A credit limit can be raised or cut without any change in how the cardholder has behaved. Limits are managed as a risk position by the issuer, not as a promise to the customer.
A limit is an exposure the issuer carries
An unused limit is a commitment to lend on demand, and issuers must hold capital against that commitment even where nothing is drawn.
Aggregate unused limits across a portfolio therefore represent a real cost, and reducing them frees capacity regardless of individual behaviour.
The exposure is also unpredictable, because the accounts most likely to draw heavily on an unused limit are those whose circumstances have deteriorated, which is exactly when the issuer would prefer not to lend.
The account agreement almost always reserves the right to vary the limit, which is what makes unilateral change possible.
Reviews run continuously in the background
Issuers refresh their view of each account regularly using bureau data, internal payment history and observed spending patterns.
Signals that carry weight include rising balances elsewhere, new applications, changes in payment behaviour and shifts in how the card is used.
Because the review is automated and periodic, a change can appear at any time and will not correspond to a single triggering event the customer can identify.
Dormancy is a signal in its own right, since an account left unused for a long period generates no revenue while still consuming capital, which makes it an easy candidate for reduction.
Portfolio conditions matter as much as the individual
When funding costs rise or losses increase, issuers reduce exposure across whole segments rather than only where individual risk changed.
A cardholder with an unblemished record can therefore see a reduction because of the segment they occupy, which is invisible from their side.
The same logic runs in reverse during expansion, which is when unsolicited increases arrive in volume.
Reductions have knock-on effects
Credit scoring models consider the relationship between balances and limits, so cutting a limit raises that ratio without the balance changing.
That can reduce a score, which may in turn prompt other issuers to review their own exposure, producing a sequence from a single decision.
The effect is strongest where a large share of available credit sits on one account, since removing it moves the aggregate figure sharply.
Increases are not unambiguously good
A higher limit improves the balance-to-limit ratio and is generally treated favourably by scoring models.
It also raises available credit, which other lenders assess as potential future borrowing when considering affordability for a mortgage or loan.
Notice requirements, the right to decline an increase and the ability to request reinstatement after a cut differ by jurisdiction and by issuer, and asking is usually free.