True and important: trade deficits are not scorecards. But under certain conditions, they’re far from benign.

March 28th, 2016 at 1:49 pm

Neil Irwin over at the NYT has apparently drawn the unenviable job of explaining what’s wrong with Trumponomics. His latest entry, on trade deficits, is framed around an important, overlooked point, but he misses some important nuances, and his conclusion re trade deficits as conditionally benign seemed pretty far off to me. As this is such an important point to my work—the economically large and persistent US trade deficits have become both a source of damaging bubbles and a steep barrier to full employment—let me elaborate. (For way more then you bargained for on the issue, see Chapter 5 in the Reconnection Agenda.)

Irwin’s point is that trade deficits are not inherently bad, nor surpluses good.

Tru dat. As long as there’s been trade, there’s been imbalanced trade, as countries invariably produce more than they consume, i.e., they’re run a trade surplus, while others, like us, will do the opposite. To somehow insist on balanced trade for all would be a huge policy mistake, one that would preclude billions of people from the reaping the benefits of trade, both as consumers and producers.

Irwin’s piece is thus a good antidote to the current trade debate, which sometimes claims not simply that trade deficits are always bad, which is false, but that trade itself is bad for America. As Paul Krugman underscores, the protectionism of today’s conservative demagogues is as ill-founded as it is predictable. When your MO is “blaming the other,” why not extend that from Muslims and immigrants to our faceless trading partners? (I’ve argued that trade agreements have become problematic and non-representative of working people both here and abroad, but that’s a different point. I’ve stressed the benefits of expanded trade, which will continue apace either way.)

But persistent deficits of the magnitude we’ve had here in the U.S. have, in fact, been highly problematic. Moreover, they are a symptom of global imbalances—a savings glut, in Ben Bernanke’s terminology—that have been, and remain, a serious problem for macro-economies across the globe. Trade deficits are not bad in a “scorecard” sense, but when they persist because they are part of a strategy of mercantilist countries to over-save, under-consume, and under-invest, big problems evolve in lots of other places.

The first problem with the persistent U.S. trade deficits (see figure below) is the extent to which they’ve displaced U.S. workers. This is not something economists, at least reasonable ones, disagree on (Irwin gets this too but gives it too short shrift here, I thought). It’s even in the models—“factor price equalization,” the idea that when you expand trade with low-wage countries, putting your relatively high-paid production workers into head-to-head competition with their low-paid workers, wages must fall here.

US Trade Deficit as Share of GDP

Source: BEA

Source: BEA

As displaced production workers move over to low-end service-sector jobs, the supply effect puts downward pressure on wages in that part of the job market as well. Economist Josh Bivens finds that non-college educated workers have lost around $1,800 in earnings per year through this channel. That’s a lot of money for them, and surely the root of a lot of the anger being stoked by the campaign.

What is less well understood is the impact of large, lasting trade deficits on investment flows. Neil argues that such flows can be either beneficial or harmful to deficit countries, but he fails to explain other key moving parts here that determine which of those outcomes are more likely. Here’s where the process—and the U.S. outcome shown in the figure above—gets far less benign.

When one country runs a trade surplus, another country must run a trade deficit. That right there tells you that the scolds who say “if only we were more frugal” are wrong about this. We are not the masters of our fates here that Neil’s rap suggests. As Bernanke intimates in the above link, when Germany runs an 8 percent trade surplus (!), other countries must consume that much more than they produce. It is in this manner that surplus countries not only export goods to deficit countries. They also import labor demand from those countries, some of whom, like peripheral Europe, could really use that demand.

Of course, we could, too. Neil’s right that surplus countries lend their excess capital to us to invest as we see fit. The happy version of this is that those capital flows put downward pressure on our interest rates, and we build all kinds of productive stuff that wasn’t being built at the higher rates that prevailed before our trade deficit got larger. The sad, and I fear more realistic version (and Bernanke got this right too in his 2005 paper that introduced the “savings glut” concept), is our rates are already crazy low and thus these flows lead to overinvestment, in junk, fiber optics, houses, oil rigs, and whatever else looks like the next big thing (the flows also push up the value of the dollar, which worsens the trade deficit).

And when I say “overinvestment,” I mean bubbles. Yes, a country can demonstrably get to full employment with persistent, large trade deficits, but they/we do it with bubbles. And bubbles must burst, meaning that we can and do get to full employment in the presence of big trade deficits, but we cannot stay there, and we will often have a hard time getting back to full employment, as has been the case in recent decades.

Bottom line, always go with Neil over Trump. But persistent, large trade deficits and surpluses concentrated in specific countries (US, China, Germany) are bad news. They cost workers in deficit countries jobs and incomes, and they are functions of man (and woman)-made savings gluts that contribute to global instability and investment bubbles, particularly when interest rates are already near zero, where they are now and where they may well be for awhile.

Print Friendly, PDF & Email

15 comments in reply to "True and important: trade deficits are not scorecards. But under certain conditions, they’re far from benign."

  1. Ed Walker says:

    TPP isn’t about tariffs. It’s about protecting patent and copyrights, especially benefiting big Pharma and Hollywood. It’s about protecting capital invested overseas by US corps moving jobs away from their fellow citizens. Why is this so hard to see? Why don’t you and Neil Irwin address the actual complaints from the left instead of hauling out Smoot-Hawley?

    • Ben Groves says:

      Right, but in this “globalized” system, trade deficits really don’t exist. The nation state doesn’t exist. The “US market” runs a deficit because that is what is required of it. To end that, means very bad things. Very bad things. The lessons will be learned harshly.

  2. spencer says:

    I have always believed that the structural trade deficit was caused by the Republican tax cuts that created a wide domestic savings-investment gap ( the federal deficit is negative savings) that required large scale foreign capital inflow to finance. At the NABE annual meeting I forecast this before it happened. This in turn lead to the structural trade deficit — remember in the early years of the Reagan administration the dollar rose over 50%.

    By ignoring this you are allowing the republican propaganda machine to win the debate.

  3. Amateur says:

    I don’t believe there has been an meaningful debate about trade in a very long time. It is easy to suspect that the problem is dishonesty, and on the part of some it certainly is a factor.

    However, trade is a complicated subject, and our models are not complicated. Models should always be simpler than reality, but they’re way too simple right now.

    We hear a lot of truths come from real knowledge of trade, such as thing like: trade deficits are not always bad, but sometimes they are bad. This is true.

    But the overwhelming problem in the debate is that nobody can define trade anymore. A lot of economists seem to believe that free capital flows are a part of free trade. Or that labor mobility is a part of free trade. Nobody seems to be able to define ‘free trade’ but everybody keeps using the term as if it is the opposite of protectionism.

    This kind of argument has been outdated since NAFTA. A lot of people knew the game changed with NAFTA. But few ever talk about it in the context of creating new models that point out what is good and what is bad.

    At this stage we shouldn’t be saying, “Free Trade Good, Protectionism Bad.” This is a nonsensical debate.

  4. Smith says:

    Stuff left out of this talk:
    1. Oil and cars – Why talk about the trade deficit without mentioning the two dominant products? More money for energy independence needed, clean renewables would also halt global warming, curb the power of oil despots, and let us breathe easier. Lazy U.S. auto manufacturers can’t sell reliability and frugal cars to Japan, or engineering to Germany.
    2. Mexico and Canada – Huge trade deficits, also related to oil and cars actually.
    3. China’s huge market and authoritarian corrupt domestic economy – tariffs are the only way because China won’t allow American imports. They insist on domestic content, shared ownership, technology transfers. There is no fair playing field.
    4. Corporate tax code – we reward overseas operations, instead tax foreign earnings before repatriation
    5. Illegal labor conditions – There should be rising minimum international standards for working conditions, child labor, wages, with real consequences.
    6. Other outcome of capital inflows – investment can mean control, half of Chinas annual U.S. trade surplus could buy the entire auto industry, lock, stock and barrel (excluding parts and labor, your milage may vary)
    7. German model – German law, regulations, and corporate governance helps deter relocation. Some say it’s related to medium size of companies also.

    Two points on the commentariat
    1. Irwin also claims the deficit is the price of being world power, usually a right wing claim about the dollar.
    2. Krugman gives capital inflows as neutral neutralizer never considering what it means when foreign governments, companies, and cultures could take control of vast segments of the American economy.

    Both Irwin and Krugman seem tone deaf to the common sense realization by American workers that trade isn’t working out for them, and both are defending deficits. Irwin’s piece was on the front page of the New York Times hard copy edition and web site, essentially a front page editorial on trade. It’s hard not to see this as part of an agenda the paper is pushing in the face of growing political opposition to the trade deficit.

    • Smith says:

      The paper of record continues it’s campaign for global trade in the guise of new stories:
      This erroneous statement is presented as fact:
      “Trade flows make up a small part of America’s economic activity. The primary explanations for the stagnation of middle-class incomes are necessarily domestic.”
      I guess the logic behind this is that a 4% trade deficit or even a 2% trade deficit is so small it couldn’t possibly affect stagnating middle-class incomes. Just like a 2% drop in GDP, or a 2% GDP output gap, or 2% extra unemployment or underemployment couldn’t possibly be what’s holding down wages. After all, what’s 2% compared to 98%. Obviously the primary explanation is necessarily somewhere else.

      This is how the establishment, both right and left, gives rise to Trump.

    • Smith says:

      Still more articles and analysis favoring trade, this one on the front page of the business section.
      Just how wrong and purposefully misleading the article is can be demonstrated by this quote,
      “For one thing, the Mexican economy is still tiny compared with that of the United States and its trade surplus has remained relatively small.”
      The Mexican GDP at 1.2 trillion vs. the U.S. 18 trillion is 15%, hardly the definition of tiny. Mexico may overall even run a trade deficit, but the trade surplus with the US is in the neighborhood of $54 billion, the largest portion of which is cars, with net oil trade accounting for about $11 billion of that (2013 numbers). Moreover, Mexico’s US trade surplus accounts for about 14% of the US trade deficit and lags only behind China, Japan, and Germany, ahead of Canada, and Saudi Arabia. Back the envelope says $54 billion is .1 to .2 percent of employment, 50,000 to 100,000 jobs, vs 800,000 jobs in U.S. auto industry alone. Still it’s like saying you’re losing 1 in 10 auto jobs. Not a big deal?
      The surplus army of labor suppresses wages until there is an actual shortage. An oversimplified model shows whether there’s 1 or 10 workers ready to take my place, there’s no shortage while there’s even 1. The last .1 of unemployed workers has a disproportionate effect.

    • Smith says:

      Yet more evidence the Times has an agenda on promoting a free trade agenda, as shown in today’s (Sunday’s) lead editorial. Although in this instance, they aren’t trying to pass off opinion pieces as factual news articles and unbiased analysis, there are other problems.
      In critiquing any effort to reverse the effects of NAFTA the Times states:
      “And the United States would gain little from undoing that and other trade deals because, as studies have found, they did not cause the huge job losses that critics feared.”

      Trouble is “huge job losses” is a link to this report:
      But that report is a study in contradictions. It says NAFTA had no substantial effect on the U.S. economy because trade, and the trade deficit only ballooned after NAFTA was passed.

      This is like defending a murder suspect by saying the victim was found dead only after my client had met with them.

      The relevant excerpts from the report (April 2015 to promote TPP no doubt) follow directly below. Two other points, the free trade argument is specifically targeted to blunt the Sanders Trump appeal (the real motivation?), and no mention made of German law, labor structure, and business culture norms that severely inhibit factory relocations.

      “In reality, NAFTA did not cause the huge job losses feared by the critics or the large economic gains predicted by supporters. The net overall effect of NAFTA on the U.S. economy appears to have been relatively modest, primarily because trade with Canada and Mexico accounts for a small percentage of U.S. GDP”

      “The overall net effect of NAFTA on the U.S. economy has been relatively small, primarily
      because total trade with both Mexico and Canada was equal to less than 5% of U.S. GDP at the
      time NAFTA went into effect. ”

      “The United States is, by far, Mexico’s leading partner in merchandise trade. U.S. exports to
      Mexico increased rapidly since NAFTA, increasing from $41.6 billion in 1993 to $240.3 billion
      in 2014, an increase of 478%. U.S. imports from Mexico increased from $39.9 billion in 1993 to $294.2 billion in 2014, an increase of 637%. The trade balance with Mexico went from a surplus of $1.7 billion in 1993 to a deficit of $74.3 billion in 2007. Since then, the trade deficit with Mexico has fallen to $53.8 billion in 2014.”

      “U.S. trade with Canada more than doubled in the first decade of the FTA/NAFTA (1989-1999)
      from $166.5 billion to $362.2 billion. U.S. exports to Canada increased from $100.2 billion in
      1993 to $312.1 billion in 2014, an increase of 211%. U.S. imports from Canada increased from
      $110.9 billion in 1993 to $346.1 billion in 2014, an increase of 212% The United States has run a trade deficit with Canada since the FTA/NAFTA era, increasing from $9.9 billion in 1989 to $74.6 billion in 2008, before falling back during the 2009 recession.”

  5. jan andreassen says:

    Keynes warned the US at Bretton Woods 1944, that this was bound to happen

  6. A says:

    With this kind of analysis, it benefits the reader if you could elaborate on the counterfactuals. For example, the point about reduced U.S. wage growth due to cheap foreign labor is ambiguous, and might switch counterfactuals multiple times in a couple of sentences. Is the counterfactual a world where U.S. labor is protected, but GDP stays on the same path, and China adjusts accordingly? Or is it that labor is protected, GDP adjusts, and China adjusts in such a way that U.S. labor isn’t diminished? It’s confusing for the reader because one could say that U.S. labor value is diminished, in a real sense, simply by Chinese labor getting over historical hurdles. Expanded trade improves pricing information, while also shaping the respective labor markets. So are these counterfactuals about inihibiting the price discovery, leaving real U.S. labor diminishment for the long-run, or about inhibiting foreign labor development, and then ignoring changes in domestic GDP paths? There are simply too many possible interpretations. It would be a great service if you could commit to an clear counterfactual, even for the purposes of a more limited discussion.

    • Jared Bernstein says:

      Good q. Counterfactual is, eg in Bivens, that trade deficit with low-wage countries does not increase post base year (1979?–check his report).

  7. dwb says:

    “Savings glut” has come to mean econ-speak for bad monetary policy. The Fed has been undershooting inflation for years, and over-forecasting growth. A few months ago they predicted 4 rate hikes, now they are back to talking about asymmetric policy.

    The “strong dollar policy” combined with the Fed errors have led to widespread importation of world unemployment. Back in the 90s when inflation was still running high, this was a feature, not a bug.

    Bottom line: Inflation target should be 2-3% combined with a policy that neither encourages not discourages the “strong dollar” – that weakens the dollar to correct the trade deficit.

    The eggheads in Washington have been getting it wrong and deserve to be fired.

  8. Supposn says:


    For a half-century, USA has been annually purchasing more goods than we produce which reduce our annual GDP, drag upon our aggregate payrolls and their numbers of jobs. I disagree with the concept that nations’ annual trade deficits can ever be net beneficial.

    Unlike most service products, goods products can and usually are produced in locations remote from their purchasers and users. When production of USA goods are moved to beyond USA’s borders, the production of goods and services supporting the final products, are similarly procured from beyond our borders.

    Due to economies of scale, the per-unit production costs of enterprises that previously supported the displaced the USA producer, are often increased to the extent that affects their other customers. This further increases trade deficits detrimental effects upon their nation’s economy.

    The per-unit costs to all remaining users of our nation’s infrastructures, highways, bridges, pipes, and cables are similarly increased when USA production of goods are moved beyond our borders.

    I’m among the proponents of the improved trade policy described in Wikipedia’s “Import Certificates” article.
    Refer to:

Leave a Reply

Your email address will not be published.