Do deficits matter?
By Scott Sumner
I am seeing a disturbing rise in “free lunch” thinking. One place this increasingly shows up is in the case of “deficits”. People seem to have trouble grasping that deficit spending implies future austerity.
Let’s start with an electric company. Suppose it plans on being in business forever. It decides to issue enough debt so that it’s current stock of debt at any given point in time is equal to 1% of the GDP of the city it serves. If the city it serves is growing over time, the electric company can basically run “deficits” forever, or so it would appear.
In fact, the electric company still faces a budget constraint. It must still service that debt with money earned from customers. The net present value of future interest and principle payments is still equal to the value of the debt issued. The debt continues to be a liability, in any meaningful sense of the terms. Ratepayers should still be concerned about the electric company issuing too much debt, and saddling them with huge future liabilities.
The same reasoning applies to the Federal government. Some might argue that if the growth rate of the economy exceeds the interest on debt, then there is a free lunch. Governments can simply issue more debt to pay interest on existing debt. That may work for a brief period, but beyond some point additional debt will push up interest rates, and the marginal cost of the new debt will exceed the growth rate of the economy. Indeed this may well have happened in the US in recent months, as some people believe that rising fiscal deficits partly explain the recent increase in interest rates on Treasury debt. The marginal cost of borrowing exceeds the average cost, and at some point it exceeds the average growth rate of the economy.
Let’s say $X represents the amount of newly issued debt that a government will issue over a period of N years. (Make X as a big a number as you’d like). In that case, the average annual deficit equals $X/N. A decision to borrow more money in any given year will lead to less borrowing (more austerity) in the other N-1 years. Fiscal policy cannot be continually expansionary, even if the government runs a budget deficit in every single year. That’s because the expansionary impact (on AD) doesn’t come from there being a budget deficit, but rather from the deficit being bigger than the year before.
In the very, very long run, there are no deficits at all. The net present value of taxes from now until the end of time should equal the net present value of government spending from now until the end of time. And the same is true of trade deficits. Other countries will not give us stuff for free, and hence we must eventually pay for all imports with exports. Even the outflow of US currency will eventually be reversed, when the world switches to electronic money.
Interestingly, there’s a certain type of person who is so swayed by recent trends that they think both trade and budget deficits can go on forever. That’s wrong. They also think that this is good news for the budget, but bad news for trade. That’s also wrong. If we could run trade deficits forever, that would be really good news. The outside world would be a big Santa Claus for the USA.
Alas, they will eventually want to be paid. Ironically, that fact (which is bad news for us) would be greeted as good news by many people who regard exports as “good for the economy”.
PS. Measured trade deficits can go on forever, as they do not accurately measure actual trade deficits. Indeed they don’t even come close. It’s not at all clear that the US has run large actual trade deficits in recent decades, properly measured.
Budget deficits are also incorrectly measured, usually quoted in nominal terms rather than real terms. In real terms the US often ran budget surpluses during the 1960s and 1970s. The real budget deficit is the change in the real stock of public debt. If the nominal stock of debt rises more slowly than inflation, the real budget is in surplus.
This post is about actual budget and trade deficits, not measured deficits.