Social networks and consumption patterns during a pandemic

Learning from friends in a pandemic: Internet sites and the macroeconomic response of consumption

Christos Makridis, Tao Wang

Aggregate consumption has seen an unprecedented drop because of COVID-19 and the resulting lockdowns. Instead of focus on techniques specific policies or the spread of the virus have altered consumption patterns through the pandemic, this column explores the impact of internet sites. Combining card transaction data with indices of social connectedness from Facebook demonstrates a 10% increase of infections in socially connected counties is connected with a 0.64% decline in local consumption expenditures, suggesting that internet sites can sharply amplify the consequences of an underlying shock.

The ongoing COVID-19 pandemic represents the biggest world-wide macroeconomic shock in at least a hundred years (Baldwin and Weder di Mauro 2020a, 2020b), resulting in substantial declines in employment (Bartik et al. 2020, Coibion et al. 2020a), consumption (Baker et al. 2020, Coibion et al. 2020b), and output (Guerrieri et al. 2020, Makridis and Hartley 2020). For example, retail sales and food service in america decreased in April by 16.4% from the prior month, and 21.6% from the same month this past year, according to a recently available release by the Census Bureau.

Social networks are actually a primary vehicle for obtaining information in the common household (Westerman et al. 2014). Individuals may adjust their consumption in response to information communicated through friends in connected regions, even if their own county has fairly low contact with the virus. Quantifying how individuals make consumption and savings decisions in response to shocks not merely with their own fundamentals but also to those of their connected friends is very important to understanding the resources of aggregate fluctuations, particularly during episodes of uncertainty and panic.

There is currently clear evidence that the pandemic has shifted household expectations about real economic activity. For instance, Coibion et al. (2020b) show that households in counties that experienced lockdown earlier anticipate higher future unemployment, lower future inflation, higher uncertainty, and lower mortgage rates over another twenty years. Similarly, Binder (2020) demonstrates greater concern about COVID-19 is connected with higher inflation expectations and more pessimistic unemployment expectations. Gleam rich literature on the role of personal experience in belief formation.

But how are individuals’ expectations and consumption decisions influenced by their internet sites? If my personal contact with the pandemic is bound, am I more pessimistic about the economy and more worried about the infection risk if others in my social networking are themselves adversely affected? And, how does that affect my consumption activity? Both are likely linked. For instance, Bailey et al. (2018) find that folks with geographically distant friends who experienced larger housing price increases were much more likely to transition from renting to owning, and paying more for confirmed house.

In Makridis and Wang (2020), we apply these suggestions to a new setting where we exploit plausibly exogenous variation in the exposure of counties to other counties (and countries) that vary within their COVID-19 experience: some areas were affected more than others. Using a mix of Facteus’s card-level transaction data and Facebook’s Social Connectedness Index (SCI), we find that being friends with heavily infected regions reduced consumption; i.e. a 10% upsurge in SCI-weighted cases and deaths is connected with 0.64% and 0.33% declines in consumption, respectively. To place this into perspective, each county block in the map below plots the SCI-weighted cases of infection on a specific date, and shows substantial variations of the social-media contact with COVID-19 experienced by different counties.

Figure 1 Number of instances per thousand on Facebook ()

To handle concerns about time-varying omitted unobservables or potential selection effects in counties which have more friendship ties with others, we condition on local infections and deaths, in addition to county and time fixed effects. Moreover, we also exclude connections to counties in the same state predicated on the prospect of mobility and/or correlation in the response of consumption to common state-level policies. Our email address details are also robust to state-by-time fixed effects, which directly control for potential state-level policies that influence both consumption and other county infection rates.

Not absolutely all consumption responds equally to peer effects. For instance, contact-based consumption goods and services could respond more elastically to changes in the infection rates among connected counties, because increases in the perception of risk linked to the pandemic will probably prompt individuals to self-isolate. We group each one of the 982 merchant classification codes (MCCs) into 17 broad categories predicated on its degree of contact with infection risks, together with its demand elasticity.

Using these new categories for consumption, we estimate separate elasticities in response to SCI-weighted infections by category. In keeping with our theory about peer effects, we find these declines are greater among social-contact-based consumption categories and activities abroad. For example, each 10% upsurge in socially-connected cases is connected with a 2% reduction in clothing/footwear/cosmetics, a 1.3% reduction in contract-based service, and a 1.1% reduction in travel. They are twice to 3 x as large as the drop in average spending. Moreover, our email address details are in keeping with Coibion et al. (2020), who find that the declines are largest in travel and clothing.

Figure 2 Elasticities by category

We also estimate heterogeneous treatment effects across various dimensions of county characteristics. For instance, our elasticities are sustained in low income counties, counties with higher shares of people beneath the age of 35, less populous countries, counties with an increase of digitally-intensive employees (Gallipoli and Makridis 2020), and more teleworking employees (Dingel and Neiman 2020).

We find an economically and statistically significant decline in consumption connected with increases in COVID-19 infections in socially connected counties, even after controlling for time invariant characteristics across space and time, together with time-varying shocks to local health outcomes (e.g. infections and deaths). However, one concern is these results are suffering from other time-varying omitted variables that jointly affect connected counties and local consumption outcomes.

A great way that people test for selection effects is by firmly taking these insights to a global setting. Specifically, we restrict our sample to 15 February – 15 March, prior to the pandemic became the centre of attention in america. We concentrate on county contact with four countries-South Korea, Italy, Spain, and France-although our results hold for a broader group of countries exposed in early stages. This enables us to purge variation which may be cor with time-varying shocks in america.

We exploit variation along two dimensions. First, counties vary cross-sectionally within their contact with these countries. For instance, whereas Maricopa County in Arizona comes with an SCI of 142,771 with France, SAN FRANCISCO BAY AREA comes with an SCI of 258,825. Second, countries vary within their intensity of COVID-19 shocks. We find a 10% rise in infections (deaths) in Italy for counties that are more closely linked to Italy is connected with a 0.07% (0.52%) decline in consumption. We see broadly similar treatment effects for every country, although they are smaller for France, perhaps as the US had already witnessed the knowledge of Asian countries, such as for example South Korea, Spain, and Italy earlier in March. Moreover, the actual fact that deaths generate a more substantial influence on consumption is in keeping with the saliency of deaths in the first amount of the pandemic, when it had been not as large important for many People in america.

As the emerging empirical literature on the pandemic has centered on the direct ramifications of specific policies and/or the spread of the virus, this paper targets the role that internet sites play in propagating the consequences on consumption. Using real-time data on consumption expenditures predicated on the transactions of 5.18 million debit card users, in conjunction with data on social connectivity across geographies from Facebook, we quantify the response of consumption to changes in a county’s COVID-19 exposure predicated on its internet sites. Our results claim that these effects from internet sites are significantly bigger than the direct ramifications of the virus on consumption.

Bailey, M, R Cao, T Kuchler and J Stroebel (2018a), “The economic ramifications of internet sites: Evidence from the housing marketplace”, Journal of Political Economy, 126(6): 2224-2276.

Baldwin, R and B Weder di Mauro (2020), “Economics in enough time of COVID-19”, London: CEPR Press, 6 March.

Bartik, A W, M Betrand, F Lin, J Rothstein and M Unrath (2020), “Labor market impacts of COVID-19 on hourly workers in small- and medium- size businesses: Four facts from HomeBase data”, Chicago Booth Rustandy Center, Working Paper.

Baker, S R, R A Farrokhnia, S Meyer, M Pagel and C Yannelis (2020b), “How Does Household Spending React to an Epidemic? Consumption Through the 2020 COVID-19 Pandemic”, Working Paper.

Binder, C (2020), “Coronavirus Fears and Macroeconomic Expectations”, Overview of Economics and Statistics, forthcoming.

Coibion, O, Y Gorodnichenko and M Weber (2020), “Labour markets through the Covid-19 crisis: An initial view”,, 14 April.

Dingel, J I and B Neiman (2020), “Just how many jobs can be achieved in the home?”,, 07 April.

Gallipoli, G and C Makridis (2018), “Structural Transformation and the Rise of IT”, Journal of Monetary Economics, 97: 91-110.

Guerrieri, V, G Lorenzoni, L Straub and I Werning (2020), “Viral recessions: Insufficient demand through the coronavirus crisis”,, 06 May.

Makridis, C A and J Hartley (2020), “The expense of COVID-19: A rough estimate of the 2020 GDP impact. Mercatus Center”, Policy Brief Special Edition.

Makridis, C and T Wang (2020), “Learning from Friends in a Pandemic: INTERNET SITES and the Macroeconomic Response of Consumption”, Offered by SSRN 3601500.

Westerman, D, P R Spence and B Van Der Heide (2014), “Social media as information source: Recency of updates and credibility of information”, Journal of Computer-Mediated Communication, 19(2): 171-183.