No, money isn’t free. Photographer: Matthew Lloyd/Getty Images
After the disease, the debt. After the plague, the pile of IOUs. It is a veritable mountain — a reminder that the original public debt in medieval Venice went by the name monte. According to the International Monetary Fund’s October Fiscal Monitor, the Covid-19 pandemic and associated lockdowns have prompted a plethora of fiscal measures amounting to $11.7 trillion, around 12% of global GDP — and that number has probably risen since it was calculated on Sept. 11. “In 2020,” according to the Fund, “government deficits are set to surge by an average of 9 percent of GDP, and global public debt is projected to approach 100 percent of GDP, a record high.”
In advanced economies, public debt relative to output has increased as much since the late 1970s as it did between 1914 and 1945. Together, the global financial crisis and the pandemic have had roughly the same doubling effect as World War II. While Covid-19 will not kill as many people globally as history’s biggest war, the ultimate U.S. death toll is very likely to be higher. The pandemic’s financial cost also looks similar to that of a world war.
The IMF’s global averages conceal huge variations between countries. The deficits of seven developed countries — Canada, the U.K., the U.S., Brazil, Italy, Spain and Japan — have each risen by more than 10% of GDP. In all these countries, gross public debt will exceed 100% of GDP this year, with Japan’s reaching 266%. The IMF’s projection for U.S. gross debt is 131%.
In many advanced countries, public debt will exceed 100% of GDP this year
Source: International Monetary Fund
Because the Federal Reserve has purchased most of the new Treasury bonds created this year, the increase in the federal debt held by the public is not so daunting. 1 The Congressional Budget Office projects that it will be just under 100% this year (98.2%), but that is still nearly three times what it was at the beginning of this century. Next year it is forecast to surpass the level in 1945.
But the story doesn’t end there, because the federal government is expected to keep borrowing as far as the eye can see, with the deficit rising inexorably from 4% in the late 2020s to over 12% by mid-century. The CBO’s baseline projection is for the federal debt in public hands to reach 195% of GDP by 2050, nearly twice as high as at the end of World War II.
U.S. debt in public hands could near 200% by 2050, nearly twice as high as at the end of World War II
Source: U.S. Congressional Budget Office
Historically, large public debts have had a terrible reputation. In his “Rural Rides,” which he began in 1822 and published in 1830, the English radical William Cobbett inveighed against the vast national debt incurred during the Napoleonic Wars. The political purpose of the debt, Cobbett argued, had been “to crush liberty in France and to keep down the reformers in England” but its principal effect after the war was redistributive. “A national debt, and all the taxation and gambling belonging to it, have a natural tendency to draw wealth into great masses … for the gain of a few.”
“The Debt, the blessed Debt,” he continued, was “hanging round the neck of this nation like a millstone.” It was a “vortex,” sucking money from the poor to a new plutocracy.
Cobbett was not wrong. According to the best estimates from the Bank of England, Britain’s wars between 1776 and 1815 drove up the debt/GDP ratio from 86% to more than 172% by 1822. In those days, government bonds were almost entirely owned by a tiny wealthy elite, while taxation was largely indirect, on imports and consumption, and therefore highly regressive. Moreover, real (inflation-adjusted) long-term interest rates were strongly positive, averaging 5.27% in the 1820s.
The “Blessed Debt”
In the 19th century, Britain’s war debts and high real rates enriched a new plutocracy
Source: Bank of England
The good bad news is that today’s public debts have a very different character. Ownership is more evenly distributed, as most bonds are held institutionally by insurance and pension funds and other financial institutions. Taxation everywhere is far more progressive than it was in the early 19th century, a time when income taxes were regarded as wartime expedients. And, as we have seen, a significant portion of today’s public debts are held by central banks, meaning that one part of the government owes money to another.
In an important new paper, economists Jason Furman and Lawrence Summers argue that public borrowing today offers something very like that rarity in economics: a free lunch. The key is the historically low level of nominal and real interest rates. On the one hand, low rates mean that “monetary policy cannot be relied on to stabilize the economy.” On the other hand, low rates also mean that “fiscal expansions themselves can improve fiscal sustainability by raising GDP more than they raise debt and interest payments.”
Today’s interest rates mean that debt/GDP ratios are a bad way to measure debt burdens. After all, the accumulated public debt is a stock, whereas GDP is a flow. If debt is measured relative to estimates of the present value of GDP or prospective tax receipts, then “current debt levels are at low rather than high levels.”
Furman and Summers are not saying — as the proponents of modern monetary theory do — that debt doesn’t matter and the sky is the limit. They are simply arguing that “traditional ideas of a cyclically balanced budget on the grounds that [high debt] would likely lead to inadequate growth and excessive financial instability” are anachronistic. Fiscal policy can support growth with ongoing deficits so long as real debt service (i.e., interest payments adjusted for inflation) does not rise above 2% of GDP over the coming decade.
If governments borrow to finance investment, then, so much the better because “many public investments pay for themselves, or come close to paying for themselves, and the risk of not undertaking these investments is larger than the risk of doing too little deficit reduction.”
“Currently,” Furman and Summers conclude, “the primary worry for policy in the United States and several other countries is doing too little to expand the debt, not doing too much.” Democrats hoping for dual victories in next month’s Senate run-off elections in Georgia will read these words with tears of joy in their eyes. If they can only replace Sen. Mitch McConnell with Vice President-elect Kamala Harris as master/mistress of the Senate, they can fulfill their campaign pledges of spending up to $4 trillion, with nothing to fear from the bond vigilantes that terrorized former President Bill Clinton’s administration in its early days.
Furman and Summers are by no means the first to argue that debt-to-GDP is the wrong way to measure fiscal sustainability. In 2001’s “The Cash Nexus,” for example, I made a similar point: What really matters is keeping the real growth rate above the real debt service rate. Like Furman and Summers, I cited the pioneering work of Laurence Kotlikoff, who focuses on the present value of projected spending and revenues, as well as on the distributional effects of fiscal policy between generations.
And credit where credit is due. In the debates on interest rates and inflation that followed the global financial crisis, Summers was the winner. His 2014 lecture on “secular stagnation” — which argued that for a variety of structural reasons (e.g. aging populations and inequality) interest rates would remain stuck close to zero for the foreseeable future — proved prescient. Those of us who worried (as I did briefly) that Fed purchases of bonds (quantitative easing) might be inflationary were wrong. So were the Fed economists who wanted to normalize monetary policy by raising rates preemptively. (Remember how well that went two years ago?)
The problem is that for there to be a free lunch, financed by borrowing that pays for itself, secular stagnation has to continue: In other words, interest rates have to stay at their present low levels, which isn’t what the CBO expects. In its most recent long-run forecasts, nominal and real rates rise over the course of the 2020s. That means that net interest payments would rise above 2% of GDP from 2030 onward and hit 8.1% in 2050.
No More Free Lunch
Rising rates would drive up the government’s net interest payments
Source: U.S. Congressional Budget Office
True, as Furman and Summers point out, the CBO has been consistently wrong about the future path of interest rates, overshooting repeatedly since 1990. True, any forecasts beyond a ten-year time horizon are subject to great uncertainty.
Even so, my Hoover Institution colleague John Cochrane has history on his side when he expresses concern. As he argued in National Review last week, the situation is very different today from the situation in 1945, the last time the U.S. had a debt mountain this big. “By 1945, the war and its spending were over. For the next 20 years, the U.S. government posted steady small primary surpluses, not additional huge deficits. Until the 1970s, the country experienced unprecedented supply-side growth in a far less regulated economy with small and solvent social programs. … [Today] we are starting a spending binge with the same debt relative to GDP with which we ended WWII.”
In any case, as noted above, around three-quarters of this year’s deficits have been financed by Fed money creation in the form of excess bank reserves. “When the economy recovers,” Cochrane argues, “people may want to invest in better opportunities than trillions of dollars of bank deposits. The Fed will have to sell its holdings of Treasury securities to mop up the money. We will see if the once-insatiable desire for super low-rate Treasury securities is really still there. If not, the Fed will have to raise rates much faster than their current promises.”
This goes to the heart of the matter. A debt mountain doesn't matter only so long as interest rates remain low. That implies that there could very well be a key role for monetary policy if market participants anticipate higher inflation and start selling their holdings of Treasury bonds. The Fed has a new framework now, which states that inflation above its 2% target is just fine, after 12 years mostly below that level, as long as it averages out around 2%. But that clearly means a prolonged period of negative returns on government bonds, made worse for foreign investors if the dollar continues to slide against other major currencies.
If market rates start to rise, the Fed will be put to the test. Will it behave as it did in World War II, intervening to keep rates low in order to avoid a rapid rise in government debt-servicing costs? There is a widespread belief that it will and that Japanese-style “yield curve control” lies ahead. But in 1945 that was a wartime expedient and it was ended with the Fed-Treasury Accord of 1951, which restored the separation of monetary policy from debt management.
Another way of thinking about this is to contrast the likely trajectory of the post-pandemic economy with the sluggish path of recovery after 2008-9. A financial crisis originates in overstretched balance sheets — in the case of the 2008-9, those of banks, shadow banks and subprime mortgage borrowers. It took the better part of a decade for balance sheets to be repaired, which was one reason for the slow pace of recovery in the Obama years — the background against which secular stagnation seemed the right diagnosis.
The post-pandemic economy will be very different — and this is the bad good news. This year, thanks to Covid-19, the U.S. household savings rate has had its most volatile year since modern data began in 1948. In the second quarter, it jumped to an unprecedented 26%, compared to 7.3% a year before. As lockdowns and other restrictions were relaxed, the rate declined to 16% in the third quarter.
To expect such high rates to persist into 2021, as the OECD does in its latest Economic Outlook, is surely wrong. This was forced saving of income boosted by government handouts, prompted by a supply-led shock (lockdowns), not balance-sheet repair as after 2008-9. According to our estimates at Greenmantle, U.S. households are now sitting on roughly $1 trillion of excess savings as a result. Many are itching to spend a large chunk of that money as soon as they can.
The best analogy for the Covid-induced economic slump is not a normal recession but a war. With vaccine distribution in sight, society is now preparing to demobilize. As World War II wound down, many esteemed economists — notably Alvin Hansen, who coined the term “secular stagnation” — wrongly predicted an enduring economic crisis. Instead, the gradual removal of wartime restrictions led to a boom in consumption. Something similar seems in prospect next year.
The key question is how inflationary that post-pandemic boom will be. Most economists seem to agree with Furman and Summers that secular stagnation is here to stay. Charles Goodhart of the London School of Economics is one of the few to predict a “surge of inflation” as soon as next year. If he is right, the promised debt-funded free lunch could turn out to be a very expensive dinner.
While I doubt Goodhart’s prediction that inflation might rise to 5% next year, inflation can come at you fast, as my Bloomberg colleague John Authers pointed out last week. The U.S. housing market has roared back. Home equity withdrawals have soared. Bank deposits are way up and household debt-service ratios are at all-time lows. We are heading for a roaring 2021, if not the full Roaring Twenties. With a weak dollar and rising commodity prices, inflation might just give the Fed a fright.
I am not a macroeconomist; I am a mere economic historian. To me, past experience is more compelling than any model. The lesson of history is indeed that there is no correlation between debt-to-GDP ratios and long-term interest rates, just as there is no simple relationship between the size of central bank balance sheets and inflation.
But history also teaches us that debt and power are connected: A great power or empire that accumulates too high a mountain of debt and fails to keep growth ahead of debt service is destined to decline. The Bourbons, the Ottomans and the British all learned this hard lesson. So the post-pandemic debt dynamics matter not just for markets but for geopolitics.
In a new book published in online installments, “The Changing World Order,” Bridgewater Associates LP founder Ray Dalio argues that the U.S. is in the wrong stage of classic debt cycle. “When the government runs out of money (by running a big deficit, having large debts, and not having access to adequate credit) it has limited options,” he writes in chapter 9:
It can either 1) raise taxes and cut spending a lot or 2) print a lot of money, which depreciates its value. Those governments that have the option to print money always do so because that is the much less painful path, but it leads investors to run out of the money and debt that is being printed. Those governments that can’t print money have to raise taxes and cut spending, which drives those with money to run out of the country, state, or other jurisdiction because paying more taxes and losing services is intolerable. If these entities that can’t print money have large wealth gaps among their constituents, these moves typically lead to some form of civil war/revolution. This late-cycle debt dynamic is now playing out in the United States.
Scary stuff. And, to judge by an essay written by Guo Shuqing, chair of the China Banking & Insurance Regulator Commission and party secretary of the Chinese central bank, Dalio has influential readers in China, the inexorable rise of which is the other big theme of his book.
Still, I am old enough to remember Paul Kennedy’s argument in “The Rise and Fall of Great Powers” that the total U.S. federal debt in 1985 (then a mere 35% of GDP) was a sign of impending American overstretch reminiscent of “France in the 1780s, where the fiscal crisis contributed to the domestic political crisis.” What followed instead was the collapse of the Soviet empire and American triumph in the Cold War, not to mention Japan’s lost decades.
Most commentators are ending the year bullish on China — the only major economy that grew this year, and forecast by the OECD to grow by 8% next year. China’s gross public debt will be just 62% of GDP this year, less than half the U.S. figure.
But it is private debt that worries Chinese officials such as Guo and Vice Premier Liu He, not public debt. Since President Xi Jinping came to power in November 2012, according to the Bank for International Settlements, credit to households has doubled as a share of GDP to 59%, while credit to non-financial corporations has jumped by 38 percentage points to 162%.
China’s “Gray Rhino” Risk
The government is more worried about private debt as a share of GDP than public debt
Source: Bank for International Settlements
Guo’s fear is that excess leverage in the property sector — which accounts for about 39% of total outstanding bank lending — is the “biggest gray rhino risk” facing China’s financial system. The enduring impact of the pandemic has created a growing problem of non-performing loans, driving smaller lenders to insolvency. Last month saw a series of defaults by state-owned companies in China. And last week S&P Global Ratings warned that local government financing vehicles could be the next casualties as the authorities clean house.
After the disease, the debt. But it’s important to look not just at public debt but also at private debt when trying to see which great power has the steeper mountain to climb.
(Adds footnote in fourth paragraph to clarify what is included in federal debt held by the public. )
The federal debt held by the public excludes bonds held by federal trust funds such as Medicare and Social Security, but includes bonds bought by the Federal Reserve.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.