Some of the ways I feel myself aging are obvious. Half of my hairs are white now. I can’t stay up past midnight. I hate loud restaurants and bars. I feel a deep urge to read long biographies of generals.
But other signs of aging are subtler. For example: I am now worried about the US budget deficit.
Coming of age as an economics reporter in the 2010s, there was no clearer generational divide than the one about government debt. Older reporters, who came of age during the exploding deficits of the 1980s and the balanced budget battles of the 1990s, tended to view deficits as obviously bad, and any cases for deficit spending — federal spending financed by borrowed money, rather than taxes — as the excuses of cowardly politicians.
Younger folks like me distinguished themselves by embracing deficits. The job market in the US after the 2008 financial crisis was a disaster. Stimulus from Congress and the Fed was too small to fix it. It was madness to worry about cutting the deficit when more spending or tax cuts or both could achieve the much more important goal of helping people get back to work. And it was especially stupid when interest rates on the debt were at record lows, which meant that taking on debt was cheaper than ever. We thought it was obsession with the debt, not the debt itself, that was destroying the economy.
I still think we were right about the 2011 economy. But the 2024 economy is not the 2011 economy. Interest rates are much higher. Unemployment is at 4 percent, a number that in 2011 we didn’t dare dream of. Wages are rising, especially for the lowest-income workers. We had the first major inflation episode in 40 years, though it finally appears to be subsiding. And the national debt is climbing higher and higher.
The most commonly cited metric has the debt load nearing 100 percent of US GDP, or $27.5 trillion. That’s not the best way to measure the burden in my view, but no matter how you look at things, the problem is significant and getting worse.
There’s no magic number at which the debt load becomes a full-on crisis. But it steadily becomes a bigger and bigger problem, and the trajectory we’re on is worrisome. It’s especially worth taking debt more seriously when other problems that deficit spending can help solve, like mass unemployment, have been fixed. Moreover, without tackling the debt problem, tackling other problems, from child poverty to housing costs to climate change, will become harder as the government has less space to spend and invest.
Here’s how I learned to start worrying and fear the debt, and why you might want to as well.
The US debt total is reported in many ways, but the most accurate, in my opinion, is a ratio: debt held by private investors as a share of gross domestic product, net of government assets.
“Held by private investors” is an important caveat: A huge share of the US debt is held by the government itself, in vehicles like the trust funds of Social Security and Medicare. Still more was purchased by the Federal Reserve as part of its “quantitative easing” programs to fight the Great Recession and the Covid downturn.
Thus, it’s money that the federal government owes to itself. That makes it fairly unimportant, economically; it doesn’t actually limit the resources available to the government. The debt that poses the biggest worry is debt owed to non-governmental investors, who can demand higher interest rates in the future.
“Share of gross domestic product” is also an important adjustment. Dividing by the size of the economy puts the debt burden in the context of the US’s resources. Just as a big mortgage can make sense if you have a high income to pay for it, a high debt load is more manageable if the US as a whole is earning enough money to finance it. It also points to an often-neglected way to reduce the budget deficit: passing economic growth-enhancing policies, like expanded immigration by STEM workers and would-be entrepreneurs or expanding support for science.
“Net of assets” means adjusting for the considerable resources the government owns, like land and loans that it made to other people (like student loan recipients).
As of this writing, debt held by private investors net of assets represents 75.7 percent of GDP; without subtracting assets, it’s 91.2 percent. That’s not full-on debt crisis levels, but it’s high. The figure before adjusting for assets (for which data is more readily available) is the highest it’s ever been since the Treasury started keeping track in 1970. It’s much higher than it was before Covid (66.4 percent) and much, much higher than before the global financial crisis (under 30 percent).
The cost this debt imposes can be measured most directly by the interest the US pays on its debt. The Congressional Budget Office projects that the amount the US will spend on debt interest will hit 3.1 percent of GDP this year. That’s a serious expense: It’s close to what we spend on defense every year.
More worrisome, the CBO projects that the debt and interest burdens are set to increase substantially. Debt held by private investors net of assets is set to rise from 75.7 percent this year to 93.7 percent in 10 years. The agency doesn’t project that figure out longer, but does provide projections for debt held by the public, which includes Federal Reserve holdings and doesn’t subtract assets. That’s set to grow from 99 percent in 2024 to 166 percent by 2054.
What’s driving this trend? It’s not defense spending or “nondefense discretionary,” a category encompassing everything from the FBI to the National Institutes of Health. The CBO projects that discretionary spending, both defense and nondefense, will fall to historic lows as a share of the economy over the next 30 years (though this is a projection I’m pretty skeptical about, especially as the US urge to counter China militarily grows). It also sees tax revenues modestly increasing over the next 30 years, due in part to the expiration of many Trump tax cuts next year and because household incomes tend to grow faster than inflation, which means the income tax raises more money over time.
Congressional Budget Office
What’s growing is spending on Social Security, Medicare, and Medicaid. As you can see in the above chart, the CBO sees spending on “major health care programs” (including Medicare, Medicaid, and Obamacare subsidies) growing from 6.3 percent of GDP in 2024 to 9.8 percent of GDP in 2054; Social Security is set to grow from 5.2 percent of GDP in 2024 to 5.9 percent of GDP in 2054.
Two major factors are, in turn, driving higher spending on these programs. One is that as baby boomers age the US population is getting older, meaning more people are benefiting from Social Security and Medicare and fewer are paying in. This adds 2.4 percent of GDP to federal spending by 2054, half from health care and half from Social Security. The second driver is that health care costs are projected to grow faster than the economy, independent of the aging population; this adds 2.6 percent of GDP to the federal budget by 2054.
Liberal and conservative budget experts love to argue about whether low taxes or high spending is driving the debt problem. Conservatives will point out that tax revenues are set to rise as a share of GDP over the next thirty years, as spending on programs for the aged explodes, so how could taxes rather than spending be the problem? Liberals will point out that if the US had not adopted the Bush tax cuts of 2001 and 2003, and the Trump tax cuts of 2017, revenue would be higher now and set to rise even more, and we would not have a debt problem at all.
I find this debate somewhat exhausting. It is trivially true that you can fill the budget gap by slashing spending, and it is also trivially true that you can fill it by raising taxes. Which you prefer has less to do with the historic trends in either than it does with your political values.
What matters is that spending on older Americans is set to spike, and we need to figure out a way either to tame that spending, raise taxes to pay for it, or do a combination of the two.
“How did you go bankrupt?” Bill Gorton asks Mike Campbell in Hemingway’s The Sun Also Rises. “Two ways,” Mike replies. “Gradually and then suddenly.”
This line is a bit of a cliché in discussions of government debt crises, for good reason. A heavy debt burden can impose at least two different kinds of costs. One, it can slowly erode economic growth. Two, it can in extreme cases lead to interest rates spiraling ever-upward because lenders no longer trust that they will be paid back, putting the borrowing nation in a crisis that can end in runaway inflation, default, recession, or all of the above. The risk of either is why considering deficit reduction now is probably a good idea.
The chronic costs of debt are a bit easier to think about, and may be familiar from an intro macroeconomics class. Budget deficits increase demand for loans, because the government needs loans on top of all the loans that private individuals and businesses are demanding. A surge in demand for loans makes loans more expensive: The average interest charged goes up.
For the government, this is an additional expense it has to incur. But the higher interest rate applies to private companies and individuals too. And that means fewer families taking out mortgages and student loans, fewer businesses taking out loans to build new factories, and just generally slower economic growth from constrained credit. This is called “crowding out.”
How big of a problem is this? It’s hard to say, because it’s very difficult to study empirically. The relationship between debt and growth goes both ways: Debt can slow economic growth, but slow economic growth can also cause countries to take on more debt, as when recessions force countries to spend billions or trillions on stimulus packages. Some factors can influence both growth and debt simultaneously: aging countries often take on more debt as pension costs rise, but also grow more slowly because less of the population is working. Economists try to work around these issues but there’s only so much you can do with statistics to disentangle them.
Here’s what we do know: a wide range of empirical evidence shows that when the US does more deficit spending, this pushes up interest rates for US government debt, which in turn pushes up interest rates throughout the economy. In the long run, this discourages business investment and hurts economic growth. The Congressional Budget Office estimates that each dollar of the deficit means 33 cents less in private-sector investment.
Of course, the economic damage depends on how that dollar of deficit is spent. And conversely, how effective deficit reduction is at promoting growth depends on the details of how it’s done. University of Pennsylvania economist Kent Smetters and the team at the Penn Wharton Budget Model recently evaluated three massive deficit reduction packages, using a mix of tax hikes and spending cuts. Sizable deficit reduction, they concluded, boosts GDP by as much as 9.8 percent by 2054, and raises wages by as much as 7.5 percent compared to the status quo where debt keeps rising. The exact numbers depend on what specific taxes you’re hiking and spending you’re cutting. These packages at best stabilize the debt; reducing it would do still more.
Crowding out is the chronic cost of debt. The acute cost is what happens if debt accumulates to such a point that investors start demanding ever-higher interest rates to service it, forcing the US into a situation where it uses the Federal Reserve to inflate away the debt by buying up bonds directly, drastically raises taxes or cuts spending to service the debt, or defaults and states it will not pay all that it owes. It’s easy to come up with foreign examples of debt crises that got to this point: Mexico in 1982, Russia in 1998, Argentina in 2001, Greece in 2015. Near-misses that didn’t quite lead to default, like Spain and Italy in the 2010s or Indonesia, Thailand, and South Korea in 1997, are even more common. If crowding out is like heart disease, this is a sudden heart attack.
The US is, of course, much richer than all those countries, and its debt plays a much more pivotal role in the world financial system. The US controls the “reserve currency”: About three-fifths of all assets in central banks around the world are denominated in dollars, meaning the whole world relies on it to do business. The dollar is used for loans and transactions everywhere, even in transactions having nothing to do with the US.
This means lots of people around the world want to buy US government debt, which has the effect of pushing interest rates down. It also means that we can always sell debt in our own currency, an ability many other countries don’t have. In the 1960s, the French finance minister, later president, Valéry Giscard d’Estaing famously called this America’s “exorbitant privilege.”
Does this privilege mean the US is immune to the dynamics that have driven debt crises in other places? Perhaps. There’s nothing in the historical record to rule out “debt crises are impossible in the US” for the simple reason that we haven’t had one in modern memory. Then again, there is no law of the universe that the dollar has to be the world reserve currency forever, and while its replacement by the Euro or the Chinese renminbi or some other rival has been erroneously predicted many times in the past, that does not mean such predictions will always be wrong.
If US debt reaches extremely high levels, such that bonds in a rival currency start to look like a safer choice, a shift could occur, which in turn would send US interest rates higher.
Smetters and his colleague Jagadeesh Gokhale have after extensive modeling concluded that if US debt held by the public exceeds 200 percent of GDP, we’ll be entering a phase where no amount of tax hikes or spending cuts will be sufficient to avoid default. We’ll either have to explicitly say we will not pay back all our debt, or else inflate it away, both of which would have massive negative consequences for American workers and consumers. They estimate that the US will reach this point in about 20 years.
You don’t have to believe that prediction is literally true to find it incredibly alarming. As Covid and the global financial crisis should have reminded us, low-probability/high-consequence shocks do happen. And continuing to add more debt indefinitely will certainly raise the probability of this kind of disaster.
So, the debt is probably a problem worth worrying about. What, then, can we do about it?
Recall that the two big drivers the CBO identified are population aging and health care costs. Aging is partially addressable with immigration. For instance, the US is currently in the midst of a surge in immigration, with 1.5 million more people entering per year than government statistical agencies had anticipated. The CBO finds that this surge will, on net, reduce the deficit by nearly $900 billion over the next eleven years, or (per my calculations) about 0.2 percent of GDP.
That’s a real and positive shift. It’s also nowhere near enough, and the result of a fairly unprecedented surge in immigration that is vulnerable to political change.
Similarly, we know that some kinds of government spending can encourage people to have more kids, but it seems unlikely that this spending is so effective as to pay for itself by creating future taxpayers. In any case, children have an annoying habit of needing to grow up before they can pay taxes, meaning this is not exactly a short- or medium-run solution.
Health care costs, by contrast, are something we can change with policy.
There’s an old saying in DC that the federal government is basically an insurance company with an army. That’s not exactly true, but it’s close. Social Security, Medicare, and Medicaid, plus defense spending, combined to make up about 66 percent of non-interest federal spending in 2023.
Within that mix, health care is becoming increasingly important. The Congressional Budget Office estimates that health insurance subsidies, encompassing Medicare, Medicaid, the Children’s Health Insurance Program (CHIP), the Affordable Care Act subsidies, and subsidies for employer-based coverage — will rise from 7 percent of GDP in 2023 to 8.3 percent in 2033. The biggest increase they project, by far, is in Medicare, which is swelling as the US population ages and whose patients tend to need more expensive care as they near the end of their lives. By contrast, defense spending, already far below where it was in the Cold War, is projected to keep falling relative to GDP.
Embedded in the projections of the CBO, and of the Medicare program’s actuaries too, is a prediction that per-person health expenses are going to increase in coming years, and increase faster than inflation. Over the past half-century or so that’s been the norm. Legislators have been making these programs progressively more generous in terms of the services covered, like extending prescription drug coverage in 2003, and new treatments and services keep getting developed that push costs higher. Add to that the fact that prices for most health services are higher in the US than in other rich countries and you have a recipe for spiraling expenses.
Except, since about 2011 or so, we haven’t seen that. In an almost miraculous development from a budgetary perspective, per-capita spending in Medicare has remained essentially constant for over a decade.
There’s considerable disagreement about why exactly this slowdown is happening. Some researchers emphasize technological change, arguing, in the words of Sheila Smith, Joseph Newhouse, and Gigi Cuckler, that “cost-reducing … innovation may have become more prominent relative to highly beneficial but expensive treatments” in recent years. That is, the ways we’ve improved health technology have cut, rather than raised, costs.
Melinda Buntin, a health economist who studied the slowdown first at the CBO and now as a professor at Johns Hopkins, told me in an interview she chalks up the slowdown to a fundamental mindset change in American health care. Partly as a result of the Affordable Care Act’s emphasis on “value-based” care, but partly also just due to an increased emphasis on paying for results on the part of doctors and hospital administrators, we’re seeing less wasteful use of new, expensive technologies than we used to.
Whatever the cause, the slowdown in medical costs is a hugely hopeful sign for the budget picture. Recall that the CBO projects that greater medical costs will add 2.6 percent of GDP to the federal budget by 2054. By comparison, it projects that the primary deficit (revenue minus spending, but excluding interest payments on past deficits) will be 2.2 percent of GDP. That suggests that the US could be in primary budget balance, a key sign of fiscal health, just by keeping health costs under control.
That said, it is by no means guaranteed that the recent slowdown will continue. Medical innovation continues apace, and new expensive drugs and treatments will continue to come on the market. For at least some of them, we will want the government to pay up, because they do so much to improve the lives of patients.
If there is a policy path that can keep the slowdown going, it is likely to be a combination of small steps rather than one big change. The 1% Steps project provides a great model for what this could look like: a crew of leading health economists have proposed measures ranging from fighting hospital mergers to changing how health claims are adjudicated to encouraging kidney donation that each save only a small amount of health spending, but together could do a lot.
Five steps to avoid the worst
Dealing with the federal debt problem isn’t exactly sexy. It will force policymakers to become less ambitious in some ways, and sacrifice for the sake of new programs or tax cuts they want. But the next few years give policymakers some real opportunities to address the deficit problem. As they do, there are a few principles they should keep in mind.
First: use tax hikes or spending cuts to offset any new spending or tax cuts. This has hardly been the norm in Congress lately, with bipartisan spending laws under both Trump and Biden costing hundreds of billions if not trillions without being remotely paid for. Those will need to be combined with tax hikes or offsetting spending cuts going forward. This also means giving up on some wilder spending visions that would require offsetting tax hikes no one is prepared to pass. Roger Wicker, the top Republican on the Senate Armed Services Committee, wants to raise spending on defense from 2.9 percent to 5 percent of GDP by the end of this decade, as part of an ambitious plan to counter China and Russia. If not paid for, this would absolutely explode budget deficits. If Wicker wants to pass the tax hikes necessary to cover the expense, god bless — but I have my doubts.
Second, take growth seriously. The point of deficit reduction is to reduce a long-term drag on economic growth, and so the best way to reduce the deficit is through actions that simultaneously enhance growth.
This is much easier said than done, of course, but expanded green cards for immigrants with science and engineering degrees seems like a no-brainer, cutting the deficit substantially over time while boosting growth. A growing literature also finds that funding for scientific research boosts productivity and economic growth. That funding doesn’t pay for itself, necessarily, but raising taxes or cutting less useful spending to boost scientific grants seems like a strong strategy. More generally, Congress should be searching for more policies in this category, ones that can boost growth however modestly at little or no budgetary cost.
Third: the coming tax fight in 2025 should be used to raise revenue, not just avoid losing it. Most of the individual tax cuts signed into law by Donald Trump in 2017 are due to expire at the end of next year. Some of these cuts, like slashing the top income tax rate from 39.6 to 37 percent, are things Democrats will absolutely want to expire. But most of them have bipartisan appeal.
Kamala Harris has pledged not to raise taxes on people making under $400,000 a year, which means keeping the expanded standard deduction and child tax credit that Trump enacted. It also means keeping his lower 12, 22, 24, and 32 percent brackets from the bill. All this and bringing back the child tax credit that Biden and Harris passed in 2021 costs about 1 percent of GDP in tax revenue by my estimations using the Committee for a Responsible Federal Budget’s tool.
Taking the debt at all seriously means that whoever’s in office next year needs to pay for whatever part of the Trump cuts they want to keep. Biden’s most recent budget contains a number of tax hikes on top earners, but notably doesn’t explicitly pay for extending the Trump cuts. Ideally, policymakers would not just pay for the extensions they want but come up with a revenue-positive package: one that doesn’t merely pay for the Trump cuts but goes further and reduces the deficit.
This won’t be easy but it’s certainly possible. Playing around with the Yale Budget Lab’s handy make-your-own-2025-tax-plan tool — which my sports-knowing editor tells me is like ESPN’s Trade Machine but for budget nerds — you can pay for the bigger standard deduction and Democrats’ desired child credit by letting all the Trump income tax rate cuts expire, letting the cuts to the estate tax expire too, raising the corporate tax rate to 28 percent, adding a new 45 percent tax bracket on the rich, and eliminating the tax break for pass-through corporations. That raises about 0.4 percent of GDP over a decade, a real bite of the long-run deficit. This does mean tax hikes on a small share of people making under $400,000 — but under the circumstances, I think that’s called for.
Fourth: Social Security should be addressed, not punted. Barring any Congressional action, the program is due for across-the-board cuts to benefits of over 20 percent starting in 2033, because payroll taxes and the trust fund will no longer be enough to pay promised benefits. This would unleash a political backlash on the part of affected seniors like nothing DC has ever seen, and Congress is almost guaranteed to do something to avoid that outcome.
The easiest way for Congress to deal with this doomsday deadline is to simply tap general tax revenues to avoid the cuts, thereby breaking the linkage between payroll taxes and Social Security benefits. They should not do this. The 2033 deadline provides one of the few forcing mechanisms that can push Congress to address the long-run imbalance between revenues and spending, and if they punt they might never get another opportunity. Democrats will push to pay for the program by raising taxes on high earners; Republicans will push for benefit cuts. Either way, the gap needs to be filled.
Congress will also get an opportunity to reform either Social Security or its sister program Supplemental Security Income to eliminate senior poverty — 14.1 percent of seniors are in poverty, a higher share than either children or working-age adults, despite the hundreds of billions the government spends on old-age pensions. That’s a travesty, and even Democratic plans to expand the program don’t address it sufficiently. It’s likely that even a relatively conservative plan could make real progress here. See, for instance, conservative Social Security expert Andrew Biggs’ plan to convert the old-age insurance program to a flat basic income for all seniors aged 62 and over, set at or above the poverty line.
Fifth and finally, Congress needs to work to ensure that per-person health spending stays roughly constant. This will likely take the form of several small reforms, along the 1% Steps suggestions, instead of another big effort like the Affordable Care Act. But precisely because the steps necessarily are more modest, the opportunities for bipartisan collaboration here are immense. See, for instance, this bill from Sens. Maggie Hassan (D-NH), Mike Braun (R-IN), and John Kennedy (R-LA) to reform Medicare billing, with support from the liberal Families USA to the Koch-backed Americans for Prosperity. Keeping health costs low might be the least unpopular way to deal with the budget imbalance, and that will hopefully spur politicians into more collaborations like this.
If Congress does all this — pays for future spending, prioritizes economic growth, raises revenue in the 2025 tax fight, fixes Social Security, and keeps per-capita health spending constant — it’ll be on track to stabilize the debt. It’s not easy. But it’s not impossible either. Bill Clinton and Congress were able to accomplish something similar in the 1990s, and now the time is ripe to try again.
