Don't Cancel Premium Cards With an Annual Fee. Downgrade Them Instead

The renewal notice arrives, the annual fee is charged again, and the honest tally of last year's lounge visits is zero. The instinct is to close the account. That instinct is usually the most expensive option available, and there is a cheaper one that most people never ask about.

What closing an account actually costs you

Cancelling does not merely stop the fee. It removes an account from your credit file, and that has two separate effects that people tend to merge into one.

It removes the card's limit from your total available credit. Utilisation is measured as balances against limits. Take a limit away and the same balances elsewhere immediately represent a higher ratio. FICO publishes that amounts owed is 30% of the score, second only to payment history at 35%, so this is not a marginal factor.

It eventually affects the length of your credit history, which carries a 15% weighting in the same model. An old account is an asset that took years to create and cannot be recreated by opening something new.

Neither effect is a catastrophe on a strong file with several accounts. Both are entirely avoidable.

The mechanism: a product change

Most issuers will move an existing account to a different card in their own range — commonly called a product change or a downgrade. Because it is generally handled as a change to the existing account rather than a new one, the account typically keeps its age, its credit line and its payment history, and the annual fee falls away with the product you left behind.

That is the whole trick. You are not choosing between "pay the fee" and "lose the account". There is a third option, and it is usually a ten-minute phone call.

What varies, and what you therefore have to ask rather than assume:

• Which cards you may move between. Issuers generally keep moves inside their own range, and often inside a family.
• How long you must have held the card. Many issuers apply a minimum holding period before a change is allowed.
• Whether a search is run. A product change is commonly handled without a new application, but this is an issuer decision, not a law.
• What happens to the rewards balance. The single most important question, and the one most easily left until it is too late.
• Whether the fee is refunded. Some issuers refund a recently-charged annual fee within a window; the window is theirs to set.

We are not going to publish a table of issuer-by-issuer rules, because those rules change quietly and a stale table would be worse than no table. Ask your issuer about your account, and get the answer in writing.

The retention conversation to have first

Before you downgrade, there is one call worth making, and it is not the one people expect.

Ask the issuer what it can do to keep the account as it is. Some issuers will discuss the fee or attach an offer to the account; some will not, and some will tell you the answer depends on the account. There is no universal script and anyone who publishes one is guessing. What is universal is that the question costs nothing and the worst outcome is the position you were already in.

Two rules for that conversation. Do not threaten to close an account you do not want to close, because an issuer may simply agree. And write down what you were told, including the date and the name, because a benefit described on a call is only as good as your record of it.

A worked example: the three options priced

Take a card with a 400 annual fee that the holder used for one lounge visit and no insurance claims last year. Available credit across all their cards is 20,000, of which this card is 8,000. Balances total 3,000, so utilisation is 15%.

Option one, keep paying. Cost: 400 a year for value they honestly assess at about 40. Net loss roughly 360 a year, repeating.

Option two, close it. Cost: nothing in cash. But available credit drops from 20,000 to 12,000, and the same 3,000 of balances now reads as 25% utilisation rather than 15%. Nothing has changed in their behaviour; the file simply looks worse. If a mortgage or car finance application is anywhere in the next year, that is a poor trade for a fee they could have escaped anyway.

Option three, product change to a fee-free card in the same range. Cost: nothing. Fee: gone. Limit: retained, so utilisation stays at 15%. Account age: retained.

On these assumptions option three dominates. The figures are illustrative — put your own limits and balances in — but the ranking rarely changes, because option three removes the cost without removing the asset.

The questions to ask, in order

1. Is a product change available on this account, and to which cards?
2. Does it run a credit search, or is it handled on the existing account?
3. Does the account keep its opening date and its credit line?
4. What happens to my existing rewards balance? Ask explicitly whether anything expires, converts, or loses transfer options.
5. Is the annual fee refunded, in whole or in part, and by when?
6. Are there benefits I am mid-claim on, or credits I have already used this year, that the change would affect?
7. Will the card number change, and do I need to update recurring payments?

Question seven is the practical one that catches people. A changed card number means every subscription and every stored payment method needs updating, and the ones you forget fail silently. If you are running a lot of recurring payments off one number, that is an argument for virtual card numbers generally, and for doing the change at a quiet moment specifically.

When closing really is right

The counter-argument, and it is a strong one in three cases.

When there is nothing to move to. If the issuer has no fee-free card you can hold, a product change is not on the table and the choice really is pay or close.

When the account carries a term you want out of. A card you are unhappy with for reasons other than the fee is not fixed by keeping it.

When the card drives spending you cannot control. This is the one that outranks every scoring consideration on this page. Premium cards are designed to increase spending, and a rewards structure that is making you spend more than you otherwise would is costing you far more than a utilisation ratio ever will. If that describes the account, close it, and do not let a 15% utilisation figure talk you out of a decision that is right for reasons arithmetic does not capture.

There is also an honest limit to how much any of this matters. On a thick file with several long-standing accounts, closing one card is a small effect that fades. The reason to prefer a product change is not that closing is dangerous — it is that the alternative is free.

The part that matters more than any of it

None of this is worth anything if a balance is revolving. The Federal Reserve reports the average rate on US commercial bank credit card plans at 20.94% across all accounts and 22.15% on accounts assessed interest. At those rates a single month of carried balance costs more than the utilisation optimisation above saves in a year.

The grace period is what keeps that cost at zero, and it is conditional. The CFPB defines it as "the period between the end of a billing cycle and the date your payment is due", notes that card companies "must establish procedures to assure that their bills are mailed or delivered to you at least 21 days before the payment is due", and warns that "if you lose your grace period by not paying your balance in full by the due date, you will be charged interest on the unpaid portion of the balance."

Pay the statement balance in full, every cycle, and everything else on this page is worth doing. Carry a balance and the fee you were agonising over is the smallest number in your account.

Where to go from here

If you are still deciding whether to keep a premium card at all, is a credit card annual fee worth it works the break-even properly. If the card is not earning its fee because you never read what it covers, how to read your card's Guide to Benefits is fifteen minutes well spent — most people are paying for cover they do not know they have.

If you are choosing between rungs rather than leaving, Visa Infinite versus Visa Signature is the comparison that usually matters, and does applying for a credit card hurt your credit score covers what a new application costs if you go that way instead. If the fee was buying travel benefits you never used because your spending is mostly in other currencies, the cheaper tool is a multi-currency account such as Wise — see multi-currency account versus travel credit card.