Branko Milanovic
March 05, 2018
It has become a truism to say that the welfare state is under stress from the effects of globalization and migration and thus may not be able to provide the same level of income support that it had provided in the past. In my previous post that looked at policies to reduce inequality in the 21st century, I mentioned that I would follow up with a discussion of the welfare state. In this post I want to go back to the origins of the welfare state to understand the origin of the current stress.
As Avner Offer has recently reminded us in his excellent book (co-authored with Daniel Söderberg), the origin of social democracy and the welfare state is in the realization (and financial ability to deal with it) that all people in their lives go through periods where they are not earning anything, but have to consume: this applies to the young (hence children’s benefits), to the sick (health care and sick pay), to those who had a misfortune to get injured at work (worker’s accident insurance), to mothers when they give birth (parental leave), to people who lose jobs (unemployment benefits), and to the elderly (pensions). The welfare state was created to provide these benefits, delivered in the form of insurance, for either unavoidable or very common conditions. It was built on the assumed commonality of behavior or, differently put, cultural and often ethnic homogeneity. It is no accident that the prototypical welfare state born in Sweden in the 1930s, had many elements of (not used here in a pejorative sense) national socialism.
In addition to commonality of behavior and experiences, the welfare state, in order to be sustainable, required mass participation. Social insurance cannot work on small parts of the workforce because it then naturally leads to adverse selection, a point well illustrated by the endless wrangles over US health care. The rich, or those who are unlikely to be unemployed, or the healthy ones, do not want to subsidize the “others” and opt out. The system that would rely only on the “others” is unsustainable because of huge premiums it would require. Thus the welfare state can work only when it covers all, or almost all, of the labor force, i.e. when it is (1) massive and (2) includes people with similar conditions.
Globalization erodes both requirements. Trade globalization has led to the well-documented decline in the share of the middle class in most western countries and the polarization of income. With income polarization the rich realize that they are better off creating their own private systems because sharing the systems with those who are substantially poorer implies sizeable income transfers. This leads to “social separatism” of the rich, reflected in the growing importance of private health plans, private pensions, and private education. The bottom line is that a very unequal, or polarized, society cannot maintain an extensive welfare state.
Economic migration, to which most of the rich societies have been newly exposed in the past fifty years (especially so in Europe), also undercuts the support for the welfare state. This happens through the inclusion of people with actual or perceived differences in social norms or lifecycle experiences. It is the same phenomenon as dubbed by Peter Lindert as a lack of “affinity” between the white majority and African Americans in the US which rendered the US welfare state historically smaller than its European counterparts. The same process is now taking place in Europe where large pockets of immigrants have not been assimilated and where the native population believes that the migrants are getting an unfair share of the benefits. Lack of affinity need not be construed as some sinister discrimination. Sometimes it could be indeed that, but more often it may be grounded in correct thinking that one is unlikely to experience the lifecycle events of the same nature or frequency as the others, and is hence unwilling to contribute to such an insurance. In the US, the underlying fact that African Americans are more likely to be unemployed probably led to less generous unemployment benefits; similarly, the underlying fact that migrants are likely to have more children than the natives might lead to the curtailment of children’s benefits. In any case, the difference in expected lifetime experiences undermines the homogeneity necessary for a sustainable welfare state.
In addition, in the era of globalization more developed welfare states might experience a perverse effect of attracting less skilled or less ambitious migrants. Under “everything being the same” conditions, a decision of a migrant where to emigrate will depend on the expected income in one country vs. another. In principle, that would favor richer countries. But we have also to include a migrant’s expectation regarding where in the income distribution of the recipient country she expects to end up. If she expects to be in the low income deciles, then a more egalitarian country with a larger welfare state will be more attractive. An opposite calculation will be made by the migrants who expect to end up in the higher deciles of the recipient countries’ income distributions. If the former migrants are either less skilled or less ambitious than the latter, then the less skilled will tend to choose countries with more developed welfare states. Hence the adverse selection.
In very abstract terms, the countries that would be exposed to the sharpest adverse selection will be those with large welfare states and low income mobility. Migrants going to such countries cannot expect, even in the next generation, to have children who would climb up the income ladder. In a destructive feedback, such countries will attract the least skilled or the least ambitious migrants and, once they create an underclass, the upward mobility of their children will be limited. The system then works like a self-fulfilling prophecy: it attracts ever more unskilled migrants who fail to assimilate. The natives tend to see migrants as generally lacking in skills and ambition (which may be true because these are the kinds of people their country attracts) and hence as “different”. At the same time, failure to be accepted will be seen by the migrants as confirmation of natives’ anti-migrant prejudices, or, even worse, as religious or ethnic discrimination.
There is no easy solution to the vicious circle faced by developed welfare states in the era of globalization. This is why I argued in my previous blog for (1) policies that would lead toward equalization of endowments so that eventually taxation of current income can be reduced and the size of the welfare state be brought down, and (2) that the nature of migration be changed so that it be much more akin to temporary labor without automatic access to citizenship and the entire gamut of welfare benefits. This last point is discussed in Chapter 3 of my “Global inequality” as well as here.
The original post can be found at GlobalInequality.
