Trump has said he wants to spend a lot of money on infrastructure. I am not dead-set against this idea, but here are three reasons to be skeptical.
First: A widespread view holds that a period of low interest rates is a great time for the federal government to borrow money to spend on infrastructure. (That view is typically expressed by liberals, but it is not exclusive to them.) This view is mistaken if interest rates are low because the market correctly regards our prospects for economic growth as low. The mere fact that interest rates are low, that is, does not mean—or even necessarily suggest—that borrowing money to pay for infrastructure will pay off in a positive return.
Second: The Keynesian argument for infrastructure spending—that it will supply fiscal stimulus—is in tension with the argument that it should be done to boost productivity. If the goal of the program is to put money in people’s pockets, inefficiency is a plus. Have the workers use spoons and forks instead of picks and shovels. To the extent policymakers believe the Keynesian argument, it will tilt them that way. So will politics.
Third: The other day I wrote about “monetary offset”: the possibility that fiscal stimulus will cause the Federal Reserve to run a tighter monetary policy and leave the size of the economy roughly where it would have been without that stimulus. In that scenario, while the total amount of economic activity would remain the same its composition would change. For example, government spending might increase (because of the infrastructure program) but private investment decrease (because the Fed moves interest rates higher than it would without that program). All the more reason to make sure that the program is designed in a way that will make Americans more productive—and, if it cannot be done, to abandon the idea.