Peer to Peer Lending. A 7m7y tool?

In my last post, I suggested that banks are profitable businesses because they have such a large mark-up. If they’re so great, my son asked, why don’t I simply plonk my cash into a safety deposit box and dole it out to willing borrowers like some kid with a lemonade stand?!

Why not, indeed?

The simple and obvious answer is risk, which the bank handles, I said to my son, with a combination of volume (to spread risk), people (to manage risk), and systems (to assess and ‘price’ risk).

However, Rick Francis offers perhaps a better-lemonade-stand-solution (?) … Peer-to-Peer Lending:

There is a fairly easy way to become the bank- peer to peer lending. It doesn’t remove the risk of default but does allow for diversification and there is a framework to asses the risk. They break loans into many small pieces that different individuals fund, so you don’t risk too much on any one loan.

Yep, P2P Lending certainly helps to address one of the banks’ three mechanisms for handling risk: you can spread your loans (the bank lend $400k many/many times over … you lend $40 many/many times over).

But, what about the experienced PEOPLE? It can take some time/trouble to sift through all of those loan apps listed on the leading P2P sites, as Jake points out:

P2P lending requires you to pick through hundreds of loan apps, and filter it to the set that you believe has the best risk / return ratio.

Then you have to diversify – invest in many loans so that a single default will not wipe you out. I think that you should invest no more than 1% of your portfolio into a given loan – so lets say you need to invest in at least 100 loans. Unfortunately, that requires you to pick through probably 1,000 applications hand-by-hand (you already discard the vast majority based on search criteria).

That’s frankly just too much work to be worth it, no?

it’s worth it for the bank, but probably not worth it for you and me (even though you can filter/sort the loan applications by various criteria) ‘just’ to get that 10% return that Rick has experienced …

And, we still haven’t addressed the risk management SYSTEMS that the bank applies, what does P2P offer there? Many sites, as Rick pointed out, offer some sort of FICO-based ranking, but banks rely on a lot more than that (for example, where’s that little thing called ‘collateral’?!) …

The only compensation for these last two (PEOPLE and SYSTEMS), that I can see, is that P2P borrowers may not want to default for a combination of:

– Getting locked out of the P2P sites … perhaps a similar mechanism to eBay’s Rating system is available?

– Perhaps it’s enough that P2P borrowers appreciate the opportunity that they have been given and don’t wish to abuse it by defaulting?

It is perhaps these two reasons that help to explain why micro-lending in 3rd world countries has such a low default rate?

But, it’s the simple logistics that Jake pointed out that put the kibosh on P2P for me …

Have you had any experience with P2P and would you use it again?

Be Sociable, Share!

29 thoughts on “Peer to Peer Lending. A 7m7y tool?

  1. You raised some good issues but I think there is something that should be clarified.

    It’s true that going through all loan applications is hard if you look at each of them carefully. However, most p2p-platforms offer some kind of rating. This eases process a lot, since now you can allocate 10% to grade A loans, 20% to grade B and so on. Of course it questionable if those rating are accurate, but so far it seems so. It is also notable that most loan applications are rejected. This info is based on Lending Club’s stats.

    P2P-lending is uncollateralized, so you should compare it with credit cards or some other same types of loans. Mortgages and other loans with collateral are bit different. As the house prices collapsed, collaterals became a burden to banks- Risks with collaterals are less probable, but more dramatic, so called iceberg risks.

    My two cents: p2p-lending is not risk-free, but I see a great potential. Especially when the industry matures. Investing in p2p-loans should be something like 10-20% of your total portfolio.

  2. As a long-time reader [and now first-time poster], I’ve greatly enjoyed the insight that has been offered by Adrian and many regulars. I think that this discussion could be further clarified, perhaps even bifurcated, if you label risk as ‘investment risk’ or ‘speculative risk’. The distinction between the two of these, at least in my mind, is based on the extent to which you can control the variables that comprise the risk, if at all. More to the point, investment risk would assume a high – or at least greater – degree of control as compared to speculative risk. In this context, the individual is left to ask himself whether he is approaching a risk as an investor or a speculator.

  3. Adrian,

    You make a good point about collateral– they are unsecured loans and that carries extra risk. However, banks make unsecured loan too- they do so at higher interest rates to compensate for the additional risk.

    As for it being too much work for a larger sum, I think it may actually be easier with a large sum for the following reasons:

    #1 I don’t think reviewing many loans and picking the “best” set decreases your chances of default.

    #2 The statistical default rates are low enough that you should still be able to make a reasonable profit.

    I don’t have any specific research to prove #1 but my intuition is that it is similar to picking individual stocks. People just don’t predict complex problems well but they think they can! There are a lot of studies on how managed mutual funds fail to beat their indices. I wouldn’t be surprised if picking loans worsened your default rate. If the statistical default rates are acceptable to you then just let their interface pick the loans automatically. With a large number of loans you should expect to get the default rates approaching the statistics. Accept that and the process is trivial.

    -Rick Francis

  4. Rick,

    I don’t have time to quote the math, but the idea is that a portfolio of uncorrelated risks reduces the overall risk.

    E.g. Assume you have two prosper loans each with a risk of 10% of defaulting. If the loans are correlated, i.e. one loan will behave very similar to the other, the risk of the total portfolio is still 10% – no change. But if the loans are uncorrelated, i.e. they have no reason to behave the same way, the total risk of your portfolio is SMALLER than 10%, e.g. 8%. Key is avoiding investments that are correlated. The underlying math and beauty of the concept of diversification is explained in great books like The Intelligent Asset Allocator.

    The further you diversify your portfolio of P2P loans and avoid correlated loans (e.g. same person, same industry, same problem etc), the lower your overall risk.

    Unlike the stock market where buying a couple of index funds will do the job, in P2P diversifying is hard work.

    Now there is a thought – an index fund for good P2P loans. 🙂

  5. is there a way to collect the defaulted p2p loans and auction them off? This happens with things like unpaid credit card debt– collection agencies pocket whatever they can get out of the debt that the card companies have already written off.

  6. The problem with P2P platforms like Prosper is that only people resident in certain states in the US can use them. No good for me.

    Banks make a lot of non-collateralised loans (over drafts, credits cards, personal loans etc) but they usually will only lend unsecured (or at all) after assessing the borrower’s ability to repay (students seem to be the exception). For non-business/non-investment non-collateralised loans this is usually done by obtaining evidence of current income (among other things). This is hard to do with P2P lending done through a an internet platform (although you could do it if you deal directly with the borrower….but then you would probably need a licence for that in most places).

    That said, looking at the sub-prime mess where the loans were supported by security, it begs the question of whether collateral is really that important? It probably depends on the collateral and the loan to value ratio.

    If you have proper collateral, the rate of return is pretty low but your risk is relatively low. If you make unsecured loans, your return is higher but so is your risk – and the amount of homework you have to do before advancing the loan.

    Also, not to put too fine a point on it, if you want to make a career out of lending money, you have to beat the banks and other professional lenders – either by being cheaper than them or by being willing to make loans that they will not or by being more responsive to applications.

  7. Yes, I(we) have had some experience in P2P lending. This is in The Philippines. Not Online 🙂
    We’re not a bank so to speak, but we lend to individuals on a regular basis. There is (as you mentioned) the reduced default rate there, as the courts really don’t believe in bankruptcy, and will take your household goods or whatever it takes to get your money back for you.

  8. We had one loan where it was for $500.00 . Had he defaulted, I don’t think I would have been real upset, as we had collateral (in the form of property ) assessed at Approx 10 times the loan amount.But collateral is not always a requirement.

  9. Great comments … looks like I’m going to have to do a f/u post instead of my usual hunt-and-peck of the various comments 🙂

    @ Kohti – Does the P2P site need to be country-specific or is language/geography the barrier to using the existing sites. If you do need a local site and regulations allow it, then you may need to find a strategic investor 😉

    @ Jake – I also love your idea about an ‘index fund’ of P2P loans …. you could have Fund A, Fund B, Fund C, etc. each focusing on one strata of ratings on each site, naturally, with a return commensurate with that risk/rating.

  10. @AJC & @Jake,

    We can then take all those loans and mix and match them so if we stick some As with Cs and Ds the rating agencies can mark them as As…

    hmmm sounds an often like the mortgage backed securities lol

  11. I’ve use Lending Club and actually like it. The process to review and manage the currently 30 loans I have is not that bad and IMHO similar to the time researching investing in stocks.

  12. @ Investor Junkie – Thanks for that @ 30 loans, it seems that you might be our P2P Expert In Residence 🙂

    So, let me ask you a few questions:

    a) How long do you expect to be invested in P2P?

    b) What % of your portfolio is dedicated to P2P?

    c) How do you feel the long term return on your loans will compare to long term returns on stock?

  13. Hey Adrain,
    I’m a fan of your site, been a lurker for awhile.. Anyways.

    I have posted my experience on my blog:

    I will be updating my Lending Club results sometime this quarter.

    To answer your specific questions:
    a. Lending Club loans are 3 year terms, though you can sell the notes earlier. I did for one where the FICO score lowered dramatically.
    b. Right now VERY small under %1 range, though I plan on increasing it this year.
    c. IMHO too early to tell, as I started mid last year. Right now lower than my stock returns, but look at the results of last year. My return is still around 10.50%. No defaults yet to speak of but based upon stats it appears they peak midway through your the loan.

    I really don’t consider them in the same class as stocks, they are more similar to bonds since you are holding debt. Overall I don’t think Lending Club is for the average Joe, but I assume were not talking about them? 😉

  14. @ Investor Junkie – Thanks for that … I am wondering, since we agree that these are not for the Av. Joe and do have an element of risk attached, why anybody would put their money into P2P for an extended period (say, 2 x 3 year ‘cycles’) and not into stocks that have a long-term average 12% return, or RE?

  15. Stocks have a long term average of 12%? Where?? Since 1926 stocks have returned 9.8%, which includes dividends. RE has typically either matched or slightly higher than inflation rate. Since 1913 inflation has been slightly higher than 3%. Of course leverage (mortgage) means your return are greater. If that’s the case an average of 9.6% return with Lending Club offers similar returns, with possible less risk than a stock. Keep in mind to own 30 stocks even with low discount brokers each would cost at last $5.00 much higher than the lending club fees.

    WSJ’s Jason Zweig just had a great article on the subject of stocks, returnsm, and people’s expectations:

    I think the bigger question you are trying to ask does Lending Club properly price default risk for the loans? For the most part, I think so. Prosper a few years ago did not.

    Also the issue is correlation for asset allocation. Does P2P lending correlate to other investments? That I cannot answer, but assume it’s closer to bonds in correlation.

  16. 9% ? 12% ? Who is right? who knows. You can read sightings from 15 different sources and come up with a different figure in each case. So What is the true average rate of return on stocks? What about Corporate Bonds? What is their average rate of return? I’ve seen a guy who claims somewhere around a 20 to 25 % return for Corporate Bonds.With some of his returns exceeding 50%

  17. @ Investor Junkie – The figures that I work on are roughly, 12% before inflation / 9% after (with no 30 year period returning less than 8% ‘before inflation) … but, it matters not to me: I don’t invest in the ‘stock market’ (e.g. index funds) nor do I lend money to others for less than 25% annualized.

  18. It is by buying these bonds below their true value or cost(the same way you buy stocks below intrinsic value ). Later sell for full value,plus interest collected during the time held.

  19. @Steve: Ok fine you are finding bonds at a discount. The way Adrian mentions it, it sounds like it’s via some other type of investment that offers 25% return on the face. To get 20+% on corporate bonds was pretty easy last year and have a few myself. Finding ones at that much of a discount today to net 25% is a pretty rare find.

  20. I should add, find 25% of bonds that aren’t of junk status. If they were would mean of course high default rates and goes back to then comparing to P2P loans and their default rates.

  21. All Good arguments. I am not saying these are easy to find investments. Just that you can find them, and my referral is not junk bonds. He does not get into junk bonds. He says the risk is less than 1 % on defaults. Certainly this would not include junk bonds.

  22. @ Investor Junkie – I cheat … I own a finance company that has a nominal rate of 25% (approx.) out of which comes:

    1. Staff, overheads, and risk – approx. 1/3

    [BTW: the risk component is virtually zero, based on 15 years history]

    2. Finance charges (like a bank, we borrow the money that we lend) – approx. 1/3

    3. Profit – approx. 1/3

    … about as active an investment, though, as P2P except for our turnover we need 3 staff.

  23. Pingback: The problem with P2P lending …- 7million7years

Leave a Reply