« Paul Krugman: The War on Warren | Main | "Global Commodity Prices Track World Demand" »

Monday, March 21, 2011

The Unemployment Problem is Mainly Cyclical, Not Structural

Cyclical and structural unemployment can be hard to tell apart. For example, suppose that a business owner would like to hire someone to operate a complicated piece of machinery, and needs someone with experience. The owner offers $10 per hour, but, unfortunately, no one applies. Interviewed by the local paper, the owner complains that qualified workers simply aren't available.

However, that is not true. There is an unemployed worker who has been running that kind of machine for 10 years. He's good at it, and only lost his job due to the fact that the place he had worked for the last 10 years shut its doors in the recession. At $15 per hour, or more, he would have taken the job. But $10 is just not enough to pay the bills and save the house, and he decides to hold out and hope that something better comes along.

So whose fault is it? Should be blame the worker for being unwilling to take a decent job due to the fact that it doesn't pay enough (perhaps unemployment compensation is helping the worker to wait for a job that will pay enough to support the household)? Should we blame the store owner for not paying enough to attract workers with families to support? Neither, the problem is lack of demand.

If times were better, i.e. demand were stronger, the business owner could afford to pay $15, and would -- problem solved. So, all that is needed is an increase in demand for the products the business sells (demand that would exist if the worker and others like him had jobs). But at current demand levels, which are depressed, it is not worth it to pay that much. The business owner would be losing money.

So is the problem cyclical or structural?  It will look like structural unemployment in the data, the owner can't find anyone who is qualified who will take the job at the wage being offered, but the heart the problem is a lack in demand.


As noted above, cyclical and structural unemployment can be hard to sort out. One way around this is to find a group of workers who should do well if the problem is structural, and see how they are faring. This Economic Letter by Bart Hobijn, Colin Gardiner, and Theodore Wiles of the SF Fed attempts to do just that. It answers the cyclical versus structural unemployment question by looking at a segment of the population that ought to be doing relatively well if the problem is structural, recent college graduates, and finds that "structural factors are of minor importance for current unemployment":

Recent College Graduates and the Labor Market, by by Bart Hobijn, Colin Gardiner, and Theodore Wiles, FRBSF Economic Letter: Although the U.S. economy is recovering from the 2007–09 recession, the labor market remains weak. The unemployment rate was 8.9% in February 2011, down more than a percentage point from its peak in 2009, but about four percentage points higher than before the recession. Some economists have concluded that this persistently high unemployment rate is due largely to structural frictions in the U.S. labor market rather than to weak demand for workers associated with the severe recession (Kocherlakota 2010). Generally, such structural frictions arise from mismatches between workers and employers. A common example of a mismatch occurs when employers are looking for skills that are different from those that available workers offer. Another type of mismatch occurs when jobs are available in geographic regions with few qualified job seekers (see Daly, Hobijn, and Valletta 2011 and Weidner and Williams 2011).
One way of testing whether such structural factors are important in the overall labor market is to examine a segment of the market that is not subject to these constraints. Recent college graduates for the most part don’t experience skill and geographic constraints because they tend to be highly educated and mobile (Pianalto 2010). Thus, if structural unemployment were the principal factor accounting for labor market weakness in this downturn, then the job market for recent college graduates would be relatively stronger than in a mainly cyclical downturn. In this Economic Letter, we analyze the extent of structural constraints on employment by comparing current trends for recent college graduates with the trends that prevailed during the recovery after the 2001 recession, a labor market downturn that was mainly cyclical in nature. We find that the labor market for recent college graduates is equally weak or even weaker than the overall market, just as in the 2001 recession and its aftermath. The weakness of the current labor market for college graduates is reflected not only in the unemployment rate for this group, but also in their part-time employment rate and earnings. This indicates that structural factors are of minor importance for current unemployment. ...[continue reading]...

Data limitations likely prevented this, but I wish they could have looked at disaggregated data to see if graduates in any subset of disciplines are faring relatively better than in 2001. Given the aggregate numbers it's unlikely that there are individual sectors that are doing substantially better than in 2001, so I don't expect this would change the picture at all, but the disaggregated data would provide useful information and rebut (or not) the charge that aggregates are hiding important information. In any case, there are many, many indications that the problem is largely cyclical, and despite the fact the legislators seem to have forgotten about the unemployed, there's still a long recovery period ahead of us. It's not too late for policy to help. [Also posted at MoneyWatch.]

    Posted by on Monday, March 21, 2011 at 12:24 PM in Economics, Unemployment | Permalink  Comments (52)


    Feed You can follow this conversation by subscribing to the comment feed for this post.