Inequality undermines growth?

Kevin Drum's blog pointed me toward this short piece, by David Callahan, suggesting that economic inequality reduces economic growth. This would commend to us a "bigger pie" argument for redistribution.

Like Drum, I don't find the case very compelling on the face of it. Why would greater inequality slow growth? One old story is that the rich save a greater share of their income than the rest of us, so redistribution upward would raise the aggregate savings rate and reduce the share of consumption spending in the economy. (Back when I was an undergrad at UMass, the notion that the owners of capital save more of their income than do the workers was called the Cambridge saving function.) If inequality lowers consumption, this could dampen aggregate demand, slowing economic growth, a la Keynes. Since inequality trends are fairly longterm trends, this claim would require the existence of "secular stagnation," another old idea that seems to be making a comeback these days.

Still, there is a long-run argument based on the Cambridge saving function that goes the other way. Since growth is partly a function of capital accumulation, more saving means more investment means more productivity and a bigger pie. This is the classic trickle-down story. Not just Reagan, but Bob Solow.

Which story is correct? That, my dears, is an empirical question.

It would be nice to have strong empirical evidence on the causal relationship between inequality and aggregate growth. Alas, the identification problem is daunting. "Identification problem" is econo-speak for "correlation does not imply causation." And just what is the raw correlation between growth and inequality? Here's the OECD. Ummm....