This is a speculative post on how to balance the federal budget. Since I would like to see a return to SC culture, the maintenance of which requires a balanced budget, this is topic than needs to be addressed. Obviously, there are political issues in doing this that I am going to ignore in this post. There are also economic issues that I will briefly discuss.

Historical budget balance
I start with Figure 1 showing a plot of Federal revenue and spending since WW II. Two things stand out. One is that revenue was pretty flat for fifty years over 1952-2001, averaging 17.2% of GDP. Since 2001 it has been significantly lower (p < 0.0025) with an average value of 16.4% of GDP. Spending, in contrast, has been much more variable and shows a distinct rising trend.

Figure 1. Government revenue and spending since WW II as a percent of GDP
Values exponentially smoothed with 0.5 weighting.

It is clear that spending has been higher than revenue since the late 1950’s in all but four years. The chart also shows that 25 years ago we actually ran a surplus. This was due to tax increases passed under Presidents Bush-41 and Clinton that raised revenue from 17.1% over 1982-92 to 19.0% over 1996-2001 and spending reductions from 21.2% to 18.0%. Three-quarters of the spending reduction came from cuts in defense spending from 5.4% to 3.0% of GDP, the so-called “peace dividend” following the end of the Cold War in 1991.

Leaders of both parties added a great deal of non-defense spending in the new millennium. New entitlements such as the Medicare Drug benefit passed under Bush-43 and the Affordable Care Act passed under Obama sent spending to 18.9% over 2003-2007 and 20.3% over 2013-19. Stimulus to deal with the 2008 crisis temporarily sent spending to much higher levels (22.5% over 2008-12). The same thing happened for the pandemic stimulus over 2020-3 during which spending averaged 26.4%.

Balancing the budget (in a technical sense): Revenue side
Looking at Figure 1 and the summary statistics provided in the previous section, it seems logical to start with the 1994-2001 period when we actually had a balanced budget. To get back to that time we need to reverse the effects of the tax cuts passed over 1997-2017. That is, we need to return to a 28% long-term capital gains tax, a corporate tax rate of 35%, a top income tax rate of 39.4%, and taxing dividends as ordinary income as was the case before 2003. I would also eliminate the carried interest loophole and define long-term for capital gains as five years instead of one year. Capital gains held shorter than 5 years would be taxed as ordinary income. The expectation is this would bring revenue levels to the 18.6% seem during the 1990’s.

Government spending
As mentioned earlier the large decline in spending in the 1990’s was largely the result of reductions in defense spending after the Cold War. Defense spending is relatively easy to cut because the spending largely does not directly benefit voters as programs like Social Security do. Defense, at 2.9% of GDP in 2023, cannot really be cut further given the administration’s decision to convert former allies into adversaries.

Social Insurance
Entitlements make up the lion’s share of Federal spending. These are Congressionally mandated benefits that go to all who qualify, regardless of their effect on the budget. They can be broken into two categories. One is social insurance, Old Age Social Insurance (Social Security) and health insurance for seniors (Medicare). These benefits go to all eligible Americans regardless of income. Because of their insurance function they are funded by a flat tax on wage and salary income as specified under the Federal Insurance Contributions Act (FICA), which may be thought of as the social insurance premium.

Figure 2 shows a plot of social insurance spending as a percent of GDP. Since 1966 spending for Social Security has increased by 2.4% of GDP largely due to rising life expectancy, falling birth rate, and a declining fraction of GDP that is paid out as wages and salaries (see Figure 2). Spending on Medicare has risen by 3.5% of GDP due to the rising cost of health care. Also shown is spending less the revenues from FICA taxes. Until 2002, FICA taxes covered social insurance despite the declining wage and salary tax base. This was the result of significant increases in the tax rate levied. The Bush-43 administration enacted new Medicare benefits, the Part D drug benefit and the Medicare Advantage program, but did not provide for any new revenue to fund them, putting the social insurance program into the red.

Figure 2. Social insurance spending before and after subtracting FICA revenue

One way to address this deficit is to expand the size of the income subject to FICA tax. In my book America in Crisis I provide an estimate of the amount that could be raised by extending the 12.6% payroll tax that funds Social Security to all earned income. My calculations suggested that doing this would raise about $0.36 trillion in annual revenue in 2021, or about 1.5% of GDP. This would close about half of the deficit. The 7.65% FICA tax levied on employee wage and salary income is currently not applied to unearned income such as rent, dividends, interest and capital gains. Were it extended to unearned as well as earned income, this would raise additional revenue equal to 1.9% of GDP, closing the gap and leaving a 0.4% surplus.

Share

Economic Stabilizers: Unemployment insurance and SNAP
Unemployment insurance and SNAP benefits (Food Stamps) fall into a special class of entitlement programs. Spending for these programs is cyclical (see Figure 3), rising during recessions and falling during expansions. By doing this they apply an automatic economic stimulus of about 0.9% of GDP during downturns which diminishes as the economy recovers. Also shown is a plot of economic volatility.1 Leaving out the Great Depression and WW II, the average volatility before the Depression was 5.4% while that after WW II was 2.2%. This finding suggests that these entitlements provide a systemic benefit even to Americans who never receive them. They can be considered as a form of economic security analogous to the physical security provided by national defense. As such it serves a core purpose of government and is a fully justified expense.

Figure 3. Counter-cyclical stimulus appears to have dampened economic fluctuations

Aid to the poor: Medicaid and income support programs like Earned Income Credit
Medicaid is the third major government benefit program. Unlike social insurance, eligibility is based on need; higher income Americans likely will never qualify for Medicaid benefits. It is funded from general revenues and not from a special fee like FICA that functions as a social insurance premium. The Earned Income Credit (EIC) program is another anti-poverty program that provides a tax credit to low-income Americans that is based on employment income. It serves as both a support for the poor and an incentive for working and has traditionally enjoyed bipartisan support. It is similar to corporations who offer matching funds for employee charitable donations as an incentive for employee giving (which is seen as good PR). Figure 4 shows a plot of spending for Medicaid+EIC, which shows it has steadily risen over time due to burgeoning health care costs and an increasing fraction of the population who qualify for benefits (reflecting rising income inequality).

National Defense and Interest
Figure 4 also shows a generally declining trend in defense spending since the Korean War. Spending remained high for the rest of the fifties and most of the sixties, declining with the gradual pullout from Vietnam after 1968 and continuing on after the 1973 end of the war. A second reduction in defense spending occurred following the end of the Cold War in 1991. Between these two was a rise for the Reagan buildup and afterward a smaller bump up for the War on Terror. Figure 4 also shows interest expenses. These rose dramatically in the 1980’s as a result of Reagan administration deficit spending in the face of extremely high interest rates and came down in the 1990’s with the elimination of the deficit and falling rates. In recent years this expense has again been rising because of a quarter century of deficit spending (which continues) and rising interest rates.

Figure 4. Medicaid+EIC, defense, interest, other spending and total spending less social insurance compared to non-FICA revenue

Other spending
The rest of Federal spending not discussed falls into the catchall term of Other Spending which also appears in Figure 4. This category (and defense) has been where budget conscious administrations (including the present one) have relied upon for spending reductions. There have been sporadic peaks in spending reflecting the changing priorities of different administrations. The first such feature was a ramp up in social welfare spending during the late 1960’s and 1970’s labeled as the War on Poverty. On top of this came programs implemented as a result of the Oil Crisis. Spending priorities changed with the conservative Reagan administration. The impetus for poverty-reduction programs dissipated and the threat of the oil crisis receded with the 40% decline in the price of gas over 1981-86 resulting in a decline in this spending category.

The next boom in spending in this category was the stimulus programs following the 2008 financial crisis. These lasted for about five years, after which spending returned to pre-crisis levels. There was another, larger boom in spending for the pandemic. This one was shorter, but ended with higher spending than before the pandemic. Significant amounts of both the pandemic stimulus and the 2021 infrastructure bill have not yet been spent and may be part of the higher spending in this category. More ominous is the recent rise in spending on interest payments, which is the chief reason why spending aside from social insurance (red line in Figure 4) is considerably higher post-pandemic. The principal cause of this is the higher interest rates resulting from Fed actions taken to address inflation in the wake of the pandemic.

A controversial way to address interest rate cost
The way the US government obtains the money it spends is by collecting taxes and issuing bonds to investors to make up any shortfall of revenue relative to spending (deficit). Over the past fifteen years we have seen a phenomenon called QE, in which the Fed buys debt securities, including government bonds. That is, it finances the government. The interest on these bonds is remitted to the Treasury. Thus, were the government to sell bonds directly to the Fed instead of investors to fund its deficits, it would get all the interest it paid remitted back to itself (and so would pay no interest on that debt).

A more direct way to do this would be to do away with bonds altogether. The Constitution empowers the government to coin money. The government can coin $1 trillion platinum coins and deposit them at the Fed, who then credits the government’s account with $1 trillion in Federal Reserve Notes (more commonly known as dollars). The Constitution empowers the government to mint coins; cannot issue bank notes (dollars are Federal Reserve Bank Notes). It must first mint coins and convert them into dollars at the Fed before spending them. Doing this over time will result in a reduction of total debt and associated interest expenses. Elimination of interest expenses would reduce spending by 3.0% of GDP. With the 0.4 percentage points of savings from the FICA modifications these two would completely offset the deficit (difference between the black and red lines in Figure 4).

Doing this is called monetization of the debt. It is what QE is, and partly how we paid for WW II. The alternative to monetization is paying down debt. We did that after the Civil War, the result was very high real interest rates that made US growth in the late 19^th century anemic compared to other countries going through industrialization, or even what the US achieved in the mid-twentieth century when real interest rates were much lower. The US and other Western nations attempted to do the same after WW I and the result was the Great Depression. Paying down debt does not seem to be an attractive option.

The big drawback to debt monetization is inflation. Historically debt monetization, whether from central bank money creation or issuing excessive coinage has led to inflation. Since debt monetization is only needed to finance deficit spending, going to direct funding of deficits with platinum coins would make the relation between deficit spending and money creation explicit. It is easier for voters to grasp how increasing the amount of money relative to stuff to buy threatens inflation making voters more likely to punish deficit spending.

Economic impact of these changes
Most of the tax increases discussed above were implemented over long stretches of time during the last century with reasonably good economic results. The big change that does not have precedent is debt monetization. I should stress that the program I suggest here would be implemented during a time of serious economic downturn and would be combined with substantial stimulus spending, which would be funded through debt monetization. The shift to minting money rather than borrowing would end the creation of new government bonds, while mature bonds would be paid off with new-minted money, resulting a trend towards lower interest rates. At first this would have little economic impact as interest rates would already be depressed due to the economic downturn. But as the economy recovered interest rates on government debt would not rise. This would make the US a less attractive place for foreign holders of dollars to stash their dollars, resulting in a fall in the value of the dollar. This would make imports more expensive in terms of dollars and exports cheaper in foreign currencies. That is, minting money could have effects similar to a tariff. It would make America poorer (US GDP would decline relative to other countries due to lower dollar value). Goods and services would become more expensive depending on the fraction of their content that is imported.

This policy would also have distributional effects. A cheaper dollar could incent foreign businesses to invest in the US to produce goods for export and for the domestic market. This would likely mean more manufacturing jobs and rising wages for factory workers and for low-wage jobs in general. On the other hand, there would be little foreign investment in professional services such as medical services, since these cannot be exported and US costs are uncompetitive compared to the rest of the world, even with cheaper dollars. Thus, the wages of people who make stuff would rise relative to providers of professional services. Thus, while the cost of stuff would rise along with the incomes of ordinary people, the cost of services such as health care and education would fall. The effect of higher tax rates on investment income and falling dollar would retard recovery of financial valuations, possibly making real estate less expensive. All told, it seems there would be some reduction in economic inequality as a result of these changes.

Another issue is the effect on Fed policy of a disappearing federal debt. Without any government debt the Fed cannot use its normal interest rate policy tools to manage the economy. The Fed has historically been able to do this with all sorts of debt levels down to less than 20% of GDP. Perhaps maintaining a low level of government debt amounting to only a few percent of GDP would be sufficient for economic policy. On the other hand, such a small government bond market could lack the heft to influence dynamics on the broader markets. Perhaps commercial paper and AAA corporate bonds would take the place of T-bill and Treasuries as reference standards for financial markets and Fed policy would involve interventions in these markets. How this might be worked out is unclear, but a way will be found.