Thursday, December 31, 2020

The unbanked, the post office, and fintech in the 1880s

"A large population of people are excluded from the financial system because they don't have bank accounts. Fintechs compete to connect them and parallel plans emanate from the government to reach the unbanked, including postal banking."

What year am I describing in the above paragraph? 

It could be 2021. But it also describes 1870s. 

It's 2021 and the U.S. still has a large population of unbanked, those who have so little money that banks would rather not serve them. An astonishing 5.4% of Americansthat's 7.1 million householdsdo not have bank accounts.

Financial technology companies (aka fintechs) like PayPal and Facebook's Libra have well-meaning plans to connect the American unbanked population. Government-run proposals abound too. Postal banking is probably the most popular option, but more exotic solutions like central bank digital currency (CBDC) have also been floated. But many economists are wary that these government efforts will cripple the private sector.

None of this is new. Concerns over the unbanked, fintech, and a government participation in the payment system were all present back in England in the 1880s. Since I enjoy when the past resurfaces in the present, I'll tell the story.


Britain in the 1870s had a very sophisticated chequing system. Because banks were the only way for people to access cheques, and banks preferred to limit accounts to rich people and wealthy merchants, the poor and middle class were often left out. 

Luckily, the 1870s version of fintech came to the rescue. The PayPal of the day was something called the Cheque Bank. Established in 1873, the Cheque Banklike PayPal todaywas a bank-on-top-of-a- bank. What do I mean by this?

PayPal is a customer of Wells Fargo, a large commercial bank. Wells Fargo provides PayPal with banking and payments services. PayPal in turn passes these services on to PayPal account holders, folks who might not otherwise qualify as customers of Wells Fargo or, if they could, prefer the way PayPal rebundles underlying Wells Fargo services.

Stock certificate for the Cheque Bank, Limited

The Cheque Bank operated on the same principles. It opened accounts at bank branches all across United Kingdom and overseas. Like PayPal, it passed through underlying banking services to its unbanked customers. The Cheque Bank's main product was cheques, which today might seem quaint. But back then they were cutting edge.

Anyone could buy a book of Cheque Bank cheques at a stationer or cigar store, the Cheque Bank redepositing the cash it received with its bankers. The customer could then spend those cheques at stores, send them to family via the mail, or hold them as a form of saving in lieu of cash (which was always at risk of being stolen). People who accepted a Cheque Bank cheque as payment could promptly take the document to any bank and cash it.

Much like PayPal does today, the Cheque Bank held 100% reserves. That is, for every $1 in cheques it issued, it kept $1 locked up with its bankers. And so its cheques were considered to be as safe as cash. Put differently, regular banks engage in both lending and payments. But fintechs like PayPal and the Cheque Bank don't lend at all. They deposit all of their assets at an underlying bank and focus on offering the payments side of the banking business to their customers.

The Cheque Bank attracted the attention of William Stanley Jevons, one of the most important economists of the day and still very much a household name among economists today. Jevons was one of three economists (along with Carl Menger and Leon Walras) to discover the principle of marginal utility, a key economic principal which had eluded even Adam Smith. 

In his 1875 book Money and the Mechanism of Exchange, Jevons devotes a full chapter to the Cheque Bank, describing it as a "very ingenious attempt" to "extend the area of banking to the masses." Here is what one of the Cheque Bank's cheques looks like:

1899 cheque issued by the Cheque Bank [source]

The cheques could only be filled to an amount printed on the document, writes Jevons. So the above cheque, which had been purchased for £5, could be written out for anything up to £5, although in this particular case the cheque writer (H.L. Stevens) chose the sum of 3 pounds 3 shillings. 

Jevons isn't the only notable economist to write about the Cheque Bank. It also pops up over a hundred years later in economist Edward S. Prescott's work, who describes it as a "highly interesting experiment in extending the use of checks to the lower and middle classes." Prescott suggests that the ability to write a specific amount on the face of one of these cheques would have greatly facilitated payments through the postal service since there was no need for change. Unlike a regular cheque, which also offered this flexibility, the recipient of one of the Cheque Bank's cheques needn't worry about it bouncing.

Jevons was excited by the Cheque Bank. But he was not a fan of a subsequent competing payments innovation, the postal order.

The British Post Office, owned by the government, had long been engaged in the business of transmitting money orders, unofficially since 1792 and officially since 1838. A customer would walk into any money order office, put down, say, £2 and 2 shillings, and get a £2 2s money order. The recipient's name was then written on the order. It could then be sent via post to a distant office, upon which the recipient could take the money order to the counter to be cashed. The officer would first confirm the payment by referring to a separate letter of advice. This letter, sent from post office to post office, served an an extra layer of security against fraud. Only then would the £2 and 2 shillings be paid out.

The problem, according to then Postmaster General Henry Fawcett, is that the money order wasn't very useful to people who only wanted to send small amounts. "If a boy wanted to send his mother the first shilling he had saved, he would have to pay twopence for the order and a penny for postage," wrote Fawcett. In other words, to send a 12 penny (i.e. one shilling) money order, three penniesa massive 25%had to be sacrificed in fees. (A shilling in 1880 was worth around US$8 today.) And so it would have been an expensive payments option for the poor.

Prior to his appointment as Postmaster General in 1880, Fawcett had been both parliamentarian and the first professor of political economy at Cambridge. And while he wasn't as illustrious an economist as Jevons (he hasn't left us any bits of economic theory), Fawcett did write what was one of the popular textbooks of the day.

But if Fawcett wasn't going to change the study of economics, he did intend to change the payments system. As Postmaster General, Fawcett proposed complementing the money order with a new product called a postal note, or postal order. (The postal order had been earlier conceived of by George Chetwynd, the Receiver and Accountant General of the Post office). Like the cheques issued by the Cheque Bank founded just seven years before, postal orders would have a fixed denomination printed on them. These increments were to start at 1 shilling and go up to 20 shillings (US$8 to US$160 in 2019 dollars).

By contrast the post office's traditional payment product, the money order, was open-faced and had no denomination. Because postal orders would be issued in smaller amounts, the Post Office needn't bother sending separate letters of advice as a security measure, which meant that they would be far cheaper to process. And so the fees could be lower for postal orders than money orders, broadening the pool of customers.

In an 1880 essay, William Stanley Jevons blasted the idea of postal orders, which hadn't yet received legislative assent. Singling out Fawcett, Jevons wrote:

"The fact of course is that not only from the time of Adam Smith, but from a much earlier date, it has always been recognized that a Government is not really a suitable body to enter upon the business of banking. It is with regret that we must see in this year 1880 the names of so great a financier as Mr. Gladstone, and so sound an economist as Professor Fawcett, given to schemes which are radically vicious and opposed to the teachings of economic science and economic experience."
So that lays out the cast of characters in 1880. It includes exclusionary banks, hoards of unbanked, a set of opposed economists in Jevons and Fawcett, fintechs like the Cheque Bank, and a post office on the verge of issuing a novel product; postal orders.

2020 seems very much like 1880. To help connect the large population of American unbanked to the financial system, a number of modern day Fawcetts (Morgan Ricks, Mehrsa Baradaran, Rohan Grey) have floated public payments solutions including a return of postal banking, central bank digital currency (CBDC), or central bank-accounts-for-all.

Our modern day equivalents to the Cheque Bank includes non-banks such as prepaid debit card issuers Walmart and Netspend, both of which are trying to reach unbanked Americans. Online wallet companies like PayPal and Chime are also in the mix. And stablecoin issuers such as Facebook's upcoming Libra project talk a big game when it comes to financial inclusion. To round things out you've got your modern day Jevonses; economists who don't buy the idea that the government should get into banking (Larry White, George Selgin, Diego Zuluaga).

So how did things end up in 1880? Despite opposition from Jevons and the Economist, Fawcett's postal order dream came to fruition. After receiving legislative approval, the world's first postal orders were issued in 1881:

Postal orders would go on to become very popular. They largely displaced money orders, except for large amounts. Other postal systems including that of New Zealand, Canada, Australia, and the U.S. would go on to copy the idea. The UK's modern day incarnation of the post, the Post Office, still offers a version of the product.

And what about the Cheque Bank? Digging through old documents, Edward Prescott discovered that the Cheque Bank failed in the late 1890s. According to liquidation proceedings reported in the Banker’s Magazine, it was plagued by forgery problems and increased competition for less wealthy depositors from banks. Perhaps the emergence of the postal order also played a part.

I'm not invoking the 1880s as a prediction of what will occur in the 2020s. Rather, it fascinates me because it reveals how old these payments dilemmas are. The same tensions between public and private payments were present then as they are now. And it's also interesting to see how economists have always been engaged in questions of financial inclusion. Not just Fawcett but Jevons too, who we know primarily for his work on monetary theory. 

And over a hundred years later, Edward Prescott delved into the topic, too. In a 1999 paper (which mentions the Cheque Bank), Prescott discusses the idea of opening up an inexpensive type of bank account called an Electronic Transfer Account (ETA) so that all Americans, particularly the unbanked, might receive Federal benefit payments digitally. (Prescott was skeptical that ETAs might work out. The program, introduced in 1999, was discontinued in 2018 and has been replaced with a prepaid debit card program.)

In closing, the topic of how to help the unbanked is a complicated one with many moving parts. Which is why we should explore how things played out in different times. Perhaps history can get us to see the debate in a new light.

Merry Christmas and a Happy New Year!

P.S. If you're interested in learning more about Jevons's thinking on payments, he was a big champion of the idea of creating an international coin standard. I wrote about it here. Think of it as a proto-version of the Euro. Jevons came up with a "tidy English solution" for fitting Britain into this proposed international coin union. The project never came to fruition.

Thursday, December 24, 2020

Dolphin Safe Tuna and Fair Trade bank accounts?

The Royal Bank, Canada's largest bank, says that it won’t lend to clients that get more than 60% of their revenue from thermal coal or coal-fired power generation. Should a bank be able to avoid providing services to businesses just because they don't engage in the sorts of activities the bank, or its depositors, approve of? 

Put differently, should the Royal Bank be able to avoid "dirty" loans so that it can offer its depositors the semblance a Fair Trade, or green, bank account?

Critics would say that the Royal Bank shouldn't be allowed to avoid banking coal-fire dependent companies because it operates within a “regime of privilege.” That is, the Royal Bank benefits from a system of government regulation, central bank lender of last resort benefits, Federal deposit insurance, and direct access to core public payments systems. Given how deeply it is fused with public infrastructure, the Royal Bank is sort of like a utility, and like any utility it has a public duty to consider all customers, even coal energy guzzlers.

A related argument is that because banks must obtain a charter in order to operate, and this is difficult, there are not enough banks competing with each other. And thus it would be unfair for Royal Bank to avoid doing business with coal-fired power generators; financial banishment could doom these businesses to failure.

I draw the above arguments from a recent paper by Brian Knight and Trace Mitchell. Hopefully I have accurately captured their views. Given their specialness, banks should not be permitted to act as "de facto regulators," say the authors.

Now for the counterargument. People should be free to enter whatever contracts they see fit, including avoiding ones they find distasteful or against their beliefs. This freedom shouldn't be available to some people but not others. For instance, say that Sarah and Tom are both worried about global warming and sustainability. Tom is a skilled cook and decides to set up a sustainable restaurant that serves only ethically sourced ingredients. Sarah, for her part, is trained in finance and wants to set up a sustainable bank. Her source material for creating 100% green bank deposits is ethical loans to green businesses.

A law disallowing banks from choosing their customers means that Tom can self-actuate his beliefs by only dealing with green food suppliers, but Sarah cannot do the same by only lending to green borrowers. That hardly seem fair.

So there are two conflicting ideals at play here: the right to receive core services vs the freedom of association.

In the U.S., the Office of the Comptroller of the Currency (OCC), a key U.S. bank regulator, is choosing a side in this conflict. In an effort to stop banks from “politically driven discrimination,” the OCC is proposing a rule that would prevent bankers from using anything other than regular credit and operational criteria for evaluating a company seeking financial services. Were the OCC's "no discrimination rule" to be applied in Canada, it would require Royal Bank to lend to companies hooked on coal-fire energy.

In proposing this rule, the OCC has adopted the same rational as Knight & Mitchell. In an op-ed for the Wall Street Journal, OCC head Brian Brooks and Chief Economist Charles Calomiris argue that government chartering and direct access to the Federal Reserve obligate banks to provide services to all companies.

I recently wrote about this "no discrimination" rule for Coindesk. In that article I took the pro-Royal Bank side, arguing in favour of fair trade bank accounts. The 21st century consumer wants to know more about the provenance of the things they buy. We don’t just want tuna, we want dolphin-friendly tuna. We don’t want our T-shirts to be made in sweatshops with Xinjiang-grown cotton. We want ethical T-shirts. So why shouldn't we get clean bank accounts?

I want to explore this tension a bit more.

If we are going to apply a no discrimination rule to any segment of the banking and payments market, I think it should be placed on the card networks, Visa and MasterCard. The card networks have the power to exercise far more de facto regulation over the economy than any bank. If a bank disconnects a business, that'll certainly a hassle for the debanked business. But at least there are dozens of other banks in Canada to turn to, and thousands in the U.S. Even if no bank is willing to step forward, there is a whole host of non-bank financial institutions that can provide a business with financing or payments services.

Not so if the two card networks disconnect a retailer. Since there is no good alternative to Visa and MasterCard (especially online), a banned business could be in very real jeopardy. For instance, the card companies currently allow gun and porn purchases across their networks. But were they to ban gun and porn sales because they deem them unsavoury, Visa and MasterCard would be doing incredible damage to both industries, far more than if two large banks were to cease providing services to gun retailers or porn sites.

For a demonstration of this power, look at Pornhub's recent reaction to the threat of being deplatformed by Visa and MasterCard. After being accused of hosting child porn, Pornhub completely redesigned its platform to try and keep the two card networks on side (it failed.) 

By the way, I wrote about this incident for the Sound Money Project. Porn is legal, but child porn is illegal. Any financial institution that knowingly allows illegal transactions to cross its platform could be accused of money laundering. So Pornhub's deplatforming wasn't a case of the card networks acting as de facto regulators of content. Rather, they were doing what the actual regulators, i.e. the law, dictate. (That doesn't mean we shouldn't be worried that card networks can engage in de facto regulation. It just means that in this case, they didn't exercise that power).

So to reiterate, card networks have more power than banks. But unlike banks, card networks don't operate within a “regime of privilege” as described by Knight & Mitchell or in Brooks & Calomiris's op-ed. They don't have lender of last resort benefits, Federal deposit insurance, or direct access to core public payments systems. Nor do they have to get to get a bank charter. 

Visa and MasterCard are powerful because they are networks. Once everyone is connected to a network, there is very little reason for any one to leave to a competing network since only the incumbent can offer a large number of connections. (A bank is not a network, it is a member of a network.) 

And so Visa and MasterCard evade—unjustifiably so, in my opinion—all of the criteria for being targeted by the OCC's no discrimination rule. Instead, it is less powerful banks that would be handicapped by it on the basis of their proximity to government infrastructure and their obligation to get a charter.

Which gets me to my final point. The OCC and Knight & Mitchell have proposed that the criteria for triggering a no discrimination rule should be the existence of a "regime of privilege" and chartering. But doesn't a wide swath of the economy operate within a "regime of privilege" and chartering? 

A restaurant, for instance, must get a restaurant license before opening its doors. It also needs to secure building, ventilation, and signage permits. It is encumbered by zoning requirements and needs to secure a license to sell alcohol. Restaurants benefit from a government-funded system of food inspection. The ingredients that a restaurateur purchases has passed through some sort of a food safety regulatory process.

In sum, I do agree that we may need some sort of no discrimination rule for financial institutions. I'm just not sure that the OCC and Knight & Mitchell have found the right criteria for applying this rule. Let's choose whatever criteria get us to a situation that the card network Visa and MasterCard are the prime candidates for a "no discrimination" rule, not banks (or restaurants).

Sunday, December 6, 2020

Judy Shelton at the Bank of Canada? No thanks

How would I feel if Judy Shelton was a candidate for Governor of the Bank of Canada? Here are my thoughts.

A bit of background first. Judy Shelton was a Trump appointee to a key spot on the Federal Reserve board, the U.S.'s central bank. A President's appointees must be confirmed by Congress, and this was probably the most heated confirmation process I've ever followed. Shelton has espoused several controversial view points, including a return to the gold standard

The reason this appointment is so important is because Federal Reserve board members determine American monetary policy. That is, they decide whether to pluck interest rates up or down in order to ensure that the central bank is hitting its mandated targets.

That's a pretty important job! Not only would Shelton have been in the monetary policy hot-seat, she would have been on track to become the next head of the Federal Reserve. But it was all for naught. Shelton narrowly lost the spot as several Republican senators, including Mitt Romney, dissented.

So Shelton for Bank of Canada? Here's a simple set of guidelines I'd suggest Canadian voters adopt when they consider what sorts of people should be at the helm of the Bank of Canada. 

First, I'd suggest that anyone being considered for the Canadian monetary policy cockpit have a PhD in economics, preferably one in macroeconomics. And second, they should have administrative experience. (I'd be willing to accept a master's degree in economics as a substitute, with sufficient time spent working up the rungs of a central bank. The administrative experience is important because candidates will be in a management position.) Once they've passed those two hurdles, the finer points of their candidacy can be discussed.

A quick scan of Shelton's background reveals that she has a PhD in an unrelated discipline and no prior central banking experience. So if she was being floated for the job of Bank of Canada governor she wouldn't have passed through my filter. (Current Fed governor Jerome Powell, who is a lawyer, wouldn't have got through either.) 

Some readers may think I'm only stating the obvious. "Of course the most important people at a nation's central bank should have high-level economics degrees." But other readers will be disappointed in my criteria. I mean, here I am, an independent blogger—one who doesn't have a graduate degree—advocating an elitist sclerotic filtering mechanism.

Note that I'm not criticizing Shelton for her controversial stance on the gold standard. If she was a trained economist whose gold standard views had survived through years of rigorous training, she'd pass through my basic filter. Nor am I criticizing Shelton because she was a Trump appointee. Christopher Waller, another recent Trump appointee, would have easily passed through my filter.

Here's why I think my filter makes sense for Canadian voters. Monetary policy is complicated. Luckily we have an institution that teaches it: economics departments. Once someone has attained a PhD (or Masters + experience) in macroeconomics, odds are they'll be better than most at understanding how to operate the levers of a central bank. 

Why not draw central banking talent from other venues like the media, activism, think tanks, law, finance, business, or the blogosphere (ahem)? These venues don't attack the problems of central banking in as disciplined a manner as an economics department does. And so the average quality of these talent pools will not be as high.

In particular, I want to comment on the idea of drawing central bankers from the business/finance community. Many voters may think that Canadian business personalities like Kevin O'Leary, investor Prem Watsa, or banker David McKay would be uniquely qualified to run the Bank of Canada. I disagree. Sure, these titans of business will have good administrative experience (although no better than anyone else at the top of their field). 

But running monetary policy has little in common with running a business. Like the litre, second, and meter, the Canadian dollar is one of Canada's most important weights & measures. We put well-trained physicists at the National Research Council (NRC) in charge of maintaining our key physical measurements, not business people. (In fact, NRC scientists recently guided us onto a new standard for the kilogram.) Likewise, we need people with proper scientific training to manage our key economic measuring unit, the $.

Will my filtering system leave out a lot of good candidates? Yes. Will it bring in some bad ones? Certainly. Economics departments have tons of problems. Trust me, I've heard stories from insiders.

But even if it's not a great filtering system, it's still the best filtering systems we've got. Imagine that the plane you're on has lost its pilot in mid-air and needs a replacement. If one of the passengers has been to flight school, that person is probably going to be the best pick for flying the plane.

The WSJ marketed Shelton's candidacy on the basis of diversity. Yes, diversity of opinion is important. We want the co-pilot on our plane to criticize the pilot when one sees him/her make a mistake. But we still want both to be trained pilots with solid skills. They need to know exactly what that little red button above and to the left of their heads does when pressed. We wouldn't put a cake decorator, a plumber, and psychologist in the cockpit, just for the sake of diversity.

It's interesting to contrast U.S. and Canadian central banks on my very simple filter. Going back to 1970 the U.S. has missed twice: Jerome Powell and William Miller. The hits include Janet Yellen, Ben Bernanke, Alan Greenspan, Arthur Burns, and Paul Volcker.

Canada hasn't missed once. Everyone from Louis Rasminsky and James Coyne to Mark Carney, Stephen Poloz, and Tiff Macklem qualify.