Debt still matters

When the government borrows, someone—eventually—will have to pay it back

John H. Cochrane

The Grumpy Economist

John H. Cochrane | Nov 12, 2020

In 2021, the United States is projected to pass a milestone: US federal debt is forecast to exceed 100 percent of GDP. In fact, according to an October estimate from the nonpartisan Committee for a Responsible Federal Budget, the debt has already eclipsed that mark. But does all this debt matter, or is worrying about debt passé?

This debate has been going on among economists for a while. Modern monetary theory has gotten a lot of attention for its arguments that debt has few consequences, but one need not go to the extremes of MMT to find support for such a view. Some in the mainstream of economics have argued that since the interest rate on US government debt may be lower than the growth rate of the economy, the US can roll over debt forever. Others have advocated that additional debt-financed spending may have so strong a multiplier as to pay for itself, a super-Keynesian version of the Laffer curve.

Unlike MMT, these are logically consistent possibilities. But are they right? 

The interest rate on US government debt is indeed slightly lower than good guesses of the economy’s growth rate, as sadly low as the latter is, so if we roll over the debt with no additional deficits, the debt-to-GDP ratio will slowly decline and the US can indeed run this slow-rolling Ponzi scheme. 

But how long will this happy circumstance of ultralow interest rates continue? More to the point, how scalable is this opportunity? Bond-market investors lend 100 percent of GDP to the US government at 1 percent interest. Will they lend 200 percent of GDP at the same low interest rate, or will they start to require higher interest rates? A government that finances itself only with money and no debt need not pay back the money—but, obviously, that government cannot double the opportunity. 

More deeply, like many good arguments, this one misses the central point. What happens when the US piles on larger and larger debt-to-GDP ratios each year, and doubles the debt-to-GDP ratio in each decennial crisis? The argument I referenced above is about sustainability of a large but steady debt-to-GDP ratio. It really is only about whether the government can run a small, steady deficit, or must run a small, steady surplus to maintain that steady debt-to-GDP ratio. It does not justify a debt-to-GDP ratio that grows forever. 

A test of bondholders’ patience

US federal debt as a percentage of GDP has gone up sharply, and is projected to continue its climb in coming years.

US Congressional Budget Office

Everyone recognizes that there is a debt-to-GDP ratio limit out there somewhere. It is not a hard and fast limit. Bondholder patience combines government debt, other opportunities, and bondholders’ view of the government’s political stability and willingness to pay back debt. What is clear is that finding the limit will be unpleasant, involving essentially a sovereign debt crisis, a sharp inflation, devaluation, and financial catastrophe. 

The view that debt-financed spending pays for itself is likewise limited to a period of “secular stagnation” with perennially deficient demand, sticky prices and wages, and other requirements of extreme Keynesianism. Are we in such a period, or is COVID-19 a supply shock? Was the economy really suffering from lack of demand when unemployment hit a 50-year low this past February? 

Washington knows no such sophistication, but our politicians have grasped the logical implications of the proposition that debt does not matter with more clarity than have economists. 

The notion that debt matters, that spending must be financed sooner or later by taxes on someone, and that those taxes will be economically destructive, has vanished from Washington discourse on both sides of the aisle. The COVID-19 response resembles a sequence of million-dollar bets by non–socially distanced drunks at a secretly reopened bar: “I’ll spend a trillion dollars!” “No, I’ll spend two trillion dollars!” That anyone has to pay for this is unmentioned. 

And who is to blame the politicians for acting this way, really? Markets offer 1 percent long-term interest rates—negative in real terms. Blowout spending financed by the Fed printing money—which is no different from debt—has resulted in no inflation so far. In the face of the deep concerns of current voters, worry that our children and grandchildren might have to pay off debt is not particularly salient. They’re either in the basement playing video games or out protesting for the end of capitalism anyway. Politicians will take the cheap money as long as markets are happy to provide it. 

The economists, even the modern monetary theorists, envision debt issued to pay for worthy investments, or valuable spending, all undertaken with a careful green-eyeshade approach. Washington has figured out the logical conclusion of the idea that federal debt doesn’t matter, in a way these economists have not: If debt and money printing have no fiscal cost, why be careful about how you spend money? Send checks to voters. Why not? It’s costless. No boondoggle project is objectionable. Send billions to prop up dying businesses. Why not? It’s costless. Why bother fixing the post office? Send them another $25 billion. Or $100 billion.  

We can go further. Why should citizens have to pay back debts if the federal government does not have to do so? Bail out student loans. Bail out bankrupt state and municipal governments and their pensions. Cancel the rent. Cancel the mortgage. Why should anyone have to pay any debt if our federal government has access to a money machine? Why work? Why should the federal government not just keep printing money and sending it to us? Other countries are not so lucky as we are. Why should emerging markets pay back debt if the US does not have to? Bail them out. 

Nobody expects a debt crisis, or it would have already happened. 

These are inescapable logical conclusions of the view that federal debt has no fiscal cost. If you’re uncomfortable with the end of the trip, perhaps you should revisit the assumption from which it inexorably follows. 

Advocates point to World War II. It is true that the US exited the war with an even greater debt-to-GDP ratio than it has today. It was not painless. Growth higher than interest rates was part of how the US managed it, but it’s not the whole story. Two bouts of inflation, in the late 1940s and in the ’70s, devalued much debt. The US ran steady primary surpluses (excluding interest costs) from the 1940s through the mid-’70s. Spending was low in the pre-entitlement economy, and the government was not totting up hundreds of trillions in unfunded promises. The war, and its spending, was over. Statutory personal taxes and actual corporate taxes were high. Financial repression and closed international capital markets kept interest rates on government bonds low, and deprived Americans of better investment opportunities—and the world’s economies of much-needed investment capital. With all that, we still had an international debt crisis in the early 1970s, prompting the abandonment of the Bretton Woods system and depreciation of the dollar. 

In short, the US grew out of WWII debt by not borrowing any more, by decades of fiscal probity, and by strong supply-side growth in a deregulated economy. We have none of these reassurances going forward. And this event, and the United Kingdom’s exit from Napoleonic War debt in the 1800s by starting the industrial revolution, are about the only historical examples of a semisuccessful repayment of this much debt. Otherwise, the history of large sovereign debts is one long, sorry tale of default, inflation, devaluation, and consequent financial chaos. The UK did not exit WWII debt successfully, leading to crisis after crisis, and everyone else did worse. 

Still, what should we be afraid of? The vision of grandchildren saddled with taxes, or even just unable to borrow more while the economy sits at its limit of, say, 200 percent debt to GDP, is indeed not a salient brake to spending. 

That is not the danger. The danger the US faces, the danger we should repeat and keep in mind, is a debt crisis. We print our own money, so the result may be a sharp inflation that wipes away the value of debt rather than an even more disruptive default, but the consequences will be almost as dire. 

Imagine that five or even 10 years from now we have another crisis, which we surely will. It might be another, worse, pandemic, or a war involving China, Russia, or the Middle East. Imagine the US follows its present trends of partisan government dysfunction, so an impeachment is going on, as well as a contested election, and militias even roam the streets of still-boarded-up cities. Add a huge economic recession, but without any reformed spending promises. 

At this point, the US has, say, 150 percent debt to GDP. It needs to borrow another $5 trillion–$10 trillion, or get people to hold that much more newly printed money, to bail out once again and pay everyone’s bills for a while. It will need another $10 trillion or so to roll over short-maturity debt. At some point, bond investors see the end coming, as they did for Greece, and refuse. Not only must the US then inflate or default, but the normal crisis-mitigation policies—the firehouse of debt relief, bailout, and stimulus that everyone expects—are absent, together with our capacity for military or public-health spending to meet the shock that sparks the crisis. 

Interest rates do not signal such problems. They never do. Greek interest rates were low right up until they weren’t. Interest rates did not signal the inflation of the 1970s, or the disinflation of the ’80s. Nobody expects a debt crisis, or it would have already happened. 

As noted above, there is no defined limit to the debt-to-GDP ratio that policy makers can use for guidance. Countries can borrow a huge amount when they have a decent plan for paying it back. Countries have had debt crises at quite low debt-to-GDP ratios when they did not have a decent plan for paying it back. Debt crises come when bondholders want to get out before the other bondholders get out. If they see default, haircuts, default via taxation, or inflation on the horizon, they get out. Sound long-term financial strategy matters.

We cannot tell when the conflagration will come. But we can remove the kindling and gasoline lying around. Reform long-term spending promises in line with long-term revenues. Reform the tax code to raise money with less damage to the economy. The Treasury and Fed should secure long-term government financing, locking in low interest rates. And spend only as if someone has to pay it back. Because someone will have to pay it back.

John H. Cochrane is a senior fellow at the Hoover Institution at Stanford University and distinguished senior fellow at Chicago Booth. This essay is adapted from two posts on his blog, The Grumpy Economist.