The Low-Skill Losers

This post was first published on the Economic Equality blog here.

As we have noted, here and here, the Fed is devoting increasing analytical – if not yet policy-maker – attention to the unequalizing impact of unconventional policy.  It’s a start – a major problem besetting central banks in countries without a robust middle class – i.e., the U.S. – is that old-school representative-agent thinking leads to unanticipated, unequal outcomes when wealth and income are disproportionately enjoyed by the very few, very rich.  It is for this reason that the Fed’s touted employment benefit and “robust” economy in the wake of post-crisis policy has done so little for so many who remain so angry.  A new Fed paper helps to show why.

The “representative-agent” model does not have anything to do with who’s selling your house.  It means that econometric assumptions, including the theories on which monetary policy is premised, assume everyone in an economy is pretty much the same as everyone else.  When the majority of the population is within large bands of similar income, wealth, and propensity to consume, then representative-agent assumptions work as a good guide to how an economy will respond to changing interest rates or other actions.  Most Fed thinking is based on representative-agent models, which is why much hasn’t worked anywhere near as well as the Fed hoped.  The U.S. is now what economists call “heterogeneous” that is, small groups of people have a huge impact on the economy because income and wealth are so unevenly distributed and marginal propensity to consume has shifted to a few people who don’t want all that much of what’s needed to stoke output growth.

Correctly recognizing all this, the new Fed staff paper goes down the heterogeneous route to see what monetary policy does to those who need growth the most.  It’s very, very econometric and model-driven, but still a helpful neo-Keynesian foray.  Instead of looking at American workers as one great big group, the analysis breaks them down from 1985-2006 by educational level (an increasingly helpful, if distressing, equality indicator).

Looked at this way, one finds that conventional monetary-policy easing reduces employment inequality – i.e., more people are working – but wage inequality gets still worse for those who can afford it the least.  Monetary-policy contraction also reduces employment among lower-skilled workers, but does not do so for more highly-educated ones, exacerbating income inequality.

Why so unequal an impact?  The paper is highly theoretical despite all its data, but postulates that more highly-educated employees are more mobile and thus less sensitive to monetary-policy shock.  We would add that, when the majority of the population is mid-to-low skilled, then monetary policy is at best a blunt-force instrument because most workers have difficulty finding full-time employment that supports consumption above the hand-to-mouth level that is similarly unresponsive to policy stimulus or constraint.

In short, this paper shows that conventional policy doesn’t work by the book when it comes to low-skilled workers.  We would add that, because conventional policy doesn’t work for workers, it also hasn’t worked for pretty much everyone else.  Is this also true for post-crisis unconventional policy?

Sadly, yes.  Unconventional policy actually tracks convention by relying in part on interest rates, albeit by driving them down to ultra-low levels in hopes of ensuring ultra-fast recovery.  The employment-mobility issues described in this paper apply at least as much in an unconventional take on conventional interest-rate changes because employment becomes still harder for low-skilled workers.  The labor-participation data shows this all too clearly.

Of course, unconventional policy also counts on a huge Fed portfolio. As we’ve shown, the portfolio largely affects wealth equality – or inequality given its clear impact boosting financial-asset valuations at disproportionate benefit to the upper stratum of the U.S. wealth distribution.  The Fed has made the same representative-agent mistake with its portfolio that it did with interest rates: when wealth was relatively evenly distributed and held in more or less the same assets, then market stimuli transmit across the economy.  But, when wealth is held heterogeneously – that is by a few people in assets most others don’t have – then market stimuli do not transmit in ways that result in increasing physical-infrastructure spending or the other economic investments that boost sustained employment growth for low-skilled and even middle-class workers.  And, so we’ve seen since 2010.

As a result, Fed policy hasn’t worked.  The Fed knows this all too well – it’s studying hard and holding conferences and even allowing its researchers to spend the time it takes to provide studies such as this happily-heterogeneous one.  There’s still, though, no sign that policy-makers are doing more than modeling new “neutral rates,” playing with “inflation targets,” or commissioning doctoral dissertations on other modifications of current policy into the old batch of representative-agent assumptions.  Unless the models change, monetary policy won’t and most will be the poorer for it.