In order to complete my Masters of Public Administration, I authored a graduate thesis focusing on the impact of a short-lived subsidy program called the Advantage program, which was particularly designed to target families who had been in shelter for several months. The average shelter stay in NYC family shelters at the time was nearing twelve months (a pattern which has continued).

If you would like a full copy of this thesis project, or would like to discuss this research, please email me.

Executive Summary

The pervasive undercurrent of Bloomberg-era rhetoric on homeless individuals implied consistently that the homeless system in New York City was of a high enough quality to induce people to enter the system who might not actually require shelter (Secret, 2011). This argument was most-often cited in the context of policies that provided subsidies to homeless individuals. The argument went that giving homeless people preference for Section 8 vouchers induced families to claim homelessness to move to the front of long Section 8 waiting list lines. The argument was echoed again when New York City instituted the Advantage rental subsidy program, providing two years of rental subsidies to families who had been homeless for 90 days or more. Advocates lambasted the Advantage program for a different reason: they felt it lasted too short a time to give families a shot at self-sufficiency, or that families were very ill-equipped to gain the types of jobs that would enable self-sufficiency, no matter the subsidy’s time frame.

This article examines these two enduring criticisms about the Advantage program and seeks to identify whether similar short- or medium-term subsidies could be effective in reducing the shelter census. First, research suggests that individuals may be induced into the shelter system by the promise of subsidies. However, research shows, that the situation is far more complex than a simple increase in shelter census as a result of subsidy availability. The level of system abuse and the ability of people to enter shelter when they are undeserving is overstated. Second, the existing literature and further research suggests that placements into programs like Advantage do reduce shelter census over time and in all likelihood do reduce homeless recidivism for a portion of those who receive the subsidies. Third, it is clear that unless New York City wants to rely on the whims of the increasingly stagnant efforts of the federal government or the very unlikely event that the Callahan vs. Carey Consent Decree (1981) will be overturned, the city must have a solution to moving individuals out of shelter, lest it risk rapidly rising shelter populations such as what we have seen in the months and years since neither Advantage subsidies nor Section 8 priority vouchers were available to shelter residents.

The question in this context is not whether to provide families a clear path from shelter but rather how. However, there are considerable gaps in our knowledge of whether short-term subsidies will, at the term of their expiration, lead the recipients back to homelessness. The same is true for how much costs to the city increase when a subsidy program is available. Evidence supports the idea that providing subsidies for those already living in shelter can raise costs for the city shelter system, not in terms of increasing the total census but rather in terms of 1) increasing the length of stay for those already in shelters, and 2) increasing the number of applications that must be processed (but which are not necessarily found eligible).

Despite the potential risks involved in providing short-term subsidies to homeless people, based on analysis subsidy placements appear to be an effective way to reduce the homeless population. In terms of cost-effectiveness and sustainability, subsidies can be better designed than the Advantage program. Subsidies should be available to homeless families in the New York City shelter system earlier in their stay. Subsidies should not be designed as a one-size-fits-all approach, but rather should be more carefully evaluated on the basis of the family’s barriers and reasons for entering the shelter system. Potentially, families could have some options to assist them with exiting shelter that will allow them to self-select between a few types of assistance.