For some of us the story of the downturn and the slow recovery is pretty damn simple. The collapse of a huge housing bubble created a big gap in demand that is not easily filled. (Name your favorite demand-filling candidates. You have consumption, non-residential investment, residential investment, government spending, and net exports -- that's all folks.) Since there was no plausible story whereby the economy could quickly fill this demand gap, continuing slow growth and high unemployment is not much of a surprise.

But no one wants to make our leading economists and policy makers look silly, so there is a considerable premium placed on efforts to make the simple story complicated. For example last week we had a discussion in the NYT raising the possibility that the economy's real problem is a skills shortage. Today, Catherine Rampell raises the troubling concern that job openings are up, but hiring isn't. Her conclusion is that firms are reluctant to actually go ahead and hire because of uncertainty about the future. She then lists causes of uncertainty which are supposed to be the cause for the delay. 

Before anyone gets too concerned about the bad effects of uncertainty on the economy, let's look at the data a bit more closely. We get our data on job openings from the Job Openings and Labor Turnover Survey (JOLTS). If we look at the most recent release and go over to the right two sets of columns we find "total separations." These are the workers who leave their jobs by quitting, layoffs, or firing. The survey shows that this number was roughly 4.5 million in the most recent month.

This is important because it means that over a six-month span we would anticipate that roughly 27 million workers will leave their job. That's a bit less than 20 percent of total employment.(The actual share will be somewhat less since some jobs will come open more than once.) This is noteworthy because it would suggest that most firms need not be too troubled about uncertainty in making a hire today since they are likely to see a substantial portion of their workforce leave in the near future in any case. In other words, if they hire someone who it turns out they didn't really need, odds are that someone will leave in the near future so they will need this worker.

This is especially true in sectors like retail and restaurant employment which have a disproportionate share of the job openings. The turnover rate in retail is 4.8 percent, which means that close to 30 percent of jobs will come open in the next six month. In restaurants the turnover rate is 5.4 percent, meaning close to one-third of the jobs will come open in the next six months. This suggests that an employer is not taking much of a risk by hiring in this time of uncertainty.


The other problem with the uncertainty story is that it implies that employers are using other ways than hiring to meet the demand for labor they are seeing. But the data won't support that part of the story either. If it were true then we would be seeing more hours per worker, we aren't. The length of the average workweek in the most recent data was 34.5 hours, the same as it was in May of 2013, before we had so many job openings. In the job opening intensive retail sector the length of the average workweek has actually declined from 31.6 to 31.2 hours over the last year, although it has risen by 0.1 hour (from 26.0 to 26.1) in the leisure and hospitality sector that includes restaurants.

In any case, there is zero evidence that employers are working their existing workforce more as an alternative to hiring. Also, productivity growth has been especially weak over the last year (the first quarter numbers were negative, but that was an anomaly resulting from bad weather), so there is no evidence that employers are forcing workers to work harder as an alternative to hiring. In short, there is no reason to think that uncertainty is causing firms to look to alternatives to new hires as a way to fill labor demand. It seems that employers are doing as much hiring as we would expect given the demand for labor they are seeing.

Then why are they listing job openings? Listing is cheap. In a weak labor market there are more potential good hires out there than would be the case in a strong labor market. This means that there is likely to be a greater return to having a greater search effort. In a strong economy and strong labor market an employer might face a substantial cost for delaying a hire, since they need to fill a position and their business will suffer from having a vacancy sit open. Furthermore, there is little likelihood that if they let an acceptable worker get away that the employer will find a better worker later.

On the other hand, in today's weak economy there is little consequence from delaying a hire and it is reasonable to believe that better workers can be found by waiting. In that context, an increase in the ratio of job openings to hiring is very plausible. 

So the long and short is that the story of the downturn can be boiled down to three simple words, demand, demand, and demand. The more complicated stories are just make work projects for economists and their friends.