A Numbers Game

San Francisco's biennial homeless count secures $18 million in federal McKinney-Vento funds for about 50 homeless programs in the city. The count has long been used as an indicator of how well the city is dealing with homelessness, which would all be fine and good, if it weren't for one small and inconvenient fact for everyone involved: It is a meaningless charade. I should know — I was one of the 500 people who volunteered to do the counting.

The city's method for counting street people guarantees inaccurate results. For instance, volunteers were told that under no circumstances were we to actually ask people whether they had a home. When asked why volunteers should rely on guesswork rather than simply talking to the people they encountered, Daryl Higashi, the city's supportive housing finance director, explained that there would be a follow-up survey of 500 homeless people that would apparently provide all the necessary additional information. He also emphasized that the volunteers should not disturb anyone's privacy, which seemed strange. Getting people off the streets involves communicating with them at some point. So why not now? 

After the training session, I got paired up with a nice guy in his 20s who worked at a social-service charity. We drove by a couple of people on 22nd Street in the Mission who had shopping carts full of stuff, but they were in a shadow, and it was too dark to figure out just how homeless they looked. My teammate seemed to think it was rude to slow down and stare, so we kept driving, unsure what to do.

The city's "guidelines" for counting people with shopping carts were fuzzy. While we were supposed to automatically count people sleeping on the street as homeless, people with shopping carts were left as a judgment call. My teammate recorded the people with shopping carts. Why? We were more inclined to count the maybes as yeses, because if these numbers were going to be used, even indirectly, to determine the services available to the homeless — the feds insist there is no direct connection between the number of homeless counted and the dollars doled out — we didn't want to risk leaving anybody out.

In the two hours we drove around our area, making sure to cruise through all the alleys and backstreets and parking lots, we encountered only seven people who looked more or less homeless. There may have been plenty more staying with friends, or hidden from view, or simply not looking homeless enough to make our count.  All I know is that we gave it our best.

In 2007, volunteers counted 6,377 homeless people in San Francisco. I have no idea what the final number will be this year, but from my experience last week I do know this: That figure will be totally wrong.