The problem with banks: the outline of a solution
TL;DR: Bank services are in the process of being unbundled. At their core, banks will retain their unique status as a government-guaranteed mortgage investment product. Card networks will also play a big role in determining just how many other services can be unbundled.
The final of a three part series. See parts one and two.
In part one I took a dig at banks. I pointed out that they weren’t popular with customers and were charging confusing, ever-rising fees. They hadn’t produced any of the major recent retail finance innovations (internet payments or tokenization) despite being innovation leaders in the late 20th century. Finally, because payment card + email = identity online and more than 17 millions Americans are excluded from bank services, banks are part of the reason segments of society are missing the cost and convenience benefits of the online revolution.
In part two I took a step back and pointed to two of the major drivers of this situation: banks aren’t earning as much as they used to because of low interest rates, and are facing a structural challenge to find a branch strategy that doesn’t lose them money. At the end I posed a question I often hear: “what would the bank of the future look like?” In this third part, I’ll address that question.
When discussing what the bank of the future looks like, people tend to whittle down the bank model in an effort to show how banks don’t need to exist. I think that this approach ignores the incredible creativity of someone with a profit motive, and the fact that services are bundled or unbundled all the time. It’s a common business cycle, and it certainly doesn’t mean that the bank business we know today has to disappear entirely. As such, it’s not an existential question, it’s more a question of just how much of banks’ current set of products will be unbundled.
With that in mind, I start by posing the question:
What do banks do for their customers that no other business can do?
The answer? They take your money, invest it, and have a government guarantee that you’ll get it back. Whatever money you give them (under $250k) is safe no matter what they do with it. Further, they don’t make these investments in government bonds, they make them in mortgages and other higher-risk instruments. Even further, these investments aren’t locked-in. You can withdraw whenever you like. So what does a future bank look like? At its core, it’s a riskless, indefinite duration investment manager, like a guaranteed money market fund, that specializes in mortgages. Let’s call them guaranteed investment managers. A portion of the interest they earn will go towards sourcing investment opportunities (making loans, as today), the rest gets passed on as “investment income” to the depositor. The greater that spread (the greater the net interest margin), the more services they can begin to offer. The more services they can offer, the more they begin to look like today’s bank. The net interest margin is a powerful thing. Whether these services are distributed through branches, I don’t know: it’s easy to envision this core ‘guaranteed investment management’ product not needing branches, but it’s equally easy to imagine hybrid models that would benefit from low-cost, kiosk-style branches. As I said, unbundling and bundling processes are dynamic, so it’s hard to know where on the spectrum you’ll end up.
Crucially, though, because their deposits are government-guaranteed, depositors in these new ‘guaranteed investment manager’ banks will need to meet all the identity checks that banks currently require, including a social security number and an address. These banks may also continue to avoid depositors with low account balances, no direct deposits, or flawed credit histories (yes, despite not offering them credit). In short, banks will likely offer their unique service to their current customer base, excluding the un- and under-banked.
If banks are now guaranteed investment managers, what happens to the other bank services I listed in the first part of this post? Here’s a reminder of what a bank does for me:
1. Keeps my money safe
2. Lets me send money electronically
3. Lets me write and deposit cheques
4. Provides me with cashier’s cheques
5. Lets me pay for things using a plastic card
6. Lets me pay for things using my phone
7. Lets me take out physical money at branches and ATMs
8. Lends to me
9. Invests my money in loans and treasuries and provides me with some interest income
10. Lets me manage my money online
So the ‘guaranteed investment manager’ entity handles parts of 1, 8, and 9, but not entirely. For example, it’s entirely possible to insure a deposit without using a government guarantee: even bitcoin processors like Xapo and Coinbase do this. So with insurance, deposits can be “kept safe” (1) for a marginal cost. Because these deposits aren’t being invested — they’re held as cash on the balance sheet — that insurance is likely to be inexpensive. The insurer isn’t covering investment risk.
Lending is also not exclusively a bank activity: lending platforms like Lending Club, specialty lenders, and, yes, banks, can all compete for the honor of giving you money. Finally, investing money — although not your “deposits” as this requires a bank license — can be done by asset managers, mutual funds, or, of course, yourself (stick to indexes!). The bank is reduced to one-of-many in each of these cases: potentially able to offer a less expensive product because of its government guarantees, but nonetheless a competitor among other providers.
What about all the other services?
Well, these services are ripe for unbundling (with one caveat which I’ll get to). Items (2) through (6) basically have to do with sending and receiving non-cash money. As Venmo and Paypal have already shown, these things can already be accomplished without relying on a bank (although it would be nice if these platforms insured the funds they hold on your behalf). ATM networks currently require a bank issuer, but deals with chain retailers (CVS, WalMart etc.) would allow for cash withdrawals and deposits in exchange for foot traffic. American Express’ original BlueBird card (issued outside of Amex’s bank subsidiaries) did exactly that, partnering with WalMart to provide cash in/out services. While non-bank membership in an ATM network may be a while off, customers could access cash without relying on a bank, which takes care of (7). Finally, many firms (Moven and Mint, to name but two) offer online money-management tools (10).
The one big caveat is cards. First, as I mentioned in Part 1, the card networks have never allowed a non-bank to issue cards. Whenever you see a non-bank issue a card, it is being issued by a partner bank (typically Metabank, GreenDot, or the Bancorp Bank). Why? The reason card networks work exclusively with banks is because all card transactions (credit, debit, or prepaid debit) involve credit risk.
Very quickly: when you swipe your card at a store, you may think that money is instantly moving from your account to the merchant’s. It’s not. It takes approximately two days for your money to reach the merchant, and during that period the merchant’s bank is effectively lending to the buyer’s bank until the funds settle. Card networks don’t want to increase the amount of credit risk in the network so the banks make guarantees to cover any losses in the network, and are in turn guaranteed by the FDIC.
Given this situation, it’s unlikely that non-banks will ever be able to issue credit cards (which is credit risk on top of credit risk). With real-time settlement, however, non-credit cards (e.g., prepaid and debit cards) don’t bring any credit risk. In fact, if I’m the card issuer, as long as I settle in real-time, there’s no credit risk involved: the merchant’s bank has its money instantly, whether it choses to deliver it to the merchant in real-time or after 3 days, I don’t care.
As a result, there is no structural reason that debit and prepaid debit card issuance should be a bank monopoly. In fact, when American Express launched its first iteration of the BlueBird prepaid card, it did so outside of its bank. Of course, being the card company in question, Amex didn’t need to ask itself for permission.. but the fact that BlueBird was issued outside of a bank shows that such a system can work.
What does the unbundled product / set of products look like?
At the end of the day, it’s possible to describe a system which, through a single product or through various products, provides:
- Safety for your funds
- Money transfers (including spending)
- Cash access (ATMs)
- Investment / lending
- Money management
Thereby replicating the services a bank provides today. By not being government-backed, however, this entity/set of entities will require only AML-KYC checks on its customers, not a full address, SSN, credit history, and the rest. This means that these services are suddenly available to anyone and everyone (same set of people using Western Union). What does this system look like?
As you can see, traditional banking is still there in the right corner, lending and providing their depositors with something akin to a government guaranteed money market fund. It is but one of the investment and borrowing services on offer — alongside traditional wealth management (mutual funds etc.) and peer-to-peer lending. Banks wouldn’t need to provide debit cards or checking accounts in their current form — they would plug into pure-play service providers who had the infrastructure to offer services without bearing the cost burden of branches. These services could be subscription based, charge a percentage of assets, sell customer data, or potentially benefit from the interchange fees paid by merchants accepting their debit cards. Sadly, nothing like this exists today — although Amex’s BlueBird came close, it was recreated inside of a bank because the government paychecks and welfare checks can only be sent to depository institutions.
What does this all mean?
Stepping outside the strictures of the regulated depository institution, these services quickly become much more flexible and diverse. Add in competition from traditional banks, and you end up with an interesting ecosystem. For the consumer, this means a wider variety of products and experiences all falling under the blanket of “banking” and, most importantly, for the un- or under-banked consumer, it means full access to the benefits of the online revolution.
/end.
If you’re curious, the ugly purple slides come courtesy of a deck I put together in 2014, available here.
