Progress for the Poor
I spent some of the long Thanksgiving weekend reading Lane Kenworthy‘s short new book Progress for the Poor (Oxford University Press, 2011). In just over 100 pages of text and graphs (plus another 50 of notes, references, and technical details), Kenworthy gives a superb and accessible analysis of anti-poverty programs in the wealthiest OECD countries.
Each of the 11 tightly written chapters is built around a single argument about the fight against poverty. I find almost every one compelling: “Growth is Good for the Poor, If Social Policy Passes It On”; “Generous Social Policy Reduces Material Deprivation”; “Low Wages Need Not Mean Low Incomes”; “Public Services Are an Important Antipoverty Tool”; and “The Tax Mix Matters Less Than We Thought,” to list just a few. Each of these short essays is well-situated in ongoing economic, sociological, and policy debates on the welfare state. And all of the arguments draw heavily on Kenworthy’s deep knowledge of internationally comparable statistics and his impressive ability to present and discuss data.
Most of the book stakes out positions in the various welfare-state debates that are consistent with my own thinking. The only significant disagreement I have with the book is the chapter that argues “Targeting May Not Be So Bad,” where Kenworthy argues that targeted anti-poverty programs may be better for the poor than universal programs. My view has long been that, more often than not, “programs for the poor are poor programs.” Think of the difference in quality between Medicaid, the government’s chronically underfunded, means-tested health-care program for the poor, for example, and Medicare, the government’s universal, single-payer health-care plan for the elderly.
Kenworthy presents international evidence for the 1980s and part of the 1990s that countries with universal social welfare systems had more success fighting poverty than was the case for countries with more targeted programs. These same data suggest, however, that by the 2000s, universal systems had lost their edge and no longer offer any more redistribution than targeted systems.
But, Kenworthy’s data don’t make an open-and-shut case on this particular point. In 1980, for example, the United States was –sensibly– deemed to be among the most targeted countries in the OECD. By 2005, however, according to Kenworthy’s measure (which updates a measure first used by Walter Korpi and Joachim Palme), the United States had become the second most “universal” social welfare system in the sample. Meanwhile, Denmark, which provides, among other things, universal health care and statutory or socially funded paid vacation days, paid sick days, and paid parental leave, moved from the second-most universal country in 1985 (close to Sweden) to the second (or third, it is hard to be sure from the graph) most targeted country. Any measure that suggests that the United States –which has a stingy welfare state that makes extensive use of means-tested programs such as Medicaid, TANF, Food Stamps, and the EITC– is more “universalist” than Denmark is not going to move me very far from my priors. To be fair, Kenworthy finishes this chapter with a nuanced discussion of the theoretical and empirical challenges that make it hard to draw strong conclusions here, but my reading is that he leans further (here and in some of his earlier work) toward targeting than I do.
That said, you are very unlikely to find a better, briefer overview of contemporary analysis of what works and what doesn’t when it comes to anti-poverty policy in the high-income OECD countries.