An almost universal consensus prevails that the record U.S.trade deficit for 2006 was a drag on U.S. economic growth. Theconsensus reflects a basic assumption that growing imports to theUnited States displace domestic production, reducing growth of realgross domestic product. But the consensus on trade deficits andgrowth ignores the actual record of the U.S. economy in recentdecades and the positive correlation of imports to domesticproduction.
The consensus was on full display in mid-February 2007 when theU.S. Commerce Department reported a record U.S. trade deficit for2006 of $763.6 million, including an unexpected jump in December'smonthly deficit. The 2006 numbers reflected a deficit in goods of$836.1 billion, which was offset in part by a surplus of $66.0billion in services.1
Stories in the major media typically greeted the record deficitas bad news for overall economic growth. As the New YorkTimes reported:
A growing trade deficit acts as a drag on overalleconomic growth. Economists said they expect that, in light of thenew numbers, the government will have to revise its estimate of thenation's fourth-quarter gross domestic product to show slightlyslower expansion.2
The Washington Post struck a similar note,reporting:
Though a record trade deficit had been expected, theacceleration in December caught economists by surprise, leading tothe revision of growth forecasts. A bigger trade deficit means moreU.S. demand for goods and services was satisfied by imports ratherthan by domestic firms.3
The Commerce Department itself reinforces the negativeconnection between trade deficits and growth in its quarterlyreports on gross domestic product. Its January 31, 2007,preliminary report on 4th quarter 2006 GDP stated that, while GDPhad been bolstered last year by exports and expenditures onpersonal consumption and equipment and software, imports were badnews. "Imports, which are a subtraction in the calculation of GDP,increased," the report stated matterof-factly.4
We can safely predict more worrying in the media and Washingtonabout trade deficits and growth later this week (March 14) when theCommerce Department's Bureau of Economic Analysis releasespreliminary numbers for the 2006 current account balance.
The Paradox of Faster Growth and Rising TradeDeficits
One glaring problem with the prevailing thesis on trade deficitsand growth is that evidence from recent decades does not supportit. In fact, the evidence more comfortably fits the alternativeinterpretation that an expanding economy promotes rising importsand an expanding current account deficit. An examination of annualchanges in the current account balance compared to economic growthsince 1980 shows that a "worsening" deficit is typically associatedwith faster economic growth, and an "improving" deficit with slowergrowth.
Table 1 shows years since 1980 arranged according to changes inthe current account balance as a share of GDP. The first group inTable 1 includes the 8 years in which the current account balanceas a percentage of GDP shifted in a positive direction compared tothe previous year (i.e., the deficit shrank). The second groupincludes the 10 years in which the balance as a share of GDPshifted modestly in the negative direction (i.e., the deficitincreased), up to 0.5 percent of GDP from the year before. And thethird group includes the 8 years in which the balance turnedsharply negative, by more than 0.5 percent of GDP.5
As the comparison shows, there is no evidence that an expandingcurrent account deficit is associated with slower economic growth.In fact, data show the opposite correlation:
- In those years since 1980 in which the current account deficitactually shrank as a share of GDP, real GDP growth averaged 1.9percent.
- In those years in which the deficit grew modestly, between 0.0and 0.5 percent, GDP growth averaged 3.0 percent.
- And in those years in which the current account deficitexpanded by more than 0.5 percent of GDP, real GDP growth grew byan average of 4.1 percent.
In other words, economic growth has been more than twice asfast, on average, in years in which the current account deficitgrew sharply compared to those years in which it actually declined.If trade deficits drag down growth, somebody forgot to tell theeconomy.
In reality, trade deficits tend to be pro-cyclical, growing whenthe economy expands and contracting when the economy slows or slipsinto recession. The trade deficit "improved" during each of thethree recessions the nation has suffered in the past quartercentury -- in 1981$ndash;82, 1991, and 2001. With the help ofpayments from Gulf War allies, the current account actually movedbriefly into surplus in 1991.5 Thus, U.S. Treasury Secretary Henry Paulson stoodon solid empirical ground when he noted in a March 1 speech ontrade, "Critics often ask: If trade is so good for America, why dowe run a trade deficit? These critics might be interested to knowthat the last time we ran a trade surplus  our economy was inrecession."7
The same forces are at work when the economy expands robustly,such as it did in the 1990s. Many Democratic leaders tout thestrong growth of the economy when Democratic president Bill Clintonand his economic team were in office, especially in the second halfof the 1990s through 2000. The U.S. economy performed well duringthat period by almost every measure: GDP growth, manufacturingoutput, net job creation and real incomes surged while unemploymentand poverty rates fell.
What is almost never acknowledged by trade-deficit critics isthat the same period also witnessed a rapid "worsening" of thetrade deficit. The current account deficit grew as a share of GDPin every year beginning in 1996 through 2000, rising from 1.5percent of GDP in 1995 to 4.2 percent in 2000. Many of today'scritics of the trade deficit claim credit for the strong economy ofthe late 1990s while demonizing the growing current accountdeficits that quite naturally accompanied the expansion.
The year 2006 was no exception to the general rule despite thedownward revision of GDP growth in the 4th quarter. On the basis ofthe first three quarters of current account numbers for 2006, thecurrent account deficit for all of 2006 will reach about 6.6percent of GDP, a modest expansion from 2005's share of 6.4percent.8 GDPgrowth in 2006 was also in the moderate range at 3.2 percent,placing last year comfortably within the established pattern.
Economic Growth Fuels Import Growth
Belief that a growing trade deficit means slower growth rests onthe enduring myth, perpetuated by the U.S. Commerce Department'sown language, that imports simply subtract from economic growth.The prevailing orthodoxy in public discussion about trade is that asurge in imports will depress growth because imports are assumed tolargely displace domestic production. Although there is anappealing plausibility to the belief, the evidence since 1980contradicts it.
Applying the same test as above to changes in imports and GDPsince 1980 shows that, contrary to popular wisdom, faster importgrowth has been associated with faster domestic economic growth. Inyears since 1980 in which imports of goods and services fell as ashare of GDP from the previous year, economic growth averaged 2.1percent. In years in which imports grew by up to 0.5 percent ofGDP, growth averaged 3.3 percent. And in years in which importssurged by more than 0.5 percent of GDP, growth averaged 3.6percent.9
The same economic expansion in the late 1990s that saw a rapidexpansion of the trade deficit also saw a rapid rise in imports ofgoods and services. For all the same reasons, recessions tend todampen demand for imports. During each of the three most recentrecessions, imports as a share of GDP have dropped sharply. Thatsimple fact alone should give protectionists pause. Their dreamscenario of sharply declining imports is frequently accompanied bythe economic nightmare of recession.
Paradox Explained: How Growth Drives TradeDeficit
How can the seeming paradox of faster growth and expanding tradedeficits be explained? The evidence certainly does not suggest thatan expanding trade deficit somehow fuels more rapid economic growthor that a trade deficit is necessarily good for the overalleconomy.
More plausibly, causation flows from economic growth to thetrade balance. An expanding economy increases demand not only fordomestic production but also for imports. It also promotes moredomestic investment as businesses seek to meet rising demand andcapitalize on new investment opportunities.
Rising investment opportunities, in turn, attract foreigncapital to the United States to fund investment over and above whatcan be financed through domestic savings alone. Those capitalinflows are the flip side of the current account deficit: thegreater the net inflows of capital from abroad, the greater thecurrent account deficit needs to be to accommodate those inflows.Thus, when GDP growth accelerates, so does domestic investment,inflows of foreign capital, and the current account deficit. Whilea growing current account deficit is not the cause of faster GDPgrowth, it is often its handmaiden.
Our simple comparison of the current account balance and GDPgrowth does not account for a host of other factors that caninfluence both growth and the trade balance. But at the very leastthe data fail to show any discernable negative effect on economicgrowth from a rising trade deficit. Absent any real evidence, thestandard assumption that trade deficits are a drag on growth shouldbe re-examined before it is repeated again uncritically.
1. U.S. Department ofCommerce, "U.S. International Trade in Goods and Services: December2006," U.S. Bureau of the Census, February 13, 2007, www.census.gov/foreign-trade/Press-Release/current_press_release/press.html.
2. Jeremy W. Peters, "U.S.Trade Deficit Grew to Another Record in '06," New YorkTimes, February 14, 2007.
3. Peter S. Goodman andNell Henderson, "Oil Prices, Imported Goods Push Trade Gap toRecord: Growth Forecasts Cut; Democrats Urge Action,"Washington Post, February 14, 2007.
4. U.S. CommerceDepartment, "Gross Domestic Product: Fourth Quarter 2006(Advance)," Bureau of Economic Analysis, January 31, 2007, Table2.--Contributions to Percent Change in Real Gross DomesticProduct.
6. Those payments,approximately $40 billion, were subtracted from the 1991 currentaccount number used in Table 1 in order to provide a more accuratepicture of the economy's impact on that year's current accountdeficit.
9. Author's calculations,based on Economic Report of the President, Tables B-1,B-3, and B-106.