The Paradox of Homelessness

Last week, I had an op-ed in MSNBC talking about the one bright spot in HUD’s annual count of the country’s homeless population: while homelessness overall continues to rise, veteran homelessness has fallen precipitously. The argument I make about why this happened should sound familiar to longtime readers of this newsletter.

As I note in the piece, veteran homelessness fell drastically between 2008 and 2016 (the period when the VA first adopted Housing First as its north star), barely budged from 2016 and 2020 (the first Trump era, when the federal government largely abandoned Housing First), and then resumed its rapid plunge from 2020 to 2024, after the Biden administration once again put Housing First at the center of federal anti-homelessness policy.

No big mystery there. But the fact that overall homelessness rose under Biden presents us with something of a puzzle. Americans saw strong wage growth over the past four years, with the largest gains going to the lowest-wage workers. So why did so many low-income households enter homelessness during the same period?

It may be tempting to wave this away as a fluke, but researchers have been noting the strange link between income growth and rising homelessness for a while. In Homelessness is a Housing Problem, Gregg Colburn and Clayton Page Aldern write: “Poverty rates are, more often than not, relatively low in places with relatively high rates of homelessness. […] This finding suggests that homelessness—at a metropolitan level—is more a symptom of affluence than of poverty.”

This general finding agrees with the evidence we can gather with our own eyes. The metropolitan areas in the United States with the highest concentrations of homelessness — and with the most visible, unsheltered homelessness — tend to be important centers of American wealth and economic productivity. The Bay Area, Los Angeles and New York are all marked by the oft-remarked-upon juxtaposition between gleaming corporate towers and urban encampments. And all three of these areas have seen decades of both strong wage growth and growth in the local homeless population.

Some readers may respond, “Sure, but strong wage growth for whom?” And it is true that the biggest income gains have gone to those who were already at the top of the income distribution. But poorer residents of these metropolitan regions have gotten richer, too. The modern American homelessness crisis started in the late 1970s/early 1980s; between 1980 and 2021, inflation-adjusted wages for the Bay Area’s bottom quintile of income-earners have grown by nearly 25 percent (though they are still below 1970 levels).

But if wage growth for the bottom quintile has outpaced inflation, so have housing costs. Over the same period, median rent in the Bay Area actually doubled. The story of the Biden economy is the story of the past 40-odd years in superstar metro areas, except more so: everyone’s take home pay went up a bit, but housing costs went up even more.

This dual trajectory isn’t a coincidence. Perversely, there is a sense in which rising incomes caused rents to go up, which in turn led to more homelessness.

That’s in part because higher median incomes attract new residents. For example, the Bay Area tech boom continues to draw in ambitious STEM majors from across the country and the world. And it isn’t just engineers who are drawn by the tech economy. As the number of affluent residents and thriving businesses grows, so does the demand for ancillary services: logistical support for the businesses themselves, but also luxury amenities like nice cocktail bars and Michelin-starred restaurants for high income earners. And while waiting tables at Kiln won’t earn you tech money, the existence of places like Kiln does drag up median food service industry wages. In theory, this is good for everyone.

Wage growth only turns into higher homelessness when you add in an additional factor: housing supply. If housing supply doesn’t keep up with population growth, then wealthier newcomers end up competing with lower-income residents for homes. Effectively, they bid up the cost of housing. And the more that population growth outstrips housing supply, the larger that delta between bottom-quintile wages and housing costs becomes.

That’s how places like the Bay Area and New York — and, increasingly, most other productive urban areas across the country — generated mass homelessness: they found a recipe for growing most people’s wages but neglected to expand their housing supply.

It’s important to understand this relationship between income, housing supply, and homelessness because of what it tells us a lot about how we should respond to the homelessness crisis. It puts the lie to a talking point I’ve heard quite frequently, including from academics who should know better: that the homelessness crisis is primarily an income crisis, and that it can simply be solved solely through redistribution of existing resources.

While I certainly support redistributive policies, there is a limit to what wealth transfers can do on their own. As long as a metro area has housing scarcity and any disparities in spending power, the relatively high-income residents will still outbid their lower-income neighbors. And in a hypothetical universe where you we were somehow able to totally equalize incomes but kept housing supply fixed, we would still need some way of rationing available units. Some people would still end up sleeping in their cars.

(At this point in the conversation, some people will volunteer the opinion that population growth is the problem, not income growth or housing supply. We simply need to keep some non-preferred classes of people from accessing the wealth of our most productive cities. I’ve written about this type of folk Malthusianism before, so I won’t spend much time dealing with it now. Suffice it to say I find this argument to be myopic and counterproductive at best, crypto-fascist at worst.)

There’s really only one effective and morally palatable way to raise incomes for most people without consigning an unlucky minority to homelessness: you need to create the land use flexibility that will allow housing supply to roughly keep pace with income and population growth. And yes, that means allowing for the construction of more expensive “luxury housing” that can absorb high-income earners. Those units will keep such households out of the lower end of the rental market, which means the lower end will remain affordable for the working class. And if we effectively get rid of housing insecurity among the working class, it means we can more effectively target public subsidies to the truly destitute. That’s the only way out of the homelessness trap.