November 13, 2023
The shrieks of the deficit hawks have been growing louder with the fiscal 2023 deficit crossing 6.0 percent of GDP, and the projected future deficits being no lower. With interest rates now close to or above GDP growth, we regularly hear about the story of a looming debt spiral, where high debt leads to high interest payments, which add to annual deficits, making the debt even larger.
To try to think clearly about the deficit and debt, it is worth pulling them apart and dealing with them as separate issues. This avoids many common sources of confusion.
Is the Deficit Too Large? Is Inflation Out of Control?
Back in the old days we used to argue about whether we needed to balance the budget. With a deficit of more than 6.0 percent of GDP, that question seems pretty much moot. If we wanted to balance the budget with tax increases, we would need to increase revenue by almost a third. If we tried to do this on the spending side, we would need cuts in non-interest spending of almost 29 percent. Neither of those seems close to politically realistic. And splitting the difference, tax increases of 15 percent, spending cuts of 14.5 percent, doesn’t look remotely plausible either.
So, we can forget about balanced budgets (thankfully), but the question is then how large a deficit is too large? There actually is a relatively simple answer to this one. A deficit that leads to inflation is a deficit that is too large.
This is the story that the Modern Monetary Theory crew keeps telling us, and it is right. If too much borrowing drives up interest rates, the Fed can counteract upward pressure on rates by buying bonds and/or targeting lower rates. The potential problem from the Fed maintaining low rates in a high deficit situation is that it can lead to too much demand in the economy, causing inflation.
We did see a surge in inflation in 2021 and 2022. Part of that story was too much demand, due to the generous pandemic support from the federal government, however the real story was the supply problems associated with the pandemic. Whatever excess demand problem we had went away when the programs ended, and supply recovered after the ending of the pandemic.
In any case, that history is behind us, the question is whether there is good reason to believe that we are seeing excess demand now. There are still important questions about the future course of inflation, but it has fallen quite rapidly over the last year and is virtually certain to fall further.[1]
Wage growth has also slowed to a pace that is close to its growth rate before the pandemic, when inflation was at the Fed’s 2.0 percent target. Other measures of the labor market, like quit rates and unemployment insurance claims, are at levels that are consistent with a strong labor market, but well below/above the levels seen when the labor market was very tight in the spring of 2022.
In short, it is certainly not clear that deficit-driven demand is so strong that it is causing inflation. Of course, the Fed has raised interest rates by more than 5.0 percentage points from their pandemic low, so arguably we are seeing a classic crowding-out story where demand would be causing inflation, if not for high-interest rates slowing demand.
This story can’t be completely dismissed, but it is hard to tell if we have a frame of reference beyond the recovery from the Great Recession. Interest rates are above their pre-pandemic level, but compared to the slightly more distant past, they still look pretty low.
The 10-year Treasury bond rate is currently roughly 4.6 percent. That is well below the rates we saw from 1998 to 2000 when the government was running budget surpluses. And, inflation is at least modestly higher today than it was in those years, meaning that the real interest rate (the nominal rate minus the inflation rate) is actually somewhat lower now with our deficits of 6.0 percent of GDP than in the balanced budget days at the end of the Clinton presidency.
It is also hard to tell the other part of this story, that investment has been crowded out by high-interest rates. Non-residential investment as a share of GDP is well above its average over the last two decades. The residential investment share of GDP is also well above its average for the decade prior to the pandemic, although down by a percentage point from its pandemic peak. This drop is primarily due to the end of the refinancing boom, since most of the expenses in refinancing mortgages count as residential investment in the national accounts.
High mortgage rates have discouraged home buying. Sales of existing homes are down by more than a third from peaks hit in the pandemic, and more than 20 percent compared to pre-pandemic rates. But part of this story is not simply high long-term rates. There is an unusually large gap between the 30-year mortgage rate and 10-year Treasury rate.
It is currently around 3.0 percentage points, giving us 30-year mortgage rates around 7.6 percent with a 10-year Treasury rate of 4.6 percent. More typically, the gap would be around 1.75 percentage points, which would translate into a 30-year mortgage rate of 6.35 percentage points. That is still considerably higher than the rates seen before the pandemic, which were generally under 5.0 percent, but not nearly as bad as what we are seeing now. (If anyone has thoughts on why we see this extraordinary spread between mortgage rates and Treasury rates, I’m happy to be informed.)
Anyhow, it would certainly be desirable to see somewhat lower interest rates. If inflation remains under control, we will likely see the Fed begin to lower the federal funds rate next year and long-term rates will follow. And, this would be the case even without any deficit reduction.
In short, it might be desirable to see somewhat lower interest rates, but the current rates are not especially high by historical standards. Furthermore, it is not clear that large budget deficits are the main culprit, as opposed to the interest rate policy of the Fed.
Should Exploding Debt Bother Us?
If current deficits are not a big deal, should the prospect of an exploding debt spiral worry us? In this story, the debt continually builds primarily as a result of higher interest payments, which in turn lead to still higher interest payments, until interest ends up taking up a ridiculously large share of the budget.
There are a few points to be made about this story. First, this is a gradual process. Even in the story where we have a deficit of 6.0 percent of GDP, the debt-to-GDP ratio rises slowly. If we have 2.0 percent growth and 2.5 percent inflation, then nominal GDP is increasing by 4.5 percent. With a debt that is roughly 100 percent of GDP, the debt-to-GDP ratio would only increase by around 1.5 percentage points. Even with relatively large deficits, we don’t have to worry about the debt-to-GDP ratio suddenly exploding.
Some of the deficit hawks raise the prospect of a crisis of confidence, where investors suddenly become unwilling to hold U.S. debt and interest rates go through the roof and the dollar goes through the floor. There are certainly examples of countries where this has happened, but not in situations where they had an otherwise healthy economy, as is the case with the United States.
Furthermore, there is no evidence that we are on the edge of anything like this. Interest rates have risen, but that was primarily because they were pushed up by the Fed. And, as noted earlier, they are still relatively low by historical standards. And the dollar is actually higher against most other currencies than it was before the pandemic.
It is also worth noting that the U.S. debt to GDP ratio is not exceptionally high by international standards. The ratio of debt to GDP in France and the United Kingdom is very comparable to the U.S. and in Italy and Japan it is far higher. In fact, using gross debt, a measure that includes debt to public pension funds, Japan has a debt-to-GDP ratio of 260 percent. The current interest rate on long-term Japanese bonds is less than 1.0 percent.
Thinking About the Future
Even if interest is a bearable burden and the dollar is not on the edge of collapse, there are still many who see the debt as imposing a major burden on future generations. This is really a case of missing the forest for the trees.
Let’s say that because of our failure to get the deficit down, a decade from now the debt-to-GDP ratio is 20 percentage points higher than if we had been “fiscally responsible.” If the real interest rate on government debt is 2.0 percent, somewhat higher than it is today, that means we would be paying another 0.4 percent of GDP in interest compared to our fiscally responsible scenario.
That’s not entirely trivial, but it’s also had to tell the story that it is an enormous burden. Our current interest payments on the debt are roughly 2.7 percent of GDP. They were 3.3 percent of GDP in the early 1990s. That burden did not prevent the 1990s from being our most prosperous decade since the 1960s.
It is also worth mentioning that if AI proves to be anywhere near as important as its proselytizers insist, then the uptick in growth over the next decade could actually create a situation where the debt-to-GDP ratio stays flat or even falls.
More importantly, if we get an uptick in growth from AI and other technologies, it will swamp the impact of higher interest burdens. If AI led to just a 0.2 percentage point increase in annual productivity growth, it would imply a GDP that is 4.0 percentage points higher after two decades than in a world without this new technology. (Productivity growth averaged 4.3 percent the last two quarters, after averaging just 1.1 percent in the decade prior to the pandemic. It is important to recognize that productivity data are enormously erratic and subject to huge revisions.)
This sort of boost to output from AI is an order of magnitude larger than any plausible burden associated with a higher debt to GDP ratio. This is especially true when we remember that the bulk of interest payments are made to people in the United States. The interest payments do limit the extent to which we can spend in other areas, without higher taxes, but they are not a net burden on the economy.
It is also important to remember that direct spending is not the only way the government pays for things. It pays for innovation and creative work by issuing patent and copyright monopolies. These monopolies impose an enormous cost in the form of higher prices – effectively a tax paid by households in the future. In the case of drugs alone, this tax is likely in the neighborhood of $500 billion a year, or 2.0 percent of GDP.
Adding in the higher prices we pay as a result of these monopolies for medical equipment, computers, software and a wide variety of other items, it is likely the cost is over $1 trillion a year. The benefits we get from the innovation and creative work fostered by these government-granted monopolies may still make them good policy (there are alternative mechanisms), but to not acknowledge the costs attributable to patent and copyright monopolies when discussing the burden of the debt is simply dishonest.
It is also important to remember that government spending plays a direct role in supporting the economy. Due to efforts to contain the deficit following the Great Recession, the recovery was much slower than it could have been. It took us a full decade to get back to full employment.
This meant both that millions of people were not working, who could have had jobs if we had not been obsessed with austerity, and that tens of millions of workers got lower pay because they lacked bargaining power in a period of high unemployment. As we have seen in the tight labor markets of the pandemic recovery, tight labor markets disproportionately benefit those at the lower end of the wage distribution. The benefits of maintaining a high employment economy are hugely larger than any downsides associated with higher interest payments on the debt.
Finally, if we are keeping generational scorecards, the success of efforts to contain global warming will have far more impact on future generations than plausible increases in the interest burden of the debt. If twenty years from now, we have completely failed in limiting global warming, but we have cut the debt in half, our children will have no reason to be thanking us.
Deficits and Spending Going Forward
Saying that we don’t need to reduce the deficit doesn’t mean that we should not look to raise more revenue to support additional spending in key areas. There are some great ways to raise revenue that would actually make the economy more efficient. My two favorites are a tax on financial transactions (effectively a sales tax on stock, bond, and derivative trades) and shifting the basis of the corporate income tax to returns to shareholders, which would put the tax gaming industry out of business. We can also have higher taxes on rich people, the big gainers in the economy over the last half-century.
There are also areas where we desperately need to spend more money. The expanded child tax credit that Biden put in place with his recovery plan should be brought back. We also need to have additional funding for child care to make it affordable. And, we have a massive problem of homelessness.
This list could be extended at some length, but the point is that there is no urgency for reducing the deficit. We have many other problems that need to be addressed, which will require additional resources. This will mean making space in the economy by raising taxes, but the focus should be on addressing real problems in society, not reducing the deficit.
[1] We can be confident that inflation will fall further because the rental indexes, which account for 31 percent of the overall CPI and almost 40 percent of the core index, are virtually guaranteed to show lower rates of inflation in the months ahead. Private indexes measuring the rent of marketed units that change hands, show sharply lower inflation. The CPI rent indexes, which measure rents in all units, not just marketed units, lag these private indexes by six months to a year.
Comments