How Big is the Employment Gap?

When trolling around blogs, I've seen some discussion about the size of the employment "gap". Put simply this is the difference between where we would be in total employment in a healthy labor market relative to our labor force compared to where we are now. The old rule of thumb that is used is that we have to create about 150,000 jobs a month just to break even on this proposition. Given that we were seeing employment contracting virtually non-stop from the beginning of 2008 until the end of 2009 by a total of nearly 8 million and have grown by less than 100,000 a month on average since then, this gap is huge.

However, there are some complications in arriving at it. One is that the proportion of the adult population involved in the labor force is not a constant even beyond the swings related to economic activity.

Excluding cyclical activity, the employment to population ratio clearly was in an up trend from the early 1970s to the late 1990s as more women entered the labor force. However, it is possible that this has started to slide into reverse, but it is hard to tell at the moment given the enormous magnitude of this recession. If the employment to population ratio is actually in decline, this somewhat lowers the threshold of necessary job creation. Let's say for the sake of argument that the average of the employment to population ratio from March of 2001 (the start of the last recession) to December 2007 (the start of this recession) is a decent proxy for a healthy participation rate. That works out to be 62.79% or so. If we take that percentage and multiply it by the adult non-institutional population, we get something like this:

This is household survey data rather than the headline establishment survey data we generally see in the press. It's harder to turn establishment survey data into these sorts of numbers, but that doesn't mean that I won't try. In any case, by this measure, the employment gap is approximately 10,712,000. This gap gets worse by about 122k a month with no job creation. 

Just for the sake of comparison, let's do something one should never do and compare establishment survey data to this and mix some data sets. This is a big no-no, but it's to try to put the headline number of +151k in context. 

By the way, the reason that the trend has kinks in it is that the non-institutional civilian population number tends to be adjusted periodically and that leads to these one-time adjustments like that. In any case, by this measure we are 11.4 million below where we need to be. What's interesting is that we barely got where we needed to be in the last expansion because it was quite weak. One could argue that outside of the housing bubble and its associated positive effects including booms in the construction and financial services industries, we didn't have much of an expansion at all between 2002 and 2007. Growth was sub-par as it was and Lord knows where we would have been in absence of the housing bubble. Going forward, it appears that the number of new jobs we need in the establishment survey to break even is around 116,000 a month.

The employment gap isn't necessary to close in order to restore growth, but until it starts narrowing a great deal you are going to have substantial downward pressures on wages. The prospects for generalized inflation with this much slack in the labor market are quite low.