Am I Really Getting a 20% Rate of Return with Prosper

No joke, 20% is the new 10%. Back when I first started p2p lending 4 years ago the goal in the back of my mind was to get 10%. Long story short, I didn’t even come close. Now I am aiming for 20% returns and so far at Prosper, according to LendStats.com, I am getting it. Granted, I’m only looking at the loans I made since Prosper relaunched in 2009, but more significantly I’m looking at the loans I made since I started doing my statistical analyses of the prosper data which I also started in 2009. Those analyses were the beginnings of LendStats.com and they have been invaluable to me as a p2p lender. As of today my returns at prosper on loans I made in the post-lendstats era are 20.9%. That is incredible and even I am astounded.

I’m sure many of you are justifiably skeptical and must think that the LendStats’ ROI calculations are incorrect or at the very least generous or maybe you think that I’ve just been lucky as a lender. I will attempt to prove all of those cases wrong by taking a deeper look into my numbers. Using LendStats.com I can track my own (or anyone else’s) returns and see a basic breakdown & analysis of my portfolio, so let’s see what LendStats.com has to say. By using this query I can see what my loan performance has been on loans made starting in 2009. This is the result:

Loan Performance Summary for lender lendstats_com: .

You can see here LendStats.com calculates a rate of return of 20.94% for the user lendstats_com, which is me. So let’s do a back of the envelope calculation to see if the LendStats calculation is legit. We can see the average interest rate on my notes is 25.93%, the average loan age is 6.2 months, there are 5 defaults for $937, there are 3 notes for $303 that are 31-120 days late and 13 notes for $2009 which are 1-14 days late. For the sake of simplicity let’s assume that all loans more that 30 days late will default. That gives us $1240 in defaults. Lendstats.com estimates that 30% of 1-14 day late loans will default which would give us an additional $603 in estimated defaults. So we add that to the $1240 and we get $1843 in estimated defaults on $167,110 lent out. That’s a 1.1% default rate, not bad. Unfortunately though, this 1.1% is misleading. We need to first subtract the $40503 in newly issued loans from the total lent because the newly issued loans have not yet been billed and they have had absolutely no effect on returns. That leaves us with $126,607 in billed notes. We also need to consider the effect of early payback and the decreasing principle of the active loans. So to keep things simple for this back of the envelope calculation, let’s just look at the current balance of all active billed loans which is $95,213. This gives us an estimated loss rate of about 2% ($1843/$95,213). Now, considering the estimated loss of 2% on my loans which are on average 6 month old, we can multiply the 2% loss by 2 to get a yearly loss rate of 4%. Since my average interest rate is 25.93% and my estimated annual loss rate is 4% that leaves me with a net 21.93% annual gain. Then we also subtract the 1% in fees that Prosper takes and we are left with 20.93%, Wow, that’s almost exactly what LendStats calculates, not bad for a back of the envelope calculation!

Now those who think Lendstats is being generous with its calculation, let’s remember that in this back of the envelope calculation we made 2 big assumptions that will always result in an underestimation of calculated returns. Those assumptions were:

#1: all loans 31+ days late will default. Although most do default, not all of them do.

#2: the average balance we used was not the average, but the current balance of active loans which is much lower than the average balance.

These two assumptions should lead us to calculate a higher rate of loss than the actual rate of loss. Since our back of the envelope return calculation matched the LendStats calculation, this shows that LendStats.com is probably not overestimating my rate of return and in fact could be underestimating it.

Now for those of you who think my returns are the result of luck, look at the number of loans. There are 660 billed loans. If there were only 50 loans, then yes, luck would be a strong possibility, but not with 660. To get that kind of return on that many loans an effective lending strategy is needed, luck would not suffice.

We can also do one last check to see how my returns compare to other similar lenders. If my returns really are so great, then I should be one of the top lenders. As of today I am the #4 lender by ROI (just 0.4% behind #1) out of 595 lenders that have at least 300 billed loans on the post 2009 Prosper platform. We can see that query here. From this list we can conclude that I am one of the top performing lenders on the Prosper platform.

So there you have it. I am getting a legit 20%+ on my loans made at Prosper post-relaunch. However, even after this validation of the LendStats.com calculation, I actually believe I will not achieve 20%. Through experience I’ve learned that returns have a tendency to slowly head downward till about the 18 month mark and I have 12 months to go till then. Nevertheless, I am still aiming for 20% and I’m sure if I don’t make 20% I will be close and in any case I’m sure I will be happy with the returns that I get.

Now for the question that everyone wants to know: How in the world am I getting 20% returns? Well, just take a good look at LendStats.com and you might just find out. Hint, start out here play around a bit and be sure to use the ‘sample filters’, that’s what I do.

Newbies beware!!!!! I recommend that newbies start off with much more modest goals, like 5% and sticking to a portfolio of 90% AA,A and B loans. Many newbies were burned in the beginning going after high returns and although the default risks are much lower now than they were 3 years ago the risks are still there. So start modest and remember patience is your 2nd biggest ally, the #1 being LendStats.com of course.

***Please note that these calculations do not reflect any trades made on the trading platform. That data is not yet available from Prosper. The analysis presented here and at LendStats.com only looks at loans funded on the Prosper platform as if no buying or selling activity ever occurred on the secondary market.