The Problem with Comparing Yourself to Other P2P Investors

Most Lending Club investors have seen this little compare link on their Account Summary screen. You can see it right underneath your Net Annualized Return (NAR) when you login. Of course, you have most likely clicked on it and seen how you “compare.”

Comparing ourselves with others is a pretty common albeit unnecessary human behavior. Take a look at the two graphics below from the Lending Club “Compare Your Investment” screen. There is no doubt where you would like to be. These are the investor percentiles for two of my Lending Club accounts.

Lending Club P2P Investor percentiles
Lending Club p2p investor percentiles 2

The one on the left is for my wife’s Lending Club PRIME account. This account is sixteen months old with several defaults and an average loan age of around eleven months. The one on the right is for my new Roth IRA that I just opened in April. It has an average loan age of less than three months with no defaults.

The One Key Missing Element

These investor percentiles compare your account with everyone else’s account at Lending Club. But it misses one key element: the age of your loans. When an account is new at Lending Club it is literally impossible to have a default (assuming you are not investing via the trading platform). This is because it takes 120 days of non-payment before a borrower is considered in default. So, you will always be in a higher percentile than most other investors with the same number of loans because you will have a perfect default record.

I have nothing against the compare feature in theory. I just would like it to be a bit more sophisticated that is all. If they could include the age of the loans in your account in the comparison tool then it could provide useful information. Come to think of  this feature should also include your average interest rate so it can take into account the grade of loans you invest in. Then it would provide more meaningful information. If you invest in A and B grade loans do you really want to compare yourself with people who only invest in E, F and G loans? Probably not. But if you can see that your performance is under par for investors just like you that would be quite useful.

What I would also like to see is some suggestions as to why your NAR is doing worse than other investors. Now, I know with all the SEC regulations that Lending Club is not allowed to be dispensing investment advice. So that is probably not going to happen any time soon. But at least I hope at some point they add some additional information to the compare feature.

Notify of
Newest Most Voted
Inline Feedbacks
View all comments
Dan B
Dan B
Aug. 29, 2011 9:04 pm

That’s funny. I’m going to ignore your rather liberal interpretation of the word “several” in characterizing the number of defaults in your Prime Account.
On second thoughts I rather like it & am going to start using it myself. Tomorrow I’m going to start getting back in shape. I figure I’ll be in tip top condition once I lose “several” pounds.
Aug. 30, 2011 12:57 pm

>If you invest in A and B grade loans do you really want to compare yourself with people who only invest in E, F and G loans?


why not? Assuming you take average loan age (weighted by volume) and then compare actual returns (ROI as calculated with XIRR) with other portfolios of same average loan age, then it should be fine.
An investor with higher ROI will be happier (I assume) – no matter what his default (in number of loans) is. And since the compared ages are equal there would be no mixing apples and oranges.


Dan B
Dan B
Aug. 30, 2011 4:45 pm

I think that a very small percentage of p2p investors understand, or in some cases, are even aware of the significance that loan age plays in the performance of their account. Present company to the contrary non withstanding, even some blog owners who write about p2p don’t understand. That’s why there’s no shortage of nonsense articles/posts about great results & positive experiences……………….after being invested for 3 months or some other silly/insignificant amount of time.

Aug. 31, 2011 7:57 am

Welcome to the real conversation. LC speedboats in its NAR calculation – counting loans that can’t possibly have defaulted yet. That is “not cool”…

Roy S
Roy S
Aug. 31, 2011 2:28 pm

I strongly dislike this “feature” for many of the reasons that have already been listed above. Yes, people always want to know how they are doing compared to everyone else, but it creates the “keeping up with the Jonses” scenario. When that happens, people make decisions based not entirely on the reality of the situation. I can see this leading to people attempting to purchase lower grade notes to seek the higher yields while not fully aware of the risks involved. It is a big issue when they begin to adopt a strategy that is not consistent with their individual risk tolerances and would otherwise keep them from a certain subset of notes or keep their portfolio with a lower exposure to risk. This does not do any favors to the investor. Nor is it good for the LC.

If the companies are putting out any sort of numbers (like Prosper’s 10.6% return), then they should try to be as real (or sobering) as possible. I understand it is all marketing. I also understand that it is really difficult to get accurate numbers, especially with the p2p industry so young and the environment changing frequently, but the 2009 loan origination cohort is only returning 8.35% for Prosper according to 8.35% is quite different from 10.6%. These loans have had longer period of time to be paid off or go into default (both of which affect returns) than the majority of loans originated in 2011. (I understand that there are other factors which make using this 8.35% not an accurate measure either, I’m just using it for arguments sake. I also thought I’d stick to the post-SEC numbers)

Ultimately, it is up to the investor: caveat emptor. Unfortunately, I don’t believe that most investors do their due diligence. For that reason alone, both LC and Prosper should not be trying to oversell themselves or provide charts and graphs like these, even if it is “for entertainment purposes only.” If someone feels the need to compare themselves to everyone else, then they should have to go to an outside site. They make their information public, and there are sites like Lendstats and others you have mentioned on your blog where they can go. With any luck, they will also run into useful sites and/or information along the way. That’s my $0.02.

P.S. It might be useful to continually provide expected returns factoring in loss expectations, age of the loans, early payment expectations, etc. rather than their current NAR. That way they don’t sign in and see 15% return on day and then sign in the following week to see three loans have defaulted and they’re down under 10%. With that lower number staring at them every time they log in, they won’t be surprised when a loan defaults and their NAR drops. Even if they were to provide this number in addition to their NAR, it would be beneficial in my opinion.

Aug. 31, 2011 7:56 pm

This is a good thread. I think Roy S sums it up well. If they going to give us the NAR numbers, then there should be other formulas available with expected losses and such. I believe with every LC order it shows the (average?) default rate, expense rate and the expected return, but they do not show this for the overall portfolio. Curious. LC generally has a good model, but they should be more open about the return numbers. If they are so transparent with their lending methods, why not for the actual portfolio returns? They should do this as a CYA if nothing else.

Roy S
Roy S
Sep. 1, 2011 3:22 pm

I’m sorry I didn’t have the facts on their 10.6% number. It is bandied about on their website, but the actual information is in fine print. As a matter of course, I ignored this number as I view it as just their marketing. I try to ignore marketing by companies except for the purpose general awareness of the company and their products/services. I prefer to get my information from other (hopefully less biased) sources.

The 10-month minimum age is better than including all loans (even those that could not have possibly defaulted yet); however, I still do not think it yields a good result. Prosper does try to state in their fine print, “Net Annualized Returns are not necessarily indicative of the future performance of any Notes.” In my opinion, using any notes where there is some outstanding liability does try to predict future results, since those notes can still default. I haven’t yet really earned a NAR of 10.6% at 10-months if the amount of principle and interest I have received is less than face value of the note. At this point, my return is still negative, and it will only turn positive after I receive my full investment back.

I should state that I do not know whether there is a better way of trying to capture the ultimate return on these notes than to wait until all have reached their maturity date, which is still a year out for the first loans originated post-SEC. Even then, I don’t think you can compare those loans to the 2010 origination cohorts, or the 2010 cohorts the 2011 cohorts. I do think that LC and Prosper are ultimately in a lose-lose situation. Either they advertise an overstatement of returns and irk people who end up not reaching that return, or they advertise an understatement and turn away people who might otherwise invest if they see an average return two or three percent higher. Either way, I get to sound good for criticizing them! 😉

Dan B
Dan B
Sep. 1, 2011 5:20 pm

An agreed upon standard is going to be the holy grail, I agree.

Roy S
Roy S
Sep. 1, 2011 6:14 pm

I also like how Lendstats uses loss factors in order to estimate ROI; however, I am not certain how Lendstats comes to those loss factors. Further, I don’t believe it factors in the grade of the loans. I would like to know how other sites are calculating loss rates and recovery rates when determining ROI. I know you featured other sites like Lendstats in August (I didn’t look too much into them because they were only set up for LC at the moment). Perhaps they would be willing to share their views on how ROI should be calculated. We may not be able to get LC and Prosper to alter the way they calculate NAR, but if someone has a more accurate way to calculate their NAR I would be interested in it for my own personal calculations.

Sep. 1, 2011 8:20 pm

Well, when the market becomes bigger, a better standard will become possible. Peter bringing up mutual funds and bonds is a good point… but that NAV is based upon assets, not ROI/future returns… for accounting of bonds, I do believe it’s based on market value (which is why there was a big mess with the mortgage crisis because companies had to value off the market price of mortgage securities which was close to nothing even for good ones)… meaning a good way to price these notes in the future could be to base notes’ prices off of folio data and what they’re actually going for. ‘Course the trouble is they don’t allow currently late loans to be sold on the marketplace, but there might be some truths hidden even in what is currently being sold.

Prosper lender “shawnw2”