Two recent movements in the structural unemployment discussion push the debate into the last redoubt of supply-side arguments.
The first is this very important new paper by Chicago economists Davis, Faberman and Haltiwanger, The Establishment-Level Behavior of Vacancies and Hiring (summarized by the Wall Street Journal here):
A new paper, though, suggests employers themselves are at least part of the problem. The authors — Steven Davis of Chicago Booth School of Business, R. Jason Faberman of the Philadelphia Fed and John Haltiwanger of the University of Maryland — take a deep dive into Labor Department data and come up with an estimate of what they call “recruiting intensity,” a measure of employers’ vacancy-filling efforts including advertising, screening and wage offers.
Their finding: Employers haven’t been trying as hard as they usually do. Estimates provided by Mr. Davis suggest that over the three months ending July, recruiting intensity was about 12% below the average for the seven years leading up to the recession. Their lack of effort probably accounts for about a quarter of the shortfall in the hiring rate.
Depressing as it might seem, the finding is in some ways encouraging. It suggests that the trouble with hiring might be more a “cyclical” function of low business confidence than a chronic, “structural” ailment that will last for years to come.
This is a crucial point. When Narayana Kocherlakota famously said that unemployment would be 6.5% without structural factors he was basing that number off his calibration of Robert Shimer’s 2007 paper Mismatch. Shimer’s model needs values of job vacancies which it gets as follows (p. 14 of the paper):
Since December 2000, the Bureau of Labor Statistics (BLS) has measured job vacancies using the JOLTS. This is the most reliable time series for vacancies in the U.S.. According to the BLS, “A job opening requires that 1) a specific position exists, 2) work could start within 30 days, and 3) the employer is actively recruiting from outside of the establishment to fill the position. Included are full-time, part-time, permanent, temporary, and short-term openings. Active recruiting means that the establishment is engaged in current efforts to fill the opening, such as advertising in newspapers or on the Internet, posting help-wanted signs, accepting applications, or using similar methods.” I measure the vacancy rate as the ratio of vacancies to vacancies plus employment.
This is a binary, thin description of a job vacancy. A firm is searching or it isn’t. What Davis, Faberman and Haltiwanger point out is that a firm’s search for a job is a thick description, one that it can do with varying amounts of intensity, that the JOLTS data isn’t capturing. If firms are posting jobs but not trying very hard to fill the spots because of weak demand it can take the model Kocherlakota is relying on and turn it inside out. Data in, data out, and if the data in isn’t what we expect it to be the data that comes out ignore the massive waste of unemployment and underutilization of real resources.
As EPI pointed out in their paper there’s a reason the Beveridge crew is walking away from that argument.
Is regulatory uncertainty actually a form of structural unemployment?
Either way health care reform now joins financial reform, two of the major policy achievements of the Obama administration, as a potential driver of “structural unemployment.”
Dave Altig at macroblog says:
The “low business confidence” part sounds right, but does that make the problem “cyclical”? I’m not so sure. Let’s say an employer is reluctant to post a job opening because, just for example, the cost of the new employee potentially will expand by an amount that is unknowable until the details of healthcare reform legislation are clarified. Would you call that cyclical or structural? If “low confidence” reduces the search intensity of businesses, wouldn’t it be reasonable to describe the resulting drop-off in job openings “structural”?
Is health care to blame for businesses not hiring? Let’s ask some businesses. We talked about the Small Business Economic Trends survey of their single most important problem. I was able to get the historical data on this. I am going to plot what are potential categories of “supply side” problems vis-a-vie health care cost worries, namely the categories of taxes, regulations, labor quality, labor costs and health insurance (which I was told is “cost and availability”). I’m going to normalize this to December 2007, when the recession starts:
Here are the same problems not normalized but their actual values:
From the regulatory and policy point of view it’s same as it ever was, with new poor sales taking a massive lead over plausible health care proxies. What are other tests to figure out if the health care bill is responsible for high unemployment rather than the obvious things we see in the interest rate and aggregate employment data?
Because we don’t see any jumps in 2010 when the bill comes back from the dead, the crucial thing in uncertainty here.