Yellen on labor market (Sept 2015):
As I said, although we’re close to many participants and the median estimate of the longer-run normal rate of unemployment, at least my own judgment – and this has been true for a long time – is that there are additional margins of slack, particularly relating to very high levels of part-time involuntary employment, and labor force participation that suggests that at least to some extent the standard unemployment rate understates the degree of slack in the labor market.
“But we are getting closer. The labor market has improved. And as I’ve said in the past we don’t want to wait until we’ve fully met both of our objectives to begin the process of tightening policy given the lags in the operation of monetary policy.”
In fact, she´s a long way from meeting both objectives! No one has any doubt about the distance we are from the 2% inflation target. On the other hand, with unemployment down to 4.9%, many could assume that we´re even “overstepped” it!
The best way to look at the unemployment rate is to analyze it from the perspective of its two constituents: The employment population ratio (EPR) and the labor force participation rate (LFPR).
That´s because the unemployment rate (UR) is, by definition, equal to [1-(EPR/LFPR)]*100. Therefore, a rise in the EPR, normally associated with a robust economy, will reduce the unemployment rate. On the other hand, a rise in the LFPR, something also usually associated with a growing economy (controlling for demographic factors, that change slowly), will increase the unemployment rate.
From this perspective, even in a strong economy the rate of unemployment could be rising (a little at least) if the rise in LFPR is higher than the rise in the EPR.
The charts below make the importance of looking at the rate of unemployment together with its determinants clear.
In the “Golden 60s” and in the “roaring 90s”, we see the unemployment rate falling with rising EPR and LFPR, with the EPR rising faster than the LFPR.
Over the last 10 years, and especially since 2008, we see unemployment first jumping from the steep drop in the EPR and then monotonically falling with the fall in the LFPR together with a reasonably level EPR.
The suddenness of the fall in the EPR and coincident falling trend of the LFPR is difficult to ascribe to sudden and big demographic changes. But they are consistent with the initially gradual and then sudden drop in NGDP growth, which even turned significantly negative (a rare event indeed).
It doesn’t look like that the labor market has in some sense, improved. What is more likely is that the perverse monetary policy of the last several years has changed its nature, maybe through hysteresis effects.
That has been the outcome of the Fed´s policy framework, which Kocherlakota aptly named “gradual normalization”. That policy framework has been instrumental in providing monetary policy tightening!
To undo the hysteresis effect on the labor market the Fed has to change the policy framework. The best alternative, and one that would do the most to reverse those effects, is for the Fed to establish a higher nominal spending target. To reach it, nominal spending growth (NGDP) would be temporarily higher, providing the right incentives for an increase in both the EPR and LFPR.