Corrine Livers was the trial lead for a study that explored cash assistance payment schedules in the Temporary Assistance for Needy Families (TANF) program, in which participants receive monthly cash assistance if they continue to meet requirements for seeking employment. The trial was designed to test whether weekly payments would be more beneficial in helping clients meet financial needs related to obtaining employment as compared to the usual monthly payment schedule. Participants were randomized to receive checks either weekly or monthly. Ms. Livers shares her insight on how to collect data, working with teams in multiple locations, and continuing to innovate in the public sector.
What inspired the idea for the trial?
One of our obstacles is that participants have to wait until the end of the month to receive the first payment, so we thought, what if we gave the payment more frequently so they don’t have to wait until the end of the month? The idea stemmed from, can we get payments to participants more frequently, and the second idea was, can we keep them in the program and are they more successful because they don’t have to wait between payments?
For our weekly participants, once eligibility was determined, we processed payments so clients received payments more quickly and more frequently. By receiving a payment quicker and more often, we wanted to know if the participants would be more successful in the program, if they withdraw their initial application less frequently, and if the weekly payments would better meet their needs.
Did you think it would be feasible to test?
BetaGov came to the department and gave the presentation; I wasn’t at that meeting, but my administrator was there and we threw down a bunch of ideas for a trial. This one came out as a more feasible study to test; that’s what the appeal was, was that we had the opportunity to run a good trial and to test it.
What were some of the challenges you encountered with your trial?
I think our biggest challenge was that it was a very manually intensive project for the weekly payments. So if somebody was in our weekly payment group, our workers had to go in every week and manually authorize the payment, because of the cost to use a software program. We already have pretty heavy caseloads, so to ask them to do this was a lot of extra work, and they were troopers about it.
How did you rally the troops?
It wasn’t that hard. We used two sites: Cheyenne and Casper, and each site chose a little different method. Our Casper site, the supervisor took on any case that drew into the weekly group, so one person did that. At our Cheyenne site, the workers kept their cases. We’ve been lucky to have staff who are willing to try different things to make programs successful.
I asked them to hang in there for six months. They’re great. They have these jobs because they care about people, so it doesn’t take a lot to get them to do something if in the long run if it’s going to help the people whom we serve.
How did you collect the data?
It was horribly manual. We put together a shared Google spreadsheet. Everybody had access to it; we set up the randomization code for whether a client got put into the weekly group or the monthly group. If it fell into the weekly group, there was another spreadsheet that tracked the case; the caseworkers had to track the case, and if it closed, they had to document what the closure reason was.
Did the results surprise you?
They did. I did not expect it at all. I thought our participants would prefer to have more frequent payments because they could meet their weekly needs. We found out they depended on the one larger check for rent.
It was the same amount of money, just a different payment schedule. Typically rent is due at the first of the month. We wondered, what did they typically use the money for? If they used it for household expenses, maybe [the weekly payments] would have been more successful, but the majority used it for rent.
What kind of feedback did you receive from participants?
I had the workers document some of the comments, and that’s how we learned what participants spent the money on. The majority said they can’t afford to pay their rent unless they got the monthly check.
What impact has your trial had on the policies and practices at your agency/organization?
So obviously we know we aren’t going to have weekly payments. I think that’s great because we can take that off the table and move on. I still want to work on moving payments to participants quicker as opposed to waiting until the last day of the month. I still think that’s a viable option, but at least we know we don’t want to break it into small payments. And we are moving to an electronic payment card, so I’m pretty thrilled about that.
What are your thoughts on innovating in the public sector?
This trial was so important because it put us on our current track; and when you test something you at least know where to direct your future efforts. We know now what works and what doesn’t.
I like looking at big numbers and seeing what kind of impact you can have. It made me look outside of the box. I participated in a certification program at the Cheyenne college to get a certification in public management. For my capstone I did a study to look at the DFS staff doing the first individual responsibility plan. We also studied unproductive billable hours for our services. Our workforce services with whom we are contracted were putting people into the work programs and services before they were determined to be eligible, but we were still billed for the hours. We significantly reduced unproductive billable hours and last month that number zero. We have since already done additional studies based on the skills and techniques we’ve learned from BetaGov.