On That Income Inequality and Income Growth Thing Out There

Ezra Klein has kicked off an expansive and useful conversation about whether reducing inequality or increasing growth (meaning a stronger recovery and driving down unemployment, not longer-term growth, as Matt Yglesias usefully points out), should be the top priority of policymakers. Klein uses my good buddy and former colleague Jared Bernstein’s recent WCEG paper as the starting point. This is good conversation to have, and is no doubt spurred by the founding of CAP’s new WCEG, headed by another former EPI colleague, Heather Boushey. Props also to Paul Krugman, Brad Delong, Matt Yglesias, Dean BakerEzra Klein, Jared Bernstein, Tim Noah, Steve Waldman for their thoughts.

I’d like to add a few thoughts to this discussion.

  1. There’s a danger to dwelling on the question of ‘does inequality hurt growth?’ if it establishes a litmus test that means addressing inequality requires a firm yes. Some lesser lights from the Manhattan Institute are already using this logic. If inequality has no effect on growth it is certainly still worth working towards more equitable growth because it would mean the vast majority—the 99 percent, you might say—would do far better. My colleague Josh Bivens (in the State of Working America) used the CBO’s comprehensive income data to calculate the middle fifth’s income was lower in 2007 by roughly $19,000 compared to a scenario where there had been equitable growth from 1979 to 2007. Josh refers to this as the inequality tax. I think this is the type of calculation that Krugman was looking for in his most recent post, when he illustrates that inequality matters.
  2. This is really a question of where the burden of proof lies in the inequality debate. Too many—even progressives who are deeply concerned about inequality—think they need to make the case that rising inequality affirmatively and mechanically hinders overall growth. They don’t. All they need to show to argue that curbing inequality will benefit the income growth of the vast majority (which is my goal) is to show that rising inequality does not affirmatively and mechanically increase overall growth. And there is no reliable evidence that more inequality leads to more growth. The last three decades illustrates that higher inequality is not associated with greater growth.
  3. There’s certainly no need to choose between the goals of reducing persistent high unemployment and generating more equal growth. As Bernstein points out, these are compatible. I’d go further and say that the very first step in any policy agenda aimed at more equitable growth would be to radically lower unemployment. Of course, as many folks are pointing out, the same political forces and economic interests stand in the way of both goals.
  4. Policy should be focused on generating higher standards of living for the vast majority (consistent with preserving our planet for the following generations). If you accept this, then addressing inequality and growth become a pretty similar enterprise. Frankly, I don’t much care what happens to the top one percent, whether their incomes fall, are flat or continue to skyrocket. This group seems to do all right without much help. In a Journal of Economic Perspectives paper that I co-authored with Josh Bivens, we argue that rising pay for executives and high earnings in finance have driven the growth of top one percent incomes, and that these groups are collecting rents above their economic value. This means we could eliminate their income surge and leave aggregate growth unaffected! See, we could do both at once—fight inequality and generate more income for the vast majority.
  5. My nominee for the ‘defining economic challenge of our time’ would be generating broad-based wage growth. After all, that’s what facilitating upward mobility and expanding the middle class is all about, since middle class families depend almost entirely on wage income. Sometimes the discussion of income inequality seems to overlook this. Of course, the immediate task for generating wage growth is to lower unemployment.