Peer to peer lending has been around for years at this point, but over the past couple months I have seen a huge surge in interview candidates mentioning it in some of their answers for where they think financial disruption will occur. Unsurprisingly this coincides with what has been a very rapid increase in loans generated by peer to peer lending platforms such as Funding Circle or Lending Club (the 2 giants that I am aware of), so I decided to take a look to see whether it was a potential asset class for an individual to consider.
The Basics of Peer to Peer Lending
So what is peer to peer lending? Effectively it’s an online platform where individuals can borrow and lend money to each other. The idea behind it is that where before if you wanted to borrow money you had to go to either a bank or rely on a credit card, now you can get a loan from another individual via the online platform. So the borrower gets a quick and easy loan, and the lender gets a higher interest rate on their money than they could get by keeping their money in a bank. The disruption of peer to peer lending stems from the basic idea of how retail banks make money. At the most basic level, retail banks take deposits from people and pay them some sort of return, they then take that money and lend it out to people who need money. The spread between what banks pay for deposits, and what they charge on the loans they make, is the profit margin for the bank. Naturally, the profit margin is slightly less than the spread because some of the loans will be defaulted on. So how does peer to peer lending step in? Essentially while big retail banks need to charge a very high spread between what they pay on the deposits and what they charge for loans due to their high costs (lots of staff, physical branches etc), a peer to peer lending’s cost are effectively just what is needed to run the website (IT staff and marketing), significantly lower. Therefore the peer to peer lending reduces the spread required between borrowers and lenders, in theory benefitting both.
Lending via a P2P Platform is not Equivalent to a Deposit at a Bank
I do want to stress that despite it being part of the reason why P2P lending is a “disruption”, lending on a P2P platform is not equivalent to having a deposit at a bank. Deposits are insured up to a certain amount (generally up to 100k), and in theory there is more protection both from a political perspective (hurting deposits generally makes the voters angry) and from a regulatory standpoint (there are a lot of buffers before depositors lose out). A bank’s asset base is in theory also more diversified (in an uncorrelated sense) than what an individual could achieve on a P2P platform. The counter argument is that you can always loan a tiny amount spread across a lot of borrowers, but I would argue that the borrower demographic on P2P platforms is not that diverse and so a lot of the diversification achieved by making more, smaller loans is a mirage, as I will expand on further below.
Either way, lending on a P2P platform is not a substitute for a deposit. The risk is substantially higher, which is why the headline return is as well. 99.9% of the time there is no free lunch, and when there is, it is usually picked up by professionals. The best we can do as individuals is making sure we are not taking risks we do not intend to take and be lulled into a false sense of security. Case in point, when UK depositors piled into high paying deposits in Iceland, and then lost out when the Icelandic government went effectively bankrupt and couldn’t honor the deposit protections. The thing people forget is that the presence of insurance always rests on someone’s creditworthiness, and often times people make the mistake that anything that is provided by the government is automatically risk-free. In hindsight, the depositors in the Icelandic banks were getting a higher return because they were taking on the credit risk of the Icelandic government, which turned out to be a bad bet as the country’s financial system collapsed.
My Conclusion: Great for Borrowers, Nots so Great for Lenders
When I first started looking into peer to peer lending, I was not biased either way. In fact, I was probably quite optimistic of its potential as a way for individuals to get access to more risk taking opportunities. However, after analyzing it in more detail, there are several issues that I believe make it very attractive for the borrower, but a lot less so for the lender.
Issue #1: There is not enough data to make default assumptions
If we look at the Lending Club total loan issuance chart below, we can see that the growth has been exponential over the past couple of years. However, the majority of loans have been made in the past year or two. But the average loan duration on these websites is to the tune of 4-5 years. So only in the next year or two are we going to start seeing principal repayments from the loans made back in 2012:
Funding Circle provides a historical breakdown of defaults:
You can see how the 2014 bad debt % is very low at 1.7%, but the 2011 bad debt % is 5.3% as expected given those loans are starting to approach maturity. This is what allows for P2P platforms to advertise statements such as “99.9% of investors experience positive returns” (Lending Club). And with regards to the 2011 loans, over the period since we have experienced some of the most accommodating monetary policies in the history of modern finance, with central banks flooding the economy with money driving asset prices higher and default rates to historical lows, which will naturally feed into lower consumer finance default rates as well. Therefore the 5.3% default rate on loans originated in 2011 should be looked at as the absolute best case in my opinion.
All we have to do is look at default rates historically to see that what part of the default cycle we have been in for the past 4 years. The first chart below shows the corporate default rates historically. So we can see that the default rates for the riskier companies are just under 2%, whereas during crises (housing crisis of 2008, tech bubble of 2001) default rates spike to 10%:
To look at consumer lending, we can look at the Experian Bankcard Default Index which is a bit more comparable to P2P lending, and that is currently 3%, and rose to 9% post the 2008 housing crisis:
So the issue is fairly clear, P2P lenders love to advertise low default rates and how everyone makes money, but that is due to a combination of historically low default rates in general due to lax monetary policy and the fact that majority of loans made have not yet gone through a principal repayment. Therefore I would be very hesitant at making any default assumptions from the limited amount of historical data, especially since the average loan maturity of these things is 4-5 years and looking at the charts above it’s not unrealistic to think that we have another default spike within that timeframe.
Issue #2: Is the marketed diversification really there when you truly need it?
I took all the loan requests on the Funding Circle website, and breaking them down by use of proceeds highlights a big concern. We can see that 89% of the loan requests are either to pay off credit cards or for loan refinancing and consolidation, which is pretty much the same thing. So essentially almost 90% of the loans you can make are to people in debt who are trying to roll over their loans. And the remaining 11% doesn’t look much more promising either:
As a general rule, I would only consider making loans where the money that was being lent was used in a way intended to make a return. There is a big difference between lending to a small business to expand operations versus an individual that is using the money to go on vacation. The former is likely to earn a return thereby making repayment more likely, whereas the latter is dead money (in investment terms) and so repayment depends on that individual finding another fool to lend to them. Now there may be some of the below loans that fall into this category, for example someone could take a loan out to pay medical bills, which would allow them to return to work and earn money, thereby providing a return, but overall I would be very skeptical.
The main pitch that P2P platforms like to use is the idea of diversification, so lending a small amount to a lot of different people, therefore reducing your risk. As I’ve talked about before, uncorrelated investing is a great way to adjust your risk adjusted returns, but I would be concerned looking at the above table that the returns are properly diversified. It’s not important how correlated investments are during the good times, it’s important that investments are uncorrelated in the bad times, and I am not sure that in a recessionary environment, lending to various highly indebted individuals that you already know have a tendency for credit debt is a proper diversified strategy.
Issue #3: The returns don’t line up with the risk taken
This is actually a problem of framing. Breaking down the loans available on funding circle by risk category and weighted returns:
So overall you are achieving 11.3%, with the riskiest loans returning 20%, and lending out money for 4-5 years on average. So let’s take a look from the perspective of the borrower, does it make sense for them to borrow at these rates. Well, what are the alternatives? They can get a credit card from a bank, with rates at 20%+, or a short term loan from someone like Wonga who charges hundreds of percent of interest. And we have already established that the majority of these borrowers are attempting to refinance existing credit card debt. So from their point of view, why wouldn’t these individuals pay sub 20% for 4-5 years as opposed to paying a higher interest credit card that is due monthly? And if it is such a great deal for the borrower, then surely it’s a bad deal for the lender.
Getting a bit more scientific, what is the implied default rate at which point you breakeven? For a 10% return, its approximately 9.1% per year:
Note that it’s actually similar to the crisis default rate on the bankcard index in the above graph. Some might say that if we are lending at the same default rate as the worst case scenario, is that not a good bet? The counter argument to that is that I would expect the crisis default rate of these P2P loans to be significantly higher than what is presented on the bank card index. For starters, the borrowers on P2P platforms already have high credit card debt (going back the the use of proceeds table).
As mentioned previously, personally I intuitively do not feel comfortable lending money for a purpose where the money isn’t used to generate some kind of return. Lending for the purposes of refinancing existing debt is just money thrown down a black hole in my eyes, bordering on a miniature ponzi scheme. And even then, I would prefer to invest money where there is equity upside rather than just fixed repayment. Additionally, on a long term timeline, I would still consider investing in a basket of larger companies (with long term equity returns of 5%-9% depending on dividend inclusion) where you are a part owner in the business (which is all buying stocks is), rather than lending money at 10% to some random Joe that has high credit card debt.