Intuitively the fact that two people are legally liable for the loan seems to be a plus, but that may not be born out in the default rates...
Bingo. There have been many things that seemed intuitive to me about credit data, but weren't validated by the data, so I'm data-driven. Wonderful that so much data is available, but on this point there seems to be no data.
Anyone know where we can get that kind of data? I looked on NSR, it ain't there 
LC hasn't issued loans like this before, so we can't get the data from LC. We have to scratch our heads to think of whether we can get it somewhere else. And then... its not just the question of whether adding a 2nd person helps the default rate, but also how one interprets the credit numbers. It appears that LC is giving us the full data on one person, and very limited data on the 2nd person. That's very strange. How should I deal with that? If the first person has a good FICO score (good stuff) but the 2nd person has a hidden FICO score, then maybe the 2nd person's FICO is horrible, in which case she isn't much of a help after the 1st person loses his job, defaults, moves to Florida, etc. So how do I even try to make intuitive evaluations if I don't have the data for both people?
Perhaps there is some data set out there from the government or some banking source or something that tells us the relative performance of 2 person loans vs 1 person loans. If that exists, perhaps we can extrapolate it and come up with a theory.
Even if we do, that's still not data-driven the way our investing in the LC INDIVIDUAL loans is data-driven. We'd need about 3 years of data before we could do that.
All very puzzling.
I do hope to force an explanation out of them so that we can at least understand what the data fields mean. That will at least allow us to guess the rest.