State-by-state decisions on shutdowns minimise Covid’s economic impact

Mario Crucini, Oscar O’Flaherty 29 May 2021



Public opinion has been divided on the policies used throughout the Covid-19 pandemic (Milosh et al. 2020). One of the most controversial of these policies has been stay-at-home orders. At the onset of the pandemic, protests over the economic impact of stay-at-home orders occurred in many US states (NBC News 2020, Fox News 2020). One protester in New Jersey said, “[b]usinesses are suffering, unemployment checks are not being sent, landlords are not getting rent. We feel like these directives are causing more suffering than is necessary” ( 2020). A poll conducted in May 2020 estimated that 35% of people “strongly or somewhat agreed” that restrictions and closures had been too severe (USA Today 2020). Goldstein et al. (2020) show that lockdowns become less effective in combating the virus over time as people experience ‘lockdown fatigue’.

The country was also divided about how stay-at-home orders should be implemented. A poll by The Economist in April 20201 showed that 61% of people believed that President Donald Trump should issue a national stay-at-home order. At the time, the president opposed a nationwide order, saying “[t]here are some states that are different. There are some states that don’t have much of a problem . . . You have to give a little bit of flexibility”.2 However, Dr. Anthony Fauci expressed a different opinion, saying “I just don’t understand why we’re not doing that [issuing a federally mandated stay-at-home order]. We really should be” (Politico 2020).

In the case of stay-at-home orders, where the health versus wealth effects are stark, achieving unanimity is likely impossible. Trade-offs do exist. Understanding and acknowledging these trade-offs from an evidence-based standpoint is a crucial first step in narrowing the public opinion divide.

Novel contributions to data dissemination and macroeconomic modelling have made it possible to empirically examine the health and economic trade-offs throughout the pandemic. Private companies, such as Homebase,3 released high-frequency data essential to measuring the ongoing economic impact of Covid-19. Previously, researchers commonly had to wait weeks or even months before seeing the most up-to-date employment numbers. Much of the data from these companies was compiled into a publicly available database by Chetty et al. (2020).4

On the modelling front, Eichenbaum et al (2020) were among the very first to integrate a conventional macroeconomic model with the standard epidemiological model of virus transmission, the Susceptible-Infected-Recovered (SIR) model. This SIR-macro model was influential in several ways. First, it allowed researchers to study how the virus, economic policies and the aggregate economy interacted. As in real life, an increase in the virus’ spread led to less spending and a decrease in employment. This reduction in economic activity had corresponding health benefits by limiting the spread of the virus. Furthermore, the model allowed policymakers to analyse the health and economic trade-offs of various social distancing policies before issuing a policy change in the actual economy.

In our recent paper (Crucini and O’Flaherty 2021), we use this new economic data to estimate the impact of stay-at-home orders on consumer spending and employment. Then we extend the SIR-macro model to incorporate multiple locations. This allows us to address whether a national stay-at-home order is better than leaving states “a little bit of flexibility”, as President Trump suggested.

Figure 1 shows the impact of the initial stay-at-home orders issued by states in March and April of 2020. Not all states issued stay-at-home orders on the same day. Our analysis uses this difference in timing to estimate what the average effect of a stay-at-home order is x days after it is issued (e.g. the day of, one day after, . . ., 15 days after).

Panel (a) shows that hours worked decreased by an additional 4 percentage points, on average, in the states that issued them. Panel (b) shows a similar decline for consumer spending in the first several weeks after a stay-at-home order was issued. Overall, these results suggest that stay-at-home orders caused a cumulative impact of $15 billion in lost earnings and a $10 billion decrease in consumer spending in the span of half a month. Furthermore, the results suggest that stay-at-home orders caused almost 6 million people to temporarily lose their jobs, which accounts for 10% to 25% of the total employment loss during this period.

Figure 1 Effect of stay-at-home orders on the economy

(a) Hours worked             


(b) Consumer spending


Were stay-at-home orders too severe or too lax? Although wider societal impacts of stay-at-home orders may prove difficult to calculate, our analysis shows that state and local orders did have real and significant effects on economic output, starkly illustrating the ‘health and wealth’ trade-offs of fighting the pandemic. 

Figure 2 presents the implications of optimal mitigation policy in our model with solid lines and no policy with dashed lines. Although the model suggests that mitigation policy is necessary, it demonstrates the need for a state-by-state approach in policy flexibility. 

Figure 2 Optimal mitigation policy

As illustrated by the bottom-right panel in Figure 2, the state with a higher infection rate (blue line) initially sets a stricter policy than the state with a lower infection rate (red line). However, as infection rates converge, so do their policies. The state that starts with the lower infection rate, therefore, avoids an early dramatic economic shock, as demonstrated in the red line in the upper-right panel in Figure 2. 

Throughout much of 2020, the Trump administration deferred decision making regarding stay-at-home orders to the state and local level. While many have castigated the former president’s team for relying too much on state decision makers, our model tests whether a national stay-at-home mandate would have resulted in less economic impact or reduced infection spread.

We next evaluated the impact of a national stay-at-home policy. As shown in Figure 3, the national policy did not have a material impact on the states’ infection rates. However, under a national policy, the mildly infected state experiences significantly larger declines in consumption at the onset of the pandemic. Simply put, a nationwide mandate slows economic activity in a mildly infected state without the intended health benefits of reducing virus spread within that state.

Figure 3 National mitigation policy

Our results strongly suggest that issuing a national stay-at-home order at the onset of the pandemic, when the virus was spreading primarily in a small group of cities, may have imposed earlier and deeper economic costs on states with relatively low case numbers without any corresponding reduction in infection rates in such states. However, as the virus spread more uniformly across the country in the last several months of 2020, a nationwide order seemed more appropriate as infection rates across states started to converge. 

The ongoing distribution and administration of the vaccine demonstrate the need for close cooperation between federal and state agencies. However, our analysis shows that informed state-level decision making about stay-at-home orders is most effective in minimising the economic costs required to achieve larger public health goals.


Chetty, R, J N Friedman, N Hendren, M Stepner and The Opportunity Insights Team (2020), “How Did COVID-19 and Stabilization Policies Affect Spending and Employment? A New Real-Time Economic Tracker Based on Private Sector Data,” NBER Working Paper No. 27431.

Crucini, M and O O’Flaherty (2020), “Stay-at-Home Orders in a Fiscal Union,” NBER Working Paper No. 28182.

Crucini, M and O O’Flaherty (2021), “State-by-State Decisions on Shutdowns Minimize COVID’s Economic Impact,” The Vanderbilt Project on Unity and Democracy, 2 March.

Eichenbaum, M, S Rebelo, and M Trabandt (2020), “The Macroeconomics of Epidemics,” NBER Working Paper No. 27141.

Fox News (2020), “Coronavirus shutdown: What states have seen protests against stay-at-home orders”.

Goldstein, P, E Levy Yeyati, and L Sartorio (2021), “Lockdown fatigue: The declining effectiveness of lockdowns”,, 30 March.

Milosh, M, M Painter, K Sonin, D Van Dijcke and A L Wright (2020). “Political polarisation impedes the public policy response to COVID-19”,, 23 December.

NBC News (2020), “’Fire Gruesome Newsom!’: Stay-at-home protests in California and across the country". (2020), “‘Open New Jersey now!’ Protesters in Trenton demand Gov. Murphy lift coronavirus lockdowns despite rising death toll”.

Politico (2020), “Fauci endorses national stay-at-home order: ‘I just don’t understand why we’re not doing that’”.

USA TODAY (2020), “Partisan divide over economy grows, but Americans more worried about their health: Exclusive poll".







Topics:  Covid-19

Tags:  COVID-19, stay-at-home orders, health vs wealth, decentralisation, US states

Professor of Economics, Vanderbilt University

PhD candidate, Department of Economics, Vanderbilt University


CEPR Policy Research