The COVID-19 pandemic has reminded us of the effectiveness of social distancing practices, voluntary and mandated, to reduce disease transmission in the presence of highly contagious infections. Even though COVID-19 has a relatively low fatality rate, slowing down the rate of transmission is critical to avoid overwhelming the health system. Thus, social distancing prevents deaths because it ensures that the infected have access to hospital beds, medical equipment, doctors, and nurses.
But the same social distancing that is so effective in preventing contagion and deaths is, at least in principle, insidious for economic activity, as social distancing usually implies economic distancing. Most industries require workers to collaborate shoulder to shoulder to produce goods (assembly lines in car factories, food processing, textile industries, etc.), while several services require close contact between clients and providers (restaurants, universities, barbershops, hotels, airlines, etc.) or among clients (sports events, casinos, amusement parks, etc.).
The ubiquitous link between social and economic distancing uncovers a notable void in the economic literature: most models and economic analyses abstract from the role of distance in economic and social interactions. Perhaps the most notable exceptions can be found in the trade literature (gravity models use distance to capture transportation costs), in urban economics (distance affects commuting costs and then the agglomeration and the shape of cities), and in environmental economics (telecommuting saves on contamination and pollution).
Distance, however, is an element that pervades all human and economic interactions more broadly. Yet, it is not explicitly modeled at a microeconomic level to capture macroeconomic implications. How does distance among coworkers affect their productivity? Which services can be provided at a distance and which require close contact? How easy is to move economic interactions to more distant activities? These questions may have sounded moot in normal times, as interacting face to face or at a distance just differs on commuting, infrastructure, environmental, or logistic costs; usually a small portion of total output.
The pandemic however, has uncovered the broad impact of shocks to distance. Indeed, a pandemic is quite a different shock than what countries are accustomed to. It is not a war, which affects the amount of capital and labor available for production (a shock to the amounts of production inputs). It is not a shock to confidence or uncertainty, which reduces demand by self-fulfilling expectations and precautionary motives. It is not a financial crisis, which affects the flow of funding and the allocation of productive resources in the economy. Pandemics constitute a shock to the production function itself, as it forces firms to redesign production practices and to recombine inputs – particularly among workers, their interaction with clients and their reach to demand – in new and creative ways.
Our Social and Economic Distancing Model
In Erol and Ordonez (2020), we propose a novel, disaggregated and flexible, production function that includes both social and economic distancing and allows us to discipline the discussion about the mapping between social and economic distancing. In the model, economic activities are performed by collaborations of pairs of individuals (a firm can be considered a collection of these pairs), who can operate in-person or at-distance. Individual pairs are heterogeneous on the relative payoffs (or costs) of operating at-distance and, as such, sort their activities into (1) no collaborating at all, (2) collaborating in-person, or (3) collaborating at-distance. The choice of operating in-person or at-distance, then, shapes both production in the economy and relations between individuals, both in their role as workers and in their role as clients.
From the eyes of the model, pandemics can be interpreted as a shock to the cost of operating in-person, as it increases the expected cost of getting infected with a virus that has neither treatment nor clear health consequences. In response to this, higher-cost individuals reshape their economic relations, both in terms of ways to operate (in-person or at-distance) and in terms of their productive network (whom to work with as a producer and whom to consume as a client). In other words, when a pandemic hits, an individual operating as a producer needs to work with some coworkers through Zoom and to have meetings with clients by phone, and while operating as a consumer stops going to restaurants or concerts. We show that this intensive margin reaction in the way activities are performed buffers the economic damage of pandemics.
Workers also react at an extensive margin, changing the number and type of people they connect to for economic activities. When a pandemic hits, the same individual above operating as a producer may change the intensity of the relation with certain coworkers and operating as a consumer may change the places from which they demand goods and services (the individual may choose to use fewer movie theaters and more streaming services, for instance).
There is recent empirical literature that shows the relevance of telecommuting and the possibility to operate from home, particularly for high-productivity activities. In a recent paper, for instance, Dingel and Neiman (2020) classify the feasibility of working at home for all occupations in the U.S. and estimated that 37% of jobs could be done at home (lawyers, teachers, finance, etc.), with a large heterogeneity across regions, much larger than the reported telecommuting happening in the U.S. before the pandemic. Saltiel (2020) extended this computation for developing countries, where, on average, 13% of jobs could be done at home in ten countries, composing the so-called STEP survey. Again, in those countries, the feasibility of working from home is positively correlated with high-paying occupations, educational attainment, formal employment status, and household wealth.
We explore an assortative setting (people prefer to connect with other people of similar characteristics) and show that the implied reduction in connections affects particularly those individuals participating in low-productivity activities. These activities tend to be characterized by more in-person relations, and then social distancing practices affect them disproportionately. This implies that pandemics not only have a greater effect on low-income individuals, but also that pandemics tend to increase inequality in the economy. Further, when the economy is technologically better suited to transfer many activities at-distance, inequality gets more pervasive as individuals that rely more on in-person activities are left further behind.
Externalities and Trade-offs
After we obtain these insights from individual reactions to pandemics, we highlight that in our model there are both health and economic externalities. Health externalities because an agent with a dense set of relations can spread the virus more rapidly and widely. Economic externalities because an agent getting infected reduces not only his own production but also that of co-workers. These effects are not fully internalized by individuals when deciding whether to operate in-person or at-distance, or when determining how many people to get in touch to operate. These externalities then, justify government intervention to impose social distancing practices.
Too stringent social distancing reduces economic output and increases income inequality, while too relaxed social distancing increases contagion and death. This obvious trade-off heavily depends on the technology that drives the mapping between social and economic distancing. We show that support for social distancing practices is heterogenous across groups in the economy. Low-income individuals tend to find social distancing mandates too relaxed. The reason is that they usually participate in low-paid jobs, and as such, the relative value of not getting infected is higher than the cost of not working. However, this changes if these individuals are liquidity constrained, as social distancing may result in not having enough resources to eat, in which case they may find social distancing unfeasible to sustain.
On the other hand, high-income individuals tend to find social distancing mandates too tough. The reason is that they usually participate in high paid jobs, and as such, the relative cost of not being able to operate are high compared to the cost of getting infected. This changes if their productivity depends heavily on their counterparts, and it is not very productive to work when other similar individuals are sheltered at home, in which case they also prefer social distancing.
Potential Long-term Benefits of the Pandemic?
While very costly in the short (and possibly medium) run, our model also suggests that the pandemic has potentially positive long-term effects. Even though digital and information technologies have advanced critically in the last two decades, adjustment costs or coordination failures may have prevented those technologies from being used more widely in the world. Adapting to the pandemic, however, is rapidly forcing economic activity to transition from technologies focused on operating in-person to technologies that allow more operations at-distance. In short, the pandemic may have arrived as a coordinating shock that pushes individuals to explore and exploit at-distance activities further. This forced adoption of new telecommuting and distanced technologies can open new avenues for economic growth in the future.
This long-term possibility becomes relevant for the question of who is likely to pay for the social plans and stimulus packages implemented by governments and central banks all over the world to face the pandemic. The previous discussion suggests that the pandemic could potentially be a negative shock that self-finances its costs if it speeds up the adoption of new and possibly superior technologies. Indeed, a recent Gartner, Inc. survey of 317 CFOs and Finance leaders on March 30, 2020, revealed that 74% will move at least 5% of their previously on-site workforce to permanently remote positions post-COVID-19. This survey also highlights that CFOs were more motivated to make the change once many competitors make the change.
Our work highlights the role of social distancing on reshaping economic distancing in the economy, with clear implications on easing the economic impact and increasing the distributional disparities. Our work also gives a message about the discussion of how to “reopen the economy.” It would be a mistake to discuss these policies as if the economic structure post-pandemic will be the same as pre-pandemic. Most likely, firms and individuals will mutate into different ways to operate, consume, and interact, and will change some, if not all, of their economic relations.
Policymakers should weigh subsidizing new technologies and reopening activities where the mapping between social and economic distancing is weak, or where such mapping has shown more flexibility. The economy has demonstrated its creative capability to maintain operations without relying on in-person activities. Restaurants are delivering meals, doctors are making consultations online, counselors are using teletherapy, school teachers are giving lessons online, and academics are sharing research through webinars. Governments should leverage these experiences to boost an economic reinvention and reopen to a new economic reality that may be more productive than before the pandemic.
 Skills Towards Employability and Productivity (STEP). STEP collects information on employment in countries across the world. The ten countries covered by Saltiel: Armenia, Bolivia, Yunnan Province in China, Colombia, Georgia, Ghana, Kenya, Laos, Macedonia and Vietnam.