Main Street Resiliency

Case Study Selection

For our detailed resiliency analysis, we selected 20 main streets in the Toronto, Montreal, and Edmonton regions. These main streets were chosen to provide a representative sample of the types of streets in each regional context, and included four downtown (workplace-oriented), four small town, and 12 neighbourhood (residential-oriented) main streets. (More main streets were included in the neighbourhood category as there is a greater variety of main streets within this group.) In addition, we intentionally included main streets with equity-deserving communities.

Resiliency Score (I)

The primary dependent variable we used to assess resiliency was based on the number of visitors to the main street on a monthly basis. The data is provided by Environics Analytics and is ultimately derived from cell phone GPS signals. We count the number of people (devices) entering buildings on a main street within certain time intervals. When a regular nighttime location (assumed to be someone’s home location) can be associated with a device, additional variables are attached to the visitor data. We derive how far visitors have travelled from home, and link demographic data (based on postal codes) to get a sense of the characteristics of people visiting each main street.

We created a single resiliency score for each main street by comparing monthly visitor levels from 2020–2022 to their equivalent months in 2019 (pre-pandemic). For each month, we calculate the number of visitors as a percentage of the number of visitors in 2019. The total amount that a main street was below its 2019 visitor levels for each month from 2020–2022 is the resiliency score. For example, if a main street experienced a sharp drop-off in visitors in 2020 and was slow to recover through to 2022, it had a relatively low resiliency score. Conversely, if a main street experienced a modest drop-off in visitors and experienced a relatively quick recovery, it had a higher resiliency score.

Resiliency Score (II)

There was a clear geography to main street resiliency through the pandemic, with downtown main streets typically suffering the sharpest drop-offs in visitor levels and experiencing the slowest recoveries.

• North York Centre in Toronto was the most resilient downtown street among our case studies. This is likely due to its function as a secondary downtown to Toronto’s central core, with a significant residential population.
• The least resilient downtown main street was 97th Street in Edmonton, which saw visitor levels drop to 15% of pre-pandemic levels in October 2021, and rising only to 33% by the end of 2022.
• Small town main streets tended to be the most resilient, although there was an extensive range between the top (Sutton in the Toronto region) and the bottom (St. Eustache in the Montreal region).
• Neighbourhood main street resiliency was in between that of downtowns and small towns (closer to the latter). Rue Beaubien in Montreal was the most resilient neighbourhood main street; Central Brampton (Main and Queen) was the least resilient.

Visitor Types

Downtown main streets were the least resilient through the pandemic, largely because people did not have to go there. The shift to working from home for those who typically worked in office buildings had, and continues to have, one of the most significant impacts on cities. Cell phone data can identify regular nighttime (home) and regular daytime (typically, work) locations, and can provide insights into the reasons people visit particular main streets. We segment visitors into three categories:

• local residents (live within 1 km of the street)
• frequent visitors (visit during the day on a regular basis)
• infrequent visitors (workers, neither local residents or frequent visitors).

There is a clear pattern: infrequent visitors (workers) outnumbered local residents on downtown main streets prior to the pandemic. The opposite tends to be true for small towns and neighbourhood main streets that relied more heavily on residential visitors.

Civic Infrastructure Provision

A core purpose of the initial Measuring Main Street project was to assess the role of civic infrastructure on main street resiliency. We created a civic infrastructure provision score for every main street in the country: the Civic Infrastructure Index. This metric combines the presence of civic infrastructure locations relative to the population density of an area. The civic infrastructure locations are weighted by how many people they employ, to account for scale of the facility. Downtown main streets tend to have the highest levels of civic infrastructure provisions, largely because they house facilities that offer central functions (main libraries, major performance theatres, city halls) rather than strictly cater to a local population. In the context of COVID-19, these facilities were not sufficient to enable downtown main streets to withstand the pandemic’s impacts on visitor levels. The connection between civic infrastructure and main street resiliency proved to be more nuanced.

Correlation Analysis

To uncover the factors that help explain main street resiliency throughout the pandemic, we performed a correlation analysis with a wide range of indicators. The factors included variables of urban form, demographics, and street content.

The analysis revealed two key indicators that had the closest associations with main street resiliency. First was the share of visitors that are local residents. During the pandemic, people tended to stay closer to home; they were less likely to travel to work or other locations further away. Main streets that relied more heavily on visitors from further away were more likely to experience sharper drop-offs and slower recoveries. Second was the presence of green spaces. While civic infrastructure generally did not predict higher resiliency, parks did. This makes intuitive sense, as people still had the core desire to congregate, but wanted to do so in a safe outdoor setting. This suggests that civic infrastructure is indeed important to the vitality of main streets and that data analysis in more “normal” times may reveal this to be the case.