Inflation remains persistent. In this note, we review one of the possible sources of this persistency: the interaction between monetary and fiscal policy. We do so by presenting 4 relevant papers in the so-called “Fiscal Theory of the Price Level” (or FTPL) and related literature. (“The fiscal roots of inflation” by John Cochrane (Stanford), A Fiscal Theory of Persistent Inflation by Francesco Bianchi (Johns Hopkins University), Renato Faccini (Danmarks Nationalbank), and Leonardo Melosi (Chicago Fed), Fiscal Influences on Inflation in OECD Countries, 2020-2022 by Robert J. Barro (Harvard) and Francesco Bianchi (Johns Hopkins), The Monetary Financing of a Large Fiscal Shock by Pedro Teles (Banco de Portgual) and Oreste Tristani (ECB), and “Do Large Fiscal Deficits Cause Inflation? The Historical Record” by Michael D. Bordo (Rutgers University) and Mickey D. Levy (Berenberg).
We begin by giving the intuition behind the FTPL. Part I of this note reviews the first two papers. Part II reviews the remaining three papers. Finally, we will circulate FRB-US based simulations on possible fiscal consolidation paths in the US.
The Fiscal Theory of the Price Level in three paragraphs
The fiscal theory of the price level starts from the intertemporal government budget constraint, an important identity in macroeconomic models about the financing need of the government. In this equation, the real value of current nominal government debt equals the present value of current and future real primary surpluses. The intuition is straightforward: The real value of nominal government debt needs to be repaid at some point, and the government has two options of achieving this. Either it inflates away the debt or it achieves real surpluses in the future.
The intertemporal government budget constraint is an identity and does not directly provide a causal interpretation of the relation between the real value of government debt and the present value of real surpluses. The fiscal theory of the price level (FTPL) takes a stance on the direction of causality: It states that government debt causes inflation. Unless the government can credibly signal that it will achieve higher surpluses in the future, agents are aware that the debt must be inflated away. The expectation of inflation in the future causes inflation today as households decide to spend before prices rise.
This interpretation is the opposite of the standard logic of New Keynesian models, the workhorse framework used by modern central banks. While the same government budget constraint holds in these models, they assume that the central bank determines unexpected inflation. Fiscal policy reacts to these changes in inflation by raising or lowering surpluses such that the inflation-induced changes in the value of government debt are consistent with the budget constraint.
The theory has received particular attention in light of the recent surge in inflation, which followed strong fiscal expansions in the United States and Europe. Whether this mechanism is truly behind the ups and downs of inflation is subject to an active debate and ultimately an empirical question. We review four papers that study the implications of the FTPL in the data.
Paper #1: Cochrane (2022)
The paper “The fiscal roots of inflation” by John Cochrane (Stanford) takes the Fiscal Theory of the Price Level to the data. It uses the intertemporal government budget constraint to derive a linear expression for the relationship between inflation rates and future surpluses, GDP growth rates, and discount rates. Each of these objects maps directly to an impulse response from a VAR, which can be used to study the quantitative importance of each of the components in explaining inflation. Discount rates emerge as the most important contributor.
What the paper does
Starting from the intertemporal government budget constraint, an unexpected increase in inflation must correspond to one of three outcomes (or any combination thereof): 1) higher surpluses relative to GDP, 2) lower GDP growth rates, or 3) higher discount rates for future surpluses. This ensures that the real value of nominal debt equals the present value of real primary surpluses.
Main results
The paper takes this theory to the data using a vector autoregression for the United States. The impulse responses map directly to objects from the theory-based decomposition of an unexpected inflation increase. Cochrane studies the response of the surplus-to-GDP ratio, GDP growth rates, and discount rates to an unexpected 1% increase in inflation. Figure 1 shows the results. The rise in inflation is moderately persistent (blue line). Negative surpluses build up with a hump shape but then turn positive (orange). The GDP growth rate falls (yellow). The real discount rate rises after four years and stays positive (purple; plotted as return for the previous period, hence the missing value at the start).
The response to an inflation shock can be decomposed as follows. The weighted sum of inflation rates is 1.59%. The rise in inflation is matched mostly by a rise in the discount rate (1.04%) and less by a decline in surplus-to-GDP (0.55%). Discount rates therefore drive a quantitatively important wedge between inflation rates and government spending. Over the horizon of the impulse response, most of the decline in surplus-to-GDP is driven by lower growth as opposed to a decline in the level of surpluses: the growth rate of GDP contributes 0.49% versus 0.06% for the surplus-to-GDP ratio.
The scenario corresponds to a stagflationary episode in which inflation and growth move in opposite directions. Instead, Cochrane also studies a recession scenario in which both inflation and GDP growth fall by 1% on impact. Government surpluses are large in this scenario, but inflation remains weak. Why? Discount rates are low because interest rates fall in recessions, which increases the present value of future surpluses.
Figure 1. Main result of Cochrane (2022). Impulse responses to a 1% inflation shock.
A critique of the paper
This paper is an important starting point in bringing the FTPL to the data. The paper provides some general facts for monetary-fiscal interaction that are a useful for a wide range of applications. The results provide a first benchmark for the quantitative importance of different channels through which inflation and the fiscal side may interact.
The theory-based decomposition allows to make sense of the state dependence of the inflation response to government spending. Inflation is absent in recessions even though government deficits are high, but this need not contradict the fiscal theory of the price level. There are multiple ways to equate the real value of nominal government debt with the present value of real surpluses, and discount rates emerge as an important player. While discount rates have been at the centre of academic attention in asset pricing, they may also require more attention in the study of the roots of inflation. Inflation surprises are cases in which discount rates and deficits do not offset.
None of the shocks are structural. The paper does not attempt to provide a deep structural interpretation of the “inflation shock”. The “shock” is simply a catch-all term for any fundamental driver that leads to a surprise increase in inflation. This is useful to describe patterns in the data. However, more work needs to be done to understand the causal drivers of the rise in inflation, and to understand in which case it is truly linked to government spending increases as opposed to other shocks.
Policy implications
Fiscal expansion can result in inflation, but only when interest rates do not fall enough. The results from this paper strongly suggest that monetary-fiscal interactions are central to understanding the inflationary consequences of government spending. If the central bank lowers interest rates, the present value of future fiscal surpluses rises, and the current level of government debt can be financed without inflation. In contrast, inflationary periods are typically characterized by a combination of government deficits and rising discount rates.
Paper #2: Bianchi et al. (2023)
The paper A Fiscal Theory of Persistent Inflation by Francesco Bianchi (Johns Hopkins University), Renato Faccini (Danmark Nationalbank), and Leonardo Melosi (Chicago Fed) studies the role of partially unfunded debt in causing persistent inflation. The authors add unfunded fiscal shocks to a workhorse New Keynesian model and find that these shocks can explain a significant share of US inflation variation. The model suggests the USD 1.9 trillion spent under the 2021 American Rescue Act Plan significantly contributed to post-pandemic inflation.
What the paper does
The paper introduces a new, general class of equilibrium models with partially unfunded debt. Within the model, the central bank can respond differently to different shocks. The central bank generally dominates the fiscal authority in that it sets the interest rate to stabilize inflation following a shock (Taylor principle) and the fiscal authority needs to adjust its debt stabilization policy accordingly. However, the fiscal authority can send partially unfunded transfers to households and the central bank does not adjust its interest rate in response to the inflation induced from these transfer payments, allowing the government to inflate away its debt. The persistence of the inflation response depends on the Taylor rule coefficient that the central bank puts on inflation arising from unfunded fiscal shocks. In contrast, funded fiscal shocks have no effect on inflation because agents expect higher government surpluses in the future (Ricardian equivalence).
Main results
Fiscal inflation accounts for the bulk of US inflation dynamics since the 1960s. Figure 2 shows a decomposition of YoY US inflation rates into the different shocks: The black bars show the inflation contribution of unfunded fiscal shocks, the grey bars refer to other policy shocks, and the white bars capture contributions from any non-policy shocks. The rise in inflation from the mid-1960s to mid-1970s is largely explained by unfunded fiscal shocks, especially due to President Johnson’s Great Society initiatives. Likewise, the Volcker disinflation can be attributed to a fall in fiscal inflation. This does not mean that fiscal policy alone drove the inflation decline: Instead, Volcker’s commitment to stable prices signaled a change in the Fed’s willingness to tolerate inflation from unfunded fiscal shocks. The decline of inflation in the 1980s should therefore be viewed as a driven by monetary-fiscal interactions.
Figure 2. Shock decomposition of YoY US inflation (red line) by Bianchi et al. (2023). Black bars show the contribution of unfunded fiscal shocks, grey bars other policy shocks, and white bars non-policy shocks.
Post-pandemic inflation can be largely explained by the American Rescue Action Plan (ARPA). While the 2020 CARES act already contributed to inflation and brought the economy back onto a recovery path, the ARPA led to a strong rise in unfunded transfers. This may partly be explained by the change in the Fed’s monetary policy framework in Q3 2020, which signaled that inflation would be tolerated above target in the short run and increased the share of unfunded transfers. Model simulations show that peak inflation would have been 3.1pp weaker under a fully funded ARPA, but also that the real economy would have recovered significantly less strongly. Out-of-sample predictions with real-time data show that the model tracks post-pandemic inflation well when the ARPA is considered, but the forecast for peak inflation falls to 3.7% without the ARPA.
A critique of the paper
The paper takes a state-of-the-art theoretical framework to the data to generate new empirical insights. Much discussion on the FTPL has either been purely theoretical or made use of anecdotal evidence without proper identification. Empirical work has often focused on government spending without being able to carefully consider the role of monetary-fiscal interactions, which are key. This paper deviates from existing approaches by estimating a quantitative model on the data, providing a direct link between theory and empirics. The model allows to decompose government transfers into funded and unfunded components and can be used for policy counterfactuals. The authors’ ability to do an out-of-sample forecasting exercise for the post-pandemic inflation period using the same framework is impressive.
The paper provides a new narrative of the history of US inflation. Fiscal inflation takes center stage in this interpretation and explains even the rise and fall of the Great Inflation through unfunded fiscal transfers that can be traced back to the 1960s. That being said, more empirical work could be done on testing the role of unfunded transfers with alternative methods that allow for causal interpretation from econometric restrictions as opposed to causality implied by theoretical underpinnings of the model equations. This would give additional confidence in the interpretation provided by the authors.
Assumptions about rational expectations and full information appear too restrictive. These assumptions are a natural starting point and allow the model to be directly comparable to standard New Keynesian models. As the authors acknowledge, policymakers may not be able to perfectly coordinate and to communicate the unfunded fraction of a fiscal expansion. The speed with which household beliefs about debt sustainability adjust may therefore be an important determinant of the transmission mechanism. Sudden exogenous belief changes could also be a new source of fiscal inflation, similar to the prominent role of sentiment in Pigouvian business cycles.
Policy implications
Monetary-fiscal interaction is at the heart of fiscal inflation and may be the key driver of US inflation dynamics. The theory carefully highlights the importance of coordination among policy institutions: Funded fiscal transfers are not inflationary, while unfunded transfers can have strong inflationary effects. Changes in the monetary policy framework can therefore have crucial implications for the effects of fiscal transfers. Policy communication also becomes key: The efficacy of fiscal policy may be limited if households do not believe in the central bank’s willingness to tolerate fiscal inflation.
Fiscal policy can be an effective tool to escape the ELB. Unfunded fiscal transfers can be highly inflationary and also come with significant real effects. Last decade’s calls of advanced economy central banks for stronger fiscal expansion may be interpreted through this lens. At the same time, the potency of unfunded fiscal shocks requires a carefully calibrated fiscal response to avoid overshooting as experienced in the post-pandemic era. Close coordination between the central bank and the Treasury would be ideal except for obvious political economy concerns. With many structural drivers of the secular interest rate decline persisting, these questions may become relevant again once the post-pandemic inflation has cooled down.