Following up on our earlier post discussing the need for a small, simple set of policy principles that every new policy must meet, our first proposed principle relates to government debt: Except to address an emergency, no policy shall be adopted that increases the ratio of the national debt held by the public (which we refer to as “public debt”) to GDP.
Causes of the current debt
In 2001, the federal government had a budget surplus of $127 billion, had $3.6 trillion of public debt, had a public debt-to-GDP ratio of 32 percent, and was projected effectively to pay off the public debt by 2009. However, it has run a deficit every year since 2001, ranging from $158 billion in 2002 to $3.13 trillion in 2020. As a result, the public debt has increased to $30.7 trillion, which is about 100 percent of GDP. The public debt does not include debt the federal government owes to itself; that intragovernmental debt is included in the gross national debt, which is about $38.7 trillion.
The Committee for a Responsible Federal Budget (CRFB) analyzed the causes of the increase from the projection of no debt by 2009 to the reality of a public-debt-to-GDP ratio of 98 percent in 2023. It concluded that tax cuts were responsible for 37 percent of that 98 percent, net discretionary spending increases and Medicare expansions were responsible for 33 percent, and responses to the 2007 recession and the 2020 pandemic were responsible for the remaining 28 percent.
The CRFB further explained that tax cuts cost about $8.3 trillion, with about $6.5 trillion due to policies that were somewhat bipartisan and $1.8 trillion due to Republican policies. Net discretionary spending increases cost about $5.9 trillion, almost entirely due to bipartisan policies. Medicare expansions cost about $1.1 trillion, of which $900 billion was due to bipartisan policies and $200 billion was due to Democratic policies. And responses to the recession and pandemic cost about $6.8 trillion, with about $4.8 trillion due to bipartisan policies and $2 trillion due to Democratic policies.
Overall, bipartisan policies were responsible for about $18.1 trillion of the $26 trillion of public debt in 2023, Republican policies were responsible for about $1.8 trillion, and Democratic policies were responsible for about $2.2 trillion. So, both parties share almost equal blame for the current debt. The federal government has had various pay-as-you-go, or PAYGO, rules in effect since 1990 that require tax cuts or spending increases to be offset by tax increases or spending cuts. However, those rules have been ineffective in limiting the national debt. First, PAYGO rules apply only to mandatory spending increases and not to discretionary spending increases through the appropriations process. Second, many programs are exempt from PAYGO rules, including Social Security, the postal service, Medicaid, the Supplemental Nutrition Assistance Program, and unemployment insurance. Third, Congress has never actually enforced the PAYGO rules, but instead has always waived or delayed their application to large tax cut and spending increase policies.
Concerns about the current debt level
Polls show that most Americans are concerned about the national debt. The Peter G. Peterson Foundation conducts surveys every month to calculate a fiscal confidence index. In December 2025, 77 percent of respondents said their concerns about the debt had increased in the last few years, 55 percent said that the country was heading in the wrong direction in addressing the debt, and 79 percent said addressing the debt should be one of the government’s top three priorities. However, 58 percent of respondents said they expect the debt problem to get worse over the next few years and only 47 percent were optimistic that the country would make progress on the debt over the next few years.
While most economists are concerned about the debt, they disagree about how urgent the problem is and about the consequences of maintaining or increasing current debt levels. The largest concern is that, at some point, bond purchasers may worry about the government’s liquidity and demand higher interest rates, which could exacerbate liquidity concerns and lead to a debt spiral. But some argue that the United States’ unique position in the world economy will prevent that result, because there will always be demand for dollar-denominated debts and confidence in the federal government to pay those debts. The difficulty is that the current situation is essentially unprecedented. Economists used to believe that a country’s debt should not exceed 90 percent of its GDP. But the U.S. debt now exceeds that ratio and Japan’s debt has grown to almost 250 percent of its GDP without serious consequences. It is impossible to predict the consequences of events that have never occurred before. Because of this uncertainty, caution is warranted and the current 100 percent public debt-to-GDP ratio should not be exceeded, except in an emergency.
What is a policy?
One important issue would be deciding what constitutes a policy for purposes of this principle. We think it’s important that every policy pays for itself; it should not be paid for through offsets in an unrelated policy, as would be the case under PAYGO if it were ever enforced. For example, the 2017 Tax Cuts and Jobs Act had a net cost of about $1.4 trillion. The most expensive changes were the reductions in the individual and corporate tax rates, the increase in the standard deduction, and the increase in the child tax credit. The policy offset about half of the lost revenue by repealing some personal exemptions, itemized deductions, and business credits and deductions. To satisfy the principle that the policy should not increase the debt, it would have needed to include additional or larger offsets and/or smaller reductions in revenues.
What is an emergency?
Another issue would be deciding what constitutes an emergency for purposes of this policy. Declaring an emergency should require Congressional and Presidential approval. The 2007 recession and 2020 pandemic were clearly emergencies. There was no emergency in 2017 that required the Tax Cuts and Jobs Act, which cost about $1.4 trillion, or in 2025 that required the One Big Beautiful Bill Act, which cost about $2.8 trillion. In order to avoid purely partisan declarations, declaring an emergency should require at least the typical 60-vote supermajority in the Senate and possibly a similar supermajority in the House.
How should a policy’s effects on the debt be determined?
Another issue would be deciding how to estimate the effects of any particular policy on the debt. The Congressional Budget Office (CBO) and the Joint Committee on Taxation (JCT) often provide both a conventional estimate, which considers only the direct impacts of the policy, and a dynamic estimate, which also considers secondary impacts on employment, gross domestic product, and other macroeconomic measures. As an example, the conventional estimate of the cost of the 2017 Tax Cuts and Jobs Act was about $1.9 trillion, while the dynamic estimate of its cost was about $1.4 trillion. We think it’s preferable to use a dynamic estimate, as long as that estimate is prepared by CBO and/or JCT. For the One Big Beautiful Bill Act, the CBO estimated a $2.4 trillion debt increase with a conventional score and a $2.8 trillion debt increase with a dynamic score. However, the Trump administration claimed that the policy would actually reduce deficits by $1.4 trillion in dynamic scoring, by making unrealistic projections about the policy’s effects on GDP. Dynamic scoring is preferable, because it includes real world effects. But estimates of those effects are heavily dependent on assumptions, so it’s important to consider estimates only from the official, nonpartisan sources.