Banks are being bamboozled by the idea that the solution to most of their problems lies in becoming more like the fintech upstarts currently threatening their turf.
This is strange because if anyone should know the degree to which fintech’s cost advantage is linked to unadulterated regulatory arbitrage, the exploitation of banks’ own subsidy systems against them, cartel-based solutions or complete disregard for par value protection of the payment float, it should be banks.
There are many facets to this story. In this post, however, we’re going to explore how and why fintech — in its rush to disrupt finance — has overlooked the critical role banks play in protecting the par value of the monetary float (usually with collateral), and why it’s not that easy or even advisable to provide payment services without the support of such a scheme.
Monetary float, of course, is tantamount to the amount of reserves the system needs to keep on hand to protect it from payment reconciliation time lags. It is the bit of the money supply, in other words, which is routinely double-counted because it can’t square up quickly enough and which — in a prudent system — needs to be backed to some degree by highly liquid securities. If double counting creates an accidental run on resources, these securities can then be liquidated to meet the shortfall. In many jurisdictions, as a consequence, the monetary float tends to equate with the reserve requirement: the idle (opportunity costing) buffer which for risk reasons cannot afford to be reinvested outside of the core system.
In a system without a reserve requirement, liquid securities circulating in the repo markets and squaring off against each other can be judged an unofficial proxy for the monetary float. As a general rule, in any functioning payments or banking system, look hard enough and you find a float eventually.
At some point in banking history, however, the act of providing payment services to customers became unbundled from the act of banking, popularising the idea that payments operators should not be regulated or controlled the same way as banks. Payments was a technology. Banking was about risk-control, customer screening, investment and of course reserve/float management. Besides, payment companies were entirely dependent on banks anyway. Right?
This is strange because hitherto payments had always been inextricably linked and connected to banking, banking networks and float management. (See, for example, the history of the Medicis).
So how and why did banks get comfortable with outsourcing critical payment services to non-banks?
Looking back it’s tempting to pin the great division on the mass introduction of credit cards in the 60s. But that would be wrong.
Credit cards do and always have depended on the underwriting role of banks as well as the interbank cooperative guarantees which allow these systems to interoperate smoothly. That means any unreconciled floats which do emerge in the course of daily business square back to the central bank ledger, and must be funded accordingly if banks are to avoid going bust.
The credit card corporations thus only represent the common ground or “glue” which brings the banks together to form the necessary standards, data-sharing protocols, message systems and cross-insurance schemes which allow individually-issued credits to be treated as equivalents in as wide a squaring network as possible.
The cost of running the cartel infrastructure, meanwhile, is funded by fees charged to the credit beneficiaries (by way of the banking relationships) or on a for-profit basis through fees charged to merchants (by way of the representative credit card company). Since banks are also the stakeholders in the credit card companies, however, any profits generated flow straight back into the pockets of banks. The credit card companies are simply profit-sharing structures devised at arm’s length to allow competitive institutions to benefit from harmonised approaches.
Credit cards accordingly don’t and never did exist separate to banks.
And while yes, sometimes credit cards were backed by just a single bank (i.e. Bank of America in the early days of Visa) most evolved into franchise networks and/or jointly controlled consortia or alliances, opening the door to the core Visa/Mastercard duopoly structure we have today.
So perhaps then we should blame the split on the emergence of charge cards like Diners Club and American Express*, which pre-empted the mass roll-out of bank-sponsored credit cards by just a few years?
Unlike credit cards, charge cards operate as common intermediaries between specific networks of merchants and customers.
Unreconciled lags within their “closed loop” systems (a.k.a the period of double-counting payments) don’t square back to the central bank system but must be funded some other way.
Take Diners Club as an example. In its initial incarnation, the card took advantage of merchants’ tendency to extend invoice-based credit to known, dependable customers. By gaining the trust of the merchants in this way, Diners Club was able to benefit from merchant credit which it was able to pass on to to the clients it was prepared to vouch for.
Diner Clubs’ clients saw value in this arrangement because it allowed them to aggregate and settle bills on a monthly basis with just one counterparty (rather than having to settle at every exchange point). Merchants, meanwhile, saw value in being able to chase down just one counterparty over missed payments at the same time as benefiting from additional ‘impulse’ business with customers they had no prior relationship with.
The scheme’s only limitation: customers were required to settle invoices in full every month and cards were only accepted with merchants signed up to the card network.
If customers missed payments the charge card companies’ credit relationship with the merchants would be directly effected, forcing them to raise financing from more expensive sources (and threatening the profitability of the model). Hence the exclusivity of the cards. The merchant credit was the lifeline of the system and it was with the implicit merchant subsidies that the power really lied.
The charge card business model in that sense was not dissimilar to PayPal‘s. Albeit there were two notable exceptions: inter-company credit agreements powered the business model, not a prepaid float; the beneficiaries of the credit arrangements were the consumers not the merchants.
Not banks, money transmitters
Going back to the big divide between banking and payment processing, it’s easy to see why the charge card purveyors thought they had good reason not to call themselves banks: they didn’t take client deposits or manage cash-floats, they didn’t originate credit and they didn’t deal in funding markets directly. When it came time to settle, they still depended on the banking network.
What they did do was vet and screen customers on merchants’ behalf and extend merchant credit to those customers at their own risk. If bills went unpaid, it was the charge card companies who were on the hook for missed payments at a cost to their own profitability and reputation.
If they failed, they failed: the losses would be by the merchants’, thus contained and removed from the banking system.
The economist Nicholas Kaldor interestingly observed in 1970 that the emergence of such closed loop systems lent credence to the endogenous theory of money, arguing that if and when official banking credit channels seized up it was clear these parallel networks could be depended for network specific money creation:
Any business with a high reputation — a well-known firm which is universally trusted — could issue such paper, and any one who could individually be “trusted” would get things on “credit”. People who can be “trusted” are, of course, the same as those who have “credit” — the original meaning of “credit” was simply “trust”. There would be a rush to join the Diners Club, and everyone who could be “trusted” to be given a card would still be able to buy as much as he desired.
This line of thinking, however, also led Kaldor to foresee the rise of a different type of payment system, one where those of us who couldn’t be trusted with credit could be paid via a system of chits, which would be issued in lieu of cash by, say, the top five hundred businesses in the country. As he predicted:
And these five hundred firms would soon find it convenient to set up a clearing system of their own, by investing in some giant computer which would at regular intervals net out all mutual claims and liabilities. It would also be necessary for the member firms of this clearing system to accord mutual “swops” or credit facilities to each other, to take care of net credit or debit balances after each clearing. When this is also agreed on, a complete surrogate money-system and payments-system would be established, which would exist side by side with “official money”.
A nice prediction of the parallel command economy states of Amazon/Ebay/Alibaba and the gift voucher, money-market-fund and payment/credit affiliates that have come to be closely associated with most of them (as they attempt to become their own banks, payments providers and float managers).
But a system funded by merchant credits is quid pro quo a system funded by a merchant’s capacity to suffer settlement time-lags, which itself is proportional to the merchant’s ability to draw down on inventory before having to seek out new financing for the replacement of his stock. It is the merchant’s float being double tapped instead of a bank’s.
If a merchant over-extends himself on this front, he is faced with only three options: replenish his stock at an additional financing cost to himself; via the credit of a friendly supplier or operate with a smaller amount of stock until payments are made, missing out on potential business opportunities.
Everything does, as a consequence, square back to banking and the core monetary float, suggesting such schemes weren’t ever as detached from banking as first thought.
The pattern continues elsewhere. Unpick the structure of non-bank wire services, traveller’s cheques providers or any other significant payments “innovation” — and as per the above, you usually end up finding your way right back to the core monetary float of the banking network.
The prepaid or zero-float exceptions
These days, however, a new trend is emerging: the rise of non-bank payment intermediaries convinced they can operate independent of the core financial system; without drawing on monetary float; or by managing a cash float more effectively than the banking sector.
For example, there are those, like PayPal or Alipay, who draw on pre-paid floats from customers to help them manage temporal discrepancies in their payment systems. When interest rates are high, the float’s investment in the banking network generates earnings which can be passed on to customers as transaction cost subsidies (increasing their competitiveness vis-a-vis the banks). When interest rates fall or when banks start charging for liquid access to cash float on less accommodative terms, many are tempted to manage their cash floats directly by way of the securities markets. They become money market funds. This works fine until a shortage of safe assets threatens their ability to defend the par value of the float. Which is when they turn to originating assets and securities directly, taking on the sort of lending risk that’s usually associated with banks.
It’s worth pointing out that PayPal has in recent years become a bank in some jurisdictions, hinting float management outside of the core banking sector may not be that easy after all.
And then there are those who think they don’t need a prepaid, collateralised or credit-supported float to manage settlements discrepancies at all. Rather that virtual “bridging currencies” or blockchains can settle transactions so precisely and perfectly that the float itself becomes unnecessary or transforms into the fixed amount of uncollateralised, units outstanding.
These entities are now influencing central bank policy. This is important because protecting the par value of the float isn’t easy, and there are major implications if — after decades of convincing everyone that floats should be as small as possible for capital to be deployed efficiently — central banks regress to large-scale float systems that amount to a return to full-reserve banking.
But more on that in our next post.
*American Express traces its origins to 1850 and the act of physically transporting money from point to point geographically before pioneering the traveler’s cheque in 1891 (an echo of what the Medicis were doing during in the Renaissance era). But it wasn’t until the 60s that the company properly pivoted into the charge card provision business we know it for today.
Related links:
RTGS, and the story of collateralised risk instead of credit risk – FT Alphaville
RTGS, and the story of batches instead of blocks – FT Alphaville
The Romans always copy the Greeks (including the repo market) – FT Alphaville
RTGS, and the story of batches instead of blocks – FT Alphaville
Copyright The Financial Times Limited . All rights reserved. Please don't copy articles from FT.com and redistribute by email or post to the web.