The topic of trade deficits can make a trade economist roll his or her eyes. Not because someone has discovered the hidden weakness in pro-trade arguments. Quite the opposite. It’s because it’s a topic on which people are so easily misled, where intuition and analysis readily part ways. There is a tendency for economists to say, “Trust me, bilateral trade deficits don’t matter.”
Yet we are in a time of very low trust. President-elect Trump has based much of his announced trade policy on the idea that bilateral trade deficits represent failure. Presumably better trade deals would deliver balance or a surplus. So, in the absence of trust, here’s a primer on trade deficits. With some simple examples, I’ll try to persuade you that bilateral trade deficits don’t matter; overall trade deficits are about borrowing and lending; and financial flows are at least as important as trade policy in determining trade deficits.
Let’s start with three countries – the United States, Brazil, and China. Initially, let’s imagine that the United States just ships $100m of high tech goods to Brazil, Brazil ships $100m of commodities to China, and China ships $100m of shoes to the United States. What do we have for trade deficits?
Each country has balanced trade. $100m of exports offsets $100m of imports for each. Yet if you look at the trade on a bilateral basis, the United States is running a $100m trade deficit with China, China is running a $100m trade deficit with Brazil, and so forth. But so what? Different countries are abundant in different goods and have different wants. This sort of ‘triangular’ trade allows for more possibilities than straight barter (each country balancing trade with every partner).
Conclusion #1: While overall trade deficits may matter, bilateral deficits don’t.
Let’s tweak the example and now allow for overall deficits. Say China ships $250m of shoes to the United States. Now the United States exports $100m and imports $250m – a trade deficit of $150m. China exports $250m and imports $100m – a trade surplus of $150m.
This raises a question: Why would China swap $250m of goods for $100m of goods? China wouldn’t. It gets something in return, some IOUs. Those could be dollar bills or they could be US Treasury bonds. In our case, China will end up with $150m of these dollar IOUs.
But what are those dollar IOUs good for? You can’t buy Chinese goods with dollars; those need RMB. You can’t buy Brazilian goods with dollars; those need reals. Dollars are useful for buying US goods. [Note – going to a bank or airport currency exchange doesn’t solve the problem of what to do with extra dollars or RMB. That just means finding someone else who wants to hold that currency. The question then becomes what they want to do with it – hold it or spend it. The same choices].
If China used those dollars to buy more US goods right away, then we’d have balanced trade. Alternatively, China could hold those dollars to buy US goods later. But sooner or later, the IOUs will be used to buy US goods. Essentially, the US borrowed from China (getting more goods than it gave, in exchange for IOUs) and has to repay at some point in the future.
Conclusion #2: Overall trade deficits are about borrowing and lending over time.
Are overall trade deficits good or bad? The answer is just as ambiguous as whether borrowing and lending is good or bad. It depends what you do with the loan. Borrowing to start a business may be a good idea; borrowing to fund a party or lavish vacation may not be.
One final twist on this example: We have assumed so far that the trade in shoes, high tech goods, and commodities drove all these transactions. Suppose, instead, that interest rates in the United States are higher than those in the rest of the world, or that US stocks look particularly attractive. Investors in China and Brazil want to invest their money here. That’s what’s driving their international behavior. So how do they do that?
To buy US stocks or bonds, or to make loans in the United States, they need to change those RMB and reals into dollars. Suppose Brazil and China each wants to buy, on net, $300m of US stocks and bonds. What does the United States then do with that pile of $600m worth of RMB and Brazilian reals? There are really only two options. The United States could just hold onto them. That’s the equivalent of investing in China and Brazil, offsetting their investments here with our investments there. Or the RMB and reals could be used to buy Chinese and Brazilian goods. It has to be one or the other.
Conclusion #3: Overall trade deficits (and currency movements) reflect a balance between the buying and selling of goods, on the one hand, and financial transactions, on the other.
If countries are going to invest in the United States on net (equivalently, if the United States is going to borrow…) then there will be an offsetting trade deficit. This is an accounting balance that has to hold, not an economic theory.
A quick corollary to this is that if President-elect Trump were successful in balancing US trade with the rest of the world, the United States would have to stop its borrowing very quickly. In 2015, that would have meant closing a gap of about $463bn.
Ok, you say. Nice simple toy models. But what about the real world? If we export more, don’t we get more jobs?
The graph shows a measure of the trade deficit (blue) and the unemployment rate (green) for the United States from 1992-2016. The vertical gray bars show recessions. If trade deficits moved in step with unemployment, we would expect these two lines to move in tandem, rising and falling together. Instead, there is a slight negative correlation.
So why might this be? How do we explain a period like the 1990s when the trade deficit was rising and the unemployment rate was falling? When the economy is booming, interest rates tend to be high, there are attractive investment opportunities, and foreign money comes flowing in. What’s more, prosperous Americans spend more on everything, including imports. All of that drives the trade deficit up, even as the unemployment rate is falling.
One final note on this graph: trade deficits here (as a share of GDP), go up and go down. Yet as of the early 1990s, US average tariffs were about 3 percent. They are currently about 1.4 percent. Thus, all this movement in the trade deficit occurred in the absence of big tariff cuts, strong evidence that it’s not trade policy but rather macroeconomic variables (growth rates, interest rates) driving trade deficits.
Either way you cut it – theoretically or empirically – trade deficits provide a poor guide to trade policy. Yet their appeal as a scorecard persists. That’s why trade economists roll their eyes.