Over the next several weeks, we’ll take a tour of the payment methods outlined in Part 1, and go into their history, which helps us understand their current position in the payments landscape.

Credit cards have a particularly rich and dramatic history, as I suggested in the following tweet:

Beginning with Diner’s Club in 1950, credit cards started as a way for businessmen to pay for lodging and meals without having to carry a lot of cash.  Credit cards also offered an alternative to store credit, which required hotels and restaurants to keep accounts of who owed them money, and press customers to settle up.  They caught on because they solved problems for customers and businesses.

Flash forward to 2020, and the situation could not be more different.  A cold war exists between credit card issuers (banks) and merchants, occasionally breaking out into open hostilities.  Both sides distrust each other, and see the market as a zero-sum game.  Merchants now see Visa and Mastercard as a duopoly that siphons their profits in exchange for very little.  Banks, for their part, see merchants as ungrateful complainers who shift blame for their own problems onto them.  Active litigation continues.

How did we get here?  And how do we achieve a détente, if not comity?  I will touch on a few key milestones that changed the course of bank-merchant relations since Diners’ Club was founded.

From Competition to Co-opetition

Following the launch of Diners Club, different consortia of banks launched what would become Mastercard and Visa.  These networks were more open, in that multiple banks could issue cards under the network brand, and multiple banks could sponsor merchants on the network.  Originally, the two bankcard networks were quite competitive, demanding that their member banks issue cards only with them, and not with the other network.  However a lawsuit by Worthen Bank and Trust Company in Arkansas led to a Justice Department decision not to defend the provisions against antitrust claims.  The result was that eventually most banks issued cards from both networks (duality), and Mastercard and Visa evolved into a state of “co-opetition.”  With major banks on both boards, there was an incentive not to compete too hard, the opposite of what the Justice Department intended.

Co-opetition also led to new accusations of price-fixing, since the same banks were influencing pricing on both networks.  As noted earlier, duality reduced the motivation to compete, except on service and features.  Pricing converged between the two networks.  In some ways, this could be seen as positive, since it created a more stable pricing environment for merchants and banks.  In others, it could be seen as artificially inflating prices, since one network no longer saw much benefit in undercutting the other.

There were attempts to launch true rivals, but until the Discover Card launched in 1985, no company succeeded in creating a competitor, and none have since.

I and many others call this “the network effect,” and it is a crucial concept for understanding how payment systems work.

Understanding the Network Effect

Payments is what David Evans and Richard Schmalensee, in their book Paying with Plastic call a “multi-sided market,” in which both the buyer and seller must affirmatively choose to participate.  I can create my own credit card network, but I will have to persuade both buyers and sellers to accept it.  The number of buyers and sellers on the network determines the number of connections, as shown in Figure 1.

Figure 1. Increasing Nodes Leads to Exponential Growth.

The number of connections increases by the square of the nodes.

Source: www.payments-101.com

As you can see, adding just one node to each side increases the number of connections exponentially; that is, by a square.  In general, we can state that, for a two-sided market where the number of nodes on each side is equal, c=n2, where c is the number of connections and n is the number of nodes.  In real life, of course, the number of nodes on each side is not equal, but we use this simplified example for clarity.

Each connection has value, which is the amount of commerce conducted over that connection.  If a single person and a single merchant do $100 in trade together, then the value of a connection is $100.  Therefore, the value of the network grows as the square of the value of each connection.  In the first example, the value would be $900, while in the last it would be $2,500.  With credit cards, you have leverage due to the lending feature, so the value increases even more.

This is the reason why so few new payment methods succeed, and while existing ones live on long after they have been rendered technologically obsolete.  Discover Card was unique because it was able to build on Sears’ gigantic store card portfolio.

The Network Effect Leads to Anticompetitive Behavior

By the 1990’s, it was clear to most observers that there were not going to be any more competitors to the “big four.”  Furthermore, Visa and Mastercard had been so successful at getting consumers to sign up that acceptance was no longer a real choice for merchants.

Cash Back and the Rewards Card

One of the biggest reasons for the networks’ success in gaining cardholders was the invention of “cash back” by Discover.  Effectively a rebate on all purchases, cash back transformed the credit card from a way of delaying payment to a way of saving money, and affluent consumers realized that they could generate substantial rewards by funneling all purchases, not just big ones, through their cards.  While there has been a lot of innovation in the rewards space, consumer surveys I conducted while at IDC and Mercator found that consumers overwhelmingly prefer cash back, and this finding has been consistent for at least the last 20 years.  There are now sites dedicated to finding the best rewards cards.

Fewer Consumers Now Pay Anything at All for a Card

As banks continued to compete for a dwindling number of uncarded consumers, they found that reducing or eliminating annual fees was an effective way to attract consumers who had less spending capacity.  If credit cards had now become gigantic collective bargaining networks for consumers, like discount cards, a prospective cardholder had to weigh the cost of owning a card with the likely benefits.  If a consumer spent $1000 a year on their card, they might see 2%, or $20, in rewards.  An annual fee of $60 or more would not make sense in these circumstances.  However, once the annual fee was removed, $10 in rewards might be attractive.  Encouraging consumers to use their cards for everyday expenses like groceries by raising the cash back rate for these purchases got the overall spend up, thereby increasing the rewards and bringing more consumers on to the platform.

The Elimination of Annual Fees Unbalances the Pricing Structure

Once consumers no longer had to pay an annual fee, the old pricing structure became unbalanced.  Now the merchants were the only ones paying for the use of the network, through interchange.  The figure below shows how interchange results in a lower payment for the merchant through a series of fees:

How fees are deducted from card payments

After the consumer submits their card, the issuer, network, and acquirer all take a cut, so that the merchant is left with the remainder.  Sources: “2017 Interchange Fee Revenue, Covered Issuer Costs, and Covered Issuer and Merchant Fraud Losses Related to Debit Card Transactions,” Board of Governors of the Federal Reserve System, 2019, and “Payments Systems in the U.S., Third Edition,” Benson, Loftesness, Jones, Glenbrook Partners, 2017.

Interchange is a way of pricing that involves the seller’s bank paying a fee to the buyer’s bank.  (This fee is passed through to the seller, plus a markup, which is why most people consider merchants to be the ones who pay interchange fees).  It was originally intended to balance costs and benefits among network participants.  If one side has less motivation to adopt the payment method, you can make it cheaper for them by charging more to the side that has more of a motivation.  Originally Diners Club balanced out an average 7% fee for merchants with a $18 annual fee for cardholders ($194 in today’s dollars).  When merchant acceptance proved harder to achieve than consumer acceptance, Diners Club adjusted the pricing structure so that cardholders paid $26 (or $281 in today’s dollars), while the merchant fee was left at an average of 7%.  American Express still largely competes on this basis today; its top-tier mass market card, the Platinum, charges a $550 annual fee. However, it is an exception, and even American Express has experimented with eliminating annual fees for some of its cards.

Once card issuers began waiving annual fees for cardholders without changing the pricing for merchants, the merchants became an even bigger source of revenue for the banks.  With the merchants locked in due to the network effects shown above, the networks began raising their interchange rates regularly. This enabled more generous rewards, which attracted more spending, forming a growth cycle.

The year 2002 was a turning point for the industry, when the Reserve Bank of Australia decided to impose caps on interchange fees.  In Part 3, we will see how this event, more than any other, brought us to the current situation.

(Note: this article was revised on November 17, 2020 to add a flow chart for interchange, and to clarify that it was an Australian regulator, not a U.S. court, that changed the way interchange was viewed around the globe).