Large unprecedented trade imbalances have bedeviled the world for the last 20 years. They are not a phantom menace. They have inflicted harm on the working and middle classes of advanced countries. They have provoked a backlash and very real trade and currency “wars,” including recently proposed U.S. tariffs. Their persistence, nevertheless, is the result of a mistake: the use of economic analysis that is almost a half-century out of date.
John Maynard Keynes famously noted that “practical men who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist.” Economists themselves are often no better. They obsess over the intellectual heritage of their ideas even when the underlying logic of older theories no longer applies. We see this fallacy in practice as mainstream economists dogmatically claim that twenty-first-century trade deficits are caused by free-spending citizens who don’t save enough to finance their economy’s domestic investment. This assertion is an egregious error of both logic and mathematics. (The technically more accurate and comprehensive term is “current account balance, but for ease of exposition, the more intuitive term “trade balance” is used here).
What do savings have to do with trade deficits? When countries save more than they invest domestically, they must either invest the surplus savings abroad or suffer recession. Conversely, if a country tries to invest more domestically than it can finance from its own savings, it needs to finance the investment with inflows of foreign savings. The term “international investment” or “capital flows” are other ways of describing flows of savings.
Without these flows, trade would always balance. The only way to pay for imports would be to export and the sole purpose of exports would be payment for imports. Countries cannot run trade surpluses without investing abroad and cannot invest abroad without running trade surpluses. These are opposite sides of the same coin.
Every transaction has two parties: buyers and sellers, lenders and borrowers, importers and exporters, etc. When these transactions are summed, they cancel each other out and total zero. The world’s net exports – all exports minus all imports – total zero. Every financial outflow from one country is another country’s financial inflow; they too add to zero. Countries can have external imbalances, but the world cannot.
Orthodox economists emphasize that a country’s savings-investment balance equals its trade balance. This is true and indisputable. But they then ironically err by treating each country’s trade balances in isolation, as if trade happened without a partner. The sum of the world’s trade balances equals zero. Therefore, the sum of all of the world’s savings-investment balances must also equal zero. Thus, national savings-investment balances are interdependent. If a country increases its savings-investment balance, somewhere else in the world, the savings-investment balance must fall by an equal amount and vice-versa; causality goes in both directions. The capital inflow that finances a trade deficit also reduces the savings-investment balance. The corresponding capital outflow has the opposite effect in a trade surplus country.
The assertion made by some economists – and unchallenged by the rest, with rare exceptions here and there – that every country’s domestic conditions determine its own trade and savings-investment balances, is not only wrong, but is also logically impossible. Since causality is, in fact, bidirectional, we can call this the “Unidirectionality Error.” This is the dominant narrative that blinds economic policy makers to the possibility that the problem originates in the trade surplus countries and obscures effective remedial policies.
This narrative may sometimes be correct if there is a happy coincidence of needs, such as when savings flow from an economy that has more savings than it needs to finance necessary investment to a country in the opposite situation. The former will have a trade surplus and the latter a deficit. Good examples are American investment in Europe after the Second World War, German borrowing to finance reunification, and even European investment in the United States during the 19th Century. But in other cases, the Unidirectionality Error rules out the possibility that low savings in deficit countries are the result of surplus savings elsewhere.
So, what caused America’s large deficits over the last 20 years? Were capital inflows sucked in by low American savings (the orthodox claim)? Or, alternatively, was surplus capital thrown off from countries that didn’t need it, didn’t want it, and couldn’t use it?
If America sucked in savings, interest rates would have been high as Americans aggressively borrowed abroad. But interest rates have been extremely low, and American businesses sit on cash hoards that they seem unwilling to invest. This clearly demonstrates that it was excess foreign savings looking for a home, and not insufficient American savings, that created global capital imbalances. Another obvious clue is that America’s persistent trade deficits are characterized as the “Exorbitant Privilege” of being a central-bank reserve currency issuer (no matter how loudly Americans complain about the deficits). Yet no one else wants to trade places and many countries insist on taking the opposite side of the transaction.
Far more plausible is the idea that the world’s unwanted savings pour into the United States due to the dollar’s role as the world’s primary reserve and safe-haven currency. Such flows are motivated by conditions in the countries of origin, not U.S. financial needs. This is consistent with former Fed Chair Ben Bernanke “Global Savings Glut” hypothesis, that some governments mobilized domestic savings to buy U.S. Treasuries, financing export-led growth to compensate for inadequate domestic demand (too much savings).
Bernanke did not address the resulting consequences to the United States. The only potentially positive outcome would have been if all the money financed additional, productive investment, such as Foreign Direct Investment in new research or production facilities. But direct observation tells us that it financed unprecedented trade deficits, drove interest rates down, funded bad and overpriced investments, and depressed savings rates. Reverse such inflows and American savings rates must rise.
So why do so many mainstream economists get the savings-trade relationship wrong? Primarily because their analysis was correct 50 or 100 years ago when the world economy rested on the gold standard or the Bretton-Woods system of fixed exchange rates after the Second World War. Economics, unlike physics, does not deal with universal or eternal laws. In these older systems, flows of gold or central bank reserves automatically financed short-term trade imbalances. Under flexible exchange rates, where there is no reserve movement and no automatic trade financing mechanism, foreign financing determines trade imbalances.
It is disappointing to see hard-earned knowledge and hard work overtaken by events. It is only human to be reluctant to revise one’s lecture notes and policy prescriptions. So, economists often cling to the right answer in the wrong time period. The old orthodoxy also has an intuitive appeal that helps explain its widespread acceptance. Trade surpluses and thrift are accepted by the public and business community as good, even virtuous, things. So, it is paradoxical to many that the reason why we have trade imbalances is because “virtuous” countries that have high savings invest them in “profligate” countries with trade deficits.
This heirloom analysis leads to wrong policy prescriptions. Unnecessary savings inflows slow the economy when a country borrows money to import things that could be produced by underutilized and unemployed resources and workers at home. Fiscal stimulus fails when financed by foreign savings. This stimulus is offset by increased trade deficits financed by the additional government borrowing.
Under the old fixed-exchange rate systems, tariffs reduced trade deficits by raising the price of imports. The smaller deficit was automatically financed because the country’s central bank reduced its gold or foreign reserve sales. Today, under current flexible exchange rate systems, this automatic financing no longer occurs. Tariffs can protect an individual industry, but they cannot remedy trade imbalances. Tariffs are the weapon of generals fighting the last trade war.
The worst policy advice from legacy economics is that trade deficit countries should increase saving and undertake fiscal austerity. That leads to recession and stagnation. Instead, trade surplus countries must reduce their savings glut and financial imbalances. Such actions will end the imbalances by eliminating the capital flows that finance the trade imbalances and will sustain stronger global growth.
Kenneth Austin is an International Economist with the U.S. Treasury Department. The views and analyses expressed herein are the authors’ own. They do not reflect those of the Department of the Treasury or the US Government and contain no nonpublic information.
Michael Pettis is an associate at the Carnegie Endowment and a finance professor at Peking University.