...top of page


May 22, 2020

A roundup of faculty activity in response to the challenges posed by the novel coronavirus pandemic


Victor Aguirregabiria, with Jiaying Gu, Yao Luo and Pedro Mira (Center for Monetary and Financial Studies)

"A Dynamic Structural Model of Virus Diffusion and Network Production: A First Report"

This work presents a dynamic structural model to evaluate economic and public health effects of COVID-19’s spread as well as the impact of factual and counterfactual public policies. An epidemiological model is combined with a structural game of network production and social interactions.

The economy comprises three types of geographic locations: homes, workplaces, and consumption places. The combination of these sets for all the individuals determines the economy's production and social network. Every day, individuals choose to work and consume either outside (with physical interaction with other people) or remotely (from home, without physical interactions). Working (and consuming) outside is more productive and generates stronger complementarities (positive externality). In the presence of a virus, however, working outside facilitates infection and the diffusion of the virus (negative externality). Individuals are forward-looking.

The authors characterize an equilibrium of the dynamic network game and present an algorithm for its computation. We describe the estimation of the parameters of the model combining several sources of data on Covid-19 in Ontario, Canada: epidemiological data, electricity consumption, and cell phone mobility data. The model is used to evaluate the health and economic impact of several public policies: subsidies for working at home, testing policies, and changes in the production and social network. These policies generate substantial differences in the propagation of the virus and its economic impact.


Gustavo Bobonis, with the Government of Puerto Rico’s Department of Education

The University of Toronto and the Puerto Rico Department of Education (PRDE) established a research-practice partnership in 2017 that supports the PRDE’s effort to increase its use of data and research in decision-making. This partnership’s relevance has come into sharp focus as the PRDE strives to make more data-driven decisions while responding to COVID-19.

The government of Puerto Rico is making short- and medium-term plans to facilitate distance education during the COVID-19 outbreak. Among other crucial information, reliable data about student Internet and computer access were not immediately available to design the distance education response.

A large-scale management training and coaching program, begun in 2019, is being implemented with all school principals in Puerto Rico’s public education system. The randomized control trial currently in progress examines whether the management training intervention can have a demonstrable impact on student educational outcomes. Given its recent transition to online, rather than in-person, management training due to COVID-19, the study is being expanded to examine whether principals recently trained using the online modules during COVID-19 are able to lead smoother transitions to virtual learning for their staff and students.


Shari Eli, Angelo Melino, Aloysius Siow and Adonis Yatchew

This group is collaborating on a series of projects on the socioeconomic consequences of pandemics.

The first project is focused on the Covid-19 epidemic in Ontario. Using aggregate daily electricity usage by census divisions, they are studying the effects of news reports on new infection cases and social distancing mandates on the decline in electricity demand. Preliminary results suggest that voluntary social distancing is affected by news reports about new cases, the current number of infections, and social distancing mandates.

That said, social distancing mandates have a first-order effect on electricity usage across Ontario. In regions with low infection rates, decline in electricity usage was primarily driven by social distancing mandates. They are also studying how the past diffusion of the disease, social distancing mandates and electricity usage affect the current and future course of the disease across census divisions in Ontario, the relationship between electricity usage and the weather, and the relationship between electricity usage and economic activity.

→ See also the website "Socioeconomic effects of the Covid-19 pandemic"


Christian Gourieroux, with Joann Jasiak (York)

The issue under consideration is that of partial observability with regard to COVID-19 – that is, those who are infected but asymptomatic and who are thus undetected and unrecorded as a result. This necessarily compromises the estimation of the SIRD numbers (Susceptible, Infected, Recovered, Deceased) and is simultaneously likely to result in the rapid spread of the virus.

While the population could be sampled daily and serological tests performed to estimate the proportion of infected/undetected/recovered individuals, the production of such tests is time-consuming as well as costly in terms of health care provider hours. The alternative proposed here is for a purely model-based approach. Loosely speaking, under the standard SIRD model, the evolution of death rates might be different depending on whether or not all infected individuals are detected. This potential difference allows for a model-based estimation of the proportion of infected undetected individuals, and the consideration of the general case of time varying Markov processes when aggregate counts are partially observed. A Markov process is a random process in which, given the present, the future is independent of the past.


Kory Kroft, with Fabian Lange (McGill), Matthew Notowidigdo (Northwestern), Jessica Gallant (PhD candidate, University of Toronto)

With the sharp downturn in the economy, Kroft along with PhD candidate Jessica Gallant, is examining labour market dynamics and the permanent (versus temporary) destruction of productive matches between employers and employees, and whether the same productive matches can be reconstituted once current “sheltering” restrictions are lifted.

The search-and-matching model in Kroft, Lange, Notowidigdo and Katz (2016) will be used, for its ability to predict shifts in the relationship between job vacancies and unemployment, and changes in the long-term unemployment rate during and after the Great Recession of 2008–09. The model will be adapted to distinguish between temporary layoffs, recalls and permanent separations, allowing the simulation of various policies that make these distinctions. A hierarchy of separations from the labour market is proposed, with a description of how they are to be measured empirically.

The goal is to identify under what circumstances a quick recovery could be expected and the likely path of long-term unemployment, including in comparison to the Great Recession.


Michael Stepner, with Raj Chetty (Harvard), John N. Friedman (Brown) and Nathaniel Hendren (Harvard)

The global pandemic is impacting nations across the world — borders are closing, schools and workplaces are shutting down and large gatherings are a thing of the past. Much of the world's economic activity is coming to a halt, hurting businesses, large and small, and workers alike. Policy-makers require up-to-date information to make decisions in this rapidly-changing crisis. But the typical economic indicators lag weeks behind the current day’s economic activity.

A new real-time tracker of American economic activity, developed by incoming faculty member Michael Stepner with colleagues at Harvard and Brown universities, will help policy-makers, non-profits and philanthropists better understand the dimensions of the COVID-19-induced economic downturn and identify targeted, effective recovery efforts.

The Economic Tracker is an unprecedented collaboration between private sector businesses and academics, opening up the insights that businesses are using daily to the public and gathering them in one place so that everyone can assess the state of the recovery – and the impact of policies.


Brian Rivard, Dhruv Sinha, Adonis Yatchew (coordinator)

COVID-19 Economics: Electricity as a Compass to Recovery

Electricity usage provides an excellent real-time indicator of economic activity. For example, April 2020 electricity demand in Ontario was about 10% below expected levels, at the same time that unemployment rose dramatically.

Ontario electricity demand Jan-Apr 2020
Ontario electricity demand, Jan-Apr 2020 vs. 2019
(Weather normalized)

The central objective of this research is to assess the relationship between changes in electricity consumption, and unemployment and GDP. Real-time tracking of electricity usage can give early indications of the nature of the recovery: will it be shaped like a U, V, W or L, or a Nike “swoosh”? Industry specific analyses are also being conducted as it is likely that some segments will recover more quickly than others.

At a time when the public purse will be stretched to the limit, real-time electricity data can provide policy-makers with more precise tools for targeting those industries that are particularly in need of support, and those regions that are especially hard-hit.

Additional information and updates on this research are posted here. Blog entries by this research team can be found here.

Brian Rivard is Adjunct Professor and Research Director, Ivey Energy Policy and Management Centre at Western University. Dhruv Sinha is a PhD student at the University of Toronto. Adonis Yatchew is Professor of Economics at the University of Toronto and Editor-in-Chief of The Energy Journal.