Interchange and Value Redistribution: Clarity in the Noise
A recent Harvard Business School working paper has drawn pushback in the payments industry. Drawing on Fiserv settlement data covering roughly one-fifth of U.S. card volume, it estimates that card rewards programs redistribute about $30 billion a year from (generally) lower-income cash and debit users to higher-income rewards cardholders. Critics challenge the methodology, the dollar figure, and the policy conclusions. Some challenges are valid. Most miss the point.
How Card Payment Acceptance Is Priced
When a consumer pays with a card, the cost to the merchant has three parts: interchange, paid to the issuing bank; network fees, paid to Visa, Mastercard, or the relevant scheme; and processor fees, paid to the acquirer. The total is the “merchant discount rate” (MDR).
Interchange is the question here: a premium card might carry interchange of 2% or more; a Durbin-regulated debit card is capped at roughly 21 cents plus five basis points. Issuing banks depend on interchange to fund their rewards programs; when Durbin capped interchange for large banks, debit card rewards largely disappeared.
What the System Actually Does
Issuing banks receive 70–90% of the MDR and use much of it for rewards: industry data suggests 40 to 70 cents of every interchange dollar fund rewards. On a 2% premium card, the rewards component alone may run 1 to 1.5 percentage points.
The trips you pay for with rewards points are not a gift from the bank; they come from increased prices everyone pays.
Interchange reaches consumers invisibly and uniformly. Unlike a surcharge, which loudly announces itself at the point of sale, interchange is silent: virtually no cardholder knows what it is.
The vast majority of U.S. consumer spending occurs at merchants that charge the same price regardless of payment method. The prices they all pay include the cost of rewards; those rewards only benefit a portion of consumers; if you pay with cash or a debit card, you don’t get rewards
One group of consumers is paying for benefits that only some consumers get: they are subsidizing other consumers—a transfer of wealth.
Hence the arithmetic. If the MDR is embedded in prices everyone pays, and only rewards cardholders receive value back, consumers who don’t earn rewards contribute to issuing bank revenue but receive nothing in return. The Harvard paper frames this as a $30 billion transfer; the precision is contestable, but the direction is right.
What the Critics Get Right — and Where They Overreach
Some comments about the Harvard Studay are valid. Rewards points are hard to value precisely: redemption rates vary, and breakage is real. These points affect the precision of the estimated value transfer, not the structural reality.
Another objection concerns the 2010 Durbin Amendment, which capped debit interchange fees for large issuers. Prices kept rising afterward, due to many factors, of which MDR is just one. What Durbin shows is that cutting interchange doesn’t reliably produce lower consumer prices. But it does not show that interchange was never in the price.
A third argument is that the reward card holders spend in areas such as luxury travel, fine dining, and high-end retail, where lower-income, non-reward card consumers are largely absent. The Harvard paper grants this, knocking its value transfer estimate down by roughly a third.
But the logic doesn’t extend to grocery, gas, pharmacy, and general retail, where all income groups shop together and rewards cards are heavily used. A significant cross-subsidy in shared categories survives intact.
What the Debate Is Really About
The cross-subsidy is real. Premium cards have widened it: higher interchange funds richer rewards, while checkout prices stay flat.
The stronger version of their argument is not that the cross-subsidy doesn’t exist — it’s that capping interchange won’t fix it.
Durbin is instructive: some, uncalculable portion of savings went to merchants rather than consumers, and there is little evidence that consumers got measurable benefit. But reduced consumer prices is not the only argument for reducing interchange rates.
Honest framing starts with what the data shows: card payment economics create a cross-subsidy, largely invisible to consumers, flowing from those who don’t earn rewards to those who do. How policymakers address it — if at all — is a legitimate question. What’s harder to defend is dismissing the structural reality the Harvard paper describes, even if it’s imprecisely measured.