The experiences in China, Italy, Spain, and all the other countries affected by COVID-19 show that the biggest challenge of the virus is to the healthcare system. The capacity of healthcare systems to take care of people in severe health conditions is limited. The sudden spike in these cases provoked by the virus pushes the system above its capacity. This is why in the last few days we have heard on various occasions how important it is to ‘flatten the curve’ (Baldwin 2020).
Many policies have been proposed to slow down the spread of the virus, and both epidemiologists and economists are sharing their expertise on how those policies might affect the spread of COVID-19 and how they will affect the economy, respectively (e.g. Baldwin and Weder di Mauro 2020). Since many of those policies have never been implemented before, advice on their effects is typically a rough guesstimate, albeit guided by the theoretical frameworks of both epidemiology and economics. Going forward, it will be important to continuously research both the epidemiologic and economic consequences of the different polices put in place, and adjust the policy mix accordingly.
At this stage it is important to think about candidate policies that could achieve both a large reduction in the transmission rate of the virus and at the same time inflict minimal damage on our economic and social lives. Targeted interventions have the potential to achieve that balance. One such policy is the contact tracing and extensive testing approach that many countries have been following over the past couple of weeks.
Where and when this policy has not been successful at slowing down the transmission rate of COVID-19 sufficiently, or capacity constraints have been hit, countries have typically shifted to broad national measures such as closing all nurseries, schools and universities, closing all public spaces, or even a national curfew, as already implemented in Italy and Spain. The idea is simple: if people do not interact with each other and stay at home, the virus cannot spread. Whether this policy will be effective in the longer run is unclear, but trying to suppress the virus seems for the moment the preferred policy intervention (Ferguson et al. 2020).
In economic terms, a main aspect of the policies put in place has been to severely restrain labour mobility. As a result of restricted mobility, the economy is in the process of collapsing. Labour immobility means that consumption is restricted to essential goods. As a consequence, large sectors of the economy have seen their demand plummet. Moreover, labour immobility has also limited the ability of firms to offer many of their services and to produce goods, leading to a fall in supply of all non-essential goods and services.
The few days under broad restrictions to labour mobility highlight how costly such policies are, and just how important it is for people to move. People need to move within cities for consumption, to gather in common workspaces (despite the fact that it is now easier than ever to communicate online), and last but not least to socialise. We need labour immobility to fight the virus, but we also need people to move for the economy not to collapse. What can we do?
A potentially useful approach is to limit human mobility in more targeted ways. In a couple of weeks’ time, if the current policies are moderately successful, there will likely be geographic areas with no known new cases of the virus, and others where this objective will not be achieved. In that scenario it might make sense to allow for human mobility within the former areas, but not the latter.
The work of economists in recent years suggests that such an approach might be particularly useful at the level of ‘commuting zones’, introduced by Tolbert and Sizer (1996) and popularised by Autor et al. (2013). Commuting zones are areas defined based on commuting flows. They are meant to capture geographic areas where there is intense commuting inside the area, but little commuting outside it. Commuting zones are sometimes referred to as local labour markets.
In a recent study, Monte and Rossi-Hansberg (2018) investigate how much commuting there is across commuting zones (CZs). As can be seen in the Table 1, adapted from Table 1 in their paper, we have that in the median CZ around 7% of its residents work outside it. Similarly, around 7% of the workers in the median CZ live outside the CZ. There is quite some heterogeneity in these distributions, but the share of residents that commute outside their residence CZ or the share of workers that commute from outside the CZ even at the 95th percentile is around 22% and 15%, respectively.
Mobility inside commuting zones is not large either. Davis et al. (2019), show consumption patterns in restaurants in New York City using data from Yelp (an online reviewing platform). As can be seen in Figure 1 using two data points, and more systematically in their estimates, Yelp users – who may perhaps be more adventurous than the average New Yorker – seem to visit restaurants that are closely located to their workplace and to their residences. This is obviously just one part of the consumption and socialising that takes part in metropolitan areas, but it suggests that mobility is quite concentrated around work and home locations, and clearly inside commuting zones. This evidence is more systematically documented in Agarwal et al. (2020) using credit card data and documenting consumption patterns across multiple industries.
Hence, commuting zones may offer a way to think about designing more targeted policies to halt the spread of COVID-19. Allowing mobility within commuting zones without COVID-19 cases – possibly very narrowly defined – but limiting it across them may dampen the spread of the virus, while limiting the effects on the economy. Making sure that commuting zones are clean from the virus before allowing mobility within them may be a price to pay to win the fight against the virus. After all, it is also inside commuting zones where the virus is most likely to spread. With time one could then allow mobility across commuting zones that have experienced enough time without any positive cases. Sealing off areas without the virus and allowing for human mobility within them has the added advantage that those areas could provide valuable resources to other areas which are affected by the virus. Trade in goods across commuting zones can help mitigate the problems caused by across-commuting zone labour immobility.
This is a drastic solution. But it would hopefully achieve much of the epidemiological benefits of curfews at a fraction of their economic costs. We could summarise it as moving from ‘red-zoning’ (i.e. the ineffective way in which governments have tried to prevent the virus from spreading by restricting labour movements outside arbitrarily defined geographic areas) to ‘green-zoning’. Among other things, the COVID-19 crisis is showing just how important it is for people to move. Hopefully, limiting this mobility in smart ways – only where it’s necessary – can help cope with the trouble that restricting labour mobility poses to the economy.
Agarwal, S, B Jensen and F Monte (2020), “Consumer Mobility and the Local Structure of Consumption Industries”, CEPR Discussion Paper 12150.
Davis, D, J Dingel, J Monras and E Morales (2019), “How Segregated is Urban Consumption?,” Journal of Political Economy 127(4): 1684-1738.
Autor, D, D Dorn and G Hanson (2013), “The China Syndrome: Local Labor Market Effects of Import Competition in the United States,” American Economic Review 103(6): 2121-2168.
Ferguson, N, D Laydon, G Nedjati-Gilani et al. (2020), “Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand”, March.
Monte, F and E Rossi-Hansberg (2018), “Commuting, Migration and Local Employment Elasticities”, American Economic Review 108(12): 3855-3890
Tolbert, C M and M Sizer (1996), “US Commuting Zones and Labor Market Areas: A 1990 Update”, Economic Research Service Staff Paper 9614.