The percentage of small businesses with at least one unfilled job opening is back near pre-recession levels. In addition, some 55% percent say they hired or tried to hire new workers, but 48% report having few or no qualified applicants.
My friend David Rosenberg made the same point last weekend when we were together in Miami, and he wrote about it again today. He thinks that the unemployment rate, rather than underestimating the amount of slack in the labor market, is actually underestimating how tight conditions are. He refers to the Beveridge curve, which is a squiggly looking timeline graph that depicts the relationship between job openings and unemployment. To most people it’s just a confusing doodle, but the point you should take away is this: The employment trendline for 2000–2009 has shifted dramatically during the years since. I find this phenomenon fascinating – just another conundrum to mull over as we think about where the labor market is going. Quoting from Rosie’s piece:
I mean, this inability of firms to fill positions from the current supply of available labor speaks to the structural issues in the labor market rather than anything cyclical.
To further this point, let’s turn to the Beveridge Curve, which plots out the relationship between job openings and unemployment – you can see that there has been a distinct outward shift in the curve in this cycle from where it was before 2009.
What this means is there is a higher unemployment rate for any given level of the job openings rate – for example, based on the prerecession relationship, the 3.7% job openings rate in September (job openings as a share of openings and payroll employment) would be consistent with a 3.0% unemployment rate, not the prevailing 5.0% rate.
This type of change in the relationship is indicative of an inefficient jobs market since the demand for labor by companies is unable to be met by the available supply, and this typically reflects structural issues on the supply side (such as skills or geographic mismatches) rather than a cyclical drop in demand for workers – again, this survey corroborates this. So in this sense, the unemployment rate is actually overstating the degree of slack that exists in the labor market.
This has two main implications. First, since the issues in the labor market are seemingly structural rather than cyclical, there is no amount of monetary policy stimulus that can reasonably be expected to fix this – this is something that needs to be addressed through skills training and related government policy; using monetary policy is a remedy means that rates will be too low for too long and result in credible risk to price stability. (Emphasis mine)
Secondly, this suggests that labor market conditions are in fact increasingly tight and that the nascent uptick that we have seen in wage rates is likely to be sustained as firms looking for qualified labor [are] forced to push wages higher to lure workers.
He has a point – up to a point – but my observation is that firms are only going to have to increase wages for jobs that demand real skills, as opposed to those that merely require warm bodies to show up for work. How, in an economy with almost 8 million unemployed workers, can almost half of our supposedly job-creating small businesses be unable to hire the workers they need?
The answer is that we have a skills gap. The workers who are looking for jobs aren’t qualified for the available job openings. Which suggests that the labor market issue is not about monetary policy. It’s an education and training issue.