We review three papers on inflation and fiscal policies. The first one “Fiscal Backing, Inflation, and US Business Cycles” by Smets and Wouters (2024) develops a model to incorporate a regime of partial fiscal backing and argues that unfunded fiscal shocks had small inflationary effects on average, excluding some exceptions such as the current episode. The second paper “Do Tax Increases Tame Inflation?” by Cloyne et al. (2024) argues that while personal income tax hikes tend to reduce inflation by lowering demand, corporate income tax increases often lead to higher prices, likely due to their adverse effects on supply-side factors such as productivity and investment. The last paper “The U.S. Public Debt Valuation Puzzle” by Jiang et al. (2024) finds that U.S. Treasuries are systematically overpriced relative to the risk-adjusted value of expected surpluses, a phenomenon they term the “government debt valuation puzzle”. This mispricing suggests that Treasury investors may not fully account for the risks associated with future fiscal surpluses.
Paper #1: Smets and Wouters (2024)
“Fiscal Backing, Inflation, and US Business Cycles” by Frank Smets (ECB) and Raf Wouters (National Bank of Belgium) develops an extended version of the Smets and Wouters (2007) model to incorporate a regime of partial fiscal backing, where fiscal policy does not fully commit to future surpluses, thereby sharing the burden of debt sustainability with monetary policy. On average, 80% of the fiscal implications of business cycle shocks in the U.S. are funded, meaning the U.S. has operated closer to a monetary-led regime. Unfunded fiscal shocks had small inflationary effects on average, though there are some exceptions such as the 1960s-1970s and 2021.
What the paper does and main results
The authors extend the Smets and Wouters (2007) Dynamic Stochastic General Equilibrium (DSGE) model by incorporating a fiscal block that includes government debt dynamics and the possibility of partially funded fiscal policies. This allows for empirically relevant intermediate cases between the extreme scenarios of either fiscal or monetary dominance. In a fiscal-led regime, inflation is influenced by fiscal policies when government debt is not fully backed by future fiscal surpluses because the central bank is forced to set interest rates to achieve the present discounted value of future government surpluses that satisfies the government budget constraint. The authors estimate this model using U.S. data from 1965 to 2019, focusing on how different fiscal and monetary policy shocks affect inflation and output.
According to the model, 80% of the fiscal implications of business cycle shocks (including fiscal shocks) are funded. The U.S. government therefore mostly finances its government debt via future surpluses instead of relying on the central bank to inflate its debt away. Because of this, most of inflation over the 1965-2019 sample has not been driven by fiscal shocks, see the black bars in the left panel of Figure 1. Notable exceptions include the 1960s and 1970s, as well as the 1990s.
Fiscal shocks played a more prominent role during the 2021 Covid inflation. As the right panel of Figure 1 shows, fiscal shocks compensated much of the disinflationary effects of demand shocks. At the same time, most of the Covid inflation is ascribed to supply shocks, with monetary shocks playing a small role.
Figure 1. Historical decomposition of U.S. inflation into contributing shocks, for 1965-2019 (left panel) and 2020-2023 (right panel).
A critique of the paper
The paper makes a contribution by offering a more nuanced understanding of the interaction between fiscal and monetary policies. By estimating the degree of fiscal backing through the lens of a workhorse macroeconomic model, the authors are able to move beyond theoretical edge cases of pure fiscal or monetary dominance.
However, difficulties in precisely estimating the degree of fiscal backing may affect the results. The authors show that the estimated backing parameter can lie anywhere between 0.5 and 0.8 according to their confidence intervals for different subsamples. This makes the choice of 80% fiscal backing for their empirical exercise less convincing and more robustness checks would be needed. In addition, any model on the fiscal theory needs to face recent empirical results suggesting that the market value of government debt does not satisfy the government budget constraint, thereby violating a core assumption of such models.
Paper #2: Cloyne et al. (2024)
“Do Tax Increases Tame Inflation?” by James Cloyne (UC Davis), Joseba Martinez (LBS), Haroon Mumtaz (Queen Mary), and Paolo Surico (LBS) investigates the impact of tax changes on inflation in the United States, particularly distinguishing between the effects of personal and corporate income tax increases. While personal income tax hikes tend to reduce inflation by lowering demand, corporate income tax increases often lead to higher prices, likely due to their adverse effects on supply-side factors such as productivity and investment.
What the paper does and main results
The study estimates the dynamic effects of tax changes on the Personal Consumption Expenditure (PCE) deflator. Exogenous tax increases are identified by focusing on those tax reforms that did not respond to current or prospective economic conditions (narrative identification). The authors use U.S. data from 1950 to 2006, focusing on how changes in average personal and corporate income tax rates influence a range of aggregate and sectoral price indices, including 190 subcomponents of the PCE deflator.
An increase in personal income taxes leads to a broad-based reduction in prices across many sectors, with the most substantial effects observed in non-durable goods, see Figure 2. This finding aligns with the Keynesian view that higher taxes reduce disposable income and aggregate demand, leading to lower inflation. Conversely, increases in corporate taxes have a limited impact on prices in the short term but tend to raise prices in the medium to long term, particularly for durable goods and capital equipment (results not shown). The paper suggests that this inflationary effect of corporate tax hikes is driven by their negative supply-side impact on productivity and investment.
Figure 2. Effect of a 1pp personal income tax increase on the PCE deflator (left panel, red line) and its subcomponents (left panel, grey lines) as well as on nondurables and durables (right panel). The shaded areas in the right panel are 90% confidence intervals.
The authors also examine the effects of these tax changes on inflation expectations and stock prices. The paper finds that personal tax increases lead to lower inflation expectations but have little effect on stock prices, while corporate tax hikes result in a persistent decline in stock prices, particularly in high R&D-intensive sectors, with limited effects on inflation expectations.
A critique of the paper
The paper shows that not all tax hikes are created equal. It makes a contribution by differentiating between the inflationary effects of personal and corporate tax increases, using an empirical approach that builds on well-established methods. The results fit nicely into the standard distinction between demand-side and supply-side effects from economic theory, providing clear insights into the mechanisms through which tax changes influence prices.
A flat-line income tax increase need not be the most relevant policy tool to study. Policymakers may want to raise a certain amount of tax income, which can be achieved through progressive tax increases. To the extent that high-income households have lower marginal propensities to consume, the demand effects of such a tax increase may be significantly weaker while raising the same tax amount, yielding weaker disinflationary effects than the authors estimate.
Paper #3: Jiang et al. (2024)
“The U.S. Public Debt Valuation Puzzle” by Zhengyang Jiang (Northwestern University), Hanno Lustig (Stanford), Stijn Van Nieuwerburgh (Columbia), and Mindy Z. Xiaolan (UT Austin) explores the apparent discrepancy between the market value of U.S. government debt and its fundamental value, which should theoretically equal the present discounted value of future primary fiscal surpluses. The authors find that U.S. Treasuries are systematically overpriced relative to the risk-adjusted value of expected surpluses, a phenomenon they term the “government debt valuation puzzle”. This mispricing suggests that Treasury investors may not fully account for the risks associated with future fiscal surpluses.
What the paper does and main results
The paper provides a framework for the valuation of the outstanding U.S. government debt. The framework is model-free and provides an upper bound on fiscal backing that supports outstanding Treasuries. The upper bound is derived as the difference between the maximum value of future tax income and the minimum value of future government spending (intuitively: the maximum the government can collect, minus the minimum it can spend). On average, the upper bound on the U.S. debt-to-GDP ratio is very low (3.15%, see the horizontal black line in Figure 3). The intuition is that simple: observed surpluses are historically close to zero, on average. Allowing for variation over time in expected growth and discount rates yields a higher upper limit indicated by the red line ( and the grey shaded areas). The main result of the paper is that since WW2, the U.S. debt-to-GDP ratio has almost always been considerably higher than the fiscal backing (the blue line has always been above the red line). Put it simply: according to the authors, the debt/GDP ratio has always been higher than the maximum future fiscal room to repay it. For this reason, the paper argues that the market value of the Treasury portfolio is too high.
Figure 3. Debt-to-GDP ratio against the present discounted value of future government surpluses over GDP.
Note: “1” on the y-axis indicates “100%”.
The authors confirm that their finding is not accounted for by convenience yields, asset price bubbles, or investor expectations of future austerity measures that correct government surpluses. In contrast, they argue it could be the case that investors price in future financial repression. Throughout U.S. history, the Treasury and the Federal Reserve have repeatedly engaged in coordinated efforts to reduce government funding costs. Anticipation of such interventions could explain the high valuation of government debt.
A critique of the paper
The paper rigorously quantifies the mismatch between the market value of U.S. debt and its theoretical value under standard asset pricing assumptions. The paper convincingly combines a model-free approach with additional modeling assumptions in later parts of the paper to draw empirically robust conclusions. The authors stress the importance of capturing the risk-adjusted value of surpluses instead of computing present discounted values using risk-free rates. While this seems sensible, it may drive a large share of the results, putting the correct estimation of risk premia into focus.
While the framework is convincing and the authors rule out several possible explanations for their puzzle, the sheer size of the mispricing makes us a bit skeptical. The empirical results suggest a gigantic difference between the debt-to-GDP ratio that the U.S. could afford and the actual debt burden. Limits to arbitrage could explain the persistence but not necessarily the size of this valuation puzzle. Two possible alternatives are that investors evaluate the probability of a U.S. austerity event conditional on it being necessary to stabilize the value of government debt as sufficiently high to justify Treasury valuations (which is not considered in the paper), or that investors face a global portfolio and simply do not see safer alternatives than dollar-paying Treasury securities (or simply must hold them, given regulations).