In 1981, American Airlines launched AAdvantage, a program that rewarded travelers for flying repeatedly with the same carrier. United quickly followed with its own. Within a few years, nearly every major airline had one. Today, loyalty programs span coffee shops, hotels, supermarkets and credit cards, and most consumers carry a phone full of them.

This creates a genuine strategic puzzle. At first glance, a loyalty program is a discount, a slice of margin handed back to customers. Firms rarely give away margin without reason. So what are they actually buying? And if every competitor runs the same playbook, does anyone come out ahead?

The answer depends on three distinct mechanisms operating at once. Rewards manufacture switching costs. Membership generates information. And competition between programs can trap an entire industry in an equilibrium nobody chose. Each mechanism changes the game in a different way. Understanding how they interact explains why some loyalty programs are strategic assets while others are expensive obligations that firms cannot abandon.

Switching Cost Creation

Economists distinguish between two kinds of customer retention. The first comes from genuine preference: you return because the product is better. The second comes from the cost of leaving. Loyalty programs are engineered to produce the second kind, converting otherwise interchangeable purchases into a sequence of linked decisions in which each choice shapes the value of the next.

The mechanism is accumulation. Each purchase adds points toward a threshold, such as a free flight, a status tier or a complimentary night. As a customer approaches that threshold, the effective price of defecting rises. Buying from a rival does not just forgo the points on that transaction. It also delays or forfeits a reward already partially earned. Nonlinear structures amplify this effect. Status tiers that reset annually, or rewards that unlock only at fixed milestones, create sharp incentives to concentrate spending with a single provider.

Economist Paul Klemperer's work on switching costs reveals the strategic implication. When customers can be locked in, firms face a two-period game. Early on, they compete aggressively, sometimes below cost, to capture customers. Later, they harvest the installed base through higher prices or thinner service. Competition does not disappear. It shifts in time, concentrating on acquisition rather than retention.

This has an uncomfortable consequence for the customer. Rewards that feel like a gift are partly a prepayment against future pricing power. A traveler loyal to one airline may pay a premium on routes where cheaper alternatives exist, rationalizing the higher fare against the miles earned. The program works precisely when the customer stops comparing prices with full attention. The more successful the lock-in, the less the firm needs to earn each sale on merit.

Takeaway

A switching cost does not eliminate competition; it relocates it. When a firm courts you generously at the start, consider what it expects to recover once leaving becomes expensive.

Customer Information Value

Switching costs explain only part of the story. Many loyalty programs offer rewards so modest, a free coffee after ten purchases, that they barely alter behavior. Yet firms persist, because the program's real product may be the membership itself. Every scan links a transaction to an identity, turning anonymous sales into a continuous record of individual choices.

Before loyalty cards, a supermarket knew what it sold but not to whom. Afterward, it could see that a particular household bought baby products on Thursdays, switched brands whenever a rival went on promotion, and stopped buying wine every January. Tesco's Clubcard, launched in the mid-1990s and developed with the analytics firm dunnhumby, became a textbook case of a retailer converting purchase histories into a strategic asset.

Information economics explains why this data is so valuable. Markets suffer from information asymmetry: the firm does not know each customer's willingness to pay. Behavioral data narrows that gap. It enables price discrimination through targeted offers, sending discounts to price-sensitive shoppers while charging full price to those who would buy anyway. It sharpens inventory decisions, store layouts and product launches. And it strengthens negotiations with suppliers, who will pay handsomely for insight into how their products actually move.

Seen this way, rewards are the price firms pay customers for their data. The exchange can be fair or lopsided, depending on how the information is used. Targeted offers can genuinely serve customers by surfacing relevant products and real savings. But the same capability that identifies who values a discount also identifies who will quietly tolerate a higher price. The data cuts in both directions, and the firm holds the knife.

Takeaway

When a reward seems too small to justify a program's existence, look for the data trade. The customer is often not just a buyer but a supplier of information.

The Competitive Loyalty Arms Race

Now consider the strategic interaction. Suppose one firm launches a loyalty program in a market of otherwise similar competitors. It gains share by raising switching costs and learning about its customers. Rivals face a clear choice: match the program or lose customers to the innovator. Matching is the rational response, and it usually comes quickly.

This produces a classic prisoner's dilemma. Each firm prefers having a program regardless of what rivals do. But when everyone has one, the advantages largely cancel. Customers divide their loyalty much as before, switching costs offset one another, and every firm now carries the expense of administering the program and honoring accumulated rewards. The Nash equilibrium, in which all firms run programs, can leave the entire industry worse off than if none had started.

The arms race also tends to escalate. Once matching is universal, differentiation requires richer rewards, bonus categories and premium tiers. Unredeemed points pile up as liabilities on balance sheets. Customers learn to multi-home, holding memberships everywhere and redeeming opportunistically, which dilutes the lock-in effect while preserving the costs. Exit becomes difficult too. The first firm to abandon its program risks signaling weakness and inviting mass defection.

Not every loyalty contest ends in mutual destruction. Programs create net value when they deliver something rivals cannot easily replicate: a broad partner network, a superior redemption experience, or data capabilities that genuinely improve the product. Airlines that sell miles to credit card issuers have turned a defensive expense into a substantial revenue stream. The decisive difference lies in whether a program builds an asymmetric advantage or merely reproduces what every competitor already offers.

Takeaway

A strategy that is individually rational can be collectively ruinous. Before matching a rival's move, ask whether it creates advantage or merely restores parity at a higher cost.

Loyalty programs look simple: spend more, earn rewards. Beneath that surface sits a layered strategic structure. Rewards build switching costs that shift competition toward early acquisition. Membership generates information that narrows the gap between what firms know and what customers will pay.

Yet the same logic that makes a single program attractive can trap an industry. When every firm matches, the advantages neutralize, leaving cost, complexity and a commitment nobody can easily abandon.

The broader lesson extends well beyond rewards cards. Any competitive tool that is easy to copy tends to become table stakes rather than advantage. The question for strategists is not whether a move works in isolation, but what happens once everyone makes it.