The Social Entrepreneurs’ Fund invests in software and services designed to help low income communities and organizations that serve them. In this episode Liz Luckett, Managing Partner, expands on the way we can use small data to evaluate investment opportunities, the risks of algorithms in large data and how organizations within the fund are already creating impact in the world during COVID-19

A Lot of the ways people are collecting information today is not transparent. For example I don’t know what goes into my credit score, I know what the number is, I don’t know what went into it, it’s very hard to figure it out and it changes all the time. People pay to have it monitored and looked at but again it’s a mystery. Because algorithms are so prevalent and opaque they are concerning and they should be increasingly concerning to people because more decisions are being made by an algorithm and I think the need for better data even if you are using algorithms, if those algorithms are alive and they are constantly updated they are probably quite good. To me small data is what is going into a decision and how frequently and often are we updating them. Small data is that constant information that is specific to you.

liz Luckett, Managing Partner

Transcript of Interview with Liz Luckett

Jenelle:
Hello and welcome. Thanks for being here with us today, Liz.

Liz Luckett:
It’s my pleasure.

Jenelle:
Today we have with us Liz Luckett from the Social Entrepreneurship Fund. She’s the managing partner and you can learn more about the Social Entrepreneurship Fund at tsef.com. Liz, can you tell us a little bit about the Social Enterprise Fund, its mission and your role within it?

Liz Luckett:
Absolutely, so the Social Entrepreneurs Fund was launched in 2012. We’re on our third fund now and we deploy private capital to addressing [inaudible 00:38] quality. We invest in early stage companies in the three verticals of FinTech, health tech and human services tech. And we’re focused on improving access and economic opportunity for traditionally underserved populations. So we’ve invested in companies over the years, we’re in our third fund, as I mentioned, that shift, we believe shift the burden of servicing low income communities off the people who can least afford it while providing high quality services to communities that are more often ignored or mistreated. So we target commercial returns while trying to make sure that communities are treated fairly.

Jenelle:
That’s fantastic and you said you’re in here third fund. Can you tell us a little bit about your portfolio?

Liz Luckett:
Absolutely. Our portfolio is really in the three areas I mentioned. So we have three investments in our third fund and it is still open those are two FinTech companies and one is in the health tech space and this is a domestic fund only. Previously we invested internationally as well. And our investments are all looking at trying to find scalable solutions that are taking out unnecessary hurdles to people’s lives improving. And we feel like financial services is one of the areas where people are treated in either a predatory or sometimes concessionary manner but neither is a great business model. So we look at companies figuring out ways to serve people responsibly.

Liz Luckett:
In health, we’re looking at we try and focus around specific disease categories that we think disproportionally affect low income like diabetes or hypertension or obesity. We also look at tele-medicine distance treatments that we think address the problem of most of the professionals being coastal or in cities and you know, health problems being universal. So we also look at mental health and addiction. And then in social services, our portfolio focuses on things like access to social services, whether that’s Medicaid or food stamps or any real government or community offered social services. We try to ease that access and we also focus a lot now on the future of work. So we have a lot of time particularly in our pipeline now of companies that are focused on remote training and reskilling of populations that we think they’re increasingly in need of affordable access to new skills.

Jenelle:
That’s a really impressive portfolio mandate and it seems like it’s actually very specific in the type of impact you’re looking for and also quite nuanced in the type of impact you’re trying to track and measure for. And recently you explored this a bit in a piece, I actually shouldn’t say recently, it was a few years ago now, and would love to hear if anything’s changed since then, but you’d written a piece in the Stanford Social Innovation review about the difference between small nuanced data and big data. Can you define these two for us and tell us your understanding of the difference between the two?

Liz Luckett:
I think my fundamental problem with big data is that it’s mostly based on historical data. Most of the algorithms that are sort of cemented into decision making are based on historical data sets that are used to be predictive of future behavior. So, the problem with that is that they replicate and propagate [inaudible 04:41]. For example, wealthy people are easier to lend into. You know, people with white sounding names tend to move up the list of job applicants. We see people who are wealthy being triaged for better health care. We see the criminal justice system is just overburdened with biases that are propagated in decision-making algorithms used by judges. So we think that all of these big data algorithms, so to entrench either racist or, you know bigoted behavior and widen the gap between wealthy and poor.

Liz Luckett:
So if you think, for example, around a basic credit score, right? Access to credit is one of the best ways to change your circumstances to get a loan to start a business, you know, you can start making investments that turn you into an asset owner, potentially a business builder and an employer of other people. That is how people in this country have worked their way from low income to high income. And because banks, increasingly banks and any lenders need an easy way to assess repayment risk, they’ve come to rely on credit bureaus as those scores over time to help them figure out and make loans to, and a Bureau score, like a FICO score, for example, are built on someone’s lifelong data trail; what your income payments look like.

Liz Luckett:
So you need pay stubs, you need credit card payments, mortgage payments. And they look at it, you know, over a decade of debt and the payment information, which means that misstep stay with you for a long time and they are punitive more often than not. And so clearly if you have a lot of money it’s easy to lend you more. So we think companies use these scores cause they’re inexpensive and that’s the best way for them to make a quick judgment. But that this is problematic. You know, we think over time they’re more nuanced and better ways to assess risk and they don’t, in general, large institutions don’t take on that cost.

Jenelle:
And so the nuanced or the small data then how is that different from big data? So that’s a really good example of outlining and I think it’s a really good anecdote, one that everyone can resonate with. But then what is the value in doing the small nuanced data?

Liz Luckett:
So if you stay on the credit score as just an analogy or as an example, you’re looking at a person’s balance sheet, right? You’re saying, okay, what are your liabilities and what are your assets? And if you don’t own anything, you’re all liabilities. And so if that’s the way I’m looking at you to decide whether you’re credit worthy, you’re not, you know, you’ve only had that in your life. If you don’t own a house, you don’t own a car, so I can’t lend you money. And we think a P and L is a better way to look at a person. So if you’re not an asset owner, what’s a better way to assess an individual’s ability to repay and look at what they earn and when they pay their bills. So we have an investment in a company called Petal, which is a cashflow based underwritten credit card.

Liz Luckett:
And it’s just an example with their going looking at, getting access to your bank account and they’re looking at how much you earn and how often you pay your bills. And this is a way to say that I’m looking at you at a much more specific way, which is one way of thinking about small data, which is what do you earn and what do you spend and the frequency with which you do that can help me determine the risk for lending.

Jenelle:
That’s really interesting. You’ve noted that we also need more and better data, so more of better data. And we absolutely couldn’t agree with you more. Can you tell myself and the audience why you believe this?

Liz Luckett:
I think because a lot of the ways people are collecting information today, first of all, it’s not transparent. So I don’t know what goes into my credit score. I know what the number is. I don’t know what went into it. It’s very hard to figure it out. And it changes all the time. People pay to have it monitored and looked at. But again, it’s a mystery. And because algorithms are so prevalent and so opaque, and this should be increasingly concerning to people because more decisions are getting made by an algorithm. And so I think that the need for better data for one, even if you’re using algorithms, if those algorithms are they’re live and they’re constantly updated, they’re probably quite good. And so to me, small data is really what is going into a decision and how frequently and often are we updating them?

Liz Luckett:
Small data is that constant information that’s specific to you. And so yes, that’s how we think about it. I use the analogy in the article of of small businesses, and both my grandfathers were small business owners. And you think about them, they both extended credit all the time because they knew the community they served really well, and they knew that if you didn’t pay this week, you’ll pay me next week. I know where you are. I know what you do, I know you’re good for it. I think that’s really what we’re looking to replicate at a bigger scale, which is to say, I know more about you. I didn’t just over time because really the credit score evolved into a centralized and national product. When people started moving more, you know, their jobs weren’t in their neighborhoods, they started moving around all the time and small businesses could no longer say, I know who you are so I can lend to you. They started just collecting data and centralizing it and then they had to summarize it and make a lot of assumptions about people, but it stopped being relevant to a lot of people. And so really what happened was there was this rift and a big part of the population just got left out.

Jenelle:
So more and smarter data means that you actually know your customers and your business better. Can you help us correlate smarter and more nuanced data with impact work and then maybe specifically through the lens of TSEF?

Liz Luckett:
Sure, so we have an investment company called Climb Credit, which is a point of sale loan that helps people get access to career in technical education continuing. So he wanted to be a truck driving certificate or learn how to be a database programmer, working in health services. You want to get some sort of continuing education. It’s usually between five and $10,000 to get that and that kind of work, but it will increase your median income significantly. It goes from around $30,000 up to upwards of 50,000 once you have this kind of certificate. So we think that, that an organization that serves the role of getting those organizations, looking into them and making sure that, that when you get that certificate, your income will in fact go up. That graduates of that program are earners and that their income has a positive trajectory. Someone who vets that before they lend you money to do it, we think is a critical step in re-skilling workers. So without that, then I’m just lending you money and saying, Hey, yeah, go get a Trump University degree and then you’re nowhere, but out of money.

Jenelle:
Yes, that’s interesting. So going back to the impact work of TSEF can you tell us maybe some stats or metrics that you feel like highlight the organization the fund and and maybe a story that highlights that?

Liz Luckett:
Sure. So, you know, for a long time now we’ve been tracking what we can quantify in our portfolio. So, since we’re speaking at a time, you know, four or five weeks now into the pandemic, it’s probably worth talking about how the fact that we’ve been focusing on low income communities for a long time this positioned us to be, to have a portfolio that’s trying to be quickly responsive to a growing community. And so we have an investment in a company called [inaudible] that for the last 10 years has been aggregating and centralizing data for every social service in the United States. They’ve really been trying to say, here’s every government service and every single foundation and free program available to you. And so by zip code you can enter and then look by a category. You’re saying, I’m looking in this specific zip code for women’s shelters, food pantries, access to free transportation, whatever it is. And they’ve aggregated that and it’s taking them a long time to build that database.

Liz Luckett:
It is now probably the largest single repository of services for low income communities and they have shifted a bunch of their staff to now focus on COVID 19 response. And so they have something called findhelp.org, which has allowed people to get online quickly and figure out by every County in the United States what’s available to them for help. So that’s one piece that we think is really a great response and really interesting right away. What is also interesting is that they’re offering this ability to then go search by those counties if you want to help and be responsive to the needs of your community, you can now see almost on an hourly basis what people are searching for and what they need. So you can see a community of young mothers are looking for free diapers. You can see that health workers are looking for childcare. It’s always been able to figure out by any geography what people need at what time.

Jenelle:
It’s really neat to see that the company is able to track so much data you said they have the largest data and then from there then pull out trends and patterns that continue to be able to provide in real time impact even though you’re going through a pandemic and a global requirement to pivot. So speaking of the data collection and the trends do you think that the ways companies are collecting and managing their data has changed since a few years back when you originally wrote your article and then twofold question, what are you expecting to see as data trends in the impact space over the next 12 months?

Liz Luckett:
So I think in the collecting and management, what’s changed? I think when you’re dealing with a low income community that has a long history of being mistreated or hoodwinked in some way or another that building a brand of trust is really important. And when people are trusting your brand that they think you’re gonna treat them well, treat them responsibly, they’re willing of course to share their data with you. So you think about that with Petal, which has given you access to their bank account now, I mean, that’s a serious level of trust. I’m willing to do that because Petal approves more than three times the average rate and an average credit score for them is a 680 which is a thin file new to credit people and they approve at more than three X rate than a traditional credit card to let people on to that platform.

Liz Luckett:
And that’s why they’re willing to share that information. And then you’re seeing with we have another investment in company called Zipongo and Zipongo is a digital nutrition platform that’s really trying to take guesswork out of healthy eating. They give you a survey and you tell them foods you like and they monitor people for pre-diabetes and hypertension, et cetera, to make sure that you’re managing the most expensive health cost everyone has his food, right? And if you’re managing what you’re eating, we believe there’s many health outcomes improved as a result. So people are willing to share a lot of their health information in exchange for that good feedback. So I think what’s changed is that you’re getting people handing you their data. You were now sort of trying to search for it, you’re getting it at a personal level, people are sharing information because it’s helping them.

Jenelle:
That’s really insightful. I mean, I think what I’ve heard from you in this conversation is, you know, increased trust in data collection from companies has been occurring over the last four years. While simultaneously there’s tension with the fact that with big data, there’s still a lack of transparency around maybe the analysis and how those algorithms are calculating and having a negative impact. So at the same time, while we’ve seen positive changes on the personal side, it sounds like we still have much more work to do on how we engage with the data that we do collect and make sure that the analysis is much more transparent. So, it’s been a really interesting conversation. Thank you so much for sharing your insights with our audience today. And thank you, Liz.

Liz Luckett:
It’s been my pleasure. Thank you.

Leave a Comment

© RIDDL Tech Inc. 2020 All rights reserved.