I was searching through the top-performing lenders, and came across Fasterfour. I remember him well, as we discussed his portfolio multiple times in the old .com forums. After I had PMed him, he was kind enough to to give his thoughts nearly one year ago. His portfolio continues to be one of the best there is. Today, he is the top lender in terms of ROI on Lendingstats for lenders with >100 loans, and >12 months at 12.82%.
http://www.lendingstats.com/lenders?loanCountFilter=3&loanAgeFilter=5&loanAmountFilter=0&submit=Filter&sort=estimatedROIColored&sortDirection=DESCA year ago, he posted this extremely well thought-out post explaining why he believes that Prosper isn't a good investment. He, to this day, still has not made another bid since his last loan nearly 18 months ago. If the top performing lender believes that he's not being compensated for risk...are the rest of us?
http://www.prosperreport.com/prosper/forums/archive/threads/2/9/8/29818.0.HTMWell, it's been 1 year since my first loan originated on Prosper on 8/17/06, so I figured this would be a good time to make my first post on the forums. As one of the more successful lenders on Prosper thus far, perhaps some of you may find my thoughts on Prosper interesting.
When I first started lending on Prosper, I viewed it as an experimental asset class. It had great potential as an uncorrelated investment in a new market showing significant indications of inefficiency. With limited information on default rates, lenders with no experience in consumer sub-prime debt, and autofunding loans, Prosper appeared to be rich with opportunities for substantial returns. I decided to commit $10,000 (<1% of my net worth) to Prosper in order to test the waters. Once this initial sum was fully invested in a diversified portfolio of loans, I would allow the loans to age for a period of time in order to evaluate default rates, before committing to a substantial allocation in Prosper loans.
So I went about determining what interest rates to bid. Like most other lenders on Prosper, I had no idea what would be a typical default rate on sub-prime consumer debt. I started by looking at the Prosper default rates, but determined that they weren?t terribly useful as they represented credit card debt default rates on borrowers with low DTI. Furthermore, there appeared to be significant differences in risk between loans within the same credit grade. More research would definitely be needed. Unfortunately there isn?t much publically available information on sub-prime consumer debt default rates. The best source I found was the graphs on p2ploananalytics website. From these I was at least able to determine which factors had the largest effect on default rates. As most lenders eventually figure out, current delinquencies, total delinquencies, # of lines of credit, and # of inquiries are way more important than DTI in determining default rates.
The other source of information I looked at was drawing parallels to low grade corporate debt. The long term variance of junk bond default rates with business/credit cycles was another area that must be factored into setting interest rates on which to bid. Chart 3 on this website: http://www.efalken.com/banking/html's/defaultcurves.htm shows the huge swings in low grade corporate debt default rates. While corporate bonds aren?t a direct proxy for unsecured consumer debt, I felt that given the currently (2006) low default rates and risk premiums for all types of debt meant that there was a risk that default rates could increase significantly in the future. Given recent events in credit markets, it appears that accounting for this risk is going to be important. I expect that going forward we?ll see increasing default rates in Prosper (as scary as that is given how high default rates have been already).
Finally, there was the Prosper business risk that needed to be factored into setting interest rates. At the time my thoughts all focused around the Prosper business failing and the resulting issues with servicing the loans that would occur in that situation. I, along with most other lenders, failed to recognize that the greater Prosper business risk to lenders was not of outright failure of the business, but of Prosper failing to operate with lenders interests in mind.
So putting all this together, how did I go about bidding? I didn?t bid on HR loans as the rate caps and high default rates made these loans unprofitable. I limited my risk by generally not bidding on loans with more than 1 current delinquency, more than 5 inquiries, fewer than 5 years credit history, fewer than 9 credit lines, etc. I scaled my bids within a credit grade according to the riskiness of the loan. I would bid much higher rates on someone with a current delinquency or 5 inquiries, than on a clean loan. I also set my interest rates to require a significant premium above the risk free rate plus expected default rates. This premium increased for each lower credit grade in order to account for the expected higher default rate variance in the lower credit grades. The end result was that I ended up with very few AA/A loans. The rates on these loans are too low respective to the risk free rate give all the risks associated with unsecured consumer loans. The majority of my loans are in the C, D, and E credit grades.
I also focused on setting up standing orders for autofunding loans as a part of my strategy to exploit inefficiencies in the Prosper market. A significant portion of my loans are autofunding loans that were posted at rates well above the typical funding rate. These loans would typical fund in a very short time period, necessitating the use of standing orders. While autofunding itself may be a risk factor and many lenders have avoided these loans for that reason, I believed that with appropriately high rates set in the standing orders, the risk would be compensated.
How has it all worked out after a year? I never reached the $10000 goal I had initially intended to invest. I loaned out $8800 in mostly $50 increments across 185 loans. In Feb. 2007, interest rates had dropped to levels where I was finding few loans to bid on and winning even fewer. I decided that it wasn?t worth the effort involved in continuing to search and bid on loans. The number of loans and amount invested was sufficient to evaluate Prosper as an asset class. On 3/8/07 my last loan originated.
Since that time I have had a number of delinquent loans, which was expected. However, what wasn?t expected is the horrific cure rate Prosper has for delinquent loans. In Prosper 1 month delinquency is pretty much the same as default. B and E credit grades have been the best performing thus far. While initially performing well, the autofunding loans have recently become the bulk of my lates. My sample size is too small to draw any definitive conclusions, but it does appear that autofunding is a risk indicator of distress that has manifested itself now that the credit cycle is contractionary. Thus autofunding loans perform like a higher risk, lower credit grade by exhibiting higher default rate volatility. Currently LendingStats has my estimated ROI at 14.8%. I expect this number to decline over time as default rates on Prosper loans continue to increase. You can look at my portfolio in detail here: http://www.lendingstats.com/memberProfile?...erId=Fasterfour
So is Prosper a valid asset class for investing? No. First, while I expect to be profitable and earn a premium over the risk free rate, there is still a significant risk that I won?t. The premium that it is possible to earn on Prosper is just too low for the risk involved. Second, as many others have posted, Prosper does not operate their business with lenders? interests in mind. So while it is possible that interest rates in the future may increase to levels that provide appropriate premiums over the risk free rate, I still wouldn?t consider it to be viable due to the misalignment of interests between Prosper and lenders.
Only time will tell how this is all going to turn out, but it has been an interesting experiment none the less.
I would like to nominate this for the lobby.
Author
Topic: One of top lenders on Prosper:This is not a viable investment class (Read 5530 times)