Below is a list of published research and working papers — preliminary versions of academic research –— by MIT Sloan faculty and PhD students related to the coronavirus/COVID-19 pandemic. This page will be updated with new papers as they become available. Papers are listed in reverse chronological order using the date of their first version. A list of projects and other resources led by MIT Sloan faculty is at the end of this page.
Andrew W. Lo, Brian Stevens, Sean P. Willems
Online teaching at higher educational institutions has become a much higher priority in the face of the COVID-19 pandemic, but most faculty and staff at these institutions are ill-prepared to adapt their teaching methods and content to this new medium. In this article, we describe our approach to dealing with the challenges and opportunities of synchronous online teaching by borrowing ideas and tools from the gaming community. The gamification of education is a well-known concept, but we found few applications in higher-education settings to rely on when we were forced to move online in March 2020. We hope to remedy this gap by providing colleagues with a step-by-step guide to setting up their own home studios, including a complete listing of the software and hardware that we use and how we use them in three distinct online teaching applications: a large (90-student) graduate healthcare finance course at MIT, an even larger (200-student) undergraduate statistics course at the University of Tennessee Knoxville, and a medium-sized (50-student) graduate operations management course at MIT.
Dhaval Adjodah, Karthik Dinakar, Matteo Chinazzi, Samuel P. Fraiberger, Alex Pentland, Samantha Bates, Kyle Staller, Alessandro Vespignani, Deepak L. Bhatt
Using publicly available data, we quantify the impact of mask adherence and mask mandates on COVID-19 outcomes. We show that mask mandates are associated with a statistically significant decrease in daily new cases (−3.24 per 100K), deaths (−0.19 per 100K), and the proportion of hospital admissions (−2.47%) due to COVID-19 between February 1 and September 27, 2020. These effects are large, corresponding to 13% of the highest recorded number of cases, 20% of deaths, and 7% of admission proportion. We also find that mask mandates are linked to a 23.4 percentage point increase in mask adherence in four diverse states, and that mask adherence is associated with improved COVID-19 outcomes. Lastly, using a large novel survey in 68 countries, we find that community mask adherence and attitudes towards masks are associated with a reduction in COVID-19 cases and deaths. Our results have relevant policy implications, indicating the need to maintain and encourage mask-wearing.
Hazhir Rahmandad, Ty Lim
Responses to the COVID-19 pandemic have been conditioned by a perceived tradeoff between saving lives and the economic costs of contact-reduction measures. We develop a model of SARS-CoV-2 transmission where populations endogenously reduce contacts in response to the risk of death. We estimate the model for 118 countries and assess the existence of a tradeoff between death rates and changes in contacts. In this model communities go through three phases – rapid early outbreaks, control through initial response, and a longer period of quasi-equilibrium endemic infection with effective reproduction number (Re) fluctuating around one. Analytical characterization of this phase shows little tradeoff between contact reduction levels (underpinning economic costs) and death rates. Empirically estimating the model, we find no positive correlation between (log) death rates and (normalized) contact levels across nations, whether contacts are estimated based on epidemic curves or mobility data. While contact reduction levels are broadly similar across countries, expected death rates vary greatly, by two orders of magnitude (5-95 percentile: 0.03-17 deaths per million per day). Results suggest nations could significantly reduce the human toll of the pandemic without more disruption to normal social and economic activity than they have already faced.
Juan Palacios, Yichun Fan, Erez Yoeli, Jianghao Wang, Yuchen Chain, Weizeng Sun, David Rand, Siqi Zheng
As the COVID-19 pandemic comes to an end, governments find themselves facing a new challenge: motivating citizens to resume economic activity. What is an effective way to do so? We investigate this question using a field experiment in the city of Zhengzhou, China immediately following the end of the city’s COVID-19 lockdown. Using self-reports and GPS trajectory data from participants’ phones, we assessed the effect of providing information about the proportion of participants’ neighbors who have resumed economic activity. We find that informing individuals about their neighbors’ plans to visit restaurants increases the fraction of participants visiting restaurants by 12 percentage points (37%), amongst those participants who underestimated the proportion of neighbors who resumed economic activity. Those who overestimated did not respond by reducing restaurant attendance, so the intervention yielded no ‘boomerang’ effect. We explore moderators, risk perceptions, and a placebo intervention for parks. All of these analyses suggest our intervention worked by reducing the perceived risk of going to restaurants.
Stefan Gavell, Mark Kritzman, Cel Kulasekaran
In light of the COVID 19 crisis, the Federal Reserve has carried out stress tests to assess if major banks have sufficient capital to ensure their viability should a new and perhaps unprecedented crisis emerge. The Fed argues that the scenarios underpinning these stress tests are severe but plausible, yet they have not offered any evidence or framework for measuring the plausibility of their scenarios. If the scenarios are indeed plausible, it makes sense for banks to retain enough capital to withstand their occurrence. If, however, the scenarios are not reasonably plausible, banks will have deployed capital less productively than they otherwise could have, thereby impairing credit expansion and economic growth. The authors apply a measure of statistical unusualness, called the Mahalanobis distance, to assess the plausibility of the Fed’s stress scenarios. A first pass of their analysis, based on conventional statistical assumptions, reveals that the Fed’s scenarios are not even remotely plausible. However, the authors offer two modifications to their initial analysis that increase the scenarios’ plausibility. First, they show how the Fed can minimally modify their scenarios to render them marginally plausible in a Gaussian world. And second, they show how to evaluate the plausibility of the Fed’s scenarios by replacing the theoretical world of normality with a distribution that is empirically grounded.
Dimitris Bertsimas, Galit Lukin, Luca Mingardi, Omid Nohadani, Agni Orfanoudaki, Bartolomeo Stellato, Holly Wiberg, Sara Gonzalez-Garcia, Carlos Luis Parra-Calderón, Kenneth Robinson, Michelle Schneider, Barry Stein, Alberto Estirado, Lia A. Beccara, Rosario Canino, Martina Dal Bello, Federica Pezzetti, Angelo Pan, and the Hellenic COVID-19 Study Group
Timely identification of COVID-19 patients at high risk of mortality can significantly improve patient management and resource allocation within hospitals. This study seeks to develop and validate a data-driven personalized mortality risk calculator for hospitalized COVID-19 patients. De-identified data was obtained for 3,927 COVID-19 positive patients from six independent centers, comprising 33 different hospitals. Demographic, clinical, and laboratory variables were collected at hospital admission. The COVID-19 Mortality Risk (CMR) tool was developed using the XGBoost algorithm to predict mortality. Its discrimination performance was subsequently evaluated on three validation cohorts. The derivation cohort of 3,062 patients has an observed mortality rate of 26.84%. Increased age, decreased oxygen saturation (≤ 93%), elevated levels of C-reactive protein (≥ 130 mg/L), blood urea nitrogen (≥ 18 mg/dL), and blood creatinine (≥ 1.2 mg/dL) were identified as primary risk factors, validating clinical findings. The model obtains out-of-sample AUCs of 0.90 (95% CI, 0.87–0.94) on the derivation cohort. In the validation cohorts, the model obtains AUCs of 0.92 (95% CI, 0.88–0.95) on Seville patients, 0.87 (95% CI, 0.84–0.91) on Hellenic COVID-19 Study Group patients, and 0.81 (95% CI, 0.76–0.85) on Hartford Hospital patients. The CMR tool is available as an online application at covidanalytics.io/mortality_calculator and is currently in clinical use. The CMR model leverages machine learning to generate accurate mortality predictions using commonly available clinical features. This is the first risk score trained and validated on a cohort of COVID-19 patients from Europe and the United States.
Olivia S. Kim, Jonathan A. Parker, Antoinette Schoar
Using detailed transaction-level data from financial accounts, this paper shows that the revenues of small businesses and the consumption spending of their owners both decline by roughly 40% following the declaration of the national emergency in March 2020. However, through May 2020, the vast majority of this average decline in revenues is due to national factors rather than to variation in local infection rates or policies. Further, there is only a modest propensity for business owners to cut consumption in response to their individual business losses: Comparing owners in the same county but whose businesses operate in industries differentially impacted by local infections and state-level policies, we show that each dollar of revenue loss leads to a 1.6 cent decline in the consumption of the owner at this early stage of the pandemic. This limited passthrough appears to be explained by three factors: (1) the liquidity of households and businesses entering the crisis – consumption is twice as responsive for small business owners who operate with low liquidity; (2) emergency federal programs — median account balances in both business and checking accounts decline in March but rebound in April and May when the transfer programs begin; (3) pandemic induced declines in the ability to spend on consumption — spending on travel, restaurants or personal services dropped dramatically.
Dimitris Bertsimas, Joshua Ivanhoe, Alexandre Jacquillat, Michael Li, Alessandro Previero, Omar Skali Lami, and Hamza Tazi Bouardi
The outbreak of COVID-19 has spurred extensive research worldwide to develop a vaccine. However, when a vaccine becomes available, limited production and distribution capabilities will likely lead to another challenge: who to prioritize for vaccination to mitigate the near-end impact of the pandemic? To tackle that question, this paper first expands a state-of-the-art epidemiological model, called DELPHI, to capture the effects of vaccinations and the variability in mortality rates across subpopulations. It then integrates this predictive model into a prescriptive model to optimize vaccine allocation, formulated as a bilinear, non-convex optimization model. To solve it, this paper proposes a coordinate descent algorithm that iterates between optimizing vaccine allocations and simulating the dynamics of the pandemic. We implement the model and algorithm using real-world data in the United States. All else equal, the optimized vaccine allocation prioritizes states with a large number of projected cases and sub-populations facing higher risks (e.g., older ones). Ultimately, the optimized vaccine allocation can reduce the death toll of the pandemic by an estimated 10–25%, or 10,000–20,000 deaths over a three-month period in the United States alone.
Dimitris Bertsimas, Alison Borenstein, Luca Mingardi, Omid Nohadani, Agni Orfanoudaki, Bartolomeo Stellato, Holly WIberg, Pankaj Sarin, Dirk J. Varelmann, Vicente Estrada, Carlos Macaya, and Iván J. Núñez Gil
The COVID-19 pandemic has prompted an international effort to develop and repurpose medications and procedures to effectively combat the disease. Several groups have focused on the potential treatment utility of angiotensin-converting–enzyme inhibitors (ACEIs) and angiotensin-receptor blockers (ARBs) for hypertensive COVID-19 patients, with inconclusive evidence thus far. We couple electronic medical record (EMR) and registry data of 3,643 patients from Spain, Italy, Germany, Ecuador, and the US with a machine learning framework to personalize the prescription of ACEIs and ARBs to hypertensive COVID-19 patients. Our approach leverages clinical and demographic information to identify hospitalized individuals whose probability of mortality or morbidity can decrease by prescribing this class of drugs. In particular, the algorithm proposes increasing ACEI/ARBs prescriptions for patients with cardiovascular disease and decreasing prescriptions for those with low oxygen saturation at admission. We show that personalized recommendations can improve patient outcomes by 1.0% compared to the standard of care when applied to external populations. We develop an interactive interface for our algorithm, providing physicians with an actionable tool to easily assess treatment alternatives and inform clinical decisions. This work offers the first personalized recommendation system to accurately evaluate the efficacy and risks of prescribing ACEIs and ARBs to hypertensive COVID-19 patients.
Megan Szasonis, Mark Kritzman, Baykan Pamir, David Turkington
The authors model COVID infections and COVID deaths, both reported and implied, for the 50 U.S. states as well as the District of Columbia, and separately for a sample of 33 countries, as a function of pre-existing circumstances that citizens have no ability to control over the short term. These models give predictions of expected COVID outcomes. They then compare their model’s predicted results with actual experience. They interpret the differences between actual experiences and the predictions across the jurisdictions as the COVID outcomes attributable to the behavior of citizens.
Deborah Ancona, Henrik Bresman, Mark Mortensen
Even before COVID‐19 we saw an evolution in team discourse that will continue long after the disease is gone. That said, COVID‐19 has been a disruptor that has shifted the trajectory of that evolution, accelerating some trends and introducing others. This is not a story of moving from one state to another, but rather shifting the ongoing arc of change. In this brief we examine the shifts before the pandemic, where COVID‐19 has taken us, and implications for future research.
Donald Berry, Scott Berry, Peter Hale, Leah Isakov, Andrew Lo, Kien Wei Siah, Chi Heem Wong
We compare and contrast the expected duration and number of infections and deaths averted among several designs for clinical trials of COVID-19 vaccine candidates, including traditional and adaptive randomized clinical trials and human challenge trials. Using epidemiological models calibrated to the current pandemic, we simulate the time course of each clinical trial design for 756 unique combinations of parameters, allowing us to determine which trial design is most effective for a given scenario. A human challenge trial provides maximal net benefits — averting an additional 1.1M infections and 8,000 deaths in the U.S. compared to the next best clinical trial design — if its set-up time is short or the pandemic spreads slowly. In most of the other cases, an adaptive trial provides greater net benefits.
Abhijit Banerjee, Michael Faye, Alan Krueger, Paul Niehaus, Tavneet Suri
We examine some effects of Universal Basic Income during the COVID-19 pandemic using a large-scale experiment in rural Kenya. Transfers significantly improved well-being on common measures such as hunger, sickness, and depression in spite of the pandemic, but with modest effect sizes. They may have had public health benefits, as they reduced hospital visits and decreased social (but not commercial) interactions that influence contagion rates. During the pandemic (and contemporaneous agricultural lean season) recipients lost the income gains from starting new non-agricultural enterprises that they had initially obtained, but also suffered smaller increases in hunger. This pattern is consistent with the idea that UBI induced recipients to take on more income risk in part by mitigating the most harmful consequences of adverse shocks.
Daniel L. Greenwald, John Krainer, Pascal Paul
Aggregate bank lending to firms expands following a number of adverse macroeconomic shocks, such as the outbreak of COVID-19 or a monetary policy tightening. Using loan-level supervisory data, we show that these dynamics are driven by draws on credit lines by large firms. Banks that experience larger drawdowns restrict term lending more—an externality onto smaller firms. Using a structural model, we show that credit lines are necessary to reproduce the flow of credit toward less constrained firms after adverse shocks. While credit lines increase total credit growth, their redistributive effects exacerbate the fall in investment.
Josue Cox, Daniel L. Greenwald, Sydney C. Ludvigson
What explains stock market behavior in the early weeks of the coronavirus pandemic? Estimates from a dynamic asset pricing model point to wild fluctuations in the pricing of stock market risk, driven by shifts in risk aversion or sentiment. We find further evidence that the Federal Reserve played a role in these fluctuations, via a series of announcements outlining unprecedented steps to provide several trillion dollars in loans to support the economy. As of July 31 of 2020, however, only a tiny fraction of the credit that the central bank announced it stood ready to provide in early April had been extended, reinforcing the conclusion that market movements during COVID-19 have been more reflective of sentiment than substance.
Christopher L.F. Sun, Eugenio Zuccarelli, El Ghali A. Zerhouni, Jason Lee, James Muller, Karen M. Scott, Alida M. Lujan, Retsef Levi
A machine-learning model trained on COVID-19 outcomes from 1,146 nursing homes identified a nursing home’s county’s infection rate, number of units, historical health deficiencies from Centers of Medicare and Medicaid Services inspections, percent of non-Hispanic White residents, and density as predictive of infection risk.
We use recent data and research results to approximate the probability that an air traveler in coach will contract COVID-19 on a U.S. domestic flight two hours long, both when all coach seats are full and when all but middle seats are full. The point estimates we reach based on data from late June 2020 are 1 in 4,300 for full flights and 1 in 7,700 when middle seats are kept empty. These estimates are subject to both quantifiable and nonquantifiable sources of uncertainty, and sustain known margins of error of a factor about 2.5. However, because uncertainties in key parameters affect both risk estimates the same way, they leave the relative risk ratio for fill all seats compared to middle seat open close to 1.8 (i.e., close to 1/4,300)/(1/7,700). We estimate the mortality risks caused by COVID-19 infections contracted on airplanes, taking into account that infected passengers can in turn infect others. The point estimates, which use 2019 data about the percentage of seats actually occupied on U.S. flights, range from one death per 400,000 passengers to one death per 600,000. These death-risk levels are considerably higher than those associated with plane crashes but comparable to those arising from two hours of everyday activities during the pandemic.
Dimitris Bertsimas, Leonard Boussioux, Ryan Cory-Wright, Arthur Delarue, Vasileios Digalakis, Alexandre Jacquillat, Driss Lahlou Kitane, Galit Lukin, Michael Li, Luca Mingardi, Omid Nohadani, Agni Orfanoudaki, Theodore Papalexopoulos, Ivan Paskov, Jean Pauphilet, Omar Skali Lami, Bartolomeo Stellato, Hamza Tazi Bouardi, Kimberly Villalobos Carballo, Holly Wiberg, Cynthia Zeng
The COVID-19 pandemic has created unprecedented challenges worldwide. Strained healthcare providers make difficult decisions on patient triage, treatment and care management on a daily basis. Policy makers have imposed social distancing measures to slow the disease, at a steep economic price. We design analytical tools to support these decisions and combat the pandemic. Specifically, we propose a comprehensive data-driven approach to understand the clinical characteristics of COVID-19, predict its mortality, forecast its evolution, and ultimately alleviate its impact. By leveraging cohort-level clinical data, patient-level hospital data, and census-level epidemiological data, we develop an integrated four-step approach, combining descriptive, predictive and prescriptive analytics. First, we aggregate hundreds of clinical studies into the most comprehensive database on COVID-19 to paint a new macroscopic picture of the disease. Second, we build personalized calculators to predict the risk of infection and mortality as a function of demographics, symptoms, comorbidities, and lab values. Third, we develop a novel epidemiological model to project the pandemic’s spread and inform social distancing policies. Fourth, we propose an optimization model to re-allocate ventilators and alleviate shortages. Our results have been used at the clinical level by several hospitals to triage patients, guide care management, plan ICU capacity, and re-distribute ventilators. At the policy level, they are currently supporting safe back-to-work policies at a major institution and vaccine trial location planning at Janssen Pharmaceuticals, and have been integrated into the US Center for Disease Control’s pandemic forecast.
Lingzhi Li, Michael, Hamza Tazi Bouardi, Omar Skali Lami, Thomas A. Trikalinos, Nikolaos K. Trichakis, Dimitris Bertsimas
During the COVID-19 epidemic, governments around the world have implemented unprecedented non-pharmaceutical measures to control its spread. As these measures carry significant economic and humanitarian cost, it is an important topic to investigate the efficacy of different policies and accurately project the future spread under such said policies. We developed a novel epidemiological model, DELPHI, based on the established SEIR model, that explicitly captures government interventions, under-detection, and many other realistic effects. We estimate key biological parameters using a meta-analysis of over 190 COVID-19 research papers and fit DELPHI to over 167 geographical areas since early April. We extract the inferred government intervention effect from DELPHI. Our epidemiological model recorded 6% and 11% two-week out-of-sample Median Absolute Percentage Error on cases and deaths, and successfully predicted the severity of epidemics in many areas (including US, UK and Russia) months before it happened. Using the extracted government response, we find mass gathering restrictions and school closings on average reduced infection rates the most, at 29.9 ± 6.9% and 17.3 ± 6.7%, respectively. The most stringent policy, stay-at-home, on average reduced the infection rate by 74.4 ± 3.7% from baseline across countries that implemented it. We also further show that a reversal of stay-at-home policies in some countries, such as Brazil, could have disastrous results by end of July. Our findings highlight that among the widely implemented policies around the world, mass gathering restrictions and school closings appear to be the most effective policies in reducing the infection rate. Given the continued spread of the epidemic in many countries, we recommend these policies to continue to the extent that they can be feasibly implemented. Our results also show that under an assumption of R0 of 2.5-3 for COVID-19, stay-at-home policies appear to be the only effective policy that was widely implemented in reducing the R0 below 1. This implies that stay-at-home policies might be necessary, for at least the vulnerable population, if an uncontrolled second wave reemerges.
Antonio A. Arechar, David G. Rand
On March 16, 2020, President Trump introduced strict social distancing guidelines for the United States, in an effort to stem the spread of the COVID-19 pandemic. This had an immediate major effect on the job market, with millions of Americans forced to find alternative ways to make a living from home. Here, we investigate the possibility that this policy also changed the pool of workers available to take part in academic studies on Amazon Mechanical Turk (MTurk) – either by influencing which existing MTurk workers participate, or by causing an influx of new ones. Specifically, we look at 10,510 responses gathered in 16 studies run between February 25, 2020, and May 14, 2020, examining the distribution of gender, age, ethnicity, political preferences, and analytic cognitive style. We find important changes on all measures following the imposition of nationwide social distancing: participants are more substantially less reflective (as measured by the Cognitive Reflection Test, CRT) and more Republican, and somewhat less likely to be white and experienced with MTurk. Most of these differences are explained by an influx of new participants who are somewhat less attentive than, and demographically different from, previous participants.
Hazhir Rahmandad, TY Lim, John Sterman
Effective responses to the COVID-19 pandemic require integrating behavioral factors such as risk-driven contact reduction, improved treatment, and adherence fatigue with asymptomatic transmission, disease acuity, and hospital capacity. We build one such model and estimate it for all 92 nations with reliable testing data. Cumulative cases and deaths through 22 December 2020 are estimated to be 7.03 and 1.44 times official reports, yielding an infection fatality rate (IFR) of 0.51% which has been declining over time. Absent adherence fatigue cumulative cases would have been 47% lower. Scenarios through June 2021 show that modest improvement in responsiveness could reduce cases and deaths by about 14%, more than the impact of vaccinating half of the population by that date. Variations in responsiveness to risk explain two orders of magnitude difference in per-capita deaths despite reproduction numbers fluctuating ~ 1 across nations. A public online simulator facilitates scenario analysis over the coming months.
Kenneth T. Gillingham, Christopher R. Knittel, Jing Li, Martin Ovaere, Mar Reguant
We explore how the short-run effects of Covid-19 in reducing CO2 and local air pollutant emissions can easily be outweighed by the long-run effects of a slowing of clean energy innovation. Focusing on the United States, we show that in the short run, COVID-19 has reduced consumption for jet fuel and gasoline dramatically, by 50% and 30% respectively, while electricity demand has declined by less than 10%. CO2 emissions have declined by 15%, while local air pollutants have also declined, saving about 200 lives per month. However, there could be a deep impact on long-run innovation in clean energy, leading to an additional 2,500 MMT CO2 and 40 deaths per month on average to 2035. Even pushing back renewable electricity generation investments by one year would outweigh the emission reductions and avoided deaths from March-June 2020. The policy response will determine how COVID-19 ultimately influences the future path of emissions.
Christos Makridis, Robert McNab
How has the COVID-19 pandemic affected state budgets? Using a quarterly panel of states between 1994 and 2019, we estimate how changes in employment affect tax revenues. We find that a one percentage point (pp) rise in employment is associated with a 1.56pp rise in total tax revenue, which is concentrated among sales taxes (a 1.19pp increase), individual income taxes (a 1.63pp increase), and corporate income taxes (a 4.13pp increase). These results are robust to a wide array of controls, such as state composition and housing price growth, and instrumental variable specifications. After estimating state-specific elasticities, and forecasting counterfactual employment within states using continued unemployment claims, we find that the average state will experience a 20% decline in their tax revenues (or $4.9 billion) and $254.8 billion nationally in 2012 prices, assuming a recovery that follows trends up to April. We are incorporating additional employment to account for the economic recovery.
Jackie Baek, Vivek F. Farias, Andreea Georgescu, Retsef Levi, Tianyi Peng, Deeksha Sinha, Joshua Wilde, Andrew Zheng
A multitude of forecasting efforts have arisen to support management of the ongoing COVID-19 epidemic. These efforts typically rely on a variant of the SIR process and have illustrated that building effective forecasts for an epidemic in its early stages is challenging. This is perhaps surprising since these models rely on a small number of parameters and typically provide an excellent retrospective fit to the evolution of a disease. So motivated, we provide an analysis of the limits to estimating an SIR process. We show that no unbiased estimator can hope to learn this process until observing enough of the epidemic so that one is approximately two-thirds of the way to reaching the peak for new infections. Our analysis provides insight into a regularization strategy that permits effective learning across simultaneously and asynchronously evolving epidemics. This strategy has been used to produce accurate, granular predictions for the COVID-19 epidemic that has found large-scale practical application in a large US state.
Eric Friedman, John Friedman, Simon Johnson, Adam Landsberg
In the face of elevated pandemic risk, canonical epidemiological models imply the need for extreme social distancing over a prolonged period. Alternatively, people could be organized into zones, with more interactions inside their zone than across zones. Zones can deliver significantly lower infection rates, with less social distancing, particularly if combined with simple quarantine rules and contact tracing. This paper provides a framework for understanding and evaluating the implications of zones, quarantines, and other complementary policies.
Dimitris Papanikolaou, Lawrence Schmidt
We analyze the supply-side disruptions associated with COVID-19 across firms and workers. To do so, we exploit differences in the ability of workers across industries to work remotely using data from the American Time Use Survey. We find that sectors in which a higher fraction of the workforce is not able to work remotely experienced significantly greater declines in employment, significantly more reductions in expected revenue growth, worse stock market performance, and higher expected likelihood of default. In terms of individual employment outcomes, lower-paid workers, especially female workers with young children, were significantly more affected by these disruptions. Last, we combine these ex-ante heterogeneous industry exposures with daily financial market data to create a stock return portfolio that most closely replicates the supply-side disruptions resulting from the pandemic.
Umair Ali, Chris M. Herbst, Christos Makridis
Stay-at-home orders (SAHOs) have been implemented in most U.S. states to mitigate the spread of COVID-19. This paper quantifies the short-run impact of these containment policies on the supply of and demand for child care. The child care market may be particularly vulnerable to a SAHO-type policy shock, given that many providers are liquidity-constrained. Using plausibly exogenous variation from the staggered adoption of SAHOs across states, we find that online job postings for early care and education teachers declined by 13% after enactment. This effect is driven exclusively by private-sector services. Indeed, hiring by public programs like Head Start and pre-kindergarten has not been influenced by SAHOs. In addition, we find little evidence that child care search behavior among households has been altered. Because forced supply-side changes appear to be at play, our results suggest that households may not be well-equipped to insure against the rapid transition to the production of child care. We discuss the implications of these results for child development and parental employment decisions.
Yan Leng, Yujia Zhai, Shaojing Sun, Yifei Wu, Jordan Selzer, Sharon Strover, Julia Fensel, Alex Pentland, Ying Ding
COVID-19 resulted in an infodemic, which could erode public trust, impede virus containment, and outlive the pandemic itself. The evolving and fragmented media landscape is a key driver of the spread of misinformation. Using misinformation identified by the fact-checking platform by Tencent and posts on Weibo, our results showed that the evolution of misinformation follows an issue-attention cycle, pertaining to topics such as city lockdown, cures, and preventions, and school reopening. Sources of authority weigh in on these topics, but their influence is complicated by peoples' pre-existing beliefs and cultural practices. Finally, social media has a complicated relationship with established or legacy media systems. Sometimes they reinforce each other, but in general, social media may have a topic cycle of its own making. Our findings shed light on the distinct characteristics of misinformation during the COVID-19 and offer insights into combating misinformation in China and across the world at large.
David Holtz, Michael Zhao, Seth G. Benzell, Cathy Y. Cao, M. Amin Rahimian, Jeremy Yang, Jennifer Allen, Avinash Collis, Alex Moehring, Tara Sowrirajan, Dipayan Ghosh, Yunhao Zhang, Paramveer S. Dhillon, Christos Nicolaides, Dean Eckles, Sinan Aral
Social distancing is the core policy response to coronavirus disease 2019 (COVID-19). But, as federal, state and local governments begin opening businesses and relaxing shelter-in-place orders worldwide, we lack quantitative evidence on how policies in one region affect mobility and social distancing in other regions and the consequences of uncoordinated regional policies adopted in the presence of such spillovers. To investigate this concern, we combined daily, county-level data on shelter-in-place policies with movement data from over 27 million mobile devices, social network connections among over 220 million Facebook users, daily temperature and precipitation data from 62,000 weather stations, and county-level census data on population demographics to estimate the geographic and social network spillovers created by regional policies across the United States. Our analysis shows that the contact patterns of people in a given region are significantly influenced by the policies and behaviors of people in other, sometimes distant, regions. When just one-third of a state’s social and geographic peer states adopt shelter-in-place policies, it creates a reduction in mobility equal to the state’s own policy decisions. These spillovers are mediated by peer travel and distancing behaviors in those states. A simple analytical model calibrated with our empirical estimates demonstrated that the “loss from anarchy” in uncoordinated state policies is increasing in the number of noncooperating states and the size of social and geographic spillovers. These results suggest a substantial cost of uncoordinated government responses to COVID-19 when people, ideas, and media move across borders.
Christos Makridis, Tao Wang
How do individuals adjust their consumption in response to information disseminated through peers and the social network? Using new micro-data on consumption, coupled with geographic friendship ties to measure social connectivity, this paper quantifies the role of social networks as a propagation mechanism for understanding aggregate fluctuations in consumption. Using the COVID-19 pandemic as a source of variation, we find that a 10% rise in cases and deaths in counties connected through the social network is associated with a 0.64% and 0.33% decline in consumption expenditures--roughly three to seven times as large as the direct effects of local cases or deaths. Counties more socially connected to epicenter countries of the pandemic also saw a bigger drop in consumption. These effects are concentrated among consumer goods and services that rely more on social-contact, suggesting that individuals incorporate the experiences from their social network to inform their own consumption choices. We are working on incorporating this micro-economic evidence into a heterogeneous agent model and social interaction to study the aggregate demand implications.
Pierre Azoulay, Benjamin Jones
As coronavirus disease 2019 (COVID-19) has spread, public health and economic well-being are increasingly in conflict. Governments are prioritizing public health, but the current solution—social isolation—is costly as commerce remains shut down. Restarting economies could rekindle the pandemic and cause even worse human suffering. Innovation can help societies escape the untenable choice between public and economic health. The world needs effective vaccines, therapies, or other solutions. But how do we achieve these solutions, and achieve them quickly?
Ran Xu, Hazhir Rahmandad, Marichi Gupta, Catherine DiGennaro, Navid Ghaffarzadegan, Mohammad Jalali
Background: Understanding and projecting the spread of COVID-19 requires reliable estimates of how weather components are associated with the transmission of the virus. Prior research on this topic has been inconclusive. Identifying key challenges to reliable estimation of weather impact on transmission we study this question using one of the largest assembled databases of COVID-19 infections and weather.
Methods: We assemble a dataset that includes virus transmission and weather data across 3,739 locations from December 12, 2019 to April 22, 2020. Using simulation, we identify key challenges to reliable estimation of weather impacts on transmission, design a statistical method to overcome these challenges, and validate it in a blinded simulation study. Using this method and controlling for location-specific response trends we estimate how different weather variables are associated with the reproduction number for COVID-19. We then use the estimates to project the relative weather-related risk of COVID-19 transmission across the world and in large cities.
Results: We show that the delay between exposure and detection of infection complicates the estimation of weather impact on COVID-19 transmission, potentially explaining significant variability in results to-date. Correcting for that distributed delay and offering conservative estimates, we find a negative relationship between temperatures above 25 degrees Celsius and estimated reproduction number (R ̂), with each degree Celsius associated with a 3.1% (95% CI, 1.5% to 4.8%) reduction in R ̂. Higher levels of relative humidity strengthen the negative effect of temperature above 25 degrees. Moreover, one millibar of additional pressure increases R ̂ by approximately 0.8% (95% CI, 0.6% to 1%) at the median pressure (1016 millibars) in our sample. We also find significant positive effects for wind speed, precipitation, and diurnal temperature on R ̂. Sensitivity analysis and simulations show that results are robust to multiple assumptions. Despite conservative estimates, weather effects are associated with a 43% change in R ̂ between the 5th and 95th percentile of weather conditions in our sample.
Conclusions: These results provide evidence for the relationship between several weather variables and the spread of COVID-19. However, the (conservatively) estimated relationships are not strong enough to seasonally control the epidemic in most locations.
Daron Acemoglu, Victor Chernozhukov, Iván Werning, Michael D. Whinston
We develop a multi-risk SIR model (MR-SIR) where infection, hospitalization and fatality rates vary between groups — in particular between the “young,” the “middle-aged” and the “old.” Our MR-SIR model enables a tractable quantitative analysis of optimal policy similar to those already developed in the context of the homogeneous-agent SIR models. For baseline parameter values for the COVID-19 pandemic applied to the U.S., we find that optimal policies differentially targeting risk/age groups significantly outperform optimal uniform policies and most of the gains can be realized by having stricter lockdown policies on the oldest group. For example, for the same economic cost (24.3% decline in GDP), optimal semi–targeted or fully-targeted policies reduce mortality from 1.83% to 0.71% (thus, saving 2.7 million lives) relative to optimal uniform policies. Intuitively, a strict and long lockdown for the most vulnerable group both reduces infections and enables less strict lockdowns for the lower-risk groups. We also study the impacts of social distancing, the matching technology, the expected arrival time of a vaccine, and testing with or without tracing on optimal policies. Overall, targeted policies that are combined with measures that reduce interactions between groups and increase testing and isolation of the infected can minimize both economic losses and deaths in our model.
Jay J. Van Bavel, Katherine Baicker, Paulo S. Boggio, Valerio Capraro, Aleksandra Cichocka , Mina Cikara, Molly J. Crockett, Alia J. Crum, Karen M. Douglas , James N. Druckman, John Drury, Oeindrila Dube, Naomi Ellemers, Eli J. Finkel, James H. Fowler , Michele Gelfand , Shihui Han , S. Alexander Haslam , Jolanda Jetten , Shinobu Kitayama , Dean Mobbs , Lucy E. Napper, Dominic J. Packer , Gordon Pennycook , Ellen Peters , Richard E. Petty , David G. Rand , Stephen D. Reicher, Simone Schnall , Azim Shariff , Linda J. Skitka, Sandra Susan Smith, Cass R. Sunstein , Nassim Tabri , Joshua A. Tucker , Sander van der Linden , Paul van Lange, Kim A. Weeden , Michael J. A. Wohl , Jamil Zaki, Sean R. Zion , Robb Willer
The COVID-19 pandemic represents a massive global health crisis. Because the crisis requires large-scale behaviour change and places significant psychological burdens on individuals, insights from the social and behavioural sciences can be used to help align human behaviour with the recommendations of epidemiologists and public health experts. Here we discuss evidence from a selection of research topics relevant to pandemics, including work on navigating threats, social and cultural influences on behaviour, science communication, moral decision-making, leadership, and stress and coping. In each section, we note the nature and quality of prior research, including uncertainty and unsettled issues. We identify several insights for effective response to the COVID-19 pandemic and highlight important gaps researchers should move quickly to fill in the coming weeks and months.
Robert S. Pindyck
I use a simple SIR model, augmented to include deaths, to elucidate how pandemic progression is affected by the control of contagion, and examine the key trade-offs that underlie policy design. I illustrate how the cost of reducing the “reproduction number” R0 depends on how it changes the infection rate, the total and incremental number of deaths, the duration of the pandemic, and the possibility and impact of a second wave. Reducing R0 reduces the number of deaths, but extends the duration (and hence economic cost) of the pandemic, and it increases the fraction of the population still susceptible at the end, raising the possibility of a second wave. The benefit of reducing R0 is largely lives saved, and the incremental number of lives saved rises as R0 is reduced. But using a VSL estimate to value those lives is problematic.
Augustin Landier, David Thesmar
We analyze the dynamics of earnings forecasts and discount rates implicit in valuations during the COVID-19 crisis. Forecasts over 2020 earnings have been progressively reduced by 16%. Longer-run forecasts have reacted much less. We estimate an implicit discount rate going from 8.5% in mid-February to 11% at the end of March and reverting to its initial level in mid-May. Over the period, the unlevered asset risk premium increases by 50bp, the leverage effect also increases by 50bp, while the risk free rate decreases by 100bp. Hence, analysts’ forecast revisions explain all of the decrease in equity values between January 2020 and mid-May 2020.
Joao Granja, Christos Makridis, Constantine Yannelis, Eric Zwick
This paper takes an early look at the Paycheck Protection Program (PPP), a large and novel small business support program that was part of the initial policy response to the COVID-19 pandemic. We use new data on the distribution of PPP loans and high-frequency micro-level employment data to consider two dimensions of program targeting. First, we do not find evidence that funds flowed to areas more adversely affected by the economic effects of the pandemic, as measured by declines in hours worked or business shutdowns. If anything, funds flowed to areas less hard hit. Second, we find significant heterogeneity across banks in terms of disbursing PPP funds, which does not only reflect differences in underlying loan demand. The top-4 banks alone account for 36% of total pre-policy small business loans, but disbursed less than 3% of all PPP loans. Areas that were significantly more exposed to low-PPP banks received much lower loan allocations. As data become available, we will study employment and establishment responses to the program and the impact of PPP support on the economic recovery. Measuring these responses is critical for evaluating the social insurance value of the PPP and similar policies.
Christos Makridis, Cary Wu
The COVID-19 pandemic represents the largest worldwide shock in at least a decade. Moreover, the spread of the virus has been highly heterogeneous. This paper investigates the role of social capital as a potential mediating factor for the spread of the COVID-19 virus. On one hand, higher social capital could imply greater in-person interaction and risk of contagion. On the other hand, because social capital is associated with greater trust and relationships within a community, it could endow individuals with a greater concern for others, thereby leading to more hygienic practices and social distancing. Our results suggest that moving a county from the 25th to the 75th percentile of the distribution of social capital would lead to a 20% decline in the number of infections, as well as a 0.28 percentage point decline in the growth rate of the virus (nearly 20% of the median growth rate). Moreover, our results are robust to many demographic characteristics, controls, and alternative measures of social capital.
Maryam Farboodi, Gregor Jarosch, Robert Shimer
We use a conventional dynamic economic model to integrate individual optimization, equilibrium interactions, and policy analysis into the canonical epidemiological model. Our tractable framework allows us to represent both equilibrium and optimal allocations as a set of differential equations that can jointly be solved with the epidemiological model in a unified fashion. Quantitatively, the laissez-faire equilibrium accounts for the decline in social activity we measure in U.S. micro-data from SafeGraph. Relative to that, we highlight three key features of the optimal policy: it imposes immediate, discontinuous social distancing; it keeps social distancing in place for a long time or until treatment is found; and it is never extremely restrictive, keeping the effective reproduction number mildly above the share of the population susceptible to the disease.
Seth Benzell, Avinash Collis, Christos Nicolaides
To prevent the spread of COVID-19, some types of stores and gathering places have been shut down while others remain open. The decision to shut down one type of location and leave another open constitutes a judgement about the relative danger and benefits of those locations. Using location data from a large sample of smartphones, nationally representative consumer preference surveys, and government statistics, we measure the relative transmission risk benefit and social cost of closing 30 different location categories in the U.S. Our categories include types of shops, schools, entertainments, and public spaces. We rank categories by those which should face stricter regulation via dominance across eight dimensions of risk and importance and through composite indexes. We find that from February to March, there were larger declines in visits to locations that our measures imply should be closed first. We hope this analysis will help policymakers decide how to reopen their economies.
Eric Friedman, John Friedman, Simon Johnson, Adam Landsberg
In the face of elevated pandemic risk, is it necessary to completely lock down the population, imposing extreme social distancing? Canonical epidemiological models suggest this may be unavoidable for months at a time, despite the heavy social and human cost of physically isolating people. Alternatively, people could retreat into socially or economically defined defensive zones, with more interactions inside their zone than across zones. Starting from a complete lockdown, zones could facilitate responsible reopening of education, government, and firms, as a well-implemented structure can dramatically slow the diffusion of the disease. This paper provides a framework for understanding and evaluating the effectiveness of zones for social distancing.
Gordon Pennycook, Jonathon McPhetres, Bence Bago, David G. Rand
The COVID-19 pandemic has become more political in the U.S.A. than in similar Western countries, allowing for a novel test of attitude polarization. Furthermore, past work disagrees about the role of cognitive sophistication (relative to ideology) in the formation of science beliefs. We therefore investigated the roles of political ideology and cognitive sophistication in explaining COVID-19 attitudes across the U.S.A. (N=689), the U.K. (N=642), and Canada (N=644). Polarization was greater in the U.S. than in the U.K., but not Canada. Furthermore, in all three countries, cognitive sophistication correlated negatively with misperceptions – and in fact was a stronger predictor than political ideology. We also found no evidence that cognitive sophistication was associated with increased polarization, contrary to identity-protective cognition accounts of motivated reasoning. Thus, although there is some evidence for political polarization, accurate beliefs about COVID-19 were broadly associated with the quality of one’s reasoning regardless of political polarization.
Qingyang Xu, Shomesh Chaudhuri, Danying Xiao, Andrew W. Lo
In the midst of epidemics such as COVID-19, therapeutic candidates are unlikely to be able to complete the usual multi-year clinical trial and regulatory approval process within the course of an outbreak. We apply a Bayesian adaptive patient-centered model---which minimizes the expected harm of false positives and false negatives---to optimize the clinical trial development path during such outbreaks. When the epidemic is more infectious and fatal, the Bayesian-optimal sample size in the clinical trial is lower and the optimal statistical significance level is higher. For COVID-19 (assuming a static R0=2 and initial infection percentage of 0.1%), the optimal significance level is 7.1% for a clinical trial of a non-vaccine anti-infective therapeutic clinical trial and 13.6% for that of a vaccine. For a dynamic R0 ranging from 2 to 4, the corresponding values are 14.4% and 26.4%, respectively. Our results illustrate the importance of adapting the clinical trial design and the regulatory approval process to the specific parameters and stage of the epidemic.
Chi Heem Wong, Kien Wei Siah, Andrew W. Lo
A key driver in biopharmaceutical investment decisions is the probability of success of a drug development program. We estimate the probabilities of success (PoS) of clinical trials for vaccines and other anti-infective therapeutics using 43,414 unique triplets of clinical trial, drug, and disease between January 1, 2000, and January 7, 2020, yielding 2,544 vaccine programs and 6,829 non-vaccine programs targeting infectious diseases. The overall estimated PoS for an industry-sponsored vaccine program is 39.6%, and 16.3% for an industry-sponsored anti-infective therapeutic. Among industry-sponsored vaccines programs, only 12 out of 27 disease categories have seen at least one approval, with the most successful being against monkeypox (100%), rotavirus (78.7%), and Japanese encephalitis (67.6%). The three infectious diseases with the highest PoS for industry-sponsored non-vaccine therapeutics are smallpox (100%), CMV (31.8%), and onychomycosis (29.8%). Non-industry-sponsored vaccine and non-vaccine development programs have lower overall PoSs: 6.8% and 8.2%, respectively. Viruses involved in recent outbreaks---MERS, SARS, Ebola, Zika---have had a combined total of only 45 non-vaccine development programs initiated over the past two decades, and no approved therapy to date (Note: our data was obtained just before the COVID-19 outbreak and do not contain information about the programs targeting this disease.) These estimates offer guidance both to biopharma investors as well as to policymakers seeking to identify areas most likely to be undeserved by private-sector engagement and in need of public-sector support.
Erez Yeoli, David Rand
Addressing public good problems typical requires people to adopt behaviors that are personally burdensome but beneficial for society. For instance, during the COVID-19 pandemic, people have been asked to stay home except in extenuating circumstances, maintain social distance, and wash their hands frequently. What is the best way to phrase one’s requests to ensure that people are maximally motivated to adhere? We distill three key insights from the behavioral science literature on social norms to create a simple messaging checklist: communicate the benefit to the community; make the ask unambiguous, categorical, and concise; and generate the impression that others expect compliance. We justify this guidance and illustrate it using practical examples, with a focus on COVID-19 prevention behaviors.
Avi Goldfarb, Catherine E. Tucker
This paper seeks to answer the simple question of what category of retail outlets generates the most physical interactions in the regular course of life. In this way, we aim to bring a marketing perspective to discussions about which businesses may be most risky from the standpoint of spreading contagious disease. We use detailed data from people's mobile devices prior to the implementation of social distancing measures in the United States. With this data, we examine a number of potential indicators of risk of contagion: The absolute number of visits and visitors, how many of the visits are generated by the same people, the median average distance traveled by the visitor to the retailer, and the number of customers from Canada and Mexico. We find that retailers with a single outlet tend to attract relatively few visitors, fewer one-off visitors, and have fewer international customers. For retailers that have multiple stores the patterns are non-linear. Retailers that have such a large number of stores that they are ubiquitous tend to exhibit fewer visits and visitors and attract customers from a smaller distance. However, retailers that have a large enough footprint to be well known, but not large enough to be ubiquitous tend to attract a large number of visitors who make one-off visits, travel a long distance, and are disproportionately international.
Ananya Sen, Catherine E. Tucker
Social distancing directives across the United States have led to school closures. Some districts are moving towards online instruction, but this requires internet access at home. We examine the factors that determine whether school children have access to the internet at home. We document that poor and non-white children still have lower access to the internet. Moreover, in areas where poor and non-white children have relatively lower test scores, such children are more likely to not have access to the internet. However, there is some evidence of positive spillovers from the historic presence of [information and communication technologies] industries in the local area in improving the access of disadvantaged children to the internet. The empirical insights highlight how the digital divide might exacerbate existing educational inequalities in the face of school closures due to social distancing.
Erik Brynjolfsson, John Horton, Adam Ozimek, Daniel Rock, Garima Sharma, Hong Yi Tu Ye
We report the results of a nationally representative sample of the U.S. population on how they are adapting to the COVID-19 pandemic. The survey ran from April 1 - 5, 2020. Of those employed four weeks earlier, 34.1% report they were commuting and are now working from home. In addition, 11.8% report being laid off or furloughed in the last four weeks. There is a strong negative relationship between the fraction in a state still commuting to work and the fraction working from home which suggests that many workers currently commuting could be converted to remote workers. We find that the share of people switching to remote work can be predicted by the incidence of COVID-19 and that younger people were more likely to switch to remote work. Furthermore, using data on state unemployment insurance claims, we find that states with higher fractions of remote workers have higher than expected UI claims.
Jillian Jordan, Erez Yoeli, David G. Rand
The COVID-19 pandemic threatens millions of lives, and an effective response will require individuals to take costly and difficult measures to slow the rate of transmission. Yet it is unclear how to best motivate preventative actions, which can be conceptualized as either self-interested or cooperative efforts. Should public health messaging focus on the benefits of prevention to individuals, society, or both? We shed light on this question across two studies conducted online via Amazon Mechanical Turk (total n = 2176 Americans) during the early days of the COVID-19 pandemic reaching the United States. We investigated the effects of three treatments, consisting of a written appeal and a flier, on intentions to engage in coronavirus prevention behaviors. We presented identical information across treatments, but varied our framing to emphasize the personal, public, or both personal and public benefits of prevention behaviors. We found evidence for the power of prosocial framing: the public treatment was more effective than the personal treatment, and the personal + public treatment was no more effective than the pure public treatment. Our results thus suggest that emphasizing the public benefits of prevention efforts may be an effective pandemic response strategy.
Catherine E. Tucker, Shuyi Yu
This paper presents some of the first evidence on the effect of the spread of coronavirus (COVID-19) in the U.S. on retail footfall traffic. The paper uses granular visit data from cell-phone tracking to estimate the shift in visits to different types of restaurants as coronavirus spread in the U.S. across the first three weeks of March 2020. The descriptive empirical work provides three useful insights. First, the precise level of coronavirus spread in the state or the timing of any in-person dining ban in the state has had far smaller effects than the pronounced nationwide overall collapse in demand. Second, there is little evidence of substitution towards restaurants focused on delivery as a result of the bans. Though dine-in restaurants suffered the largest drop in customers as a result of state-imposed restaurant bans, quick-service restaurants experienced a steep decline. Last, the biggest individual effects of these state-specific bans appears to have been [on] top-ranked brands focusing on full-service dining. Both top and non-top-ranked brands suffered drops for restaurants not focused on dining in, with top brands suffering a slightly smaller decline.
Lesley Chiou, Catherine E. Tucker
This paper measures the role of the diffusion of high-speed internet on an individual's ability to self-isolate during a global pandemic. We use data that tracks 20 million mobile devices and their movements across physical locations, and whether the mobile devices leave their homes that day. We show that while income is correlated with differences in the ability to stay at home, the unequal diffusion of high-speed internet in homes across regions drives much of this observed income effect. We examine compliance with state-level directives to avoid leaving your home. Devices in regions with either high-income or high-speed internet are less likely to leave their homes after such a directive. However, the combination of having both high income and high-speed internet appears to be the biggest driver of propensity to stay at home. Our results suggest that the digital divide — or the fact that income and home internet access are correlated — appears to explain much inequality we observe in people's ability to self-isolate.
Sergio Correia, Stephan Luck, Emil Verner
What are the economic consequences of an influenza pandemic? And given the pandemic, what are the economic costs and benefits of non-pharmaceutical interventions (NPI)? Using geographic variation in mortality during the 1918 flu pandemic in the U.S., we find that more exposed areas experience a sharp and persistent decline in economic activity. The estimates imply that the pandemic reduced manufacturing output by 18%. The downturn is driven by both supply- and demand-side channels. Further, building on findings from the epidemiology literature establishing that NPIs decrease influenza mortality, we use variation in the timing and intensity of NPIs across U.S. cities to study their economic effects. We find that cities that intervened earlier and more aggressively do not perform worse and, if anything, grow faster after the pandemic is over. Our findings thus indicate that NPIs not only lower mortality; they may also mitigate the adverse economic consequences of a pandemic.
Jonathan Vu, Benjamin Kaplan, Shomesh Chaudhuri, Monique Mansoura, Andrew Lo
Recent outbreaks of infectious pathogens such as Zika, Ebola, and COVID-19 have underscored the need for the dependable availability of vaccines against emerging infectious diseases (EIDs). The cost and risk of R&D programs and uniquely unpredictable demand for EID vaccines have discouraged vaccine developers, and government and nonprofit agencies have been unable to provide timely or sufficient incentives for their development and sustained supply. We analyze the economic returns of a portfolio of EID vaccine assets, and find that under realistic financing assumptions, the expected returns are significantly negative, implying that the private sector is unlikely to address this need without public-sector intervention. We have sized the financing deficit for this portfolio and propose several potential solutions, including price increases, enhanced public-private partnerships, and subscription models through which individuals would pay annual fees to obtain access to a portfolio of vaccines in the event of an outbreak.
Christos Makridis, Jonathan Hartley
This paper provides an early estimate of the economic effects of the national quarantine policy associated with the COVID-19 pandemic in the United States. Using a measure of digital intensity from Gallipoli and Makridis (2018), we exploit counties' exposure to industries that vary in their digital intensity as a proxy for the adverse effects of the decline in physical activity and in-person shutdown. Our baseline estimates suggest that the average county experiences a 5% decline in real GDP growth for two months of economic shutdown, which amounts to $2.14 trillion. Our results also suggest that the effects are highly heterogeneous across countries: for example, the most adversely affected counties may experience as much as a 15% decline in real GDP growth. Counties with higher shares of college graduates, digital workers, and higher median household income are less adversely affected, whereas those with higher shares of non-tradables employment are more affected.
Navid Ghaffarzadegan, Hazhir Rahmandad
Background: The 2019 Coronavirus (COVID-19) has turned into a global pandemic with unprecedented challenges for the global community. Understanding the state of the disease and planning for future trajectories relies heavily on data on the spread and mortality. Yet official data coming from various countries are highly unreliable: Symptoms similar to common cold in majority of cases and limited screening resources and delayed testing procedures may contribute to under-estimation of the burden of disease. Anecdotal and more limited data are available, but few have systematically combined those with official statistics into a coherent view of the epidemic. This study is a modeling-in-real-time of the emerging outbreak for understanding the state of the disease. Our focus is on the case of the spread of disease in Iran, as one of the epicenters of the disease in the first months of 2020. Method: We develop a simple dynamic model of the epidemic to provide a more reliable picture of the state of the disease based on existing data. Building on the generic SEIR (Susceptible, Exposed, Infected, and Recovered) framework we incorporate two behavioral and logistical considerations. First we capture the endogenous changes in contact rate (average contact per person) as more deaths are reported. As a result the reproduction number changes endogenously in the model. Second we differentiate reported and true cases by including simple formulations for how only a fraction of cases might be diagnosed, and how that fraction changes in response to epidemic's progression. In estimating the model we use both the official data as well as the discovered infected travelers and unofficial medical community estimates and triangulate these sources to build a more complete picture. Calibration is completed by forming a likelihood function for observing the actual time series data conditional on model parameters, and conducting a Markov Chain Monte Carlo simulations. The model is used to estimate current "true" cases of infection and death. We analyze the future trajectory of the disease under six conditions related to the seasonal effects and policy measures targeting social distancing. Findings: The model closely replicates the past data but also shows the true number of cases is likely far larger. We estimate about 493,000 current infected cases (90% CI: 271K-810K) as of March 20, 2020. Our estimate for cumulative cases of infection until that date is 916,000 (90% CI: 508K, 1.5M), and for total death is 15,485 (90% CI: 8.4K, 25.8K). These numbers are significantly (more than one order of magnitude) higher than official statistics. The trajectory of the epidemic until the end of June could take various paths depending on the impact of seasonality and policies targeting social distancing. In the most optimistic scenario for seasonal effects, depending on policy measures, 1.6 million Iranians (90% CI: 0.9M-2.6M) are likely to get infected, and death toll will reach about 58,000 cases (90% CI: 32K-97K), while in the more pessimistic scenarios, death toll may exceed 103,000 cases (90% CI: 56K-172K). Implication: Our results suggest that the number of cases and deaths may be over an order of magnitude larger than official statistics in Iran. Absent extended testing capacity other countries may face a significant under-count of existing cases and thus be caught off guard about the actual toll of the epidemic.
Ian W. R. Martin, Robert S. Pindyck
Most of the literature on the economics of catastrophes assumes that such events cause a reduction in the stream of consumption, as opposed to widespread fatalities. Here we show how to incorporate death in a model of catastrophe avoidance, and how a catastrophic loss of life can be expressed as a welfare-equivalent drop in consumption. We examine how potential fatalities affect the policy interdependence of catastrophic events and “willingness to pay” (WTP) to avoid them. Using estimates of the “value of a statistical life” (VSL), we find the WTP to avoid major pandemics, and show it is large (10% or more of annual consumption) and partly driven by the risk of macroeconomic contractions. Likewise, the risk of pandemics significantly increases the WTP to reduce consumption risk. Our work links the VSL and consumption disaster literatures.
Gordon Pennycook, Jonathon, McPhetres, Yunhao Zhang, Jackson Lu, David G. Rand
Misinformation can amplify humanity's greatest challenges. A salient recent example of this is the COVID-19 pandemic, which has bred a multitude of falsehoods even as truth has increasingly become a matter of life and death. Here we investigate why people believe and spread false (and true) news content about COVID-19, and test an intervention intended to increase the truthfulness of the content people share on social media. Across two studies with over 1,600 participants (quota-matched to the American public on age, gender, ethnicity, and geographic region), we find support for the idea that people share false claims about COVID-19 in part because they simply fail to think sufficiently about whether or not content is accurate when deciding what to share. In Study 1, participants were far worse at discerning between true and false content when deciding what they would share on social media relative to when they are asked directly about accuracy. Furthermore, participants who engaged in more analytic thinking and had greater science knowledge were more discerning in their belief and sharing. In Study 2, we found that a simple accuracy reminder at the beginning of the study — i.e., asking people to judge the accuracy of a non-COVID-19-related headline — more than doubled the level of truth discernment in participants’ sharing intentions. In the control, participants were equally like to say they would share false versus true headlines at COVID-19 whereas, in the treatment, sharing of true headlines was significantly higher than false headlines. Our results — which mirror those found previously for political fake news — suggest that nudging people to think about accuracy is a simple way to improve choices about what to share on social media. Accuracy nudges are straightforward for social media platforms to implement on top of the other approaches they are currently employing, and could have an immediate positive impact on stemming the tide of misinformation about the COVID-19 outbreak.
Projects and resources led by MIT Sloan faculty
The COVID-19 Policy Alliance rapidly generates actionable data intelligence and operational recommendations that can help government entities and organizations make better policy decisions regarding the health care system and economy.
The MIT community has been activated to fight the pandemic. From ways to combat feelings of isolation to open-source ventilators — students, faculty, staff, and alumni from academic departments and labs across campus are living the MIT mission to, “…work wisely, creatively, and effectively for the betterment of humankind.” We’ve aggregated the many MIT efforts here for you to explore, learn more, and get activated.
We are a group of researchers from the MIT Operations Research Center, led by Professor Dimitris Bertsimas. We aim to quickly develop and deliver tools for hospitals and policymakers in the U.S. to combat the spread of COVID-19. This work represents a collaborative effort with Hartford HealthCare and ASST Cremona, which have been providing us with data and support through the model creation process.
Despite the threat that infectious diseases pose to global health and security – exemplified by the current COVID-19 pandemic – there are few incentives for biotech and pharmaceutical companies to develop vaccines and other treatments for such diseases due to the high costs of R&D and the uncertain future demand. Furthermore, although policymakers and global actors seek to diminish the danger these diseases pose to the wellbeing of nations, regions, and the world, they are often forced to choose between priorities based on limited information – there is large range of potential biological threats that are unpredictable in nature, and limited resources available to address them. Given these challenges, it is clear that new business models and financing structures are urgently needed. Do you have a working paper that should be included here? Faculty members and PhD students can submit papers for inclusion to Zach Church, editorial director, at firstname.lastname@example.org.