Suppose you had, back in 1992 or 2004 or indeed any other time since 1990 and before 2013, asked a question of Charlie Evans—or, indeed, any of the other non-Neanderthal participants in the Federal Reserve's Federal Open Market Committee meetings. Suppose you had asked whether a 25-54 employment-to-population ratio in the United States of today's 78.5% was anywhere near "full employment". What would they have said?
They would have said "of course not!"
They would have observed that the post-baby boom decline in fertility, wage stagnation among male earners, and the coming of feminism had greatly increased the share of 25-54 year old women who wanted paying jobs outside the home.
They would have pointed out that upward pressure on core inflation at an 80% 25-54 employment-to-population ratio in 1990 was small, that upward pressure on core inflation at an 82.5% 25-54 employment-to-population ratio in 2000 was minimal, and that upward pressure on core inflation at an 80% 25-54 employment-to-population ratio in 2007 was minimal.
They would have pointed out that today's 78.5% was between between the 78.1% 1992 recession trough and the 78.6% 2003 recession trough.
They would have concluded that, by the standards of the post-feminist revolution era, a labor market with a 25-54 employment-to-population ratio of 78.5% was a labor market as bad from a business cycle perspective as the labor market got during the Great Moderation.
So why is Charlie Evans now saying that today the United States has "essentially returned to full employment"? Why no qualifiers? Why no "if you look only at the unemployment rate, and put the shockingly-low labor force participation statistics to one side...? Why no "it may well be the case that the U.S. has essentially returned to full employment? Why this certainty on the part of even the non-Neanderthal members of the FOMC—in public, at least—that the unemployment rate is the sole guide?
And why the puzzlement at the failure of core inflation to rise to 2%? That is a puzzle only if you assume that you know with certainty that the unemployment rate is the right variable to put on the right hand side of the Phillips Curve. If you say that the right variable is equal to some combination with weight λ on prime-age employment-to-population and weight 1-λ on the unemployment rate, then there is no puzzle—there is simply information about what the current value of λ is:
Charles Evans: Lessons Learned and Challenges Ahead: "These policies... produced results. Unemployment began to fall... https://www.chicagofed.org/publications/speeches/2017/05-25-lessons-learned-and-challenges-ahead-bank-of-japan
...more quickly than anticipated in 2013.... We were able to scale back the QE3 purchases.... Today, we have essentially returned to full employment in the U.S.
Unfortunately, low inflation has been more stubborn, being slower to return to our objective. From 2009 to the present, core PCE inflation, which strips out the volatile food and energy components, has underrun 2% and often by substantial amounts. This is eight full years below target. This is a serious policy outcome miss...
Charlie says a lot of good things in his talk. His discussions of "outcome-based policies... symmetric inflation target[s]... [and] risk management" are wise. But wisdom can be usefully applied only if you know where you start from. And we start from a position in which we really do not know how close the U.S. economy is right now to "full employment"—how much headroom for catch-up growth and catch-up employment remains, and how powerful and useful more aggressive policies to stimulate spending would be right now.
In 25 years my students are likely to ask me why the FOMC was so certain the U.S. was "essentially at full employment" today. What will I tell them?