Attacks on worker wages have kept benefits from middle class

Attacks on worker wages have kept benefits from middle class
© Getty Images

For years, our work at the Economic Policy Institute (EPI) has highlighted an ever-growing wedge between growth in economy-wide productivity and the hourly pay (including benefits) of typical workers.

Simply put, productivity is the income generated in an average hour of work in the U.S. economy. Over time, it should increase, as technology advances, as workers are given more and better tools to do their jobs (for example, construction workers replacing shovels with backhoes) and as the workforce becomes more skilled and better educated.


Productivity defines the ceiling on how fast living standards can grow and provides an obvious benchmark for healthy growth in hourly pay.


The measure of hourly pay we examine in our work is the average hourly compensation of production and non-supervisory workers — a group that accounts for roughly 80 percent of the private-sector workforce. For decades following World War II, our measures of productivity and pay tracked each other in lockstep. But since 1979, productivity has risen by more than 64 percent while hourly pay rose just 11 percent.

The conclusion we draw from this analysis is that policymakers need to implement measures that not only boost overall economic growth, but also those that give typical workers the leverage and bargaining power they need to claim their fair shares of this growth.

Last week, economists Anna Stansbury and Larry Summers released a paper that many interpreted as challenging our work. This interpretation is wrong — their findings are completely consistent with ours.

We both agree that productivity grew many times faster than hourly pay for typical workers over the 36 years between 1979 and 2015. Their study, however, looks at the correlation between pay and productivity growth over much shorter chunks of time (one to five years) within this period.

Additionally, their findings look at the effect of productivity on pay while holding other factors constant. They find that, since 1979, a 1-percent gain to productivity growth over a short time-horizon boosted wages by between 0.4 and 1 percent — holding everything else equal.

But, of course, all else has not held equal since 1979. We know this is true simply because pay lagged productivity so badly over the full 36 year period they examine.

An intuitive way to understand what the Stansbury and Summers results are telling us is that productivity growth has been “trying” to raise wages in recent decades, but influences besides productivity growth have successfully suppressed wage growth and neutralized productivity’s effect.

What are these influences that have suppressed wage growth? Obvious candidates include higher average unemployment rates post-1979, an ever-shrinking share of American workers in unions, a declining inflation-adjusted value of the minimum wage, growing competition with low-wage nations and a cocktail of tax and regulatory changes that have empowered corporate managers to claim a greater share of their businesses’ income at their workers’ expense.

Stansbury and Summers emphasize the finding that productivity growth tried to raise pay in recent decades and claim that research like ours has aimed to convince people that productivity is utterly irrelevant for pay growth. In our case, this latter claim simply isn’t true.

While our work emphasizes influences that have suppressed pay, we’ve been clear over and over again that productivity growth is a necessary condition for raising pay. Necessary, but not sufficient. 

In remarks at a conference where they presented their findings, Summers expressed concern that work like ours had convinced too many on the progressive side of economic debates to look skeptically at policy ideas aimed at spurring growth if they aren’t accompanied by measures to ensure equitable distribution as well.

To be really blunt, we’re not concerned that the policy pendulum has swung too far away from growth and toward equity. Policymakers have ignored the problem of wage suppression for too long, and if they have recently started to focus on this problem then we think it’s a good thing. 

Furthermore, most policy ideas that would reliably spur overall growth actually are progressive in distribution. Public investments in both infrastructure and early childcare and education, for example, are both pro-growth and pro-equity.

So are efforts to target very low rates of unemployment until we actually see some wage growth, and efforts to impose a small tax on financial transactions.

In the end, I think a lot of the difference concerns how to interpret the wage suppression that occurred after 1979. Stansbury and Summers might think this was a discrete historical episode, and productivity growth going forward should be expected to raise wages smartly once we have gotten past it.

We don’t think it was one-off episode, and we expect that going forward, all these instruments of wage suppression will be dialed up and new instruments will be invented by corporate managers and capital owners, unless policy intervenes.

In short, yes, productivity growth remains necessary for typical workers to achieve healthy wage growth. But it cannot be the only response to the imperative to raise Americans’ pay.

Josh Bivens is the director of research at the Economic Policy Institute (EPI).