Print

[ira01]

Excellent. You should send a link to the Washington Post.

Actually, a better place to send it would be the SEC -- Presumably, the Washington Post reporter got that number from Prosper.  If so, that sounds an awful lot like a material misrepresentation in connection with the sale of a security -- otherwise known as securities fraud -- to me.   Of course, TINLA. 

[Senator]

Maybe it should read as follows:

http://www.ericscc.com/lenders/anton

CL himself is a lender on 1,040 loans, and 200 of them, roughly 10% per year, have defaulted.

Were you in the same math class with the Washington Post reporter?

[lenderguy]

Excellent. You should send a link to the Washington Post.

Actually, a better place to send it would be the SEC -- Presumably, the Washington Post reporter got that number from Prosper.  If so, that sounds an awful lot like a material misrepresentation in connection with the sale of a security -- otherwise known as securities fraud -- to me.   Of course, TINLA. 

My "off the cuff" attempts at annualizing the default rate yield a site-wide 6.3% or so.  Not sure how much the SEC is going to bust Prosper's balls over that one.

[xraider]

The easiest way to fudge the numbers on defaults: Change what you call 'em to "charge offs."

[bamalucky]

The easiest way to fudge the numbers on defaults: Change what you call 'em to "charge offs."

That could be it.. They are saying 5% have no chance of recovery due to debt sales & bankruptcy

[112233]

The easiest way to fudge the numbers on defaults: Change what you call 'em to "charge offs."

is a default now a contractual event strictly between prosper and the borrower? in other words, can they charge off a loan to the lender but not declare it default to the borrower? it doesnt make sense those would not go together, but how else could you claim 5% default?

[ira01]

Excellent. You should send a link to the Washington Post.

Actually, a better place to send it would be the SEC -- Presumably, the Washington Post reporter got that number from Prosper.  If so, that sounds an awful lot like a material misrepresentation in connection with the sale of a security -- otherwise known as securities fraud -- to me.   Of course, TINLA. 

My "off the cuff" attempts at annualizing the default rate yield a site-wide 6.3% or so.  Not sure how much the SEC is going to bust Prosper's ballsover that one.

As Fred93 explained in his OP blog post, the annualized rate is about 16% -- more than TRIPLE the rate claimed in the WP article.  The SEC most certainly should bust Prosper's balls over that.

[lenderguy]

Excellent. You should send a link to the Washington Post.

Actually, a better place to send it would be the SEC -- Presumably, the Washington Post reporter got that number from Prosper.  If so, that sounds an awful lot like a material misrepresentation in connection with the sale of a security -- otherwise known as securities fraud -- to me.   Of course, TINLA. 

My "off the cuff" attempts at annualizing the default rate yield a site-wide 6.3% or so.  Not sure how much the SEC is going to bust Prosper's ballsover that one.

As Fred93 explained in his OP blog post, the annualized rate is about 16% -- more than TRIPLE the rate claimed in the WP article.  The SEC most certainly should bust Prosper's balls over that.

I should have clarified: My "off the cuff" attempts at annualizing the default rate from Prosper's 20.1% figure yield a site-wide 6.3% or so.

[Fred93]

I should have clarified: My "off the cuff" attempts at annualizing the default rate from Prosper's 20.1% figure yield a site-wide 6.3% or so.

Yea.  Many people may do a similar thing.  Unfortunately, the 20.1% number is for a mixed-age portfolio which includes many very young loans, in fact some so young that they could not have possibly defaulted yet no matter what the behavior of the borrower.  There is really no way to "annnualize" an aggregate statistic for such a mixed-age portfolio.

[lenderguy]

There is really no way to "annnualize" an aggregate statistic for such a mixed-age portfolio.

Although what you said is true, I will take issue with this part.  Much of the mathematical modeling done with real world data is never going to be perfect.  It's a fact of life.  We can only do the best we can with models, and accept that there are flaws, imperfections, or limitations to them.

[HollowOak]

I think at issue here is a new venture like Prosper and an established venture. In a new venture, it is not possible to present the same, steady-state data as in a venture that is in steady state and has (in the Prosper case) loans in all stages of the process.  Prosper's "pipeline" if you wish is still full of product of varying descriptions, quality and other variances that make steady-state comparisons and annualized rates less reliable than, for instance, bonds or other finnacial instruments.