Tuesday, September 15, 2020

Fintech in chains

"Fintech can come out of the shadows" is the title that Wall Street Journal editors gave to  Brian Brooks and Charles Calomiris' oped last week. I have not in a long time seen a title that more utterly contradicts the content of the essay.  For what they advocate is exactly the opposite: Fintech in chains, hemmed in by  the sort of regulatory stranglehold that fintech was created to escape. 

What is fintech? Basically companies that offer  

services—consumer loans, credit cards or payment processing—that banks have traditionally offered.

but, crucially, fintech does not accept deposits. 

The issue? 

The Office of the Comptroller of the Currency determines which companies qualify for charters as national banks or federal savings associations and supervises the activities of those banks.

But not fintech companies, because fintech companies don't take deposits. And that is the legal issue prompting the oped -- Brooks and Calomiris, coincidentally acting comptroller of the currency and chief economist of the OCC, want the OCC to regulate fintech just like banks. (Calomiris is a topnotch economist who normally writes very good papers. ) 

So what's so awful about fintech? 

..Fintech unicorns such as Square, PayPal and Stripe have similarly captured a large and growing segment of the payments business that national banks previously dominated.

...technological changes have created new fintech competitors that can form [?] banking transactions.... Many customers—whether consumers with minor credit blemishes, or small retailers [that banks don't serve]—have voted with their feet and taken their business to specialized (and often cheaper) marketplace lenders and payments providers.

Traditional bank lending and payments have earned low or negative returns in recent years, compared with returns of 50% to 150% for specialized fintech providers in payments, bank and capital markets technologies, information services, and payroll services.

Technological change, customer preferences and market performance tell us that unbundling is inevitable—services that banks have historically provided will increasingly be offered by specialty firms, often with a fintech flavor. 

...banks must adapt or die. In a market economy, customers will obtain loans and receive the payment and other financial services they need in the manner they choose.

This sounds like a classic free-market success story! A sector (regulated banks) is sclerotic, over-regulated, complacent, and protected from competition by regulators.  New competitors innovate their way around regulations to provide better, cheaper, more tailored services. Consumers flock. Read Uber vs. Taxicabs. Why should we not celebrate? And why should the taxi commission regulate Uber? 

What is the theory of regulation here? There is enough free-market economics left in the world that the baseline remains freedom and competition. You need some problem, some market failure, some reason for regulators to exist.

For banking there is a reason for regulation, and it is precisely deposits. Deposits are open to runs. Runs are bad for all sorts of reasons. Financial crises are runs. Hence we have banking regulations. That's not a defense of current banking regulation. But there is a problem and a reason for some regulation.

But fintech doesn't take deposits. The one central problem with banking is gone. Why should they not get a gold star, a thank you for not endangering the financial system plaque, and be sent on their way to better serve customers in a ruthlessly competitive market? (Much more on banking without deposits here. )

So what is Brooks and Calomiris' theory of regulation? Why do we need the OCC to regulate fintech just like banks, and deliver us the hearty innovation and competition their regulated banks have shown (sarcastic tone)? 

The Office of the Comptroller of the Currency determines which companies qualify for charters as national banks or federal savings associations and supervises the activities of those banks to ensure fair access to financial services and the safety and soundness of America’s federal banking system. The commitment to safety and soundness has a record: National banks fail less often than state-chartered banks and lose less money when they do.

That's it as far as I could see. "Fair access'' and less failure. 

Why do we care about failure? Is the purpose of federal regulators to make sure businesses never fail? 

Sure, there is a reason to worry about failure in banks that take deposits. Such failures cause runs. But no deposits, no runs.  So what's the problem if a fintech company fails? 

Hot dog stands are regulated by cities, largely for cleanliness. Do we need a federal hot-dog stand regulator, to make sure hot-dog stands are "fairly" distributed, and evaluated by making sure hot-dog stands never fail? Should the OCC give federal hot-dog-stand charters? 

The number one way that regulators stop businesses from failing is by limiting entry and protecting them from competition. Is this not exactly what the OCC has done to banks, which is why shadow banks are growing?

Look through your economics textbook on justifications for regulation. Protecting business from failure is not one of them. 

What about "fairness?" In their first paragraphs Brooks and Calomiris pointed out that fintech is growing and  making a ton of money by lending to and serving people that regular banks wouldn't touch! So much for "fairness" imposed by regulators vs. fairness that comes from a competitive market searching for under-served customers. 

This is hilarious: 

So why are New York and other states fighting the OCC’s authority to charter national banks that engage in many core banking services but happen not to take deposits? They could simply be protecting turf. Or they could be worried about losing the money they earn from licensing shadow banks. In 2019, New York alone oversaw 113 state-licensed money transmitters, 18 nonbank lending companies, 92 sales finance companies, and other companies, some of which might qualify as national banks. Regulatory assessments alone earned Albany more than $100 million.

Recall that the authors work for the OCC, which is... well, protecting turf and at least the budget and bureaucracy it gets by regulating banks!  

No. An ideal OCC should be working to encourage migration to fintech companies, that are not financed by short-term debt, and hence  pose no risk to the financial system, and are not regulated by the OCC. And then, like the Rural Electrification Administration, its job complete, it should fade away. 

(Almost all mortgage originations now go through fintech companies not banks. A good paper: Buchak Matvos Piskorski and Seru.) 


Updates. My benevolent treatment of fintech presumes that they are financed by equity and long-term debt, not short term debt. Buchak et al find that to be the case, but obviously a fintech that financed itself 99% by overnight borrowing in financial markets is systemically dangerous. 

Macro di Maggio and Vincent Yao have a very nice paper on fintech. From a quick read, I see confirmation that fintech does lend to people that regular banks will not or cannot for regulatory reasons. Isil Erel and Jack Liebersohn  also show "that FinTech is disproportionately used in ZIP codes with fewer bank branches, lower incomes, and a larger minority share of the population, as well as in industries with little ex ante small-business lending." 

There is an elephant in the room, which we might as well bring up in this context. The Fed is moving into further into inequality and racial justice. The OCC seems to be moving in that direction as well, see for example their (regulatory, that's what they do)  effort to "promote investment and lending in underserved areas and to highlight activities supported by the Office of the Comptroller of the Currency’s (OCC) new Community Reinvestment Act (CRA) rule." The Democratic Party Platform states "Democrats support making racial equity part of the mandate of the Federal Reserve." 

Whether you like or dislike the idea, clearly a movement is building for regulatory agencies and Congress to steer credit in certain directions, to specific areas, businesses, and organizations, and away from others. This includes where the Fed sends its prodigious money, but more deeply how the Fed and other regulators use regulatory tools to push banks and other institutions under their control to lend.  A parallel movement for central banks to squeeze out fossil fuel companies and subsidize green bonds is underway in other countries. 

You can see a storm brewing: politicians want to control where credit goes and squeeze some organizations out of credit, and they will want to control the fintech flow as they do the flow from regulated banks. That academics find fintech on its own reaching out to serve minorities, low income etc. will not persuade them, because politicians like to take credit for it, they will object that companies are (heavens) making profits by doing it, and they also like to subsidize organizations not people.  

Another big unanswered question for fintech: It uses AI algorithms to screen borrowers. AI is notoriously hard to unpack -- just why did the algorithm give a loan to Joe but not to Sally? You can see a storm brewing over whether the AI uses statistical discrimination -- conditional probabilities of repayment based on a huge range of characteristics -- in ways regulators do not like, or that we all might not like. This is not directly about race or other protected characteristics, which fintech does not use because it is illegal. But there are lots of statistical facts that people might not like being used, and are correlated with race and other protected characteristics. Just because I live in a zip code, work at a job, eat kinds of foods, buy certain products, am of a gender, own a car model, and browse the internet in a way people who don't pay back their loans do, why should I not get a loan? The DOJ suit to automobile finance is a tip of this iceberg. 

My opinions on whether these moves will promote racial harmony, income equality, or a cooler climate are irrelevant to the present discussion. Let's just put above board that these trends are big important drivers of financial regulation today.  

Also, a commenter points out that one uniform national regulatory standard for fintech might not be a bad idea, and suggests that Brooks and Calomiris have such a proposal. If someone has seen it, send a link. Their oped certainly did not mention such. 

The pros and cons of one uniform regulation vs. multiple competing regulations are obvious once you state the question. And there are both pros and cons. 

If the one uniform regulation were just immense capital standards, then do what you want I  might be for it. 


  1. The problem with runs comes from a debt dominated financial system. As long as these shadow banks are issuing debt and lending onward, runs (or the effect of one) can still happen even though no deposits are involved. For whatever reason, if investors decide these shadow banks are not safe and refuse to lend more at some point, that can result in a fall in consumption which causes a fall in income in others, and then defaults to shadow banks, thereby reinforcing the thesis that shadow banks are unsafe. If you need a lender of last resort, then you need regulation. The problem is debt, not deposits.

    Allowing equity investment in small businesses, individuals etc. and eliminating the preferential tax treatment of debt seems like a better solution to me than unregulated lending through shadow banks. For example, Lambda school goes in this direction and is way better than education loans in my opinion. (Oh and get rid of those regulations that stop small, "non-accredited" investors from investing in startups and small businesses.)

  2. My preference would be for federal regulation of some* Fintech, though for different reasons than the article you're critiquing. Specifically I think federal preemption would be beneficial in this area.

    Any business that exists primarily to hold and/or move vast amounts of other people's money is going to be regulated by someone. I'm not arguing whether that's right/wrong, just pointing out a fact. So the only issue worthy of debate is who that "someone" ought to be. Given that the current patch-work of state "money transmitter" rules are truly a slog for companies, my own take is that pragmatically federal involvement would (counter-intuitively) result in less overall regulation. Often when the feds step into an area, states either step-out or conform to the federal rules.

    I get that's perhaps not so good as nobody regulating, but nobody regulating isn't a choice. Companies moving 9-figure sums of currency around will be regulated. My pick would be the feds.

    *I'd draw a distinction between non-banks that move cash (e.g., payroll processors, retail payment processors, etc.) vs. those that do things like mortgage origination. To me at least, it's the money movers that are the issue.

    1. What kind of regulation are you thinking of? We already have laws against stealing.

    2. From 10,000 feet, most state regulation in this area tries to impose some combination of the following three requirements: solvency (can't be a money transmitter if you're bankrupt), co-mingling of funds (can't keep the money you're moving about in the same bank account as your earnings) and surety (you need some insurance if you'd like to be a money transmitter).

      We could debate whether the above are necessary given we already have laws against stealing, but my point is why bother? Such regulations already exist in a complex patch-work at the state level, at least in part because there has been a vacuum in regulation at the federal level.

  3. My only observation: in order for your argument to be valid users of fintech solutions should have no expectation whatsoever of receiving any compensation from an eventual financial loss in their fintech operations other than those stemming from litigation. I wonder whether they have that expectation and whether the pricing they face reflects that. In my country (Portugal), certainly that is not the case. An example. After the demise of Banco Espírito Santo, holders of junior debt campaigned years for full reparation and in the end they obtained a generous partial reparation with taxpayer money. I see the same type of attitude with Revolut and other fintechs. People believe that their funds are protected. They aren't, at least ex ante. Ex post, it's another story.

  4. This is about Wirecard https://www.ft.com/wirecard
    There will be a call for more regulation after it
    But the whole business is quite dirty. It is getting better,

  5. Very timely and insightful post!
    One question: does fintech really not take any deposits?

    I have a recent paper about the explosion of shadow credit driven by Basel III in Korea (uh.edu/~rpaluszy/banking.pdf). We find that, for example, about half of all shadow lending is provided by insurance companies. And indeed, they have zero deposits in their balance sheet data. But isn't a "cash value" life insurance policy de facto a deposit?

  6. I think that the underlying theory is something like: "Deposits are risky, in fact the inherent bank business of borrowing short term to lend long is risky, therefore banks need cross-subsidization in other ancillary services, if fintech exists and is not regulated the cross-subsidization will go away."

  7. I’m not sure what deposits are...it feels like lots of these phone apps let me hold positive balances on them

  8. Is this similar to the turf protection the Fed exhibited when it denied access to a "narrow bank?

    1. That was even more explicitly to deny entry and defend profits of existing banks, and feds ability to command cross-subsidy. At least fed was open about it!

  9. It reminds me of Merton Miller's thesis that every important financial innovation was designed to circumvent regulation e.g., Euro dollars (dollar denominated savings accounts offered by European banks) were designed to circumvent Regulation Q, the 5.25% price ceiling on what US banks were allowed to pay depositors; similarly, Hedge funds got around the short selling ban on mutual funds, and Private equity thrived after SOX.

    Now we see money flowing into betting markets to circumvent the ban on short-selling financial stocks.
    BetsForTraders.com, a financial bookmaker, has reported volumes up by more than 400 per cent since last Thursday's ban, with almost all the increase in activity in bets against banking stocks.

  10. "The Office of the Comptroller of the Currency determines which companies qualify for charters as national banks or federal savings associations and supervises the activities of those banks to ensure fair access to financial services and the safety and soundness of America’s federal banking system." Really? I suspect this a classic case of regulatory capture whereby the Comptroller of Currency is heavily influenced by the banks they ostensibly regulate. The timorous fear competition.

  11. I'd distinguish between the WSJ op-ed (aimed at a specific audience) and the proposal itself, which as far as I know is to give fintechs an alternative to 54 (with states & territories) money transmitter licenses. Later on, they'd like the Fed to give fintechs a master account, too. So, they want to give fintechs more options but not obviously take anything away. If innovation is easier at the state level (which it often is), fintechs can continue to undertake it there.

    1. Do you have link to proposal? They didn’t mention it. A uniform very light regulation would make sense.

    2. Sorry - just saw this. There hasn't been a formal proposal yet. While I'm not a lawyer, some people have said the reason there is no proposal is that states would then sue (like they did with the OCC's previous attempt to offer special-purpose fintech charters) and this would discourage any firm from applying for a charter. If that hypothesis is true, then the OCC might prefer to charter fintech firms first (if they apply) and deal with lawsuits afterward. States can't sue unless the OCC formally proposes something OR issues a charter.

  12. The issue is cross-subsidization. The OCC wants to subsidize deposit-taking. It should make a case for why this is worthy.

  13. The legitimate regulatory failure that is driving this proposal is the fact that non-bank fintech lenders and money transmitters (at a minimum) are at a pronounced and unjustified regulatory disadvantage versus their bank competitors. This frequently forces fintechs to partner with banks, adding cost and further regulatory uncertainty. At best the OCC is using its (debatable) exiting authority to create a mechanism to address this through a non-depository national bank charter. This is not an ideal solution but is the solution available under current law. Of course, there may also be empire building/government control issues at play as well, but there is an underlying "there" there.

    At the risk of shameless self promotion here is an article discussing this dynamic in greater detail: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2928985


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.