But will financial democracy work in a downturn?
SAVERS DO NOT get much in the way of interest from their banks these days. But a different logic seems to apply to borrowers, who still often pay double-digit rates for credit—if they can get it at all. That has attracted a number of outfits offering to connect those who need cash with those who have a surplus of it. The rapid growth of such “peer-to-peer” lenders has been one of fintech’s most visible successes. The biggest such firm, Lending Club, based in San Francisco, listed its shares in December to a clamour reminiscent of the 1999 tech boom.
Fans compare peer-to-peer lenders to other pioneers of the “sharing economy”. Like Uber with cars and Airbnb with accommodation, the newcomers are making available a commodity they do not provide themselves: in this case, money. Instead of a bank intermediating between savers and borrowers, the two parties deal with each other directly. The platforms do the credit-scoring and make a profit from arrangement fees, not from the spread between lending and deposit rates.
The sector has grown rapidly: the five biggest platforms for consumer lending—Lending Club, Prosper and SoFi, all based in San Francisco, and Zopa and RateSetter in London—have so far issued nearly 1m loans between them and are generating more at the rate of well over $10 billion a year. The Anglo-Saxon countries are the spiritual home of credit, and so of peer-to-peer lending, but smaller platforms exist in mainland Europe and China.
Those loans are still dwarfed by the $3 trillion of consumer debt outstanding in America alone. But the sector is doubling its lending roughly every nine months, and almost everyone expects it to go on growing rapidly. Having started as a provider of unsecured consumer credit, competing mainly against banks’ credit cards, it has expanded into lending to small businesses, student loans and now mortgages.
Though most of the lenders were established before the financial crisis, none thrived until its aftermath. This was partly because the banks’ rapid retrenchment after 2008 created unmet demand for loans. In America, even those who could still borrow from conventional sources soon found that peer-to-peer providers offered better deals. Credit-card rates tend to remain stable through the economic cycle, so they have looked especially uncompetitive as central banks pushed interest rates to record lows. Lots of borrowers paying 18% on their credit-card balance found they could take out a peer-to-peer loan charging 14% instead. On the other side of the equation, low interest rates meant savers were open to new investment opportunities, including lending their money to perfect strangers on the internet.
Knowledge is power
More broadly, says Hans Morris, a venture capitalist who sits on Lending Club’s board, the declining cost of information-gathering is pushing consumer credit the way corporate credit has gone over the past three decades. In 1980 only a few hundred blue-chip firms could borrow from investors other than banks, by issuing bonds. By the end of that decade, all creditworthy firms could do so, and by 2000 “junk”-rated firms were at it, too. But whereas the incumbents, through their investment-banking arms, played a key part in the lucrative business of helping firms issue bonds, they have no role in peer-to-peer lending.
Those pining for the democratisation of finance have been disappointed by one notable development: most of the money for peer-to-peer no longer comes from the general public but from institutional investors such as hedge funds. The industry makes no secret of this; in America many firms have dropped the peer-to-peer label and instead describe themselves as “marketplace lenders”. The shift has increased the supply of money to borrowers, but also made it harder for the newcomers to present themselves as markedly different from the banks.
Yet from a regulatory point of view, they are indeed very different. There is much to like about peer-to-peer, no matter whether the money is being put up by a hedge fund or by the general public. A bank is fragile by nature: when it faces a slew of defaults on its loans, it rapidly runs into trouble. That is because it cannot pass on losses to its main creditors, often the bank customers who deposited their money on the firm understanding that they would get it back. Even when capital cushions designed to absorb lending losses are bolstered after crises, as happened after 2008, the risk of a taxpayer-funded bail-out or some other state support is ever present.
By contrast, those who lend money through peer-to-peer platforms explicitly accept that they may suffer losses. Unlike bank deposits, their investments are not guaranteed by the state. And whereas banks are subject to runs when too many fickle depositors demand their cash, lenders on peer-to-peer platforms know they will get their money back only when borrowers repay their loans.
A core task
Not all peer-to-peer lenders work the same way. Some platforms allow potential lenders to pick their borrowers, others oblige them to lend to all those approved for credit. British platforms typically feature protection funds, designed to compensate lenders exposed to loans that have defaulted. This twist makes them far more akin to banks. For all their differences, the peer-to-peer platforms perform one of the core tasks of the banking system: they pick the applicants who get credit, and at what interest rate. Many claim to be doing a better job than traditional lenders.
A common refrain is that banks are on the defensive, trying to keep risk-averse regulators happy. The peer-to-peer crowd do not have to contend with that, giving them scope to try new things. All of them start their assessment of potential borrowers by looking at a raft of readily available consumer data from credit bureaus such as FICO and Experian, which track who has welched on past bills or car payments (banks use these too). They overlay that with whatever information they can get their hands on, from employment history to verifying pay cheques directly with employers. Borrowers may be asked to provide their online banking details so their financial history can be downloaded from their bank’s website. That means the incumbents no longer have much of an information advantage over anyone else.
Any data can be mined for insights, says Martin Kissinger of Lendable, a British newcomer: how often someone has used a credit card to withdraw cash, say, or whether he makes minimum monthly repayments. Zopa tracks the applicants it has turned down for loans to see if they turned out to be good credit risks when they found another willing lender. “We don’t necessarily have better data, but we are far better at analysing what we have,” says Giles Andrews, its boss. Social-media activity was once touted as the new frontier for credit-scoring, but is no longer considered so useful except, crucially, to help prove an applicant’s identity. In America, rules intended to ensure that credit is allocated fairly—by protecting minorities whose neighbourhoods used to be “red-lined” by bankers—make it harder to use novel techniques.
Kreditech, a German startup which makes short-term loans in countries from Peru to Poland, says it uses 20,000 data points to extend high-interest credit at a rate of $120m a year. Beyond using Facebook data, it says it can “triangulate the truth” about a customer’s creditworthiness by using behavioural data such as the way its online application form is filled in. How often a customer uses capital letters, say, or the speed at which he moves his mouse during the process are useful clues. “We are a tech company that happens to be doing lending,” says Lennart Boerner, its head of strategy. If Silicon Valley dismisses the idea that bankers can gauge their customers’ creditworthiness by meeting them face to face, bankers may consider fintech’s method as sorcery.
Some credit-scoring is more intuitive. SoFi has carved out a niche pitching credit to what the industry calls HENRYs: high income, not rich yet. It built a franchise refinancing student loans for asset-poor but high-potential graduates of top universities, whom it sees as good credit risks. Those loans run to around $75,000, against the $10,000-$15,000 more typical on other platforms. “Our credit assessment looks to the present and the future, not just the past,” says Mike Cagney, its boss. That has a harsh flipside: those who default on their loan risk having their name broadcast to the lenders, “so the whole community knows you’re a deadbeat.” It is the first established platform to branch out into mortgages, offering loans worth up to 90% of the value of a house—much more than a bank.
Many people will feel it is too soon to encourage innovation in underwriting, let alone higher loan-to-value ratios, given what happened in 2008. Sceptics argue, rightly, that divorcing the party which authorises credit from the party which will suffer from a default has proved disastrous in the past. Was the financial crisis not triggered by borrowers being given too much credit by mortgage-brokers who cared little if those loans were repaid? How are peer-to-peer platforms different, given that they immediately offload the loans they have approved?
The comparison is unfair, says Renaud Laplanche, Lending Club’s founder. Before 2008 subprime mortgages had long, diffuse chains of intermediation. By the time a mortgage was brokered, sold, sliced, diced, repackaged and resold into the market, few cared or even remembered who had issued it. With peer-to-peer, the chain is much shorter. “If loans we issue do not perform, we have nobody else to point the finger to,” says Mr Laplanche. A platform that issues dud loans will struggle to attract bidders, be they hedge funds or the general public.
The bigger question is what happens when economic conditions turn. Peer-to-peer lending, though enabled by technology, would not have flourished without the benign credit conditions of recent years. For all the talk of superior underwriting, the industry’s claims of beating banks at their own game will be tested only when interest rates rise or the economy tanks. The industry is aware of this. “My daughter could come up with an underwriting model based on what band you like and it would work fine right now,” says SoFi’s Mr Cagney. But for how long?
At best, peer-to-peer lenders may find their advantage over banks becomes eroded. As interest rates rise, credit cards will probably become more competitive (though they may be pricier for less creditworthy borrowers). Peer-to-peer marketplaces will probably have to raise their own rates to attract investors lured by improved returns elsewhere. So the opportunity to arbitrate credit mispriced by banks may narrow, particularly in America.
At worst, a credit shock or a recession will leave existing borrowers unable to repay their loans. One worrying feature as the industry matures is that many borrowers are return customers: they are using peer-to-peer loans to refinance peer-to-peer loans taken out earlier. That is particularly true for riskier borrowers. If the industry were to contract even slightly, those unable to refinance would be pushed to default. If banks were to tighten lending criteria at the same time, the customers’ problems would multiply.
That might cause a downward spiral as withdrawals creep up: even a modest rise in dud loans might spook lenders, particularly flighty hedge funds. In the absence of fresh money to repay old loans, more defaults would be inevitable, followed by more exits by investors. That is one reason why most peer-to-peer lenders are eager to keep some of their loans funded by retail money. Mom-and-pop investors are thought to be “stickier” in a downturn, so their money will remain available for future loans .
All platforms vaunt their superior underwriting skills and boast of having “prime” borrowers, but they are also under pressure to show rapid growth in their loans. The temptation—which all claim to be resisting—is to relax their rules and pitch loans to those at the shadier end of the credit spectrum. This may be encouraged by apparently low default rates, but these are flattered by the rapid growth in lending: a 10% default rate will become 5% if a loan book has doubled in the meantime.
On the other hand, if peer-to-peer can weather the next downturn it should get a fillip. Big-money institutions such as insurance companies and pension funds have so far only dipped their toe into the sector. Many of them need better returns, and have long-term liabilities they are keen to match with long-term assets such as mortgages. If unsecured consumer loans perform as well in a downturn as their boosters hope, some investment titans will be tempted to buy paper from peer-to-peer platforms directly, dwarfing the hedge funds that are already there. A few might buy pools of mortgages from peer-to-peer lenders instead of tapping Wall Street for complex securities whose performance tracks the performance of those same pools of mortgages.
A more surprising investor in this field is the banking sector itself. Small local lenders in America have turned to peer-to-peer marketplaces to gain exposure to consumer credit; Citigroup said in April that it would lend $150m through Lending Club. This might bemuse observers: why would a bank buy a loan rather than issue it itself? Mr Laplanche points out that although banks’ cost of capital is lower, its cost of operation is higher. A bank spends roughly 7% of the value of a loan on administration, against Lending Club’s figure of just 2.7%. Still, some might question the business model of a bank that admits it cannot successfully underwrite loans itself.
A piece of the action
Peer-to-peer is the most established of all fintech’s branches. Lending Club is listed on the New York Stock Exchange, and has John Mack, a former Morgan Stanley boss, and Larry Summers, a former Treasury secretary, on its board. Goldman Sachs estimates that when peer-to-peer comes of age, it could reduce profits at America’s banks by $11 billion, or 7%. That would be troublesome but not unmanageable. Bankers point out that, leaving aside credit cards, unsecured loans to consumers are a fiddly business that is not particularly close to their hearts. The risk, though, is that a graduate who turns to a marketplace for her first loan then also shops there for services banks do care about, such as mortgages or investment advice.
Peer-to-peer lenders have their own problems, even when the economy is steaming ahead. Acquiring customers, which is often done through mailshots, is expensive and erodes margins. Overheads are rising steadily. But regulators have kept reasonably clear so far because the risks around this form of lending are borne by those who put in the money, not by the general public. As long as that remains the case, the challenge they present to banks should be heartily welcomed.