The easiest part of a card rate cap is cutting the rate

The week

Late on the night of January 9, 2026, the President posted on social media that credit card interest rates should be capped at 10% for one year. He said he hoped the cap would begin on January 20, eleven days later. For an industry where many cards carry rates well above that level, it was a startling message to wake up to.

The post left the most important question unanswered: how would the cap actually happen? It did not say whether it would come through an executive order or through a new law passed by Congress. Commentators were quick to point out that no general federal authority exists that would let the executive branch impose such a cap on its own. In practice, a binding cap would need Congress.

The banking industry reacted on two fronts. Bank trade groups, the associations that represent banks in Washington, put out a joint statement opposing the idea. Their argument was that if lenders cannot charge enough to cover the risk of lending to some customers, they will simply lend less, and credit will become harder to get. Investors reacted too. Bank shares fell as markets weighed what a cap could do to card profits.

There was already a version of the idea on the table. A bipartisan Senate bill introduced in 2025, S. 381, proposed a similar 10% cap. It had not moved out of committee, which is the stage where most bills stall.

During the week of January 19, the target date arrived and passed. No binding cap took effect. On January 21, the President shifted course and called on Congress to enact a cap. The next day, reports said that Bank of America and Citi were considering card products with a 10% rate. That was a notable twist: even without a law, at least some issuers appeared to be exploring whether to offer lower rates by choice.

So the week ended with no new rule, no new law and no order. But it left every card issuer with a practical question worth answering before the next time it comes up: if we had to lower our rates quickly, what would that really involve?

The rate-cap episode is easy to dismiss as political noise. It produced no law, no rule and no order. Nobody’s card statement changed. Yet it handed every card issuer a concrete question, and the reports that some of the largest banks were weighing voluntary 10% products showed that some were already working on an answer. If an issuer had to lower rates, whether by mandate, by political pressure or by its own product choice, what would actually happen inside the organization?

The intuitive answer is that the hard part is the rate. Changing a number is in fact the simplest piece. The harder pieces are the offsetting changes an issuer would make to protect its economics, and the record it would need to defend those changes later. The rate, the offsets and the record are taken in that order below.

How a card earns its interest

A credit card is a revolving loan. The bank agrees to lend up to a set amount, the credit limit, and the customer can borrow, repay and borrow again. Customers who pay their full balance each month typically pay no interest at all. Customers who carry a balance from month to month, known as revolving, pay interest on it. That interest is a major source of card revenue, alongside fees and the small charge merchants pay each time a card is used.

The interest rate is expressed as an annual percentage rate, or APR, a standardized way of stating the yearly cost of borrowing so that customers can compare offers. Many card rates are variable. They are tied to a benchmark called the prime rate, which moves with the Federal Reserve’s policy rate, plus a margin that reflects the customer’s credit risk. Riskier customers usually get a higher margin.

A cap would put a ceiling on that number, much like a speed limit added to a road that never had one: the bank’s systems may never have been built to check for it.

Two more terms come up throughout. The first is the card processor. Most banks, especially community banks and credit unions, do not run their own card computer systems. They rent them from specialist companies that handle the accounts, calculate interest, produce statements and connect to the card networks. The bank sets the terms; the processor carries them out. The second is the regulator’s toolkit: a set of consumer protection laws covering what a lender must tell customers, how it must treat them fairly and what notice it must give before changing the deal.

The rate: mostly a configuration problem

Lowering an annual percentage rate is legally the least fraught change an issuer can make. Federal advance-notice rules for cards are aimed mainly at changes that hurt the cardholder, such as rate and fee increases, so a reduction is generally simpler. Customers rarely complain that their rate went down. Counsel should still confirm the details for any specific program.

The real constraint is operational. Most community banks and credit unions run card programs on third-party processors, and many large issuers run several platforms after acquisitions, each with its own quirks. A portfolio-wide rate change touches far more than one field in a database. It reaches pricing tables, the calculation of interest, promotional balances, penalty rates, monthly statements, required disclosures and regulatory reporting. Processors work in scheduled change windows, often a few times a year, with testing requirements before anything goes live. A variable rate tied to the prime rate may need a ceiling the platform was never configured to apply.

Suppose a regional bank with a few hundred thousand cardholders runs its cards on a processor shared with dozens of other banks. Its chief executive asks on a Monday how quickly it could cap every account at 10%. The honest answer from operations might be: we need a change request, a slot in the processor’s next release, a test cycle, new statement language and a check that our variable-rate accounts will stop rising when the prime rate does. A project like that is measured in months, not days.

The test is simple: ask the processor how long a portfolio-wide rate ceiling would take, and what it would break. The answer is useful even if no cap ever comes.

Federal credit unions provide a working reference. They already lend under an interest rate ceiling set by the NCUA board under the Federal Credit Union Act, currently 18%. Their card programs and processor configurations are built around a cap. That shows the mechanics are manageable when they are planned rather than improvised.

Voluntary low-rate products raise a different configuration question. A new card with a capped rate is a product launch, with its own disclosures, underwriting criteria and marketing review. It is usually easier than repricing an existing portfolio, because it touches only new accounts and leaves the existing book alone. It also lets an issuer decide who qualifies. That choice carries its own fair lending questions, covered below, but it helps explain why the first reported responses were new products rather than across-the-board cuts.

The offsets: where the legal exposure sits

A lower rate reduces revenue on revolving balances. A business that loses revenue in one place looks to recover it somewhere else, and card issuers have a short list of ways to respond. Each carries more risk than the rate cut itself. They can tighten approvals for new customers, lower credit limits on existing accounts, close inactive or marginal accounts, raise or add fees, or reduce rewards.

Each lever has its own rules, governed by laws that many executives know only by their acronyms.

• Cutting limits or closing accounts. Reducing a credit line or closing an account can count as adverse action under the Equal Credit Opportunity Act and its implementing rule, Regulation B: a decision against a customer’s interest, which generally requires a notice explaining the reasons.
• Raising or adding fees. Fee increases generally require advance notice under Regulation Z, the rule that implements the Truth in Lending Act. The CARD Act, a 2009 law aimed at card practices, limits how fees and rates can change on existing balances.
• Trimming rewards. Rewards changes depend on the program’s terms. They are judged against the standards for unfair, deceptive or abusive acts or practices, often shortened to UDAAP. The question is whether customers were treated fairly and told the truth.
• Tightening approvals. Stricter underwriting needs fair lending testing, because credit score cutoffs and other thresholds that move quickly under pressure are the ones most likely to fall unevenly across protected groups.

The trade groups’ warning that a cap would reduce credit availability is, in practice, a description of these offsets. Whatever one thinks of the policy, an issuer that responded by cutting limits for customers with thin credit files, people with short or sparse credit histories, would be making a fair lending decision as well as a pricing one. Each offset should therefore go through the same review as any underwriting change: model risk, fair lending, complaints and legal sign-off.

Things can go wrong quickly. Suppose the same regional bank decides to protect its margin by lowering credit limits for every account below a certain credit score. The decision is made in a week, by a small group, using a single number. Months later, an analysis shows the cut fell much more heavily on customers in certain neighborhoods. The pricing decision has become a fair lending problem, and the bank cannot show it tested for that before acting.

Partnership programs add a layer. In co-brand cards, the ones that carry an airline’s or retailer’s name, and in arrangements with financial technology companies, pricing, fees and rewards are often set in the partner agreement. Neither side may be able to change them alone, and some agreements specify how the cost of regulatory changes is shared. Those clauses are worth reading before a crisis rather than during one.

The record: explaining who lost access and why

Public pressure on pricing tends to produce later scrutiny. That can take the form of letters from lawmakers, supervisory data requests from bank examiners, inquiries from state attorneys general or press reports on named issuers. The question will rarely be why the rate was what it was. More often it will be who lost access to credit, who paid new fees, and whether the institution’s explanations to customers were accurate.

An issuer should be able to show three things.

1. How its rates relate to risk. How rates relate to credit risk, funding cost and loss experience, documented before anyone asks.
2. How each offset was decided. How each offsetting change was decided, tested for disparate impact (whether a neutral-looking rule falls harder on protected groups) and approved.
3. What customers were told. What it told customers, and when.

That third point is easy to underestimate. When a pricing story is in the news, customers call, chat and post. The contact center, chat tools, branch staff and social media team should all use one approved message, logged with dates. In January, the accurate message was that nothing had changed. Inconsistent answers, with one channel implying a rate cut and another implying a credit squeeze, create their own deceptive practices risk. A well-meaning branch employee who tells a customer that rates are about to fall has, in effect, made a promise the bank did not make.

That record serves the customer as much as the regulator. A cardholder who sees a lower limit or a new fee deserves a clear reason. An institution that can give one has kept the customer relationship even while changing the terms.

How a pricing shock moves through an organization

When a pricing question lands suddenly, the risks inside a bank appear in a predictable order.

Day one: the question arrives. The chief executive, the board chair or a major investor asks what the proposal means. Finance produces a rough estimate of lost interest income. That number is useful, but it is also where the danger begins, because it frames the problem as a revenue gap to be filled.

Week one: options appear. The card business sketches the levers: fewer approvals, lower limits, new fees, thinner rewards. Each looks like a straightforward commercial choice on a slide. Few of those slides mention adverse action notices or fair lending testing, because the people drawing them up are thinking about margin, not compliance.

Weeks two to six: technology and vendors weigh in. Operations discovers which changes the processor can make quickly and which must wait for a scheduled release. This is often where plans change shape. A lever that is easy to configure, such as cutting limits by a fixed percentage, can become attractive simply because it is fast, not because it is fair or wise.

Months later: the questions come back. If anything changed, examiners, lawmakers, journalists and customers start asking who was affected and why. At that point, the institution can only show what it documented at the time.

The answer is to bring compliance, fair lending and customer communications into the room on day one, without slowing everything down, alongside finance and the card business, so that the options considered in week one already carry their legal and customer consequences.

Earnings, fair lending and the customer whose limit is cut

Every credit-limit cut and fee change a bank makes to recover lost interest will be judged under fair lending and consumer protection law.

Money. Card interest is a meaningful revenue line for many issuers. Any change to it, forced or chosen, ripples into earnings, capital planning and the bank’s share price, as the January market reaction showed.

Risk. The biggest legal exposure does not sit in the rate. It sits in decisions made quickly to compensate for it. Those decisions touch fair lending, consumer protection and model risk, the areas where enforcement actions and remediation costs tend to concentrate.

Customers and reputation. Credit cards are among the most personal products a bank offers. A customer whose limit is cut without a clear explanation does not think about margin pressure. They think the bank let them down, and they may say so publicly.

Operational readiness. The question of how fast a bank can change a core pricing setting is a test of its technology and vendor relationships. A bank that does not know the answer for cards probably does not know it for deposits or loans either.

Habits of the prepared issuer

Well-prepared issuers share a few habits.

• They know, in writing, how long their processors would take to apply a rate ceiling, and which reports, statements and disclosures would need to change.
• They have a pre-built map of every likely offset and the notice, adverse action and fair lending rules each one triggers.
• They treat pricing changes made under pressure as underwriting changes, with full review, not as quick commercial adjustments.
• They know which partner agreements constrain their pricing and who must agree to a change.
• They have a single, dated customer message ready for use across every channel whenever pricing becomes public news.

The failures mirror those habits. Some issuers assume a rate change is a quick switch and discover the processor’s release calendar only when they need it. Some make offsetting changes in a hurry, by a small group, without fair lending testing. Some let each customer channel improvise its own answer. And some forget the partner agreement until a co-brand partner objects to a change it never approved.

How fast the portfolio could be repriced

• If we had to cap card rates across the portfolio, how long would our processor take, and what else would it affect?
• If we needed to make up lost revenue, which levers would we pull first, and what notices would each require?
• Who reviews credit limit cuts or tighter approvals for fair lending impact, and how quickly can they do it?
• Which partner agreements limit our ability to change pricing, fees or rewards, and what do they say about regulatory change?
• If a lawmaker or examiner asked next month who lost access to credit and why, could we answer clearly?

Answering these requires no view on whether a rate cap is good policy. They are questions about whether the institution can change its own pricing safely, fairly and on time. That capability is worth having whatever Washington decides.

With hindsight

The January 20 date passed without a binding cap, and the legal obstacles commentators identified at the time held. By the time of writing, no federal cap had been enacted, and we found no record of the Senate bill moving out of committee. Analysts generally treated the odds of a cap as low because it would require Congress, and Congress did not act.

Separately, in February the NCUA extended the temporary 18% ceiling for federal credit unions through September 10, 2027. That decision was a reminder that interest rate ceilings are not hypothetical in American consumer lending. One part of the industry already lives under one, and its systems, products and compliance programs are built around it.

The question the episode raised has outlived the proposal. Political attention to consumer credit costs did not disappear. Voluntary low-rate products also show that rate ceilings can arrive through product strategy as well as through law. A bank may decide on its own, for competitive or reputational reasons, to offer a lower rate to some customers, and it then faces every one of the operational, fair lending and disclosure questions a mandate would have raised.

Pricing agility is a capability, not a given. The issuers best placed know how long a rate change takes on their processors. They have mapped the fair lending and notice consequences of each offset. They keep a dated record of what they told customers. That work does not depend on predicting what Congress will do next.

Those that treat pricing as a fixed configuration, something set once and rarely touched, will find out its limits under pressure. The time to learn how fast and how safely a card portfolio can be repriced is during an uneventful quarter, not during a week when the President, the press and the bank’s own customers are all asking the same question at once.