While the majority of the general public thinks drug addiction is the main reason for homelessness, the reality is quite different.
On any given night in Australia one-in-200 people are homeless.
Few would disagree that this number is far too high, but there is much debate about why people become homeless in the first place.
Rough sleepers are the most visible of the homeless population, but it is actually much larger than just those sleeping on the streets.
Being ‘homeless’ includes anyone whose housing conditions don’t qualify as a decent ‘home’ – like squatting in abandoned buildings, having to stay with relatives or friends temporarily when there is no alternative, or staying in a caravan park, boarding house, hotel or crisis accommodation.
If we’re going to really address the huge challenge of homelessness and design policies that really work, we need to better understand why people become homeless in the first place.
And there is a significant perception gap between what the general public think about why people become homeless, and what people who have experienced homelessness say – particularly when it comes to substance use.
A survey conducted by Hanover Welfare Services in 2006 found that 91 per cent of Australians believe that drug addiction is one of the main reasons people become homeless. This view is widely shared in the community, including among ‘experts’ in public policy and academia.
But if you ask people who have experienced homelessness, only 10 per cent say this is the case.
So, while homelessness and substance use are indeed commonly associated, is it actually the case that people become homeless because they use drugs?
Substance use and homelessness: what comes first?
Our team investigated the reality of the relationship between substance use and the onset of homelessness.
Our research, which is published by the Royal Statistical Society, found that illicit drug use was only linked to homelessness in young men, not young women. And even then, only daily cannabis use increased men’s likelihood of becoming homeless, with no effect shown for the additional use of harder drugs.
We used the Journeys Home dataset – the largest and most comprehensive longitudinal study of homelessness and housing insecurity internationally. In this sample more than 75 per cent had experienced homelessness by the age of 30, and almost 50 per cent of the sample had used drugs regularly (cannabis daily and/or hard drugs weekly) by the age of 30.
This high occurrence of homelessness and substance use offers a rare opportunity to analyse the relationship between them in greater detail than with other data.
Consistent with general opinion, our research showed substance use and homelessness are associated. Of people who used drugs regularly by age 30, 86 per cent had experienced homelessness. Among those who had not used drugs regularly, this figure is 70 per cent.
But is that enough to conclude that using drugs increases the likelihood of homelessness?
There are other things that could explain this link.
In some cases homelessness may lead to drug use. Also, other characteristics and events, like risk-taking behaviour or adverse circumstances in childhood, may make people more prone to both homelessness and substance use.
To rule out these alternative explanations, we took into account the timing of events – did a person’s drug use start before or after they became homeless?
We then accounted for all the permanent differences between respondents that may have affected their experience with substance use and homelessness (like personality traits or childhood circumstances). Because Journeys Home holds detailed information on whether and when respondents started using drugs, and whether and when they became homeless, we were able to interrogate these questions.
The real causes of homelessness
Interestingly, our results are much more closely aligned to the beliefs of people who have experienced homelessness than to public opinion.
The use of illicit substances other than cannabis didn’t increase the likelihood of someone becoming homeless. And when it came to cannabis use, women under the age of 30 who use it daily are no more likely to become homeless than those who don’t. For men, using cannabis daily increases their likelihood of becoming homeless by age 30 by 7-14 percentage points.
In contrast, we found in previous research that the effect of parental separation on homelessness is substantial for both genders – specifically, it is six times that of drug use across both men and women.
When you break it down, the effect of parental separation is about twice as large as that of using drugs regularly for men (combining the use of cannabis daily and illegal/street drugs weekly) and ten times as large for women.
Again, this research also shows that people who have experienced homelessness have a more reliable sense of why they found themselves in that situation than the general public.
They cited ‘relationship breakdown and conflict’ as the main cause for homelessness six times more often than substance use (64 per cent vs 10 per cent). In contrast the general public cites ‘marriage or relationship breakdown’ as being the main cause for homelessness less often than that of substance use.
This shows how valuable people’s insights into their own experiences can be when designing policies that work.
Ultimately, our findings demonstrate that while substance use does increase the risk of homelessness for young boys and men, the effect is nowhere as high as what is generally believed. Our research suggests that early interventions to reduce cannabis use may be effective in reducing the number of boys and young men who become homeless but will unlikely have a similar effect for young women.
But policy interventions supporting the housing needs of families that break down, can effectively reduce the transitions of children and young adults into homelessness, possibly breaking the path into lifelong extreme disadvantage.
These results constitute a critical step in understanding how people, in particular children and young adults, become homeless.
This article was first published by the Pursuit blog, from the University of Melbourne.