If we define our essential debt challenge not in absolute terms, but relative to GDP, then robust economic growth would cover a multitude of fiscal sins. Simply put, fast growth in the denominator of the debt-to-GDP ratio would solve the problem without the hard choices of spending less or raising taxes. This solution is alluring in its appeal, but, alas, easier said than done. Over the past 50 years, outstanding federal debt has grown at an annualized pace of 8.8%, implying that the economy would need to expand faster than that (in nominal terms) for growth alone to solve the problem. The economy has grown this fast for a quarter or two, almost always in a rebound from a recession, but never on a sustainable basis.

In the long run, economic growth is a function of two variables: growth in the labor force and productivity. Labor force growth is almost a fixed variable and has declined over the past few decades as birth rates have dropped. Other nations (China, notably) have tried to increase birth rates through policy, but to no avail. Policies of the current administration make it highly unlikely that immigration will help to boost labor force growth, so productivity and technology are all that is left.

History offers a curious lesson in the economic benefit of technology: The promise of productivity rarely translates into reality. Consider, for example, the introduction of computing technology into corporate America in the 1970s and 1980s. As businesses and individuals integrated computers into their professional and personal lives, the measurable impact on productivity was negligible, or even negative. The quarterly data shown in the nearby chart is noisy, but from the 1940s through the 1960s, productivity growth (measured as output per unit of labor input) grew at an average annual pace of 2.8%. As computers permeated the economy in the 1970s and 1980s productivity still grew, but at a more modest average pace of 1.7%.

This absence of measurable productivity gains prompted economist Robert Solow to wryly observe, “You can see the computer age everywhere but in the productivity statistics.” This so-called “productivity paradox” has been explained away as a delayed benefit (productivity growth averaged 2.1% in the 1990s), or a mismeasurement, but the historical record speaks volumes that technology is not, or at least has not been, the panacea that the evangelists of Silicon Valley would have us believe.

The promise of artificial intelligence may very well make these historical examples moot. It is possible that we are on the precipice of an unprecedented boom in productivity and therefore economic activity. It would be imprudent, nevertheless, to rely solely on this promise of outsized GDP growth to solve our debt problems. Hope is not a strategy.

Note that we have so far written only in nominal terms, ignoring (up until now) the effect of inflation on these debt, deficit, and economic dynamics. We’ve done that in order to focus specifically on inflation as a potential “solution” for our debt challenge. We employ scare quotes here to underscore that the inflation isn’t an appealing approach to addressing our growing debt burden, particularly for investors.

Regular readers will be familiar with the miracle of compound interest: how a small amount of money can compound, even at a modest rate of interest, into a great fortune. Inflation is the evil twin of compounding. Inflation can turn a great fortune, even at a modest rate of inflation, into a small amount of purchasing power. Inflation is therefore a debtor’s best friend, as it reduces the real liability of debt over time.

This is not to argue that inflation would ever become the explicit policy of the U.S. government in an effort to reduce the debt burden, but benign neglect can go a long way. Yet interest rates provide a self-correcting market mechanism that responds to rising inflation rather quickly. If lenders (bond investors) conclude that the government or the Federal Reserve is growing complacent about inflation, they will demand higher rates of interest to compensate for this rising risk. This ultimately leads to higher debt service and deeper deficits, thereby exacerbating the very problem that inflation was expected to address. The law of unintended consequences is not subject to repeal.

Modest inflation is likely to remain one factor restraining the growth in the real burden of debt, but the market will provide painful discipline through higher interest rates if inflation fears escalate too much.

Here we are, deep into this commentary, and we finally come to perhaps the most obvious solution: If our national fiscal problem is a function of overspending, maybe we should simply spend less. As with the other solutions covered herein, cutting spending is easier said than done.

Government outlays fall into three categories:

• Mandatory spending is determined by existing law and is not subject to annual budget appropriations. This category includes Social Security, Medicare, Medicaid, federal military and civilian retirement plans, and veterans’ disability benefits, among others. This is not to say that this spending can’t be changed, only that cutting this category requires a change of law distinct from the annual budget process. As the nearby graph illustrates, mandatory spending accounts for almost two-thirds of annual federal government outlays.

• Discretionary spending is easier to cut, as it is set by the annual budget process and does not require legislative change. This category includes defense spending, education, international affairs (such as USAID), government agency operating budgets, research and medical grants, community and regional development, and transportation. Discretionary spending represents 24% of total federal outlays.
• Interest on the national debt, the final category, is mandatory for all intents and purposes. In the last full fiscal year, debt service amounted to $882 billion, or 13% of total government spending. Notably, interest expense last year exceeded spending on national defense for the first time in our history.

The challenge is painfully obvious: So much of federal spending is mandatory that cuts are hard to find. About 75% of government spending is locked in even before the annual budget process starts. The four big spending categories of Social Security, Medicare, Medicaid, and defense accounted for 61% of all spending in 2024, and this has been true for decades.

Peter Fisher, former undersecretary of the Treasury, once remarked that the U.S., at least financially, resembles a giant insurance company with an army. There seems to be little political will to make serious cuts in any of these categories, as politicians would no doubt pay a price at the ballot box come election day.

The Department of Government Efficiency (DOGE) under the leadership of Elon Musk learned these difficulties firsthand earlier this year. According to the department’s own website, DOGE has identified $190 billion of savings as of June 29, 2025. This seems on first glance like a great deal of money. Leaving aside how much of this claim can actually be accomplished, this is equivalent to a mere 2.8% of the outlays detailed in the pie chart nearby. All else being equal, this would have reduced last year’s fiscal deficit from $1.8 trillion to $1.6 trillion. We should by all means identify and rectify waste, fraud, and abuse where it exists, but there is not enough to make more than a dent in the size of our debt and deficits.

The DOGE experience raises another general point. We human beings are not naturally conversant in such large numbers. Many of us deal in millions, particularly in matters of finance or business. Trafficking in billions is rare territory, and unless you are an astrophysicist or molecular biologist, you almost certainly don’t deal with measurements in the trillions. It is easy to think subjectively that all of these units of measurement represent “really big numbers,” but they are vastly, vastly different.

To illustrate the point, consider that a million seconds is about 12 days. A billion seconds ago, however, was 1993. A trillion seconds ago? The year 30,000 B.C. Saving millions – or even billions – in pursuit of a debt and deficit challenge measured in trillions is quixotic at best.

If spending less won’t get all the way toward our goal of fiscal prudence, maybe we should find ways to make more – that is, to increase the sources of federal income. Similar to what we saw with the analysis of government spending, a few categories account for the vast majority of receipts as well, which makes the task difficult.

By far the largest source of federal revenues is individual tax payments, followed closely by payroll tax payments into various social insurance programs (again, the dominance of Social Security, Medicare, and Medicaid in our national finances is evident). Corporate taxes amount to only 11% of total receipts, and all other sources of revenues (including customs duties, tariffs, and estate and gift taxes) are close to rounding errors. Note that the total for federal receipts in 2024 was $4.9 trillion, compared to the $6.7 trillion of spending detailed in the last few paragraphs. The difference of $1.8 trillion is last year’s deficit.

Let’s focus on the largest category of individual income taxes and see if we can close this $1.8 trillion gap through this revenue source alone. Our individual income tax structure is already quite progressive. The nearby graph illustrates the trend in effective (not statutory) tax rates by quintiles of household income, with the top 1% of households detailed in red. This data aggregates and averages individual income tax payments as a percentage of total income, so incorporates tax credits and deductions, but does not include payroll taxes. Note that the data extends only through the 2019 tax year, as the tax benefits associated with the pandemic wreak havoc with the data starting in 2020.

More than 40% of households (the bottom two quintiles in the nearby graph) have negative effective tax rates, due largely to benefits such as the earned-income tax credit and the child tax credit. The top quintile of households pays an average individual income tax of 15.3%, and the top 1% of households (making more than about $800,000 per year, on average) pay an average rate of 23.3%.

Here again a simple calculation illustrates the intractability of the problem. The top 1% of U.S. households comprise roughly 1.3 million taxpayers, with average annual income of about $780,000. If we were to tax all of this income at an effective (not marginal!) rate of 100%, it would generate revenues of a little over $1 trillion, leaving the deficit $800 billion in the hole. This is, of course, a Swiftian proposal. No one is proposing a tax hike to 100%, or, indeed, tax hikes at all given the recent passage of the OBBB.

Placing the U.S. tax experience into a global context offers an interesting avenue of inquiry. Comparing tax burdens across borders is difficult, because there are as many different tax regimes on the planet as there are taxing authorities, but at the highest level – and taking into account all forms of taxation, not just individual income taxes – the tax burden in the U.S. relative to GDP is relatively modest.

This graph won’t surprise most readers. Europe (and the Nordic countries, in particular) is known for imposing higher taxes on just about everything. Tax revenues to GDP in Canada are almost 10 percentage points higher than in the U.S., and in the nearby graph, only Ireland and Mexico have lighter tax burdens relative to the size of the economy than we do in the U.S. (For the record, Turkey, Chile, Colombia, and Costa Rica also have lower tax burdens than the U.S. relative to GDP.)

There is a simple reason for this apparent disparity in tax burdens. Every country on this graph except the U.S. has a national value-added tax (VAT) or a goods and services tax applied to consumption. Indeed, 175 countries and over 90% of the world’s population live with a national consumption tax. The U.S. is an outlier.

Herein lies an intriguing thought experiment. Personal consumption is the largest driver of economic activity in the U.S., amounting to more than $20 trillion over the past year. A 10% VAT applied to this level of consumption would generate around $2 trillion of tax revenue, all else being equal, more than covering our current level of deficit. Furthermore, a national consumption tax at this level would still leave us far below similar consumption taxes applied elsewhere in the world.

As with so much else in this article, this is easier said than done. The math and economics work all day long, but politics is a different matter. Many states and cities already have sales or use taxes. Would a national tax supplant those existing taxes or be incremental? The cost of compliance is meaningful at the point of sale, although other countries have figured it out, and technology should mitigate this challenge. From a social standpoint, perhaps the biggest obstacle is that consumption taxes are regressive: Poorer households bear a greater burden relative to income than wealthier households.

Yet the point remains: Leaving aside the societal issues of a more regressive tax regime, if we need to substantially raise revenues, a broader tax base such as that offered by a consumption tax fits the bill.

None of the challenges outlined in these pages is new. As long ago as August 2011, Standard & Poor’s referenced them in lowering the U.S. sovereign debt rating to AA+ from AAA: “Our opinion is that elected officials remain wary of tackling the structural issues required to effectively address the rising U.S. public debt burden in a manner consistent with a ‘AAA’ rating and with ‘AAA’ rated sovereign peers.”² This downgrade took place almost 15 years ago, in the wake of the GFC and as Congress played one of its regular games of economic chicken with the debt ceiling.

It took some time, but Fitch followed suit with a rating downgrade from AAA to AA+ in August 2023, and Moody’s rounded out the trifecta by lowering its rating of the U.S. from Aaa to Aa1 in May 2025. Moody’s noted that the downgrade “reflects the increase over more than a decade in government debt and interest payment ratios to levels that are significantly higher than similarly rated sovereigns.”³ The press release went on to observe that:

Nothing in these downgrades speaks explicitly of default, and the U.S. remains one of the highest-rated sovereign debt issuers on the planet, bested only by Canada, Denmark, Germany, the Netherlands, and Switzerland. The real import of these downgrades is to sound the alarm that, unless addressed, debt and deficits can rise to the point that default does become more likely, and that somewhere out there is a tipping point at which the options outlined in this piece are no longer available. At that undefined point, some sort of default becomes inevitable.

Financial markets provide some transparency into the waning and waxing of sovereign risk through credit default swaps (CDS), which allow investors to insure against the possibility of default. Swap spreads rise with anxiety, and trends over time illustrate aggregate market sentiment toward individual issuers. In the nearby table, the figures represent in basis points (bps)⁴ how much it costs to protect against the default of a sovereign issuer. A simple way to read the table is that the number in the second column represents how many dollars it currently costs to hedge $1 million of underlying value. For example, to hedge a $1 million five-year U.S. government bond costs a little over $40 per year at present.

CDS in the U.S. have ranged from a low of about 10 bps to a high of around 80 bps over the past few decades, indicating that anxiety about U.S. sovereign credit rises and wanes from time to time, but remains at a generally low level.

We typically associate sovereign defaults with spectacular economic collapse, social unrest, and political upheaval, but default need not be binary. There are variations on the theme, and rising default risk in the U.S. will more likely present itself as rising inflation and a weaker dollar. Inflation is itself arguably a soft form of default, albeit not in a contractual sense. Repaying debts with devalued purchasing power might not violate the legal terms of a bond contract, but it does seem to violate some sort of social contract.

There are examples throughout history of stronger forms of default, such as forced swaps. Greece in 2012, Argentina in 2005, and Ecuador in 2008 all forced bond holders to exchange their current bond holdings for newly issued debt with lower par values, lower interest rates, or longer maturities. Any of these changes represents a unilateral change to the terms of a contract. Of course, owners don’t have to accept a swap offer, but they are faced with the legal and opportunity costs of lengthy court battles.

We do not believe that the U.S. will ever explicitly default on its debt obligations, even though the previously unthinkable and unspeakable is spoken about a bit more often these days. It is hard to be forced into default when the U.S. enjoys the exorbitant privilege of having a monopoly on the production of the global reserve currency. Phrased differently, it’s easy to stay in business when you can always print more of the currency you need to service debt.

And yet the mere suggestion of default – whether soft or hard – would create enormous volatility in financial markets. Investors would quickly incorporate such risks into interest rates, with concomitant effects throughout the real economy as well. Complacency is the enemy of the investor, and change can happen quickly. In Ernest Hemingway’s “The Sun Also Rises,” a down-on-his-luck character is asked how he managed to wind up bankrupt. “Two ways,” he replies, “gradually, then suddenly.”⁵ Markets can remain complacent for a long time and then be quickly shocked into a new reality. There are decades when nothing happens, and there are weeks when decades happen.