Monday, March 31, 2008

Prosper At Finovate 2008

It looks like Prosper will once again be at Finnovate this year, amongst 40 young and upcoming financial startups:

Already 40 leading startups have committed to demoing their products (no PowerPoint allowed) including: Andera, Aradiom, Authentium, Boulevard R., Buxfer, Cake Financial, CAPS, ClairMail, Credit Karma, Diversinet, Expensr, First ROI, FindABetterBank, Guard ID, Guardian Analytics, IP Commerce, Jwaala, Lending Club, Loanio, Mint, Prosper, SmartyPig, SmartHippo, Simple Tuition, SocialPicks, TradeKing, TrustedID, Tyfone, Unified Money, VaultStreet, Vestopia, Vidoop, Wesabe, Wonga, WorkLight, Zecco and 4 more still in stealth.

This could be most interesting since Loanio is on the list. They've a supposed-P2P lender that has been running ads all over the web for the last year. They have yet to put up any substance, earning them the vaporware entry of the year.

And, I say to you Prosper, please have your media strategy determined before hand. We'll find out about the features that you'll be showing this year, like we did last year. Please, on please, talk about the features and don't go all cathedral on us by hiding the proposed features. Think of us lenders as partners here, after all, and we'd like to see where things are going too.

Prosper Big In The UK? Who Knew?

I found this little oops in the Warsaw Business Journal (English translation):

"In Poland, the difference between interest rates for loans and those on bank deposits is so large that [this situation] presents a great opportunity for social lending services like Monetto.pl," said Ɓukasz Banach, the president of Prender, the firm which operates Monetto.pl. ...

Social lending has already become very popular elsewhere. The largest social lending web pages are Zopa.com in the US and the UK's Prosper.com. In Poland, Kokos.pl and Finansowo.pl also offer social lending.

Let the Polish jokes begin.

Friday, March 28, 2008

More On Prosper Loan Diversification

A while back, I had written to encourage Prosper lenders to truly diversify their loans. It's not enough to spread your loans across different credit grades if they're all in California, for example. Such a lender will still be vulnerable to local economic conditions in California (as the current housing mess illustrates). There are two parts to understanding diversification for lenders.

To be successful, it's important to understand the nuances of diversification. One of the best books on the topic of diversification I've seen is "The Intelligent Asset Allocator" (it was recommended by RGF from the old forums). It covers the how and why of diversification in the context of mutual funds and describes the math in easy terms. The principles translate into the Prosper domain without too many headaches. There is a problem, though. Prosper is too young and the statistical reporting tools are too immature.

To succeed in diversification, it is necessary to identify different groups of investments and spread investment dollars between these groups. Prosper makes it easy to verify performance for different credit grades and then for listings, but it stops there. The borrower's state is the second-best grouping criteria that's Prosper verified. There are no readily available tools, Prosper or otherwise, to succinctly report performance by state or a lender's loan distribution over states. Beyond this, there are self-reported criteria from borrowers like occupation and the borrower's reason for getting the loan. If you assume that the borrowers self-report accurately (a big IF), there is insufficient statistical information available to determine the performance for these sub-groups, and it will take a while to build a sufficient loan history to produce meaningful information.

This situation, however, is an opportunity for either Prosper or the independent sites like Lending Stats and Erics Credit Community to step up and provide the next layer of analysis to aid lenders in truly diversifying.

Thursday, March 27, 2008

FDIC Thinks Bank Failures Coming

Yeah, it's a double-post day. I just spotted this on Calculated Risk. The AP is reporting that the FDIC is staffing up for an anticipated rush of bank failures.

The Federal Deposit Insurance Corp. wants to add 140 workers to bring staff levels to 360 workers in the division that handles bank failures, John Bovenzi, the agency's chief operating officer, said Tuesday. ...

There are 76 banks on the FDIC's "problem institutions" list — which would equate to about 10 expected bank failures this year, though FDIC officials declined to make projections. Historically, about six banks fail per year on average, FDIC officials said.

Since 1981, total failures per year averaged about 13 percent of the number of institutions that started the year on the agency's list of banks with weak financial conditions.


Now here's a topic that I'd like to see Chris Larsen address in one of the monthly marketplace summary commentaries. He's been very upbeat about otherwise ugly financial events, and I do have to admit that there is a good "cup-is-half-full" for Prosper in this otherwise troublesome bit of financial fortune telling. I'm sure it'd go something like this:

The recent turmoil in the banking sector has provided great opportunities for Prosper lenders. As the number of banks shrink, borrowers should be more willing to look toward Prosper for loans, seeing how they have so few other alternatives.

The other thing to note is that Prosper, being a marketplace instead of a traditional bank, will most likely not be brought down by the banking implosion. For good or bad, Prosper is a conduit that directly transfers the borrower's risk to the lenders, bypassing itself (this is why it's important for lenders to lend conservatively). As long as Prosper can make borrowers appear and find some return-hungry lenders, they'll survive. And I do believe that more borrowers will appear as banks tighten up their lending standards.

What A Difference A Month Makes

February was not Prosper's best month, with hitting a low for loan dollars originated. This shortfall was easily anticipated by the lower number of active listings (~1700) Prosper had during mid February. A fair amount of favorable press and Prosper Days appears to have jump started their originations. Prosper has about 2500 active listings as of this moment (9:01 am CST - your mileage may vary) and is on track for another $7.1M month. In other words, they were successful at hitting a critical March milestone of attracting more borrowers. Prosper has a large uptick in active borrowers, approaching almost 7k this month thus far compared to 6.3k in February.

I did a quick survey of the portfolio plan criteria, and it appears that lenders are still pouring more money into the portfolio plans than there are borrowers to fund. There aren't any loans inside the portfolio slices that I checked that aren't fully funded and at the minimum portfolio bid. It's a good place for Prosper to be operating. Now, if AmSher could pull up those collections statistics...

Monday, March 24, 2008

Prosper Pitching To Hedge Funds

I almost overlooked this little nugget. Prosper has taken to pitching it's lending platform as the perfect solution for hedge funds to get into the consumer credit market.

“Prosper is America’s largest peer-to-peer marketplace with over 600,000 members,” stated Kirk Inglis, CFO of Prosper. “As credit markets experience unprecedented changes, institutional lenders, including hedge funds, are using Prosper to diversify portfolio returns without the lack of transparency and fees associated with structured consumer debt products.”

Bear Sterns has shown how well hedge funds handle risky marketplaces. I'm not sure I like the idea of these guys throwing weight around in the Prosper bidding process.

Update: If you're coming over from P2P-Banking.com (Wiseclerk's site) and can't figure out the thing about Texas, try this link. Now if I could find Wiseclerk's e-mail address...

Friday, March 21, 2008

Office Linkfest

Hello, Mike. What's happening? Uh… we have sort of a problem here. Yeah. You apparently didn't put one of the new coversheets on your Prosper linkfest reports. Mmmm. Yeah. You see, we're putting the coversheets on all the Prosper linkfest reports now before they go out. Did you see the memo about this? Yeah. If you could just go ahead and make sure you do that from now on, that will be great. And Uh, I'll go ahead and make sure you get another copy of that memo Mmmm, Ok?

In other words, work's been kickin' my tush this last week, hence few posts. Fortunately, it's nothing that a good linkfest can't fix up!

LoanChimp continues to finish out his increasingly ill-named 3-part Prosper Days summary:
HollowOak walks off the field. Sniff...

Motherdust teases out statistics for social lending factors.

The social factors that seem to be most relevant are multiple friends bidding, and/or friends who bid substantially ($300+). These factors are at least as important as autofund, low DTI, and even current DQ’s. When we combine the fundamental and the social factors, the default rate improves dramatically.

The Prosper Blog had their own version of the linkfest.

Prosper Lending Portfolio wonders about loans in Pennsylvania

What were these lenders thinking? Interest rates of 1% and less when Prosper also takes a percentage point of the interest paid. 47411 was a money loser automatically — even before default. These loans do not make any logical sense. So, I must check… How are BorowersGL and MuleShoes doing on their rate of return on Prosper peer-to-peer loans

Tom at Prosper Lending Review did a survey on whether people should dump tons of money into Prosper as a conservative investment strategy (want to guess at my take on the situation?)

Tuesday, March 18, 2008

What Should My Target Prosper ROI Be?

With the Fed dropping interest rates like they're going out of style, it's an appropriate time to discuss what average annual return to expect from Prosper. The easiest way to look at it is to say "I'd like to get 10%". Unfortunately for us individuals, this is not how the finance world operates, and, in an bidding system like Prosper, this matters.

We're competing for borrowers who have alternatives like traditional banks. These banks set the rates borrowers pay based on the prime rate plus some fixed offset to compensate for an individual's credit risk. This effective indexes the bank's return to the prime rate, which is, in turn, indexed to the Fed Funds rate (that quickly dropping number). In this same vein, lender should index their expected return instead of always anticipating a certain return amount.

Ok, so what should lenders use as an index? I have often looked toward a number that corresponds to the risk-free rate of return I can get on my money. As in, how much can I get for no effort and virtually no risk? My index of choice for Prosper loans is my bank's 3-year CD, though this can be approximated by Bankrate's 3-year CD list. I then add a percentage, typically 5%, on top to provide additional cushion for the risk. This is my target average annual return on Prosper.

My long-time target was 10%, back when the economy was fine and CDs were 5%. And, despite some rather bonehead newbie bids, this was still achievable. Since that time, I've revised my target down to 8% as interest rates have fallen. I suspect that I'll be revising downward again real soon.

To use this number, I make sure that my estimated return for every bid exceeds my target average annual return. Happy bidding.

Monday, March 17, 2008

Does Idle Cash Matter?

There has been constant griping about idle cash not earning interest on Prosper. I had always assumed that the effect was small and that Prosper had better things to be doing with their time. Like fixing collections. I had never bothered to calculate it out. j9359 did and proved that it doesn't matter too much:

For my account the difference is minimal, one or two basis points once past the initial period. But I don't put a whole lot of cash into Prosper and then let it sit idle.

That's right. If there was interests on his idle cash, it'd increase his ROI 0.02%. I can understand why Prosper is trying to improve the pool of borrowers, improve collections, implement a secondary market, enhance statistical reporting tools, or just about anything else ahead of getting interest on idle cash.

Friday, March 14, 2008

Friday Fun: Gordon Bailey Quotes

Prosper CEO Chris Larson has been pushing the idea that Prosper combines aspects of George Bailey and Gordon Gekko to do something in the financial marketplace. What insightful quotes might we expect to hear uttered around this Mr. Gordon Bailey?

  • Look, Daddy. Teacher says, every time a Prosper loan funds an angel gets his wings.
  • A toast! A toast! A toast to Mama Dollar and to Papa Dollar, and if you want to keep these loans out of collections, you better have a family real quick.
  • Greed is good. Now, come on, get your clothes on, and we'll stroll up to my car and get... Oh, I'm sorry. I'll stroll. You fly.
  • I don't throw darts at a board. I bet on sure things. Read Sun-tzu, The Art of War. Every battle is won before it is ever fought. And I'm gonna build airfields, I'm gonna build skyscrapers a hundred stories high, I'm gonna build bridges a mile long.
Update: And after watching the stock market today... (hat tip)

Thursday, March 13, 2008

Final Prosper Days Roundup

In the final roundup of Prosper Days commentary (previous ones), LoanChimp did a 7 part recap of his Prosper Days experience. LC was notable for being the most experienced lender on the high-ROI panel, and spread his chimpy-wisdom to all who would listen. Go bananas.

Wednesday, March 12, 2008

Post-Prosper Days Interview In WSJ

This is the first post-Prosper Days interview I've seen with Chris Larsen (which is not to say that there havn't been others, just that I havn't seen them). He's picking up several themes that were rolled out at Prosper days and taking them around. This includes benevolent greed:

The Wall Street Journal: How would you describe your business?

Chris Larsen:
An eBay for money and credit. Borrowers can make listings and any American with $50 or more are free to bid. It's combining community banking with the high finance of Wall Street -- bringing together George Bailey with a little Gordon Gekko.

And stricter enforcement on fraud (ahem):

WSJ: What steps has Prosper taken to address concerns about fraud or identity theft?

Mr. Larsen:
We have a 100% fraud ID theft guarantee. If there's a fraudulent borrower who gets through our screens, we'll buy that loan back. Beyond that, we're also very aggressively prosecuting people and we did have our first conviction.

They're good themes for talking to the press, so expect to hear more of them in the future.

Tuesday, March 11, 2008

Prosper's February Lending Market Survey

Prosper has released their February lending market survey, and, once again, my great enjoyment is perusing Chris Larsen's take on the marketplace:

In February 2008, the percentage of prime loans funded on Prosper once again hit record levels, accounting for 43% of originations. At the same time, the percentage of prime listings on Prosper hit an all time high of just over 18% a big jump from 9% in February 2007 and the previous high of 12% in December 2007 while the percentage of sub prime listings hit an all time low of 33% and accounted for a mere 6% of funded loans.

Some other key metrics we watch closely include the type of listings that are created and funded in the Prosper marketplace. Very broadly, we look at listings that, based on historical Prosper loan performance data, can be made at an attractive risk-return tradeoff and those that can only be made at an unattractive risk-return tradeoff. By providing more robust information to lenders on the expected returns of listings, we have seen an increase in originations from attractive risk-return listings of over 200% and a decrease in originations from unattractive risk-return listings of 80% over the course of the last year.

As we discussed at our Prosper Days community conference, these dramatic and constructive shifts in the marketplace have been driven by three key factors: the pervasive credit crunch and sub prime mortgage meltdown; recently introduced performance data-driven tools and features; and, increasing mainstream acceptance of Prosper as an attractive funding source and asset class.


In other words, Chris Larsen is pulling an anti-Rodney Dangerfield. He thinks that Prosper is getting respect and becoming a viable alternative to banks. He also mentions that Prosper is gaining acceptance as an asset class, but this is not Prosper's primary problem - Prosper needs good borrowers.

Also of note what looks like solid indicators that the average loan rate has very poor correlation with the Fed Funds Rate. Chris was big on this concept back in September and October. The summary notes that the average "Prime" borrower rate was 10.32% in February 2008 while it was 9.23% in February 2007. Of course, the Fed Funds rate was 5.25% in February 2007 and it's floating around 3.0% now (get ready for that to change). The irony of this is that the falling Fed Funds rate will decrease Prosper's competitiveness. Banks tend to index their personal loan rates to something that derives off of the Fed Funds rate. If Prosper lenders do not follow suit, Prosper's loan rates will not drop to match the banks. Quite the sticky wicket.

Friday, March 7, 2008

Ninja Stealths Prosper, Prosper Responds

As Tom over a Prosper Lending Review has noted, a ninja has managed to spot a CSS vulnerability in Prosper's website. Prosper has provided an official reply to this:

We were contacted by this blogger about this vulnerability in Prosper’s site on February 15, 2008. Since we were contacted, we have made the code change that will eliminate this vulnerability; although it has not yet been rolled out (a release is expected this weekend). We appreciate the blogger’s help in finding these vulnerabilities.

XSS attacks can introduce significant security issues. We are investigating right now whether this kind of attack can actually do anything malicious on the Prosper site (many security mechanisms are already in place). Nonetheless, there are no known cases of hackers exploiting these vulnerabilities to date. As I mentioned, we are planning to release a fix shortly.

Discussions will probably continue on the prospers.org forum (reg required - sorry)...

Update: I'll note that it looks like the security hole is already closed, but that could be my version of Firefox.

Update 2: Revised Prosper response

Thursday, March 6, 2008

Simple Receipe For 100% Funded Prosper Loan

As I had previously observed, Prosper seems to have more lenders chasing loans than there are borrowers to take the money for some classes of loans. One such class encompasses just about anything that's inside one of Prosper's standard portfolios. If a borrower wants to get funded, all they need to do is qualify for one of the portfolio slices (I count 21 slices spanning AA - D loans) and choose a starting interest rate above their appropriate slice's minimum bid rate, and they'll be 100% funded.

And this is without all the inconvenience of having to write a description, prepare a budget, have a plan for what to do with the money, or have much of an income. Someone could sign-up, verify their ID, write "I promese to paid it too u back", and they'll be fully funded. The only requirement is to have a sufficient credit score and be able to prove your income, should Prosper challenge the stated income. The second problem can be avoided by claiming 1 dollar a year in income a low income.

This is why I don't use the portfolio plans and why I think the portfolio plans will have degraded performance relative to the predicted performance. Manual bidders won't bid on the outliers who pull the slice's estimated return numbers down. Eventually, the statistics may even out, but it'll take more time, and the current portfolio users will get frustrated in the process. The size of this problem will be the guiding factor for how much of a performance delta can be expected.

Update: ZCommodore has noted that most slices have a DTI requirement, typically DTI <= 40% or DTI <= 70%. Income will have to be slightly more than $1, but $10k - $20k should do nicely. The downside to all this is that, at Prosper Days, Prosper indicated that they're less likely to verify small incomes because they find it very likely to be true.

Wednesday, March 5, 2008

Prosperous Land Survives One Year

Yup, it's been a year since I got started with this whole thing. I was naive then. I'm not much better now. So, crack your favorite beverage in celebration of my unwillingness to go away quietly.

Carnival Of P2P Lending #5

I'm hopping onto the Carnival of P2P Lending Bandwagon. This week it's hosted by The Amatureist Financial Journey

This week’s Carnival of Peer to Peer Lending is a little more intimate than the other super huge carnivals. Only six articles were submitted this week, and they all were ether background on lending, or lending news. Enjoy the reading, and let be sure to leave an note on the blog, letting them know how you found their article.

Introducing P2P Lending

One of the first questions you need to ask yourself is “are peer to peer loans guaranteed?” (Posted at Cash Money Life.)

Once you decide you want to lend, now what? Pinyo has the answer an in depth Lending Club and Prosper comparison (Posted at Moolanomy.)

Brip Blap discusses the specifics of Lending Club? (Posted at brip blap.)

P2P Lending News:

Mike noticed January, Prosper’s monthly loan totals have dropped to about $5.7 million., and discusses the impact of decreasing membership and Prospers new emphasis on portfolios. (Posted at Prosperous Land.)

Prosper’s decline isn’t surprising considering Lending Club loans just reached 8 figures. (Posted at Lazy Man and Money.)

Tom shares highlights from Prosper Days 2008: Keynote from CEO Chris Larsen (Posted at Prosper Lending Review.)

Monday, March 3, 2008

Prosper API And XML Workie

So sayith the Prosper:

Dear So-And-So,

Thank you once again for your patience as we worked to resolve the problems with the API. We apologize for any inconvenience this may have caused you.

The XML was fixed this weekend and we consider the issues resolved and expect no further updates.

We are committed to adding the Prosper calculated ROI values. We will update you with more information regarding the Prosper calculated ROI values as soon as possible.

Sincerely,

Prosper

But, most importantly, LendingStats is updating their information again.