Interpreting Stiglitz

Mark Thoma points to a New York Times commentary by Joseph Stiglitz, in which he suggests a solution to the US trade deficit:

There is one way out of this seeming impasse: expenditure cuts combined with an increase in taxes on upper-income Americans and a reduction in taxes on lower-income Americans. The expenditure cuts would, of course, by themselves reduce spending, but because poor individuals consume a larger fraction of their income than the rich, the “switch” in taxes would, by itself, increase spending. If appropriately designed, such a combination could simultaneously sustain the American economy and reduce the deficit.

There is only one problem with this solution (as Greg Mankiw points out, a bit more politely than here): it doesn’t make any sense.

If the decrease in government spending is offset by an equal increase in consumption due to the change in tax policy, we end up just where we started. As a matter of basic national income accounting, output minus government spending minus consumption equals domestic investment minus the trade deficit. Shifting output from government spending to consumption can’t reduce the trade deficit unless it increases total output or reduces investment. The Fed won’t tolerate an increase in total output (except for normal growth), so that leaves investment as the only possible offset for the trade deficit. But reducing investment doesn’t, in principle, help matters: we end up borrowing less (a lower trade deficit) but also building less productive capacity to pay back what we do borrow.

However, judging by Prof. Stiglitz’ Nobel Prize (as well as by some papers I remember reading by him), he seems like a pretty smart guy. I have to believe there is some method in his madness. The missing piece of this puzzle must have to do with the composition of investment. Some investment (plant, equipment, software) clearly does increase our ability to pay back what we borrow, but other investment (residential housing) probably not so much (although you can try to make an argument that the pleasure of living in better houses will substitute for other material consumption in the future, enabling us to send abroad more of what we produce). Largely as a matter of arithmetic, therefore, I conclude that Prof. Stiglitz expects his proposal somehow to reduce residential investment.

But how, exactly? It would have been nice if he had spelled it out. It seems like a vaguely reasonable conjecture that richer people are more likely to use their marginal income to buy bigger houses, compared to poorer people. (Housing, above some minimum threshold, is a “superior good,” perhaps.) But it isn’t obvious that this would be the case. If Prof. Stiglitz wants to convince other economists, he’s going to have to fill in the rest of the argument here.