We review the second set of papers on inflation and fiscal policies (part I is here). The first paper (Fiscal Stimulus with Supply Constraints by Luca Fornaro (CREI)) examines the macroeconomic implications of fiscal stimuli. Fornaro (2024) argues that the effects of fiscal policies on inflation are large when the stimulus is short-term and magnified by supply disruptions, but the effects can be the opposite in the long-term. The second paper (Deficits and Inflation: HANK meets FTPL by Angeletos et al.) explores the relationship between fiscal deficits and inflation within the frameworks of Heterogeneous Agent New Keynesian (HANK) models and the Fiscal Theory of the Price Level (FTPL). The authors show that unfunded fiscal deficits in HANK models can trigger a boom in real economic activity and expand the tax base, reducing the inflationary impact of deficits by about half compared to the textbook FTPL prediction. The third paper (Inflation’s Fiscal Impact on U.S. Households by Altig et al.) argues that, contrary to previous research, the fiscal channel of inflation is progressive. The authors estimate that a permanent increase in inflation reduces median lifetime spending, with more pronounced effects on wealthier households.
Paper #1: Fornaro (2024)
Fiscal Stimulus with Supply Constraints by Luca Fornaro (CREI) examines the macroeconomic implications of fiscal stimuli when firms face supply constraints. Supply constraints significantly amplify the inflationary effects of fiscal stimulus, particularly when the stimulus is large and short-term or when the economy experiences supply disruptions. Conversely, a persistent fiscal stimulus can lead to increased investment and productivity growth in the medium term, while generating only temporary inflationary pressures.
What the paper does and main results
The paper presents a model that integrates supply constraints into the analysis of fiscal stimulus. These constraints, such as shortages of intermediate inputs or technological limitations, prevent firms from scaling up production in response to demand surges, causing prices to rise disproportionately relative to wages. The model features non-linear supply curves, where inflation rises once production hits certain capacity limits.
There are several key conditions under which fiscal stimulus leads to higher inflation. First, when fiscal stimulus is large but temporary, it can quickly push firms to their production limits, resulting in rising prices. The left panel of Figure 1 shows this mechanism pushing the economy from point A to point C. Second, when there are external supply disruptions—such as shortages of intermediate goods—supply constraints exacerbate inflationary pressures. This is because the supply disruption moves the kink of the Phillips curve, see the right panel of Figure 1.
Figure 1. Inflation response to fiscal stimulus in Fornaro (2024)
When government spending targets specific sectors, the sector-specific fiscal stimulus can concentrate demand in areas already operating near capacity, further intensifying price increases. When the fiscal stimulus is persistent, the paper shows that firms may respond by investing in capacity expansion and technology upgrading, thereby alleviating supply constraints over time. As a result, while inflation may rise initially, the long-term effects on productivity can lead to lower inflation and higher output.
A critique of the paper
The model offers a valuable framework for understanding how supply constraints can magnify the inflationary impact of fiscal stimulus. The model offers an alternative transmission mechanism to the canonical New Keynesian model, in which fiscal stimulus increases inflation via wages because it induces labor market tightness. Instead, binding supply constraints mean that wages do not fully reflect firms’ marginal production costs, so inflation can rise even if wages do not.
These dynamics may become even more important in an economy with production linkages across sectors, which realistically captures that certain goods require specific inputs from other sectors. Since the current model abstracts from this idea, it may understate the importance of supply constraints in quantitative exercises. The biggest drawback of the current paper is that it does not validate the model with data, so it remains unclear how much supply constraints contribute to inflation dynamics.
Policy implications
The Phillips curve can move due to adjustments in supply constraints. Central bankers should therefore aim to track these changes in real time to improve their nowcasts of the slope of the Phillips curve. The ongoing reshuffling of global supply chains due to shifting geopolitical power, national security concerns, and the green transition mean that the Phillips curve may move more strongly than in previous years making it more difficult to forecast inflation dynamics. Translated: the current inflation forecasts are way more uncertain than pre-Covid.
Paper #2. Angeletos, Lian, and Wolf (2024)
Deficits and Inflation: HANK meets FTPL by George-Marios Angeletos (Northwestern University), Chen Lian (UC Berkeley), and Christian K. Wolf (MIT) explores the relationship between fiscal deficits and inflation within the frameworks of Heterogeneous Agent New Keynesian (HANK) models and the Fiscal Theory of the Price Level (FTPL). Both frameworks can predict similar inflationary outcomes when fiscal adjustment is sufficiently delayed, though via different mechanisms. An empirically disciplined HANK model suggests much weaker inflationary effects of unfunded deficits than found in standard FTPL calculations.
What the paper does and main results
The paper investigates how fiscal deficits influence inflation, contrasting the predictions of the FTPL with those of HANK models. FTPL posits that deficits not backed by future surpluses must result in higher nominal prices to maintain the government’s intertemporal budget constraint. However, FTPL relies on assumptions about equilibrium selection that are often criticized. In contrast, HANK models predict that deficits drive inflation through increased aggregate demand because (at least some) households are not Ricardian, a mechanism independent of the controversial assumptions of the FTPL. Non-Ricardian behavior typically arises from financial constraints, the effects of which the authors proxy through mortality risk.
A benchmark HANK model can predict as much inflation as the FTPL if fiscal adjustment is slow enough, even when debt is fully financed through inflation. These similarities allow to study the inflationary effects of unfunded deficits through the lens of an estimated HANK model. A quantitative exercise shows that unfunded fiscal deficits in HANK models likely trigger a boom in real economic activity and expand the tax base, thereby reducing the need for inflation-driven debt erosion. This channel cuts the inflationary impact of deficits by about half compared to the textbook FTPL prediction. Figure 2 shows this result by plotting the output (left panel) and inflation (middle) response to a 1% deficit shock. The different lines indicate different slops of the Phillips curve (kappa). Steeper slopes give a stronger inflation response, but the cumulative inflation response always remains far below the cumulative response implied by the FTPL (see dashed line in the right panel).
Figure 2. Output and inflation response in HANK to 1% deficit shock, depending on the slope of the Phillips curve.
A critique of the paper
The paper makes a significant contribution by bridging HANK and FTPL frameworks, offering a robust alternative to the latter’s controversial assumptions. The use of a simplified HANK model is a strength, allowing for clear theoretical insights and practical applications. Since the FTPL relies on inherently untestable assumptions about equilibrium selection, the authors’ approach to use HANK as a bridge that can be empirically tested is smart. Ideally, we would like to have a model with equilibrium selection and HANK features to see how much each component contributes to the inflationary effects of unfunded fiscal deficits. Since this appears infeasible given the state of the literature, this paper emerges as the state-of-the-art approach. It is an important theoretical contribution which should be tested thoroughly in future empirical work.
Policy implications
The inflationary effects of unfunded fiscal deficits may be significantly smaller than documented by the FTPL. This is because of an endogenous expansion of the tax base, that provides an inherent self-financing mechanism of deficits. In practice, the quantitative effects of unfunded fiscal deficits will therefore depend on the strength of the tax base expansion and the slope of the Phillips curve. The former is a function of type of fiscal spending, the speed of fiscal adjustment, and underlying structural factors determining the economic response to government stimulus. The latter depends on several factors such as the strength of wage-price pass-through and markup changes. This makes assessing the inflationary effects of unfunded deficits no easy task: Its effects may be strongly time-varying and differ across countries.
Paper #3: Altig et al. (2024)
Inflation’s Fiscal Impact on U.S. Households by David E. Altig (Atlanta Fed), Alan Auerbach (Berkeley), Erin Eidschun (Boston University), Laurence Kotlikoff (Boston University), and Victor Yifan Ye (Opendoor Technologies/BU) explores the effects of inflation on U.S. households through the fiscal channel. Contrary to previous research, the paper argues that the fiscal channel is progressive. Inflation affects households through nominal rigidities in the fiscal system, with effects varying across income levels and age groups. A permanent increase in inflation from 0% to 10% can reduce median lifetime spending by 6.8%, with more pronounced effects on wealthier households.
What the paper does and main results
The study addresses the hidden fiscal impacts of inflation that arise due to unindexed or lagged indexation of tax and benefit provisions in the U.S. fiscal system. These provisions include the taxation of nominal asset incomes, Social Security benefits, and income thresholds that are not adjusted promptly for inflation. The authors employ the Fiscal Analyzer (TFA), a life-cycle consumption-smoothing model, to analyze data from the 2019 Survey of Consumer Finances (SCF) under various inflation scenarios (0%, 5%, and 10%). The authors begin by neutralizing non-fiscal effects of inflation and then simulate the impacts under different inflation rates.
The authors show that rising prices are considerably more progressive than implied by prior research. A permanent 10% inflation rate reduces median lifetime spending by 6.8%, with the most significant reductions observed among the wealthiest 1% of households, who experience a 15.9% decrease. These results are shown in Figure 3 (right), along with the results for a 5% permanent inflation rate (left). The effects persist even when accounting for lags in cost-of-living adjustments (COLA), which explain only a third of the reduction. The study also reveals substantial heterogeneity in the fiscal impacts of inflation, with middle-aged households (ages 50-59) being particularly affected due to their higher financial wealth holdings.
The authors highlight the complexity of inflation’s impact, showing that it is highly progressive, with wealthier households bearing a larger share of the burden. The variation in impacts across households is driven by factors such as state-specific tax structures and differences in asset holdings, which influence how nominal income taxation and benefit adjustments play out.
Figure 3. Estimated effects on median lifetime spending due to permanent inflation rate of 5% or 10%.
A critique of the paper
The paper provides a thorough analysis of how inflation affects household welfare through the fiscal system. The use of the Fiscal Analyzer model allows for detailed simulations that capture all major federal and state tax and benefit programs. However, the model’s reliance on the 2019 SCF data may not fully reflect more recent economic developments, particularly those following the COVID-19 pandemic. Not only but an obvious question mark is whether measurement of spending at the top of the distribution is accurate.
The simulated change in inflation is permanent, which may require additional modelling features to allow for endogenous responses. A 10% permanent inflation rate generally lead to faster adjustments in tax and benefit programs to avoid distortions. Additionally, household expectations and firm behavior may considerably change in such an environment, therefore limiting the conclusions that can be drawn from such simulations. While the model is rich on the tax and benefit side, its limitations in this regard may require further work or a change in the experiment to be studied.
Policy implications
The paper has three implications, far from obvious. First, accepting a higher inflation target imposes non-trivial costs on households beyond the typical price and wage effects. A 3.5% long-term inflation target would, in expectation, lower US households’ standard of living by nearly a percentage point on average compared to the current 2% target. Second, the regressive wage- and price-channel impact of inflation is at least partially offset by progressivity of the fiscal channel. Therefore, it is likely that rising prices are considerably more progressive than implied by prior research, with the largest losses hitting households with the highest levels of lifetime resources. Third, aggregate estimates of inflation’s impact – both price-wage and fiscal – mask striking heterogeneity, both across and within age and resource groups. The complexity of the US fiscal system imposes huge costs – and, in rarer cases, gains – to some households.