Modeling the Consumption Response to the CARES Act

Finance and Economics Discussion Series Divisions of Research & Statistics and Monetary Affairs Federal Reserve Board, Washington, D.C. Modeling the Consumption Response to the CARES Act Christopher D. Carroll, Edmund Crawley, Jiri Slacalek, Matthew N. White 2020-077 Please cite this paper as: Carroll, Christopher D., Edmund Crawley, Jiri Slacalek, and Matthew N. White (2020). “Modeling the Consumption Response to the CARES Act,” Finance and Economics Discussion Series 2020-077. Washington: Board of Governors of the Federal Reserve System, https://doi.org/10.17016/FEDS.2020.077. NOTE: Staff working papers in the Finance and Economics Discussion Series (FEDS) are preliminary materials circulated to stimulate discussion and critical comment. The analysis and conclusions set forth are those of the authors and do not indicate concurrence by other members of the research staff or the Board of Governors. References in publications to the Finance and Economics Discussion Series (other than acknowledgement) should be cleared with the author(s) to protect the tentative character of these papers.

ConsumptionResponse,2020/05/04,15:50 Modeling the Consumption Response to the CARES Act April 24, 2020 Christopher D. Carroll1 Edmund Crawley2 Jiri Slacalek3 JHU FRB ECB Matthew N. White4 UDel Abstract To predict the effects of the 2020 U.S. CARES Act on consumption, we extend a model that matches responses of households to past consumption stimulus packages. The extension allows us to account for two novel features of the coronavirus crisis. First, during the lockdown, many types of spending are undesirable or impossible. Second, some of the jobs that disappear during the lockdown will not reappear when it is lifted. We estimate that, if the lockdown is short-lived, the combination of expanded unemployment insurance benefits and stimulus payments should be sufficient to allow a swift recovery in consumer spending to its pre-crisis levels. If the lockdown lasts longer, an extension of enhanced unemployment benefits will likely be necessary if consumption spending is to recover. Keywords Consumption, Coronavirus, Stimulus JEL codes D83, D84, E21, E32 econ-ark.github.io/Pandemic HTML version of paper Interactive-Jupyter-Notebook Allows user to modify some assumptions github.com/econ-ark/Pandemic Full codebase; explore all assumptions LaTeX subdirectory of ↑ PDF version of paper LaTeX subdirectory of ↑ Presentation slides 1Carroll: DepartmentofEconomics,JohnsHopkinsUniversity,http://econ.jhu.edu/people/ccarroll/,ccarroll@jhu.edu 2Crawley: FederalReserveBoard,edmund.s.crawley@frb.gov 3Slacalek: DGResearch,EuropeanCentralBank,http://www.slacalek.com/, jiri.slacalek@ecb.europa.eu 4White: DepartmentofEconomics,UniversityofDelaware,mnwecon@udel.edu ThankstotheConsumerFinancialProtectionBureauforfundingtheoriginalcreationoftheEcon-ARKtoolkit,whoselatest versionweusedtoproducealltheresultsinthispaper;andtotheSloanFoundationforfundingitsextensivefurtherdevelopment that brought it to the point where it could be used for this project. The toolkit can be cited with its digital object identifier, 10.5281/zenodo.1001067, as is done in the paper’s own references as Carroll, Kaufman, Kazil, Palmer, and White (2018). We are gratefultoKiichiTokuoka,whoprovidedvaluablefeedbackandinputasthisprojectprogressed,andtoMridulSeth,whocreated the dashboard and configurator. The views presented in this paper are those of the authors, and should not be attributed to the FederalReserveBoardortheEuropeanCentralBank.

“Economic booms are all alike; each recession contracts output in its own way.” — with apologies to Leo Tolstoy I Introduction In the decade since the Great Recession, macroeconomics has made great progress by insisting that models be consistent with microeconomic evidence (see Krueger, Mitman, and Perri (2016) forasurvey). Fromthenewgenerationof models, wetakeonespecificallyfocusedon reconciling apparent conflicts between micro and macro evidence about consumption dynamics, as documentedinHavranek,Rusnak,andSokolova(2017),andadaptittoincorporatetwoaspectsofthe coronavirus crisis. First, because the tidal wave of layoffs for employees of shuttered businesses will have a large impact on their income and spending, assumptions must be made about the employment dynamics of laid off workers. Second, even consumers who remain employed will have restricted spending options (nobody can eat dinner at a shuttered restaurant). On the first count, we model the likelihood that many of the people unemployed during the lockdown will be able to find work again quickly by assuming that the typical job loser has a two-thirds chance of being reemployed after each quarter of unemployment. This ‘normal’ unemploymentisthesameasexperiencedinthemodelwithoutapandemic. However, weexpect that some kinds of jobs will not come back quickly after the lockdown,1 and that people who workedinthesekindsofjobswillhavemoredifficultyfindinganewjob. Wecallthesepeoplethe ‘deeply unemployed’ and assume that there is a one-third chance each quarter that they become merely ‘normal unemployed.’ The ‘normal unemployed’ have a jobfinding rate that matches average historical unemployment spell of 1.5 quarters (as a ‘normal unemployed’ person). Thus a deeply unemployed person expects to remain in the ‘deep unemployment’ state for three quarters on average, and then remain unemployed for another one and a half quarters. When the pandemic hits, we assume that 10 percent of model households become normal unemployed and an additional 5 percent become deeply unemployed; in line with empirical evidence, the unemployment probabilities are skewed toward households who are young, unskilled and have low income. (All of these assumptions can be adjusted using our dashboard; changing several parameters simultaneously requires installation of the software toolkit). On the second count, we model the restricted spending options by assuming that during the lockdown spending is less enjoyable (there is a negative shock to the ‘marginal utility of consumption.’) Based on a tally of sectors that we judge to be substantially shuttered during the ‘lockdown,’ we calibrate an 11 percent reduction to spending. Thus households prefer to defer some of their consumption into the future, when it will yield them greater utility. (See Carvalho, Garcia, Hansen, Ortiz, Rodrigo, Rodriguez, and Ruiz (2020) for Spanish data already showing a strong effect of this kind in recent weeks, and Andersen, Hansen, Johannesen, and Sheridan (2020) for similar evidence from Denmark).2 In our primary scenario, we assume that this condition is removed with probability one-half after each quarter, so on average remains for two quarters. When the ‘lockdown’ ends, the buildup of savings by households who did not lose their jobs, but whose spending was suppressed, should result in a partial recovery in consumer spending. However, without the CARES Act, total consumer spending remains below its pre-crisis peak through the foreseeable future. 1Thecruiseindustry,forexample,islikelytotakealongtimetorecover. 2A shock to marginal utility captures, in a reduced form, the essence of what depresses consumption spending, and is a kind ofshockcommonlystudiedintheliterature. AppendixCshowsitisequivalenttoanincreaseinthequality-adjustedpriceofsome typesofgoods(e.g. eatingoutandvacations). 2

Our model captures the two primary features of the CARES Act that aim to bolster consumer spending: 1. The boost to unemployment insurance benefits, amounting to $7,800 if unemployment lasts for 13 weeks. 2. The direct stimulus payments to most households, up to $1,200 per adult. Weestimatethatthecombinationofexpandedunemploymentinsurancebenefitsandstimulus payments should be sufficient to expect a swift recovery in consumer spending to its pre-crisis levels under our default description of the pandemic, in which the lockdown ends after two quarters on average. Overall, unemployment benefits account for about 30 percent of the total aggregate consumption response and stimulus payments explain the remainder. Our analysis partitions households into three groups based on their employment state when the pandemic strikes and the lockdown begins. First, households in our model who do not lose their jobs initially build up their savings, both because of the lockdown-induced suppression of spending and because most of these households will receive a significant stimulus check, much of which the model says will be saved. Even without the lockdown, we estimate that only about 20 percent of the stimulus money would be spent immediately upon receipt, consistent with evidence from prior stimulus packages about spending on nondurable goods and services. Once the lockdown ends, the spending of the always-employed households rebounds strongly thanks to their healthy household finances. The second category of households are the ‘normal unemployed,’ job losers who perceive that it is likely they will be able to resume their old job (or get a similar new job) when the lockdown is over. Our model predicts that the CARES Act will be particularly effective in stimulating their consumption, given the perception that their income shock will be largely transitory. Our model predicts that by the end of 2021, the spending of this group recovers to the level it would have achieved in the absence of the pandemic (‘baseline’); without the CARES Act, this recovery would take more than a year longer. Finally, for households in the deeply unemployed category, our model says that the marginal propensity to consume (MPC) from the checks will be considerably smaller, because they know they must stretch that money for longer. Even with the stimulus from the CARES Act, we predict that consumption spending for these households will not fully recover until the middle of 2023. Even so, the Act makes a big difference to their spending, particularly in the first six quarters after the crisis. For both groups of unemployed households, the effect of the stimulus checks is dwarfed by the increased unemployment benefits, which arrive earlier and are much larger (per recipient). Perhaps surprisingly, we find the effectiveness of the combined stimulus checks and unemployment benefits package for aggregate consumption is not substantially different from a package that distributed the same quantity of money equally between households. The reason for this is twofold: first, the extra unemployment benefits in the CARES Act are generous enough that many of the ‘normally’ unemployed remain financially sound and can afford to save a good portion of those benefits; second, the deeply unemployed expect their income to remain depressed for some time and therefore save more of the stimulus for the future. In the model, the fact that they do not spend immediately is actually a reflection of how desperately they anticipate these funds will be needed to make it through a long period of low income. While unemployment benefits do not strongly stimulate current consumption of the deeply unemployed, they do provide important disaster relief for those who may not be able to return to work for several quarters (see Krugman (2020) for an informal discussion). 3

Inadditiontoourprimaryscenario’srelativelyshortlockdownperiod,wealsoconsideraworse scenario in which the lockdown is expected to last for four quarters and the unemployment rate increases to 20 percent. In this case, we find that the return of spending toward its no-pandemic path takes roughly three years. Moreover, the spending of deeply unemployed households falls steeply unless the temporary unemployment benefits in the CARES Act are extended for the duration of the lockdown. Our modeling assumptions — about who will become unemployed, how long it will take them to return to employment, and the direct effect of the lockdown on consumption utility — could prove to be off, in either direction. Reasonable analysts may differ on all of these points, and prefer a different calibration. To encourage such exploration, we have made available our modeling and prediction software, with the goal of making it easy for fellow researchers to test alternative assumptions. Instructions for installing and running our code can be found here; alternatively, adjustments to our parametrization can be explored with an interactive dashboard here. There is a potentially important reason our model may underpredict the bounceback in consumer spending when the lockdown ends: ‘pent up demand.’ This term captures the fact that purchases of ‘durable’ goods can be easily postponed, but that when the reason for postponement abates some portion of the missing demand is made up for. (We put ‘durable’ in quotes because ‘memorable’ goods (Hai, Krueger, and Postlewaite (2013)) have effectively the same characteristics.) For simplicity, our model does not include durable goods, because modeling spending on durables is a formidable challenge. But it is plausible that, when the lockdown ends, people may want to spend more than usual on memorable or durable goods to make up for earlier missing spending. Existing Work on the Effects of the Pandemic Many papers have recently appeared on the economic effects of the pandemic and policies to manage it. Several papers combine the classic susceptible–infected–recovered (SIR) epidemiologymodelwithdynamiceconomicmodelstostudytheinteractionsbetweenhealthandeconomic policies (Eichenbaum, Rebelo, and Trabandt (2020) and Alvarez, Argente, and Lippi (2020), among others). Guerrieri, Lorenzoni, Straub, and Werning (2020) shows how an initial supply shock (such as a pandemic) can be amplified by the reaction of aggregate demand. The ongoing work of Kaplan, Moll, and Violante (2020) allows for realistic household heterogeneity in how household income and consumption are affected by the pandemic. Glover, Heathcote, Krueger, and Ríos-Rull (2020) studies distributional effects of optimal health and economic policies. Closest to our paper is some work analyzing the effects of the fiscal response to the pandemic, including Faria-e-Castro (2020b) in a two-agent DSGE model, and Bayer, Born, Luetticke, and Müller (2020) in a HANK model. All of this work accounts for general equilibrium effects on consumption and employment, which we omit, but none of it is based on a modeling framework explicitly constructed to match micro and macroeconomic effects of past stimulus policies, as ours is. A separate strand of work focuses on empirical studies of how the economy reacts to pandemics; see, e.g., Baker, Farrokhnia, Meyer, Pagel, and Yannelis (2020), Jorda, Singh, and Taylor (2020) and Correia, Luck, and Verner (2020). 4

II Modeling Setup A The Baseline Model Ourmodelextendsaclassofmodelsexplicitlydesignedtocapturetherichempiricalevidenceon heterogeneity in the marginal propensity to consume (MPC) across different types of household (employed, unemployed; young, old; rich, poor). This is motivated by the fact that the act distributes money unevenly across households, particularly targeting unemployed households. A model that does not appropriately capture both the degree to which the stimulus money is targeted, and the differentials in responses across differently targeted groups, is unlikely to produce believable answers about the spending effects of the stimulus. Specifically, we use a lifecycle model calibrated to match the income paths of high school dropouts, high school graduates, and college graduates.3 Households are subject to permanent and transitory income shocks, as well as unemployment spells.4 Within each of these groups, we construct an ex ante distribution of discount factors to match their distribution of liquid assets. Matching the distributions of liquid assets allows us to achieve a realistic distribution of marginal propensities to consume according to education group, age, and unemployment status, and thus to assess the impact of the act for these different groups.5 B Adaptations to Capture the Pandemic To model the pandemic, we add two new features to the model. First, our new category of ‘deeply unemployed’ households was created to capture the likelihood that the pandemic will have long-lasting effects on some kinds of businesses and jobs (e.g., the cruise industry), even if the CARES Act manages to successfully cushion much of the financial hit to total household income. Each quarter, our ‘deeply unemployed’ households have a two-thirds chance of remaining deeply unemployed, and a one-third chance of becoming ‘normal unemployed.’ The expected time to employment for a ‘deeply unemployed’ household is four and a half quarters, much longer than the historical average length of a typical unemployment spell. Reflecting recent literature on the ‘scarring effects’ of unemployment spells, permanent income of both ‘normal’ and ‘deeply’ households declines by 0.5 percent each quarter due to ‘skill rot’ (rather than following the default age profile that would have been followed if the consumer had remained employed). Second, a temporary negative shock to the marginal utility of consumption captures the idea that, during the period of the pandemic, many forms of consumption are undesirable or even impossible.6 The pandemic is modeled as an unexpected (MIT) shock, sending many households into both normal and deep unemployment, as well as activating the negative shock to marginal utility. Householdsunderstandandrespondinaforward-lookingwaytotheirnewcircumstances (according to their beliefs about its duration), but their decisions prior to the pandemic did not account for any probability that it would occur. 3ThebaselinemodelisveryclosetothelifecyclemodelinCarroll,Slacalek,Tokuoka,andWhite(2017). 4Householdsexitunemploymentwithafixedprobabilityeachquarter—theexpectedlengthofanunemploymentspellisone andahalfquarters. 5ForadetaileddescriptionofthemodelanditscalibrationseeAppendixA. 6Forthepurposesofourpaper,withlogutility,modelinglockdownsasashocktomarginalutilityisessentiallyequivalentto notallowingconsumerstobuyasubsetofgoods(whicharecombinedintocompositeconsumptionbyaCobb–Douglasaggregator). However,thetwoapproacheswouldyielddifferentimplicationsfornormativeevaluationsofeconomicpolicies. 5

Calibration The calibration choices for the pandemic scenario are very much open for debate. Here we have tried to capture something like median expectations from early analyses, but there is considerable variation in points of view around those medians. Section III.B below presents a more adverse scenario with a long lockdown and a larger increase in unemployment. Unemployment forecasts for Q2 2020 range widely, from less than 10 percent to over 30 percent, but all point to an unprecedented sudden increase in unemployment.7 We choose a total unemployment rate in Q2 2020 of just over 15 percent, consisting of five percent ‘deeply unemployed’ and ten percent ‘normal unemployed’ households. We calibrate the likelihood of becoming unemployed to match empirical facts about the relationship of unemployment to education level, permanent income and age, which is likely to matter because the hardest hit sectors skew young and unskilled.8 Figure 1 shows our assumptions on unemployment along these dimensions. In each education category, the solid line represents the probability of unemployment type (‘normal’ or ‘deep’) for a household with the median permanent income at each age, while the dotted lines represent the probability of unemployment type for a household at the 5th and 95th percentile of permanent income at each age; Appendix A with Table A2 detail the parametrization and calibration we used. To calibrate the drop in marginal utility, we estimate that 10.9 percent of the goods that make up the consumer price index become highly undesirable, or simply unavailable, during the pandemic: food away from home, public transportation including airlines, and motor fuel. We therefore multiply utility from consumption during the period of the epidemic by a factor of 0.891. Furthermore, we choose a one-half probability of exiting the period of lower marginal utility each quarter, accounting for the possibility of a ‘second wave’ if restrictions are lifted too early — see Cyranoski (2020).9 The CARES Act We model the two elements of the CARES Act that directly affect the income of households: • The stimulus check of $1,200 for every adult taxpayer, means tested for previous years’ income.10 • The extra unemployment benefits of $600 for up to 13 weeks, a total of $7,800. For normal unemployed, we assume they receive only $5,200 to reflect the idea that they may not be unemployed the entire 13 weeks. Wemodelthestimuluschecksasbeingannouncedatthesametimeasthecrisishits. However, only a quarter of households change their behavior immediately at the time of announcement, as calibrated to past experience. The remainder do not respond until their stimulus check arrives, 7AsofApril16,about22millionnewunemploymentclaimshavebeenfiledinfourweeks,representingalossofover14percentof totaljobs. JPMorganGlobalResearchforecast8.5percentunemployment(JPMorgan(2020),fromMarch27);TreasurySecretary Steven Mnuchin predicted unemployment could rise to 20 percent without a significant fiscal response (Bloomberg (2020a)); St. LouisFedpresidentJamesBullardsaidtheunemploymentratemayhit30percent(Bloomberg(2020b)—seeFaria-e-Castro(2020a) fortheanalysisbehindthisclaim. BasedonasurveythatcloselyfollowstheCPS,BickandBlandin(2020)calculatea20.2percent unemploymentrateatthebeginningofApril. 8See Gascon (2020), Leibovici and Santacreu (2020) and Adams-Prassl, Boneva, Golin, and Rauh (2020) for breakdowns of whichworkersareatmostriskofunemploymentfromthecrisis. SeeadditionalevidenceinKaplan,Moll,andViolante(2020)and modelingofimplicationsforoptimalpoliciesinGlover,Heathcote,Krueger,andRíos-Rull(2020). 9The CBO expects social distancing to last for three months, and predicts it to have diminished, on average and in line with ourcalibration,bythree-quartersinthesecondhalfoftheyear;seeSwagel(2020). 10Theactalsoincludes$500foreverychild. Inthemodel,anagentissomewherebetweenahouseholdandanindividual. While we do not model the $500 payments to children, we also do not account for the fact that some adults will not receive a check. In aggregateweareclosetotheJointCommitteeonTaxation’sestimateofthetotalcostofthestimuluschecks. 6

Figure 1 Unemployment Probability in Q2 2020 by Demographics 0.20 0.15 0.10 0.05 0.00 ytilibaborP All education (mean) Dropout Unemployed Deep unemp 0.20 0.15 0.10 0.05 0.00 30 40 50 60 Age ytilibaborP Unemployment probability after pandemic shock High school College 30 40 50 60 Age which we assume happens in the following quarter. The households that pay close attention to theannouncementofthepolicyareassumedtobesoforwardlookingthattheyactasthoughthe payment will arrive with certainty next period; the model even allows them to borrow against it if desired.11 The extra unemployment benefits are assumed to both be announced and arrive at the beginning of the second quarter of 2020, and we assume that there is no delay in the response of unemployed households to these benefits. Figure 2 shows the path of labor income — exogenous in our model — in the baseline and in the pandemic, both with and without the CARES Act. Income in quarters Q2 and Q3 2020 is substantially boosted (by around 10 percent) by the extra unemployment benefits and the stimulus checks. After two years, aggregate labor income is almost fully recovered. (See below for a brief discussion of analyses that attempt to endogenize labor supply and other equilibrium variables). III Results This section presents our simulation results for the scenario described above. In addition, we then model a more pessimistic scenario with longer lockdown and higher initial unemployment rate. 11See Carroll, Crawley, Slacalek, Tokuoka, and White (2020) for a detailed discussion of the motivations behind this way of modeling stimulus payments, and a demonstration that this model matches the empirical evidence of how and when households have responded to stimulus checks in the past — see Parker, Souleles, Johnson, and McClelland (2013), Broda and Parker (2014) andParker(2017),amongothers. 7

Figure 2 Labor and Transfer Income 2850 2800 2750 2700 2650 2600 2550 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$ noillib( emocni refsnart dna robal etagerggA Aggregate household income under alternate scenarios Baseline Pandemic, no policy Pandemic, CARES Act Figure 3 Consumption Response to the Pandemic and the Fiscal Stimulus 2850 2800 2750 2700 2650 2600 2550 2500 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$ noillib( noitpmusnoc ylretrauq etagerggA Aggregate consumption under alternate scenarios Baseline Pandemic, no policy Pandemic, CARES Act 8

A Short-lived Pandemic Figure 3 shows three scenarios for quarterly aggregate consumption: (i) the baseline with no pandemic; (ii) the pandemic with no fiscal response; (iii) the pandemic with both the stimuluschecksandextendedunemploymentbenefitsintheCARESAct. Thepandemicreduces consumption by ten percentage points in Q2 2020 relative to the baseline. Without the CARES Act, consumption remains depressed through to the second half of 2021, at which point spending actually rises above the baseline, as a result of the buildup of liquid assets during the pandemic by households that do not lose their income. We capture the limited spendingoptionsduringthelockdownperiodbyareductionintheutilityofconsumption, which makes household save more than they otherwise would usual during the pandemic, with the result that they build up liquid assets. When the lockdown ends, the pent up savings of the always-employed become available to finance a resurgence in their spending, but the depressed spending of the two groups of unemployed people keeps total spending below the baseline until most of them are reemployed, at which point their spending (mostly) recovers while the alwaysemployed are still spending down their extra savings built up during the lockdown. Figure 4 decomposes the effect of the pandemic on aggregate consumption (with no fiscal policy response), separating the drop in marginal utility from the reduction in income due to mass layoffs. The figure illustrates that the constrained consumption choices are quantitatively key in capturing the expected depth in the slump of spending, which is already under way; see Baker, Farrokhnia, Meyer, Pagel, and Yannelis (2020) and Armantier, Kosar, Pomerantz, Skandalis, Smith, Topa, and van der Klaauw (2020) for early evidence. The marginal utility shock hits all households, and directly affects their spending decisions in the early quarters after the pandemic; its effect cannot be mitigated by fiscal stimulus. The loss of income from unemployment is large, but affects only a fraction of households, who are disproportionately low income and thus account for a smaller share of aggregate consumption. Moreover, most households hold at least some liquid assets, allowing them to smooth their consumption drop — the 5 percent decrease in labor income in Figure 2 induces only a 1.5 percent decrease in consumption in Figure 4. Figure 5 shows how the consumption response varies depending on the employment status of households in Q2 2020. For each employment category (employed, unemployed, and deeply unemployed), the figure shows consumption relative to the same households’ consumption in the baseline scenario with no pandemic (dashed lines).12 The upper panel shows consumption without any policy response, while the lower panel includes the CARES Act. The figure illustrates an important feature of the unemployment benefits that is lost at the aggregate level: the response provides the most relief to households whose consumption is most affected by the pandemic. For the unemployed — and especially for the deeply unemployed — the consumption drop when the pandemic hits is much shallower and returns faster toward the baseline when the fiscal stimulus is in place. Indeed, this targeted response is again seen in Figure 6, showing the extra consumption relative to the pandemic scenario without the CARES Act. The dashed lines show the effect of thestimuluscheckinisolation(foremployedworkersthisisthesameasthetotalfiscalresponse). For unemployed households, this is dwarfed by the increased unemployment benefits. These benefits both arrive earlier and are much larger. Specifically, in Q3 2020, when households receive the stimulus checks, the effect of unemployment benefits on consumption makes up 12Householdsthatbecomeunemployedduringthepandemicmightormightnothavebeenunemployedotherwise. Weassume that all households that would have been unemployed otherwise are either unemployed or deeply unemployed in the pandemic scenario. However,therearemanymorehouseholdsthatareunemployedinthepandemicscenariothaninthebaseline. 9

Figure 4 Decomposition of Effect of the Pandemic on Aggregate Consumption (No Policy Response) 0 50 100 150 200 250 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$ noillib( noitpmusnoc etagergga ni egnahC Decomposition of change in consumption from baseline Total effect Income component Marginal utility drop component about 70 percent and 85 percent of the total effect for the normally and deeply unemployed, respectively. Figure 7 aggregates the decomposition of the CARES Act in Figure 6 across all households. In our model economy, the extra unemployment benefits amount to $544 per household, while the stimulus checks amount to $1,054 per household (as means testing reduces or eliminates the stimulus checks for high income households). Aggregated, stimulus checks amount to $267 billion, while the extended unemployment benefits amount to just over half that, $137 billion.13 The figure shows that during the peak consumption response in Q3 2020, the stimulus checks accountforabout70percentofthetotaleffectonconsumptionfortheaveragehouseholdandthe unemployment benefits for about 30 percent. Thus, although the unemployment benefits make a much larger difference to the spending of the individual recipients than the stimulus checks, a small enough proportion of households becomes unemployed that the total extra spending coming from these people is less than the total extra spending from the more widely distributed stimulus checks. The previous graphs show the importance of the targeted unemployment benefits at the individual level, but the aggregate effect is less striking. Figure 8 compares the effect of the CARES Act (both unemployment insurance and stimulus checks) to a policy of the same absolute size that distributes checks to everybody. While unemployment benefits arrive sooner, resulting in higher aggregate consumption in Q2 2020, the un-targeted policy leads to higher aggregate consumption in the following quarters. The interesting conclusion is that, while the net spending response is similar for alternative ways of distributing the funds, the choice to extend unemployment benefits means that much more of the extra spending is coming from the people who will be worst hurt by the crisis. This has obvious implications for the design of any further stimulus packages that might be necessary if the crisis lasts longer than our baseline scenario assumes. 13SeeAppendixBfordetailsonhowweaggregatehouseholds. 10

Figure 5 Consumption Response by Employment Status 14000 13000 12000 11000 10000 9000 8000 7000 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$( noitpmusnoc ylretrauq egarevA Consumption among working age population (no policy) Employed after pandemic Unemployed after pandemic Deeply unemp after pandemic 14000 13000 12000 11000 10000 9000 8000 7000 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$( noitpmusnoc ylretrauq egarevA Consumption among working age population (CARES Act) Employed after pandemic Unemployed after pandemic Deeply unemp after pandemic 11

Figure 6 Effect of CARES Act by Employment Status 1000 800 600 400 200 0 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$( tcA SERAC morf esnopser noitpmusnoC Consumption effect of CARES Act among working age population Employed after pandemic Unemployed after pandemic (Stimulus checks only) Deeply unemp after pandemic (Stimulus checks only) Figure 7 Aggregate Consumption Effect of Stimulus Checks vs Unemployment Benefits 50 40 30 20 10 0 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$ noillib( esnopser noitpmusnoc etagerggA Decomposition of CARES Act effect on aggregate consumption Checks and unemployment benefits Stimulus checks only Unemployment benefits only 12

Figure 8 Effect of Targeting the CARES Act Consumption Stimulus 60 50 40 30 20 10 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$ noillib( esnopser noitpmusnoc etagerggA Effect of targeted stimulus on aggregate consumption Means tested checks and unemployment benefits Equal checks to all households B Alternative Scenario: Long, Deep Pandemic Given the uncertainty about how long and deep the current recession will be, we investigate a more pessimistic scenario in which the lockdown is expected to last for four quarters. In addition, the unemployment rate will increase to 20 percent, consisting of 15 percent of deeply unemployed and 5 percent of normal unemployed. In this scenario we compare how effectively the CARES package stimulates consumption, also considering a more generous plan in which the unemployment benefits continue until the lockdown is over. We model the receipt of unemployment benefits each quarter as an unexpected shock, representing a series of policy renewals. Figure 9 compares the effects of the two fiscal stimulus scenarios on income. The persistently high unemployment results in a substantial and long drop in aggregate income (orange), compared to the no pandemic scenario. The CARES stimulus (green) provides only a short term support to income for the first two quarters. In contrast, the scenario with unemployment benefits extended as long as the lockdown lasts (red) keeps aggregate income elevated through the recession. Figure 10 shows the implications of the two stimulus packages for aggregate consumption. The long lockdown causes a much longer decline in spending than the shorter lockdown in our primary scenario. In the shorter pandemic scenario (Figure 3) consumption returns to the baseline path after roughly one year, while in the long lockdown shown here the recovery takes around three years; that is, the CARES stimulus shortens the consumption drop to about 2 years. The scenario with extended unemployment benefits ensures that aggregate spending returns to the baseline path after roughly one year, and does so by targeting the funds to the people who are worst hurt by the crisis and to whom the cash will make the most difference. 13

Figure 9 Labor and Transfer Income During the Long, Four-Quarter Pandemic 2900 2800 2700 2600 2500 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$ noillib( emocni refsnart dna robal etagerggA Aggregate household income, long pandemic Baseline Long pandemic Long pandemic with CARES act Long pandemic, CARES act and continued unemployment payments Figure 10 Consumption Response to the Long, Four-Quarter Pandemic 2850 2800 2750 2700 2650 2600 2550 2500 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2020 2021 2022 2023 Quarter )$ noillib( noitpmusnoc ylretrauq etagerggA Long Pandemic Aggregate Consumption Baseline Long pandemic Long pandemic with CARES act Long pandemic, CARES act and continued unemployment payments 14

IV Conclusions Ourmodelsuggeststhattheremaybeastrongconsumptionrecoverywhenthesocial-distancing requirements of the pandemic begin to subside. We invite readers to test the robustness of this conclusion by using the associated software toolkit to choose their own preferred assumptions on the path of the pandemic, and of unemployment, to understand better how consumption will respond. One important limitation of our analysis is that it does not incorporate Keynesian demand effects or other general equilibrium responses to the consumption fluctuations we predict. In practice, Keynesian effects are likely to cause movements in aggregate income in the same direction as consumption; in that sense, our estimates can be thought of as a “first round” analysis of the dynamics of the crisis, which will be amplified by any Keynesian response. (See Bayer, Born, Luetticke, and Müller (2020) for estimates of the multiplier for transfer payments). These considerations further strengthen the case that the CARES Act will make a substantial difference to the economic outcome. A particularly important consideration is that forwardlooking firms that expect consumer demand to return forcefully in the third and fourth quarters of 2020 are more likely to maintain relations with their employees so that they can restart production quickly. The ability to incorporate Keynesian demand effects is one of the most impressive achievements of the generation of heterogeneous agent macroeconomic models that have been constructed in the last few years. But the technical challenges of constructing those models are such that they cannot yet incorporate realistic treatments of features that our model says are quantitatively important, particularly differing risks of (and types of) unemployment, for different kinds of people (young, old; rich, poor; high- and low-education). This rich heterogeneity is important both to the overall response to the CARES Act, and to making judgments about the extent to which it has been successfully targeted to provide benefits to those who need them most. A fuller analysis that incorporates both such heterogeneity, which is of intrinsic interest to policymakers, and a satisfying treatment of general equilibrium will have to wait for another day, but that day is likely not far off. 15

References Adams-Prassl, Abi, Teodora Boneva, Marta Golin, and Christopher Rauh (2020): “Inequality in the Impact of the Coronavirus Shock: New Survey Evidence for the US,” mimeo, Oxford University. Alvarez, Fernando, David Argente, and Francesco Lippi (2020): “A Simple Planning Problem for COVID-19 Lockdown,” BFI working paper 34, University of Chicago. Andersen, Asger Lau, Emil Toft Hansen, Niels Johannesen, and Adam Sheridan (2020): “Consumer Responses to the COVID-19 Crisis: Evidence from Bank Account Transaction Data,” Covid Economics, 7, 88–111. Armantier, Olivier, Gizem Kosar, Rachel Pomerantz, Daphne Skandalis, Kyle Smith, Giorgio Topa, and Wilbert van der Klaauw(2020): “CoronavirusOutbreakSendsConsumer Expectations Plummeting,” URL link retrieved on 04/07/2020 here. Baker, Scott R., R. A. Farrokhnia, Steffen Meyer, Michaela Pagel, and Constantine Yannelis (2020): “How Does Household Spending Respond to an Epidemic? Consumption During the 2020 COVID-19 Pandemic,” working paper 26949, National Bureau of Economic Research. Bayer, Christian, Benjamin Born, Ralph Luetticke, and Gernot J. Müller (2020): “The Coronavirus Stimulus Package: How Large Is the Transfer Multiplier?,” discussion paper 14600, CEPR. Bick, Alexander, and Adam Blandin (2020): “Real Time Labor Market Estimates During the 2020 Coronavirus Outbreak,” Working paper, Arizona State University. Bloomberg (2020a): “Mnuchin Warns Virus Could Yield 20% Jobless Rate Without Action,” Press article, Bloomberg. (2020b): “U.S. Jobless Rate May Soar to 30%, Fed’s Bullard Says,” Press article, Bloomberg. Broda, Christian, and Jonathan A. Parker (2014): “The Economic Stimulus Payments of 2008 and the Aggregate Demand for Consumption,” Journal of Monetary Economics, 68(S), 20–36. Brown, Jeffrey, Jeffrey B. Liebman, and Joshua Pollett (2002): “Estimating Life Tables ThatReflectSocioeconomicDifferencesinMortality,”inTheDistributionalAspectsofSocialSecurity andSocialSecurityReform,ed.byMartinFeldstein,andJeffreyB.Liebman,pp.447–457.University of Chicago Press. Cagetti, Marco (2003): “Wealth Accumulation Over the Life Cycle and Precautionary Savings,” Journal of Business and Economic Statistics, 21(3), 339–353. Carroll, Christopher D. (2006): “The Method of Endogenous Gridpoints for Solving Dynamic Stochastic Optimization Problems,” Economics Letters, 91(3), 312–320. Carroll, Christopher D., Edmund Crawley, Jiri Slacalek, Kiichi Tokuoka, and Matthew N. White (2020): “Sticky Expectations and Consumption Dynamics,” American Economic Journal: Macroeconomics, Forthcoming. Carroll, Christopher D., Alexander M. Kaufman, Jacqueline L. Kazil, Nathan M. Palmer, and Matthew N. White (2018): “The Econ-ARK and HARK: Open Source Tools for Computational Economics,” in Proceedings of the 17th Python in Science Conference, ed. by Fatih Akici, David Lippa, Dillon Niederhut, and M Pacer, pp. 25 – 30. doi: 10.5281/zenodo.1001067. 16

Carroll, Christopher D., Jiri Slacalek, Kiichi Tokuoka, and Matthew N. White (2017): “The Distribution of Wealth and the Marginal Propensity to Consume,” Quantitative Economics, 8, 977–1020, At http://econ.jhu.edu/people/ccarroll/papers/cstwMPC. Carvalho, V.M, J.R. Garcia, S. Hansen, A. Ortiz, T. Rodrigo, J.V. Mora Rodriguez, and J. Ruiz (2020): “Tracking the COVID-19 Crisis with High-Resolution Transaction Data,” Discussion paper, Cambridge University. Correia, Sergio, Stephan Luck, and Emil Verner (2020): “Pandemics Depress the Economy, Public Health Interventions Do Not: Evidence from the 1918 Flu,” mimeo, MIT Sloan. Cyranoski, David (2020): “‘We need to be alert’: Scientists fear second coronavirus wave as China’s lockdowns ease,” Nature. Eichenbaum, Martin S., Sergio Rebelo, and Mathias Trabandt (2020): “The Macroeconomics of Epidemics,” working paper 26882, National Bureau of Economic Research. Faria-e-Castro, Miguel (2020a): “Back-of-the-Envelope Estimates of Next Quarter’s Unemployment Rate,” Blog post, Federal Reserve Bank, St. Louis. (2020b): “Fiscal Policy during a Pandemic,” Covid Economics, 2, 67–101. Gascon, Charles (2020): “COVID-19: Which Workers Face the Highest Unemployment Risk?,” Blog post, Federal Reserve Bank, St. Louis. Glover, Andrew, Jonathan Heathcote, Dirk Krueger, and José-Víctor Ríos-Rull (2020): “Health versus Wealth: On the Distributional Effects of Controlling a Pandemic,” Covid Economics, 6, 22–64. Guerrieri, Veronica, Guido Lorenzoni, Ludwig Straub, and Ivan Werning (2020): “Macroeconomic Implications of COVID-19: Can Negative Supply Shocks Cause Demand Shortages?,” working paper 26918, National Bureau of Economic Research. Hai, Rong, Dirk Krueger, and Andrew Postlewaite (2013): “On the Welfare Cost of Consumption Fluctuations in the Presence of Memorable Goods,” working paper 19386, National Bureau of Economic Research. Havranek, Tomas, Marek Rusnak, and Anna Sokolova (2017): “Habit formation in consumption: A meta-analysis,” European Economic Review, 95, 142–167. Jorda, Oscar, Sanjay R. Singh, and Alan M. Taylor (2020): “Longer-run Economic Consequences of Pandemics,” Covid Economics, 1, 1–15. JPMorgan (2020): “Fallout from COVID-19: Global Recession, Zero Interest Rates and Emergency Policy Actions,” Blog post, JPMorgan. Kaplan, Greg, Benjamin Moll, and Giovanni L. Violante (2020): “Pandemics According to HANK,” mimeo, Princeton University. Krueger, Dirk, Kurt Mitman, and Fabrizio Perri (2016): “Macroeconomics and Household Heterogeneity,” Handbook of Macroeconomics, 2, 843–921. Krugman, Paul (2020): “Notes on Coronacoma Economics,” mimeo, City University of New York. Leibovici, Fernando, and Ana Maria Santacreu (2020): “Social Distancing and Contact- Intensive Occupations,” Blog post, Federal Reserve Bank, St. Louis. 17

Parker, Jonathan A. (2017): “Why Don’t Households Smooth Consumption? Evidence from a $25 Million Experiment,” American Economic Journal: Macroeconomics, 4(9), 153–183. Parker, Jonathan A, Nicholas S Souleles, David S Johnson, and Robert McClelland (2013): “Consumerspendingandtheeconomicstimuluspaymentsof2008,”TheAmericanEconomic Review, 103(6), 2530–2553. Sabelhaus, John, andJaeSong(2010): “TheGreatModerationinMicroLaborEarnings,”Journal of Monetary Economics, 57(4), 391–403. Swagel, Phill (2020): “Updating CBO’s Economic Forecast to Account for the Pandemic,” Blog post, Congressional Budget Office. 18

Appendices A Model Details ThebaselinemodelisadaptedandexpandedfromCarroll,Slacalek,Tokuoka,andWhite(2017). The economy consists of a continuum of expected utility maximizing households with a common CRRA utility function over consumption, u(c,η) = ηc1−ρ/(1−ρ), where η is a marginal utility shifter. Households are ex ante heterogeneous: household i has a quarterly time discount factor β ≤ 1 and an education level e ∈ {D,HS,C} (for dropout, high school, and college, i i respectively). Each quarter, the household receives (after tax) income, chooses how much of their market resources m to consume c and how much to retain as assets a ; they then it it it transition to the next quarter by receiving shocks to mortality, income, their employment state, and their marginal utility of consumption. Foreacheducationgroupe,weassignauniformdistributionoftimepreferencefactorsbetween ` ` β − ∇ and β + ∇, chosen to match the distribution of liquid wealth and retirement assets. e e Specifically, the calibrated values in Table A1 fit the ratio of liquid wealth to permanent income in aggregate for each education level, as computed from the 2004 Survey of Consumer Finance. The width of the distribution of discount factors was calibrated to minimize the difference between simulated and empirical Lorenz shares of liquid wealth for the bottom 20%, 40%, 60%, and 80% of households, as in Carroll, Slacalek, Tokuoka, and White (2017). When transitioning from one period to the next, a household with education e that has already lived for j periods faces a D probability of death. The quarterly mortality probabilities ej are calculated from the Social Security Administration’s actuarial table (for annual mortality probability) and adjusted for education using Brown, Liebman, and Pollett (2002); a household dieswithcertaintyifit(improbably)reachestheageof120years. Theassetsofahouseholdthat dies are completely taxed by the government to fund activities outside the model. Households who survive to period t+1 experience a return factor of R on their assets, assumed constant. Household i’s state in period t, at the time it makes its consumption–saving decision, is characterized by its age j,14 a level of market resources m ∈ R , a permanent income level it + ppp ∈ R , a discrete employment state ‘ ∈ {0,1,2} (indicating whether the individual is it ++ it employed, normal unemployed, or deeply unemployed), and a discrete state η ∈ {1,η} that it represents whether its marginal utility of consumption has been temporarily reduced (η < 1). Denote the joint discrete state as n = (‘ ,η ). it it it Each household inelastically participates in the labor market when it is younger than 65 years (j < 164) and retires with certainty at age 65. The transition from working life to retirement is captured in the model by a one time large decrease in permanent income at age j = 164.15 Retired households face essentially no income risk: they receive Social Security benefits equal to theirpermanentincomewith99.99%probabilityandmisstheircheckotherwise; theirpermanent income very slowly degrades as they age. The discrete employment state ‘ is irrelevant for it retired households. Labor income for working age households is subject to three risks: unemployment, permanent incomeshocks,andtransitoryincomeshocks. Employed(‘ = 0)households’permanentincome it grows by age-education-conditional factor Γ on average, subject to a mean one lognormal ej permanent income shock ψ with age-conditional underlying standard deviation of σ . The it ψj household’s labor income y is also subject to a mean one lognormal transitory shock ξ it it 14Householdsenterthemodelaged24years,somodelagej=0correspondstobeing24years,0quartersold. 15Thesizeofthedecreasedependsoneducationlevel,veryroughlyapproximatingtheprogressivestructureofSocialSecurity: ΓD164≈0.56,ΓHS164≈0.44,ΓC164≈0.31. 19

with age-conditional underlying standard deviation of σ . The age profiles of permanent and ξj transitory income shock standard deviations are approximated from the results of Sabelhaus and Song (2010), and the expected permanent income growth factors are adapted from Cagetti (2003). Normal unemployed and deeply unemployed households receive unemployment benefits equal to a fraction ξ = 0.3 of their permanent income, y = ξppp ; they are not subject it it to permanent nor transitory income risk, but their permanent income degrades at rate Γ, representing “skill rot”.16 The income process for a household can be represented mathematically as:  ψ Γ ppp if ‘ = 0, j < 164 Employed, working age  it ej it−1 it ppp = Γppp if ‘ > 0, j < 164 Unemployed, working age , it it−1 it Γ ppp if j ≥ 164 Retired ret it−1  ξ ppp if ‘ = 0, j < 164 Employed, working age  it it it y = ξppp if ‘ > 0, j < 164 Unemployed, working age . it it it ppp if j ≥ 164 Retired it A working-age household’s employment state ‘ evolves as a Markov process described by the it matrix Ξ, where element k,k0 of Ξ is the probability of transitioning from ‘ = k to ‘ = k0. it it+1 During retirement, all households have ‘ = 0 (or any other trivializing assumption about the it “employment” state of the retired). We assume that households treat Ξ and Ξ as zero: 0,2 1,2 they do not consider the possibility of ever attaining the deep unemployment state ‘ = 2 from it “normal” employment or unemployment, and thus it does not affect their consumption decision in those employment states. Wespecifytheunemploymentrateduringnormaltimesas = 5%, andtheexpectedduration of an unemployment spell as 1.5 quarters. The probability of transitioning from unemployment (cid:48) back to employment is thus Ξ = 2, and the probability of becoming unemployed is determined 1,0 3 as the flow rate that offsets this to generate 5% unemployment (about 3.5%). The deeply unemployed expect to be unemployed for much longer: we specify Ξ = 0 and Ξ = 1, so that 2,0 2,1 3 a deeply unemployed person remains so for three quarters on average before becoming “normal” unemployed (they cannot transition directly back to employment). Thus the unemployment spell for a deeply unemployed worker is 2 quarters at a minimum and 4.5 quarters on average.17 Like the prospect of deep unemployment, the possibility that consumption might become less appealing (via marginal utility scaling factor η < 1) does not affect the decision-making it process of a household in the normal η = 1 state. If a household does find itself with η = η, it it this condition is removed (returning to the normal state) with probability 0.5 each quarter; the evolution of the marginal utility scaling factor is represented by the Markov matrix H. In this way, the consequences of a pandemic are fully unanticipated by households, a so-called “MIT shock”; households act optimally once in these states, but did not account for them in their consumption–saving problem during “normal” times.18 16Unemploymentissomewhatpersistentinourmodel,sotheutilityriskfromreceiving15%ofpermanentincomeforonequarter (asinCarroll,Slacalek,Tokuoka,andWhite(2017))isroughlythesameastheriskofreceiving30%ofpermanentincomefor1.5 quartersinexpectation. 17Ourcomputationalmodelallowsforworkers’beliefsabouttheaveragedurationofdeepunemploymenttodifferfromthetrue probability. However,wedonotpresentresultsbasedonthisfeatureandthuswillnotfurtherclutterthenotationbyformalizing ithere. 18Our computational model also allows households’ beliefs about the duration of the reduced marginal utility state (via social distancing)todeviatefromthetrueprobability. Thecodealsopermitsthepossibilitythatthereductioninmarginalutilityislifted asanaggregateorsharedoutcome,ratherthanidiosyncratically. Wedonotpresentresultsutilizingthesefeatureshere,butinvite thereadertoinvestigatetheirpredictedconsequencesusingourpublicrepository. 20

The household’s permanent income level can be normalized out of the problem, dividing all boldface variables (absolute levels) by the individual’s permanent income ppp , yielding non-bold it normalized variables, e.g., m = m /ppp . Thus the only state variables that affect the choice of it it it optimal consumption are normalized market resources m and the discrete Markov states n . it it After this normalization, the household consumption functions c satisfy: e,j h i v (m ,n ) = max u(c (m ,n ),η )+β (1−D )E Γb 1−ρv (m ,n ) e,j it it e,j it it it i e,j t it+1 e,j+1 it+1 it+1 ce,j s.t. a = m −c (m ,n ), it it e,j it it m = (R/Γb )a +y , it+1 it+1 it it n ∼ (Ξ,H), it+1 a ≥ 0, it where Γb = ppp /ppp , the realized growth rate of permanent income from period t to t + 1. it+1 it+1 it Consumption function c yields optimal normalized consumption, the ratio of consumption e,j to the household’s permanent income level; the actual consumption level is simply c = it ppp c (m ,n ). it e,j it it Starting from the terminal model age of j = 384, representing being 120 years old (when the optimal choice is to consume all market resources, as death is certain), we solve the model by backward induction using the endogenous grid method, originally presented in Carroll (2006). Substitutingthedefinitionofnextperiod’smarketresourcesintothemaximand, thehousehold’s problem can be rewritten as: h i v (m ,n ) = max u(c ,η )+β (1−D )E Γb 1−ρv ((R/Γb )a +y ,n ) e,j it it cit∈R + it it i e,j t it+1 e,j+1 it+1 it it it+1 s.t. a = m −c , a ≥ 0, n ∼ (Ξ,H). it it it it it+1 This problem has one first order condition, which is both necessary and sufficient for optimality. It can be solved to yield optimal consumption as a function of (normalized) end-of-period assets and the Markov state: η c−ρ−β R(1−D )E h Γb −ρ vm ((R/Γb )a +y ,n ) i = 0 =⇒ c = va e,j (a it ,n it )!− ρ 1 . it it i e,j t it+1 e,j+1 it+1 it it it+1 it η | {z } | {z } it =∂ ∂ u c ≡va e,j (ait,nit) To solve the age-j problem numerically, we specify an exogenous grid of end-of-period asset values a ≥ 0, compute end-of-period marginal value of assets at each gridpoint (and each discrete Markov state), then calculate the unique (normalized) consumption that is consistent with ending the period with this quantity of assets while acting optimally. The beginning-ofperiod (normalized) market resources from which this consumption was taken is then simply m = a + c , the endogenous gridpoint. We then linearly interpolate on this set of market it it it resources–consumption pairs, adding an additional bottom gridpoint at (m ,c ) = (0,0) to it it represent the liquidity-constrained portion of the consumption function c (m ,n ). e,j it it The standard envelope condition applies in this model, so that the marginal value of market resources equals the marginal utility of consumption when consuming optimally: vm (m ,n ) = η c (m ,n )−ρ. e,j it it it e,j it it The marginal value function for age j can then be used to solve the age j−1 problem, iterating backward until the initial age j = 0 problem has been solved. When the pandemic strikes, we draw a new employment state (employed, unemployed, deeply 21

unemployed) for each working age household using a logistic distribution. For each household i at t = 0 (the beginning of the pandemic and lockdown), we compute logistic weights for the employment states as: P = α +α ppp +α j for ‘ ∈ {1,2}, P = 0, i,‘ ‘,e ‘,p i0 ‘,j i0 i,0 where e ∈ {D,H,C} for dropouts, high school graduates, and college graduates and j is the household’s age. The probability that household i draws employment state ‘ ∈ {0,1,2} is then calculated as: , 2 Pr(‘ = ‘) = exp(P ) X exp(P ). it i,‘ i,k k=0 Our chosen logistic parameters are presented in Table A2. B Aggregation Households are modeled as individuals and incomes sized accordingly. We completely abstract from family dynamics. To get our aggregate predictions for income and consumption, we take the mean from our simulation and multiply by 253 million, the number of adults (over 18) in the United States in 2019. To size the unemployment benefits correctly, we multiply the benefits per worker by 0.8 to account for the fact that 20 percent of the working-age population is out of the labor force, so the average working-age household consists of 0.8 workers and 0.2 non-workers. With this adjustment, there are 151 million workers eligible for unemployment benefits in the model. Aggregate consumption in our baseline for 2020 is just over $11 trillion, a little less than total personal consumption expenditure, accounting for the fact that some consumption does not fit in the usual budget constraint.19 Aggregating in this way underweights the young, as our model excludes those under the age of 24. Our model estimates the aggregate size of the stimulus checks to be $267 billion, matching the the Joint Committee on Taxation’s estimate of disbursements in 2020.20 This is somewhat of a coincidence: we overestimate the number of adults who will actually receive the stimulus, while excluding the $500 payment to children. The aggregate cost of the extra unemployment benefits depends on the expected level of unemployment. Our estimate is $137 billion, much less than the $260 billion mentioned in several press reports, but in line with the extent of unemployment in our pandemic scenario. We do not account for the extension of unemployment benefits to the self-employed and gig workers. Households enter the model at age j = 0 with zero liquid assets. A ‘newborn’ household has its initial permanent income drawn lognormally with underlying standard deviation of 0.4 and an education-conditional mean. The initial employment state of households matches the steady state unemployment rate of 5%.21 We assume annual population growth of 1%, so older simulated households are appropriately down-weighted when we aggregate idiosyncratic values. Likewise, each successive cohort is slightly more productive than the last, with aggregate productivity growing at a rate of 1% 19PCE consumption in Q4 2019, from the NIPA tables, was $14.8 trillion. Market based PCE, a measure that excludes expenditures without an observable price was $12.9 trillion. Health care, much of which is paid by employers and not in the household’sbudgetconstraint,was$2.5trillion. 20The JCT’s 26 March 2020 publication JCX-11-20 predicts disbursements of $267 billion in 2020, followed by $24 billion in 2021. 21Thisisthecaseevenduringthepandemicandlockdown,sothedeathandreplacementofsimulatedagentsisasecondorder contributiontotheprofileoftheunemploymentrate. 22

Figure 11 Concave Cost of Consumption Units 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Cost stinu noitpmusnoc fo rebmuN Quality Adjusted Cost of Consumption Units Normal Times C Lockdown per year. The profile of average income by age in the population at any moment in time thus has more of an inverted-U shape than implied by the permanent income profiles from Cagetti (2003). C Marginal Utility Equivalence We model the ‘lockdown’ as a reduction in the marginal utility of consumption. This can be interpretedasanincreaseinthequality-adjustedpriceofgoods, wherethequalityofbasicgoods such as shelter and housing has not decreased, but more discretionary goods such as vacations and restaurants have decreased in quality. Figure 11 shows how this works. In normal times, the cost of a consumption unit is equal to one, represented by the blue line. During the lockdown, the cost of a unit of consumption is increasing in the number of units bought. As shown here, the number of consumption units that can be bought follows the lower envelope of the blue and orange lines, where the orange line is equal to Costα. As long as the household is consuming above the kink, their utility is log(Costα) = αlog(Cost), exactly equivalent to the reduction in marginal utility we apply. Taking this interpretation seriously, the drop in marginal utility should not be applied to households with very low levels of consumption, below the kink. Our implementation abstracts from this, taking the marginal utility factor to be the same for all agents. An alternative interpretation is that consumption is made up of a Cobb-Douglass aggregation of two goods: C = cαc1−α 1 2 During the lockdown, the second good is replaced by home production at a fixed level c¯. A 2 log-utility function gives log(C) = αlog(c )+(1−α)log(c¯), equivalent to our model in which 1 2 we reduce marginal utility by a factor α. 23

Table A1 Parameter Values in the Baseline Model Description Parameter Value Coefficient of relative risk aversion ρ 1 ` Mean discount factor, high school dropout β 0.9637 D ` Mean discount factor, high school graduate β 0.9705 HS ` Mean discount factor, college graduate β 0.9756 C Discount factor band (half width) ∇ 0.0253 Employment transition probabilities: – from normal unemployment to employment Ξ 2/3 1,0 – from deep unemployment to normal unemployment Ξ 1/3 2,1 – from deep unemployment to employment Ξ 0 2,0 Proportion of high school dropouts θ 0.11 D Proportion of high school graduates θ 0.55 HS Proportion of college graduates θ 0.34 C Average initial permanent income, dropout ppp 5000 D0 Average initial permanent income, high school ppp 7500 HS0 Average initial permanent income, college ppp 12000 C0 Steady state unemployment rate 0.05 Unemployment insurance replacement rate ξ 0.30 (cid:48) Skill rot of all unemployed Γ 0.995 Quarterly interest factor R 1.01 Population growth factor N 1.0025 Technological growth factor ג 1.0025 24

Table A2 Pandemic Assumptions Description Parameter Value Short-lived Pandemic Logistic parametrization of unemployment probabilities Constant for dropout, regular unemployment α −1.15 1,D Constant for dropout, deep unemployment α −1.5 2,D Constant for high school, regular unemployment α −1.3 1,H Constant for high school, deep unemployment α −1.75 2,H Constant for college, regular unemployment α −1.65 1,C Constant for college, deep unemployment α −2.2 2,C Coefficient on permanent income, regular unemployment α −0.1 1,p Coefficient on permanent income, deep unemployment α −0.2 2,p Coefficient on age, regular unemployment α −0.01 1,j Coefficient on age, deep unemployment α −0.01 2,j Marginal Utility Shock Pandemic utility factor η 0.891 Prob. exiting pandemic each quarter H 0.5 1,0 Long, Deep Pandemic Logistic parametrization of unemployment probabilities Constant for dropout, regular unemployment α −1.45 1,D Constant for dropout, deep unemployment α −0.3 2,D Constant for high school, regular unemployment α −1.6 1,H Constant for high school, deep unemployment α −0.55 2,H Constant for college, regular unemployment α −1.95 1,C Constant for college, deep unemployment α −1.00 2,C Coefficient on permanent income, regular unemployment α −0.2 1,p Coefficient on permanent income, deep unemployment α −0.2 2,p Coefficient on age, regular unemployment α −0.01 1,j Coefficient on age, deep unemployment α −0.01 2,j Marginal Utility Shock Pandemic utility factor η 0.891 Prob. exiting pandemic each quarter H 0.25 1,0 25

Table A3 Fiscal Stimulus Assumptions, CARES Act Description Value Stimulus check $1,200 Means test start (annual) $75,000 Means test end (annual) $99,000 Stimulus check delay 1 quarter Fraction that react on announcement 0.25 Extra unemployment benefit for: Normal unemployed $5,200 Deeply unemployed $7,800 Note: Theunemploymentbenefitsaremultipliedby0.8toaccountforthefactthat20percentoftheworkingagepopulationis outofthelaborforce. SeeaggregationdetailsinAppendixB. 26