Savings From Supportive Housing for the Homeless

There's been an 86 percent reduction in the incidence and cost of jail time in Knox County for formerly homeless individuals who've been placed in permanent supportive housing (PSH).

That is one of the most striking findings of an intensive study of the costs of homelessness conducted by University of Tennessee professor of social work David Patterson in collaboration with the Knox County Health Department. The study, titled "Comparative Costs and Benefits of Permanent Supportive Housing in Knoxville, Tennessee," compiles an array of costs incurred by a sample group of 41 individuals during a year of chronic homelessness, compared to the year following their placement in PSH. The study draws upon extensive data that Patterson maintains as overseer of the Knox Homeless Management Information System, but also entailed delving deeply into the physical and mental care, social services, and other costs of caring for the sample group.

The 41 individuals who consented to examination of their medical, incarceration, and other records are believed to be representative of the larger group of 246 individuals whom HMIS shows as residing in PSH at the present time. And Patterson further believes they are demographically similar to the much larger number (1,362 in a recent census) of people whom HMIS classifies as Knox County's chronically homeless population—at least 1,000 of whom he believes are candidates for permanent supportive housing.

So the reduction in jail stays from 626 days to 89 days, and in jail costs from $45,072 to $640 for the sample group, could be multiplied by six if extended to all 246 residents of PSH. And the resultant annual savings of $267,000 could be multiplied by another factor of four to a total of more than $1 million if all of the candidates for PSH placement were taken into account.

That is, of course, a very iffy proposition because such placement is dependent upon 1.) a supply of suitable housing; 2.) funding of rental subsidies that mostly take the form of federal (HUD) Section 8 vouchers dispensed by Knoxville's Community Development Corp.; and 3.) enough case managers to provide the support services that are considered crucial to getting and keeping the once chronically homeless housed.

Unfortunately, none of the above is presently available in Knoxville on anything like the scale needed to serve the entire population that could benefit.

The addition of 57 units at Minvilla on Broadway and 48 more through renovation of the former Flenniken School in South Knoxville have augmented the city's stock of PSH. But the proposed sites for development of anything like that size—especially in West Knoxville—have drawn a firestorm of opposition from surrounding residents and effectively put what was billed as the city's Ten-Year Plan to End Homelessness on hold.

Even if more acceptable housing sites could be identified, their financing and operation depends upon a stream of subsidized rental income. KCDC's president, Alvin Nance, reports that he doesn't foresee any increase in Section 8 voucher funding from HUD to create that stream. Already, there's a waiting list of 1,200 for KCDC's static stock of some 3,700 vouchers. And while there's a turnover rate of 300 to 600 a year, it's not enough to serve many more homeless applicants anytime soon. A recent boost has come from a new Veterans Administration voucher program for homeless veterans. Nance now has 25 of these to dispense, and he believes they will go a long way toward filling Flenniken, which only has 10 residents at present.

When it comes to case managers, new Mayor Madeline Rogero's budget continues funding two positions—one each at Minvilla and Flenniken. But County Mayor Tim Burchett has cut off the $50,000 grant that used to go to Volunteer Ministry Center for case management services at other locations—at least until it no longer allows alcohol to be consumed on their premises.

Patterson's study places the annual costs of housing subsidies at $3,531 per capita. So covering a much larger number of residents would obviously cost much more than the savings in jail costs, whether derived from federal or local dollars.

It would be nice to think that the comparative cost study identifies other savings resulting from PSH. But such is not the case. While inpatient mental health costs come down, this reduction is more than offset by increases in other health-care costs, including hospitalization, primary care, and (most surprisingly) emergency room services.

Patterson believes these increases are an anomaly attributable to the statistical phenomenon known as "outliers." "A few very sick individuals in our sample group accounted for a disproportionate share of the costs," he says. And the study goes on to note that, "Studies of post-PSH health care for individuals experience [sic] chronic homelessness in Chicago, San Francisco, Denver, Massachusetts, Maine, Seattle, Rhode Island, and Portland, Ore. found notable decreases in services and cost associated with PSH." All of these studies involved much larger sample sizes, which reduce the impact of outliers, he says.

In an appendix, Patterson presents results with outliers removed that show an aggregate cost reduction from PSH, including housing costs of $1,145 per capita. But the true costs of homelessness, as he well recognizes, should be measured less in dollars than in the degradation of human lives that PSH can help reclaim. And it's fortunate that Rogero has retained a holdover from the moribund Ten-Year Plan, Mike Dunthorn, to develop a new one. "Madeline has said we've got to find a path forward that prevents and ends homelessness in a way that's acceptable," Dunthorn relates. And I would add that it's also got to be affordable, which probably means incremental rather than transformative change.

Corrected: The spelling of Mike Dunthorn's last name.