS. Pearson andJamesRiedel(Cambridge, Massachusetts: Basil Blackwell, 1990).

    • Search Google Scholar
    • Export Citation

    Feldstein,Martin, andCharlesHorioka, “Domestic Saving and International Capital Flows,”Economic Journal (London), Vol. 90 (June1980), pp. 314–29.

    • Search Google Scholar
    • Export Citation

    Frenkel,JacobA., andAssafRazin,Fiscal Policies and the World Economy (Cambridge, Massachusetts: MIT Press, 1987).

    Gramlich,EdwardM.,“Budget Deficits and National Saving: Are Politicians Exogenous?”Journal of Economic Perspectives (Nashville, Tennessee), Vol. 3 (Spring1989), pp. 23–35.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund, World Economic Outlook,May1990 (Washington, 1990).

    Kemp,MurrayC.,“The Benefits and Costs of Private Investment from Abroad: Comment,”Economic Record (Sydney), Vol. 38 (March1962), pp. 108–10

    • Search Google Scholar
    • Export Citation

    Kenen,PeterB.,Managing Exchange Rates (London and New York: Rout-ledge, 1988).

    Makin,John,“International ‘Imbalances’: The Role of Exchange Rates,”The Amex Bank Review, Special Paper No. 17 (London: American Express Bank Ltd., November1989).

    • Search Google Scholar
    • Export Citation

    Marris,Stephen,Deficits and the Dollar: The World Economy at Risk (Washington: Institute for International Economics, 1985).

    McKinnon,RonaldI.,“Exchange Rate Instability, Trade Imbalances, and Monetary Policies in Japan and the United States,”in Issues in International Economics,ed. by PeterOppenheimer (London: Oriel Press, 1978).

    • Search Google Scholar
    • Export Citation

    Mellor,Patricio,“Chile,”in Latin American Adjustment: How Much Has Happened?ed. by JohnWilliamson(Washington: Institute for International Economics, 1990).

    • Search Google Scholar
    • Export Citation

    Obstfeld,Maurice,“International Finance,”in The New Palgrave,Vol. 2, ed. by JohnEatwell and others(London: Macmillan, 1987).

    Organization for Economic Cooperation and Development, Economic Outlook (Paris, December1989).

    Pitchford,J.D.,“Optimum Borrowing and the Current Account When There Are Fluctuations in Income,”journal of International Economics (Amsterdam), Vol. 26 (May1989), pp. 345–58.

    • Search Google Scholar
    • Export Citation

    Pitchford,J.D.,Australia’s Foreign Debt: Myths and Realities (Sydney: Allen & Unwin, 1990).

    Polak,JacquesJ.,“Comment,”in Macroeconomic Policies in an Interdependent World,ed. by RalphC. Bryant and others(Washington: Brookings Institution, Centre for Economic Policy Research, and International Monetary Fund, 1989).

    • Search Google Scholar
    • Export Citation

    Salop,Joanne, andErichSpitaller,“Why Does the Current Account Matter?”Staff Papers,International Monetary Fund (Washington), Vol. 27 (March1980), pp. 101–34.

    • Search Google Scholar
    • Export Citation

    Stockman,AlanC.,“On the Roles of International Financial Markets and Their Relevance for Economic Policy,”journal of Money, Credit and Banking (Columbus, Ohio), Vol. 20 (August1988), pp. 531–49.

    • Search Google Scholar
    • Export Citation

    Williamson,John,The Exchange Rate System, Policy Analyses in International Economics, No. 5 (Washington: Institute for International Economics, 2nd ed., 1985).

    • Search Google Scholar
    • Export Citation

    Yellen,JanetL., and others,“Symposium on the Budget Deficit,”The journal of Economic Perspectives (Nashville, Tennessee), Vol. 3 (Spring1989), pp. 17–93.

    • Search Google Scholar
    • Export Citation

Proposals for coordinating current account targets were especially common after the first oil shock, but also since. Jacques has been unsympathetic to such views, which suggests that he would tend to agree with what is called the new view here. It is also worth noting that he does not believe that countries have current account objectives anyway. See Polak (1989, pp. 378-79).


stated it in Corden (1977) and elaborated it further in the third (1985) edition of that book, also exploring some qualifications. The present paper is really a development of the arguments in that book. See also Congdon (1982) where the main point was taken up, developed further, and popularized in the United Kingdom.


I am indebted for this line of thought to some comments made by John Williamson at the conference.


In Polak (1989, p. 378) Jacques questions whether governments have views or objectives on the current account at all. In a footnote he remarks perceptibly that, “If countries did attach great importance to their current account balances as guides to policy, they would be sure of having good statistics for them. The fact that in recent years the world’s current account statistics summed up to a combined deficit of about $100 billion instead of the theoretical zero is proof in itself that the current account is not uppermost in governments’ minds as they plan their policies.” This is an interesting thought, but is difficult to relate to the Artis and Bayoumi (1990) conclusion cited here.


See, especially, Bruce and Purvis (1985, pp. 844-54), Frenkel and Razin (1987), Obstfeld (1987), Stockman (1988), and Pitchford (1989), These papers contain further references. Incidentally, earlier indications of the new view can be found in McKinnon (1978) and Cooper (1981), and most recently it has been advanced by Makin (1989) who put his references to “imbalances” always in quotation marks.


See the extensive discussions in the United States of this fiscal policy issue, for example, in the symposium on the budget deficit in the journal of Economic Perspectives(Yellen and others (1989)).


If there is no moral hazard problem, the suggestion that an “unsound” boom may justify public policy action to restrain it implies a belief that the official authorities may have more knowledge or judgment than the private agents creating the boom. Our modern Cartesians, working from standard first principles currently accepted by many economists, would query this: why should governments know better, when would we know that they know better, and can they be relied upon to act appropriately even when they do know better? 1 do not subscribe to the Cartesian position as religiously as some, but it was certainly implied in my own statement of the new view, in Corden (1977).


This does not refer to the effects of exchange rate expectations on the domestic interest rate nor to the effects of exchange rate risk resulting from exchange rate uncertainty. It can be assumed here that all debt is denominated in foreign currency, so that the relevant interest rate is the foreign one. In the case of debt, the concern is thus with default risk. If borrowing is in the domestic currency (as in the case of the United States), one must separate the exchange rate and default risk elements, the focus here being on the latter. (But the risk of unexpected inflation would, in that case, be a form of default risk.)


Various reasons for seeking to manage exchange rates are discussed in Kenen (1988). One concern of advocates of exchange rate targets or intervention of some kind has been with the effects of exchange rate “bubbles” resulting from speculation, these being quite distinct from fluctuations or “misalignments” that are caused by changes in investment, savings propensities, or fiscal policies. This is one element in the analysis in Williamson (1985).


See Marris (1985), which was for a time a very influential study.


The largest body of literature has come from Canada. I hope that the recent U.S. preoccupation with this problem will not involve a complete rediscovery of old ideas in new language.


There are some complications here; these are discussed in Corden (1974, pp. 335-47), which reviews various factors determining the optimal taxes on foreign investment, taking into account also taxes on domestic capital.


Barro (1989) is a convenient statement of this point of view. For criticisms, see Gramlich (1989).


This is not the actual situation. Italy’s deficit is much higher and partly it has been financed by an inflation tax.

    Other Resources Citing This Publication