So it's been awhile since I've published any thoughts on Prosper.com. I used to be hugely in favor of the website and the idea of person-to-person lending. Now I'm taking much more of a wait-and-see attitude. I've been through my first default (a loan that the some guy took out and then never made a payment on, I got about 1/6th of my money back at the debt sale) and now I have two loans that are Late. If both of those loans go bad, my overall performance on Prosper.com will probably lag a savings account such as ING Direct.
Not surprisingly, the missed payments and defaults have come from the lower credit grades in my portfolio. Ideally, the amount of interest that you receive from lower credit grades would make up for the increased risk of default. As you can see from Prosper.com's handy little "Estimated ROI" tool, this is not the case (scroll to the bottom of the page).
Essentially, from the lender's point of view, there are three types of people - people where you can get a much higher ROI than from a savings account (credit grades AA, A, and B), people who will give you about the same returns (C and D) and people who will not pay you back (E and HR).
Looking at those statistics is scary. Over 1 in 4 'E' credit grade loans made between June 1, 2006 and June 1, 2007 has defaulted. Since it's actually impossible for loans made between February 1, 2007 and June 1, 2007 to have already reached default, that means that 'E' grade credit is ridiculously risky. It'd be the dumbest thing you could invest money in, if it weren't for the existence of HR grade credit.
Looking at October 1, 2006 to December 1, 2006 (a period of time after Prosper.com had kind of caught on and started reaching a larger audience, but long enough ago that loans have aged a bit), only 80% of loans to 'E' grade credit are current. Based on the Estimated ROI chart from Prosper.com, in order to break-even loaning to people with 'E' credit, you'd have to get an interest rate of 36% APR.
Herein lies the true usefulness of Prosper.com from an academic/policy standpoint. Prosper.com the company does not make money from pricing credit, they make money on small transaction fees. Therefore, it is in their interest to publish as much data as they can to allow individual lenders to price loans. The side effect is that the world can see exactly how risky unsecured loans can be. Whenever policy makers see interest rates on loans reaching or exceeding 30% APR murmurs of "predatory lending" begin to become audible. However, there is growing evidence from the experience of Prosper.com lenders that there is a large population that you can not profitably loan to at rates less than 36% APR. Finally, we have a set of publicly available data that shows that the interest rates on high-rate loans are a reaction to the risk of those loans.
Actually, I should qualify that last statement. While rates on 'E' grade credit are much higher than the good grades (AA to B), over almost any period that I choose, the "Estimated ROI" based on default rates is negative for 'E' grade credit. In essence, even though the interest rates are higher, they still do not accurately price in the risk of the lowest credit grades. Prosper.com lenders are giving away huge sums of money to people with lousy credit.
In fact, if you take a look at the "Estimated ROI" tables for almost any time period you will see that if you lend to AA to B credit, you can expect a 8-9.5% ROI after fees and defaults. C and D credit are 4-7%. E credit is usually single-digit negative (-4% to -8%) and HR credit is hugely negative (-25%)!! This is not just irrational, it's downright crazy.
I think this has a lot to do with a learning curve. When Prosper.com started out there was a lot of talk about "community lending" and helping people with poor credit get back on track. There wasn't much for historical data to go on and 25% APR looks awfully attractive. Also, new lenders seem to hit Prosper.com in waves. Once the early adopters figured out that loaning to HR credit was equivalent to flushing their cash down toilet, they moved to the higher credit grades. However, many more people were just discovering Prosper.com and getting seduced by the high APR's that people with low credit scores will accept.
It's easy to see why people would get seduced. I would guess that most lenders are probably 'AA' to 'B' credit people who, upon entering into a legal contract, tend to fulfill their obligations. There is just a different mindset between 'A'' credit and 'E' credit. One would expect the market to rationalize a little and charge higher rates to 'C' and 'D' credit and stop making loans to 'E' and 'HR' credit altogether (because the 'E' and 'HR' break-even loan rates are above most states "anti-predatory lending" rate caps). Ideally you would expect the "Estimated ROI" data to converge, probably to something around 8% (seems like people feel like the unsecured loan market needs to offer at least 3% over a risk-free investment to take part at all).
So what conclusion do I reach? I'm not adding any new money to my Prosper account, but I am also not taking any money out, for now. I get enough money in payments each month to make one or two $50 loans, so I will stay involved, but less active than I had been. I will not loan to anyone with 'C' or lower for a credit score (unless I personally know the borrower and think they'll pay the loan back).
Obviously, I do feel some sort of guilt about writing off large swathes of the population as too risky for lending, but at the current going rate for their loans, it's just a losing proposition. Am I to believe that there is some inalienable right to credit access?
Still, the guilt persists. Why can't people just repay their debts or learn enough math to figure out a monthly budget? I suppose I might as well ask that no one drop out of high school or grow up in an environment that rewards illegal behavior.
I really ought to get a little more involved in some sort of community volunteering I guess. Try to explain to some children from disadvantaged backgrounds how the majority of the U.S. economy operates. But how do you actually explain to children that everything they've been told is a half-truth? How can I explain to them that the secret to financial success is delayed satisfaction, when all they see on television (probably their only access to the wider world) promotes instant gratification? I suppose that's not something you can just pronounce to a room of teenagers, it's one of those things you have to fool them into discovering themselves.
That's probably a subject for another post, perhaps I'll think of some innovative program for youth to discover compound interest and how it can affect their lives.
Sunday, July 01, 2007
Subscribe to:
Post Comments (Atom)
Posts
7 comments:
As a young American who has had little in terms of financial advice from his schooling or his parents, I would be very interested in a post about basic financial savvy. I plan on graduating with my bachelor's within the year, and I want to feel secure in my skills when I have real money to lose.
I couldn't agree more with your comments. One way I got fooled by prosper was looking over their marketplace performance data. If you have a look at that you can view the stats on all of the loans active to date. The numbers there indicate defaults in 87 of 1837 in the E credit rating and 153 out of 2155 in the HR group which is quite a bit lower than what I've seen in my small amount of lending.
It would be interesting to know, even anecdotally, why different people default. Do they have different expectations about their loan than the lender had? Did they just plan poorly? Did they really intend to pay back the loan? Is their default part of any larger economic problems in their life, or isolated to the scope of the loan?
Great post. I suppose that before you lend money to someone with "E" or "HR" credit scores, you need to briefly contemplate the fact that people don't earn themselves credit ratings like that by reliably repaying their debts.
I also agree with Mr. Bicking's post. I think it would be fascinating to see the thought process that leads up to defaulting on a Prosper.com loan. Also, it would interesting to see the relative default rates for Prosper.com vs. other, more traditional unsecured loans.
Great post. I wonder if there would be some criteria to mine for within the E/HR category that would show a more consistent payment record. For example, one blog I read suggested people with recent delinquencies were more risky than those who had none. But there may be other criteria, like someone who just got married is more likely to pay, etc. We could at least then direct loans to the subset that would yield the higher APR without default.
You could lure them into playing one of the many Flash-based games out there that incorporate compound interest in their game play mechanisms.
No, seriously.
The "tower defense" (TD) games are a prime candidate for this kind of accidental learning. A great example (and fun to play as well) is Vector TD. The game play is basically to place defensive towers that can defeat waves of enemies; the towers cost money, and you make more money as you defeat enemies, but you also accumulate interest after every wave. There are also occasional waves with bonus points that allow you to purchase upgrades, one possible upgrade being an increase in the interest rate.
In my experience, the best way to reliably win this game is to keep yourself well in the black, spending only as much as you need to, and always taking the increased interest rate upgrades. If someone doesn't realize this on their own, encourage them to try it. By the time you get 2/3 of the way through the total number of waves of bad guys, you can see a HUGE difference in what you're capable of buying.
You make some great points. Unfortunately many lenders learn these lessons after they have lost money. Now that Prosper has been around for over a year I think that there is enough information to learn from the experiences of others. Here's another good article that makes many of the same points you made:
Prosper: A hands-on education in risk management
Post a Comment