The long-term impact of the COVID-19 Recession on homelessness in Canada

by admin | Dec 10, 2020 | Cities, Demographics, Determinants of Homelessness, Employment, Employment Insurance, Ending Homelessness, Fiscal policy, Government, Government Funding, Homeless, Housing, Housing First, Income Support, Labour Market, Macroeconomics, municipalities, Pitching new programs to government, Point in time Count, Policy, Poverty, Poverty Reduction, Research and Data, Social Assistance, Social Policy, Unemployment

The long-term impact of the COVID-19 Recession on homelessness in Canada

La version française de ce billet se trouve ici.

I’ve written a report for Employment and Social Development Canada (ESDC) that assesses the likely long-term impact of the current recession on homelessness. The link to the report is here.

Here are 10 things to know:

1. The current recession may contribute to rising homelessness across Canada, but that matter is complicated by several factors. Those factors include: a lag effect of up to five years from the time a recession starts until its impact fully plays out; the many unknowns that lie ahead (e.g., whether there will be future waves of the pandemic, when and if a vaccine is developed, what types of new social benefits are announced, etc.); and differences from one community to another (with respect to both the labour market and housing market, for example).

2. A recession’s lag effect stems in part from a strong desire of households to avoid absolute homelessness. When faced with reduced income or outright job loss, a household may try to arrange a rental arrears plan with their landlord; they may also borrow money from family and friends. They may try to move into cheaper housing as well, or move in with family or friends. The lag effect also stems from Canada’s elaborate social welfare system. For example, Employment Insurance (and more recently the Canada Emergency Response Benefit) can cushion the blow from job loss and help households hang on to their housing. Social assistance, while not as generous, can also delay homelessness onset.

3. This lag effect means there is time for senior orders of government to plan homelessness prevention initiatives. Since it could be a few years before we see rising homelessness in some communities as a result of the current recession, there is time for preventive measures to be designed, implemented and to take effect. Those measures could target households that are either at serious risk of becoming homeless or that have just become homeless.[1]

4. The recession’s impact on homelessness will vary from one community to another. Housing markets, income assistance systems and homelessness system planning frameworks vary across Canada. What is more, migration patterns over the next several years will be hard to predict. As a result, it is challenging to say which Canadian communities will see rising homelessness at what junctures in time. We do know that, thus far, the following types of workers in Canada have been most directly affected by the COVID-19 Recession: young people, women, nonmarried persons, and persons without high school accreditation.

5. In order to monitor the many complex factors involved here, policy-makers needs to track various indicators. The report recommends that ESDC track the following indicators as the recession unfolds: the official unemployment rate; the percentage of Canadians falling below the Market Basket Measure (and especially those falling below 75% of the Market Basket Measure);[2] social assistance benefit levels; median rent levels; the rental vacancy rate; the percentage of households with extreme shelter cost burdens; evictions; and average nightly occupancy in emergency shelters.

6. This tracking will require some nuance. As much as possible, such tracking should emphasize both how these indicators have changed since the start of the pandemic, and how this change varies across both geographical areas and specific populations (e.g., women, youth, Indigenous peoples, etc.).

7. The report recommends that the federal government enhance the Canada Housing Benefit (CHB). This benefit provides financial assistance to help low-income households afford rent. It is expected that half of this money will come from the federal government, and the other half from provinces and territories. The CHB was supposed to launch nationally on 1 April 2020; however, just five provinces have formally agreed to terms regarding the CHB. The federal government could increase the value of this benefit, which could encourage other provinces and territories to sign on. For example, the federal government might offer 2/3 or 3/4 cost-sharing.

8. The report also recommends that the federal government take a soft approach to recovering CERB overpayments from social assistance recipients. This is important in light of the considerable confusion that existed as the CERB was being rolled out. Such an approach might include not trying to fully recover the value of the CERB from these individuals (via the tax system). Even complete amnesty should be considered in some cases.

9. The report recommends that ESDC introduce a new funding stream for Reaching Home (i.e., the federal government’s main funding vehicle for homelessness). The report discusses the successful implementation of prevention efforts in the United States following the 2008-2009 Recession, and encourages ESDC to introduce something similar for Canada. A new prevention stream could focus on time-limited financial assistance directed at households who are either still housed (but at risk of becoming homeless), are in the process of losing their housing, or who have just begun to experience absolute homelessness. Targeting can evolve over time, in light of changes seen in the aforementioned indicators (e.g., the official unemployment rate, the percentage of persons with incomes below the Market Basket Measure, etc.).

10. The report identifies policy changes that could be made by provincial and territorial governments. These include increases to social assistance benefit levels, the reinstatement of social assistance eligibility for recipients who became ineligible due to the CERB, and the encouragement of housing-focused practices at emergency shelters.

In sum. Since we know there is serious risk for more homelessness in Canada as a result of the current recession, senior orders of government need to limit the damage. Well-designed prevention efforts can be more cost-effective than emergency responses after the fact.

I wish to thank Susan Falvo and Vincent St-Martin for assistance with this blog post.

[1] It is also very important to continue addressing existing homelessness. I’ve written about that here.

[2] For more on the Market Basket Measure, see this blog post.

Lifting singles out of poverty in Canada

by admin | Dec 2, 2020 | Canada, Demographics, Determinants of Homelessness, Employment, Ending Homelessness, Fiscal policy, Government, Government Funding, Health Care, Homeless, Housing, Income, Income Distribution, Income Inequality, Income Support, Labour Market, Policy, Poverty, Poverty Reduction, Social Assistance, Social Policy, Unemployment

Lifting singles out of poverty in Canada

I’ve written a report for the Montreal-based Institute for Research on Public Policy making the case for higher social assistance benefit levels for employable single adults without dependants. The link to the report is here.

Here are 10 things to know.

1. In Canada, most employable adult singles without dependants who receive social assistance get less than $10,000/yr. in benefits. This amount of money is ridiculously low (keeping in mind that this figure includes all forms of tax credits received by the recipient). A person with this income must use it to pay for housing, food, transportation and other basic necessities (to see benefit levels in every province and territory, check out Welfare in Canada).

2. In relation to Canada’s official poverty line, social assistance benefit levels for this household group are dismal. ‘Welfare income’—which includes social assistance benefit levels, child benefits and all forms of tax credits—brings couples with two children to between 75% and 95% of the federally-defined poverty line, depending on the province (see figure 1 below). However, welfare income for employable singles without dependants typically comes to about 50% of the poverty line for this particular household type.

3. In most provinces and territories, $10,000 is less than half of what a minimum wage earner would earn in one year working full-time hours. Historically, policy-makers and economists have often been nervous about setting social assistance benefit levels high enough to make paid work unattractive. However, that shouldn’t be a major concern right now in most parts of Canada, as the differential between welfare incomes and minimum wage rates is currently quite substantial.

4. Increases to social assistance benefit levels could help Canada’s federal government achieve its poverty reduction targets. In Canada, we say a household is in ‘deep income poverty’ if it makes less than 75% of the official poverty line. Canada’s Poverty Reduction Strategy, unveiled in October 2018, seeks to track progress on this indicator. Increases in social assistance benefit levels would be a very easy way for progress to be made in this respect.

5. Doing so could also help provincial and territorial governments achieve their poverty reduction targets. All provinces and territories now have their own poverty reduction strategies; many of these strategies include targets pertaining to reducing the number of people under the poverty line (New Brunswick’s strategy actually seeks to reduce deep income poverty by 50%). Increasing social assistance benefit levels would help all provinces and territories achieve their targets.

6. More than half of people in Canada who are in ‘deep income poverty’ are singles. Not only do singles receive very low social assistance benefit levels relative to other household types, but they also do not realize many of the economies of scale that come with cohabitating (e.g., shared rent, shared utility costs, etc.). This reality makes this household group all the more worthy of policy attention.

7. Higher social assistance benefit levels can result in less homelessness. It’s intuitive for many of us that higher social assistance benefit levels would both reduce the likelihood of a person losing their housing and also increase the likelihood of a person experiencing homelessness to obtain rental housing on the private market. Research by Ron Kneebone and Margarita Wilkins confirms this, estimating that a $1,500/yr. increase in social assistance benefits for an employable single without dependants would (in 2011) reduce the use of shelter beds on any given night by nearly 20%.

8. Higher benefit levels can improve food security. A recent study in British Columbia confirms this, finding that overall rates of food security improved among social assistance recipients after a one-time increase in social assistance benefit levels in that province.

9. Less homelessness and improved food security would almost certainly result in public cost savings. The costs of homelessness to the taxpayer are well documented, as are the healthcare costs associated with food insecurity. Put differently, increasing public expenditure on social assistance would likely result in public savings elsewhere.

10. While higher benefit levels would likely lead to more takeup, this increased takeup would be modest. That is precisely the finding of a recent Canadian study that I co-authored with Ali Jadidzadeh. We found that a 10% increase in the real value of social assistance benefit levels for this same household group would likely result in an increase in caseloads of less than 5%.

In sum. When it comes to social assistance across Canada, employable single adults without dependants are a very neglected subgroup. Increasing their benefit levels would likely result in less poverty, improved food security and less homelessness.

I wish to thank Susan Falvo, Lynn McIntyre, Vincent St-Martin and Val Tarasuk for assistance with this blog post.

The use of homeless shelters by Indigenous peoples in Canada

by admin | Oct 3, 2019 | Aboriginal Peoples, Alberta, Calgary, Cities, Data Sharing, Demographics, Determinants of Homelessness, Ending Homelessness, Government, Harm Reduction, Homeless, Housing, Indigenous peoples, municipalities, Research and Data, seniors, Toronto

The use of homeless shelters by Indigenous peoples in Canada

The Canadian Press recently gained access to results of analysis of the use of homeless shelters across Canada by Indigenous peoples. The results are summarized in a March 2019 slide presentation obtained by Jordan Press through an Access to Information and Privacy (ATIP) request, and are discussed in this Canadian Press article. They are based on a research project conducted by Employment and Social Development Canada (ESDC).

Here are 10 things to know:

The analysis draws on data gathered from homeless shelters across Canada. ESDC has data on persons using homeless shelters from roughly half of the country’s homeless shelters. This includes data gathered via the Homeless Individuals and Families Information System (HIFIS) software, as well as data gathered via data sharing agreements with the City of Toronto, the Government of Alberta and BC Housing. The data used for this analysis was gathered in 2016 and is based on approximately 133,000 unique individuals (approximately 41,000 of whom are Indigenous).
According to the slide presentation, Indigenous peoples in Canada are more than 11 times more likely to use a homeless shelter than non-Indigenous people. Results of the analysis are succinctly summarized in a memo prepared for the Minister which accompanies the slide presentation: “Results show that Indigenous peoples are consistently overrepresented in homeless shelters in all 46 communities examined [Indigenous peoples accounted for more than 30% of people using homeless shelters over the course of 2016, while representing less than 5% of the total population]…The degree of overrepresentation is particularly high for Indigenous women, seniors, and Inuit. Indigenous shelter users experience more shelter stays each year, and are less likely to exit a shelter because of finding a residence.”
As noted in the accompanying memo: “A report is being written to expand on the results shown in the deck.” Neither the memo nor the slide presentation indicate when the report will be ready, which people and groups will be invited to provide input on drafts of the report, or whether the report will be made public. Put differently, the federal government seems to be holding its cards close to its chest on this.
I was personally surprised to see the extent to which Indigenous peoples experience high episodic shelter use. In other words, Indigenous peoples (compared with non-Indigenous peoples) tend to cycle in and out of shelters with high frequency, rather than stay for long periods of time.[1]
I find it remarkable that these very important research findings had to be obtained via an ATIP request. In light of the federal government’s stated commitment to reconciliation, I would have thought it would have proactively released these findings and engaged Indigenous stakeholders in the process. This raises the following questions: To what extent were Indigenous peoples part of this analysis (beyond the fact that their data was collected and analyzed)? To what extent will Indigenous peoples be invited to discuss the implications of the analysis? Why does the slide presentation make no mention of the Canadian Housing and Renewal Association’s Indigenous Caucus?
The findings pertaining to shorter shelter stays raise at least two questions. First, why do Indigenous peoples have short, but frequent, stays in homeless shelters? Second, what could be done to address this? Frequent migration between urban centres and First Nations communities may help explain this. This may also speak to the need for more low-barrier and culturally-appropriate housing options in Canada—both emergency and permanent options.
The findings showing Inuit over-representation are disturbing, though not surprising. Canada’s Inuit experience high rates of unemployment, poverty, and housing need. For more on this, see: the Inuit Statistical Profile 2018, published by Inuit Tapiriit Kanatami; this Statistics Canada report from 2018; and this profile of Inuit living in Ottawa.
The findings pertaining to women and seniors merit reflection. The analysis finds that, while Indigenous men are more than 10 times more likely to use a homeless shelter over the course of a year than non-Indigenous men, Indigenous women are more than 15 times more likely to use a homeless shelter than non-Indigenous women over the course of a year. Meanwhile, Indigenous seniors are more than 16 times more likely to use a homeless shelter over the course of a year than non-Indigenous seniors. Why would rates of Indigenous over-representation in Canada’s homeless shelters be higher for women and seniors than for other categories?
I was disappointed to see Canada’s largest city dropped from the analysis. According to the slide presentation, City of Toronto data could not be included in the analysis because the City of Toronto lacked Indigenous status data for more than 40% of unique individuals using its homeless shelters. I hope it serves as wake-up call to the Toronto’s homeless-serving sector.
Among the communities studied, Calgary’s rate of Indigenous over-representation was especially high (see screenshot below). The analysis found that, in Calgary, shelter users are about 16 times more likely to be Indigenous than are members of the city’s total population. The rate for Edmonton appears to be about half of that, with only York Region (near Toronto) having a higher rate among the cities studied. It is worth noting that the Calgary Homeless Foundation recently commissioned a research project assessing flow between Treaty 7 First Nations and Calgary’s Homeless-Serving System of Care. Hopefully that project can shed light on this matter. (Full disclosure: I’m one of the research consultants working on this project, along with Gabrielle Lindstrom, Steve Pomeroy, and Jodi Bruhn.)

In sum. Many years of work go into this kind of analysis. Thousands of people were involved in the data collection, including shelter staff and officials both inside and outside of government. Further, people using homeless shelters patiently answered questions about their lives in order for this analysis to happen. However, it is clear that much work remains. It would appear that ESDC up their game in terms of working with Indigenous peoples and groups on such research projects, and some cities clearly must do a better job of data collection. We must all work collectively to understand what specific policy measures are required to address the over-representation of Indigenous peoples in Canada’s homeless shelters.

The following individuals provided me with assistance in preparing this blog post: Jodi Bruhn, Kathy Christiansen, Damian Collins, Dan Dutton, Ron Kneebone, Diana Krecsy, Eric Latimer, Katelyn Lucas, Jenny Morrow, Jennifer Robson, Vincent St-Martin, and two anonymous reviewers. Any errors are mine.

[1] Federal definitions of chronic vs. episodic homelessness are available here.

Ten Things to Know About Homelessness in Canada

by admin | Sep 15, 2015 | Aboriginal Peoples, Alberta, Canada, Cities, Demographics, Employment, Employment Insurance, Fiscal policy, Homeless, Housing, Income, Income Support, Indigenous peoples, Labour Market, Macroeconomics, municipalities, Nunavut, Ontario, population aging, Poverty, seniors, Social Policy, taxation, Toronto, Unemployment

Ten Things to Know About Homelessness in Canada

This afternoon I gave a presentation at Raising the Roof’s Child & Family Homelessness Stakeholder Summit in Toronto. My slide deck can be downloaded here. To accompany the presentation, I’ve prepared the following list of Ten Things to Know About Homelessness in Canada.

1.Efforts to enumerate persons experiencing homeless have generally been spotty, but it is reasonable to assert that homelessness in Canada saw substantial growth in the 1980s and 1990s. On a nightly basis in Toronto, there were about 1,000 persons per night staying in emergency shelters in 1980. By 1990, that figure had doubled. And ten years later, there were 4,000 persons per night staying in Toronto’s emergency shelters. The Toronto figure of 4,000 per night has remained relatively constant for the past 15 years, though it has edged up in the aftermath of the 2008-2009 recession a phenomenon which I’ve previously written about here. (Admittedly, the number of persons living in emergency shelters on a nightly basis is a rather narrow gauge of homelessness. According to Canada Mortgage and Housing Corporation, approximately 13% of Canadian households are in core housing need; for Nunavut, the figure is a whopping 39%.)

2. Though it’s difficult to establish causation, I think relatively safe assumptions can be made about some of the major contributors to homelessness. Researchers are generally careful about using the term causation in fact, there are long-standing tensions among academic disciplines as to what methodological approaches are required to establish it. Statisticians, for example, generally believe that randomized controlled trials (RCTs) are needed to establish causation; but as David Freedman has argued, RCTs are often impractical or unethical (Freedman, 1999, p. 255). Rather, careful researchers are more likely to say things like these factors have likely contributed to this effect,” or “I think it’s likely that this effect caused this to happen And with that in mind, I’d like to suggest that there are probably three major factors that have contributed to homelessness in Canada: 1) macroeconomic factors (especially unemployment); 2) changes to our social welfare system (including a decrease in the availability of government-subsidized housing); and 3) the design and administration of policies whose specific intent is to respond directly to homelessness (often referred to as ‘systems responses’ to homelessness).

3. Homelessness has profound ramifications on the lives of children. As I wrote in 2012: Two studies have been done in Toronto looking at the role of housing with respect to children in care. Results of both studies indicate that the state of the family housing was a factor in one in five cases in which a child was temporarily admitted into care. Results from the Toronto research also indicate that, in one in 10 cases, housing status delayed the return home of a child from care (Falvo, 2012, p. 14). Other research estimates that, on an annual basis in Toronto alone, approximately 300 babies are born to mothers who are homeless. (Of course, homelessness can have profound ramifications on the lives of adults as well. For more on this, see this 2007 study.)

4. The role of Canada’s federal government in funding both housing for low-income persons and programming for homeless persons has varied considerably over time. Provinces and territories spend much more of their own money on housing for low-income persons when the federal government leads. Thus, a considerable amount of subsidized housing for low-income Canadians was built from the mid-1960s through to the early 1990s. Since the early 1990s, comparatively little subsidized housing has been built for low-income persons in Canada. I should also note that the annual, inflation-adjusted value of federal funding for homelessness today is worth just 35% of what it was worth in 1999.

5. Not every province/territory responds to homelessness in the same way. While much mores subsidized housing for low-income persons gets built when the federal government leads, provinces and territories don’t always respond to federal funding initiatives in the same way. For example, between 2002 and 2013, three times as many subsidized housing units were built in Alberta (on a per capita basis) than in Ontario. I would argue that a driving force behind this differential stems from Alberta’s strong economic performance during this same period relative to that of Ontario’s.

6. Though a careful researcher will be cautious in discussing what causes homelessness, I think we know a lot about what solves it. In many cases, a person who stays in an emergency shelter will exit homelessness without substantial public resources. In some cases, they might find housing on their own; in other cases, family and friends may provide them with short term assistance e.g. some financial support, a couch to sleep on, etc. (To learn more about lengths of stay in homeless shelters in a sample of Canadian cities, see this 2013 study.) Researchers and advocates for the homeless generally don’t view such short-term stays as a major public policy challenge the bigger challenge is in the case of persons who stay in emergency shelters (and outside) for longer periods of time. Even here though, I would argue that it’s hardly a mystery as to what constitutes an effective policy response.

Indeed, as early as the mid-1980s, small non-profit organizations in Ontario (and possibly in other provinces as well) found success in building subsidized housing for persons who had experienced long-term homelessness they did so by providing professional staff support to help such tenants live independently in those units. This was (and still is) known as supportive housing. The emergence of supportive housing in Ontario happened in large part due to strong advocacy by community-based groups. This included: the Singles Displaced Persons Project; the consumer/survivor movement; the slogan homes not hostels the founding of Houselink Community Homes; and the founding of Homes First Society. Conditions of eligibility for such housing varied from one provider to the next. In many cases, the tenant did not have to prove housing readiness before being offered a unit. In fact, Homes First Society got its name because its founders believed that its tenants needed homes first before addressing other challenges (i.e. mental health, substance use, employment, etc.).

Today, researchers, practitioners and advocates refer to this approach ashousing first. And very recently, a successful RCT of housing first was conducted in five Canadian cities; I’ve previously written about that study here.

7. There are several ways of making housing available to low-income households; all of them involve the private sector to varying degrees. Sometimes when government subsidizes housing for low-income persons, it provides money to a non-profit entity that develops, owns and operates the units. Other times, government provides a subsidy to landlords (either for-profit or non-profit); in exchange for the subsidy, the landlord agree to rent units at a reduced rate for a specified period of time (e.g. in some cases, for 10 years). And other times, government provides money (often known as a housing allowance) to low-income tenants who then rent a unit from a for-profit landlord. Of the three possible approaches, I personally have a preference for the option where a non-profit entity develops, owns and operates the units (and I have previously written about this here). Having said that, I think there’s a place for all three approaches, depending on local context.

8. Some jurisdictions have used sophisticated information management systems as part of their efforts to respond to homelessness. Many organizations serving homeless persons in Calgary enter client information into a database called the Homelessness Management Information System, a system that is also used in many American cities. Client-level information (such as age, health status, employment status and housing status) is entered into the database when an initial intake is done. While the client is receiving services, updated information is entered again; in the case of some programs, follow-up assessments are done every three months. In the case of some program types, there are both exit and post-exit follow-up assessments completed. All information-gathering is subject to provincial privacy legislation. There are many uses for the data once it’s gathered. For example, some organizations use the data to provide case management services to clients. Also, funders are able to assess each organization’s performance against benchmarks (i.e. percentage of clients who receive housing after a specific period of time).

9. When it comes to both preventing and responding to homelessness, the capacity of government to generate revenue matters a great deal.Governments typically use revenue generated from taxation to finance both subsidized housing and other important social programs. When tax revenue decreases, many governments have less ability to spend on such programs. Since the mid-1990s, tax revenue in Canada (measured as a percentage of our Gross Domestic Product) has decreased substantially. If this trend doesn’t reverse itself soon, it will be very challenging for many governments (especially provincial, territorial and municipal governments) to invest in important social programs. There is currently a move afoot by some Canadians to increase taxes; it is led by Alex Himelfarb, former Clerk of the Privy Council. Alex and his son Jordan recently co-edited a book that calls for the need for higher taxation in Canada. (Note: according to some schools of thought, it isn’t necessary for a sovereign government with its own currency to tax more in order to finance more social spending. While keeping in mind that such an approach would be most relevant to Canada’s federal government and much less relevant to provincial, territorial and municipal governments readers can read more about one such school of thought here.)

10. Over the course of the next decade, Canada will likely see substantial increases in homelessness among both seniors and Indigenous peoples (First Nation, Metis and Inuit). Seniors and Indigenous peoples are growing as a percentage of Canada’s total population. Further, the percentage of seniors living below Statistics Canada’s Low-Income Measure has grown substantially since the mid-1990s. I think all of this makes it likely that both of these groups will begin to grow as a percentage of Canada’s homeless populations.

–

The following individuals were very helpful in helping me prepare the present blog post: Maroine Bendaoud, Lisa Burke, George Fallis, Greg Suttor, Francesco Falvo, Louise Gallagher, Ali Jadidzadeh, Lisa Ker, Jennifer Legate, Kevin McNichol, Richard Shillington, Blake Thomas and Mike Veall. Any errors are mine.