America's Real Homeless Hotspots
The one upside of the uproar in Austin over using homeless people as wireless hotspots at this year's South by Southwest Interactive festival is that it's focused a bit more attention on the very real and very disturbing tragedy of homelessness in America.
An estimated 600,000 Americans are currently homeless, including nearly 70,000 veterans, according to the recently released The State of Homelessness in America from the National Alliance to End Homelessness. That's a small drop-off from 2009, but U.S. rates are alarmingly high: 21 homeless per 10,000 people across the country.
The spread of homelessness is not uniform. Some cities have been hit much harder than others. To get at the geography, my colleague Zara Matheson of the Martin Prosperity Institute (MPI) mapped homelessness across the 100 largest U.S. metros. It's based on data covering the period 2009-2011 provided by Pete Witte of the National Alliance to End Homelessness (click here for the metro-level data on homelessness).
These are the homeless hotspots we all should be concerned about.
As you can see, the metros with the highest rate of homelessness are Tampa, New Orleans, Fresno, Las Vegas, Honolulu, Los Angeles, San Jose, and Seattle.
With the help of my colleague Charlotta Mellander, I took a quick look at the factors that might be associated with the rate of homelessness across metros, including crime, weather, and economic conditions. As usual, I point out that correlation does not imply causation; other factors that we didn’t take into account might also come into play.
Still, our analysis undercuts some well-established myths about homelessness and may well provide a starting point for future discussions of the topic.
Not surprisingly, housing costs play a role in homelessness. We found a positive correlation between the homelessness rate and the cost of housing (.37).
While the popular impression is that homelessness breeds crime, we found none. In fact, we found homelessness to be negatively associated with crime (with a correlation of -.33 to property crimes and -.27 to the overall crime rate).
As for economic conditions, our findings were mixed. The main economic factor that plays a role in homelessness is unemployment (with a correlation of .36). But we found no correlation between homelessness and productivity, income, human capital, poverty, or inequality, and just a weak positive correlation between homelessness and wages.
We also found no correlation between homelessness and the share of African-Americans in a population, while the correlation between homelessness and the percent of Hispanics was positive (.34).
It's commonly thought that homelessness is a problem of larger cities and metros. We found a modest correlation between homelessness and metro size (measured as the log of population, .21). The report notes that the rate of homelessness is only slightly higher in the most populated metro areas (22 per 10,000 people) than it is nationally (21 per 10,000 people). New York ranks 13th, Washington, D.C. 21st, and Chicago 75th.
Unsurprisingly, one of the strongest correlations of all was between warm winters and homelessness. The correlation between homelessness and mean January temperature is .47, the highest of any in our analysis.
Homelessness affects far too many Americans, but it's a mistake to conflate it with crime or see it as a breeder of other social problems. It is not the exclusive province of large urban centers, but affects cities large and small, rich and poor alike.
Top image: Reuters/Lucas Jackson