Decide on a P2P Lending Strategy

Last time I described the performance of P2P loans I’ve invested in on the Harmoney platform. Today, I want to describe my thoughts behind loan selection.

Diversify

Definitely, take advantage of the fractionalisation of loans and go for maximum diversification. Lending club history shows that you need to hold at least 100 different notes and no loan should make up more than 1% of your portfolio. Otherwise, returns will be more volatile and are more likely to be negative. It’s fine to start small, but try to get up to 100 in the first year.

Diversification can reduce volatility of returns Owning 100+ notes reduces risk in your returns
From: https://www.lendingclub.com/info/statistics-performance.action

Choosing Risk Grades

First, I wanted to see how the interest rates and default rates at Harmoney change across the risk grades. The interest rate charged to customers can be broken down to an expected default rate, a service fee and the Net Annualised Return or NAR which is what’s left for the investor. This is how loans are presented in the marketplace. This graph shows the three components adding up to the published interest rate charged to borrowers.

Harmoney Interest Rates and Returns by Risk Grade

The blue line for NAR shows a positive relationship between risk and return. The kink in NAR at F5 does draw attention because it has a lower predicted NAR than F4 but is higher risk. Default rates are estimates only, so F5 loans might actually perform better than expected and outperform F4, but I still wonder if investors are being adequately compensated for the risk in F5 loans. Personally, I am not investing in any F3-F5 loans so it’s not a worry for me. We won’t know if these default estimates are accurate until Harmoney publishes actual default rates.

Although we have no idea how the credit processes and grading systems between Lending Club (USA) and Harmoney compare, it’s worth looking at the historical results. First the entire history of Lending Club.

Lending Club: Historical Returns by Grade

From: https://www.lendingclub.com/info/demand-and-credit-profile.action

As you can see, the NARs (coloured bars) don’t differ very much between the C and F+G grades. The NAR doesn’t really go up much as risk increases. Should you be indifferent regarding credit grade, or target a specific area? I personally try to hold fewer of the higher risk loans because I feel that in an economic downturn, these are more likely to default.

It is worth exploring the Loan Statistics page at Lending Club to get a feel for how loans there performed. For example, grade F+G loans issued in 2007 returned -9%. Was this due to poor economic conditions, a poor credit model (because Lending Club was brand new), or some other reason?

My Harmoney Porfolio

Thirty credit grades are just too many for me to try to target specific grades to choose to invest in. I’d be waiting ages for a particular grade to appear on the marketplace.

Outstanding Principal by Grade

This is the actual outstanding principal in my account for each of the thirty credit grades. It does look very messy. I don’t use this graph. Instead, I’ve grouped the thirty grades into ten deciles.

Outstanding Principal by Decile

Ahhh! Much nicer looking. Within a decile category I’m indifferent between grades.

I’m trying to target those loans in the middle that should have better returns than the low risk A grades. I also mentioned earlier that I’m avoiding the riskiest decile, F3-F5. I want to see how those loans perform in a downturn before investing. I’m currently targeting an average of a C2 or C3 grade.

I guess this is how I’m cautiously testing P2P lending. I want to see some history behind Harmoney’s credit grading abilities before funding the riskiest loans.

Choosing the loan term

Lots of new investors wonder how they get their money back. I think it’s best to remember that investing in a loan directly pays differently from bonds and term deposits. The monthly payments are both interest and principal (like most home mortgage payments) so your money is returned gradually and it may take 36 or 60 months to get it all back. Borrowers have no prepayment penalties so they may pay you back faster than expected. I don’t see this as a negative because the funds can be easily reinvested.

Investors’ money is locked in until it is repaid by the borrower. There is currently no market for selling notes to cash in early. For this reason, I don’t think it is a good idea to try to invest large sums or redeem large sums at once (like a traditional bond or term deposit). If that’s what you need, this may not be for you. It is much better to drip feed the money in over the long term, reinvest and then take the cash flows later on. You are able to withdraw the cash in your account at any time.

Initially I had the thought that I should lean towards 36 month loans with the logic that the borrower had actively chosen a shorter term, has the ability to make higher monthly payments and thus may have a firmer financial footing. However, I have no evidence of a difference. 60 month loans do pay more interest over the life of the loan and require less effort to reinvest, but they don’t pay at a higher rate per annum. Assuming the cash flows are reinvested, the 1.25% service fee takes a bigger bite of 36 month loans than 60 because the principal is being reinvested more often or at a faster pace. The 1.25% stretched over 5 years is less per year than 1.25% over 3 years. The difference is pretty small however.

Based on these assumptions, I’m not favouring one term or the other and I’m investing in both. For me, I am investing money I don’t need for more than five years.

So, how are you using the Harmoney platform to achieve your goals?

5 thoughts on “Decide on a P2P Lending Strategy”

  1. What a great post! Thank you for writing this one!

    I have a different strategy.. I actually choose the F graded loans because even with increased default risk I expect the overall return of my portfolio to be better by doing this. I believe Mr Money Mustache also chooses higher risk loans.

    Particularly with rewrites being introduced recently, you want a higher interest rate to allow for the fact that you might only have that loan for a few months before it is paid off in full. There is also the time factor before placing any paid off funds into new notes.

    I agree totally with diversification. I choose one note in each loan for this reason.

    Very interesting point about the difference between F4 and F5. To be honest I hadn’t looked at that very closely. I may now choose less F5 loans.

    My preferred loan purpose is debt consolidation because I figure that means the person already has the debt and is already making payments towards it. I don’t invest in weddings or holidays because I don’t see much incentive to pay off the debt after the event.

    Cheers,
    Meg

  2. Hi Meg, Thanks for sharing.

    Yes in some years F should have the highest return and in some it won’t. I think it depends on general economic conditions at the time.

    Your strategy probably puts more faith in Harmoney’s credit scoring ability and forecast of default rates. Let’s hope they’re doing a good job!

    Interesting point on screening for loan purpose. I know that some Lending Club and Prosper investors have sophisticated screens and they are able to earn yields much higher than the average. Though those platforms also have the problem of desirable loans getting snapped up very quickly via computer program.

    It would be interesting if Harmoney exposed users’ portfolios (anonymised of course) so we could compare performance on the platform.

  3. Hi Hamish,
    I have reviewed my position on investing in Harmoney. I’m starting to get a bad feeling about it. I have written a post on why I have started withdrawing funds.
    I think because the peer-to-peer lending concept is so new to New Zealand it is great for Kiwis to discuss their own thoughts and perspectives.
    Cheers,
    Meg

  4. “Borrowers have no prepayment penalties so they may pay you back faster than expected. I don’t see this as a negative because the funds can be easily reinvested.”
    Would you be making much less return when loans are paid off quickly as you still have to pay 1.25% on the capital but don’t get any interest earnings? That might be a good reason not to invest in low risk loans where the person needs bridging funds. For example, a builder who needs money to buy tools for a job.
    Also, I wonder if people who have rewritten their loan are more or less likely to rewrite their loan in the future? I would think they would be less likely to rewrite it but possibly more likely to default. In that case it would make sense to invest in mid risk loans like c’s and only invest in loans that have been rewritten.
    Thanks for the great blog post.
    Andrew

    1. Thanks for the comment Andrew. Yes, the effect of service fees on rewritten loans is certainly a concern. I made some comments on rewrites in my Results Issue 1 commentary for line 2 and in my Results Issue 2 commentary for line 2. I agree that if you can identify a borrower who only wants short term bridging finance, you may want to avoid. I did do some modelling of the effect of the service fee on return using different periods (i.e. 3 months, 6 months, etc), but I saw that http://www.monetary-meg.com had already published something similar, so I never wrote it up. Now I can’t seem to find her article. My results showed that a rewrite after 3 months of normal payments reduces the expected return by about 4% (e.g. 12% reduces to 8%).

      Your questions on rewrite patterns and likelihood of default are interesting and I’m hoping that Harmoney one day makes raw data available so that we can test those sorts of questions.

Leave a Reply to Andrew Hughes-Games Cancel reply