Frequent-flyer currency
A Socratic walk-through of frequent-flyer currency — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why can an airline earn more from giving its seats away than from selling them?
An airline hands a passenger a seat and charges nothing for it — a giveaway funded out of ticket revenue, justified by a vague hope the recipient comes back. Yet at several large carriers the loyalty programme is reported as more reliably profitable than flying aeroplanes, and in the borrowing of 2020 a number pledged their programmes as collateral at appraised values widely reported as exceeding the airline's own stock market value. I recall that episode; I will not quote a figure, since those were appraisals rather than observed prices.
So how can what you give away be worth more than what you sell?
Reasoning it through
REASONING #Start by questioning the word "give". Somebody paid for that seat — who, and when? Follow the miles backwards rather than forwards.
Where did the member's miles come from? On most large programmes the majority were not earned by flying at all, but by spending on a co-branded credit card. And the issuer did not conjure them: it bought them, in bulk, for cash, at a negotiated price per mile, then gave them away as a reward.
Stop there, because the business model is visible. The airline is not running a discount scheme; it is running a mint. It creates a unit of account at essentially zero marginal cost, sells it to a third party, and undertakes to honour it later in seats. The revenue arrives years before the obligation is discharged, and it arrives from banks rather than travellers.
Why would a bank pay real money for something an airline invents? Because the card earns it interchange fees from merchants on every transaction, interest on revolving balances, and annual fees. Miles are how it buys spending volume. Notice where the money originates — card spending across the whole economy, on groceries and fuel and school fees. Almost none of it comes from aviation.
Now the cost side, where the giveaway looks less like one. When the member redeems, what does the airline forgo? That depends on which seat it releases — and the airline chooses. Award availability is controlled inventory, released overwhelmingly onto flights not expected to fill for cash, where the cost of one more passenger is small: fuel to carry their weight, a meal, and government charges, which programmes commonly bill to the member anyway.
Two further margins sit on top. Some miles are never redeemed — accounts lapse, people die, balances sit below the cheapest award forever. That is breakage, pure margin; published estimates vary widely and are commercially sensitive, so I will not offer a rate. And the issuer sets the exchange rate: an airline can raise the miles an award costs whenever it likes, cutting the real value of every mile outstanding with no default or renegotiation. No ordinary debtor has that power.
So the seat was never given away. It was sold in advance, to a bank, and delivered later to whoever the bank chose — with the airline keeping the right to decide which seats are eligible and what they cost.
The analogy
THE ANALOGY #Think of a shopkeeper who prints his own scrip. The local employer buys a stack of it for cash to hand out as bonuses; the workers spend it in the shop, where honouring a note costs the shopkeeper only the wholesale price of the goods. He keeps the gap, keeps the value of every note lost in a drawer, and can raise his shelf prices in scrip whenever he likes.
The shopkeeper's stock is storable and saleable to anyone, whereas an airline seat perishes at departure — which cuts both ways. It makes redemption cheap when the flight would have flown empty, and it makes the airline unwilling to honour scrip at all on the flights people most want, so this shopkeeper also decides day by day whether his shop is open to his own currency.
Clarifying the model
THE MODEL #Three refinements, and one story to abandon.
First, this is not free money without limit. Every mile issued is a liability to be settled in a seat, and the accounts treat it so: bank cash is booked as deferred revenue and released as miles are redeemed or estimated to lapse. Issue too many relative to the seats you will release and the currency devalues in members' eyes, the reward weakens, and the bank pays less next time. The programme is constrained as any currency issuer is — by confidence.
Second, the claim in the question needs narrowing. The airline does not earn more overall from award passengers than from paying ones; it plainly cannot, since the paying ones fund the aeroplanes. It is that the programme, as a separate business, earns a wide and stable margin selling miles, while flying is thin-margin and cyclical. Two businesses sharing a brand.
Third, the structural point most travel writing misses. Programme profitability rests on what a bank can afford to pay for miles, which rests on payment-card economics, which are regulated. Where interchange fees are capped, as in the European Union, co-brand cards generate far less to share and the attached programmes are less lucrative. The gap between a hugely profitable loyalty programme and a modest one is often a payments regulation rather than anything about the airline.
The story to abandon is the one the programmes are named after: that they exist to make travellers loyal. Loyalty effects are real but contested and modest, and the largest programmes belong to carriers with dominant hubs, where alternatives are limited by the network rather than by affection.
Which gives the test. If the currency account is right, programme revenue should track co-branded card spending rather than the airline's own passenger numbers — and in 2020, when traffic collapsed far more than card spending, that is roughly what appeared. It also predicts award availability concentrated on flights that would not have sold, and weaker economics under capped interchange. The refuting observation would be a programme whose revenues rise and fall with its own ticket sales instead.
A picture of it
THE PICTURE #How to readThis is a system-context diagram repurposed as a map of who pays whom, so read every arrow as money or value moving in the direction it points, not as a sequence in time. Start at the merchants: card fees flow to the bank, the bank converts them into cash paid to the programme, and only then does the programme credit miles to the member. Follow the member's own arrows and notice the asymmetry — the member supplies spending and receives miles and eventually a seat, but never pays the programme money. The link from programme to airline operation is the real cost of the giveaway, small because the seat released is one the operation did not expect to sell.
What became clearer
WHAT CLEARED #The award seat is not a gift; it is the delivery leg of a sale made years earlier to a bank. An airline with a loyalty programme runs two businesses: a capital-heavy cyclical one that flies aeroplanes, and a currency business that mints a unit at no cost, sells it, controls what it buys, and can devalue it at will. The second is profitable because its revenue comes from card spending in the wider economy while its obligations are settled in perishable inventory the first could not sell — and because it depends on what banks can pay, it is ultimately a creature of payments regulation.
Where to go next
ONWARD #- How breakage estimates are recognised in accounts, and what changed when the standards tightened.
- Why airlines devalue their currencies in steps rather than continuously, and what that does to member behaviour.
Key terms
TERMS #| Term | What it means |
|---|---|
| Co-brand agreement | a contract under which a bank buys miles in bulk from an airline to award to its cardholders. |
| Interchange fee | the charge a merchant's bank pays the card issuer per transaction; the source of most co-brand revenue, and capped in some jurisdictions. |
| Breakage | miles issued that are never redeemed, and therefore become margin. |
| Award inventory | the seats an airline releases for redemption, controlled separately from seats sold for cash. |
Every term the collection defines is gathered in the glossary.