According to a recent HUD report (pdf):
Khimm considers the possible reasons for this decline:
Over the last 10 years, the federal government has funded more than 100,000 housing placements for chronically homeless Americans under an initiative called “Housing First” that President Bush began early in his presidency and that President Obama continued, according to Dennis Culhane, a social policy professor at the University of Pennsylvania. State and local officials across the country have now embraced such strategies.
During the financial crisis, Obama redoubled efforts to combat homelessness, putting $1.5 billion in the stimulus for programs like “rapid re-housing,” which moves people out of shelters and into permanent housing as soon as possible. With bipartisan support from Congress, Obama has vastly expanded housing vouchers for homeless veterans, whose numbers have dropped 24% since January 2010.
Plumer worries that this trend will reverse:
There were huge increases in homelessness in some cities last year. Los Angeles saw a 27 percent jump in homelessness in 2012, while New York had a 13 percent increase. (New York City’s homelessness is now at levels last seen during the Great Depression.) Those two cities alone account for nearly one-fifth of all homelessness in the nation.
The Department of Housing and Urban Development has also warned that budget cuts from sequestration could further strap various housing programs. This year, the agency is set to announce a 5 percent cut in federal funding for programs to help the homeless, due to budget constraints.
Danielle Kurtleben casts doubt on the report:
“I think the numbers are questionable,” [Maria Foscarinis, executive director of the National Law Center on Homelessness & Poverty] says. “I don’t think that they can be viewed with a lot of confidence.” Foscarinis characterizes the government’s total count of the homeless as “far off,” though she is hesitant to speculate as to what the real count might be, only that it is much higher than the government says.
A number of aspects of the report make her doubtful. One is the department’s method of counting – having localities choose a night in January (when the cold is most likely to send the homeless to shelters) and conduct their sampling. This means missing the total number of people who become homeless in a year, she says, favoring instead a “point in time” estimate. In addition, different cities might conduct their counts differently, Foscarinis says.
A HUD spokesman says the department stands by its numbers, though he adds that even the number of volunteers available to count can make a difference in a city’s estimate.