Free exchange | Imbalances

The American periphery

How do American regions adjust?

By R.A. | WASHINGTON

IN LOOKING at Jay Shambaugh's paper on the euro crisis, it seems fairly clear why America doesn't suffer from internal balance-of-payments crises in the way that the euro zone has: the truly national banking system (and regulatory apparatus) breaks the link between local bank troubles and local government-debt crises. As one speaker quipped in discussion on Friday, if Delaware were forced to stand behind the debts of the banks headquartered in Delaware, then the state of Delaware would long ago have been bailed out by the federal govermment (or would have defaulted, or would have been subjected to torturous austerity in a hopeless effort to cut its way out of insolvency before ultimately defaulting anyway).

But just because American states are less vulnerable to joint debt-banking crises doesn't mean that they never develop persistent imbalances. Indeed, it is almost certain that there are metropolitan areas or regions across America that are persistent net exporters with counterparts that are persistent net importers. Which makes one of Mr Shambaugh's points—that internal devaluation has become almost unheard of since the achievement of low and stable inflation in the 1990s—quite interesting:

Shambaugh (2012) uses price data for 27 U.S. metro areas from 1961 to 2010 (not all regions are available at the beginning of the sample) to see if metro areas can have falling prices relative to the rest of the U.S. currency union. Using the same standards for an internal devaluation, but comparing each metro area to the median inflation rate for the nation, Table 1 shows that in the U.S., internal devaluations do take place prior to 1991, albeit rarely. With 2 exceptions, though, they do not happen at all after the U.S. moved to a lower inflation period post 1990. U.S. inflation averaged over 5% from 1968 to 1990 but averaged 2.5% from 1991 to 2010. There were also no internal devaluations in the period 1961-8 when inflation averaged just 1.7%. Just as nominal devaluations may be a path of least resistance, labor mobility in the U.S. may take place before internal devaluation is needed.

The last sentence is particularly interesting. In most international cases, flexible currencies resolve imbalances relatively quickly, so a grinding process of internal devaluation isn't necessary. Mr Shambaugh suggests that within countries, higher levels of geographic mobility may have a similar effect. When wage growth in one region outstrips productivity growth for a long period, leaving an economy with an uncompetitive export sector, job loss should be the result. That job loss should—slowly and painfully—lead to falling wages, until competitiveness is restored. In America, however, underemployed households may simply pack up and move, say from Ohio to Texas.

What's curious about this, however, is that one might then expect to see an increase in geographic mobility from 1991, as households compensate for the increased difficulty of regional adjustment by moving more often. Instead:

By most measures, internal migration in the United States is at a 30-year low. Migration rates have fallen for most distances, demographic and socioeconomic groups, and geographic areas. The widespread nature of the decrease suggests that the drop in mobility is not related to demographics, income, employment, labor-force participation, or homeownership. Moreover, three consecutive decades of declining migration rates is historically unprecedented in the available data series. The downward trend appears to have begun around the 1980s, pointing to explanations that should be relevant to the entire period, rather than specific to the current recession and recovery—that is, the decline in migration is not a particular feature of the past five years, but has been relatively steady since the 1980s.

The above is from a recent paper by Raven Molloy, Christopher Smith, and Abagail Wozniak. They speculate that the decline in mobility could be due, in part, to a return to normal after a wave of rapid, southward, postwar migration associated with catch-up growth in the American South. They're disinclined to attribute the shift to homeownership, since mobility fell among nearly every sub-category, including among renters.

I take their point, but migration rates are lower among owners than renters (1.5% versus 5.9% in the 1980s, 0.9% versus 3.5% in the 2000s). Meanwhile, America's homeownership rate rose from under 64% in 1984 to over 69% in 2005. That change explains at least some of the country's falling mobility. And rising homeownership would seem to be at least partially due to the dramatically lower interest rates that were the norm after Paul Volcker's victory over inflation in the early 1980s. It's possible that one little macroeconomic subplot of the past three decades is that the Fed's achievement of low and stable inflation make cross-regional wage adjustment more difficult, while simultaneously encouraging homeownership—and geographic immobility.

Discover more

Religious competition was to blame for Europe’s witch hunts

Many children are still persecuted as alleged witches in Africa for similar reasons

Has BRICS lived up to expectations?

The bloc of big emerging economies is surprisingly good at keeping its promises


How to interpret a market plunge

Whether a sudden sharp decline in asset prices amounts to a meaningless blip or something more depends on mass psychology