As headlines go, it’s a grabber.
“Nine Out of 10 Families at Risk of Homelessness in Toronto’s Aging Rental Highrise Buildings,” declares the title of a recent report from the University of Toronto’s Factor-Inwentash School of Social Work.
An accompanying press release from the university’s media office explains the study comes from a “survey of more than 1,500 families with children living in rental high-rises built between 1950 and 1979,” a housing category that covers half of all rental units in metropolitan Toronto.
An academic study from the country’s largest university thus proclaims almost everyone in Canada’s largest city in an apartment from the disco-era or earlier may soon find themselves out on the street. If that sounds like a crisis, you’re right—it’s a crisis in what passes for social policy research these days.
A quest for Really Big Numbers has taken over social policy advocacy lately. We’re constantly barraged with enormous figures allegedly detailing various catastrophes in our midst: poverty costs the Canadian economy $80 billion a year, heart disease and stroke costs $21 million, obesity $7 billion and so on. Sometimes, instead of dollar figures, we’re told of the legions of people suffering from various afflictions. To ensure appropriate Bigness, advocates add-in the curious caveat of being “at risk of.” Anyone “at risk of poverty” is not actually poor, of course. “At risk of hunger” means you’re eating regularly. And anyone “at risk of homelessness,” currently has a place to live.
These figures rarely get a fact check, but they shape public opinion, drive donations and become arguments for government intervention. The bigger the number, the better the chance it will affect public policy.
Which brings us back to the dubious claim that 90% of families in apartments of a certain age in Toronto are on the verge of living under a bridge. With the university document part of a campaign calling for federal action on “the housing and homelessness crisis,” (and given that the research was funded by Ottawa) this is a Really Big Number in real need of a closer look.
According to the report, families are considered “at risk of homelessness” if their situation meets any one of six indicators. These include spending over half household income on housing, sleeping two or more people per bedroom (excluding couples and same-gender children), being behind on the rent in the past year and/or living in a building that suffers from any two of: frequent elevator breakdowns, pests or broken entrance locks.
These may be legitimate housing problems. But trumpeting a broken elevator as portent of looming homelessness stretches logic. Even a family forced to make do with substandard housing would surely prefer such inconvenience to life in a shelter. The only category that properly indicates an immediate threat of eviction is rent in arrears. But this includes anyone who was previously behind but had paid in full by the time of the survey—and thus at no pressing risk of eviction.
The survey itself is described as “randomly sampled.” Yet over two-thirds of the interviews were conducted in seven specifically selected poor neighbourhoods in downtown Toronto—what the author calls “over-sampling.” The rest were scattered across the city. This renders the entire report meaningless from a statistical point of view. It’s not an exclusive look at poor households in Toronto. Neither is it a random sample of all residents. It’s simply a haphazard look at a bunch of renters, most of whom happen to live in low-income neighbourhoods. The public policy implications of this are nil.
None of the above disputes that homelessness is a problem, in Toronto or elsewhere. But our understanding of important social issues—and governments’ response to them—is not improved by inflated numbers or trumped-up crises. If housing advocates want Canadians to worry about homelessness, they should offer up the unvarnished facts, however big or small they may be. Otherwise they’re “at risk of” looking stupid.
Peter Shawn Taylor is a writer specializing in economic issues