The Economist carries a story about Brazil's reform of its antipoverty programs. Given the marked disparity between classes and regions in Brazil, I'll be watching with interest to see if this works. What really caught my eye, though, was this:
The state already spends an ample 16% of GDP on social programmes at the federal level alone. But little of this reaches the poor. Pensions consume two-thirds of social spending; health and education, universal programmes that offer little extra benefit to the poor, account for much of the rest. Social assistance specifically for the poor accounts for just 0.4% of GDP.
In a sense, we've got the worst of both worlds in a lot of government programs: everyone seems to love huge redistributive programs like Social Security, but people are so wigged out at the thought that they might be, yes, welfare recipients (which they are) that it's politically impossible to do the kind of means testing that would at least lower the cost of serving the program's nominal purpose of reducing elderly poverty.
The delusion that Social Security isn't a welfare program, but some kind of "social insurance" seems to be one of the major obstacles to reform. "I paid in to the system," people say, "no fair changing the rules on us now!" Except someone has to get shafted here, eventually. It's just a question of who. It can be in 2017 and on, when outlays exceed FICA reciepts, or it can be sooner (and a smaller amount) if we fund the transition to private accounts up front. But because people don't think of Social Security as welfare, they're under the (legally unsupportable) impression that they've got some kind of divine right to the same "rate of return" that the last generation saw.