Estimating the inflation impact of the 2021 U.S. fiscal stimulus
Jonathon Hazell and Stephan Hobler, in this paper, use a novel “high frequency narrative approach” to estimate the causal effect of the 2021 U.S. fiscal deficits on inflation. The key elements of the methodology are:
- Identifying an event that released news about deficits: The authors use the Georgia Senate runoff elections in January 2021 as the key event, as a Democratic victory was expected to lead to larger fiscal stimulus.
- Measuring the size of the deficit shock: Using new narrative data from investment bank reports, the authors estimate the shock to expected deficits from the Democratic victory was about 2.1% of GDP.
- Measuring the high-frequency response: The authors examine how inflation expectations from asset prices (inflation swaps) changed in a narrow window around the election results.
- Combining narrative and high-frequency data: This allows the authors to estimate an “inflation multiplier” – the effect on inflation of a 1% of GDP deficit shock.
The main findings are:
- The estimated inflation multiplier is 0.18% over 2 years (0.37% over 10 years) for a 1% of GDP deficit shock.
- This implies the 2021 deficits (estimated at 13% of GDP) caused around 30% of the excess inflation in 2021-2022.
- A standard Heterogeneous Agent New Keynesian (HANK) model, calibrated to pre-2020 data, can quantitatively match the estimated inflation multiplier when the measured deficit shock is incorporated.
- Loose monetary policy in response to the fiscal shock appears important for explaining the size of the inflation response.
- The results suggest deficits were an important but not the sole cause of post-pandemic inflation.
The authors argue this approach allows for better identification of the causal effect of deficits on inflation by separating it from other confounding shocks through the use of high-frequency data around a specific event.
Do Deficits Cause Inflation? A High Frequency Narrative Approach
Authors: Jonathon Hazell, Stephan Hobler
From: LSE
The deficit-inflation nexus: bridging HANK and FTPL
In this paper Angeletos et al. compare and contrast two theories of how fiscal deficits drive inflation – the Fiscal Theory of the Price Level (FTPL) and Heterogeneous Agent New Keynesian (HANK) models. The key methodological approaches include:
- Developing analytical models of both FTPL and HANK frameworks to compare their predictions about the relationship between deficits and inflation.
- Extending the baseline models to incorporate features like long-term government debt, heterogeneous household types, and hybrid Phillips curves.
- Calibrating a quantitative HANK model using empirical evidence on key parameters like marginal propensities to consume, wealth distribution, and price rigidities.
- Conducting numerical simulations of the calibrated model to quantify the inflationary effects of deficit shocks under different assumptions.
The main findings are:
- Despite different underlying mechanisms, HANK can replicate key predictions of FTPL about how deficits drive inflation when fiscal adjustment is sufficiently delayed.
- The HANK framework provides a more robust foundation for the deficit-inflation relationship, avoiding some theoretical controversies surrounding FTPL.
- In HANK, the inflation response to deficits tends to be more front-loaded and short-lived compared to FTPL.
- Quantitative analysis suggests deficits are likely to be less inflationary in practice than predicted by simple FTPL arithmetic, with the cumulative inflation response cut by about half.
- The dampened inflation effect is due to factors like the tax base “self-financing” channel and the interaction of inflation front-loading with long-term government debt.
- An application to post-COVID fiscal stimulus predicts material but dampened inflationary effects compared to simple FTPL predictions.
Overall, the paper argues HANK provides a more empirically grounded framework for understanding deficit-driven inflation, while still capturing key insights from FTPL. The quantitative results suggest meaningful but more modest inflationary effects from deficits than simple FTPL arithmetic would imply.
Deficits and Inflation: HANK meets FTPL
Authors: George-Marios Angeletos, Chen Lian, Christian K. Wolf
From: Northwestern University, UC Berkeley, MIT
