The Trade Balance and Winning at Trade

So what is the equivalent scorecard for trade?

Here’s where economists part ways from most other people. Economists will answer the question by mumbling about national welfare functions and terms of trade gains. It will not be something that comes out in a monthly statistical release.

Most noneconomists answer, “The trade deficit!” By their reasoning, if a country sells more than it buys (that is, exports more than it imports), that country is winning; similarly, a country that buys more than it sells is losing. So trade surpluses are good, and trade deficits are bad.

As it turns out, the trade balance is a particularly bad measure of national well-being. That’s not just because of problematic reasoning behind the argument but because it’s not usually what shows up in the numbers. We’ll start with the numbers and then work to understand them.

We can start by examining two popular hypotheses surrounding the trade balance and the US economy. First, let’s see whether higher trade balances correlate with higher economic growth. To do this, we look at 50 years of quarterly data. For the trade balance, we take net exports (exports − imports) as a percentage of GDP and plot that against GDP growth (Figure 1).

If that hypothesis were correct, these two series would be rising and falling together. If that isn’t apparent in the graph, you have a discerning eye. The correlation between the two series is 0.01, which is statistically indistinguishable from being uncorrelated. Put simply, there is no evidence that higher trade surpluses (or smaller deficits) accompany higher GDP growth.

Second, we can examine the trade balance and US unemployment. Here, the hypothesis would be that larger trade deficits accompany higher unemployment rates—a version of the assertion that imports cost jobs. So, we flip our trade balance measure and look at net imports (imports − exports) as a percentage of GDP. We compare that with a quarterly average of monthly unemployment rates (Figure 2).

Again, if this hypothesis were right, the two series would rise and fall together. Here the two series do seem to have something to do with each other, but the movement is the opposite of what is predicted by the hypothesis. The correlation is 0.31 and statistically significant. In other words, when the trade deficit goes up, the unemployment rate goes down.

This does not mean, of course, that a higher US trade deficit causes lower US unemployment. Nevertheless, the relationship here is a clear point against the common argument that trade deficits cost jobs. More proof of that claim is needed but rarely offered.

As we’ll discuss next, there are good economic reasons for the omission.

How could popular intuition differ so sharply from the data? Beyond the yearning for simplicity, two more subtle missteps often play into trade balance reasoning.

First, when we look at what goes into the GDP calculation, the accounting is as follows:

GDP = C + I + G + (X − M)

In this formula, C is consumption, I is investment, G is government spending, and (X − M) is exports minus imports (that is, net exports). Setting aside questions about levels versus changes, this accounting identity is commonly understood to indicate that a decrease in net exports results in a reduction in GDP. It seems obvious, right? But there’s a trap in the interpretation.

Imagine if the United States were to import 100 additional cars from abroad. This would increase M by the value of 100 autos and decrease (X − M) by a corresponding amount. If nothing else happened, GDP would have to go down. But something else does happen. Those cars were imported to be consumed. So C goes up by the value of 100 autos, exactly offsetting the move in M and leaving GDP unchanged. If the imports were not subtracted from the GDP figure, it would wrongly attribute goods made abroad to those made at home.

Furthermore, this basic math does not reveal the economic activity associated with the imports at issue. For example, imports often complement rather than replace domestic production. In fact, more than half of US imports are intermediate goods, raw materials, and capital equipment, which American companies use to make their final products. Even imported consumer goods can complement domestic output by reducing retail prices and thus freeing consumer dollars for spending on domestic goods and services, boosting their output.

Thus, the common characterization of imports and the trade deficit as a “drag on growth” reflects a misunderstanding of the accounting identity. Net exports are indeed subtracted from the nominal GDP figure, but they tell us nothing about how trade (imports and exports) actually affects US economic growth in the real world.

Second, the conventional wisdom about imports and unemployment is driven by a popular computation that aims to determine how many jobs we gain for a certain dollar amount of exports. Such figures often stem from a strong desire by elected officials to quantify the benefits of a trade agreement in terms of jobs gained. That figure doesn’t usually pop out of trade models, which generally show that trade will affect the composition (type) of jobs, not the total number. However, the economic models often do predict the change in trade flows that will result from an agreement. So eager staff accommodate elected officials with a dubious bit of math:

1. Determine how much the country exported in a given year ($1.694 trillion in 2008).
2. Determine how many jobs were supported by exports in that year (10.293 million).
3. Divide the answer in step 1 by the answer in step 2 to calculate that roughly 6,000 jobs were gained per every $1 billion of exports.
4. Finally, multiply the jobs figured in step 3 by the predicted increase in exports, and—voilà!—you have (again, dubiously) calculated the jobs gained by trade.

This math is problematic for several reasons (often detailed and ignored in the reports that generated the number). Here, we can focus on just one: There is a difference between marginal and average. The ratio described above was an average number. Now, suppose we were to increase exports by 5 percent. Do we need to increase the number of lawyers, accountants, janitorial staff, production line workers, and executives all by 5 percent? No. Maybe some of them, but at the margin, the same people could probably make more just by increasing their input orders. So the marginal number need not equal the average number.

However, once one adopts this faulty “exports equal jobs” logic, someone else can apply it to imports and thus predict jobs “lost” to imports in an equally flawed way. And this person also has a ratio that can say how many jobs supposedly come or go from a change in the trade balance.

One can see the flaw in this logic from a different angle too. At this writing, the United States is at 3.8 percent unemployment—a historically low figure that usually indicates full employment. Now, imagine if exports were to increase significantly and imports were to decrease significantly from here. How much lower would the unemployment rate go?

You don’t need to know the ratios to answer this one. If the unemployment rate were to push much below 3.8 percent, the Federal Reserve would grow even more concerned about inflation and raise interest rates to slow the economy down. If anything, the unemployment rate would likely rise (given the tighter monetary policy) because of all those exports.

If it seems like cheating to bring in another argument—monetary policy—when we were happily considering just trade balances and unemployment, this is exactly the problem. GDP and unemployment are predominantly determined by factors other than the trade balance.

Before turning from what the trade balance isn’t to what it is, we have one more distinction to draw. Until now, all our examples of trade balances (net exports) have been for the United States in its trade with the rest of the world. These are global figures.

But some of the most prominent trade balances in popular discourse are bilateral, such as the US trade imbalance with China. These data bring all the problems of interpreting trade balances discussed above plus an important additional problem. To explain this problem, we will once again start with the data.

To make it simple, we will look only at the share of US imports. Figure 3 shows the share of US imports coming from several Asian countries (also a collection of countries). They sum to the total share of US imports from Asia.

Look at China’s rise as a trading partner. In 1989, it accounted for 4.83 percent of US imports. By 2008, it accounted for 16.80 percent, a dramatic increase. Yet over that same period, Asia’s percentage of US imports barely budged, moving from 34.92 percent to just 32.57 percent.

This exercise demonstrates one big problem with bilateral trade figures: They are not for value-added but rather for the value of finished goods. Thus, suppose a US import product was 100 percent made in Japan in the early 1990s. Suppose the final stage of production then shifted to China in the late 1990s, so the product is now made 70 percent in Japan and 30 percent in China. In trade statistics, such as those shown above, this would not reduce Japan’s share by 30 percent of that value or increase China’s share by 30 percent of the value. Instead, it would reduce Japan’s share by 100 percent of the value and increase China’s share by the same amount. The product now counts completely as a Made in China good.

Figure 3 shows how misleading it would be—given the increasing presence of global supply chains in Asia and other regions, where goods are assembled in one country but contain inputs from several other countries—to see the long-term increase in China’s share of US imports, and an increase in the US-China trade deficit, as representing a new dominance of imports from Asia overall. Instead, Chinese goods most likely displaced goods from other Asian countries in the US market and probably contained inputs from those same nations. And the changing US-China trade balance would completely hide this fact.

This is not the only problem with bilateral trade balances. There is also the problem of triangular trade. To oversimplify, let us imagine that the United States sells $100 billion of wheat to Saudi Arabia. Saudi Arabia sells $100 billion of oil to China. China sells $100 billion of consumer manufactures to the United States. If this is all the trade they do, each country will have balanced trade ($100 billion in exports and $100 billion in imports). But the United States will run a $100 billion trade deficit with China. Even if you’re worried about how trade balances might affect the US economy, that US-China trade deficit tells you nothing.

It’s thus difficult to find any economic meaning in the bilateral trade balance, even if it causes a great deal of political excitement. Next, we will discuss what a country’s global trade balance might mean.

Understanding the trade balance requires us to revisit accounting identities. In accounting for international transactions, the trade balance is part of the balance of payments, which summarizes a nation’s international transactions in exports and imports of goods and services, foreign direct investments (FDIs) and portfolio investments (i.e., noncontrolling shares of public companies), transfer payments, and the change in foreign reserves of the central bank.

From this accounting emerges the following math, shown in Box 1:

• National income (mostly expressed as GDP) is created by producing a range of different products (including services): consumption goods (C), government spending (G), investment goods (I), and export goods (X). Each category is produced by using imports (M) as inputs, which—as previously discussed—must then be subtracted to avoid double counting (hence, the minus in Equation 1 in Box 1).
• That income is then spent on either private consumption (C), government consumption (G), or savings (S), as shown in Equation 2.
• From these two equations emerges the relationship expressed in Equation 3. For every nation engaged in international commerce:
  
  • exports (goods and services) − imports (of goods and services) = savings minus investment,
    
• which alternatively can be expressed as
  
  • current account = capital account + the change in foreign exchange reserves.

This relationship holds for all trading nations because it is based on fundamental equations of national accounting that capture how national income is created (Equation 1 in Box 1) and how it is spent (Equation 2).

Consequently, there are two measures of international transactions that must be equal: the capital account (S − I) and the current account (X − M). If domestic savings do not satisfy the domestic investment appetite, investors (or their banks) must borrow capital from abroad. If they can do so, it means that foreign savers (or their banks) trust the host country—in this case, the United States—to be a safe and profitable haven for their investment. Note that the mechanism by which these two measures equate in the US case is the value of the dollar. For the most part, exchange rates are set in global markets, where savers, investors, exporters, and importers all meet. For investors, factors such as expected exchange rate appreciation play into expected returns. For traders, exchange rate movements determine whether domestic goods appear cheap or expensive compared with those produced abroad.

The United States is a net borrower of foreign capital, meaning that foreigners invest more in the United States than people here invest abroad. According to Equation 3 in Box 1, this net capital inflow must equal the net outflow of payments for goods, services, investment income, and unilateral transfers (i.e., the current account deficit). In other words, a US current account deficit is matched by a US capital account surplus (i.e., a net inflow of foreign capital into the country).

Because net capital inflows (e.g., foreign direct investment, portfolio investment, etc.) can fund US businesses and create or support US jobs, they can partly explain the negative correlation between a trade deficit and unemployment in Figure 2. And if the foreign capital is invested well, it can even lead to new jobs or better ones than those potentially displaced by imported goods or services.

Usually, when another country invests in the United States, it demands a return, such as dividends on stock, interest rates on loans, and so on. However, there is an instance in which other countries provide loans for free—when they wish to hold dollars as their reserve currency. While there is an obvious upside to free loans, these dollar holdings add to the demand for dollars and, thus, the capital account surplus that the current account deficit must offset. Many economists have therefore concluded that as long as the US dollar maintains its long-standing role as the world’s main reserve currency, the United States will run perpetual trade deficits irrespective of other factors.

Understanding the balance of payments helps us see how the trade balance works in practice. First, the balance of payments is not comparable to a company’s or a bank’s balance sheet as it does not measure stocks of assets and liabilities but flows: specifically, trade flows, remittances to friends or family abroad, official development aid, FDIs, portfolio investments, income generated by past investment flows, and changes in national currency reserves. Second, and equally important, the national balance of payments is the aggregate of transactions between individuals. It is not fundamentally about one nation trading with another but about individuals and firms within those countries engaging in mutually agreeable transactions. Thus, the trade balance has a microeconomic foundation.

There are several sub-balances that form the balance of payments, with the current account and the capital account being its main components (Box 2).

The current account consists of the merchandise (or goods) trade balance, the services trade balance, the primary income balance (generated through income payments from foreign assets), and the secondary income balance, which primarily consists of transfers. The current account should equal the capital account plus the changes in foreign currency reserves. In total, the national balance of payments should be zero. If this is not the case, the difference is booked under errors and omissions. Typically, the difference is small, and the current account and capital account are effectively mirror images of each other.

As we now see, these technical definitions and accounting identities can help explain what drives a nation’s current account balance—not trade policy (e.g., tariffs) but savings and investment.

Particularly, a nation’s current account balance (X − M) must be equal to its capital account balance (S − I), meaning any change in the trade deficit must be accompanied by a change in the capital account balance. Thus, any real attempt by policymakers to fundamentally change the trade balance must somehow alter the balance between national savings and investment. This further implies that a nation’s trade balance has little to do with trade policy and instead is caused by underlying macroeconomic factors affecting levels of national savings and investment. Because different factors can cause changes in the trade balance, one cannot judge the desirability of a trade surplus or trade deficit without fully considering its underlying macroeconomic causes.

The first step to understanding a country’s balance of payments is to view it as the sum of the individual transactions its citizens make with partners abroad. To be sure, it is not the United States that runs the deficit; rather, it is American citizens and firms that trade and invest with foreign partners. The one exception is the Treasury Department, which sells government-issued bonds to investors abroad. However, as it is only one of many actors in determining a macroeconomic balance, we can conclude that the balance of payments and its sub-balances are fundamentally based on an individual—economists call it a microeconomic—foundation.

This microeconomic foundation of the balance of payments is also determined by the intertemporal calculus of citizens and firms making decisions about their savings and investments (at home or abroad). These decisions occur mostly simultaneously with their other decisions to buy or sell goods and services. However, common sense suggests that financial decisions precede the real transactions of the trade account.

This logic implies that in a country with a relatively low savings rate on the one hand and a relatively high level of investment and public consumption on the other hand, there will be a net capital inflow (and a resulting trade deficit).

The question of why savings are low can often be answered by examining an economy’s demographic structure (Box 3).

Net inflows or outflows of capital then lead to an international movement of purchasing power as well as movements of real and nominal exchange rates. A country with net capital inflows will experience an increase in demand for its currency, which causes a real appreciation of its currency. The appreciation encourages imports and discourages exports (everything else being equal). The opposite happens in a country with net capital outflows.

Here, the typical economist’s answer is correct: It depends. A nation’s imbalance in the current account cannot be judged without a closer look at the underlying macroeconomic drivers. Consider two scenarios:

• A country with many young people (and a relatively low savings rate) and a good investment climate will attract foreign capital, expanding productive capacity while creating more productive and better-paying jobs. The resulting current account deficit would thus be sustainable.
• On the other hand, a country with high social spending and low investment may also attract capital (e.g., through government bonds with the promise of high interest rates). No productive capacity is built, and only consumption goods are imported on net. In this case, the current account deficit would be less sustainable.

Thus, whether a current account deficit is good or bad strongly depends on how a nation’s capital inflows are used. If the inflows are invested productively and in a way that produces positive economic returns, the deficit will tend to be benign. However, if the current account deficit is primarily financing increased consumption due to higher indebtedness, the current account deficit will be more of a concern.

This framework requires us to judge a nation’s trade deficit by examining the structure of its foreign net capital inflows.

• Private investment flows, especially FDIs, as well as investments in stocks or long-term private debt, are typically productive. Foreign capital inflows not only entail new debt but also create new jobs and income, generating tax revenues. If it is an investment (and not a loan), it adds to the domestic capital stock.
• Assessing government-issued foreign debt is more complex. If the US Treasury sells its bonds abroad, they reflect public debt. Nevertheless, public debt can be spent in ways that create real returns for the economy (for example, by investing in better infrastructure or national security) or in ways that create no real returns and are thus unsustainable (such as hiring more public staff without investing or increasing subsidies in election years). The United States’ public debt can also satisfy global demand for safe assets. In this case, the repayment of such debt may not even be desirable: If savings rise, domestic savers may purchase treasuries from foreigners, thereby formally exporting capital (and reducing the current account deficit).

In summary, countries with a relatively young population and a need for domestic investment capital should expect net capital inflows, thus producing a trade deficit. That situation only becomes a problem if these capital inflows are not well-invested; for example, when Greece borrowed money after joining the eurozone, it was mainly spent on government salaries and not investments. Aging societies should invest part of their savings abroad to generate a net capital outflow and automatically achieve a trade surplus. Needless to say, an aging society with significant unemployment should still invest the bulk of its savings at home, but it can still potentially export capital. The trade surplus of such a country is caused by neither unfair behavior nor the country’s export competitiveness—instead, it is caused by the intertemporal decisions of its aging population.

The next question concerns the consequences of a trade deficit and whether governments should seek to combat a deficit or even strive for a surplus. We will use the US example to demonstrate that the answer to this question depends on several factors and that the United States has generally benefited from its trade deficit because it has used the capital inflows wisely.

The outcome of a trade deficit can vary because it is determined by the use of capital inflow, which, at first glance, is an increase in negative net wealth. If a country with a current account deficit does not invest its net capital inflow productively, the deficit is unsustainable and leads to long-term problems. It should soon be reversed. If a country with high unemployment has a high net capital outflow, its current account surplus is problematic as it signals poor investment conditions (and not any form of competitiveness).

However, if capital allocation is efficient, neither a net capital outflow (i.e., a current account surplus) nor a net capital inflow (i.e., a current account deficit) is a problem. That said, it is essential to analyze the underlying causes of current account imbalances before making a judgment.

We will now use this logic to examine the US balance of payments more closely since the end of the Bretton Woods system. During the Bretton Woods system, capital flows were highly restricted; disparities in the balance of payments were limited, and a change in foreign currency reserves typically offset any trade surplus. In such a setting, the intertemporal logic does not apply; a nation’s investment is mainly financed through its domestic savings. International lending and borrowing are the exception rather than the rule.

However, as soon as international capital flows are possible, the intertemporal calculus—which entails borrowing from the future to finance current investment opportunities—drives the balance of payments. The United States is an example of this economic logic. Since the early 1980s, the US current account has been almost permanently in deficit. This is mostly explained by intertemporal decisions about savings and investments.

As seen in the figures below, the ebb and flow of the US current account deficit in recent decades has been exactly mirrored by the net inflow of foreign capital (Figure 4) into the United States. When net capital inflows rise, as they did in the early 1980s and again from the mid-1990s to the mid-2000s, the current account deficit increases in tandem. When net investment declines, the current account deficit contracts. This inverse relationship between net investment inflows and current account deficits is driven by the inescapable logic of the national income account (Box 1).

In the early 1980s, the US government started a deregulation program and pursued tax reforms that reduced rates and, at least in the short run, also revenues. Both measures encouraged private investment. Since the US government did not reduce public spending but increased it mainly due to heightened national defense spending, the public budget went into deficit. Both of those developments—the attractive investment conditions and the government’s thirst for savings—led to an increased inflow of foreign capital.

This capital inflow caused the US dollar to appreciate compared with leading currencies such as the German mark. The stronger dollar contributed to a surge in imports and a slower increase in exports than possible without the capital inflows or the subsequent appreciation of the dollar, causing a large increase in the current account deficit. The United States experienced so-called twin deficits: Both the federal budget and the current account balance were in deficit.

The US production structure changed as it became clear that the capital inflows would be lasting. Many sectors of manufacturing faced increased international competition, while services sectors grew—these developments benefited the US economy even if specific industries were forced to undergo painful transitions. Once this quite substantial structural change was mastered, future capital inflows did not need a strong dollar appreciation to maintain the current account deficit.

Although most observers pay attention to trade in goods and its balance, this episode already shows that it is not enough to focus only on goods. To fully understand the current account, we must add the balance of services trade as well as the primary income balance, which includes interest and dividend payments on foreign investments. For years now, the United States has run a trade surplus in services, where its economy has a strong competitive advantage. US services even account for a high share of the value of goods assembled and imported from foreign countries, such as smartphones. Focusing on trade in goods alone is thus misleading.

Similarly, from the 1990s until the global financial crisis in 2008–2009, the United States experienced large net capital inflows and enormous investment, partly in services industries and partly in housing. When the housing boom ended, international capital flows became much smaller—and the so-called global imbalances decreased significantly. Nevertheless, the United States has remained a haven for foreign capital. Public debt rose sharply between 1990 and 2022, from $3.25 trillion to almost $31 trillion.

Meanwhile, the US population grew from 230 million inhabitants in 1980 to more than 335 million in 2020, an increase by nearly half in 40 years. Such a growing population has stimulated demands for both private and public investment to create jobs and secure education, health care, infrastructure, and housing.

The United States’ continued role as a haven for foreign investment also plays a role. Figure 5 shows net income from US investments abroad and foreign investments in the United States. It shows that American investors who invest overseas earn significantly more than their foreign counterparts who invest in the United States. This is true even though the United States has a high negative balance of net foreign assets, which has been accumulated through the enormous net capital inflows since the 1980s by US citizens, firms, and the government. The strong primary income surplus is good news for the sustainability of the US current account.

If the return on all assets were equal, the US primary income balance would be in deficit. What the figure tells us is that US investors are much more successful in securing higher returns on their investments abroad than their foreign counterparts are in the United States. This in turn implies that foreign investors trust the US capital market and are willing to accept lower interest and dividend payments for that security, whereas US investors can collect much higher returns (interest and dividends) for their investments in foreign destinations. It shows the strength and desirability of the United States as an investment location as well as the success of US investors abroad.

If analyzed through the lenses of economic theory, trade deficits are much less of a problem than many observers and policymakers think. That, however, does not imply that every balance of payments outcome is equally welcome. There may well be problems with a trade deficit or trade surplus, but these are different from the problems the conventional wisdom usually portrays.

Germany and the United States illustrate the misunderstood aspects of the balance of payments. It has often been rightly stated that Germany has underinvested in its domestic economy and that investment abroad, in combination with high domestic savings (in an aging society), drives the German current account and accompanying trade surplus. Indeed, although Germany should have a moderate surplus, it is far too high—but not because of beggar-thy-neighbor trade and industrial policies or the high competitiveness of German firms. Instead, the country urgently needs more private and public investments.

In the United States, which has a relatively younger population, people save a smaller share of GDP than people in Germany, large federal deficits further reduce national savings, and business investment remains high. These factors drive the United States’ large current account deficit (and trade deficit). Again, the US trade deficit may be judged too high—but not because of other countries’ unfair trade practices or US trade policy but because of low government and private savings in the United States. The chief policy goal should not be to restrict trade but to promote a higher domestic rate of savings, especially by the heavily indebted US government.

In summary, the US trade deficit is caused by the savings and investment decisions of American consumers and firms and by high federal government debt. The answer to whether the US trade deficit is bad greatly depends on whether the United States is borrowing and investing wisely.

The findings, interpretations, and conclusions expressed in this report do not necessarily reflect the views of the World Bank, its executive directors, or the governments they represent.