Rethinking “Imminent Risk of Eviction” Homelessness Prevention

I want you to research two statistics in your community:

1. The number of evictions in the last year.

2. The number of unique households accessing shelter in the last year.

Let me guess – the number of evictions is higher than the number of unique households accessing shelter, and it isn’t even close.

Many communities have financial assistance programs to assist households at imminent risk of evictions. The thinking is that through these efforts homelessness is prevented. It is commendable in theory. It is most often an incorrect use of resources in practice. Why?

Even most households that get evicted (not just at risk of “imminent eviction”) never become homeless. They figure it out. They transition to another place to live. It is bunk to think that all of these households end up as doubled up or hidden homeless.

This is one of the reasons why I think it is important to put “imminent risk of eviction” in quotes. The best and strongest indicator of whether a household will become homeless is previous homelessness. I think there are loads of households that face eviction. I think there are too many households that become evicted and that government benefits do not keep adequate pace with average rental costs. Nonetheless, drawing a straight cause and effect line from eviction notice to homelessness is precarious at best, and extremely naive.

Look at the data of unique households in a community that make use of eviction prevention (including rental arrears and utility arrears programs) longitudinally and you will see something interesting (which staff that operate these programs can tell you intuitively): whenever there is a time-out period for accessing the resources, the same households find their way back time and time again as soon as they are eligible to receive assistance again. What happened in the intervening period? Were they stellar tenants making all payments in between and coincidentally needed assistance as soon as they were eligible? No.

Another interesting thing you will see if you examine the data is rates of homelessness that occur when the pool of funds to assist households runs out in any given month or any given year. There is not a spike in the number of households that become homeless. There isn’t even a small blip.

Financial assistance programs like this type of eviction prevention assistance for those at “imminent risk of eviction” are a form of charity. They are NOT a form of effective program design or permanent problem solving. Most often they are a band-aid (and a lousy one) for structural issues related to poverty, and do not solve the permanent reason why the household needed the assistance in the first place.

Some cities have used Rent Banks for almost 20 years. The idea is that a household in need because they are at imminent risk of eviction for rental arrears borrows from the Rent Bank and pays back a nominal amount each month (say $20-$50) thereafter until it is all repaid. Unless the repayment is automatically deducted from some type of income benefit, the amount borrowed is very rarely repaid. Most often the money received is written off.

Furthermore, it is interesting to examine how communities structure their financial assistance programs for households at “imminent risk of eviction”. Most often the risk tolerance is very low. Screened out are very high need households with multiple, co-occurring factors. Screened in are very low need households with single risk factors. This runs contrary to how these programs should be considered. One would think that lower need households are more resourceful and perhaps have enhanced problem solving skills compared to higher need households that have more limited options, and perhaps diminished problem solving skills. Some communities even require that the household can prove employment income in order to qualify for assistance.

I would be remiss if it was not pointed out that there is another bias in the types of households that are assisted. Time after time in my travels I see “imminent risk of eviction” programs reserved exclusively for families. It is more rare to see assistance for single adults or unaccompanied youth – and when it does occur for these two groups the resource allocation is often considerably smaller.

Why do these programs exist? Hard to say with absolute certainty. I believe providers of these programs feel they are making a bigger difference than they are actually having by doling out the money. They can speak to the number of households assisted. They can speak to the amount of money allocated. They think this data shows how much homelessness has been prevented. It does not. Why? Because they do not have a control group of households that did not get assistance that they also monitored to see if they become homeless. If they had, they would find the ones that do not get assistance rarely become homeless. Again, they think there is cause and effect in the program where nonesuch exists. But it certainly is a feel good story to think homelessness is being prevented, and I believe that is the most compelling reason why organizations, faith groups, utility companies, etc. continue to do it.

So what should be done? I believe the first big step is to reallocate a whack of the money to more direct services like Rapid ReHousing and increasing vouchers and other supportive housing options. Then, with the smaller amount of money that remains, I would recommend it be used only with those households that have previously been homeless. That would be the primary eligibility criteria. While that may be sufficient enough, if you wanted to dig deeper (and maybe you do not) households with more co-occurring issues should be prioritized for assistance before others. Keeping people from returning to homelessness probably makes the most sense for “imminent risk of eviction” financial resources.

Iain De Jong

About Iain De Jong

4 Responses to “Rethinking “Imminent Risk of Eviction” Homelessness Prevention”

Read below or add a comment...

  1. Natalie McCleskey says:

    Thank you Iain. This is exactly where my thinking has been on the matter..thank you for explaining it so clearly.


  2. Keith Root says:

    Ian you have made very good points and prioritizing is a vital component. However, if better communication existed between community/private/faith-based/and governmental organizations existed then issues could be addressed more clearly. If an agencies wants to assist a particular group and states that is their mission that is fantastic i.e. (working poor-at risk-people needing temporary assistance-veterans) but making it clear what they hope to accomplish and sharing what assistance has been offered would greatly enhance everyone’s ability to be sure resources are being used appropriately and not allow program surfing (my word). Just a thought.

  3. Jeff Guyett says:

    Very provocative and clearly “myth-busting” perspective that is hopefully drawn from valid research and data analysis–which would be good to cite for all to see.

    In the same way that prevention cannot be linked as effective in actual homeless prevention without the control group of unassisted households, we would like to see the research studies that indicate prioritizing “previously homeless” households will indeed move the needle down on homelessness in the long term. It seems the data from our work illustrate to us the previously homeless are more likely to be homeless again in the future, with or without housing assistance.

    While the eviction prevention approach’s effectiveness in reducing homelessness may not be supported by the data, in our service area the local housing is so un-affordable that when eviction prevention and other assistance is unavailable, a significant number of those households simply move to a more affordable housing market–and look for assistance in that new market. Ironically, in that move, they have often become homeless, as well.

    Please share the data that inform the prioritization of “unsheltered” and previously unsheltered, for as the field becomes more evidence-based, us skeptics need to see the documented research to demonstrate the prioritization, especially when big, provocative pokes are made at the balloon of common thinking and prior practice. As another perspective, these prevention programs exist because at some point the data seemed to indicate prevention as an effective means to reduce homelessness, not just based on a “feeling.” If the data say otherwise, we welcome the opportunity to change the approach–just want to also look at the ongoing affordability of housing through a long-term solution lens, as well.

    Jeff Guyett
    Executive Director
    Community Action Center
    Pullman, WA

  4. Iain,

    Exactly right. Shelter folks are rarely coming from leases. They are doubled up or the like.

    If prevention should exist it should probably not draw on homelessness dollars. HPRP already showed us how to give away over a billion homelessness dollars to people who would never have become homeless (and how to then use that data in, um, creative ways to justify spending more.)

    Interestingly however prevention is exactly the homelessness question where we actually have peer-reviewed and published research (I learned at NAEH15 that SSVF will publish shortly). Here is the Shinn/Greer screener that, far better than humans without such a screener, predicts whether a family or single person will enter a shelter after seeking prevention services. However, it’s still a rough art and one has to choose how severe a score qualifies and doesn’t qualify for assistance.

    Screener here:
    Journal article here (they have tested the work in CA since publication and the results are reflected in the screener):

    If, for better or worse, you are implementing a prevention program, and not using Shinn/Greer, you cannot claim to be using evidence based best practices.