Holy S___, Batman, Look at the Foreign Exchange Market!

In reaction to my last post about the s___ that may soon hit the fan and why labor cost targeting would help, Gabriel M. suggests that I discuss the nature of the s___ in question and how it came about. In a nutshell, the issue is that most economists (not all, but I would say the vast majority) think the dollar is significantly overvalued. For the moment, the market seems to disagree, inasmuch as interest rates on dollar-denominated bonds are not much higher than those on bonds denominated in, for example, euros. In theory, if people believed the consensus of economists, it would be a no-brainer for anyone with an international bond portfolio to dump their dollar assets and replace them with other assets such as euro bonds. (It’s implausible to me that the slight difference in nominal creditworthiness between the US treasury and, say, the German treasury, is enough to make this less of a no-brainer.) If everyone were doing that, dollar interest rates would already be significantly higher than non-dollar interest rates.

I don’t think I’m going too far out on a limb to say that, historically, when economists and the market disagree, the economists usually eventually turn out to be right (though the economists are usually wrong about how long it takes to get to “eventually”). A case in point is the last time the dollar was regarded as severely overvalued, in 1984. It may have taken an international agreement (the 1985 Plaza Accord) to get the dollar on a downward path, but once the decline started, it was beyond what international policymakers seemed able to control. Anyone who listened to economists in 1984 about the dollar, and then changed his mind when they turned out at first to be wrong, would have been extremely unhappy by late 1987.

I should note in passing that there were factors present in the mid-1980s that mitigated, and even reversed, the supply shock from the falling dollar. In 1985, when the dollar began to decline, the US was still recovering from worst recession since the 1930s, so there was considerable slack in the economy. (There may be considerable slack today, too, but that’s not the consensus.) Also, as a consequence of having shown itself willing to induce and prolong the worst recession since the 1930s, the Fed had a surplus of inflation-fighting credibility. And since the dollar had been continuing to rise until early 1985, there was something of a favorable import price shock already in the pipeline to offset the subsequent unfavorable one. Then in 1986, OPEC members were unable to reach agreement on new quotas, and the price of oil dropped dramatically, providing a favorable supply shock to (probably more than) offset the unfavorable shock from the falling dollar. Another factor was that many goods sold to Americans were (either explicitly or implicitly) priced in dollars, so the import price shock from the weak dollar was not as strong as it otherwise would have been. The same is true today, but less so.

Part of the reason that the dollar remains overvalued is that much of the investment in fixed income dollar assets comes from sovereign entities, such as the People’s Bank of China (PBoC) or the Saudi Arabian Monetary Authority (SAMA), that are less concerned about profit and loss than private investors. But even such entities are not entirely oblivious to profit and loss, and lately there are increasing signs of their desire to diversify away from the dollar. In doing so, they would also probably have to give up the currency pegs that have kept the dollar overvalued relative to their own currencies. SAMA gave the world a bit of a shock recently, when it uncharacteristically failed to echo the Fed’s interest rate cut. From China there are vague noises about the “nuclear option” of divesting of US bonds, which would entail dropping the dollar peg entirely. And China’s rising inflation rate is, one may presume, making it clearer to the Chinese authorities that continuing the peg in its current form is not in their national interest. So there are obvious cracks developing in the structure that has supported the overvalued dollar.

Another factor is the low national savings rate in the US, evident in both the negative personal savings rate and the federal fiscal deficit. With Americans not saving, the nation as a whole needs to attract capital from abroad, and that demand tends to keep interest rates high enough (relative to foreign rates) to keep the dollar from weakening too much. (My Keynesian self is telling me that is a big oversimplification, but I don’t want this post to get too long.) The risks here are (1) that (because of liquidity constraints due to falling home prices, or because of demographics, or just because of an attack of prudence) the savings rate could rise, taking away this support for the dollar, and (2) that, if the savings rate does not rise, foreign investors will lose confidence in the creditworthiness of the US and dump dollar assets until the dollar weakens or US interest rates rise.

The low savings rate in the US is necessarily offset by an excess of savings over investment abroad – the so-called “savings glut.” There’s a bit of a chicken-egg debate about which of these is the first cause, but from the point of view of the exchange rate, it doesn’t much matter. If the savings glut is the primary cause (a view to which I’m sympathetic), then the glut itself is arguably what has supported the dollar. However, the scenarios described in the previous paragraph still apply: (1) if Americans become no longer willing to absorb the excess savings, then the dollar will drop; (2) if international savers begin to think that the US is not a good place for their savings, then the dollar will drop. (The first of these scenarios – which I will call the “ice” scenario – is particularly troubling because, while it could be an inflationary shock for the US, it could at the same time be a particularly strong deflationary/recessionary shock for the rest of the world.)

So, whether or not it happens immediately, chances are there will be downward pressure on the dollar in near future, and we cannot assume that the market’s response will be orderly. If the dollar drops quickly, the price of imports would likely rise much faster than the US economy can adjust by shifting demand and production to domestic producers. That’s a classic example of an import price shock, and it would likely cause a rise in aggregate US consumer prices out of proportion with its relatively mild stimulative initial effect on US output (though the latter should increase over time). The shock will likely be exacerbated by rising commodity prices (including oil), as demand by stronger-currency countries pushes up the dollar price of commodities. (It’s debatable whether the last point has any real substance: most commodities are priced in dollars by convention, and their price in terms of some hypothetical “average” currency should not be affected by exchange rates. In theory, rising dollar commodity prices are just part of the original exchange rate shock, but when oil goes to $100, it will certainly get separate coverage in the media.)


Unfortunately, the prospects for anything even vaguely resembling explicit labor cost targeting are dismal at best. So I have left two great hopes for avoiding an unpleasant outcome for the US. First, that the rise in import prices will continue to be slow enough to avoid having much inflationary impact. Second, that there is more slack in the US labor market than the consensus recognizes, and this slack will absorb much of the price shock. Even if my hopes are realized, the outcome will be less than optimal, but then, when do we ever get an optimal outcome?