Note: This is the third in a series of posts where I’m offering my thoughts as to the role of peer-to-peer (or “direct”) lending within the broader context of the investment landscape.
Over the past few weeks, we’ve been discussing the role of peer-to-peer lending in an investment portfolio. The conversation started around the advantages of directly holding debt to maturity; then we focused on the essentiality of capital preservation. Next, we’ll drill down a bit further into the theme of protecting principal by analyzing the reasons for how this asset class minimizes downside. What are the underlying factors that make peer-to-peer lending – something seemingly risky and definitely “quirky” – a comparatively safe investment?
Fundamentally, it’s an issue of the spread between two basic, primary variables: a) gross interest rates and b) default rates. The spread between them is “net yield.” Sure, there are lots of other factors that come into play in determining a net bottom-line return. Servicing fees charged by the platform (Lending Club, Prosper, etc.) will reduce yield. Timing of defaults – as opposed to a flat absolute default rate – can either negatively or positively effect returns. Borrower pre-payments, idle cash, and asset manager fees are other factors that can repress returns below the fundamental spread between interest rates and charge-offs. In the spirit of keeping it simple, though – and to illustrate the point of how the characteristics of this asset class are akin to capital preservation – let’s focus on the delta between these two simple parameters.
Data from a variety of sources (US Federal Reserve, US Bureau of Labor Statistics, BankRate.com, CardHub.com, among a host of others…) generally lead us to coalesce around one straightforward conclusion. Over the past 25 years, average gross interest rates that consumers pay on unsecured revolving credit lines range from 13% to 19%. There are ebbs and flows in these rates, and there is some correlation to shifts in broader interest rate climates. Regardless, at no point do they fall below the teens, and it’s very realistic that peer-to-peer lenders can construct a diversified loan portfolio of notes that mirror that 13% – 19% range.
Conversely, a high level analysis of historical credit card charge-off rates by banks and other financial institutions shows that default rates seldom even remotely approach the teens – even in the worst of consumer economic cycles. Over the past 25 years, average reported charge-off rates are roughly 4.5%. Dating back to the beginning of 1991, default rates exceeded 6% in only 16 of 90 fiscal quarters. The highest reported charge-off in a single quarter was 10.97% in Q2 of 2010 – a time when unemployment rates were also historically high (between 9-10% throughout 2010). At that time, credit card interest rates were approximately 14% – indicating a still positive net yield of approximately 3%, in what was by far the worst consumer credit cycle in the past quarter century. The net relationship between these two variables has never been such that charge-off rates exceed interest rates, and, in a normalized environment, will range anywhere from 6% to 12%.
The fact that the spread between the two is positive in both good times and bad is remarkable. It signals opportunity. To be clear, credit card interest rates and charge-offs of bank credit card receivables are not entirely analogous to peer-to-peer lending interest rates or charge-offs in online lending portfolios. Risk averse brick-and-mortar institutions will typically lend at a lower price point in exchange for limiting losses; on the other hand, peer-to-peer lenders can play anywhere on the borrower risk curve – an investor can, if he so chooses, mirror what banks have historically done and lend to a subset of higher quality borrowers at lower interest rates with lower estimated write-offs. A peer-to-peer lender can alternatively construct a portfolio at a yield point much higher than would be typical of a book of credit card receivables. The comparison is admittedly not apples-to-apples for obvious reasons, but the data are a relevant starting point to illustrate the compelling characteristic of persistently positive absolute yields in various climates. (Lending Club and Prosper have published historical loan performance data, so by back-testing various strategies we are able to conclude that there is a strong relationship between the referenced data sources and actually peer-to-peer loan performance.)
My belief is that this opportunity may contract over time, and peer-to-peer lending is actually a step in that direction. In a market this big, however – with roughly a trillion eligible dollars – it’s going to take a while before direct online lenders become the lion’s share of the market and push yields down further. Eventually the mis-priced risk, the faulty assumptions of consistently astronomical default rates, and the stigma of high-interest consumer borrowers will start to fade, and the opportunity for investors to obtain such upside with minimal downside will be more difficult to find. In the meantime, though, those of us who understand these fundamentals have a unique opportunity to capture significant yields with the peace of mind that comes from a historical track record of surprisingly low default rates. Say what you will about the presumed risks of peer-to-peer lending, but this consistently positive net yield – access to which was traditionally reserved only for brick-and-mortar institutions – warrants serious consideration for each of us as a mechanism for capital preservation.