On the Sustainability of American Public Debt
This past weekend, the U.S. debt ceiling, which was suspended in June of 2019, became operational once again. Inevitably, concerns about U.S. debt have resurfaced, and reasonably so, given the sudden spike in public debt that followed recognition of COVID-19 as a global pandemic in February of 2020. Federal debt in the U.S. as a percentage of gross domestic product has remained elevated at an average of 130% since COVID-19 prompted vast amounts of government spending and debt buybacks, facilitated by both monetary and fiscal levers. While the technicalities of the debt ceiling are probably nothing to lose sleep over, considering Congress has revised the debt ceiling upwards 78 times since 1960, the real implications of high public debt are worth examining.
For perspective, as high as current US debt-to-GDP seems, it’s only several percentage points higher than the 123.9% average at the end of 2019, taken across 25 advanced economies. Triple-digit debt-to-GDP ratios have become unusually common in the developed world and were prevalent even before the global pandemic. Broadly, this growth in public debt is associated with a slew of economic and demographic trends including lower economic growth and productivity gains, larger amounts of automatic fiscal spending tied to social insurance programs like Social Security, depressed nominal interest rates, and a shift in investment demand towards safer assets. Though real risks surround moderate to high levels of public debt, some of the more fearful consequences, notably mass default, are unlikely to play out. In fact, it’s likely that these debt figures are more sustainable than politicians make them out to be.
But what does it really mean for public debt to be sustainable? Literally, public debt is sustainable if total public debt today is less than or equal to the projected discounted stream of government revenues. Additionally, I argue that sustainable implies an economy can operate at full productive capacity, without imposing additional constraints on growth or investment during non-recessionary periods. There are both theoretical and empirical reasons to believe that the debt we see today meets the first of these criteria. In an influential 2019 paper, Olivier Blanchard of MIT showed that as long as the risk-free rate of return, r, is less than the growth rate of the economy, g, debt rollovers have trivial fiscal costs and can be sustained; further, this condition is welfare enhancing, or increases the total wellbeing of both individuals in this generation, and in the next generation. In other words, even if debt-to-GDP lingers in the triple-digit range, it can be sustained as long as the economy grows at a quicker rate than the risk-free rate of return, which is the rate at which a sovereign government must repay accrued debt. This condition has been borne out by the data, as shown below where I use the 1-year Treasury bill yield as a proxy for the risk-free rate of return. It’s clear that g > r since 2010, barring a trend deviation in early 2020 as COVID-19 cases surged. From a mechanistic standpoint, U.S. federal debt is feasible, even if the absolute level of debt increases over time. Thus, the first criteria of sustainability is satisfied. I have not yet demonstrated that the second criteria, which is arguably more important from a growth perspective, is satisfied theoretically or empirically.
Blanchard went on to show that when the marginal product of capital, which is a technical term that can be thought of as the average risky rate of return (i.e. return on private capital) and is denoted m, is greater than g, we observe a reduction in total welfare. In essence, this condition is welfare reducing because it signifies lower than expected returns to capital and labor as a possible result of public debt crowding out private investment. The data shows that m has indeed been greater than g for the past ten years; I use year-over-year S&P returns as a conservative estimate of m. It should be noted that this condition doesn’t preclude feasibility of the public debt, it only comments on the potential disruptions to welfare.
So what do we make of an environment where r < g < m, in which we’re supposed to observe welfare gains and reductions simultaneously? The net effect depends on which difference is larger — net welfare will be positive given r is sufficiently low and m is not considerably higher than g. I admit that these qualifiers are somewhat vague, but only because it has yet to be determined what the true thresholds on debt growth are.
These results have several implications for current public debt and the trajectory of future public debt in America where r < g < m. Precisely, r has hovered near zero, g has averaged 3.39%, and m is conservatively estimated to be 9.09% (Lowe et. al 2019) over the past ten years. On the one hand, public debt is likely feasible in the short to medium term since r < g, meaning we won’t see public debt imploding in on itself anytime soon, regardless of what mainstream media outlets might say. Additionally, increases in the primary surplus, or government revenue less non-interest government expenditures, can further offset costs borne by looming interest payments on old debt; however, we’re unlikely to see any increases in the primary surplus during Biden’s term, given the massive infrastructure plan looming in Congress. Biden’s initiatives to revise both the capital gains tax and the corporate tax rate upwards, as well as impose greater taxes on the ultra-wealthy, will somewhat assuage the deficits generated by the probable passage of the bill.
On the other hand, I’m unconvinced that the current rate of public debt accumulation has had, or will have no adverse consequences on private investment and growth, which I’ve argued as the second criteria for sustainable public debt. For one, the low-rate environment the Western world has become accustomed to is due not only to a diminished appetite for risky assets by investors, but also by more than a decade of quantitative easing and loose monetary policy by central banks around the world. Though I will not delve into the rationale and potential implications of recent monetary policy, a byproduct of low interest rates is an artificially low r, which may drive m up by fattening the risk premium for investors hungry for higher returns. Though it is yet to be determined, this could manifest in the form of potential asset bubbles in equity and housing markets. The formation and eventual collapse of an asset bubble could ruin growth and investment prospects for years. Moreover, high debt-to-GDP ratios come with real risks and tradeoffs that aren’t fully appreciated such as lower returns to public investment projects, inflationary pressures, and higher probabilities of fiscal crises. A wealth of empirical and theoretical research has documented significant negative associations between debt-to-GDP and output growth beyond certain thresholds (Rogoff and Reinhart 2010; Kumar and Woo 2010).
What should one make of all this? Right off the bat, I think any concerns over an explosion of the public debt or of the U.S. government’s inability to repay past debt obligations are overblown. Admittedly, a sovereign government is never required to default as it can always “print more money”, though doing so is clearly unpalatable as it comes with high inflation and reduced future credibility. There is, however, a second option which involves the sovereign government running a Ponzi game, where it issues new debt to cover past debt ad infinitum. To some extent this is what many advanced economies are doing, and it’s even feasible given the r < g condition I previously mentioned. Additionally, Blanchard shows how contrary to standard economic theory, this condition is the rule rather than the exception.
However, just because high public debt is feasible does not mean it’s desirable, for the reasons I’d mentioned. From a monetary perspective, the Federal Reserve should focus on slowly hiking rates and scaling back asset purchases once the worst of the pandemic has subsided, however it may be a good while till we see any mention of either of these in Fed meeting minutes. On the fiscal front, it would clearly make sense to reduce public expenditures, though this is neither realistic, given the average annualized growth rate of government expenditures since 2000 is 9.67%, nor even desirable, given the plenitude of market failures that pervade America’s modern economy. Congress should continue to fund public investments with high returns on investment, such as clean energy, higher education, and STEM research and development, that make the United States globally competitive. These public investments could spur significant increases in g, which would naturally raise the primary surplus and help alleviate the debt burden.
It’s worth scrutinizing announced political concerns over the public debt on the right, as they’re often little more than concealed partisan efforts to curb the level of public expenditures and thus the “overreach” of the public sector; somehow, we never hear these concerns when public debt rises as a result of massive tax cuts, or when the primary deficit goes further negative under Republican presidents. That isn’t to say that all discussions related to public debt are necessarily ideological. A growing public debt should be monitored and may have potentially adverse consequences on growth and investment. However, policies that substantially boost economic growth are likely to assuage the adverse consequences of a rapidly growing public debt, which in practice is much more difficult to control than we’d like.
