[PeerSocialLending]

I found this Forbes article pretty fascinating and would love to hear some thoughts from others.  It basically talks about how all loans are created equal and that there is no evidence that any investor can consistently generate better than average returns. Obviously, this doesn't include the folio folk. 

http://www.forbes.com/sites/stevenbertoni/2014/05/12/no-hedge-funds-arent-hijacking-peer-to-peer-loans-says-lending-club-ceo/

[SBryantMS]

I disagree on one major point -- not 45 risk buckets -- but seven risk buckets for me (A - G).

I do believe that there are better loans (lower default potential) in each risk bucket.  I use P2P-Picks to identify those loans.

I also believe that statistically that there are better loans (do not payoff early) in each risk bucket.  I don't have a tool for that loan type. 

With P2P-Picks, I am riding about 1.5% higher on the wave when I look at LC's graph that compares me to similar lenders.
     

[daniel2023]

from: PeerSocialLending on May 12, 2014, 01:26:10 PM

[Zach]

That's a great article. Thanks for sharing!

[Lovinglifestyle]

"There is no evidence that any investor has generated better than average returns on the platform consistently."  (from the article)

Are Folio users excluded from all that "evidence"?  Are their Folio gains and losses excluded and considered to average out, since their return statistics can't be calculated?

[rawraw]

I think his point is that that rank order risk predictably.  Not quite the same thing as inferring that this means there are no better or worse loans -- perhaps he got confused lol

[Fred]

from: Lovinglifestyle on May 12, 2014, 06:15:17 PM

[TravelingPennies]

I have often rejected loans that are approved and verified.  It makes me grit my teeth, but my view of risk doesn't match theirs.  Ditto for the Build A Portfolio choices.  If their statistics were more transparent there might be some evidence to the contrary. It's pretty convenient to just say there isn't any.

[TravelingPennies]

from: Lovinglifestyle on May 12, 2014, 06:15:17 PM

[Rob L]

from: Fred on May 12, 2014, 08:50:02 PM

[Booleans]

from: Fred on May 12, 2014, 08:50:02 PM

[TravelingPennies]

My comment was specifically to counter-argue the "There is no evidence ..." statement in the article.

[BruiserB]

There might not be a best grade of loan.  I can see where LC attempts to have their underwriting criteria and interest rates balanced to provide an even return to investors for any given grade.  However if someone else could do better predictive analysis and even weed out 25% of the loans that will probably default or find a way to predict the ones that won't, then the lower grades offer more potential for higher overall return.

This seems to be what Bryce is doing with P2P picks and what Emmanuel is doing with the "Expected Return" feature of LendingRobot.  Even Peter's relatively simple filters seem to weed out some of the higher risk loans.  It remains to be seen how they perform over the full term of the loans, but I do believe their algorithms can lead to improved performance vs simply investing in random loans.

[lascott]

from: BruiserB on May 14, 2014, 09:41:10 PM

[Fred93]

from: lascott on May 14, 2014, 10:26:43 PM