All right. So what I want to do now is talk about the Evolution of Peer to Peer Lending, from its early days back in 2006, to what we see now, and how it's really a business plan that's really pivoted from one thing to another which is why as I mentioned, the name is changing. Okay. So the original idea was peer-to-peer lending, right. You have some people who have money that they want to just put to work for a bit, and some people who are in a situation where they need to borrow, let them find each other, let them discover the terms on which each side is happy, on the transaction and the platform, the Internet platform is just going to help them make this happen, right. The big flat forms in those days are the same as today. The lenders to look for here if you want to browse the Internet would be Lending Club, and Prosper Marketplace. I mean there are plenty of others. I don't mean to say there are some how better than others, but they've been there since the beginning, and they have pivoted along with the rest of this industry towards marketplace lending. Okay. So you were there to facilitate the discovery of the borrowers and lenders with this sort of idea was there all retail. There's just all, just people consumers living their lives, who either have money to put to work or they need some money right now. Okay. So how would this work? Well, you have borrowers, the borrowers if you wanted to borrow in this marketplace, you'd have to provide some financial information, some standardized financial information that they would need from you. So you would provide that, for people to look at. You could post a picture, and some people did of yourself, maybe you and your family. You would say well, here's how much I want to borrow, and here is the most I'm willing to pay. The term alone will be pretty much fixed. It would just be a mere on a three-year loan or five -year loan. So that would be pretty much standardized. Why would you go to this market? Well, then and now the main reason people go to borrow in this market is their credit card debt. All right. They've run up a balance on their cards. They're paying that big rate that, you'd see in the fine print, on your credit card statement, to pay in that big rate, and they want to refinance into something more manageable. Okay. This is by 80 percent of the borrowers. There some other reasons, but that's the big one. Okay. They could get the home equity line. They could go that route. Maybe some of them already have, and of course maybe the others are thinking well, I don't want to risk my house like that. I don't want to ramp up the risk of foreclosure especially after we've seen that with our own eyes so much 10 years ago. So okay. So they're trying to meet up each other and find the rate at which they're willing to transact. Let me just sketch just for a moment here, how as an economist would look at this problem of trying to find the rate at which we could transact. There's a fundamental problem that lending institutions everywhere encounter every day, and to an economist, the term of art for this problem is what they call credit rationing. The credit rationing. Let me give you just without any numbers or figures, just how to think about this problem of credit rationing, and how it can be a barrier to finalizing transaction on this marketplace. So here's the basic idea. Let's say we're thinking you maybe you're going to loan to me. Okay. Maybe you're going to loan to me. But the problem that you're going to encounter, is that we both know that, I know things about my risk that you don't know. All right. I know things about my job, my health, my plans, all sorts of other things, my investments, whatever it is, I know things about my risk that would be difficult or maybe even impossible for you to figure out for yourself. Okay. So the issue is that, if I let's say, I put a high probability on the outcome where I go bust. All right. That I just I can't pay it back, I privately know that that's a high probability. You know it's possible right, but I happen to know that actually I put high probability on that. Now, let's say you offer me, offer to loan to me at a high interest rate. Well, if I look at that loan at a high interest rate, I might look at that and say, well, you know what, it's on paper it's a high interest rate, but in fact if I put a high probability on not repaying the loan, then it doesn't really seem like such a high interest rate to me because I put a big probability on paying you back zero, right. Or something small right. So a high interest rate to me to someone who privately knows his risk is high, might not seem so big, might not seem so onerous to me, because I see that big chance of not paying you back. Whereas, someone who views, who privately knows that he's low-risk, then puts a very high chance on repaying the loan. So if it's a high interest rate loan, that person looks at is very expensive that I expect to repay this loan, and paying it back whatever, 15-20 percent, whatever it is you said, to me that's expensive, right. So the issue is that, if you loan, if you offer to loan to me at a low interest rate, then I would accept that whether I'm low risk or high risk. Okay. But the higher the interest rate that you offer, the more likely it is that the guy who privately knows he's low-risk is going to drop out and say, forget it. But the guy who privately knows that he's high risk is just still going to accept it. As you increase the interest rate that you offer, the average risk of the person who would accept that offer, is going up. Okay. So whereas you might think that if we're trying to find a rate to transact, there's got to be some, there's got to be some rate that at which we can agree. The problem is that, you might only be willing to go so far with the interest rate that you offer because you know that if you go above that rate, then the effect of going above that rate on the expected risk of the person who would accept that rate, is just too high. So you're willing to go so far but not above that in the rate that you'll offer me, and that's why people talk about credit rationing because normally in any market you think well, if there is excess demand to buy something at some price, then the price is going to go up to make supply equal demand. But in the lending market, if there's excess demand to lend at this rate, that doesn't mean that the price is going to change. There's only so much. People are only going to raise the rates so much. They're not going to go higher than that because of this negative effect of the rate that you offer on the risk of the kind of person who would accept it. So you get that kind of problem playing out in this market, and it makes it hard to transact efficiently. But people tried, that's how this market started. So in the original days, if I want to borrow, I'd say, "I want to borrow this much at a rate no higher than this." Then the lenders would say, " Well, I'm willing to loan whatever each one myself along a 100 bucks, 200 bucks, whatever and at a rate no lower than this." So a bunch of people making bids to lend and then you have the borrower saying, "How much he wants to borrow?" If everyone who wants to loan adds up to the amount that the person wants to borrow, then that's it. You've got a transaction and they set the interest rate at the market clearing rate, and they say, "That's it, okay. You guys have a loan." So the lender supply the cash, the borrower gets the cash, and then the idea is to borrow, then ultimately pays back the loan, and this is all intermediated by the platform. Okay. So that's how it started, and there was some of this in a small-scale. Of course, it didn't help very much that this started in 2006, and then very soon, well, we were in the recession which naturally didn't help the repayment capability of the people who borrowed. So what happened over the next few years is that the lending platform got more and more involved in the decisions. So they would start the lending platform would make its own credit risk evaluation and then set a floor on the interest rate for this particular borrower. After that, then the lending platform would start not to putting a floor on it but just say, "Okay, this guy we've looked at the credit, we've looked at the financials for this person, the interest rate is going to be 10 percent or 12 percent or eight percent." Whatever it is, they would set the rate. Then as an investor, you're not really bidding anyway, you just say, "Okay, I'll take some of that or I won't. I'll buy a hundred bucks of that loan or I won't." Then over the years, the platform has gotten more involved in setting the terms of the transaction, and the role of the investors, the lenders has gotten more and more passive. Just, okay yeah, it can't begin on that or no leave me out. So that's has evolved over time, whereas when you get to today, by now, about 95 percent of all lending on this platform is by not retail investors, but just institutional investors who just post a bunch of capital, they say, "Okay. Prosper or Lending Club. We want this 10 million bucks to just go to borrowers who show up on your platform and will just automatically buy whatever loans you approve within this or that risk category." The platforms themselves have greatly ramped up their ability to do this risk evaluation with this big data machine learning approach, so they've got all this data now. These days if you want to borrow on Lending club, you supply all sorts of financial information. They can browse all around your bank account and all sorts of other things, and they've got these algorithms they've developed for looking around your bank account and your other financials, and coming up with a risk score. They've gotten very good at this. One thing I've noticed is that back in the early days you might think, "Well, we don't need the lending platform to give us a risk score. We already have what's called the FICO score, everyone has a FICO score." The FICO score is a score developed by the Fair Isaac Company, and FICO scores are everywhere. You've probably seen your FICO score, you've used your FICO score when you bought a house or a car. So we have FICO scores, why don't they just use the FICO score? Well, they can do much better than the FICO score. Over the years, the correlation between the score that the lending platforms come up with in the FICO score has just gone down, and the lenders credit scores is actually now do a lot better in this context of predicting the outcome of these loans. If you talk to these lenders in this peer-to-peer or marketplace lending space, and you ask them, what's the social benefit of what you guys are doing? Well, they'll have various things they would bring up but one thing that comes up a lot is what they referred to as the invisible prime borrower. What they mean by that, let me show you some graphs here and I'll walk you through them. What they mean by invisible prime is that, if you look at all the people who are currently rated subprime due to their FICOs, and there's different cut-offs, but one popular cutoff when you think about is your FICO score to low the 680. If your FICO score is below 680 many people would say, "Well, you're a subprime borrower." FICO score which is calculated from your credit report is below 680, you're subprime. What the people at the Lending Club and Prosper and so on will say, "Well, okay. We can look at that population of people below 680, and we can find good credits. We can find people who really are prime even though their credit score says otherwise, and we're willing to loan to them at rates that reflect our belief that they really are prime." So here is a couple of graphs. Here, I got off of a paper written by some economists at the Federal Reserve Bank of Philadelphia. They got the data from Lending Club. If you look at these graphs here, let's look at the one on the left. This is showing you, what they're saying here is, look, looking just at those people with FICOs below 680. When we look at that group of people and we put them into risk categories based on our algorithm, from A risks which are the best risks to the G risks which are the worst risks. Look at the A risks, and remember, this is only from those people who have subprime FICOs, those people we do find these A risks are in that population and their chance of default over the course of the two years is just five percent, right? Five percent. Well, that's a very low chances of default, and that's a prime borrower. We found that borrower in everyone else we call subprime. Of course, we find some very high risks in that category too looking out at the G risks, they have a chance of default over 35 percent, that's very high, right? So yes, there are high risks there, but look, we can find the low risks. Look at the graph on the right-hand side. Here, what they're doing is they are within each of their risk categories, or A through G, that's Lending Club's risk categories. Within each of them, they're showing you the performance of those accounts that they put in those A through G buckets, broken out by what was their FICO score. So if you look on the very left, the borrowers above A, this is showing you that when they call you an A risk that you have a low probability default whether your FICO was below 680, which is the blue bars, or above 750 which is the purple bars. Yes, the people with the worst FICOs are in fact doing a little worse expose but still all of them are doing well as far as the risk of the mortgage. So it's basically showing you that as you sort people by our credit risks, we are capturing the lion's share of the actual variation in how risky people really are in the FICO score. On top of that, this isn't telling you very much. So they've developed an algorithm. Of course, they're looking at things your FICO score can't look at, your FICO score is calculated off your credit report. They are calculating this off of all sorts of stuff that the FICO score cannot be calculated off of. Your bank account statement isn't on your credit report. All sorts of things that they can see are not on your credit report. So this big data coming through for them allowing them to make credit decisions which are far more informed. All right. So that's the credit decision and the source of the funds as they say institutional investors just post cash and that cash is what's used to fund the loans. You also have loans that are securitized, which means that you have a big bundle of Lending Club or Prosper loans which are all pooled together, and then they issue securities backed by that pool of loans. Just because we had all sorts of securitization going into the financial crisis, didn't always necessarily work out that well. But that can be a great source of funds, is a way that institutional investors can buy these Prosper loan backed bonds and invest in those. Now, one thing people often say about this market and it's true is that, this market in this current business model hasn't been tested by a big downturn. So we'll see what happens when the economy turns out to the performance of these loans. It's going to be a test, and maybe they pass that test, maybe they don't, that's yet to be seen. So to summarize we have so far these Peer-to-Peer platforms, they're offering an alternative to consumers. An alternative way to refinance out of your credit card debt. You already had helocks that can give you a nice rate but there's a risk to that. You can go to this other platform instead, maybe can borrow at a better rate. The platforms have very accurate credit scoring algorithms that can find that invisible prime borrower, and bring to that person the kind of rate that they in some sense earned with their management or finances that the platform can see with all the browsing around that they do. This really isn't a retail investment play anymore. That's really not the way this works. It's not peers loaning peers anymore. It's big institutions posting the cash that is going to be invested based on these very sophisticated algorithms. Okay? So that's the peer-to-peer lending, and the next thing we're talking about in the next module is going to be the student loans.
