When you cap the fee, you don't abolish the cost

The Reason article puts the essential point plainly: lawmakers cannot change the fact that expenses must be offset somewhere. I would not ordinarily quote a libertarian outlet with unqualified approval — our disagreements on the proper role of the state are considerable — but on this narrow, mechanical point, they are correct, and the correction matters.

Credit card networks are not charities. The interchange fee — the fraction of each transaction that flows from the merchant's bank to the cardholder's bank — funds fraud protection, payment infrastructure, and, yes, the reward programmes that middle-class cardholders have come to treat as a kind of private dividend. Cap the fee by statute, and the issuing bank faces a choice: thin its rewards, raise its annual fees, tighten its credit standards, or find some combination of all three. The cost does not vanish; it migrates. Whether it migrates toward the consumer who carries a balance, or toward the small merchant who cannot negotiate bespoke rates, or toward the lower-income cardholder who never had rewards to lose in the first place — that is an empirical question, and Congress would do well to ask it before the bill passes rather than after.

This is, at its root, an instance of the most durable mistake in economic policy: the confusion of a price with the underlying reality the price is measuring. A price ceiling on bread does not produce bread; a price ceiling on rent does not produce housing; a price ceiling on interchange does not produce cheaper credit. It reallocates who bears the burden. Sometimes that reallocation is morally justified — society may decide that landlords should bear more of the cost of housing scarcity, and reasonable people can argue that case. But the argument must be made honestly, not concealed behind the fiction that the cost has been legislated away.

I will acknowledge the rational case on the other side, because intellectual honesty demands it. Interchange fees in the United States are, by international comparison, notably high — this is inference on my part, drawn from the structural logic of the market rather than recollection of post-1946 data I cannot possess. A market where two networks hold oligopolistic sway over a payments system that has become as essential as the electrical grid is not a market in the textbook sense, and there is a legitimate argument that some regulatory correction is warranted. The question is whether a blunt fee cap is the instrument, or whether structural remedies — interoperability requirements, network access rules, the kind of deliberate public architecture I always believed financial systems deserved — would reach the disease rather than merely rearranging the symptoms.

The deeper macroeconomic concern is one of distributional consequence. If caps compress rewards and tighten credit standards simultaneously, the households most likely to lose access to credit are those at the margin — not the affluent cardholder who will simply move to a premium product, but the worker whose card is a thin buffer between a car repair and a missed rent payment. Aggregate demand is not indifferent to whose purchasing capacity contracts. A policy that redistributes the cost of credit downward along the income scale is a policy with macroeconomic consequences, and those consequences deserve at minimum a paragraph in the committee report.

What would I build instead? A payments system designed as a public utility, with transparent cost accounting, regulated access, and a genuine competitive floor — not a fee cap that leaves the oligopoly intact while redistributing its costs onto the least powerful participants. That is a harder political lift, I grant you. But it is the policy that addresses the actual problem, and in my experience, the settlements that ignore the actual problem tend to generate their own sequels.