Most young adults do not become independent on a birthday. They lean on family well into their twenties: a place to stay between leases, a loan when the car dies, someone to call when a plan falls apart. We treat that long, soft landing as normal, because it is. Then there are the young people who turn a certain age in government care and are expected to be fully independent overnight, with no one to call. We have built a system that withdraws support at the exact moment most families lean in. That is the cliff, and the fall is measurable.
The support most young people get now stretches into their thirties; the care cliff has not moved
01 · The scaleAn incomplete count
Aging out happens on different timelines across Canada, because provinces and territories define the end of care and the supports beyond it differently.10 The country still cannot produce one complete, comparable count of who leaves care each year and what happens afterward. That absence is not a footnote. A state acting as parent should be able to say whether the people it raised are housed, connected, learning, working, and well after its formal responsibility ends.
Because child welfare is run province by province, with no mandatory national outcome reporting and no federal role except for First Nations children, Canada does not fully know how many young people it sets adrift, or what happens to them afterward.3 We are running a national policy without a national dashboard.
For practitioners: why the numbers come in ranges, and why the comparison above is not clean +
The figures above start from people experiencing homelessness and ask how many had a care history. They do not start from everyone leaving care and calculate how many later became homeless. Those denominators answer different questions and cannot be exchanged. The point-in-time method also undercounts hidden homelessness, which matters especially for young people staying temporarily with others.1
The discipline: state the definition, state the window, and never let a cleaner-sounding number travel without them. When you see a single dramatic statistic with no denominator, treat it as a claim to verify, not a fact to repeat.
02 · The moving baselineIndependence at thirty, care at eighteen
The cliff is usually described as support ending at eighteen while families keep helping. The truth is worse, because the age of independence itself has moved. Developmental researchers now treat the late teens through the twenties as a distinct life stage, "emerging adulthood," a period of exploration, instability, and semi-dependence that for most young people runs to roughly twenty-five and often to the end of the decade.8 Today's young adults reach the traditional markers of adulthood, stable work, a home of their own, a partner, children, years later than their parents did, and they get there leaning on family the whole way: a room between leases, a loan when the car dies, and a soft place to fail and try again. The median age of a first marriage now sits in the late twenties.8
So the standard a care leaver is measured against is not "independent at eighteen." It is "supported, still forming, and allowed to be in-between until the mid-to-late twenties." A young person aged out at eighteen to twenty-one is not simply losing help a little early; they are being asked to skip an entire developmental stage that their peers get to spend inside a family. Against a baseline that keeps climbing, the fall is not only early; it gets steeper every year the rest of us take longer to grow up.
And not everyone even reaches the statutory age inside the system. Placements break down, relationships end, some young people leave or are moved on, and others are discharged early; so for a meaningful share the exit does not arrive at twenty-one, or even at eighteen, but at sixteen or younger. That is the steepest version of the cliff: support ends not late in emerging adulthood but before it has properly begun, at an age when a young person in a family is still years from being asked to stand on their own. The earlier the exit, the longer the unsupported stretch, and the further behind the line they start.
The earlier the exit, the longer the stretch with no one behind you
We moved the finish line for everyone else and left it at eighteen for the young people with the least behind them.
For practitioners: measuring against a moving norm +
If independence is now a process that runs to the mid-twenties for the general population, an outcome measured at nineteen or twenty-one for a care leaver is being taken at a point where almost no one is expected to be settled. Compare like with like: judge transition supports against what a reasonably resourced twenty-five-year-old looks like, not against an idealized, self-sufficient eighteen-year-old who never really existed. The right benchmark is developmental, not administrative.
03 · The cascadeLinked outcomes, not separate misfortunes
What we know about outcomes is consistent across studies. Care leavers face sharply elevated rates of homelessness, interrupted education, unemployment and under-employment, poverty, mental-health struggles and post-traumatic stress, substance use, and early parenthood, with employment roughly twenty percent below that of their peers.2
These are not separate misfortunes; they cascade. Lose stable housing and education becomes nearly impossible. Lose education and the labour market closes. And the family supports that would normally break that chain, a spare room, a co-signer, a ride, are exactly what a care leaver does not have. A single early shock propagates because there is no buffer behind it.
A system should be judged not only by how it treats the people in its care, but by how it treats the people it is done with.
04 · The clearest signalFrom placement to shelter bed
Homelessness is the clearest signal because it is the hardest to explain away. Looking from the homelessness system back toward childhood, 45 per cent of youth respondents in the federal point-in-time dataset reported experience in government care.1 That does not tell us the probability that any individual care leaver will become homeless. It tells us that the care system is dramatically over-represented among young people the housing system later finds without a home.
For practitioners: reading a two-directional statistic +
"How many care leavers become homeless?" and "how many young people experiencing homelessness were once in care?" are two different measurements. Canada has stronger national evidence for the second than the first. Say that plainly. The disciplined argument does not need an unsupported majority claim; the verified 45 per cent finding is already a profound systems signal.
05 · Why it happensA date, not a readiness
The cliff is not caused by a lack of effort from these young people, who often show extraordinary resilience. It is a design flaw. Support that should taper instead stops. A relationship that should continue is legally ended on a schedule. A date, not a readiness, decides when help ends, and the date is chosen for administrative convenience, not developmental fit.
Jurisdiction makes it worse. Because thirteen systems set their own rules, a young person's support can depend on which border they happen to age out inside, and a good reform adopted in one province can die at the next one's line. The problem is rarely a lack of knowledge about what to do. It is will, funding, and fragmentation. It is also expensive: in Ontario alone, the cost of inaction for youth aging out has been estimated at roughly two billion dollars over their lifetimes.7 A ramp is not only kinder; it is very likely cheaper.
06 · What a ramp looks likeThe interventions are not subtle
What works is well established and unglamorous: extending care past eighteen, supported and gradual transitions rather than a hard exit, and, above all, a trusted adult who does not disappear.5 That is, more or less, what stable families provide by default. When jurisdictions raise the age of support and keep a relationship intact, housing stability, education, and employment all tend to move in the right direction.5 The barrier is rarely the evidence; it is the decision to fund a ramp instead of a cliff.
If independence now runs into the mid-twenties, the most honest reform is to make support match the developmental reality rather than an administrative date; extend care and its relationships toward the mid-twenties, the way a family would. And the ramp does not have to be invented; parts of it already exist and work. Education is the clearest example: British Columbia waives post-secondary tuition for current and former youth in care, at any age, and layers grants on top for living costs, so a young person from care can attend a public college or university tuition-free.9 That is a funded, running ramp; the question is why it is the exception rather than the floor across the country.
The leverage is not only at the edge. Much of the fall is preventable upstream, through health promotion and prevention long before the exit: stable placements instead of many moves, mental-health promotion rather than only crisis response, and keeping one trusted adult in place through the transition and past it. A bursary catches a young person who is already falling; a stable placement and a lasting relationship mean fewer of them reach the edge at all. Both matter, and the second is the cheaper investment.
The current Canadian direction: the Senate's national study calls for a pan-Canadian framework, comparable data, extended and equitable supports, and transition planning built around the young person's needs rather than a jurisdiction's birthday rule.10 The important word is equitable. A benefit that exists on paper but requires stable housing, perfect records, or sustained administrative navigation may systematically miss the people who need it most.
For practitioners: evaluating a "raise the age" reform without fooling yourself +
Extending care tends to look good in the data partly because the young people who stay are, on average, different from those who leave; selection can masquerade as effect. If you are evaluating one of these programs, guard against it: compare like with like, prefer a developmental or quasi-experimental design over a simple before-and-after, and measure the mechanism, not just the outcome. Did a trusted-adult relationship actually persist? Did the transition taper, or just get relabelled? A reform that raises the age on paper but keeps the hard exit in practice will show weaker results, and the weakness will be in the implementation, not the idea.
This is the difference between an evaluation that judges a program and one that helps it steer, the distinction I keep returning to in these notes.
07 · Try itAudit the ramp, not the young person
This short audit asks whether a transition plan provides what a reasonably supported family would provide: someone who stays, somewhere safe to land, material backing, health continuity, and real control over the plan. Complete it about the system's offer, never as a score of a young person's readiness.
Transition-ramp audit
Answer each condition based on evidence, not intention. The report treats the weakest domain as the first design priority.
08 · The measurement problemWe cannot fix what we refuse to count
Underneath all of this sits a quieter failure: we barely measure it. Without mandatory, comparable national reporting on what happens to young people after they leave care, every province optimizes in the dark and no one is accountable for the aggregate.3 Better outcomes start with the unglamorous work of counting honestly, disaggregating by the communities most affected, and publishing what the numbers show even when they are uncomfortable. Measurement is not the opposite of compassion; it is how compassion learns.
The cliff is not a force of nature. We built it, on purpose, with a birthday. We can build a ramp instead, and we know how. What is missing is the will to fund the boring, durable things, a longer runway, and a person who stays, and the discipline to measure whether they worked.
Sources & notes
- Youth homelessness in Canada (Data Snapshot). Housing, Infrastructure and Communities Canada, 2024. Youth exiting care are over-represented among those experiencing homelessness.
- The link between foster care, homelessness, and criminalization. The Homeless Hub / Canadian Observatory on Homelessness. Source for the ~10% (about 6,700) who age out annually, the 18-21 range, employment ~20% below peers, and the "nearly three-quarters" figure.
- Rates of out-of-home care among children in Canada. Public Health Agency of Canada, 2024. On the limits of national administrative child-welfare data.
- The Outcomes of Indigenous Youth Aging Out of Care and Exiting Care in Canada (Environmental Scan). Canadian Child Welfare Research Portal. On the disproportionate impact on Indigenous youth.
- Independent living programs and services for youth aging out of care in Canada and the U.S.: a systematic review. Children and Youth Services Review. On what supports show promise.
- Homelessness: how does it happen? Statistics Canada. Lifetime prevalence in the general population: ~2.2% of households have experienced absolute homelessness; ~11% some form, including hidden homelessness.
- The cost of inaction for youth aging out of Ontario foster care is estimated at $2 billion. Social Policy in Ontario, 2022.
- Arnett, J. J. (2000). Emerging adulthood: a theory of development from the late teens through the twenties. American Psychologist, 55(5), 469-480. On the distinct 18-to-29 developmental stage, and the later, family-scaffolded transition to adulthood.
- Provincial Tuition Waiver Program. Province of British Columbia. Tuition waived for current and former youth in care at B.C. public post-secondary institutions, with additional grants for living costs; expanded to former youth in care of all ages in 2023.
- Standing Senate Committee on Human Rights. Nothing to Celebrate: The Crisis of Youth Aging Out of Care. National evidence, testimony, and recommendations on extended supports, equitable access, data, and a pan-Canadian framework.