Demonology, Part I

by Reihan Salam

In recent conversations with friends and friendly acquaintances, ranging from distinguished scholars to accomplished young professionals, I’ve been struck by the vehemence and intensity of the anti-Republican and anti-conservative sentiment. I should note that this sentiment is generally directed against the Tea Party movement, which seems to have become a proxy for, as I understand it, white conservatives. I noticed a few tics that strike me as fairly common in these exchanges. I’ll start by discussing just one of them. 

Conservative political success is inextricably linked to conservative racism. This is a claim that is essential to anti-conservatives. It aims to delegitimate conservative political success, to cast aspersions on advocates of small government (“you might not hold these views, but unnamed conservative voters do, hence … the only morally appropriate thing to do is to advocate government policies that in your view harm the long-term prospects of vulnerable people or, more to the point, shut up”), and, perhaps most importantly, to save time.

For busy people who work long hours, time devoted to gaining high-quality information about the political scene is costly, as it subtracts from time that might otherwise be spent working, indulging in more entertaining activity (which could range from classic leisure activities to the consumption of “infotainment” aimed at like-minded people, etc.) Dismissing those who disagree with you as racist, or subtly motivated by racism, or as apologists for a racist political movement, or otherwise associated in some delicately indirect but no less salient way with racism in the present or even the distant past, allows you to save the time it would take to engage with arguments that might otherwise cause cognitive dissonance. (“But if they actually make a decent point about this …”) 

The most sophisticated version of the racism claim rests on the Glaeser-Alesina thesis, i.e., that high levels of racial diversity are associated with low levels of redistribution. The takeaway is that people are less willing to redistribute to members of perceived outgroups. 

One could also argue, however, that in societies defined by low levels of racial diversity, voters overinterpret racial homogeneity. That is, they too readily assume that people who look like them embrace the same normative ideals. This rejoinder is rarely taken seriously, perhaps because it is considered a good thing to project one’s normative ideals onto others, provided it means a willingness to support redistribution. 

Moreover, people underestimate the extent to which Americans have immediate experience of public transfers. Large numbers of working Americans have siblings, children, and other relatives who have at some point been the beneficiaries of various social programs. This is taken as evidence of hypocrisy: aha, you like the programs that benefit you, but not those that benefit others! A subtler point is that at least some people actually recoil from their experiences with public bureaucracies and transfers, e.g., one might have at one point used a transfer program, like food stamps, as a bridge, yet the experience didn’t inure them to the sense that it wouldn’t be ideal to permanently subsist on said program. I might see how a brother or sister has fared as a dependent on the state, and sense that an alternative mode of arranging society, in which civil society groups played a larger role in meeting the needs of the less affluent and (in some cases) the less capable, would be preferable. Indeed, there are so many mental models that seem at least as plausible, if not more so, than “I don’t like members of outgroups” that it’s a wonder that the racism model has such a purchase on the anti-conservative imagination — or rather it’s a wonder if you don’t believe in the power of time-saving rhetorical devices.

The Economist has made a recent contribution to this discussion:

Social divisions also play a role in determining who within a society prefers greater redistributive taxation. In America blacks—who are more likely to benefit from welfare programmes than richer whites—are much more favourably disposed towards redistribution through the fiscal system than white people are. A 2001 study looked at over 20 years of data from America’s General Social Survey and found that whereas 47% of blacks thought welfare spending was too low, only 16% of whites did. Only a quarter of blacks thought it was too high, compared with 55% of whites. In general (though not always), those who identify with a group that benefits from redistribution seem to want more of it.

This might complicate our ingroup-outgroup view. Few would argue that the affluent African Americans in the sample are “racist” for favoring policies that serve the interests of their coethnics, yet non-poor whites are presumably more suspect for not mirroring the preferences of non-poor blacks. 

The article cites a paper by Erzo F.P. Luttmer, which ends with the following discussion:

Social psychologists have documented perception biases in which poor outcomes of “in-group” members tend to be attributed to adversarial external circumstances but poor outcomes of “out-group” members tend to be attributed to characteristics of those out-group members (Brown 1986; Brewer and Miller 1996). This might explain why respondents perceive welfare recipients of their own racial group as more deserving. Many have argued that idleness, out-of-wedlock births, and other behaviors of welfare recipients that conflict with mainstream values influence public support for welfare spending (Heclo 1986; Will 1993; Kull 1994; Bowles and Gintis 1998). There is no definitive evidence on mechanisms through which interpersonal effects operate. Investigation of these mechanisms is left for future research.

I would argue that the latter stream of research is important to keep in mind. I’d also caution that research agendas in this space are shaped by the norms established within a discipline.

The Economist goes on to cite recent work on “last place aversion,” from economists Ilyana Kuziemko, Ryan Buell, Taly Reich and Michael I. Norton. Here is the magazine’s discussion of the paper:


Instead of opposing redistribution because people expect to make it to the top of the economic ladder, the authors of the new paper argue that people don’t like to be at the bottom. One paradoxical consequence of this “last-place aversion” is that some poor people may be vociferously opposed to the kinds of policies that would actually raise their own income a bit but that might also push those who are poorer than them into comparable or higher positions. The authors ran a series of experiments where students were randomly allotted sums of money, separated by $1, and informed about the “income distribution” that resulted. They were then given another $2, which they could give either to the person directly above or below them in the distribution.

In keeping with the notion of “last-place aversion”, the people who were a spot away from the bottom were the most likely to give the money to the person above them: rewarding the “rich” but ensuring that someone remained poorer than themselves. Those not at risk of becoming the poorest did not seem to mind falling a notch in the distribution of income nearly as much. This idea is backed up by survey data from America collected by Pew, a polling company: those who earned just a bit more than the minimum wage were the most resistant to increasing it.

Poverty may be miserable. But being able to feel a bit better-off than someone else makes it a bit more bearable. [Emphasis added]

Note the dismissive if not contemptuous tone of that last sentence. And here is a brief passage from the paper itself:

Last-place aversion would predict that those making just above the current minimum wage would face a trade-off : on the one hand, they may receive a raise if the new minimum wage is above their current wage; on theother hand, they would then join the “last-place” group. In data we collect ourselves, we find that support for a minimum wage increase is lowest among those making just above the current minimum. Using data from the Pew Research Center, we also find that support for a minimum wage increase is relatively low among groups whose income would suggest they themselves make close tothe minimum wage.

Another possibility, which rest assured will be discounted, is that people who make close to the minimum wage are more attuned to who earns the minimum wage, e.g., that it is often not the primary earner in the household, that many young workers earn the minimum wage, etc. And so these workers might be more attuned to the consequences of the minimum wage for hours worked. That is, workers who earn close to the minimum wage might have more and better information about the dynamic consequences of a minimum wage increase than those who earn many multiples of the minimum wage.

I’ll end by noting that there’s something interesting about interrogating the secret reasons one might advocate certain policies. Are there people who advocate redistribution because they are racists, and assume that members of outgroups are feeble and incapable of self-help? I assume that the answer, in at least some small number of cases, is yes, yet this doesn’t in itself discredit the views of others. In my limited experience, this view is more pervasive than is commonly understood, though of course it is characterized in different language that is more pleasing to the ear.  

This subject merits much lengthier treatment. But I hope I’ve given you some food for thought.