About 11,600 people are experiencing homelessness in King County, according to the latest point-in-time count.
About 11,600 people are experiencing homelessness in King County, according to the latest point-in-time count. City of Seattle

New stats released by the U.S. Department of Housing and Urban Development (HUD) this week showed that the national count of people experiencing homelessness increased for the first time in seven years. Those same numbers showed that King County has the third largest homeless population in the country.

But an analysis from a national homelessness advocacy group argues that HUD's data may be underestimating the problem. In fact, the numbers may be between two and 10 times higher, according to a report from the National Law Center on Homelessness and Poverty, which does advocacy and litigation on housing and homelessness. The report, titled "Don't Count on It," details the way HUD counts the homeless is "severely flawed" and can result in a "significant undercount of the homeless population."

Like King County, cities and counties around the country conduct one-night counts of people living in shelters and on the streets. They report those numbers to the federal government, which are used as both a snapshot of the problem and a benchmark of whether local efforts to address homelessness are working.

But the numbers are imprecise. HUD releases guidelines on how to do the counts, but those instructions change from year to year and are not always applied in the same way by different jurisdictions, according to the report. Beyond methodology, there are people experiencing homelessness who won't be counted. Visual street counts are imprecise, especially as people sleeping outside seek more secluded places to avoid police. And HUD's definition of homelessness is "narrow and does not measure the real crisis," according to the report. The counts do not include people staying with family or friends or people who are homeless, but in jail or the hospital. According to the report, when Houston did an "expanded" count and included people in county jails who were homeless before arrest, their total homeless count increased by 57 percent.

According to the report, a national 2001 study used records from service providers and other data to create one-day, one-month, and one-year estimates of homelessness. That method found that the total number of people experiencing homelessness could be between 2.5 and 10.2 times greater than the point-in-time count. The report recommends that HUD create a "more rigorous methodology" for its count and try to include people who are in jails and hospitals and those staying with friends or family.

This year, King County revamped its count methodology to cover more area and include surveys with people experiencing homelessness. The count did not include people in jails or hospitals who reported being homeless before they were booked or admitted. But it did include some people who were being released from the emergency room and had no place to go, according to Mark Putnam, director of All Home King County, which oversees the count.

The report concludes that undercounting homelessness could hurt our attempts to solve the problem. "Legislators frequently rely on the results of the counts to determine whether public policies are reducing homelessness," the National Law Center on Homelessness and Poverty writes in its report. "Rather than understanding that the [point-in-time] count represents only a portion of the homeless population, many interpret the count as a comprehensive depiction of the crisis and rely on it to inform policy design and implementation decisions. This can lead to policies that fail to address the homelessness crisis or may even exacerbate it."