Rowe on Government Debt

Nick Rowe gives an explanation of why government debt imposes a burden on future generations. He even mentions James Buchanan. Rowe's argument is one of his all apple consumption economies. Still, I think I agree with his argument.

My view, which follows one of Buchanan's simple formulations is that financing government consumption spending by debt allows the current generation to receive the services from public goods at a lower tax price.

While those buying the government debt give up private goods and services (to provide the resources to needed to produce the public goods,) they receive government bonds in exchange and are no worse off. They can receive the principal back in the future, along with interest.

If they had not purchased the government bonds, they could have instead purchased capital goods, either directly or else equity or debt claims. These capital goods would have increased future output, and provided a stream of future income. Further, if the capital goods are not replaced when they wear out, that provides additional consumer goods to pay back the principal.

When they instead purchase the government debt, the stream of income and principle payments come from future taxpayers. Either future taxes are higher than they otherwise would be, and future taxpayers have fewer private goods and services than they otherwise would have, or else, the government pays interest out of given tax revenues and provides fewer public goods in the future than otherwise. Those in the future receive fewer services from public goods than they otherwise would have.

Obviously, this argument is an alternative framing of crowding out of investment and so reduced future production. To me, that only means that one should be careful not to try to make the arguments additive.

Rowe makes the argument without public goods (it is a straight transfer) and without capital goods (it is all consumption loans.) Assuming the interest rate is greater than the rate of growth of the economy, Rowe argues, future taxpayers must give up consumption to pay off the loans, and are worse off. As in the more complicated version I described, the loss to the taxpayers is not offset by the payment bondholders, who are being compensated for the consumer goods they gave up when they purchased the bonds.

Rowe argues that if the interest rate is less than the growth rate of the economy, this no longer holds. Then it is possible to fund government spending by debt without there being any burden on future taxpayers.

In my view, that isn't exactly correct. Interest expense is part of the future budget, and taxes are higher than they otherwise would be, or the provision of public services are less than they would otherwise be. With a growing economy, some of the benefits of future growth are transferred from the future to the present. From my perspective, rather than future generations being able to enjoy lower tax rates because essential government services can be provided with a lower fraction of growing incomes, they must pay interest on past debts.

But Rowe's framing is that the government runs a deficit today to provide public goods (or transfers in his example) and then runs deficits in the future to fund the interest payments on the national debt. If the interest rate on the national debt is less than the growth rate of the economy, the national debt shrinks relative to income. As the centuries pass, the burden of the national debt becomes smaller and smaller. In the year 10,000, perhaps a philanthropist could pay it off with spare change. Of course, that is a burden--just very small. But in that scenario, it never need be paid off.

If this is true, and the interest rate is less than the growth rate of the economy, then why have taxes? Why not provide public goods until their marginal value is zero? Well, presumably private consumer goods and services would still have value, so why not transfers? Apparently, the government should run a deficit and create a national debt high enough that the interest rate is equal to the growth rate of the economy. Outside of Rowe's apple world, this would presumable work at both margins--raising the interest rate and reducing the growth rate of the economy.

What troubles me most about these sorts of thought experiments is that they assume perfect knowledge about the indefinite future. For example, suppose the government can borrow today at a rate that is lower than the growth rate of the economy. We decide to take advantage of the free lunch. What happens if the economy grows more slowly or the interest rate rises? For example, suppose the population begins to decline in the U.S., while great investment opportunities in China pull up interest rates? Maybe those future generations should impose capital controls and outlaw birth control?

Can we really say that there is no burden on future generations?