Redistribution and National Identity
What looks to me like one of the most important articles in political science over the last several years is out in the new American Political Science Review under the rather unprepossessing title, “A Model of Social Identity with an Application to Political Economy: Nation, Class and Redistribution” (available here for APSA members; ungated earlier version available here). Moses Shayo briefly lays out a model of identity that borrows insights from both economics and social psychology. What is interesting is that the model’s predictions are (a) starkly counter-intuitive about the relationship between national identity and preferences over redistribution, and (b) appear to be born out by the data.
Most people tend to assume that strong national identity and strong preferences for redistribution go hand-in-hand - the plausible intuition here is that we are more likely to give to our fellow citizens if we identify strongly with them. This intuition (I’ll get back to this in a little bit) underlies a significant chunk of political theory argument about the relationship between the nation-state and redistributive obligations. Shayo’s argument points in a very different direction - he argues that strong national identity goes together with high income inequality and low desire (among working class voters) for redistribution.
Why? The more similar someone is to other members of a group, the more likely she is to identify with the group, and the higher status the group is, the more likely an individual is to identify with it. This suggests that working class voters are less likely to identify with their class in situations of high inequality (being working class is less high status), and this has consequences for voting behaviour. When working class voters identify more strongly with their class, they are likely to push for redistribution (which will favor their class interests). When working class voters identify less with their class (perhaps because that class is ethnically heterogenous), and more with their nation, they are less likely to want redistribution. Furthermore, these mechanisms are partly recursive - lack of redistribution may increased inequality which makes class identification less attractive still for working class voters, or alternatively, exogenous decreases in inequality may make workers more likely to identify with their (now higher status) class rather than with the nation as a whole. Eventually, an equilibrium is reached:
two types of equilibria may emerge. In the first, the members of the lower class (who constitute a majority) identify with their class. Hence, they vote for a relatively high level of redistribution. A high level of redistribution can in turn help strengthen that class identity by endowing it with a higher status. In the second type of equilibrium, members of the lower class tend to think of themselves more as members of the nation as a whole than as members of a low-status part of it. They are thus less concerned with income redistribution and vote for a lower level of redistribution than they would under class identity. Again, low levels of redistribution can in turn help make identification with the lower class less attractive.
This model seems to be born out by the (admittedly low N) data.
in most economically advanced democracies, national identification reduces support for redistribution. This effect appears to be very large when compared to the effect of economic self-interest. … the model implies that regardless of whether differences in redistributive systems arise from exogenous factors or from multiple equilibria, we should observe a negative relationship between the prevalence of national identification and the extent of income redistribution. A cross-country analysis reveals a very strong negative relationship between these two variables. Indeed, when looking at well-established democracies, the R2 is between 60% and 72 %.
The key graph is here:
Shayo here graphs the percentage of income gained by the lowest quintile of the population via redistribution against a six item national identity scale for advanced industrialized democracies. The negative relationship is striking - the more people identify with their country, the less redistribution is likely to occur, and vice-versa (the relationship is fuzzier when less established democracies are included in the data set, but still looks to be there.
If this all holds up (and I note that this is not a debate that I am directly involved in), it has a number of interesting implications. Here’s one. Some left-leaning political theorists such as David Miller (see his On Nationality ) have sought to reconstruct social democracy around the model of the nation state, arguing that in the absence of other solidaristic identities, nationalism is the most likely way to generate mass support for e.g. welfare states. If Shayo is right, then Miller is wrong on this - the more nationalism, the less solidaristic redistribution we are likely to see (perhaps Miller might rescue his position by arguing that he wants to see a more civic nationalism which is distinct from the kind of nationalism picked up in opinion polls - but this still looks like a quite damaging finding to me). And there are plenty more suggestive lines of argument like this - I imagine that this piece is going to spur a lot of debate.
Comments
Didn’t Evan Lieberman make this argument six years ago in his book on Brazil and South Africa?
Posted by: Luther van Pumpernickel | June 30, 2009 03:15 PM
This reminds me of Jost’s system justification theory.
Posted by: chrismealy | July 1, 2009 01:25 AM
It’s a fascinating claim, but I have to wonder if it hasn’t uncovered a real correlation, but reversed the underlying causal relationship.
To wit, I’m suggesting that nations which have historically harbored strong socialist movements tend, consequently, to be particularly internationalist in orientation. That class-based politics has historically emphasized trans-national solidarity and denigrated jingoistic pride, and that it has often framed itself in opposition to conservative parties that have waved the flag vigorously in defense of tradition. In other words, it’s the historical influence of the labor parties that accounts for the attenuated state of national pride, and not the other way ‘round.
So what do the data actually show?
I think it’s fairly simple. When a nation convinces its citizenry that the nation-state equitably serves the interests of all of its citizens, they tend to feel proud. They tend to embrace and identify with the nation. They’re willing to shoulder individual hardships, and to tolerate large measures of inequality. On the other hand, when citizens become convinced that a nation serves the interests of a restricted economic or social class, they tend to become alienated from the state. They form class-based parties, and advocate for redistributionist policies.
Here’s the curious part. Even after such policies are implemented, the class-based frame doesn’t recede. The nation-state, having been recast as an instrument of class struggle, maintains that role. So a state handing out large gains to the bottom quartile will also tend to experience high levels of disaffection and low levels of pride. Its citizens see themselves in class terms, and since inequality persists despite the steeper redistribution, remain aggrieved. Conversely, a nation that produces extremely high levels of inequality, and does little to ameliorate them, may nevertheless continue to experience high levels of nationalism - and that may be particularly pronounced among the least advantaged. They don’t divide their society into competing quartiles, and they don’t expect the state to level the gaps among them. And while the affluent may point to their material success as evidence of national supremacy, the indigent, lacking such success, tend to point to their membership in the national group itself. It’s a concept equivalent to the racial ‘Wages of Whiteness’: group identity and pride is often fiercest among those for whom it is their most valuable possession, because they have relatively few others.
Posted by: Cynic | July 6, 2009 03:29 PM
Marx made this same argument almost 150 years ago. At the core of Marx’ critique of ideology is that nationalist ethics do not allow for the emergence of class consciousness.
Posted by: Ari | July 6, 2009 11:41 PM
The model appears under-specified. National identity is very strong in East Asian democracies. Are they in the original paper? The variables in S.M. Lipset’s classic POLITICAL MAN and FIRST NEW NATION? I suspect that a more nuanced model would easily reduce the R-square to below .4.
Posted by: larry wade | July 7, 2009 01:38 AM
David Goodhart made exactly the opposite argument in Prospect magazine back in 2004:
http://www.prospect-magazine.co.uk/article_details.php?id=5835
While he didn’t do any actual number crunching or deep theorising like Shayo has, his argument is that the more racially and culturally diverse a society becomes the less empathy its members feel for each other and the less open they are to talk about economic redistribution.
On East Asia national identity is indeed very strong - but is in most countries a profound fault- line within society as Malay or Indonesian or Han Chinese nationalism radically excludes big national and religious minorities - who in SE Asia are often the most economically dynamic segments of the population.
Posted by: Roger McCarthy | July 7, 2009 07:46 AM
I read through the paper and it has many interesting points. But I think the correlation it shows between national identification and redistribution is likely to have a causal factor that is not included in the calculations but is mentioned in the theoretical statement of the model. Specifically, I suspect that ethnic/racial heterogeneity is positively correlated with intensity of national identity: i.e. the more diverse a society, the more like it is to have a higher level of national identity among the “majority” ethnic community. As social scientists John Roemer, Wojin Lee and Karine Van der Straeten have shown, attitudes towards ethnic and racial minorities have had a measurable impact on redistributive policies in western democracies in the last 30 years.
As an example, we have Sarah Palin and her “real America.” What is this “real America?” It is, as best I can tell, hyper-patriotic and composed of non-hispanic whites. This would also fit with the observation that states with the highest self-identification of “American” when asked about their origins were the only states to vote more Republican in 2008 than in 2004. (See Nate Silver’s analysis.) It seems fairly clear that those who identify their ethnicity as “American” were most strongly against Obama because they did not perceive him to be a fellow American, any more than they perceive other ethnic and racial minorities to be Americans.
So, while the paper is interesting, it does not show why certain nations have a higher level of national identification. If this is explained in terms of ethnic/racial heterogeneity, then the explanation is not so much identifying with one’ nation over one’s class, but not wishing to see ethnic/racial minorities benefited by redistributive policies (think about the famous welfare queen).
Posted by: Cap and Gown | July 7, 2009 08:39 AM