Unconventional fiscal policy to exit the COVID-19 crisis

Francesco D'Acunto, Daniel Hoang, Michael Weber 08 June 2020



As many economies are planning a slow reopening after the COVID-19 containment measures, the question of how to jumpstart the economy and stimulate aggregate demand has taken centre stage (Furman 2020). When conventional monetary policy is constrained and high debt burdens limit the scope of conventional fiscal policy, unconventional policies enter the policy toolbox.

On 3 June 2020, the German government released the measures that will constitute an unprecedented €130 billion stimulus plan. The most unexpected and prominent measure is a form of unconventional fiscal policy (Feldstein 2002, D’Acunto et al. 2016, 2018), whereby the government changes VAT rates to create a future path of increasing sales taxes and hence stimulate inflation expectations and spending at impact. Germany will implement an unexpected, sudden, and temporary drop in VAT, starting in July 2020 until the end of 2020, which will increase households’ expected inflation for 2021 and hence should stimulate spending for the rest of 2020, until prices increase again. The same exact effect could also be obtained by pre-announcing a future increase in VAT, without first cutting the tax. Both measures incentivise households to move their spending forward as long as VAT is lower, although the measures differ in terms of implications for government budgets. 

Unconventional fiscal policy is part of a set of non-conventional policies designed and implemented to tackle the main conundrum of the euro area since the Great Recession:  stimulating inflation expectations and ultimately aggregate demand in a setting in which traditional monetary policy measures are not viable, and governments can barely support growth with fiscal spending because of their large debt-to-GDP ratios. Another popular unconventional policy that has entered the standard toolbox of many central banks is forward guidance – central banks’ explicit commitment to keep interest rates low until after the end of a liquidity trap, which will generate inflation then and hence should increase inflation expectations today. 

In theory, unconventional fiscal policy and forward guidance act through the same channels. They both aim to increase households’ inflation expectations in times in which the effective lower bound on nominal rates binds, and hence changes in inflation expectations should translate one-to-one into changes in perceived real rates by households. Because households perceive lower real interest rates, their incentives to save decline and their willingness to consume should increase based on the consumer Euler equation.  These channels are unconventional in the sense that policymakers aim to manage households’ behaviour directly, as opposed to changing the incentives of financial intermediaries and firms and transmitting policy measures indirectly to households. 

Despite the same theoretical channels, in the aggregate, forward guidance seems to have been ineffective (Del Negro et al. 2015), and only scant evidence exists to assess the effectiveness of unconventional fiscal policy (D’Acunto et al. 2019). Given the prominence of these measures to jump start the economy after the COVID-19 crisis, in this column, we discuss whether any of these policies appears to have been successful using the same identical microdata and surveys for both policy measures.

Figure 1 reports the central findings in D’Acunto et al. (2020) using data from the harmonised European Commission Consumer Confidence survey for Germany. The figure compares German households’ reaction to a measure of unconventional fiscal policy in 2005 and two forward guidance announcements in 2013-2014.

Figure 1 Inflation expectations and readiness to spend on durables: German households

The top panels report the fraction of the German population expecting higher inflation over the following 12 months; the bottom panels report their average willingness to purchase durable goods. The left panels plot these measures for the unconventional fiscal policy episode. In November 2005, the German government unexpectedly announced a rise in VAT from 16% to 19%, but the increase was only effective in January 2007 and hence, the government engineered a path of future price increases. The right panels plot these measures for the first two explicit forward guidance announcements by former ECB President Draghi in July 2013 and January 2014. 

Although theoretically both policies should raise households' inflation expectations and spending on impact, only unconventional fiscal policy has produced these outcomes in the micro data. Forward guidance announcements do not appear to manage expectations or spending plans. The raw data time-series evidence in Figure 1 is interesting, but any unobserved shock contemporaneous to the policy announcements could explain those dynamics.

Methodologically, we need a scientific approach that helps reduce the endogeneity concerns inherent to these policy measures. We propose an identification strategy in the spirit of Poterba (1996), in which other EU consumers not exposed directly to the German VAT shock and/or to the ECB measures of forward guidance act as counterfactual for the behaviour of Germans had the policy announcements not happened. Because different households with different demographics react differently to changes in inflation expectations (Bachmann et al. 2015), we match German households to similar foreign households based on observables. Moreover, our data allow us to control directly for important determinants of spending plans, such as households' income expectations, financial outlook, employment status, and housing choices.

Figure 2 Average treatment effects on durable purchases

Figure 2 plots the average treatment effect (ATE) of the unconventional fiscal policy announcement on durable purchases (top panel) and of forward guidance (bottom panel). We find no differences in the readiness to spend on durable goods between German and matched households before the announcement of the VAT change. Starting in December 2005, after the future VAT change was announced, German households' willingness to spend increased relative to matched households: German households were 3.8 percentage points more likely to declare it was a good time to purchase durable goods after the announcement relative to before and to matched foreign households. The effect increased in magnitude throughout 2006 and peaked at 34 percentage points in November 2006. The average treatment effect dropped to zero in January 2007 once the VAT increased and higher inflation materialised. 

For forward guidance, instead, we find no difference in the readiness to spend on durable goods between German and matched households before the first forward guidance announcement in July 2013. Germans' propensity to spend did not change after the first forward guidance announcement relative to before and relative to matched foreign households. Even around the second announcement in January 2014, the announcement had no effect on the willingness to spend. The propensity to spend did not move after the second announcement either. Ultimately, forward guidance announcements had no effect on impact nor delayed indirect effects through financial intermediaries, such as due to an effect of lower long-term interest rates on credit-financed consumption.

Unconventional fiscal policy

Given the novelty of unconventional fiscal policy, we now discuss in more detail its features and intellectual background. Feldstein (2002) introduced the notion of unconventional fiscal policy measures at times of liquidity traps. Among several possible interventions, he proposed a series of pre-announced increases in VAT to generate consumer price inflation, and hence increase private spending via intertemporal substitution. In his words: “This [VAT] tax-induced inflation would give households an incentive to spend sooner rather than waiting until prices are substantially higher.” The intuition for this proposal is based on a simple logic: announcing higher prices in the future – either with a sudden but temporary cut, or with an outright future increase in VAT – will increase current inflation expectations. Higher inflation expectations at times of fixed nominal interest rates should reduce real interest rates (Fisher equation), and lower real interest rates should increase households’ incentives to consume rather than save (Euler equation). Because imposing higher VAT reduces households’ wealth (especially poorer households’ wealth) and might affect households’ labour supply, lower income taxes (or transfers to those households that do not pay any income tax) should accompany the increase in VAT. 

Designing such a measure with a sudden but temporary VAT cut with a pre-specified date in which the VAT will revert back to its original level is exactly what the German administration announced as part of their €130 billion stimulus package. A pre-announced tax increase and a sudden temporary drop in VAT differ in terms of budget neutrality: the former policy is budget neutral for the government if paired with income tax cut or even generates revenues, whereas the latter policy creates budget deficits. In both cases, unconventional fiscal policy incentivises households to consume immediately, jump-tart the economy, and hence help the economy exit the slump. 

In his presidential address to the 2011 American Economic Association Annual Meeting, Bob Hall (2011) reiterated Feldstein’s ideas, and encouraged further research to understand the viability and effects of unconventional fiscal policy, both theoretically and empirically. In the US, the proposal of announcing a national sales tax to take effect on a specified date in the future to quicken up the recovery at times of economic downturns was advanced as early as 1991, in an op-ed for the New York Times by Matthew Shapiro (Shapiro, 1991). In Shapiro’s words, “How would such a proposal work? […] Consumers, anticipating the tax increase, would accelerate their purchases, particularly of durable goods. This would stimulate the economy immediately, though there would be no immediate direct impact on the deficit."

The first theoretical formalisation of Feldstein’s proposal was in Correia et al. (2013). The authors show unconventional fiscal policies, including higher future consumption taxes, can completely offset the zero lower bound constraint on nominal interest rates by generating consumer price inflation without distorting production decisions. When higher future consumption taxes are paired up with cuts to income taxes, the economy can reach a first-best outcome, despite the distortionary nature of sales taxes on the intra-temporal margin of households. Overall, unconventional fiscal policies counteract deflation and spur economic growth at times of high output gaps. Crucially, they do not worsen the governmental budget, and keep constant the tax burden on households. The only mechanism at play is the inter-temporal substitution of consumption. These measures should be more relevant for durable and storable goods, which represent the most elastic component of households’ consumption bundle.

Forward guidance

Eggertson and Woodford (2003) introduce forward guidance theoretically as an unconventional monetary policy tool to circumvent a lower bound on nominal interest rates. Promises to keep interest rates at zero until after the end of the liquidity trap will be inflationary in the future. Households, anticipating this future inflationary pressure will therefore already update their inflation expectations today which, in times of constant nominal rates, translates into lower real interest rates. According to the consumer Euler equation, lower real then lower the incentives to save and increase the incentives to consume. 

One concern with this type of policy is time consistency. Central banks will have to deviate from their conventional reaction function in the future and tolerate higher inflation which might limits its effectiveness today if agents do not perceive this promise credible. During our sample period, both financial markets and professional forecasters reacted to the two forward guidance announcements, which makes lack of credibility an unlikely explanation for the ineffectiveness to stimulate households’ expectations and spending plans. 

What should the ECB and European governments do?

The theoretical and empirical research discussed above has clear-cut implications for policymakers, especially as they plan measures to jumpstart the economy after the COVID-19 crisis.  Unconventional fiscal policy can successfully increase consumer spending and hence growth today based on its effectiveness from earlier episodes. This reaction is in stark contrast with forward guidance measures, which appear rather ineffective, at least in part arguably because households do not understand their implications and hence, contrary to experts, do not react to them (D’Acunto et al. 2019). 

On the political economy side, unconventional fiscal policy might also be easier to implement in a consensus-based institutional setup like the current European Council relative to other fiscal and monetary proposals.  They could satisfy both the central and northern European countries which would benefit relatively more than southern countries given the higher levels of VAT evasion in southern countries, as well as the southern countries which want to exploit fiscal policy to spur growth without further increasing the high tax burden on households. Because these measures do not involve monetary easing and open a path to future interest rate increases because of the inflation it generates, they would perhaps not be opposed by the banking and corporate sectors of central European countries either. 


Bachmann, R, T Berg, and E Sims (2015), “Inflation expectations and readiness to spend: Cross-sectional evidence”, American Economic Journal: Economic Policy 7(1): 1-35.

Bernanke, B (2010), “Monetary Policy Objectives and Tools in a Low-inflation Environment”, speech at the “Revisiting Monetary Policy in a Low-Inflation Environment” conference, Federal Reserve Bank of Boston, 15 October.

Blanchard, O, G Dell'Ariccia, and P Mauro (2010), “Rethinking Macroeconomic Policy”, Journal of Money, Credit and Banking 42(6): 199-215.

D’Acunto, F, D Hoang, and M Weber (2016), “The Effect of Unconventional Fiscal Policy on Consumption Expenditure”, NBER Working Paper 22563.

D’Acunto, F, D Hoang, and M Weber (2018), “Unconventional Fiscal Policy”, AEA Papers and Proceedings 108(5): 519-23.

D’Acunto, F, D Hoang, M Paloviita, and M Weber (2019), “Human Frictions in the Transmission of Economic Policies”, Working Paper.

Draghi, M (2016), “How Central Banks Meet the Challenge of Low Inflation”, Marjolin Lecture. 

Furman, J (2020), “Protecting people now, helping the economy rebound later”, VoxEU.org, 31 May.

Correira, I, E Farhi, J P Nicolini and P Teles (2013), “Unconventional Fiscal Policy at the Zero Bound”, American Economic Review 103(4): 1172-1211. 

Feldstein, M (2003), “A Role for Discretionary Fiscal Policy in a Low Interest Rate Environment”, in Rethinking Stabilization Policy, 2002 Federal Reserve Bank of Kansas City Annual Conference volume.

Hall, R. (2011), “The long slump”,  American Economic Review 101(2): 431-469.

Mian, A and A Sufi (2012), “The Effects of Fiscal Stimulus: Evidence from the 2009 ‘Cash for Clunkers’ Program”, Quarterly Journal of Economics 127(3): 1107-1142. 

Poterba, J (1996), “Retail price reactions to changes in state and local sales taxes”, National Tax Journal 49(2): 165-176.

Shapiro, M (1991), “Economic Stimulant: Sales Tax”, New York Times, 16 December.



Topics:  Covid-19 Monetary policy Taxation

Tags:  COVID-19, coronavirus, unconventional fiscal policy, forward guidance, VAT

Assistant Professor of Finance at the Carroll School of Management, Boston College

Assistant Professor, Karlsruhe Institute of Technology

Associate Professor of Finance, University of Chicago Booth School of Business


CEPR Policy Research