I’ve finally made it through the Prosper.com signup process (thanks to Andrew and Tim!). The root of my problem was with their verification system, which is the same one used by my brokerage firm. When I want to transfer or withdraw money, my broker verifies my identity by asking me personal questions from my past that only I should know. The problem is I’ve got a bad memory for details so when they ask me to, say, name the street I lived on for three months when I was a sophomore in college (actual question), I fail the test. It’s scary, some of the odd personal details they have in their files.

For anybody who hasn’t read my last post, Prosper.com is a way to borrow or lend money that completely bypasses banks. It uses a auctions to establish borrowing rates based on borrower’s credit scores and other personal information they care to provide. Lenders can spread their capital across numerous borrowers, minimizing the impact of a single default.

I transferred in $1,000 and started hunting for borrowers. There were thousands of people looking for money and I decided to do some basic analysis first. How do I know what rate to offer? Should I lend to riskier credit grades or stick with borrowers who have good credit?

Prosper gives lenders some great tools for researching their database. I calculated the average annual return for the different credit categories over two periods. Since Prosper is so new, I’m restricting my analysis to the last 18 months.

Here are the average annual returns by credit rating:

For June 2006 – June 2007:

For January 2006 – January 2007:

The data puts the sweet spot in low-risk loans to people with excellent credit (duh). According to this, I could invest my $1,000 with AA- and A-rated borrowers and expect to see an 8.7% - 9.3% average return. Practically this will be a 10% - 12% actual return, with a 20% chance of having a default on at least one loan. And the average return number seems to be rising.

Since Prosper has been in business for a short amount of time, much of this data could be misleading. Their default rate is only for a year’s worth of 3-year loans, so I’m assuming the defaults may end up being 3x what’s listed. Plus, we’ve been sailing through some relatively good times lately with low unemployment, great market returns, and reasonable inflation. If we hit a downturn, the default rates could end up being much higher.

Ok, I think I’ve made my decision. I’m going to do 10 loans @ $100 each to borrowers who have A or AA credit ratings. I lack the patience to analyze the loans on their merits so I’ll be randomly selecting them based on gut feel and potential amusement value.

I think I see why people are turning to online loans. I’d cross the street to avoid meeting some of these guys with AA credit ratings. If you were a lending officer at a bank, would you lend money to this guy?

The loan process isn’t instantaneous. I bid on each individual loan then wait a few days to see whether my bid was accepted. I’m putting bids in on loans for self-publishing, debt consolidation, satellite modem prototyping, startup funding, and boob jobs.

I’ll give you an update soon.

UPDATE:

As I was proofreading the submitted post, I took a closer look at the expected return numbers for high-risk categories. Almost *half* of the High Risk borrowers default on their loans. That’s just amazing.

When Congress eliminated the 18%/year annual cap on credit borrowing rates, I assumed it was due to undue influence from credit card company lobbyists. Now I’m looking at it in a different light. Allowing the credit card companies to charge >18% seems reasonable when you consider the outrageous default rates you see among people with bad credit.