Adventures in Personal Finance: P2P Lending

My buddies and coworkers will tell you I’m a nut about personal finance. I spend more waking hours than I care to admit searching for the best way to allocate money. I think it’s because I am, at heart, an optimizer, and finance is rife with optimization opportunities.

I’m a young person investing for the long-run, which means the conventional wisdom is that the stock market is where almost all my money should be. I’m worried about structural risks in the stock market, though, and with the Shiller P/E ratio at 24.5 today, this seems like a particularly good time to look for non-correlated diversification opportunities. That’s why I’m poking at everything I can in search of a better way to invest my money.

Today’s adventure is in peer-to-peer lending. I’ve just opened my first account ($1000 in a taxable account at Prosper). Here’s what I learned along the way.

TL;DR: Attractive top-line returns, crappy taxes, frustrating implementation.

A Pedantic Aside

I wouldn’t be me if I didn’t stop right near the beginning to say something pedantic. In this case, I want to point out that as a personal finance geek, “peer-to-peer lending” is a marketing term that I really don’t care about. What I care about is diversifying into a high-return asset that is not highly correlated with my other assets, and in this case, the actual asset is consumer credit. From mortgages and car loans to credit cards and margin accounts, credit is the lifeblood of the consumer economy, but it’s usually the exclusive domain of massive banks and shady loansharks.

My advice is to ignore anything touchy-feely about the peer-to-peer part. We’re here in search of an asset class that should generate stock-like returns even when the stock market isn’t doing well, and that class is consumer credit. It just so happens that our way in is through peer-to-peer lending companies.

How It Works

The business model for the peer-to-peer lending networks is pretty straightforward: the lending network attracts consumers who want loans, most often for debt consolidation. The network vets the potential borrower and offers him or her a loan with an interest rate based on their expected riskiness. If the borrower accepts, then the network turns to its investors (that’s you and me), and sells the investors slices of the loan, called “notes.” As the borrower pays off the loan, the network takes a cut and distributes the remainder to the investors who hold the notes for that loan.

These are generally unsecured loans, so the interest rates are pretty high: the lowest interest rate Prosper offers to borrowers right now is for 6%, and those rates top off around 28%.

These rates are also risky, and can go unpaid. That 28% interest rate corresponds to an 17% default rate, meaning the real return is more like 11%. If an individual loan goes unpaid, the investors who hold notes in the loan get nothing. That’s why investing peer-to-peer lending requires diversification through small notes held on many loans to be successful.

It’s also worth noting that your returns on a given set of loans will deteriorate over time: almost everyone has the cash to make the first payment on a loan, but 35 months in, the defaults on your riskier loans will drag down your overall returns. The number you care about most as an investor is the “seasoned returns,” meaning the equilibrium that you settle into as the notes age. For Prosper, that range is from 5.5% to 11.3%, depending on the riskiness of the loan.

You can’t directly liquidate a note with the lending-network, so in some sense your money really is locked in to the notes and the network for the lifespan of the loan (3 or 5 years). Both Lending Club and Prosper offer second-hand markets, though, where users can sell their notes-in-good-standing to other users, allowing some retrieval of capital if necessary. I haven’t played with these features myself yet, though.

Meet the Competitors

There are two big boys in peer-to-peer lending that are accessible to a decent slice of the population: Lending Club and Prosper.

There are a few others that publish appealing numbers (like Avant), but they require you to be an accredited investor, which is to say, you need $1,000,000 in the bank or an income to the tune of $200,000 to invest through them. Since I’m just a little rich and not a lot rich, that didn’t work for me.

It should be noted that the table stakes for Lending Club and Prosper do require you to be decently-pocketed from the get-go: “Financially Suitable” investors have an income of $70,000 as well as at least that much in the bank. If you’re a sucker like me living in California, bump that up to $85,000. And then there are a number of states where you’re not allowed to participate at all (that list of states varies by lending network). There are a few asterisks and exceptions around these limitations, so it’s worth checking for yourself if you’re under that limit, but the point is that it’s not a poor man’s game.

Lending Club and Prosper are incredibly boring to compare. They were founded in 2006 and 2005, respectively, offer the same broad set of loans to borrowers, take the service fees, have the same $1000 minimum investment, and offer notes in the same minimum $25 increment. They both rebooted their models after the recession in 2008/2009.

The biggest difference from the outside is that Lending Club is significantly larger, and its publicly traded. Prosper continues to gobble up venture capital ($70M in 2014, $165M in 2015). Since both companies say other banks would service the outstanding loans in the event of a bankruptcy, I don’t think there’s much reason to see a difference in security between the competitors, but it’s worth highlighting.

The biggest difference that I see is in expected returns: Prosper’s past returns are just plain higher, to the tune of 1% or more. My guess is that this is because Lending Club’s larger scale means they face more pressure to have more notes available to investors, and that means they need to offer more loans to borrowers at lower interest rates.

The Good

I won’t lie: there are a number of bad and ugly parts to peer-to-peer lending that I’ll get into soon. But far away the best thing about it is the respectable returns with a weaker correlation to the stock market. When you buy into notes, you get the option of what level of risk you want: a portfolio full of top-rated AA loans gets you a pre-tax expected return in the 6% range, while a riskier portfolio can hit 9 or 10%.

Bear in mind that these returns aren’t truly uncorrelated: a recession that damaged consumers’ ability to repay loans would hurt lending network returns and also hit the stock market. I have a friend who invested in the early days of Lending Club in 2008 and saw comparable losses there as he saw in the stock market.

The business model also hasn’t been around that long, so we don’t have a lot of historical data to compare. And since 2008–2009 when Lending Club and Prosper developed their current business models and started tracking returns, the stock market has had an amazing positive run.

But all that said, peer-to-peer lending portfolios have demonstrated a remarkably low-volatility positive return during quite a volatile time in the stock market. If the topic engages you, I’ll direct you to this piece on the reliability of consumer credit as an investment instead of stealing all its graphs myself.

The Bad

There are a lot of frustrations associated with peer-to-peer lending. Let’s get into it.

Limited Liquidity

There just aren’t that many people taking loans compared to the number of people who want to invest in them. That low volume, combined with the automated third-party API-driven investment tools used by high-rollers sucking up attractive loans, means it takes a while to invest money. My $1000 into my account arrived on Friday. I wanted to invest in the higher-risk, higher-return funds so that I could match my expected stock market returns. But when I applied the corresponding filter to the available loans, I was only able to invest $400 that day without double-dipping with multiple notes in the same loan. I loosened my filter to accept more lower-risk, lower-return loans. Even then, Prosper’s auto-invest tool only bought me notes in three more loans Saturday. At this rate, it will take more than a week to fully invest $1000 in a portfolio with an expected return of 8%.

There is higher availability at lower risk levels, but it would still take weeks to invest even $10,000 without buying multiple notes in a single loan.

Nasty Taxes

The tax treatment for peer-to-peer lending really sucks.

The interest payments that investors receive are considered taxable income and are reported on multiple 1099 forms, so that brings a whole bunch of taxes. I haven’t figured out yet exactly which state payroll taxes are applied, but at the very least you’ll be paying your full marginal income tax rate. For a sucker like me in California, that cuts out 38% of the return on the investment.

Compared to long-term capital gains (24.3% for me in California, 15% in sane and reasonable places), the tax treatment of peer-to-peer lending returns is quite a bit worse than shares of your favorite index fund.

Unfortunately, the losses from loan defaults don’t offset that income; instead, losses are treated as capital losses for tax purposes. That means they can offset capital gains on stock sold this year (but if you’re like me, you’re rarely selling stock anyway). You can offset $3000 of regular income with capital losses each year, and any excess capital losses can be carried forward into future years, but it’s still not an ideal situation.

IRA Accounts

If you’re like me, you saw that taxes are ugly and immediately thought of using your IRA as a way around them. Good impulse! Unfortunately, the peer-to-peer networks are not your friend here.

Both Lending Club and Prosper offer IRA accounts, which eschew the tax problem, but they require a serious commitment. There’s a minimum investment of $5000 for an IRA account, and they’ll charge you a $100 fee for each year after the first that your balance is less than $10,000. In other words, if you don’t commit nearly two full years of IRA contributions to peer-to-peer lending, you need to pay a fee that chews up a big part of your expected returns. I don’t know about you, but I’ve only got about $25,000 in my Roth IRA account: committing such a big slice of it to a new asset class that has some new uncertainties is a move I’m not ready to make.

These fees aren’t the only problem for IRA accounts on lending networks: the minimum investment level also has a nasty intersection with the liquidity problem of these networks. If I’m on track to spend a week deploying just $1000 of capital, investing $10,000 could take more than two months! The opportunity cost of the thousands of dollars sitting inactive for a month or two cuts the expected return in the first year further.

Moreover, notes can’t be moved from taxable to IRA accounts, so there’s no way to build up a taxable account and then make the switch. IRA accounts are all-or-nothing.

As best I can tell, the only reason for these fees and limits is so that some bank that hosts the IRA accounts for the lending networks can get greedy on the fees. If these fees change and volume improves, I’ll be thrilled, and with a solid 8 or 9% tax-free return, I could happily move $5000 into it.

The Ugly

Opacity and Misleading Returns

My biggest frustration in moving into these networks has been their opacity. Their published return numbers date back to 2009, which creates the illusion of consistent data, when in fact the system lending networks use to evaluate the risk of borrowers is always changing. Similarly, the interest rates that they charge are always shifting. Since the track record for the sector is only about 8 years long, and a lot has changed, the “steady 9% return” should not be taken as gospel. The general consensus I find in reading around the web is that average returns are dropping, mostly as the networks try to expand their volume with lower interest rates.

That’s a fine business strategy for the networks, but when all the historical data put front and center on the investor information pages includes returns from years with much higher interest rates and a different clientele, I can’t help but feel I’m at risk of getting hoodwinked.

False Sense of Security

The networks harp hard on the stability of their returns, and how no investors with more than 100 notes have had negative returns. We shouldn’t take that too seriously, though: it would have been hard to have a negative 3- or 5-year return if you bought 100 random stocks in 2008 too. These notes are an investment that face their own structural risks, and that needs to be borne in mind: they’re not magical 9% bonds.

The “no negative returns” number also brushes aside the problem of taxes: if the returns are highly taxed and the losses aren’t deductible in the same category, then a positive gross return can be a negative post-tax return.

What I Did

I’d been toying with the idea of peer-to-peer lending for so long that I needed to stop punting and just try it. At the same time, I’m very aware of the various “bad” and “ugly” components I highlighted above, and I didn’t want to be a sucker by going in too hard on a new sector.

So, I ended up buying into Prosper (for the higher returns) at $1000 (because I can afford for it to be a dud) in a taxable account (because the IRA requirements are outrageous).

I’m certainly not an evangelist yet: I don’t think more than 2% of investments will be in consumer credit until I’ve gotten quite a bit more experience with these notes and I really understand the nuances of their messy tax treatment. But, for the time being, this feels like an experiment worth running.

Have questions? I’ll do the best I can to answer them at @jordanp.