Modernizing LIHEAP: Rob & Roger Pavey on the Future of Energy Assistance

0:01Hello everyone and welcome to the Fortell podcast and today I'm joined by Roger Pavey the president and CEO of Illinois Association of Community Action Agencies. Now Roger spent over 30 years working in community action and around 25 directly involved with lie including administering the program at a local level both in Illinois and Iowa. So I think that gives him a really interesting perspective on the conversation that we want to have today because you've been seeing light people on the front front line from the CEO's chair and now from a statewide policy position. So Roger, thanks so much for joining me. >> Thank you, Rob. Yeah, it's great to be here. Actually, my experience goes back even farther than that because one of my childhood memories is having our power

0:41disconnected. My mother attempting to cope with that in the days right before lie became a reality. This would have been in the 1970s. And I remember sleeping under the kitchen table. She had made a tent for it with a blanket and ran an extension cord to the neighbor's house and tried to keep us warm using a toaster by pushing the button all night, which is a terrifying and scary thing. Although, when you're, you know, 11, 12 years old, it's also okay. >> Yeah. Kind of an exciting experiment. What could go right? Yeah. >> Yeah. Right. Yeah. >> But no, that's a great story about your mother. And so you came to this from a really personal perspective, right? I mean, h how did you get into community

1:21action? Maybe let's start there. >> Yeah, that's that's also part of the story. Uh you know, my my community action experience even goes back farther than that. It was a Head Start kit in 1971. The Head Start program was created as part of the war on poverty. It's older than the LI He program. It's one of the original community action programs and started in 1960s, mid1 1960s. So, I was in one of the early cohorts of children who benefited from Head Start. Had a large family. I'm the only one that had the opportunity to go to Head Start. And I think that really gave me a a push that I needed at a very young age. So, I I uh ended up in foster care and that was a horrifying experience until it wasn't. I I was fortunate after getting shuffled around

2:05a little bit, I was fortunate to end up with a great set of foster parents who decided I was going to go to college, even though that surprised me when they told me that. And then I, you know, when I graduated, I went to work for a water company, a municipal water company, and just couldn't do it. You know, I needed to do something that I felt like I was giving back and and Community Action was the place that felt like home. Yeah, that's really that's really interesting that you took what probably was quite a safe normal job and it just felt completely wrong, >> right? Yeah. >> Yeah. I think a lot of people that we interviewed feel like that for sure. And then and to go back to what lie he is, I think you probably have to from the

2:47reading I've done, you have to go back to the 1970s energy crisis. Right. >> Right. which is all around sort of after the OPEC oil shot, government started to experiment with assistance for households who were struggling with a rapidly rising energy price, right? Which I don't know if you know people say that history doesn't repeat, but it rhymes. There's certainly some resonance today, right? But take us back there. Why why did the federal government at that point decide that they needed to step in? >> Yeah, it's a really a story about utilities in the United States in general. We've had cheap utilities up until that point. Utilities um were not a thing that put a strain on household budgets because they were quite low as

3:27you know as a household cost for so many decades. And when when the OPEC crisis hit, I I was in elementary school and so it it was visible throughout everyone's life. Jimmy Carter went on television wearing a sweater telling everyone to turn down their thermostats. my lunchroom in the school. They turned off the lights. We ate in the dark as some kind of energy saving measure. You know, it was a very visible time and it it really brought utilities into the public dialogue as an affordability issue which has echoes today of course and it gets to Engle's law of economics which says that you know as a household's income

4:09increases while it might spend more money on necessity goods as a percentage it's less. And and so you know the utility prices when they rise if utility prices go up 10% let's say um for a low-income household you might only be able to adjust your usage by 2%. So it compounds the affordability problem and um it it has lately been pressured by multiple fronts. So the OPEC crisis was you know an international relations type of an event. We have that now with the the war in Iran, but we also have data centers that transition to clean energy. There's so many factors going into affordability of utilities today.

4:51>> Yeah, for sure. And obviously, you know, being a company that thinks about artificial intelligence and its data usage. It's one of those questions we get at conferences all the time and there's not an easy answer to it to be brutally honest which is I don't think anyone's putting the genie back in the bottle when it comes to AI and we think we've got an interesting use case but definitely you know some of my family saying thank you every time to chat GPT when they ask the 18th time about their pension plan probably isn't the best use of uh of our country's energy but but just just go back again so so the oil shocks happened there's a there's a weatherization assistance program, right? Arriving in 1976, I think. >> Yes.

5:34>> And which is the predecessor to the energy assistance program in 1980. So, and then in 1981, we come up with what we now know as LIE, right? So, >> that's it. Yeah. >> I I assume there's a lot of wrangling and behind the scenes, but just just take us back to what was lie then and maybe what does it become today? >> Yeah. Yeah. That's that's a that too has a lot of layers to it. uh you know in in the 19 late 1970s and early 1980s there were a lot of policy issues around energy that were out there including price fixing. You know like we had government locking down prices on gasoline and utilities and the the really the genesis of this was out of

6:14the northeast where households tend to still heat a lot with uh fuel oil and so more direct tie to petroleum you know like it's not a step or two removed. So it came out of the northeast and initially it was well we'll just you know do the conservation approach but it became clear that that is more expensive. It still is today to weatherize a home costs more although the savings are more durable but it costs more in any given year than giving someone some assistance with their utility bill. So Congress created the forerunner to lie heap but really lie hasn't changed much since 1981. Uh, and I think we're at a point where it's going to need to change to be enduring. If you applied for lie in 1981, 1982, you went in and you filled out a paper

6:58application that listed who was in your household and what your income is and you provide your power bill. Same today. You might do that at a, you know, an agency and they type it directly into a computer, but it's the same questions that have been on a paper form forever. Really hasn't modernized, >> right? And the interesting thing with that is also because the world has well I think the world's fundamentally changed right so in 1981 it's largely about heating costs and energy shocks right but but we think energy poverty is a different problem right in 2026 >> yeah I mean the the climate impact here has really brought cooling to the forefront from an empirical standpoint

7:38but maybe not from a political standpoint. We're still using an old allocation that favors the northern states and favors, you know, the winter season. That is in part legacy politics, the coalition behind lie heap, but it's also in part just a function of the calendar. The federal budget starts on October 1st and lie has almost always operated as a first come, first serve, get it before the money runs out kind of program. >> Yeah. Yeah. For sure. And I think there's that's that's one of the things we've struggled with, right? because we're we're taking all these phone calls and you essentially create a wait list for the next next round, >> right? >> And so and there some agencies decide,

8:20okay, we're going to spend a certain amount of money per month. some let's say if you're in Dallas or you're in somewhere else in Texas, you get this enormous spike January 1st and then it kind of you know there was one agency we spoke to that took 20,000 calls on the first day of SEAP which is their lie heap down in down somewhere in Texas and it just melted their whole phone lines. Right. So this it's it's interesting though because I think maybe before we say about what we want to change or what is potentially broken about it, what has it actually got right? because I mean the scale of it is pretty phenomenal, right? It's it's 6.7 million households. It's $4 billion. You know, what are the things that it does get right? >> Yeah. Well, it does make energy

9:04affordable by realizing that uh energy costs are regressive. You know, they they just are they're regressive. Every time you have an increase, it's going to hit a lower income household harder because they have no discretionary income than it would a moderate or higher income household. So, it does get that mechanism correct. Even though it's $4 billion, the best estimate that we have out there is that it only reaches about 16% of the eligible households nationwide, which is alarming. And so, you know, I'm I'm working with a couple of colleagues to research why that is. And we think there are five mechanisms that that we're testing. There may be more, but the five mechanisms are administrative burden. And that's where you, you know,

9:46really focus is to try to reduce that administrative burden. Administrative burden can be broken down into compliance cost. I have to have all this bureaucratic paperwork. A psychological cost. I'm taking government assistance. It might be shaming, especially if I have to go into an office and wait or try to, you know, schedule to get an appointment. Uh learning cost. I don't know what the program is or does. So that's psycho. That's the administrative burden. A lot of people just don't know about HEAP. That's another mechanism. There's a policy feedback loop. If I tried to apply but I couldn't get through because there was 50,000 people trying well then maybe I'd learn that this just is never going to work and that that travels through social networks. And then the last two are really fascinating and haven't been studied. One's supply side rationing and that

10:28says that that uh you know the demand follows the supply not the other way around. So if there's only so much money then only so many people are going to apply. More people would apply if there was more money. that 16% is artificial, in other words. Um, and then finally, it could be a rational choice. Households could look at this and say, "It's not for me in a rational way." So, we're we're going to analyze all of that by essentially identifying people that research has never identified before, and that's non-applicants. We're going to cross compare databases to find people who applied for Head Start or workforce innovation that didn't apply for Life and find out why.

11:09It reminds me a little bit of the the work that LA County's been doing around housing and homelessness actually, which is essentially that that what they were doing is building an algorithm to predict which people are most likely to become homeless in a period of time and then build an outreach mechanism to get in touch with those people and see if they could prevent them from eventually needing homeless services. And so in your in your big five that you just discussed, obviously, yeah, you're right. we're focused on on the administration cost and just to all or the and when I say cost I mean the sort of mental cost of of >> yes >> I mean one of one of the fascinating examples is in one of the states we work

11:50where someone has sent a often it's a paper application or it can be online but that you have to write a letter to that person and they have to resp if if they've made a mistake and you have to respond to that letter in 10 days but the problem is the letter will take four to five days to get And so while I think the people who built that probably were well-intentioned and well actually I've met a couple they are well-intentioned clearly that that is not fit for purpose for a number of reasons right because it just is so difficult for someone to actually correct something that was a mistake but if you were to talk and I don't want you to hypothesize too much you're going to actually find out maybe you can come back and tell us but which of the ones that you've just mentioned do you think

12:31are the biggest barriers for only 16% of people are eligible taking the benefit? I I think supply side rationing is the one that I'm chasing the hardest at the moment because it it has a it has a policy solution to it. >> Mhm. >> You can lower the amount you give to each household until the point where you send a price signal to that household that it's not worth it >> right >> to apply. So, you know, like where is the sweet spot for a state to set benefits? That's that's a question that we want to try to figure out. And then how do you layer other utility assistance programs in with lie heap to keep keep it worth it for the household? Um, for example, getting community solar

13:13for a household that gets lie or low-inccome discount rates, which I'm sure we'll talk about in a moment. If you can stack those three things, then maybe you can reach more households and stretch that lie heap dollar a little further. And then of course administrative burden. And administrative burden is not all bad. There has to be some level of administrative burden because this is public funds and we got to treat these public funds. Well, those are the two that that I think get shorted in the literature quite a bit. I just don't think that they that >> has there ever been in another program any any examples where let's say they've you know increased funding by 10% or they doubled the funding and you've seen a massive increase in the level of people who were was able to help

13:54>> as we just went we just went through a quasi experiment with that through co right we got all of this one time money and applications went way up >> right so >> that's true that's true >> we have a quasi experiment right there >> yeah well especially in rental assistance was the big one there, right? So, >> I remember talking to an organization in Pennsylvania who had literally all these dollars and yeah, it was a one time and clearly they get the same number of u maybe you know they still get the same number of phone calls but they aren't able to help in any >> in any degree. So, so what we're saying there, okay, but go back. So, if you're if you're on the right, and I don't want to talk politics too much, but if you're on the right of of politics, what would

14:36be the benefit to saying we're going to double lie budget next year? >> Yeah. So, that I think that gets into issue networks like who who does LI heat benefit in the end, >> right? Yeah, >> it benefits the household, but it also benefits the utility provider and all of the rateayers because if I get disconnected for non-payment, all rateayers are paying that cost. >> And and so through our work when we were advocating on tariff, on rate increases for utilities, we did a lot of work on bill signaling. If I'm a low-income household and you give me an unaffordable utility bill, let's say my get my open my none of us know it's a

15:17trust market. So, you open your utility bill. You don't know what it's going to be until you pull that envelope out or you open that email. You might have an idea. It's been hot or it's been cold, but you don't really know. So, you open it up. I have $100 in my bank account and the bill is for $500. Well, bill signaling tells me like it's futile. I can't pay that. So, I just throw it aside and don't worry about it today because they're not going to shut me off today. I'll figure it out. But, if I open that bill and the bill is $50, it's suddenly affordable and I'm more likely to pay it. And we we were able to statistically model that using utility data to show that yes, people will pay what they can pay. And I and I think that's an argument that works on a bipartisan level that that sensible utility assistance levelizes the cost

16:01not only for the low-income household, but for all rateayers. It's going to levelize that cost down. >> Yeah. And that's that's exactly the same argument that they were making in in the in the homeless pilot. So it's it is way cheaper to prevent someone going into this homeless you know coordinated entry program than it is to give them a lump sum or a certain amount of money. Now the hard thing for them actually was not so much predicting which people were likely to to become homeless but the outreach associated with that and that some people to to your earlier point did not want to be contacted and there was a shame associated with falling into a group that was going to going to go into poverty. >> Yeah. Yeah. And that that shame piece uh

16:44you know like that gets into the dignity of the process. And you when I when I fill out my FAFSA for my kids to go to college or I do my tax return, I don't need to gather a bunch of stuff and go sit in a government office, take a number, wait for two hours, and then be told, "No, you got to come back. You didn't bring quite the right thing." >> I have a more dignified way to to do this. >> Lie is still operated mostly the way it was in 1981. you're gonna go sit in an office and wait or be on hold on a phone forever and you'll have multiple trips that you have to make. >> Yeah. And that's what your your point about the administrative process there needs to be some hurdle to to it. But um I mean again when we're when we're doing

17:27it in Tennessee for example 50% of the phone volume is status update. So it's >> am I getting my money? Has it happened? did you get my you know it's those basic questions and because people are in what what's difficult to explain when I've talked to policy makers about this is if you're in a crisis you're not thinking as you are sitting in a nice desk when you've had your ham sandwich and you've patted your dog on the head right it's like you're you're going in a circle and it's very difficult to rationalize and so people will call multiple multiple times to check something and they're on hold and you know it it is it's difficult so the bit that's interesting thing for me is lessening the burden

18:08somehow by giving people the appropriate information at the right time and then your point I think is if you allow the supply side if you allow people to believe there will be money available when they would need money it would be a more successful process. Do have you worked on almost means testing so you know slightly rich families might get a little bit more and you know families on the edge might get a little less or or multiple times. >> Yeah. Yeah. We are uh we designed our low-inccome discount rates to to give priority to lower income households to you know like better levelize the bill. >> Just explain just explain that as a whole as from the beginning. >> Sure. Yeah. So a household that is at you know 0 to 50% of poverty very very

18:49poor is going to get a better discount a higher discount on their power bill. So they have less to pay than a household that sits closer to the poverty line or closer to the cutoff point for lie heap. So it's graduated in terms of the level of assistance. Now as we reduce administrative burden, there is some research out there that um this is also nent, but there's some research out there that says that when you reduce administrative burden, then you're going to overcentivize the people that are higher levels of income applying that they're going to appear more in the application process because again, this is costly. You know, you hinted at this earlier. I tell people applying for lie

19:32isn't free. It's not free. You might not have to write a check, but you're going to do a lot of work to get that lie heap benefit. So if I'm closer to the cutoff point, I've got a job and I'm work, you know, like I can't take time off of work and I'm not going to get that much lie heap anyway because my income's a little higher, well maybe I don't apply. But if we make it easier to apply, maybe those households then are over represented in the benefits. So it's there is a balance with all of these uh these metrics that we're looking at in terms of you know like where do you put things I'll give you an example from the pandemic in another area another area I was at an agency that was doing personal hygiene items you know the pandemic breaks we can't find toilet paper diapers everyone's you know stressing over the supply chain. So, we ordered semi-truck

20:15loads of stuff, you know, diapers, shampoo, like we ordered as much as we could from everywhere and filled up all our offices because our staff were working from home. We created a little mini warehouse essentially. And uh everyone got the same thing. So, you filled out an application and everyone got a standard box and you know I'm in the office with a few staff spaced but we've got a little conveyor belt and we're filling these boxes and a guy wraps on the window and motions for me to come outside and then just rips into me for giving a bald guy who lives alone shampoo and maxi pads and wasting government money. He had a great point. So, we then created a website that looked a little bit like a Walgreens site and the only way you could shop on

20:58it with was a gift code. You couldn't use a credit card or pay cash or anything. You had to have a gift code. Turning that whole process into a shopping experience. So, we removed the administrative burden and we put choice into there. You applied online. If you were qualified, you got a gift code and you could go shop and you get your items and we would pack them and get them to your house. Great service. It did all of the things that you mentioned that lie heap doesn't do. You got an email that said like, "Thank you for your order and here's your order and here's your status." And then it would they get another order saying your order has been packed and will be delivered on Tuesday. And then you got another email saying it's on your front step. So it did all of the things that closed the loop and it crashed within two days because

21:38everybody wanted it. You know, like there were plenty of qualified households that suddenly decided this is worth it. So where is the sweet spot there is a great question on administrative burden. We we really worked hard to remove the administrative burden. Did we go farther than we should have? I don't know. I don't know the answer to it. >> It's a really interesting point. I It's just reminded me of an article I'm sure you've read which is it's out of the University of Bloomingham. I think it's by Graph and Perog. It's called the red tapers is not so hot. >> Yeah. Yeah. >> Yeah. So So I'm I'm paraphrasing. It's something like they they were they were doing asset

22:18tests right so where they were reducing like heat participation just and disproportionately excluded the very lowest income households and increase the administrative cost right so >> the estimate is basically stating that so using asset tests in 2014 removed them the average state could have served approximately 9,000 something something like that with with heating assistance rate so so if I was if I was to now phrase that as a question so The research is basically suggesting that lie asset tests actually increase admin costs while excluding the poorest households. Right. That's what that's what that's talking about. >> Yes. Yeah. And and not very many states use asset tests in part just because it adds complexity to the process. But the

22:59other reason is that that very reason right there though is incomebased and most states bizarrely use 30-day income, the most recent 30 days. And we know low-income households have a lot of income volatility, but so do other households. >> True. So if I'm a farmer and I apply in February, I probably qualify. >> Right. Right. Right. That's super interesting, isn't it? Because that's one of the things that we find from an admin perspective one of the hardest things because what we're trying to do is we're obviously doing the phone call, but what we're interested in now is getting the documents and then using an algorithm to look at the documents to say in real time, is this person eligible or not? And if they're not, if they haven't provided two jobs, right? We try and tell them

23:43while it's happening, but that like actual income is very difficult because of Exactly. because of volatility. >> Yeah. >> Yeah. My my goal, and I don't know if I can get this to happen here in Illinois, but my goal is to do what the FAFSA does, the free application for federal student aid to create a connection for that LIE HEAP worker to the Illinois Department of Revenue to pull their adjusted gross income. >> Yeah. Because that, >> you know, gives fidelity, but it reduces burden. Fidelity and burden both. Yeah. >> Right. Yeah. Yeah. Yeah. And the and the nice thing about that is it's it's pinging a database, finding a number, and not putting the burden so much on them because,

24:23>> right, >> the the the paylip problem, you know, you can get an OCR to read it from a technology point of view, but actually for someone to collect all the right paylips and not do three months back or the month before, and also there's this other thing because it's household income, right? So, what does my significant other earn? Well, she's a music teacher and she had a bad week. And what did I earn? Well, I work construction and you know, we got a big job here, but we didn't have it over there to your farmer analogy. I suppose it's one of those things. It's like it's an imperfect system that currently works and we're thinking, okay, how can you how can you improve that? When you're talking to policy makers, let's talk about Illinois. Are they is this an obvious thing that they know is an imperfect system and they're interested in changing it or is there do you have

25:05to whack them over the head with these types of studies to for them to be interested in trying to do something different? >> Yeah, I think my answer now is different than it would have been five years ago. Five years ago it was a lot of just obscure got to get their attention uh you know why would we want to deal with this kind of thing. It was not on the you know it was not an open policy stream. it wasn't an open window where we could get something done. >> Right now though, the you know utilities are everything. So the state is dealing with uh utility issues around the the markets you know the the auction markets on this the supply side. They're dealing with clean energy goals. Uh every almost every state has clean energy goals even

25:48the red states and uh data centers you know prices spiking. So, when we show up and say, "Hey, we saw that you're running this bill about data centers. We want to make sure low-income households are not going to be harmed by this in any way." >> Yeah, >> it resonates. So, we we're finding an open and active policy cycle right now on utilities politically in the states. Federally, not at all. This is a devolution to the states kind of thing. on the federal level, we're not going to probably have an AI policy or a data center policy or much of any other kind of energy policy on the federal level. >> Yeah, I can see that. Tell me about let's talk about data sense for a moment because I know it's a thorny issue from

26:31your perspective looking at, you know, with the lens that you're interested in, which is essentially using energy policy and energy benefit schemes to help lowincome households. What has the data centers in Illinois or the use of, you know, the the the use because data center been around for a long time, right? They're not just new for AI, but what is the use of AI meant for energy costs in Illinois? >> I don't think we know yet. Um, you know, and our governor has presidential aspirations, so the current policy is freeze everything. You know, like don't take a position, just freeze everything. There is a bill that's going to be addressed, be brought back into our veto session this fall. It's called the Power Act, which is an acronym, but the the

27:13the bill is basically saying that these data centers will need to bring their own energy. You know, like come up with a private deal with a provider or something uh to get their own power so that it doesn't cost all of the rateayers. So, pay your own way almost provide your own way. We have a nuclear plant that uh you know, we have a big company has said like we're going to buy all the energy from that nuclear plant. That's all symbolic. You know, electricity doesn't work that way. Electrons don't discriminate where they come from. But that act has an implication for low-income households because we're now seeing data centers that want to connect

27:53directly to the power grid or to the to the generator, not through the power grid. We get utility bill money on the state level through the grid. So, you know, anything anything purchased on the grid is going to contribute to lowinccome assistance that matches the federal lie heap. Right now, the way that works is it's 80 cents per residential meter, 10 times that for commercial, 375 times that for industrial. But if you take these big providers, well, that 375 times thing, 375 time 80, the welding shop around the corner is going to pay the same as a data center. Um, and that's we don't like that metric, of course. So, we're trying to figure out

28:36how do we get the low-income households represented in this movement towards these very large power users >> and that power act would assess an annual fee that would go into this fund. >> Yeah, I see that. So if we bring it down to a slightly more local level just for a moment, but again in in Illinois. So you know, you represent 36, I think, community action agencies, 102 counties, big surface area, right? You've clearly got Chicago, but some pretty rural areas, too. What are you talking to your community action agencies about? What are they what can they do better? How can they influence this? How can they serve their communities better as well? >> Yeah, great question. We always say the

29:19number one thing that you can do to communicate your mission, you know, we talk about elevator pitches and all those kinds of things, but the number one thing is just run a great program. That's it. You know, like run a fantastic program. And I think we especially with lie heap here in the state, we've got that. Now, the one thing that has been true for lie nationwide is that every agency runs it a little differently. >> I don't know if that's good or bad. I still haven't decided. Well, I can tell you it's not great from my perspective, but you know, but that but I'm I'm less important on this. Yeah. >> But technology does give a a a path that can make it more standardized >> and yet not negate that local feel. >> Sure. you know like that that omni channel approach if you have a good

30:01mechanism behind the line heap application a good technology I can log into a site and do my own or a worker can do that with me >> and and yet we can make it easier for everyone so you know the the technology isn't isn't uh going to erode any of that local feel but doing delivering a great program is number one making it more accessible to households would be the second one you know Illinois is a big state uh if you have to get into an office that's only open during working hours and you're a working person. Are you going to take time off to go there and do this? So, you know, like making this making it all more accessible would be the the next thing I would say. >> That's interesting. The bit that we're really obsessed with is this idea of a

30:43front door. So, the truth is if you're applying for light, you've got other things that you probably are eligible for too, right? And I think part of the problem is that most organizations are not all work in silos, right? because you've got to get a person and push them through a process. And so the thing I'm obsessed with is well, if you're eligible for this, what else are you eligible for? Or if you're not eligible within this agency, let's punt you over here and try to get you introduced there, right? Because it's all people on access in the in the end. So that's the that's the one thing I think would be the one we try and bang on about. The second one to your point about 16% of people take the benefit but actually you know there are a bunch of other people

31:26far more who could who could get it. Outreach for me would be something that would be super interesting because from a technology perspective because we can't expect community action agencies to be going out knocking on doors saying you know come into our offices. They're already mostly overwhelmed. And so but the idea would be if you can find some technology mechanism to actually say hey you are more likely than not to be eligible for these things. Do you know this could you know I think that would be that that those are the two things I think would be levers against your if there was more money pot that would be helpful too. >> Yeah and I think uh you know I've thought about this a lot and even tested some things but I think I I think I've

32:08always looked at it the wrong way. We're in this era of government often referred to as new public management where you know we we treat everything as a business. Run government like a business, treat people like a customer. And when you do that, the problem with doing that is that then your funer gets into counting things. And so the customer at the counter becomes cargo. So you know like you're you're saying that agencies operate in silos. In the business world we would call that specialization. And it wouldn't necessarily be a bad thing, >> right? And then when you talk about getting out there and doing outreach, we would call that customer acquisition cost, right? You know, like business development, right?

32:48>> Right. So, I've always approached it that way that like we're in these silos, we got to break down the silos and also we got to go out and acquire a customer and so we're going to do that from a customer acquisition standpoint. A few years ago here in Illinois, we we took a million dollars of state lie heat money and did targeted marketing, web and otherwise, billboards, you know, print media, everything. a million dollars and we paired zip codes. So, we did match zip codes where, you know, this zip code is demographically similar to this one. One's treatment, one's control. Our customer acquisition cost in that million dollars was over $400 per household. Almost exactly what the lie he benefit was. Like, we're we're spending almost the entire lie benefit to acquire a customer because it's an approach that didn't work. And instead,

33:31you have to humanize the the whole process. And to your point, I I think it would be fascinating if these programs did not all have fractured databases. If a household has come in for lie heap, but they have a preschooler, well, you know, that Head Start becomes a logical tie and you've already acquired the customer. But if you have them in separate databases, you don't even know you've acquired the customer. >> Yep. So, you know, unifying all of that, then you can have a human conversation with that customer, that person about, you know, how these other services might benefit them instead of doing all of these isolated approaches that really just don't work because this is not a

34:11business. People are not looking to buy something. >> Yeah, exactly. I mean, there there was an agency I just talked to who has 16 separate databases just for their agency. So, this may be an interesting challenge. But what what amused me about your story about the outreach was a million years ago I did a I did a documentary around how the Barack Obama campaign and the Mitt Romney campaigns used data analytics to do outreach and the best the best voter turnout mechanism was told to me by a guy called Hal Malchow who's a savant in this whole area and he was saying that the way you did it was you could do issues based mail clearly and you could be sort of

34:52more ragy based male and there's plenty of that. But the thing that got the most voter turnout wasn't saying you should vote for this issue. It was saying your neighbor has voted the last three elections. >> Right? >> This other neighbor had only voted in two. We after the election we will let everyone know whether you voted or not. And it turned out the vote mattered. Now, it really enraged everyone because clearly, you know, that feels like quite a personal thing to know that Joe blogs down the road hasn't voted and you have, but it was it's a fascinating thing in the sort of behavioral economics or nudge theory on how to get someone to do >> Yeah. Yeah. That that is a fascinating article and uh you know, they they were only sharing public data,

35:35>> right? It was it was available. It was it was available. Yeah. Absolutely. >> Right. And but it but I think you know it uh it gets into the the nudge of okay my neighbor's doing this and I I don't want to look like a cretton so I'll do it. So if you flip that around for lie you know normalizing getting lie heap you like that it's not a bad thing that you're not sucking on the government or taking charity or whatever it might be you know how do you nudge in that direction to say this is a perfectly acceptable and normal thing to levelize your bill. I'm uh for the first time ever we have longitudinal data available in Illinois. The states made deidentified data sets available for seven years in counting. And so I'm

36:16working on a panel analysis because the myth is that the people that get this come back every year and they just, you know, apply. The whole welfare queen thing applied to lie heap. And my initial pass at that data, I've got to do some robustness checks on this. But it's not true. In fact, most people don't apply every year. And the households that are more likely to apply every year are senior citizens. So, so it's if you set that aside and you look at like, well, what caused households to apply? If you put them into groups and you do this sorting, then then you'll find that crisis is it, you know, like I'm going to be disconnected. So, I come in because I have a crisis, not because I need to apply every year. >> Yeah. Yeah. I see that. I see that. What

36:58What do you would What would your So, you talked about an agency running a better program, for example. when you did it yourself, what was when was it humming? When was it working best? >> Yeah. And I think you'll like my answer here because uh this is what you you do. This is your passion. I was in Iowa and it was a couple of years before the pandemic and we just we leaned into CRM and the whole omni channel marketing kind of approach. You want to text us, you want to call us, you know, what do you want to do? Uh and really leaned hard into that. Even if someone walked into the office, we put them into the CRM because then we had transparency. I did it to create a learning organization. Not really. That wasn't my

37:39priority was not to make customer service better. It was I I thought it would and I hoped it would, but I wanted to create a learning environment where our staff could all learn from each other by watching these tickets in this CRM. And then we would all own customer service. So, you know, if you have an iate customer, it's transparent to everyone and we all work on it. We all see how each other handles it. and you know just created a better team and then the pandemic hits and we're all trying to work from home and I'm like oh thank goodness you know that we've got this CRM and then we layered an online application very rudimentary but we layered an online application on top of that and suddenly our services became asynchronous you know you did not have to come during the daytime you did not have to talk to me in lifetime and then

38:21I had a a staff person who was fantastic who said I've got to quit because my child care options have fallen through because of the pandemic I need a second shift shift job so my my mother can watch my children at night and I'm like no it's okay just work second shift. So you know like being taking off the way we've always done it and thinking about the way technology and internal business choices dynamics can be different I think would be the the way I would want agencies to begin to look at lie heap. >> Yeah. Well, clearly yes, you are talking to the converted because obviously we like 24/7, we like lots of languages, you know, those are the types of things. It it reminds me a little bit of the

39:02last company I worked in before we started this, which was that so this was equivalent of Zillow in the UK and people generally started searching for properties on Boxing Day and it was because they were desperate to get away from their family and so they they really wanted to, you know, you couldn't legislate for when people wanted to do something, right? They wanted to do it when they had a trigger. >> So, so no, that's great. But, but Roger, thanks so much for for coming on and and talking to me about this. It's been it's been a real joy and I hope we're going to talk more about this because, you know, clearly you've got a fabulous background in this and now, you know, we're trying to learn as much as possible. So, I really appreciate it. >> Thank you. It's been fun visiting with

39:45you and yeah, I look forward to working with you. >> Yeah. X