Breakage is not profit

The argument

Breakage begins as a liability, not income. When a member earns a point, the programme has received consideration and owes future performance. Until redemption or expiry, that point is deferred revenue. The fact that some members will never redeem does not mean the earning process is complete for the unredeemed pool; it means the programme is estimating how many obligations will lapse. An estimate is a modelling input, not a realised gain. The register profiles 217 programmes and 87 vendors, and across all of them no count of outstanding points can change that timing.

Among the 214 programmes with known tier status, 112 run tiers. A tier is a promise of future recognition. Those 112 programmes have formalised obligations that outlive any single point expiry. Breakage booked while tiers remain live double counts the same member relationship: the programme records profit from points that may still be used to qualify for a tier, while still advertising that tier as available.

The sector pattern makes the temptation visible. All 61 airline programmes and all 18 hotel programmes run tiers. These are the programmes with long-lived points, frequent expiry, and transfer complexity, where breakage rates are highest. Yet none of those 61 or 18 can call breakage profit without first extinguishing the tier promise. In grocery, where 0 of 18 programmes run tiers, points are used quickly and breakage is lower. The register's own counts therefore show that breakage is concentrated exactly where the future obligation is strongest, not where it is weakest.

Even the supply side is opaque. Of 87 vendors profiled, only 48 publish full or partial pricing. Breakage assumptions are usually embedded in vendor contracts and point expiry rules. If breakage were straightforward profit, the pricing and assumptions would be transparent. Their absence suggests the industry treats breakage as a contested estimate, not a settled earning event.

The strongest counter-argument

The strongest counter-argument is contractual expiry. A point with a fixed expiry date becomes worthless to the member on that date. If a programme has a long history of stable non-redemption, then an expected breakage rate is not a guess; it is an actuarial fact about the member pool. Under accrual accounting, that expected breakage can be recognised as revenue before every individual point lapses because the obligation to the pool has fallen. A blanket refusal to treat breakage as profit would force a programme to carry liabilities for members who will never return, overstating debt and understating earnings. This is not a straw man: it is the standard justification in accounting standards.

What would change our mind

A public, audited reconciliation covering a full point expiry cycle would change our mind. We would need to see a programme with a fixed expiry date publish a line-by-line account showing that every point classified as breakage never later generated a redemption, reinstatement, refund or upheld member claim. If that reconciliation shows zero reversals and no increase in the outstanding liability after the breakage was recognised, then breakage was profit for that programme. That is the observable test.

What follows if we are right

A programme should treat breakage as a reduction of the redemption liability only when points lapse or are redeemed, not when a model predicts they will. It should fully fund the outstanding point balance, disclose its breakage rate and expiry assumptions, and never use expected breakage to inflate current profit or justify weaker redemption offers. Running the programme this way keeps the balance sheet honest and protects member trust.