First off, John Hussman is still bearish.
OK, now that we’ve gotten that out of the way, his latest weekly letter takes a look at some of the supposedly “good” jobs data we’ve gotten lately — both on the monthly jobs number, and with the weekly initial claims reports.
In his note, he introduces readers to the notion of an impulse response — basically the tendency of economic data to echo down the road:
To provide some perspective on this, below is a simple estimate of what economists call an “impulse response” profile for the U.S. labour market. When we deal with economic variables – such as employment – that are subject to positive or negative “shocks,” it is often helpful to estimate how those shocks tend to “propagate” over time. For employment, a 1% shock in job creation or destruction (versus trend growth) tends to be followed over the following year by an additional 1% movement in jobs in the same direction. After that, the impulse gradually attenuates over a larger period of years, as the initial positive or negative burst is followed by a trajectory back toward trend growth. In effect, positive and negative “shocks” to job creation have very strong tendency to “cluster,” propagating in the same direction for a period of about 12 months, and then gradually attenuating toward the long-term trend.
Here’s what the echoes look like for a job shortfall:
So as you can see, the impact of a 1% jobs shortfall initially would be expected to lead to more job shortfalls, but then the impact fades.
Now, what does that have to do with the current employment situation? Everything.
With respect to the employment situation, given the massive job losses we observed in 2008 and 2009, we are already past the point where the impulse response curve should propagate additional job losses. Rather, in a normal post-war recovery, the normal impulse response profile suggests that we ought to be observing rapid employment gains on the order of 460,000 to 500,000 jobs a month.
Based on typical impulse response, very robust job growth would normally have been expected following the massive job losses of 2008 and 2009. From this perspective, the past three employment reports have not simply been bad – they have been among the worst job creation shortfalls on record. While not every shortfall results in a fresh wave of propagating job losses, we are observing this shortfall in the context of leading economic indicators that have already turned down clearly.