Assume that the stimulus bill has two types of provisions: Things we would do anyway, and things we would not do but are doing because we need to figure out how to spend $800 billion. The house-flipper tax credit is in the latter category: Paying me $15,000 to sell my house to my brother is not the sort of thing we would do outside of stimulus mania. But school construction is in the former category. We will, presumably, build the schools we need and repair the ones we have. We may not do it quickly, but in theory at least, we will do it. The question, then, is "when." And there's a particular upside to doing it now: At the moment, it's unusually cheap for the government to borrow money to fund investment. The reason is fairly simple: Investors fear the instability of the stock market so they've moved a lot of their money to "safe" investments like government bonds. Put another way: A lot of people want to loan the government money right now, which has, in turn, made those loans cheaper. The government can borrow at a lower interest rate today than it will likely be able to in 10 years. As such, it's cheaper to make spend on school construction now than later. if you're worried about total government spending, the smart thing to do is to do as much inevitable spending now, when it's cheap to spend, as opposed to later, when it'll be more expensive. Meanwhile, while conceiving and writing what I thought was a very trenchant post, I went googling for some numbers and found that Ben Furnas wrote much the same thing in late November. So credit to him.
