I think what we’re seeing is that the NAIRU (non-accelerating inflation rate of unemployment) model continues to be correct in some cases but fall apart in othe

I think what we’re seeing is that the NAIRU (non-accelerating inflation rate of unemployment) model continues to be correct in some cases but fall apart in others. This model has traditionally been based more on empirical data than a strong theoretical basis (although anybody could tell you that is reasonable for low unemployment to lead to wage inflation) and, most notably, is an aggregate measure. Certainly you can argue that Japan proves an exception to this rule, or that the current low unemployment is simply a lagging indicator of our previously dovish monetary policy that will eventually normalize. But I think that NAIRU simply needs to be refined: wage inflation occurs not during low unemployment but during times of increasing employment. That is to say, an economy may arrive at a steady state of employment at a variety of employment rates, which needn’t cause wage inflation, but when employers in aggregate are creating new jobs and attempting to fill them even if they have to raise compensation, you do get wage inflation. This also doesn’t necessarily need to be done in aggregate: you can argue that wage inflation in the technology industry during the 2010s is partially what created the localized inflation in the Bay Area during that time, while in the rest of the country without heavy tech presence, this didn’t happen. There are of course many other wrinkles: for example, the size of the workforce has a degree of elasticity, where higher wages or lower hiring standards may draw some people not considering participating to again participate. There is also elasticity to hiring: depending on any individual employer’s demand for workers they may be more or less willing to raise compensation vs only partially meeting hiring goals. And of course there is stickiness to employment on both the employer and employed. The elasticity and stickiness of these things seems like it can have tons of factors that influence their magnitude: from culture norms (like in Japan where job hopping is less common), to employment regulations (how hard is it to hire-fire), ageism and other *isms (if a culture discriminates against X group, they may be more inclined to drop out of the labor force, but happy to rejoin it if employers need to hire so badly that discrimination gets lessened), to the size of the social safety net. Plus, unemployment is a very coarse metric for skilled workers. Skilled workers are less likely to simply become “unemployed” and more likely to switch directly from one job to the other. During a broad hiring binge you could see large wage growth with little change in unemployment. Anyway, I think basically economics has the tools to explain everything we are seeing already - it’s only perplexing when reasoning based on very simplified models like NAIRU. https://en.m.wikipedia.org/wiki/NAIRU https://en.m.wikipedia.org/wiki/NAIRU