Addressing Homelessness
London Breed, the Coalition on Homelessness, and Yet Another Unproductive Public Debate
Hey all, thanks for reading. Rolling Thoughts is a personal publication where I write about topics ranging from my career in Tech, to life in the Bay Area, to personal stories and memoirs. If you enjoy reading this piece, let me know by liking and commenting below. If you really enjoy it, help me reach new subscribers by sharing with a friend 👇🏼
One of San Francisco’s most contentious issues is homelessness.
I wrote a bit about it when I first moved here, and how jarring it was to see pervasive drug use on city sidewalks. But in recent days, the issue has shown itself to be more newsworthy: Mayor London Breed is fighting with San Francisco’s Coalition on Homelessness. The Coalition has worked to file a federal injunction, blocking the removal of homeless encampments from city sidewalks. Breed is fighting the injunction.
San Francisco politics is a long-running list of low brow, low-rung political mudslinging. It’s anti-productive, something this battle between the Mayor and the Coalition on Homelessness epitomizes.
In this essay, I’ll address the following:
Thoughts on the root cause of San Francisco’s homelessness.
Potential policy solutions to alleviate the issue.
Why the battle between the Mayor and the Coalition on Homelessness is a waste of time and city resources, as it completely ignores the two items above.
A quick note before beginning: this topic is dense. There are many different angles and prisms through which it can be viewed. I will omit a lot of critical information, because I’m writing an abbreviated essay, not a book.
With that understood, we proceed.
Defining the Problem
As best I can tell, no major city in the United States has an effective solution to homelessness. During a trip to New York City back in April, I watched a homeless man defecate on the wall of a Subway station as I found my way to work one sunny morning. Let’s accept an assumption: New York City, San Francisco, and all major cities in throughout the United States struggle with homelessness.
Most people will agree that we have a moral imperative to help folks like these. A high-functioning government should focus on the weakest and most vulnerable, and look for ways to improve their quality of life. Every major religion preaches something similar, and you’d be hard-pressed to find someone who would accept their son or daughter living the way the homeless in America live.
Setting this moral imperative aside, though: homelessness is expensive. It is a fiscal burden to the cities that experience it, and everyone benefits if we alleviate that cost.
There are the first-order costs of homelessness:
Government-subsidized food
Government-subsidized housing
Unemployment/cash handouts (a la San Francisco local policy1)
And there are some staggering second-order costs:
Increased crime (car break-ins, home break-ins, property lost due to theft) and increased police staffing to address these issues.
Increased prison/jail costs, as evidence suggests the homeless are more likely to be incarcerated, and more likely to become re-incarcerated2.
Worsening public health (literal piss and shit on city sidewalks, worse mortality rates across the homeless population relative to others, high rates of violent crimes perpetuated by and against the homeless, often due to lack of safety or chronic mental illness, etc.)
Increased healthcare costs as hospitals treat sick and injured individuals without employer-sponsored health insurance.
Depressed GDP, as chronic unemployment is typical across this segment.
Reduced tax revenues as fully employed/housed individuals migrate away from homeless-dense areas.
In summary: there’s a moral imperative to help the weak and helpless. Additionally, there are staggering public health and financial costs that are critical to address: solving homelessness would reduce crime, organically reduce demand for police, reduce prison costs to taxpayers, reduce healthcare costs to taxpayers, improve public health, and prevent migration away from major cities. Problems like these make cities less healthy. Let’s solve them.
Sticking Data to the Issue
In May of 2022, San Francisco had a ~$1.1 billion budget to address homelessness3, with approximately 19,000 homeless people (a ~36% increase since 2016, and about 2.2% of the overall population of San Francisco)4. This means San Francisco spends about $58,000 per homeless person per year, and that population is growing at a 7.7% compound-annual growth rate.
Translation: as San Francisco throws massive sums of money toward the problem of homelessness, the problem is growing exponentially.
The 36% increase in homeless people begs a question: where did they come from? Are these poor San Franciscans who’ve been priced out of their apartments, or are these homeless people from neighboring cities? Is this a problem of poverty and desperation and unfairness, or is this a problem of migration and osmosis?
The answer: it’s hard to tell. But we can test these theories with available data.
Hypothesis: if the recent increase in homelessness is due to San Francisco residents becoming homeless, we should see some correlation between volume of annual evictions and count of homeless people.
Hypothesis: if homelessness is not a result of net migration of homeless people into San Francisco, then it is not more attractive to be homeless in San Francisco than it is in nearby cities, like Oakland and San Mateo.
Looking at our first hypothesis: San Francisco publishes data on annual evictions5, and those data are publicly available. Aggregating everything in a quick spreadsheet, we see that evictions have been decreasing substantially, year over year, between 2015 and 2022. You can find these data + links to their annual reports here. Based on these data, it appears highly unlikely that recent growth in homelessness is due to local residents becoming homeless.
In 2016, San Francisco launched the Department of Homelessness and Supportive Housing. According to data from the San Francisco Department of Public Health and the Mayor’s Budget and Legislative Analyst’s Office, the budget for this department has grown exponentially, seemingly in lockstep with the city’s growth in homelessness. These funds are used in really attractive ways: cash handouts up to $687 per month, for example, as cited in footnote 1.
From this, we can infer: the Department’s staggering budget and cash-heavy use of funds makes it more attractive to be homeless in San Francisco - more attractive than other nearby cities - and it causes a net migration into the city of homeless people. This net migration is a key driver to the homeless issue.
Based on these data, we can define a successful policy-based approach to homelessness in San Francisco: spend less than $58,000 per homeless person, shrink the growth rate of homeless people below 7.7% per year, and do not displace homeless people to surrounding cities. Lift them out of it, instead. Accomplishing any of these will be more effective than current solutions.
Ideating Solutions
When I think about homelessness, I picture four quadrants: people with mental illness, people without mental illness, people with addiction, and people without addiction. See the matrix below.
Homeless people are not homogenous. Some are homeless due to family instability (more pervasive among teenagers), some are homeless due to massive financial debt (i.e. from gambling), and some are homeless due to drug addiction or mental illness. We can identify these as quadrant 1, 2, 3, and 4, as done below. People in quadrant 1, without mental illness or addiction, are the ones that can most easily be lifted out of homelessness, while quadrant 4 will have the most difficulty.
It’s important to note: these quadrants are dynamic. The longer someone stays homeless, the more likely they are to drift from quadrant 1 to quadrant 4. Mental illness may develop over time, and probability of addiction increases over time. Without intervention, quadrant 1 decays into quadrant 4.
An effective policy response to homelessness, then, will be focused on moving people from quadrants 2, 3, and 4 closer to quadrant 1, where success rate is presumably highest.
Policy Suggestions
So long as the total cost of these programs averages to $58,000 per person (or less), and the gross volume of homeless individuals decreases year over year, these solutions will improve existing solutions.
Halt the Homeless Migration into San Francisco.
If housing accessibility leaves impoverished San Francisco residents without a place to stay, the city must help them. But the city cannot help these residents if they’re competing with homeless people from other cities and other states for resources. A city only has so much tax revenue before they can’t afford their expenses. When that happens, they can’t support anyone. Therefore: in order to receive homeless assistance, you cannot have been a resident of another city or state within the past three years. Those people deserve help, but that help needs to come from elsewhere.
Estimated reduction in homelessness, YoY: 7-8%
Halt all importing and distribution of Drugs into the City.
Data is difficult to find on the fraction of homeless in San Francisco who are also addicted to opioids, crack-cocaine, or others. If you supply hard drugs to people, you’re hurting them, and you’re making this issue bigger, impacting the city’s ability to solve these issues. Zero tolerance for people caught importing or dealing fentanyl, crack-cocaine, and other highly addictive substances contributing to San Francisco’s homeless problem.
Estimated reduction in homelessness, YoY: 4-5%
Resources for the First Quadrant
For people in the first quadrant, who’ve passed a drug test, have no signs of addiction, and who do not need medical attention for chronic mental illness, shelter them immediately. Give them a cot and a shower and a door with a lock. Offer resources for job placement and classes on financial literacy. As this population is lowest risk for theft and crimes of violence, be more lenient with how long this population remains housed. Allow anywhere from six months to one year of fully sponsored housing in order to re-establish themselves professionally.
Estimated success rate: 50+% of this population, so long as jobs are available to them, should successfully transition to low-cost, more independent housing within 6-12 months.
Resources for the Third Quadrant
Offer access to the same housing facilities as the first quadrant after undergoing 14 days of drug rehabilitation at a city-sponsored rehab center, followed by 14 days in public housing where sobriety is monitored. Keep this housing separate from the housing facilities in quadrant 1, as this population is more likely to relapse, and crimes of theft and violence are more likely in this environment. After a full 28 days of sobriety, allow this quadrant to transition into First Quadrant housing.
Drug rehab programs have mixed rates of success. To begin, it’s reasonable to estimate 10-20% of this population would transition to First Quadrant housing without relapsing, etc.
Resources for the Second and Fourth Quadrants
This is the most difficult population to address. Depending on the severity of the illness (i.e. schizophrenia or BPD), it will be a lifelong battle. This population should first enter a drug rehabilitation program, and simultaneously should have access to physician care where treatment options can be discussed. Because housing stability is critical for drug rehabilitation and managing chronic mental illness, this population will struggle significantly should they relapse. Government-subsidized hospital care is necessary until this population is able to manage their mental illness. As this is extremely costly, funds saved from policy solutions 1 and 2 should be reinvested into these solutions.
Summary
The way I’ve defined, explored, and offered solutions to this issue is inherently flawed. It’s full of estimates and approximations that, at the margins, will not hold true. But the problems identified are real ones, and the city’s recent focus on where the homeless population can reside - on city sidewalks or elsewhere - is wholly unproductive.
The solutions above should accomplish two things: halting the migration of homeless people into the city, and pulling a significant portion of the existing population out of homelessness. It isn’t perfect, but it’s better than what exists today, in San Francisco and elsewhere.
Here’s to better problem solving in the future.
https://www.sfhsa.org/services/financial-assistance/county-adult-assistance-programs-caap/caap-benefits
https://californialocal.com/localnews/statewide/ca/article/show/6215-homelessness-crime-california/#:~:text=While%20unsheltered%20persons%20are%20statistically,of%20crime%2C%20particularly%20violent%20crime.
https://www.sfchronicle.com/projects/2022/san-francisco-sros/
https://www.hoover.org/research/despite-spending-11-billion-san-francisco-sees-its-homelessness-problems-spiral-out
https://sf.gov/information/rent-board-annual-eviction-reports
Loved reading this Dan, great analysis!
Great in depth statistical analysis with recommendations. If only a portion of your ideas got implemented the situation would improve.