-A A +A

From: Paul Jenkins and Jeremy M. Kronick

To: Policymakers across the country

Date: October 30, 2020

Re: Dealing with the second wave

The recent resurgence of COVID-19 infections in many parts of the country, with a predictable increase in hospitalization and deaths, represents a renewed challenge in the public health battle.

The good news is that we have everything in our power to reverse this trend through strategic interventions, responsible behavior, transparency, and clear, direct communications.

Epidemiological-economic models can assist us in this challenge with policy insights that can be used to help communicate and implement a winning strategy.

These models can usefully simulate a wide range of possible interventions that can turn infection rates back down to our summertime levels, and perhaps better, while minimizing economic losses. In other words, with the knowledge we have gained to date, we can ask what mix of interventions will give us both better health and economic outcomes.

To gain those insights, we re-calibrate the epidemiological SIR model of Acemoglu et al. for Canada using data from our experiences these past seven months. The key innovation in their model is the ability to treat age groups differently under different intervention strategies. We replicate this for Canada treating three age groups differently – young (20-49), middle-aged (50-64) and older (65+).

Our initial focus on age groups is because of the differences in transmission and mortality rates – a clear characteristic of our experiences to date. This approach differs from the current lockdown practice of the provinces, which targets by region based on already-existing infection hot zones. (In a subsequent Memo, we will look at how targeting policies by industry affects our results, where more of our reliance on trade can be reflected.)

Any such model is a simplified description of a process of cause and effect, with a wide confidence band around its parameters. Recognizing these limitations, our objective is not to be prescriptive, but to help identify how a combination of interventions, undertaken in a coherent, strategic way, can turn the situation around and guide us to both better health and economic outcomes.  

In our base case scenario, policymakers do not differentiate by age group and initiate similar lockdowns for all three age groups, conceptually similar to Canada’s springtime closure of all non-essential businesses. We ask what would be the economic loss from this lockdown strategy, which would bring infection rates back down, and where lives lost over the next year, as a percentage of the population, are no greater than they have been from March to present day.

We then allow for a targeted approach to interventions where we engage in more targeted lockdowns by age groups, and look at ways to improve on non-economic interventions such as better testing and tracing.

In the base case scenario, a similar lockdown for all three age groups results in an economic loss (in terms of annual GDP) of 26 percent.

Allowing for a more targeted approach, the older population is locked down more than the other groups, and the ensuing economic loss is reduced to 15 percent. A significant improvement.

But we can go further with non-economic interventions. Starting with the targeted approach:

  • Improving social distancing, specifically by lowering interactions between the older population and the rest of society by 20 percent, reduces the GDP loss from 15 percent to 13.5 percent;
  • Introducing better testing and tracing, with only a 5 percent improvement, we can reduce the economic loss from 15 percent to 11.75 percent;
  • Lowering the transmission rate by 10 percent through, for example, more adherence to mask wearing, lowers the economic loss from 15 percent to 11.4 percent; and
  • Combining all three measures lowers the economic loss to 9.5 percent, a big improvement over the original 15 percent loss.

We take the following insights away from our model and results:

  • Targeting: Targeted interventions are clearly better than across the board interventions as they can be directed to vulnerable populations and high risk areas (e.g., the elderly in long-term care homes) and encourage broad-based preventative behavior (e.g., social distancing, testing and tracing, and mask wearing).
  • Intervention improvements: The capacity to provide an appropriate and timely mix of interventions is the key to better health and economic outcomes. Leveraging data, better use of the tools we have available (e.g., tracing Apps) and a focus on high risk areas improves outcomes considerably.
  • Vaccine availability: Given the uncertainty about when a vaccine will be widely available, policymakers need to stay the course with timely, targeted intervention.

We are gaining the knowledge and developing the tools to knock this virus down. What we do not yet have, and what Canadians want to hear, is a clear, consistent strategy from all of our leaders. Formal modeling, while only one tool, provides a framework to inform coherent messaging and a narrative that Canadians can understand and trust.

Paul Jenkins is a former senior deputy governor of the Bank of Canada and a senior fellow at the C.D. Howe Institute, where Jeremy M. Kronick is associate director, research.

To send a comment or leave feedback, email us at blog@cdhowe.org.

The views expressed here are those of the authors. The C.D. Howe Institute does not take corporate positions on policy matters.