Are America’s problems cyclical or structural?
Recently there has been some discussion concerning whether the economic malaise Americans are currently undergoing is structural or cyclical in nature. The former view holds that current high unemployment rate about 9.5% is here to stay whereas the latter holds that after a period of painful credit de-leveraging by America’s indebted households businesses, things will simply be back to business.
Now personally not too long ago I believed the problem was cyclical, namely that after the recession is over we should see an uptick in employment along with output and unemployment will decrease back to below 5%. But lately, some worrying signs have emerged that structural unemployment may be here to stay.
What evidence is there that the problem is likely structural rather than cyclical? The WSJ Econs blog highlighted a paper (from 2005) recently which showed that one can’t always count on an extended period of unemployment to exert sufficient pressure on inflation:
But the ECB paper found a wrinkle in that argument: the longer the duration of unemployment, the smaller the impact on inflation. In the current context, that suggests a higher share of long-term unemployment means less of a damping effect on inflation, especially in Europe, giving the ECB less of an anti-inflation cushion to work with.
John Hopkins economist Laurence Ball explains it this way in a recent NBER paper: “The key idea is that these [long-term unemployed] workers become detached from the labor market, both because they appear unattractive to employers and because they don’t search vigorously for jobs. Consequently, while a high level of short-term unemployment puts downward pressure on wage inflation, a high level of long-term unemployment does not.”
How long is long term unemployment? One of Krugman’s recent posts shows a graph which compares current median unemployment duration of unemployment with historical trends and it’s not a pretty sight:
This finding, if true conflicts with what economists have long observed with what is known as the Philips curve, that unemployment and inflation are inversely related. It shows that the basic idea of inverse correlation requires some qualification that only short term unemployment affects inflation. Building on this idea, Tim Duy of Fed Watch blog wrote a somewhat alarming post quoting several influential economists such as Krugman, Mankiw and Delong that this finding may imply a rising NAIRU, or a natural unemployment rate. Simply put, if inflation returns to normal (about 2% and let’s not even talk about hyperinflation), while high unemployment still persists, we have evidence that the natural unemployment rate has shifted.(The natural unemployment is simply the unemployment rate for which attempts to reduce through policy will result in undesirable consequences such as rising inflation.) This idea has also recently been gaining steam amongst other analysts and economists. Just to note, Krugman has a post today on this idea.
Is it likely that the US is experiencing this at present? Whilst I cannot say for sure, it may be worth examining the experiences of other countries. For much of the late 1990s and mid 2000’s before the current recession hit Europe, European countries were already experiencing high single-digit unemployment rates. Take Germany, Italy, France, Greece, Spain for example. Why is this the case? While I can’t deny that unemployment benefits might have played a role in causing extended unemployment, they do not tell the full story because no unemployment benefits would last a decade, whereas high unemployment has been observed longer than that. Clearly some structural factor(s) must be responsible.
In support of this, Olivier Blanchard, current chief economist of the IMF concluded in a 2006 paper on the causes of long term unemployment in the EU was due to a rise in natural rate of unemployment:
How much of the increase reﬂects an increase in the natural rate, and how much reﬂects an increase of the actual rate over the natural rate? The answer to that question is relatively straightforward: since 2000, EU15 inﬂation has been indeed roughly constant – around 2% using the CPI index. If we take a stable inﬂation rate to be an indication that unemployment is roughly at the natural rate, this suggests that, today, the EU15 actual unemployment rate is close to the natural rate. It follows that the increase in the actual unemployment rate since 1970 reﬂects, for the most part, an increase in the natural rate.
So far from being an alien concept that the natural rate can and does shift, policy makers ought to be less complacent that the problem may be structural rather than cyclical in nature. Additional evidence from Japan’s lost decade also support the view that one can’t always attribute long-standing episodes of deflation to persistent unemployment:
After the burst of the bubble boom in 1990, the era of “a lost decade” began, in which the growth rate was nearly zero and in some years below zero. In 1997, Japan was on the verge of a great depression, caused by financial crisis that led to a series of financial institutions’ bankruptcies, including a big city bank and one of the three biggest brokerage houses. Quite naturally, many Japanese were seriously afraid of losing jobs.
In 2002 and 2003 the labor market deteriorated further, recording the highest jobless rate after the war. How high was the worst jobless rate? It was 5.5%. Now reflecting a little bit the improving economy, it was 4.6% in June 2004.
Taken together, does this imply that the empirically observed trade-off between unemployment and inflation is invalid? Of course not, because it’s also possible for the Philips curve to shift. But determining exactly why this is the case is more of a challenge. Part of the explanation may lie with the fact that corporations are hoarding more and more of their profits instead of hiring workers; see here for example for evidence of a trend of increasing corporate savings. The Economist also ran an article on Jul 1st highlighting this issue.
But such an explanation should also take into account that this is a long term trend in the making, rather than a recent newfound thrift which developed in the wake of the great recession. Corporations have been increasingly hoarding cash and liquid assets since the early 1980s. It is hardly a new phenomenon, as Yves Smith explained here:
Unbeknownst to most commentators, corporations in the US and many advanced economies have been underinvesting for some time.
The normal state of affairs is for households to save for large purchases, retirement and emergencies, and for businesses to tap those savings via borrowings or equity investments to help fund the expansion of their businesses.
But many economies have abandoned that pattern. For instance, IMF and World Bank studies found a reduced reinvestment rate of profits in many Asian nations following the 1998 crisis. Similarly, a 2005 JPMorgan report noted with concern that since 2002, US corporations on average ran a net financial surplus of 1.7 percent of GDP, which contrasted with an average deficit of 1.2 percent of GDP for the preceding forty years. Companies as a whole historically ran fiscal surpluses, meaning in aggregate they saved rather than expanded, in economic downturns, not expansion phases.
The big culprit in America is that public companies are obsessed with quarterly earnings. Investing in future growth often reduces profits short term. The enterprise has to spend money, say on additional staff or extra marketing, before any new revenues come in the door. And for bolder initiatives like developing new products, the up front costs can be considerable (marketing research, product design, prototype development, legal expenses associated with patents, lining up contractors). Thus a fall in business investment short circuits a major driver of growth in capitalist economies.
If I may be so bold as to speculate, perhaps such increased hoarding instead of investment since the 1980s may help explain why growth has paled in comparison to similar time periods before that:
Take the United States, which wasn’t damaged in the war. Take per capita real GDP. Give hostages by taking data from 1950 to 1980, which means including the 1980 recession, but stopping at 2007, so that the current slump isn’t included. Then here’s what you get:
Growth in per capita real GDP from 1950 to 1980: 2.2 percent per year
Growth in per capita real GDP from 1980 to 2007: 2.0 percent per year
Oh, and if we look at real median family income instead, we get:
Growth from 1950 to 1980: 2.3 percent per year
Growth from 1980 to 2007: 0.7 percent per year
Sorry: there’s no measure I can think of by which the U.S. economy has done better since 1980 than it did over an equivalent time span before 1980. It may be something you’ve heard, it may be something you’d like to believe, but it just didn’t happen.
In other words, if it’s a structural problem America faces, it may be a long term problem, one which the recent recession did much to expose. Just as the financial crisis showed the world the evils of deregulating obscure derivatives finance, despite the fact that such derivatives were in existence since the 1990s. And unless this problem is addressed, workers should settle in for the long haul.