There's a better way to fight homelessness than emergency shelters -

There’s a better way to fight homelessness than emergency shelters

OPINION: Housing First initiatives are better at combatting homelessness than shelters, and make more fiscal sense for taxpayers, too

(David Cooper, Getty Images)

(David Cooper, Getty Images)

Recently, the Salvation Army of Canada announced its intention to build a 350-bed facility a literal stone’s throw away from my family home in Ottawa. Given the lack of public consultations that preceded the announcement, it came as a significant shock to both my family and the rest of the neighbourhood.

After the initial shock of the announcement wore off, many of us began researching what this facility would mean for our community. It would be comprised of 140 emergency-shelter beds plus an additional 210 beds for various types of residential treatment. A facility of this magnitude would not only have the typical impacts associated with condo developments, such as added traffic, noise, etc., but would also bring many more people struggling with addictions into the neighbourhood. And as the Salvation Army does not engage in harm reduction programs, such as needle exchanges or supervised injection, drug use, drug dealing, and alcohol consumption in public spaces can be expected to increase.

I initially felt as though commenting on this facility was not in my wheelhouse. My background is in economic and fiscal modelling and forecasting. And, although I have worked on evaluating public policy more than the average Canadian, evaluating models of effective service delivery for the homeless is not an area in which I have any expertise.

However, the more I researched this issue, the more I realized how central funding to support homelessness and addiction is to social policy at every level. And, as such, it has a significant impact on the fiscal frameworks of federal, provincial, and municipal governments.

At the federal level, the previous-Conservative government committed “nearly $600 million over five years (2014-2019) starting in April 2014 to renew and refocus the Homelessness Partnering Strategy (HPS) using a Housing First approach.” Then, the current-Liberal government committed an additional $112 million over two years in Budget 2016 to Housing First initiatives, and a further $2.1 billion over 11 years in Budget 2017.

Housing First, as defined by the federal government, involves “moving primarily individuals who are chronically or episodically homeless from the streets or homeless shelters directly into permanent housing. Permanent housing is complemented by the provision of services to assist clients to sustain their housing and work towards recovery and reintegration into the community.”

The reason for this multi-partisan support for Housing First is due to the results of the federal-government sponsored research project, At Home/Chez Soi. The largest project of its kind in the world, it established pilots in five Canadian cities (Vancouver, Winnipeg, Toronto, Montreal and Moncton) in 2009 to examine outcomes of the Housing First approach. The At Home/Chez Soi Final Report “demonstrated that a Housing First approach can be effectively implemented in Canadian cities both large and small, to rapidly reduce homelessness while alleviating pressure on shelters and health and judicial services. Once stable housing is obtained, the focus shifts to more enduring issues, such as addiction and mental health.” Indeed, in the last six months of the study, “62 per cent of Housing First (HF) participants were housed all of the time … whereas 31 per cent of treatment as usual (TAU) participants were housed all the time …” Further, “Among participants who were housed, housing quality was usually better and more consistent in HF residences than TAU residences.” So, we know from this work that “Housing First rapidly ends homelessness.”

But, while it’s clear that Housing First is much more effective than the emergency-shelter model in combatting homelessness, what is the fiscal impact? Fortunately, this too is a good news story.

According to researchers at McGill, using the data from the At Home/Chez Soi project, “Support services for homeless people with mental illness in Canada’s biggest cities cost more than $55K a year per person on average.” To put that in perspective, the average Canadian earns just shy of $50K annually before taxes.

Meanwhile, researchers in Regina found that Housing First created cost savings of nearly $75,000 per person annually through lower costs of policing, days in hospital, emergency room visits, and detox visits. Further evidence is provided by the MaRS Centre for Impact Investing. In their ‘Housing First Social Impact Bond Feasibility Study’, the authors found that Housing First was consistently cheaper across measures of ‘health and justice use’ as well as ‘overnight stays’ due to lower incidence. Still further research in other jurisdictions, such as Victoria, has confirmed the same thing.

In all, I have been unable to find any evidence that does not support Housing First over emergency shelters, either from a cost or outcome perspective. This begs the question: Why are tax dollars being spent on expensive emergency shelters when we know they are ineffective at combatting homelessness and addiction?

In this case, as in all others, evidence matters, and citizens should demand politicians base their decisions on it. When they do, we are all better off.


Randall Bartlett is the chief economist of the Institute for Fiscal Studies and Democracy



There’s a better way to fight homelessness than emergency shelters

  1. Hi Randall – Good piece. I have some experience with the conundrum of why Housing First hasn’t become the standard treatment for all homelessness. Like you, my background was as a policy analyst but negotiating large scale collaborations among dissimilar partners. I was asked to develop Saskatoon’s Plan to End Homelessness after a group of agencies had reached the conclusion this was the best next step.
    In my experience, the barriers to uptake are as follows:
    – there are widespread savings (shelter costs, policing, health and hospitalizations and more) but in practice, these are all born by different levels of government (municipal, provincial and sometimes federal) and within a province by different departments. So the savings are widespread but the cost of implementation is usually born by Social Services. Unless senior leadership accepts a systems cost born by the province, Housing First wallows without champions.
    – A cabinet minister told me this: “I can’t tell people who are working hard to make ends meet that we’re giving free houses to drug addicts”. It would take a lot of explaining to breech that gap.
    – Many of the savings are illusory. Yes, fewer people will use the ER as a default winter shelter (at $1300/visit) but in fact even if homeless people stopped coming in, noone would reduce staff levels. The benefit would come in shorter ER wait times, more time/better quality visits. Similarly, fewer homeless people being picked up by policy for public intoxication (they can drink in their own homes like the rest of us) wouldn’t mean laying off cops. Instead, there would be better responsiveness by police and the ability to focus on policing issues that might not get a lot of attention.
    – a rethink by society is required. Many, many agencies, churches and more are invested in charitable projects to help homeless – free socks, soup kitchens, fighting city hall for the right to tent cities, winter coat drives. In fact, this means all these resources are dedicated to keeping people comfortable on the streets when the resources should go to the central problem of providing housing, thereby solving the problem.
    – Finally, 85% of homelessness is short-term and people solve the problem themselves. A young couple moves to the city for jobs, break up so he leaves the apartment but has nowhere to live and no network to rely on, for example. Studies show anything other than a short-term shelter bed (programmatic responses) can actually hinder these people. The focus of Housing First and other responses is on the 15% of chronically, year-over-year homeless.
    That said, I remain a champion of Housing First. Individual case management gives people a chance to move lives forward on a different trajectory. There are numerous stories of crack cocaine use dropping by 80% just because of housing: the addiction levels were a response to living on the street. Or of people finding the ability to work part-time despite FASD or mental health challenges. It is a reflection of us that when most chronically homeless people have mental health problems, we don’t do more to find compassionate solutions that are often beyond their ability to solve themselves.

  2. Excellent article and good comments from cityparker. Housingfirst is both humane and economical – an appealing win-win situation. 40 years of experience has taught me that, without a home, services do not have a lasting impact. and without a home we as a society are forced to ‘shelter’ the un-sheltered. We need to find a way to realize the savings in the various systems and to acknowledge the improvements in services that result from reducing the draw on those services – that’s the challenge for our political and community leaders. We often need to do the difficult thing because its the right thing to do.
    The temptation to provide shelter and then bring in services to help those that are sheltered is a slippery slope designed primarily as a convenience and efficiency for the providers and not for the recipients.
    I’d also like to note that sheltering people on a large scale simply creates an environment in which people are trapped by the nature of the environment and the people living there. Whether its a jail, a hospital or a shelter these are environments that present their own unique problems for people to overcome. Whether in the animal, plant or human world, an integrated, community is a stronger, more resilient community.
    And when i see a shelter that will house 300+ individuals I wonder if we are not simply recreating the institutions for the mentally ill that we all were happy to de-commision just a decade or two ago.