Brad DeLong has posted a draft statement on fiscal policy for the IMF conference on “rethinking macroeconomics” — and I’m shocked, in a good way. As regular readers may have noticed, Brad and I share many views, so I expected something along lines I have also been thinking. Instead, however, Brad has come up with what I believe are seriously new ideas — enough so that I want to do two posts, following different lines of thought he suggests.
What Brad argues are two propositions that run very much counter to the prevailing wisdom, especially among Very Serious People. First, he argues that we should not only expect but want government to be substantially bigger in the future than it was in the past. Second, he suggests that public debt levels have historically been too low, not too high. In this post I consider only the first point.
So, how big should the government be? The answer, broadly speaking, is surely that government should do those things it does better than the private sector. But what are these things?
The standard, textbook answer is that we should look at public goods — goods that are non rival and non excludable, so that the private sector won’t provide them. National defense, weather satellites, disease control, etc.. And in the 19th century that was arguably what governments mainly did.
Nowadays, however, governments are involved in a lot more — education, retirement, health care. You can make the case that there are some aspects of education that are a public good, but that’s not really why we rely on the government to provide most education, and not at all why the government is so involved in retirement and health. Instead, experience shows that these are all areas where the government does a (much) better job than the private sector. And Brad argues that the changing structure of the economy will mean that we want more of these goods, hence bigger government.
He also suggests — or at least that’s how I read him — the common thread among these activities that makes the government a better provider than the market; namely, they all involve individuals making very-long-term decisions. Your decision to stay in school or go out and work will shape your lifetime career; your ability to afford medical treatment or food and rent at age 75 has a lot to do with decisions you made when that stage of life was decades ahead, and impossible to imagine.
Now, the fact is that people make decisions like these badly. Bad choices in education are the norm where choice is free; voluntary, self-invested retirement savings are a disaster. Human beings just don’t handle the very long run well — call it hyperbolic discounting, call it bounded rationality, whatever, our brains are designed to cope with the ancestral savannah and not late-stage capitalist finance.
When you say things like this, libertarians tend to retort that if people mess up on such decisions, it’s their own fault. But the usual argument for free markets is that they lead to good results — not that they would lead to good results if people were more virtuous than they are, so we should rely on them despite the bad results they yield in practice. And the truth is that paternalism in these areas has led to pretty good results — mandatory K-12 education, Social Security, and Medicare make our lives more productive as well as more secure.
Now, Brad argues that we’re going to need even more of this kind of paternalism. An aging population and the demand for a more highly educated work force certainly push in that direction. It’s less clear, I’d say, that health care will be a big driver, since the rate of growth of health spending seems to have slowed.
But he certainly has the principle right. To think about the growth of government, we need to look at the range of things government does well, a range that goes well beyond the narrow concept of public goods.