October 28, 2013

Cohort effects in homelessness: Lifelong higher risk for those born near beginning of crime wave

Tonight I ran into an unfamiliar situation -- a panhandler under the age of 40. I hadn't thought about their age structure before, but meeting someone not too different from me was strange. Subtle speech patterns, mannerisms, slang, and so on made this guy stand out as someone from my group.

Most homeless folks these days do seem to be older than they used to be. I didn't live in an urban area as a kid, but whenever I did go into the city or visit other cities, I don't recall the homeless being that old back then. And sure enough, demographers have found that the homeless population has been aging since the start of their data in the late 1980s. See here for a short review article, or here for an NPR story with an accompanying info-graphic.

No matter when the data were collected, the peak of the age distribution consistently points to people born in the later '50s and earlier '60s, ground zero being around 1960. And it's not just a reflection of the general population's age curve shifting in the older direction. Even accounting for what chunk of the overall population is made up of late Boomers, they always show higher relative risk for being homeless. That's for the single male population, who are the majority.

Homeless families headed by single mothers (the other sub-population) show women aged 18-23 as having the greatest risk, no matter what year, hence no cohort effects. Their turn at homelessness is temporary and facultative, not part of some deeper set of traits that they have and will carry with them through life.

Before trying to figure out what about their upbringing and maturation was distinct, we should see if there's another time when the homeless were old. It keeps showing up, so you shouldn't be surprised to hear that it was the mid-century. The article above refers to studies done on the skid row phenomenon of the mid-century:

Researchers of skid row found that the population consisted almost exclusively of older, single white male households, with three-quarters of the men over the age of 45 (Blumberg et al., 1960).

So, the median age must have been higher than 45, perhaps around 55 or 60, which would place their births around the turn of the century. Just as the homicide rate was beginning to rise (peaking in 1933).

My vague impression of the tramp and hobo phenomenon during the 1920s was that it was younger and more footloose men in their 20s and 30s, not middle-aged and borderline elderly men living on skid row. This would have been similar to the situation of the 1980s, when the homeless population had grown much younger since the mid-century, and before the return back to the mid-century pattern during the past 20-25 years.

Here are a few pictures of hobos from the Jazz Age, and they look relatively young and fit, probably in their 30s, and here is a gallery of mostly mid-century skid row in San Francisco. Notice how they get older and older, with baldness or white hair becoming common by the '50s and '60s. Here is a gallery of homeless folks in San Francisco in the late '80s, and notice how younger they've gotten. Here is a representative shot of today's much older homeless, not too different in age from the man in this shot of Detroit's skid row in the '50s.

The main difference between Millennial and mid-century fates of the lowest of the economic ladder is that their standard of living was much better back then than today. Inequality was low and still falling back then. At least they were probably living in a flophouse rather than having nowhere to live, and being vulnerable to theft and violence from living out in open spaces all the time. Doesn't seem like they got harassed as much either -- "go get a job you bum!" -- and the winos were left to drink their misery away. I wouldn't want either fate, but drinking and falling asleep in a doorway sounds better than getting hostile looks in between having my stuff stolen by other bums out in the open.

What makes people born around the start of a crime wave the most likely to live a transient lifestyle, whether they were born circa 1900 or 1960? They're the first cohort not to be exposed even somewhat to the sheltering and smothering culture of a falling-crime period (Mom-ism was the norm in Victorian and mid-century times). Parents start to put away their Dr. Spock manuals and instead tell their kids that once they're 18, they're out of the house. Independence becomes valued over pampering.

And as everyone begins to emerge from their cocoons, those born around the time of the shift will assume that life is supposed to be lived out in public, not holed up indoors all day. Those born too much earlier will have experience with the cocooning norm pulling them the other way, and those born much later will have experience with the next wave of cocooning. But if you were born around 1960, your entire formative experiences -- up through age 30 -- were untouched by mid-century or Millennial cocooning. They'll feel the most at home outside the home.

Then there's the drug culture, whose trends nearly overlap the trends for crime and outgoing lifestyles. Homeless people, winos, bums, denizens of skid row usually have some degree of substance abuse problems, some very serious, making it hard to pick themselves up and lead a productive life again. Someone born around 1960 grew up unaffected by the mid-century and Millennial anti-drug zeitgeist. Drugs were just becoming popular when they were small children, and would only surge in popularity throughout their formative years. The drug culture seems to peak about 5-10 years after the homicide rate does, so that's well beyond age 30 for the 1960 cohort.

Both of these factors appear necessary -- you could always shoot up in a place where somebody you knew was letting you stay, particularly if you didn't like venturing outside. Homelessness requires not only some kind of drug habit or severe mental problems, but also the push from others and the desire from oneself to want to get out of the house and stay out there.

Nobody seems to care about homelessness as an issue anymore, as social isolation closes other people off from your view, and out of sight, out of mind. People dwell on their own troubles more, and a sense of existential malaise begins to set in. In outgoing times, people are too focused on others, including how they can help them through their troubles.

Woodstock fans who tuned out before pop culture reached its peak in the '80s might not even know about Live Aid and the other zillion national benefit concerts for famine relief in East Africa in the mid-'80s. And in fact, political content (including "social issues") in pop music lyrics reached a peak in the late '80s and early '90s. "Luka" by Suzanne Vega, "Man In the Mirror" by Michael Jackson, "Something to Believe In" by Poison, "Runaway Train" by Soul Asylum, and so on. All those touch on homelessness except "Luka," which is about child abuse.

Incidentally, Suzanne Vega was born in '59, Michael Jackson in '58, Bret Michaels from Poison in '63, and Dave Pirner from Soul Asylum in '64. It's easier to see a soaring problem when it's striking your generation-mates, and that also makes it easier to try to reach out to those who need it most. It's harder when you have to cross generation lines.

That effect is amplified by which generation is getting hit -- namely, the one that experiences rising social connectedness all throughout their formative years. If it were Millennials who had to help other Millennials, or Silents helping Silents, forget about it. But for the generation that were children in the '60s, teenagers in the '70s, and young adults in the '80s -- it's second nature to tune in to what other people's situation is, help if they need it, or ask if you need it from others.

I suppose that's another key difference across the generations too -- how comfortable are you asking others for help? Silents and Millennials are withdrawn and would see it as a sign of weakness, an embarrassment, and an awkward thing that would need to be repaid, making them feel in debt. You aren't going to get by very well as a wino, homeless, or whatever, unless you can reach out to strangers for assistance.


  1. When I think "homeless" and "song from late 80s/early 90s" it's "Rockin' in the Free World" (a serious Young fan wouldn't list it as their favorite song by him, but it might be mine). In contrast, I always found Everlast's "What It's Like" undermined by its annoying self-righteousness. although he's perfectly capable of being earnest as with "Put Your Lights On".

  2. As long as we're talking music, another entertaining "the music was better when I was in the targeted demographic" discussion broke out at Steve's due to Lou Reed.

  3. Seems kind of surprising in light of when you search for hits on mean age of the homeless, for present day samples, they generally talk about people between the age of 30-50, with a mean of 42-44 years old, which seems fairly cross nationally consistent.

    That doesn't suggest under 40s should be rare at all. Most of the beggars I meet seem young, in late 20s to 30s.

  4. Parents start to put away their Dr. Spock manuals and instead tell their kids that once they're 18, they're out of the house. Independence becomes valued over pampering.


    Virtually all earlier studies assert that the age of home-leaving was declining from as early as it could be measured until 1970. Our results say otherwise. From 1880 until 1940 for males and 1950 for females, the age at leaving home did not decline, it rose (see Figure 1). The decline came later. Led by men in 1940, the generation of the Second World War experienced a sharp decline in the age at which they left home, one that continued for everyone until 1960 and for white men and women until 1970. Beginning with 1970, the age of home-leaving rose again, reaching relatively high levels by 1990.

    There are certainly ways you could explain this in your model - during cocooning periods, kids want to leave home earlier to form their own cocoons - but it still doesn't seem to map great. It maps a little with inequality, but again not great.

    This isn't to say that you're wrong about people living their life in the open, just that that may not have meant getting their own place and setting up a household.

  5. Phil Collins "Another Day in Paradise"


    "Think Twice, cause its another day for you and me in paradise..."


  6. A large number of homeless people tend to be military veterans, so expect to see spikes in homeless men a few years after wars or major military campaigns take place.

  7. Very interesting observations.

    Though, for the record, let's establish that the appeal you're making to parenting is nonsense. To be clear, behavioral genetic data make it clear that parenting has no impact on how children turn out:

    Taming the “Tiger Mom” and Tackling the Parenting Myth « JayMan's Blog

    That out of the way, there may be a simpler explanation for this apparent cohort effect you see: reproduction was far more universal during the Baby Boom:

    Some guys get all the babes – not exactly | JayMan's Blog

    Throughout most of the century, around 15-20% of people failed to leave any descendants (with reproduction being more even among women than men). However, during the boom years, that fell to less than 10%. We can imagine that many of the individuals who reproduced during that time were of, in one way or another, were of substandard genetic quality. If so, all manner of physical and mental ills would have become more common among people born during that period.

  8. You've misread the post. I said that when parenting norms shift toward encouraging independence, children will get pushed out the door and be on their own more than when parenting norms are more toward smothering.

    Parenting may not affect the personality of the child, but it does reflect the personality of the parent. And when the parent says "out," the child is "out," and must adapt to that fact whether he's genetically inclined or not.

  9. @ Jayman,
    You wrote:
    " parenting has no impact on how children turn out:"

    If data says that, the data is wrong! You obviously haven't been around with open eyes very long.


You MUST enter a nickname with the "Name/URL" option if you're not signed in. We can't follow who is saying what if everyone is "Anonymous."