U.S. Trade Deficits Aren't a Problem; R. Folsom and R. Gonzalez

Isn't our trade deficit evidence of shameful profli gacy? If foreigners own more here than we own abroad, how will we pay them a return on their assets? Could our trade deficit continue indefinite ly-will it ever become a surplus? Wouldn't a shift from trade deficit to surplus require the dollar's foreign exchange value to fall even more than it has in the recent past? And what about the federal government's budget deficit? Professors Folsom and Gonzalez teach in the Department of Economics, San Jose State University. They are partic ularly indebted to Heather Folsom, who at age 10 posed the fundamental question that gave rise to this essay. They wish to thank Kirk Blackerby, Betty Chu, Mario Escobar, Anna K.N. Folsom, David Henderson, Franz Hirner, J. Paul Leigh, John Navas, Geoffrey Nunn, Michael Pogodzinski, Tim Sass, and David Saurman for their sug gestions; Ralph Kozlow and Russell Scholl for explana tions of international economic data; Ann Arlene Mar quiss Folsom for her editing;and participants at the March 1989 National Social Science Association convention at Reno, Nevada, where they presented an earlier draft, for critical comments. The authors alone are responsible for the views expressed here.

Deficitsand Debt A lO-year-old at dinner posed the first, funda mental question: "Even if we import more than we export, if the exports and imports are paid for, where does the debt come from?" The question implies the answer: Of course debt comes only if our trade deficit isn't paid for. And we are paying for our net imports by selling all sorts of assets (real and financial, non-debt and debt) to foreign ers. These asset sales need not put the U.S. in debt internationally, just as domestic asset sales need not put us in debt domestically. (Some examples: Selling your house doesn't put you in debt. If you own a mortgage on someone else's house, selling that mortgage won't put you in debt, and won't increase his debt, either. And kicking down all the "for sale" signsin your neighborhood, because you don't want your neighborhood to be indebted to outsiders from other neighborhoods, makes no sense.) Trade means exchange. If we export less to for eigners than they export to us, they must be getting something else from us to compensate. Our "cur rent account" goods and services trade deficit is inevitably balanced by our foreign investment sur plus. If our (broadly defined) exports are less than our imports, then foreign private individuals, firms, or governments must be accumulating more U.S. assets (either real or financial assets, short or long term) than U.S. individuals, firms, or govern ments are accumulating foreign assets.

Although conventional wisdom assumes that our trade deficit causes our foreign investment sur plus, our trade deficit could just as well result from our foreign investment surplus. Actually, our trade deficit and investment surplus are determined simultaneously, as people choose to export, import, and invest at home or abroad. Incidentally, the dollar value of U.S.-owned assets abroad has increased in every year since at least 1960, although a decrease would mean merely that Americans preferred to invest at home. Our for eign investment surplus occurs because foreign owned assets in the U.S. are increasing even more rapidly. VoluntaryTransactions That we are financing our trade deficit by selling assets to foreigners may sound disastrous. But our trade deficit and foreign investment surplus reflect voluntary market transactions from which each party expects to benefit, or else the transaction wouldn't occur.We have a trade deficit because we think we benefit from the transactions that gener ate it, and the foreigners with whom we trade think they benefit. Both sides generally do benefit.

Note that if we were running a trade surplus, for example by exporting more automobiles than we imported, we still would be selling assets-auto mobiles-to foreigners, again to the benefit of both sides. Also note that the automobile assets that we could be selling abroad, and the real and financial assets that we in fact are selling abroad, all use or represent scarce economic resources. This similarity is obscured under the definitions currently used in international "balance of pay ments" accounting. An American-made automo bile sold abroad counts as an export and hence reduces our trade deficit (and foreign investment surplus), while a building constructed in the U.S. but sold to a foreigner counts as foreign invest ment in the U.S. and hence increases our trade deficit (by increasing our foreign investment sur plus). Yet there is no real difference between these transactions, other than that the automobile phys ically moves abroad while the building remains here.

Even though both sides benefit from their vol untary market transactions, there may be some different set of transactions, other than the ones actually agreed to and carried out, that would yield even greater benefits. We don't make the Polly anna claim that any particular set of voluntary market transactions generates the best of all possi ble worlds; the most advantageous trades may be 311 overlooked. Voluntary market transactions simply make both sides better off than they would be without the transactions. ThoseProfligateAmericans Despite many wails to the contrary, neither our many imports nor our net trade deficit show us to be frivolous and profligate. Most of our imports are consumer goods, but we also import many cap ital goods such as industrial machinery, trucks, and construction equipment. Moreover, many "con sumer goods" imports such as automobiles and even home electronics could be considered capital goods because, just like a lathe or millingmachine, they last and produce services for a long time.

Even short-term imports such as food, flowers, or quickly broken toys can add to the stock of real capital in the United States. The more we import of anything, the more domestic resources we have available for production of other commodities, including real capital goods. Some of what we buy may be judged foolish, but purchases don't become foolish merely because they are imported. Foreign-OwnedAssets Won't we have to pay back the foreign invest ments now being made in the U.S.? No. Foreign ers are buying many kinds of real and financial assets. If they buy real estate, they own it now, so we won't have to pay anything back. If they buy equities in businesses, they own those equities now, so again we have nothing to pay back. If they buy private debt, for example General Motors bonds, General Motors will have to pay neither more nor less than if the bondholders were Amer icans. If foreigners buy U.S. government debt, the U.S. government willhave to pay neither more nor less than if the debt were owned by Americans. If foreigners hold U.S. bank accounts (denominated in either dollars or foreign money), the bank's lia bilities are no greater than if these accounts were owned by Americans. All of these assets pay a return (an implicit return in the case of non-inter est bearing bank accounts) to whoever owns them, but there is nothing additional to be paid back, paid off, or paid out.

Admittedly, foreign willingness to lend to Americans may induce us to borrow more than we would otherwise. In this sense, some of our trade 312 THE FREEMAN • AUGUST 1990 deficit is being financed by new borrowing. But new borrowing from foreigners should cause no more problems than would new domestic borrow ing. If some Americans borrow and waste the pro ceeds, they become worse off (as do the lenders if the borrowers default), but whether the lenders are domestic or foreign makes no real difference. Of course, if exchange rates change, speculators who hold portfolios of net assets denominated on balance in moneys that unexpectedly depreciate, or net liabilities denominated on balance in mon eys that unexpectedly appreciate, will lose, but those losses will be balanced by others' gains. (Even a growing international debt wouldn't imply impoverishment because the proper mea sure of wealth is assets minus liabilities, not assets or liabilities alone. U.S. wealth continues to rise, because u.s.domestic saving-even after deduct ing all government budget deficits-remains posi tive.) Where will the output come from to pay the returns on the assets in the U.S. now owned by for eigners? This is an irrelevant question, since, if a foreign-owned asset is productive, its return accrues to its foreign owner; if it isn't productive, that is the foreign owner's problem, not ours. And the foreign investment was accompanied by enor mous inflows of resources (remember our huge trade deficit) resulting from exchanges to which we would not have agreed unless we expected to benefit, presumably by increasing our productive capacity or at least our economic welfare.

No other society coerced us to import more than we export and to accept huge volumes of for eign investment. We aren't a pre-perestroika East ern European nation "trading" with the Soviets. Voluntary foreign investments accompanied by resource inflows can pay their own returns. For eign purchases of U.S. assets aren't a zero-sum activity, since increases in foreign-owned assets require neither a decline in U.S.-owned assets nor a rise in U.S.-owed liabilities. Descriptions of the U.S. as a "debtor nation" are unwarranted.! TradeDeficitor Surplus? As foreigners reap their returns from owning U.S. assets, our current trade deficit could be fol lowed by a trade surplus if foreigners choose to consume their returns or invest them outside the U.S., but these choices and a resulting trade surplus aren't inevitable. Capital that flowed in need not flow out again. Foreigners could continue to reinvest their returns here. Many U.S. assets are owned by foreigners who want not to repatriate profits but to accumulate even more assets in the U.S., where private ownership rights are relatively more secure than in their home countries.

Thus our trade deficit and foreign investment surplus could persist indefinitely. Real capital flows to wherever the expected real rate of return is highest, and apparently it has been and contin ues to be higher in the U.S. than elsewhere. Even tually,in a static world, the inflow of capital would reduce U.S. rates of return to equal those else where, and the inflow would cease, but "cease" doesn't mean "reverse." In any case, the world isn't static. Even ifrates of return around the world eventually did equate, additional saving and investment would upset these equalities. The high est of the new rates of return would attract the new investment, creating new trade patterns in which the U.S. conceivably could have either a trade deficit and foreign investment surplus, or a trade surplus and foreign investment deficit. A U.S. trade surplus willfollowthe current U.S. trade deficit only to the extent that foreigners con sume or invest abroad their U.S. assets' returns, instead of reinvesting them here. Even then, if for eigners move or sell title to their U.S. assets across national boundaries, our current trade deficit with one country could be followed by a trade surplus with another country-or by no trade surplus at all, if foreigners follow their capital and migrate here, or sell their U.S. assets to other foreigners who migrate here.

Fora TradeSurplus, Mustthe Dollar'sValueDecline? For the U.S. to develop a trade surplus (and for eign investment deficit), the value of the U.S. dol lar relative to foreign money need not decline. A drop in the dollar's international value does make our exports more competitive and our imports more expensive, but it also makes our assets more attractive to foreigners and foreign assets less attractive to us. The net effect on our trade deficit and foreign investment surplus is ambiguous. We could develop. a trade surplus without the dollar falling at all, or even if the dollar's value rose. Recall that the U.S. trade deficit (imports into U.S. TRADE DEFICITS AREN'T A PROBLEM 313 the U.S. minus exports from the U.S.) necessarily equals the U.S. foreign investment surplus (foreign investment inflow into the U.S. minus U.S. invest ment outflow abroad). A drop in the dollar's inter... national value encourages U.S. exports by making them cheaper to foreigners, and discourages U.S.

imports by making them more expensive to us. If the dollar drops, the dollar value of our exports will certainly increase, but the dollar value of our imports willdecrease only ifwe cut them enough to compensate for the higher dollar prices we pay. Thus a drop in the dollar's value will reduce our trade deficit (measured in dollars) only if our exports increase enough to offset any increase in the dollar value of our imports. That is, a drop in the dollar's value will reduce our trade deficit only if either our exports or our imports are sufficiently responsive to exchange rate changes.2 Otherwise, if a drop in the dollar's value decreases our imports too little or increases our exports too little, then our trade deficit will increase instead of decrease. So much is wellknown, at least among those who have spent some time thinking about the effect on trade deficits of exchange rate changes.

Whenthe Valueof aDollarDrops... Less frequently considered is the effect of ex change rate changes on the components of our for eign investment surplus. A drop in the dollar's val ue affects international investment flows as it affects exports and imports. A drop encourages foreign investments in the u.s. by making them cheaper to foreigners, and discourages U.S. invest ments abroad by making them more expensive to us.3 It could be argued that foreign investment is rel atively insensitive to exchange rate changes. For example, with a drop in the dollar's international value, expected to be temporary, foreigners would be especially eager to buy u.s. assets but Ameri cans would want to postpone asset sales until the dollar returned to a higher "normal" value. If these motivations offset each other, asset sales wouldn't change. Even a permanent drop in the dollar's interna tional value could have little effect on foreign investment, because the demand for an investment asset presumably depends on its expected rate of return, which-it often is supposed-is unaffected by a permanent changein exchangerates. A perma nent drop in the dollar'sinternational value reduces the price of u.s. assets in terms of foreign money, but it also reduces the future income that will be earned by that asset in terms of foreign money.

However, if foreigners have any expectation that they will spend any of their future returns in the U.S., then a drop in the value of the dol lar-even if expected to be permanent-does make U.S. assets more attractive to foreigners. And if the resulting increase in foreign demand raises the dollar price of U.S. assets, Americans are encouraged to sell (despite the decline in the dollar's international value) if they expect to spend any of the proceeds of their asset sales in the Unit ed States. Only if Americans expected to spend all asset sale proceeds abroad would they be indiffer ent to the higher dollar prices for U.S. assets offered by foreigners when the dollar's interna tional value goes down. Thus if the dollar's international value drops, the dollar value of foreign purchases of U.S. assets will almost certainly increase, and the dollar value of our asset purchases abroad will decrease if we cut them enough to compensate for the higher dol lar prices we pay. A drop in the dollar's value will increase our foreign investment surplus, unless we cut our asset purchases abroad so little that their dollar value increases, and increases enough to off set foreigners' increased asset purchases in the United States.

A drop in the dollar's value will decrease our foreign investment surplus (and hence our trade deficit) only if foreigners increase the dollar value of their asset purchases here less than we increase the dollar value of our asset purchases abroad. A drop in the dollar's value will reduce our trade deficit and foreign investment surplus only if either foreign purchases of U.S. assets or our pur chases of assets abroad are sufficiently unrespon sive to exchange rate changes. 4 Otherwise, if a drop in the dollar's value increases foreign pur chases of U.S. assets too much, or decreases our purchases of assets abroad too much, then our trade deficit and foreign investment surplus will increase instead of decrease. Those who forecastthat the dollar's internation al value will drop farther on the assumption that the trade deficit must end, and also those who advocatea further drop in order to force our trade deficit to end, are assuming not only that exports 314 THE FREEMAN • AUGUST 1990 or imports are highly responsive to drops in the dollar's value, but also that net foreign investment flows aren't highly responsive to drops in the dol lar's value. When both trade and investment flows are considered, the effect of exchange rate changes on foreign trade deficits and foreign investment surpluses becomes much less obvious.5 Ultimate ly, the issue becomes an empirical question.

A drop in the dollar's international value could reduce our trade deficit and foreign investment surplus, by making our exports more competitive and our imports more expensive. But a drop in the dollar's international value could instead enlarge our trade deficit and foreign investment surplus, by stimulating foreign investment here and dis couraging U.S. investment abroad. Although recent trade statistics suggest that the drop in the dollar's international value since February 1985 is beginning to reduce our trade deficit and foreign investment surplus, an end to our trade deficit cer tainly doesn't require the dollar's value to drop more. For example, without any drop in the dollar's international value, our trade deficit could end and even become a surplus either because foreigners simply decide to buy more of our exports while investing less here, or because we decide to import less while investing more abroad. (Such decisions could result from changes in weather patterns and agricultural productivity, industrial productivity, new inventions and technologies, reliability of alternate suppliers, safety of investments in vari ous countries, government domestic and trade policies, perceived goods' quality, consumer tastes and preferences, and so forth.) Regardless whether we sell foreigners more commodities and fewer assets, or we buy from them fewer com modities and more assets, neither the demand for nor the supply of dollars on foreign exchange mar kets need change, so the dollar's foreign exchange value need not change.

All parts of the U.S. use the same money, yet resources have flowed from New England to real investments in southern and southwestern states, and then have stopped flowing and even reversed direction. The fixed exchange rate between the New England dollar and the rest-of-the-U.S. dol lar, constant for more than 200 years, has facilitat ed rather than impeded such resource movements within the U.S. by eliminating the risks offluctuat ing exchange rates. The nonhuman capital we have been discussing can move even if the owners of that capital stay put; human capital cannot move unless its owners migrate. Somehow that difference muddles our thinking and prevents us from seeing the similari ties. Propositions that apply neither to labor nor non-labor resources we correctly reject for human capital but wrongly accept for nonhuman capital. Not only foreign non-labor resources but also foreign people have come here. Do we need to "pay back" this labor inflow, by sending our chil dren abroad against their will? No. Must we have extra children, in order to create a surplus to pay off our "immigration deficit?" No. Is our trade deficit an imaginary problem worried about by hallucinating minds? Yes.

GovernmentBudgetDeficits What about our high government budget deficits and low (but still positive) saving rate? In years past, when almost all government debt was owned by Americans, government budget deficits seemed less threatening not only because they were smaller (and saving was relatively higher) than now, but also because "we owe it to our selves." Now that foreigners own about 20 percent of U.S. government debt, that saying is less accu rate and less comforting. But even in years past, "we" and "ourselves" were different people. Regardless how much gov ernment debt is held domestically or by foreigners, government debt and the interest on that debt can be paid only by taxes or by defaulting-either out right, or by inflating the debt's value away, or, in the case of foreign-held debt, by a drop in the dol lar's international value. Each of these alternatives would affect different people differently. Some will gain while others lose; nobody's wealth is likely to be unaffected.

Foreign trade "deficits" and government bud get deficits are entirely different concepts. Foreign trade deficits and investment surpluses result from voluntary market exchanges of goods, services, and assets; government budget deficits arise from government spending financed by fiscal and mon etary policies that government coercion imposes on individuals in the society. And unlike foreign trade deficits financed by foreign investment sur pluses, government budget deficits really do gen erate debt (either interest -paying bonds or nonu.s.TRADE DEFICITS AREN'T A PROBLEM 315 interest-paying money) that will be financed by coercion (taxation or some sort of default). Com pared with trade deficits, there is less assurance that government budget deficits are benign. The U.S. government budget deficit and foreign trade deficit are often described as "twin deficits," implying that the budget deficit's adverse conse quences are worsened by the foreign trade de ficit-that is, by the foreign investment surplus that helps finance the budget deficit. But this view is seriously misleading because it forgets that the trade deficit results from voluntary market trans actions, while the government budget deficit does not. Given the magnitude of the budget deficit, it isn't more serious merely because foreigners finance part (or even all) of it, since, without a for eign trade deficit and investment surplus, gross investment in the U.S. would be less. (If anything, the government budget deficit's consequences are alleviated, not worsened, by the trade deficit.) If the government budget deficit is too high, it is too high no matter whether it is financed by U.S. resi dents or by foreigners, no matter what the size of the foreign trade deficit. The trade deficit doesn't compound the government budget deficit.6 D 1. Incidentally, although asset ownership estimates state that for eigners own more assets here than we own abroad ($1,786.2 billion versus $1,253.7billion at the end of 1988), this comparison is ques tionable because other data show the V.s.receiving $2.2 billion of net foreign income in 1988.Unless Americans are consistently more sagacious investors than are foreigners, something is wrong-if they own so much more here than we own there, our net foreign income should be negative rather than positive. All these data are suspect, but the asset ownership data are especially suspect, because they undervalue V.S.-owned assets abroad by not fully allowing for ap preciation since those assets were acquired, many of them long ago.

"V.S. assets abroad are primarily direct investments that have been accumulated much earlier than foreign direct investment holdings in the V.S. and are recorded, for the most part, at their acquisition val ue, not at their current market price. As a consequence, the recorded value of foreign investment in the V.s. is less understated relative to its market value than isthat of V .S.investment abroad .... It has been estimated by two State Department economists that V.S. foreign direct investment was undervalued by between $400billion and $600 billion as of the end of 1987." Mack Ott, "Trade Deficit Myths," The Wall StreetJournal,January 19, 1990. 2. More precisely, a drop in the dollar's value will decrease our trade deficit if and only if the weighted sum of the elasticities of our export quantities plus the elasticities of the dollar prices we receive for our exports exceedsthe weighted sum of the elasticities of our import quantities plus the elasticities of the dollar prices we pay for our imports, with each elasticity weighted by the dollar value of its (export or import) transaction. These elasticities are the percentage change in quantities or dollar prices with respect to a percentage change in the international value of the dollar. (If the trade deficit is close to zero, and if the elasticity of the dollar prices we receive for our exports is zero while the elasticity of the dollar prices we pay for our imports is unity, this condition becomes simply that the weighted export and import quantity elasticities sum to more than one in absolute value.) 3. We ignore the effect of exchange rate changes on expectations, particularly expectations about future exchange rates and prices.

4. More precisely, a drop in the dollar's value will decrease our foreign investment surplus if and only if the weighted sum of the elas ticities of the quantities of our asset sales to foreigners plus the elas ticities of the dollar prices we receive for selling these assets is less thanthe weighted sum of the elasticities of the quantities of our asset purchases abroad plus the elasticities of the dollar prices we pay for assets abroad, with each elasticity weighted by the dollar value of its (asset purchase or sale) transaction. (If the investment surplus is close to zero, and if the elasticity of the dollar prices we receive for our asset sales to foreigners is zero while the elasticity of the dollar prices we pay for our asset purchases abroad is unity, this condition becomes simply that the weighted quantity elasticities of our asset sales to foreigners plus our asset purchases abroad sum to less than one in absolute value.) These elasticities are the percentage change in quantities or dollar prices with respect to a percentage change in the international value of the dollar.

5. Some readers may remember some variant of the condition in note 2 (or note 4) above as a foreign exchange market stability con dition, necessary and sufficient to make the elasticity to acquire and hold dollars, with respect to the international value of the dollar, neg ative. However, this is a stability condition only in models in which a money such as dollars is the only asset. In the present context, with exports, imports, non-money assets, and foreign money all trading against dollars, the stability condition would be that the weighted sum of the elasticities of the quantities of our exports andassets (oth er than dollars) sold to foreigners plus the elasticities of the dollar prices we receive for these sales exceeds the weighted sum of the elasticities of the quantities of our imports andassets (other than dol lars) purchased abroad plus the elasticities of the dollar prices we pay for these purchases, with each elasticity weighted by the dollar value of its (export, import, or asset purchase or sale) transaction.

If money were the only asset, it might be reasonable to suppose that any trade deficit eventually would end. As the residents of the trade surplus country received more and more of the trade deficit country's money, eventually their demand for it would begin to become satiated and hence highly inelastic. Rather than continue accumulating trade deficit country money, trade surplus country res idents would reduce their sales to the trade deficit country, or else increase their purchases from it, enough to end the trade imbalance. But this argument doesn't apply to a world containing a variety of assets, for which the total demand isn't likely to become satiated. 6. The opposite argument-that a large government budget deficit enlarges the foreign trade deficit and foreign investment sur plus by raising U.S. interest rates-is plausible. However, the con nection, if any,between budget deficits and interest rates depends on why the deficit is large-for example, whether an enlarged deficit results from an economic recession, a tax cut, an expansion of gov ernment spending on transfer payments, or an expansion of govern ment spending on goods and services. (A recession tends to lower interest rates; economic expansion tends to raise them.) Similarly,an attempt to reduce the budget deficit could either lower or raise inter est rates, depending on whether the deficit were reduced by a tax increase, lower government spending on transfer payments, or lower government spending on goods and services. It also would depend on how people reacted to the deficit-reducing policy, specifically whether V.S. private saving, and foreign investment in the V.S., rose or fell. Ultimately, interest rates are determined by the relative mag nitudes of total saving (supply of loanable funds) and investment (demand for loanable funds), although many policy discussions seem to forget this fundamental.

Empirically, most deficit increases and decreases result from a combination of causes. Consequently, and not surprisingly,empirical studies generally don't support a straightforward "larger deficits raise interest rates" hypothesis.