The WealthBoy Strict ROI for Prosper Lenders

As I was writing my Rule of 72 on Prosper article for the official Prosper Blog, I began to think about developing my own ROI calculation based on what I had written. I had attempted creating an ROI calculation once before that was based on actual payments received, but I became frustrated with the lack of the detail payment data in the LoanPerformance table of the private export. I took another crack at developing an ROI calculation based on actual payments, and think I’ve come up with something that’s reasonable. If you’re interested in the actual implementation, you may want to check out the technical details and link to the SQL code here:

http://wealthboy.com/wbsroi-technical-details/

Once I’ve constructed the tables necessary to calculate the payments that a lender has received, I have all of the information necessary to calculate the WealthBoy Strict ROI. The WBSROI performs two return calculations: TotalROI and AnnualizedROI. The TotalROI calculation is calculated by dividing the total profit (interest less servicing fees) by the total amount invested (which excludes reinvested loans). The calculation does not take into account the declining balances, hence the “strict” designation. If you have been lending successfully for a long time, it is certainly possible to have a TotalROI more than 100%. Here is the formula in a nutshell:

TotalROI = (Total Interest Received – Fees – Losses on Defaults) / (Total Loan Originations – Reinvested Loans)

The AnnualizedROI is calculated by dividing the TotalROI by the weighted average loan age and multiplying by 12. The weight for each bid is the amount lent as a percentage of the total originations. This may not be the best way to perform the AnnualizedROI calculation, but it was the best I could come up with. I believe the TotalROI is relatively indisputable, barring the errors in the payment calculations. The AnnualizedROI could probably use some enhancements.

I like the idea of having a strict ROI calculation that doesn’t account for the declining balances. Many lenders on Prosper may not even be aware of what a declining balance is. Others may know about declining balances, but they just want to know what kind of return they’ve received on the total amount they’ve invested. That is what the WealthBoy Strict ROI attempts to do, and I believe it does it reasonably well. If you are reinvesting loans, your strict ROI should be reasonably close to your average interest rate less fees and your default rate.

So what about late loans? Why aren’t they part of the calculation? Well, one of the nice things about of my calculation is that it really doesn’t take much more effort to account for the probability of late loans eventually defaulting. With the information provided in the calculation, you know the total investment and you know the total profit. All you need to do to account for late loans is to incorporate the loss estimation into the profit and presto! You have your new ROI including the probability of late loans defaulting.

I decided to exclude any default projections from the initial announcement of my ROI calculation. Although it probably wouldn’t take much more effort to incorporate it, I think there is something to be said for an ROI calculation that doesn’t make any kind of suppositions. The WealthBoy Strict ROI calculates how much went in and how much came out. It makes no assumptions about the future value of loans. I have left it to others to make whatever assumptions they wish to make about estimating defaults.

I do realize that not everyone has the expertise and/or resources to put together a Microsoft SQL Server database for analyzing Prosper data. Unfortunately, I don’t have a web application connected to my database so that people can see their WBSROI. If you would like me to provide you with your WBSROI, just post your screen name into a comment here. I’ll post the data in a responding comment. If I become overwhelmed with responses to the post, I may not be able to respond to requests any longer. If it does become a popular metric, perhaps someone with a popular stats site may be willing to add my calculations to their site. Here is what the WBSROI on my account looks like:

Screen Name:WealthBoy
Total Bid Count: 45
Total Reinvested Bids: 3
Total Originations (total amount loaned): $2,250
Total Investment (total amount loaned excluding reinvested bids): $2,100
Total Income (total principal and interest less fees): $209.22
Total Profit (total interest less fees and defaults): $60.53
Total ROI: 2.88%
Average Loan Age: 1.76 months
Annualized ROI: 19.62%

Bookmark and Share
Blog Traffic Exchange
Related Websites

  • Another Error in a National Newspaper on P2P Lending
  • lending-club-shaking-hands-over-a-loan-or-promissory-noteMy Plan to Beat the Lending Club, Peer-to-Peer Investing Average
  • Factors to Consider When Consolidating Debt
  • Related Posts:

  • The WealthBoy Strict ROI Technical Details
  • Prosper Investing Tips and Testimonial
  • My First Post on the Official Prosper Blog
    • That's right, it was a defaulted loan that had been bought buy a debt buyer, not one that was in the lawsuit.

    • The loan we were looking at was defaulted and had nothing to do with my repurchased loans... I have 6 loans in the lawsuit... so similarly my numbers are also lower.

    • Stephen:

      Interesting... I was looking at one such loan for RateLadder that he told me had been sold to a debt buyer. On the site it appeared it had been marked as a repurchase, I could have sworn that in the data export it showed up as a default. Therefore it behaved as it should have within the ROI calculation. However, I took a look at the detail for your account, and they have indeed been marked as repurchased loans. In fact, none of your loans show up as defaults and according to the data you have three repurchased loans.

      I suppose it's something else that needs to be added to the list of issues with my ROI calculation. Of course, this one isn't my fault. :) There's nothing I can do about Prosper flagging them as repurchased loans when in reality they are defaults. Perhaps they might be able to do something in the future to resolve this issue. Perhaps a new status to add (Repurchased by Debt Buyer)? Even if they do want to mark them as repurchased loans, they could at least display the proper NetDefaults value in the LoanPerformance table. Thanks very much for bringing it to my attention.

    • Chrisfs:

      There are some subtle differences in the syntax between MySQL and SQL Server, but it should be doable. You'll need to use regular tables instead of temp tables and table variables as I've done. It may take a bit of work to get the reinvestment cursor coded for MySQL as well.

    • Chrisfs

      Looks good, will it port easily to MySql ? that's what I have access to (no pun intended...)

    blog comments powered by Disqus