In prepared remarks today, Brainard clarified the meaning of maximum employment as a precondition for liftoff. Brainard’s appearance was bookended by a couple of appearances by Clarida.
Chair Powell also spoke today, appearing for a second day for the semiannual monetary policy testimony, this time before the House Financial Services Committee. We covered yesterday’s appearance in a note. Today, Powell’s testimony adhered to the same policy stance. In particular, he pushed back gently against several questions that cited his comment that 2021 GDP growth “could be” 6%. He is not willing to commit to 6%–or a rate “in that range” or any other rate–as his baseline. The market also took note of his remark that reaching 2% inflation “may take more than three years, but, you know, we’ll update that.” He said that in response to a question about the forecast horizon of the SEP (which is “arbitrary”).
As we thought might be the case, the appearances by Clarida and (especially) Brainard were more intriguing.
While Clarida did not go into the topic in as much detail as Brainard, both covered the key part of the definition of maximum employment. That definition is in line with my interpretation since the new monetary policy framework was announced: Ultimately, maximum employment is about inflation, not a broad range of labor market indicators. That means that the maximum employment and price stability objectives are consistent and that the preconditions for exit are about realized and expected inflation, and only that.
When I saw the title of Brainard’s speech, “How Should We Think about Full Employment in the Federal Reserve’s Dual Mandate,” I was a bit disappointed that she did not use the term “maximum employment” rather than “full employment.” Are full employment and maximum employment synonyms? That’s a good place to start.
Full Employment and Maximum Employment
Maximum employment has always been the term for the employment mandate in the Federal Reserve Act. As such, it had no empirically operational or theoretically well-specified definition. Rather, I preferred to define the employment mandate as full
employment. Of course, that isn’t well defined either, which leaves the same problem. For the purpose of monetary policy, and to ensure an empirically operationally and theoretically relevant measure, I define full employment as maximum sustainable employment and, in terms of the unemployment rate, the minimum sustainable unemployment rate. “Sustainable”–that’s the key word. That meant without upward pressure on inflation. So lurking behind maximum employment has always been inflation. Indeed, this must be the case to ensure that maximum employment and price stability hold at the same time.
Maximum Employment and Social Welfare
Still, the refocus on maximum employment as the employment mandate is very important, as can be seen in the evolution of how FOMC participants talk about the unemployment rate as an objective. It seems as though participants–as though by a decree from above–are not allowed to say “unemployment rate”—that is, the U-3 unemployment rate—without explicitly mentioning the wide disparities between unemployment rates for different demographic groups. So maximum employment carries this spirit.
A Broad and Inclusive Goal
To be sure, Brainard also says that U-3 is too narrow a measure to assess labor market conditions.
She is apparently in search of a set of indicators that are collectively relevant to maximum employment. She first discusses U-3 and the labor force participation rate. She notes key limitations of the U-3 rate, which does not count job losers that leave the labor force. She notes that the employment-to-population does not have this problem, which makes it “a strong indicator of the extensive margin in the labor market.”
Having covered “extensive” indicators, she says, “there is also important information in the intensive margin.” For this, the indicator she points to is those employed part-time for economic reasons. That cohort is one of several additional that are added to the number of unemployed to generate the U-6 rate, BLS’s “most expansive” measure of labor underutilization.
But she is not done. Perhaps most revealing is that she adds disaggregated measures of the unemployment rate and other indicators, especially those related to different groups. For example, taking into account minority unemployment rates is an important part of the spirit of maximum employment, especially given that gaps in unemployment rates for minorities have tended to narrow later in expansions, when aggregate unemployment rates are very low.
She also delves into how, in this pandemic episode, there is special reason to look at changes in labor market indicators across genders. That’s partly because “A decline in participation by prime-age women is an important contributor to the overall participation decline [and] Some portion of the decline reflects the increase in caregiving work at home with the shutdown of schools and daycare due to COVID-19.”
Then, for good measure, she adds quit rates and the ECI.
But we do not have a collective and empirical measure of maximum employment. How do we identify when we are at maximum employment, that is, have met the employment precondition for liftoff? We can’t, of course. We are just back to the problem I had with maximum employment in the Federal Reserve Act.
Maximum Employment as the New NAIRU
Not to worry. We can resolve this problem. The NAIRU, of course, is about inflation, and only inflation. In the end, so is maximum employment. This is how it is ensured that maximum employment is consistent with the price stability objective.
If the estimate of the NAIRU does not well identify when inflation will be inconsistent with the price stability objective, let’s come up with a term that does: maximum employment, the highest sustainable level of employment, or, again, the minimum sustainable level of the unemployment rate, the U-3 measure. We can’t measure it directly, but we know it when we see it. We see it in realized inflation, and, in a forward looking sense, with respect to lift off preconditions, in inflation being projected to overshoot!
Brainard eventually does get to defining maximum employment as something that can be measured and that is indeed all about inflation. Clarida has been in the same place. That was my view from the very beginning in my discussion of the new monetary policy framework.
Our new monetary policy framework recognizes that removing accommodation preemptively as headline unemployment reaches low levels in anticipation of inflationary pressures that may not materialize may result in an unwarranted loss of opportunity for many Americans…the shortfalls approach means that the labor market will be able to continue to improve absent high inflationary pressures or an unmooring of inflation expectations to the upside. (Brainard, emphasis ours)
A low unemployment rate, in and of itself, will not be sufficient to trigger a tightening of monetary policy absent any evidence from other indicators that inflation is at risk of moving above mandate-consistent levels. (Clarida, emphasis ours)
The NAIRU is admittedly an unreliable estimate of the level at which inflation is at risk of moving above mandated-consistent levels. Treating it as such might avoid sacrificing employment opportunities that would otherwise be created without compromising the inflation objective for exit: realized and expected inflation.
In any case, for the stance of monetary policy overall, the implication is clear: Don’t tie the conduct of monetary policy to the point estimate of the NAIRU. Keep pushing employment higher (the unemployment rate lower) until you observe realized inflation at 2% with a forecast that inflation is on track to overshoot 2%, modestly and for some time.
So liftoff is about inflation, and only inflation.