Slate columnist Jamelle Bouie writes:

Nearly twice as many whites as blacks favor the death penalty. There is a simple, and disturbing, reason why. . . if you needed a one-word answer to why whites are so supportive of the death penalty, “racism” isn’t a bad choice.

Actually, I do think it’s a bad choice, and this graph shows why:

Screen Shot 2014-03-30 at 10.55.35 AM

The four curves (each with gray uncertainty bounds) show estimated support for capital punishment among four groups of Americans: nonblack men, nonblack women, black men, and black women. (The surveys we are analyzing start in the 1950s and back then there were so few respondents outside the “white” and “black” groups so for simplicity we lump all non-African-Americans into a single “nonblack” category.) All the curves are shown, relative to the national average. The pattern is clear: blacks’ and whites’ attitudes on the death penalty have steadily diverged over the decades.

Here’s the full graph with caption:

Screen Shot 2014-03-30 at 10.54.10 AM

And this all comes from the paper, Hierarchical models for estimating state and demographic trends in US death penalty public opinion, by Kenny Shirley and myself.

Why does our graph contradict Bouie’s claim? He’s attributing the difference in attitudes to historical racism among white southerners. But the gap in attitudes has increased during recent decades, when racism has declined. Whites have become more politically conservative, but that’s not the same as becoming more racist.

To accept my argument, you don’t need to believe that there is no racism among American whites. All you have to accept is that white racism is much lower than it was in the 1950s, which seems clear enough to me, given survey evidence on direct questions about racism. In their 2005 paper “Old Times There Are Not Forgotten: Race and Partisan Realignment in the Contemporary South” from the American Journal of Political Science, Nicholas Valentino and David Sears summarize their analysis of GSS racial attitudes questions as follows: “There is much evidence that Jim Crow or ‘old-fashioned’ racism has declined greatly, but we doubt that the more contemporary symbolic racism has.” Here are the time series:

Screen Shot 2014-03-30 at 11.26.27 AM

So, according to the data: yes there’s still racism but it has on balance decreased, not increased, in recent decades. Hence I don’t think it makes sense to attribute the big black-white gap in death penalty attitudes to racism.

What I do think is happening is an increase in conservatism among whites, not just in the south but in some other parts of the country as well.

Consider this map from my paper with Kenny Shirley showing estimated trends in death penalty support by state during the period 1955-2005:

Screen Shot 2014-03-30 at 11.42.10 AM

This is not so different from a map of strong Republican states, and it includes states such as the Dakotas which are not southern but whose voters are strongly conservative and Republican.

The short story is that capital punishment has become a partisan attitude, associated with general conservatism. I agree with Bouie that racism is part of the story, but it’s coming in a context in which racial attitudes have an increasingly partisan flavor.