SPEAKERS
Kevin Moss, Rich Alterman

Rich Alterman  00:04
You're syncing up and tuning in to The Lending Link Podcast, powered by GDS Link, where the modern day lender can dive deeper into the future of data decisioning and Credit Risk Solutions. Welcome to the show everyone. I'm your host Rich Alterman and today we're syncing up with Kevin Moss. Kevin was my first guest on The Lending Link back in October 2022 and I'm very pleased to have him join me for my 20th Lending Link podcast. As a refresh, Kevin is a Lending Payments Credit Risk and Compliance Management Veteran, having held key roles with various companies including Wells Fargo and with SoFi, where he was their Chief Risk Officer. Kevin today runs his own advisory firm providing consulting and advisory services to various companies in the lending and credit risk management space. In his spare time, Kevin sits on the advisory boards of several companies including Plaid, Peddle, FairPlay and fraud platform providers, including Centrelink secure and NeuroID. In this episode, Kevin and I will be discussing the economy, its impact on lenders, fairness and lending, the use of cash flow data and so much more. But before we dive into the interview, please head over to our LinkedIn and Twitter pages at GDS Link that's G D S L I N K and hit those like and follow buttons. If you have not done so already, please subscribe to our podcast on Apple podcasts, Spotify, or wherever you prefer to listen to your podcast. All right now let's get synced with GDSL Link. Good afternoon Kevin, I hope you have a great week so far. And thanks for returning for my 20th Lending Link podcast. Very excited to have you here today. Are you joining us from the Bay Area?

Kevin Moss  01:46
Yes, I am Rich.

Rich Alterman  01:47
Good. So Kevin, when we we met in October. One of the things that you shared was that your big warriors fan, I see that they had a bit of a drop in the win rate for the 22/23 season compared to the 21/22 season. If you are their coach, is there anything different you would have done to drive a better outcome?

Kevin Moss  02:05
Well, last year the warriors had some injuries. They had Andrew Wiggins out for quite a while towards the end of the season. And they had the Draymond Green Jordan pool event which I think upset the locker room and probably the normal camaraderie that has been so well orchestrated over the last years that the Warriors have been such a great team. But Jordan Poole was traded, Draymond Green was signed to a longer term contract. I think the picking up Chris Paul was an interesting trade. And I'm optimistic that at least for a season or two, he might bring some real value to the second team. So bottom line, last year was disappointed. It was a very frustrating season. Clay didn't play very well in the playoffs. But I'm optimistic they picked up some height in the trades, and signed some seven-foot player to a contract. So I think this year will be better. Do the Warriors have what it takes? I would say I'd like to see the first 10 games of the season. And I'll tell you what I think.

Rich Alterman  03:17
Maybe you'll come back after that. And hopefully, Coach Kerr will drive home a good win for you guys. So Kevin, it's been 10 months since we last spoke. And you know, getting ready for today's podcast, I took the time and went back and listened to our prior podcast. And there are a handful of discussion points from our first podcast we're touching on today in addition to some new items we have not covered. So let's get started. A key discussion item in October was the state of the economy and your view on what we would see in 2023. There's still debate as to whether or not we're in a true recession or not. But there is no doubt that consumers and even businesses are really struggling to make ends meet. Since we last spoke, there have been a lot of key events and trends in the financial markets. I'd like to spend a few minutes and just list a few here. But earlier this year, we had the collapse of Silicon Valley Bank, Signature Bank and First Republic Bank. According to the Federal Reserve Bank in New York, household debt balances set a new record at $17.05 trillion during the first quarter of 2023. Throwing $148 billion or 0.9% for the fourth quarter of last year. The debt load has spiked with additional $2.9 trillion since the end of 2019. According to the Federal Reserve data credit card debt has surpassed $1 trillion dollars for the first time ever. In the past 16 months if it has increased the Fed funds rate more than five percentage points, with six of those taking place since you and I last spoke in October. In August this month, Moody's downgraded 10 small to midsize regional banks across the US and put several others on notice. And Fitch Ratings downgraded the US credit rating from the top tier of triple A to double A plus. And lastly, hot off the press today just before I joined the podcast with the federal student loan pause that took effect in 2020 coming to an end this October of whole by the Harris Poll done on behalf of Credit Karma found that some 56% of federal student loan borrowers said they will need to choose between making their monthly payments or buying necessities such as rent and groceries, because they cannot afford both. According to the study, the percentage of those forced to make that choice swells to 68%. For those with annual incomes of $50,000, or less. Clearly 2023 has seen its share of challenges in the financial markets. And as we sit here today, it's hard to see a soft landing anywhere in our near future. Thinking back to when we spoke in October, which of the items I outlined above, or any I did not mention are the most surprising to you. And at the same time, maybe which are the most concerning to you.

Kevin Moss  05:47
So Rich, I think back in October, I was thinking there was a better than 50% chance that we might be in recession in the second half of this year. And I think of late economists are more optimistic about avoiding a recession, Goldman in July put the odds at 20%. In the next 12 months, Moody's phrase the term called "slow session", instead of recession, putting the odds over 50%. So what I see is, you know, second quarter growth was 2.3%. In the latest read on it better than what was expected. Unemployment is still in the mid threes claims have been coming up, but not at a fast rate. So my view is that, while much of the six increases that you've identified since we last talk, still have to play through the economy because of the land effect. Like I think the economy's handled, the Fed rate increases pretty well. So far. We've seen some slippage in the stock market over the last two, three weeks. I think the downgrade was a surprise, but it shouldn't be, you know, I think Fitch what they did made sense to me, but the bottom line is bringing it back to credit and performance. I mean, we have seen significant changes in delinquency and particularly in the nonprime space, personal loans have come back the last four to six months where delinquency at at the aggregate level has come down the most recent villages that that we're following, which is Q4 of 22 and Q1 of 23 look better. A credit card continues to deteriorate as a asset class. Auto is softening particularly in the used car market and real estate's held in pretty well. Underneath it, there's a little bit of softening. So I would say that credit overall is still doing okay. And bank portfolios are better than FinTech portfolios, because what's what's hurt FinTech volumes, has been warehouse financing rising of costs. And banks have a competitive advantage, because they're using deposits to fund themselves. So what you see is you've seen a little bit of a share shift, like in personal loans between banks and fintechs, because they can offer better rates because they're not as affected by the rate environment as warehouse financing. So overall, I'm not tremendously pessimistic. I'm cautiously pessimistic. I think that when I talked in October, I talked about a handful of things that were driving credit deterioration, we talked about score inflation. We talked about credit builds, certain credit builder trades, and obviously, interest rate increases and inflation. So let's talk about each of those. So TU actually released a study, I think, a summary of which was in the Wall Street Journal. And you know, what they did is they looked at Vantage 4.0, or they looked at pre COVID. And then they took a period of time during COVID and 21. They looked at credit card performance. And what they did is they shifted the COVID performance 35 points to get the bad rates to kind of line up for most of the score ranges on top of each other. So that meant roughly that Vantage 4.0 was inflated by roughly 35 points relative to the pre COVID.

Rich Alterman  09:46
And that was a result of the forbearance and deferment.

Kevin Moss  09:50
Correct and the natural disaster. The way Vantage treat natural disaster trades where they remove all the payment history they neutralize the payment at the trade line level. So that was more than I actually thought it was. Okay, I think Vantage probably is affected a little bit more than FICO and because of how they treat those natural disaster trade lines, but all scores, who use credit bureau data, and payments is always a big component of most scorecards are going to be inflated. And now, I know we'll talk about student loans later. But the next phase of this is a bunch of 27 million student loans that haven't had a payment made in three years, are going to start payments again. And the government is not reporting delinquency status for 12 months. So that's creating another potential score inflation that creditors need to think about. Okay, on the credit builder trade front, not much more to report, I still think that there are some product structures out there that are problematic. And then on inflation and interest rate increases. You know, this week, the minutes suggested that the feds not done. So we might see another one or two increases, and then inflation is coming down, but it's pretty sticky. Right! So I think we're going to be in a higher inflation rate environment for another year, or 18 months. And what people don't really comprehend very easily is that even if inflation comes down to 2%, it's coming down off of a much higher level. So going out to eat isn't getting cheaper, it's still gonna go up, yeah, supermarket, going to the supermarket, food is still way more expensive than it was a couple of years ago. And even if inflation comes down to 2%, it doesn't mean that prices are gonna go back to where they were. So I think the impacts of both higher rates and inflation are going to have a sustainable impact on people's cash flow. And hopefully, with above average wage increases for a while, maybe that could help cover some of the gap. But that's going to continue inflation at a higher level for a while. So bottom line is we still have some challenges ahead. We're in baseball terms, we're probably in the middle innings of this, right?

Rich Alterman  12:25
Yeah, I mean, not only are groceries more expensive, but getting there here in Georgia in the last month, gas prices have gone up a quarter and part of that summer, right. But people absolutely feel that in their pocketbook when they're standing there before that pump. And I saw an interesting illustration the other day, Kevin, I think it was on LinkedIn where it showed three shopping carts. And it showed what $20 bought I don't know how many years ago, but the shopping cart was overflowing with things. And then it showed what it buys today. And there were like six items in the shopping cart. And, you know, those are, you know, real real impacts. And to your point going out for for meals, and I think people kind of rationalize that they've got to spend money on food anyway. So going out for dinner is something that they're more willing to do. But yeah, price is, it's pretty hard to walk away from a decent restaurant for for people now where you don't drop 125 bucks.

Kevin Moss  13:18
Or more depending on where, depending on where you live.

Rich Alterman  13:21
Exactly. Yeah. So thank, you know, thanks for the revisit on on some of those things. You know, one of the interesting things that happened earlier this year was Goldman made a decision to wind down their markets Personal Loan offering, you know, as you think about that, and think about the efforts they made in that space, are there any lessons that can be learned from what we saw with Goldman that you might want to share with the other industry people?

Kevin Moss  13:46
Well, I mean, I don't know all the inner workings of what went on from Goldman, I have a bunch of friends from there that are looking for jobs that I've been trying to help. But I think the lessons for people who are looking at organizations, Goldman is historically been an investment bank that tried to reimagine itself, as a consumer, like a bank for the average person. And I think culturally, retail banking, and investment banking are very, very different from each other. The second thing was that they were building this up from the ground up, they didn't really have a portfolio to start with. And anybody who's been in consumer lending for a long time knows that it's going to take a little while to cover your acquisition cost and get get your policies right and your processes right. And I guess it feels to me like the company needed longer term perspective, because they were losing money. And I don't know what the long term nature of what they did. I can't talk about their credit or their marketing or or anything. But the bottom line is that anybody who's looking at an organization, who's new in this type of business, needs to make sure that there's a real commitment and understanding about it for the long run. And that's true, whether you're, you're looking at a early stage startup, and how well they're funded, or even financial institution who wants to get into a new business. So, you know, I think there's a lot of smart people running around that organization. You know, they still have the Apple partnerships, they're still committed committed to savings, they've gotten out of personal loans. So right now, they've just refocused some of those efforts. But you know, there's been talk of selling green sky, and Amex wanting that wanting to step in, or someone wanting Amex to step in, to take the partnership. So it could be a complete exit from what all the talk is. And for a company like that, it might be the best thing, because that's not their core business. It was never their core business. So I think it's important for people who enter new businesses to really be committed for the long run.

Rich Alterman  16:15
Right. So that's a great segue, you know as we think about diversification. And you know, when we think about most FinTech lenders, they began their life offering just unsecured personal loans. And many have expanded their products offering to include credit cards, auto equity loans, home equity loans to name but a few, can you share your view on the importance for lenders to become multiproduct? And what are some of the considerations related to risk operations and compliance that they really need to account for, as a, you know, have expanded or plan to expand beyond personal loans?

Kevin Moss  16:45
Yeah, I mean, you know, I worked at SoFi as their CRO, SoFi started off as a student lender, and then got into personal loans and mortgage. And I think the rationale was that the HENRYS's as they were called the high high income, but not rich yet, kind of target customer where the student loan refinance is almost two thirds graduate students, lots of doctors, dentists, MBAs, lawyers, $160-$170,000 income, that the personal loan business there, they targeted, larger personal loan balances than the typical prime and super prime market, higher end customers, again, the HENRY type, and then mortgage, which is the single biggest credit transaction that most people do in their lives. That's as these student loan customers mature in their careers, many of them are going to end up being homeowners, so. So I think there was a logical flow for how SoFi expanded its lending products. And then it got into deposits and brokerage. And the whole idea is to be like a super financial app. And I think they did it a very logical way. So for a lot of these single product companies, let's say they're a personal loan lender, a bunch of them have gotten into the card space, the credit card space, unsecured, similar underwriting approaches, leveraging of machine learning and alternative data, all all that kind of stuff. So it feels to me like many of these personal loan lenders are diversifying into like the car business, or some of like, Prosper went into the home equity business. So I think, I think that you just got to think about what is the adjacency of the products that you're offering? When you look at your core talent and core capabilities? How good a fit, are they and your underwriting and servicing aspects. So it does make sense to me, because over the years, like there was a time in our history where there were a lot of monoline credit card issuers and those then it was like now you got to be. You got to be a supermarket of products. Right. That was the Citicorp merger and and then I think over time, FinTech has kind of reemerged with monoline companies. So I think that's a cyclical kind of thing. Where I think with the internet and digital lending, if you're really good, you can be a successful monoline company. If you have great products, great experience, great delivery, great pricing, you have great analytics, which I think is the thing that really differentiates the use of the data. You can be a successful single product company, I think in the digital world. But over time, I think growth comes from product development and product expansion too. And so looking at other products other adjacent channels, other things that you can cross sell or satisfy your target segment of customers with makes a lot of sense to me. Like, if I love you as a personal loan customer, and you have a great credit card product, I'd probably give me a shot, right? I think it all boils down to being really good at whatever you do. Being really customer focused and and value oriented.

Rich Alterman  20:29
Thanks for sharing that. But I mean, certainly from a risk perspective, you know, if you're providing some guidance, and working with a monoline player that is looking at, let's say, getting into credit cards, because it seems to be a typical, next logical step, when we look at some of the fintechs Are there any just key words of advice from a risk management perspective that someone who's just delving into credit cards, a lot of success in personal loans, but it's a very different product? What are some of the nuances that they you know, if you were advising them that they really need to be thinking about as they shift or expand, especially if it's not, maybe with the same customer base, but it's a brand new customer base that they pick up?

Kevin Moss  21:09
Well, hire someone who has deep credit card experience to run the business, and also to run the risk for that business. Because mining credits is very different than a loan. There's lots of different dynamics around customer management, you know, bust out payment risk, you know, what happens over time when people's credit starts to migrate up or down line management authorizations. Quite candidly, credit card is much more complex a product than a personal loan product isn't a personal loan, your key decision is: Do I lend to them and how much? And then once the money is gone, it becomes a payment processing and collection game. For the most part, when it comes to credit card, it's about, I tend to say that you can reduce everything that everybody does in a credit card, or line of credit business to two simple things, either focused on growing good balances, or focused on reducing or controlling bad balance accumulation. And the most successful players are good at both things. So you got to have a risk leader to start with, and a risk team that really understands line of credit risk, transaction risk, fraud risk, all that kind of stuff. And then around the rest of the organization, you have to have people who are focused on loyalty and usage and retention and account activation and utilization programs, merchant focus things to try to spur usage, where people like to transact, like there's just so many different marketing and product aspects to a transactional product, that a loan product just doesn't have to worry about.

Rich Alterman  23:07
Good advice. So before COVID hit, you know, there was definitely a lot of talk in the FinTech space, where companies were considering or looking at, could they acquire a bank charter? Should they become a bank? Certainly, we saw LendingClub, SoFi were used to work, and Upgrade became banks. We sit here today, given the current environment, do you see that as a steal as an opportunity for some of the fintechs to potentially become banks? Or do you think that's really been kind of put on the shelves at this point?

Kevin Moss  23:36
So first off, FinTech grew up in a low interest rate environment. So the last 12-18 months is one of the first times where we've been in a sustained higher interest rate environment. And I think, like I said, in the personal loan space, banks are taking some share, and fintechs will lose, actually, credit unions were also doing pretty well, too. And I think that banks have a real funding advantage. So if you want to be big, you want to grow to be a big lender. I think getting a banking charter makes sense. But it comes with a real commitment to compliance, you're gonna have to regulators get very concerned about very, very high and fast growth rates and things that typical FinTech or early stage FinTech companies are very enamored with historically, like a bank regulator bikes to see predictable growth generally in the prime and area. Most banks are not in subprime deeply Capital One is in subprime credit card, Subprime Auto, but they are known to have the best analytics for most banks. Most banks are not in subprime lending. So I think it really depends on like, who your target market is, what your long term ambitions are. I think the FinTech companies like LendingClub and so far that have become banks have been enjoying, I think better growth than other FinTech companies have. And, and a part of the reason at least, is that they've been growing deposits aggressively. And those deposits give them a competitive advantage over the other non bank fintechs, because they have more pricing spread to actually work with, right. So I do think it's an advantage. So, in my view, the marriage of banking, and fintech, like a Varo.Varo, is still relatively small in lending, but they're growing and they're investing in it. Like, I think, in the long run a bank that has deposit funding available to it. And also has the FinTech platform and experience without the overhead of legacy systems and, and in some cases, branches. I think in the long run, if that's managed well, it could be a winning model. In the long run, I think what we're seeing is, banks have been reducing branches, as more and more adoption of the digital channel has happened. And I think that's going to continue, I don't think branches will ever go away. Because they'll always be people who have branches are good acquisition tools for deposits. And people like sometimes to be dealing with people in situations whether it's my generation, or even others. But digital, remotely adaptable channels are going to be the growth engine for I think financial services into the future. And I think banks have the journey ahead to replace some of their core legacy platforms, and become more digital and cloud based.

Rich Alterman  27:00
Great, thanks. So I want to kind of bring couple things together, we've met, we've talked about credit cards, we've talked about personal loans, we talked about coming to bank and the value of deposits. So as I mentioned in the podcast and you reiterated we've seen six increases in the Fed funds rate since we last spoke in October. Despite this, right, we've seen credit card debt break through the trillion dollar mark, as I mentioned. And according to a study by TransUnion, in Q1 2023, we saw 26.3% year over year increase in personal loan volume to a new high of $225 billion, with the average loan per consumer being the highest on record since 2005 at $11,300 versus $9,900. Given that a large majority of unsecured personal loans are used to consolidate credit card debt, do you feel that we might see some type of a "yo yo" effect with credit card balances increasing, then they get consolidated via personal loans with the risk that cash flow pressures, then result in run ups on the credit cards that they had done the debt consolidation with. And in fact, you're probably familiar with a study that was done. It was called winners and losers in marketplace lending that came out many, many years ago, that actually went through that process and showed that as people actually did pay off their credit card balances, people that had lower FICO scores tended to get a bigger bump in their scores than people that had a higher FICO score. And they found that in the study, 12 months later, a lot of people that had taken that debt consolidation and had in fact paid off their credit cards, had now basically gotten to a position where now they had this new installment loan, and they had racked up the balances again on their credit cards. That's where they actually better off. So do we see that phenomenon potentially happening again?

Kevin Moss  28:47
Well, there's no question that two thirds to three quarters of the loan purposes for from my experience for personal loans, or debt consolidation, or debt and loan consolidation, most of it is credit card debt. And the reality is, from my experience, something like a third of it is successful. A third of it is people who kind of run back up. And then a third. I mean, the losses are are nothing like this. But But then there are people who fail completely. So a good chunk of people in the long run, don't change their behaviors. Now, here's what's different today. So first off, let's talk about home equity. Mortgage cast out refi is essentially dead. Mortgage rates just hit the highest rate they have in 20 years. So over 7% and 91% of the people from his TU study that I just saw through Q2 have rates below where the rates are right now. So home equity has been doing okay, so people are using home equity to consolidate debt. But some people are still using personal loans because they don't necessarily want to take out a second or put a lien on their home. So I still think debt consolidation in an environment where credit card rates have risen a lot. They're around 20%, because the most of them are variable. And as the Fed rises, so do the credit card rates, okay. And so there's the benefit of consolidating to a fixed rate, personal loan is still there, where, you know, if you pay your minimum payments, it's a 10 plus year type pay down on a credit card. But if I can take out a four or five year personal loan, and I really stick to that, then I can be out of debt, you know, when I pay off those loans, so I said, there's still a strong value prop. And as the economy goes, so goes the risk associated with that down the line. So I think with higher rates and higher inflation, the risk of people running up their credit card balances, again, is probably higher now than it was pre COVID. But I also think that the underwriting, like I said, the 21 and 22, Vintage curves, through the third quarter of 22, for personal loans, were bad. Fourth quarter looks a lot better. First quarter, it's probably too soon to read. So I think the industry like would tightened up substantially in subprime. And even on the prime side, I think people have really tightened up. So my view is that the industry underwriting has gotten more conservative, probably offsetting the risk that the environment has created. So I'm gonna say that going forward, I think we're gonna see we're always going to see successful consolidations, we're always going to see people who run some level of debt backup. And we're always going to see people who, who don't successfully consolidate in any environment, you may see more or less of one or the other. But I think that underwriting is a bit more conservative now. And while people got burned in 21, and 22, I think we've now mostly addressed those gaps in performance. Now 23 has the student loan payment resumption going on? Yeah. And that's something that risk people should be worried enough. So we got 40 million student loans out there. 27 million roughly, have not made a payment over the last three years. And either they're making their first payments, because they were supposed to over the last three years, or they're resuming payments that they were making before the COVID stop. And like, median amount is 180, 19%, over 500. So there's going to be some payment hierarchy shifts. And so, you know, there's been some articles coming out about it. I'm working on an interesting one, which I can't really talk about, but hopefully in the next 30 to 45 days, that'll come out. And, you know, what I can say is that payment shocks. From this, the resumption of student loans will play through everybody's portfolio, you have that many people, a typical credit card portfolio has 20 to 25% of people who have government loans, for example. So it's good to play through. And the other thing we have to watch out for is, since the government has made the decision not to report student loan delinquency, were in effect, creating a score inflation again. If you've got people applying to you that are heavily their trade lines are heavily weighted with government loans. The potential score bias could be very significant, like at points in the future. So I'm very concerned about we're essentially recreating the Cares ACT score biases for the right public policy reasons. But it's it's going to be a challenging environment, again, for risk people to understand what the how they should adjust their credit policies and the credit underwriting guidelines to make sure that we don't get another bad set of vintages.

Rich Alterman  34:40
That's a really good segue into my next question. And in our podcast back in October, we discussed the use of alternative data with the emphasis on data secure from applicants bank account. As part of this we discussed how the use of data can help with one of the CFPB's key missions which is improvement in the area of financial inclusion. On July 27, so last month, the CFPB released a new blog post putting forward the conclusion that cash flow data broadly defined as the various inflows outflows and accumulated amounts in the customers checking savings account, may provide lenders with a better picture of a consumers ability to repay their loans than using just a credit score. While their study validates the value of cash flow data. The conclusions in my mind really should not be surprising to the reader. Given your involvement with both Plaid and PrismData, can you share some insights on the adoption rates of the use of bank transaction data, and perhaps highlight some of the finding clients of yours have observed through its use?

Kevin Moss  35:39
So the biggest hurdle for using cash flow data is that it's permissioned. So what people like about the credit bureaus is if I have your name, address, social and birthdate, I can, unless you put a freeze on your credit file. I can this is friction free. Okay. So a lot of people are hesitant about, well, if I asked somebody for their banking credentials, historically, depending on the population 50% opt out, right? Because they don't want to share the banking information. So right now, where does it make sense? It makes sense for thin or no hit credit files, for people who don't have much of any credit established, it makes sense for people who have who might have tarnished credit. But when you look at the way they're handling their primary checking account, because that's where their income comes in, they're able to pay all their bills every month, and and have a positive balance at the end of the cycle. So the question is, you know, how do you find those people? So I think right now, when I talk to clients about cash flow in the consumer side, people are worried about the friction that it creates. And a lot of the discussion is about where does it make sense? Where could it be a swap in opportunity? Where does it provide really the most added lift? I think it's in those segments. But now we got 1033, which, later this year, we're going to have some guidance from the CFPB, probably effective at some point in the next couple of years. And, you know, that essentially, says that a consumer controls their deposit transaction system of history with with a financial institution.

Rich Alterman  37:42
And that's gonna include credit card as well, right?

Kevin Moss  37:45
It may, it may. And then the other thing is this thing called the financial data exchange, which has the format by which people share the data between institutions, and ultimately, will tokenize the process, right. So what that means is, like if your account is at Wells Fargo, and you want to apply to Sofi, SoFi can get a one time use case, you know, without sharing credentials, from your bank, and online banking, to get the transaction history that they need for their purposes. Once that happens, then the potential for this to replace, and complement traditional credit scores, I think goes through the roof, like what I can tell you is from looking at PrismData and comparing it to traditional credits, like generic credit scores, or even custom credit scores, is you can build models as powerful as you can with the credit. But the friction is really the big hurdle. Right? So I think, if we can get to the point where it's tokenized, where roughly, theoretically could double the use of this from the abandonment we have today. I think that this in five years, could end up as big as credit scoring. Interesting. But, but that's probably an ambitious timeframe. But either way, what I'm here to tell you is the transaction data, there's really two big steps. One is taking the raw transactions, creating attributes or features. And then good data science teams can build models, once they have those features. I know you guys have a set of features at GDS presume has them some other folks. Truth has them. There's other providers that have them. And so data scientists who are good at building models can get the raw ingredients to build good credit models using that data. So I don't know if it's five years, maybe it's longer. But ultimately, I think the best models will likely combine information from both the credit bureau, because you have things like open dates and limits and loan amounts that we don't have in the deposit account, right. And you can combine that with things from the deposit account. And I think you'll, the models that you build will be stronger than the models you could build individually with either transaction data, or just credit bureau data.

Rich Alterman  40:36
And like you say, GDS has done that, we've built hybrid models taking in both the credit bureau data and the banking data. So it's definitely a powerful lift. So I'm keeping an eye on our time here. gonna wrap up with one, one last question. So back in December, when I did a podcast with Kareem Saleh, who is the CEO of FairPlay, and FairPlay, offers a fairness as a service platform, which helps lenders identify potential biases in their decisioning systems and provides options to increase profitability and fairness. And one of the things that, you know, we talked about when I was speaking to Kareem, was how potentially there's some lenders that really may struggle with their decision to leverage a platform like FairPlay, perhaps adopting a little bit of a head in the sand approach, recognizing that you're an advisor to FairPlay and putting on your unbiased consulting hat, what guidance would you provide lenders you consult on the value of adopting a more proactive approach toward disparate impact and disparate treatment identification?

Kevin Moss  41:38
So we could spend the, an entire hour on this, but here's the message that I would share: The world that we're in today is migrating towards the use of more advanced modeling techniques. And what comes with machine learning and AI is a higher bar around things like fairness, making sure that models are explainable, that we can provide adverse action reasons, all the things that have been well ingrained for 50 plus years, in like logistic regression models. So to me, if you want to play in more advanced model techniques, which, over logistic regression, probably lift between 15 and 25%, the separation, whether it's AUC or KAS, or whatever your metrics of choice, the advanced modeling techniques, which handle interaction effects more effectively, particularly with transaction data, it provides substantial lift. So the business justification is you can approve more loans, we can make more loans to people using especially the alternative data to people who are underbanked or unbanked, at fairer prices instead of going to payday lenders or paying, you know, triple digit loan rates. Many of these people hopefully can qualify for under 36% lending. But part of what model governance and model fairness advocates have to accept is that there's a higher bar for validation. And so what I sat in on Karim's pitch to Nika, and I hit it off with them. Because I believe that the future is here today, that the opportunity to use some of these advanced techniques is there for banks. Capital One's been using it, others have been using them for for many years now. But it requires you to, you know, Patrice Ficklin, from the CFPB basically said, if you're going to use these techniques, you better be using fair lending evaluations very carefully. And so, so what I think is that, like, the traditional way of doing fair lending analysis was, it's time for me to do an annual fair lending test. Your lawyer sends requests, you go to a Charles River, maybe you have a big bank has their own teams, you find some differences. Maybe you do a match pair test. You say, Oh, it's all explainable. Somebody writes a memo, it goes into a locked drawer somewhere. Okay. And, you know, and then you deal with it when the when the regulator's come and do their fair lending testing. I think that's an old model. The new model that we have available to us is with Fairplay. You can do continuous monitoring. So you can know, every day, theoretically, what you did whether a group of loan applicants that had the wrong pricing. Did someone get approved or denied at the wrong rate. And you can correct that virtually in real time. So what I think is that for companies that want to be completely open, and fair to all protected classes to all of their customers, FairPlay offers the opportunity for you to do continuous monitoring, continuous remediation, if there's any thing that has to be addressed. And then you could frankly, share that publicly, which will give lawyers audio. And class action attorneys will want to go after that. But what a marketing advantage that a firm could have to say, XYZ bank, where the fairest you could be, we're going to share with you what we've seen, and what we've done with them. So literally, I had 50 loans last month, that I lowered their rates, 25 basis points, because our monitoring identified that there was this little pricing disparity, it's taken care of, we've already lowered the payments for the customers before they even made a payment. I just think the opportunity for a financial institution to differentiate themselves, rather than lock this thing up in a drawer and not talk about it, I think, represents like: Who wouldn't want to work with a company who is completely open about the fairness that they bring to the table? So I think this is a huge opportunity for the industry to reduce the impression that so many people in protected classes may have that we discriminate against.

Rich Alterman  47:06
Yeah, one of the things that Kareem and I talked about was the goal, the vision right of having like the United laboratories stamp of approval with FairPlay right, that this is my, my certification. So Kevin, we're, we've come up on the top of the hour, I want to take this opportunity to thank you for joining me today on my 20th podcast was great to have you as my first guest and great to have you as my 20th guest. This is Rich Alterman, we've been sinking up with Kevin Moss, a 40 year veteran in the lending risk management, FinTech and banking areas, who is proudly wearing his warrior shirt. Once again, thank you Kevin. No one can see that but he's proudly displaying it. We've hope you've enjoyed this podcast and please stay connected with GDS Link in The Lending Link to listen to future podcast and catch up on the ones you've missed. That's a wrap and make it a great day. Thank you. Thanks for listening. If you've enjoyed today's episode, please be sure to subscribe on Apple, Spotify, Google, or wherever you listen to your podcast. And be sure to leave us a review. Follow us on LinkedIn and connect with us on Twitter at GDS Link that's at G D S L I N K. Have a question for the show or have a specific topic you want us to cover. Hit the link in the description to drop us a note. Thank you for lending us part of your day. Make it a great one.