Last month in the New York Times, Columbia sociologist Shamus Khan published an essay, “We are Not in This Together.”  Khan’s op-ed tries hard to be fair to the rich and to those who don’t favor greater income redistribution.  But there are problems with his analysis.

Consider part of Khan’s account of the last six decades:

The rich thought, not incorrectly, that high tax rates were handicapping their capacity to advance. And they found common ground with suburbanites who didn’t see social spending as something that enhanced their lives and neighborhoods, but as something that transferred their tax dollars to a different kind of American — urban, of a notably darker hue — who had only recently gained political legitimacy. Through a tax revolt these groups went to work dismantling social programs.

They were terribly successful and they helped turn America on its head. Since the late 1970s, it has been average Americans who have experienced comparative wage stagnation and who are more likely than their parents to stay in the same economic position. For the rich, the story is the exact opposite.

Khan’s assumption that suburbanites who opposed redistribution were motivated by racism is very likely false.  Those who oppose greater income redistribution have tended to be slightly less traditionally racist and slightly more tolerant than redistributionists. That, of course, is not surprising because the rich and successful tend to be both less redistributionist and better educated than the poor—and tolerance increases with education.

Khan’s causal account for income inequality is also highly questionable.  He asserts that a tax revolt by the richest 1 percent and somewhat racist suburbanites was “terribly successful” in “dismantling social programs” and “helped turn America on its head.” Thus, Khan attributes increasing income inequality in substantial part to the dismantling of social programs.

Khan offers no evidence for the causal side of his tale.  In fact, means-tested government spending, which had briefly stagnated in the late 1970s, grew strongly in the 1980s, 1990s, and 2000s.  The relatively minor retrenchments of government programs over the last three decades were swamped by a flood of redistributionist government spending.  As Ron Haskins of the (liberal/centrist) Brookings Institution shows in Figure 2 below, means-tested government spending has increased five-fold.

Since 1980, by which time all but two of the ten programs that spent the most money in 2011 were in place, spending has increased by about $500 billion, from $126 billion to $626 billion after adjusting for inflation. Similarly, spending per person in poverty between 1980 and 2011 increased from about $4,300 to $13,000 or more than $3 spent per person in poverty in 2011 for every dollar spent in 1980. More recently, means-tested spending increased from about $477 billion to $626 billion in the first three years of the Obama administration, an increase of about 31 percent.


Ron Haskins, Combating Poverty, Brookings, 2012

(The solid line shows means-tested spending per person in poverty in 2011 dollars. The dashed line shows total federal means-tested spending in 2011 dollars. The chart is easier to see at the link.)

And that’s just the 10 largest federal means-tested programs.  If you add the smaller federal programs and the state’s contributions to means-tested programs, the total for 2011 was about a trillion dollars a year, which approximates the entire annual receipts of the federal government from individual income taxes.

So, on balance, do bigger redistribution programs increase inequality, reduce inequality, or have no net effect?  Frankly, I don’t know for sure.

But the result that seems most unlikely is that massive increases in means-tested government spending would have such a major effect on reducing disparities that the results would satisfy people like Khan.  I say this because we’ve already tried this approach over the last three decades and it didn’t work.  Further, at the moment we are ramping up a huge expansion of Medicaid and a huge new subsidy program for private health insurance under the Affordable Care Act.  While these efforts might (or might not) achieve short-term redistribution, their unintended consequences (economic distortions and probably fewer jobs and thus lower wages) may offset or even reverse the net effects of these attempts at redistribution.

If you think that redistribution programs couldn’t be a significant cause of increasing inequality, you might want to read “The Redistribution Recession: How Labor Market Distortions Contracted the Economy,” by University of Chicago economist, Casey Mulligan.  Mulligan is a first-rate economist and his statistical analysis is quite sophisticated.

Here is part of the Amazon blurb on the The Redistribution Recession:

Economist Casey B. Mulligan argues that while many of these changes were intended to help people endure economic events and boost the economy, they had the unintended consequence of deepening—if not causing—the recession. By dulling incentives for people to maintain their own living standards, redistribution created employment losses according to age, skill, and family composition. Mulligan explains how elevated tax rates and binding minimum-wage laws reduced labor usage, consumption, and investment, and how they increased labor productivity. He points to entire industries that slashed payrolls while experiencing little or no decline in production or revenue, documenting the disconnect between employment and production that occurred during the recession. . . . [The Redistribution Recession] reveals the startling amount of work incentives eroded by the labyrinth of new and existing social safety net program rules, and, using prior results from labor economics and public finance, estimates that the labor market contracted two to three times more than it would have if redistribution policies had remained constant.

Thus, Khan’s causal story is almost surely false: cuts in redistributionist government programs couldn’t have caused increasing inequality since such programs have increased five-fold.

Yet just because Khan is wrong doesn’t mean that Mulligan’s causal story is right.  Mulligan might be right that redistribution causes recessions and massive job losses or (perhaps more likely) the truth might lie somewhere in the middle between Khan and Mulligan.  The causal paths may go in both directions: redistribution causing job losses, and these losses leading to increased efforts at redistribution.