In an article for Defining Ideas in February 2019 titled “Who’s Afraid of Budget Deficits? I Am,” I took issue with two well-known economists, Jason Furman and Lawrence H. Summers. Furman is an economics professor at Harvard’s Kennedy School and was previously the chairman of President Obama’s Council of Economic Advisers. Summers, who is president emeritus of Harvard, was head of the National Economic Council under Obama and had been treasury secretary under President Clinton. They had written a piece in Foreign Affairs (which is, unfortunately, gated) titled “Who’s Afraid of Budget Deficits?” They claimed, as their article’s title suggests, that we should not fear budget deficits and that we should even be willing to support even higher government spending. Indeed, the two authors did not specify even one government spending program that should be cut.
Unfortunately, because of actions by President Trump, President Biden, and Congress, federal government spending is out of control. The solution is not higher taxes, which are already quite high as a percent of gross domestic product; the solution is to restrain the growth of government spending.
Furman and Summers revisited
One of Furman’s and Summers’s arguments was that the main culprit behind higher deficits was tax cuts, not increases in government spending on entitlements. They wrote:
The tax cuts passed by Presidents George W. Bush and Donald Trump totaled 3 percent of GDP—much more than the projected increases in entitlement spending over the next thirty years. Those cuts meant that in 2018, the federal government took in revenue equivalent to just 16 percent of GDP, the lowest level in half a century, except for a few brief periods in the aftermath of recessions.
But that argument no longer holds water. In 2023, according to the Congressional Budget Office, federal revenue will be 18.4 percent of GDP. That’s 1.2 percentage points above its 17.2 percent average in the thirty years from 1993 to 2022. And, more relevant to Furman’s and Summers’s argument, it’s over 2 percentage points above the 16 percent on which they grounded much of their argument.
On the spending side in 2023, the CBO estimates, federal government outlays, which averaged 21.0 percent of GDP between 1993 and 2022, will be a whopping 24.2 percent of GDP. In short, the higher deficits are largely a result of spending increases, not tax cuts.
The problems with raising taxes
There are three main problems with raising tax rates. One is ethical; the two others are practical.
The ethical problem is that when the federal government raises tax rates, it uses force or the threat of force to take more of our money. For those who don’t think we have a right to that money, this is not a problem. But for those of us who do think we have a right to our money, it’s a huge problem.
Whatever your view on the ethics of raising tax rates, there are two practical problems. The first is that higher tax rates discourage productivity and, therefore, lead to less of an increase in government revenues than would be predicted based on a static analysis. The second is that it’s difficult politically to get revenues much above 18 percent of GDP.
Consider the first. Increased marginal tax rates reduce the incentive to earn income. This applies to very high-income people and even to people making an income of $100,000 or more. I’ll focus on high-income people because so many of the proposals by Democrats, the main advocates of tax increases at the federal level, are to increase tax rates mainly on high-income people. Consider a married couple with a current income of $700,000 who file their taxes jointly. They are now in a 37 percent federal income tax bracket. But that 37 percent substantially understates their marginal tax rate. If their incremental income is from working and they are self-employed, they also pay a Medicare tax rate of 2.9 percent; if they are employees, their Medicare tax rate (listed as “HI” on a pay stub) is 1.45 percent—the employer pays the other 1.45 percent.
We’re not done. If our high-income couple lives in a state with a high state income tax rate, it would pay a marginal tax rate of 10.3 percent (California), 8.97 percent (New Jersey), or New York (6.85 percent). For simplicity, let’s assume that both members of the couple are employees and live in New Jersey. The couple’s marginal tax rate would be 37 + 1.45 + 8.97 = 47.42 percent. So the couple’s incentive to make an additional $100 is not $100 but, rather, $52.48. Now imagine that the federal government raises this couple’s marginal tax rate 3 percentage points to 40 percent. The couple would then pay a marginal tax rate of 50.42 percent. Their incentive to earn an additional $100 would fall to $49.42, a drop of 6 percent. That’s not large, but it matters.
The other main alternative way for the federal government to raise money, of course, would be to raise marginal tax rates on income that is well below our hypothetical $700,000. Why do that? Because, as the famed bank robber Willie Sutton is said to have stated, when asked why he robbed banks, “That’s where the money is.” If the government were to, say, collapse the current 22 percent and 24 percent income tax brackets into one 24 percent bracket, it would raise more revenue from people in the 22 percent bracket and would also raise more revenue from people in higher tax brackets, because the income they made in the 22 percent bracket would now be taxed at 24 percent.
Moreover, it too would substantially reduce the incentive to earn income. Why? Because of Social Security taxes.
Consider a couple currently in the 22 percent bracket making a combined income of $150,000 annually. Imagine that the couple live in California. Both members of this hypothetical couple work, and imagine that one of them is self-employed. His current marginal tax rate is 22 + 15.3 + 9.3 = 46.6 percent. (The 15.3 percent is the sum of the tax rate for Medicare, 2.9 percent, and the tax rate for Social Security, 12.4 percent.) His incentive to earn an additional $100 is $53.30. Now if the federal government raises the 22 percent rate to 24 percent, his marginal tax rate rises to 48.6 percent and his incentive to earn an additional $100 falls from $53.30 to $51.30, a drop of 3.8 percent. This is almost two-thirds of the fall in incentive for our previously considered couple making $700,000 annually.
Because of the reduced incentive to make income, the economy’s output would be somewhat smaller than otherwise, making not only income taxes but also Social Security taxes and corporate income taxes somewhat smaller than otherwise. For these reasons, the federal government’s increase in tax rates would not yield the amount of revenue that a static analysis would predict.
The other practical problem with raising taxes is that overall federal tax revenues as a percent of GDP appear to be a political constant. Since the end of the Korean War in 1953, except during deep recessions, they have rarely dropped below 16 percent and have rarely increased beyond 19 percent of GDP. Indeed, when they do pierce the 19 percent ceiling, we get tax cuts, as with Ronald Reagan in 1981 and George W. Bush in 2001.
It's difficult to know why this is a political constant. My own view is that the majority of Americans still object to large government and they see taxes as the price of government. They really should see government spending as the main price of government, but they don’t see government spending on their pay stub. But we don’t necessarily need to know why this political constant exists to know that it does. So there’s a good case to be made that we are “stuck” with federal tax revenues being below 19 percent of GDP.
That leaves spending.
Spending
In the CBO report that I referenced earlier, the CBO’s analysts project government spending to be 24.2 percent of GDP in 2023, 24.4 percent in 2033, and 26.7 percent in 2043. It projects an even higher percent in later years but the further out we go, the less sure we can be. By 2033, the CBO predicts, the deficit will be 6.4 percent of GDP and the debt held by the public will be 115 percent of GDP. By 2043, those numbers, the CBO predicts, will be 8.1 percent of GDP and 144 percent of GDP, respectively.
A large part of the increase, not surprisingly, is projected to be due to increases in Social Security spending (from 5.1 percent of GDP in 2023 to 6.0 percent in 2033 to 6.2 percent in 2043) and in Medicare, net of offsetting receipts (from 3.1 percent of GDP in 2023 to 4.0 percent in 2033 to 5.1 percent in 2043.)
Discretionary spending is actually projected to fall, from 6.5 percent of GDP in 2023 to 5.6 percent in 2033 to 5.4 percent in 2043.
While cutting government spending as a percent of GDP is a tough problem politically, just as increasing tax revenues as a percent of GDP is a tough problem politically, it’s not hard to find ways to cut. In 2010, the New York Times introduced something on its website called “Budget Puzzle: You Fix the Budget.” It gave participants ways of reducing the deficit, both by increasing taxes and by reducing spending. I wrote about it here and my bottom line was that reducing the deficit without major tax increases except for one was easy. I wrote:
After I was done, the Times announced that I had solved the deficit. How did I do so? Entirely with budget cuts, with one exception. On spending, I took all the budget cuts offered, and the most radical version of each, except that I didn’t cut Social Security benefits for people with higher incomes. I probably should have. The only exception is that I did increase taxes by having the favorable tax treatment of employers’ contributions to employees’ health insurance phase out gradually.
Again, this is politically difficult. But the answer that there isn’t much room for budget cuts is completely wrong.
Conclusion
“May you live in interesting times.” That old saying is often said to have come from China, although there’s no evidence that it does. Whatever its source, the idea is that “interesting times” are times of trouble and uncertainty. With regard to US government deficits and debts, we live in interesting times. We could make them less interesting by beginning a process now of cutting the rate of growth of government spending.
