It’s often asserted that cultural diversity is incompatible with expansive social spending. “Diversity or the welfare state: choose one”, declared The Economist.
The incompatibility thesis is sometimes advanced by liberals and libertarians as an argument against social democrats, implying a contradiction between the latter’s cosmopolitan and redistributive commitments.
The thesis goes back to pundits’ interpretation of a series of influential publications by political economist Alberto Alesina of Harvard University and his colleagues in the early 2000s.
Alesina is probably better known for pioneering “expansionary austerity” – a disastrous concept used to justify severe spending cuts in recession in the belief they would promote growth – which John Quiggin has called “probably the most-refuted piece of economic analysis put out in recent decades”, with the work “ exceptionally poor” and “full of silly errors and spurious claims”.
The point here is not to banish Alesina, but rather to remind us to take care of the compounding exaggeration that can occur when pundits and policy practitioners overegg results from research reports that are themselves overegged. (Alesina has just revisited the topic of immigration and redistribution in a working paper fresh off the press. I’ll look at that in the future.)
In his original research, Alesina was seeking to explain why the USA evolved to have less public goods provision and welfare spending than did Europe. In some basic regression analysis he found an inverse correlation between national social spending in one (but not another) index of cultural diversity. Cultural diversity wasn’t the only association; also significant were variables on electoral systems, judicial institutions, and beliefs about personal responsibility.
So even Alesina’s work implies there is something more going on than a simple “choose one or the other” between diversity and social democracy.
Clash between diversity and social democracy: is it a thing?
Associations between this or that come a dime a dozen. A political scientist in this area could probably rattle off a dozen other ones without pause. The questions to ask are whether the association is 1) causal, 2) important and 3) contingent on other variables.
Let’s assume the association is causal – a big assumption – and move to 2). Is it of sufficient magnitude to dominate other factors, as the dichotomous rhetoric implies?
It is thought to be better for life expectancy to be short than to be tall – but nobody would ever say “be short or die young, choose one”. Simple observation reveals that plenty of tall people live to an old age. Other factors – like smoking, alcohol, diet, exercise – are obviously much more important.
Moreover, we would want to know whether any association that does exist is mediated or moderated by other variables. For example, it might (hypothetically) be that tall people are just more accident prone, with the correlation driven by workers in risky occupations rather than tall people in general. Equivalently, it might be that high immigration only undermines the welfare state in countries with certain institutions, or a history of a particular type of racism – such as the American experience of slavery, segregation and ongoing oppression. It might only apply to certain classes of immigrants.
Social democrats vs liberals: migration head-to-head
To examine the incompatibility thesis, let’s compare migration rates in the big Nordic welfare states – we’ll call them social democracies – with what some people like to call the “Anglosphere” states, which have tighter, more targeted welfare systems. (I use the term Anglosphere with misgivings but I can’t think of another label to describe this group of countries with their historical and institutional links.)
As the graph shows, every Nordic social democracy substantially outspends every Anglosphere liberal state on social expenditure.
The incompatibility thesis would predict the Nordic social democracies to be relatively homogenous and closed. Indeed, that’s a widespread perception about Nordic countries. But actually Sweden is more diverse than Australia according to the cultural diversity and ethnic fractionalisation indices. Denmark, Finland and Norway aren’t far behind.
Migration is a mixed bag. Both the Nordic social democracies and the Anglosphere liberal groups include countries with relatively high migration rates and countries with more modest migration rates:
An eyeball of the graph suggests there’s no clear and consistent pattern here. On an unweighted average of the five-year net migration rates, the Nordic social democracies at 28.2 edge out the Anglosphere liberal countries at 22.1.
Here’s the CIA net migration estimates for 2017:
Again, the story is that there is no story. On average, the Nordics appear to have about the same or even slightly greater rates of migration than the Anglosphere. This kind of crude analysis cannot conclude much, but at the very least it does show there’s no automatic, absolute incompatibility between high migration and generous social spending.
What about other advanced economies? The Nordic social democracies compare favourably here too. Of the countries with the OECD’s ten highest net migration rates, three are Nordic social democracies and two Anglosphere states:
Perhaps, it might be argued, high immigration can be compatible with social democracy but only if social democracy comes first. Perhaps a large welfare state needs homogeneity to incubate, to attain widespread support and to institutionally entrench itself, before becoming exposed to the strains of diversity.
The standout refutation of this modified incompatibility thesis is Belgium, which is the world’s third highest social spender (29 percent of GDP). Belgium is characterised by literally thousands of years of deep cultural conflict, with a chasm between its French and Flemish (and German) ethnolinguistic communities. Not only are the major political parties all ethnolinguistically aligned, but so are basic social institutions like hospitals, banks and trade unions. Yet despite profound divides and mistrust it evolved one of the world’s largest welfare states.
So we have good case studies to show there is nothing set in stone to preclude a diverse country from a large welfare state.
Back to Alesina
What about the negative association Alesina identified? Is there anything to see here?
Alesina was trying to explain a specific phenomenon: the lack of welfare in the USA. He suspected, correctly I believe, that this was profoundly influenced by US racial division, and looked for international evidence in support of this hypothesis.
Alesina found no significant correlation between the ethnolinguistic fractionization index, which was then the predominant index of cultural diversity in the literature, and national social spending. Arguing that linguistic diversity was not relevant to his focus – white-black attitudes in the USA – he constructed a new index. He ran a regression, and this one did find a correlation.
Subsequent studies by other researchers have been equivocal. In summarising the literature, economists Stichnoth and Straeten write:
our main conclusion from this survey is that the evidence is mixed at best. In some studies, the level of ethnic diversity is unrelated to public spending or to individual attitudes and behaviour. In other studies, there is evidence of an association, but the association is weak. For instance, in many studies on individual attitudes towards redistribution or public spending, the association with ethnic diversity is much weaker than for other factors such as own income (current or expected) or beliefs about the role of effort versus luck in determining this income.
From an examination of European Social Survey responses in 16 Western European countries, researcher Steffen Mau concludes:
The number of foreigners matters but is outweighed by factors like federalism, unemployment rate or the welfare regimes… Interestingly, the people in social-democratic countries are more in favour of granting equal rights to foreigners compared to the respondents in liberal or conservative regimes…
[O]ne can surmise that the effect of societal heterogeneity on the welfare state’s ability to sustain its legitimacy is limited, and that other factors play a more significant role such as institutional factors and the politics of interpretation.
(Interesting to note the suggestion that people in social democracies might be more supportive of granting equal rights to foreigners. I suspect there’s something to this, and it arguably gains support from the Swedish responses in Alesina’s most recent survey. Social democratic welfare states are built on an explicit public ideology of citizenship equality. There is no equivalent in a liberal welfare state, which explicitly stresses differentiating citizens based on need and desert, thus requiring continuous scrutiny towards the legitimacy of welfare recipients, an arrangement which appears to feed into prevalent attitudes and anxieties about welfare in liberal countries.)
Back to Alesina. The twisted historical dynamic of white-black racial oppression in the USA is unlikely to generalise in any simple way. Alesina hinted as much:
The history of American welfare suggests that enemies of welfare often used race to defeat attempts at redistribution in the post–Civil War period…
It could be argued that ethnolinguistic heterogeneity within some European countries (such as Belgium) is as great as racial heterogeneity in the United States. Furthermore, it is at least possible that this heterogeneity creates antipathy that is as robust as the race-based animosity observed in the United States. However, in no European country is there a minority that is as poor, relative to the rest of the population, as blacks in the United States.
The idea is that racial majorities may turn against the welfare state in contexts where they perceive it benefits minorities more than themselves. In the USA, African Americans have indeed long been more likely to receive welfare benefits due to their oppression and disenfranchisement, so welfare effects a large and visible interracial transfer of wealth (much less than it should effect in my opinion). It is no surprise that American welfare debates have been overtly racialised for decades, and its dominant tropes about welfare recipients figure African American “welfare queens”. This was the dominant public rhetoric that Bill Clinton responded to when he promised to “end welfare as we know it”, delivering the largest welfare cuts in US history.
Australia’s racism and xenophobia don’t play out in welfare discourse to any comparable extent. Given Australia has more than its share of moral panics about both welfare and immigrants, it’s perhaps surprising that these anxieties are not intertwined more than they are. Immigration is much more likely to be attacked based on perceptions of immigrants taking jobs, committing crime, pushing up house prices, stressing infrastructure, siding with foreign interests, and undermining Australian cultural values (whatever these might be).
The comparative lack of racialised welfare discourse probably relates to the reality that nothing in Australia’s migration program has or will lead to the kind of racialised economic disparity seen in the USA. On the contrary, Australia’s skills-centric migration program selects precisely for those migrants most economically secure and least likely to receive social security. Indeed, the economic anxiety over immigrants in Australia seems to be that they might work too hard – and take all the jobs – rather than work too little. That’s no less morally messed up than the converse, but it does create fewer problems for the welfare state.
In summary the association between cultural homogeneity and welfare is not strong if it exists, and less important than several other factors. Where heterogeneity does appear to undermine welfare, it may be related to large-scale, long-term racial disparities in economic standing, such as what is seen with African Americans relative to whites in the USA.
Social democratic welfare states and high migration can and do co-exist, so the incompatibility thesis should be abandoned. Get the institutions right, and the two are totally compatible.