Last year I wrote a post on my initial experiences with Peer-to-Peer lending through Funding Circle. I initially got an 8.8% gross return, diminishing to 5.1% after fees and losses. Now I have been using the site a year longer, so its is time to update the returns. Over the last year Funding Circle has continued to expand. The website has improved with a better design and tools (still not sure about the secondary market though). There are many more loans available (33 live at the moment) asking for more money (£2 million last week). Anecdotally, this meant for most of the year the rates available rose. Although they have fallen recently as more lenders and even the UK government join the site.
So, drum roll, the year’s results are:
After Fees: 7.6%
After Bad Debt & Recoveries: 2.3%
Ugh, 2.3% is not a very good before tax return. However, still better than any UK bank account. Those rates are calculated with the actual cashflow entering my account. It is interesting to note that the Funding Circle website says my gross rate is 9.2%, but I can’t see how it calculates that over a single year (unless it is including compounding rates). Anyway, it is clear that the poor performance is due to bad debt – the recovery figure over the year is immaterial (0.04%). None of my loans are larger than 1.2% of total capital, and most are around 0.7%. The bad debt is made up of multiple defaults. Using the suggested bad debt rates by loan rating on Funding Circle’s website, I calculated my expected default rate at around 2% per year by volume (my portfolio skews riskier than average). Over the year I saw double that and thus earnt a rate far less than suggested. In Funding Circle’s defence the difference in suggested and observed default rates translates into only a few loans, so it is not statistically significant (yet). Interestingly the bad debt seems to arrive in two waves roughly 6 months apart – one at the end of summer, and another at the end of the tax year.
After these results I won’t be putting any more money into Funding Circle over the coming year, but neither will I be taking any out – I’ll give the system another year.
Postscript: In the last few days and after the above calculations were made, one of the defaulted loans was “recovered”. This resulted in a 61% loss on the original loan amount becoming a 39% loss, with no chance of further recoveries. Better than a poke in the eye I suppose. Had it occurred a week earlier, this event would have made the final return 2.5%. Instead it will provide a boost to this year’s return – see the next update in 12 months time!