The Second Year, When the Bonus Is Gone and the Fee Is Not

The number that decides it is a difference, not a rate

The instinct is to compare the card's earn rate against its fee. That is the wrong comparison, and it is wrong in a way that consistently flatters the card.

You are not choosing between this card and no card. You are choosing between this card and the card you would otherwise have used, which also earns something. So what the card returns is not its rate — it is the gap between its rate and the alternative's.

reward rate = points per dollar × cents ÷ 100 edge = reward rate − what the other card returns break-even spend = (annual fee − credits you'd have bought anyway) ÷ edge

Watch what that does. A card earning 1 point per dollar in a currency worth 2.05¢ returns 2.05%, which sounds respectable. Against a plain 2% cash-back card you already own, the edge is 0.05% — five hundredths of a cent per dollar. Covering a $95.00 fee at that edge takes $190,000.00 of spending a year.

Nothing about that card is bad. The currency is genuinely valuable and the published figure is not in dispute. The card is simply not better enough than the thing it replaces to pay for itself, and no amount of enthusiasm about 2.05¢ per point changes it.

The multiplier is the whole game

Same card, same fee, same currency, same 2% alternative. Only the number of points per dollar changes:

The requirement collapses from $190,000.00 to $2,289.00 — a factor of 83 — for a change from 1x to 3x. That is the entire second-year question in one line: not what the card earns, but how much of your actual spending lands in the categories where it earns well.

It is also worth seeing how fragile the 1x case is. Using the lowest published figure for the same currency (2.00¢) instead of The Points Guy's 2.05¢, a 1x card returns exactly 2.00% — identical to the alternative. The edge is zero, and there is no amount of spending that covers the fee at all. The break-even is not a large number; it does not exist. A conclusion that flips between “$190,000.00 of spend” and “impossible” depending on which publisher you read is a conclusion you should hold loosely.

Informational only, not professional advice. This guide explains arithmetic on figures you supply and cites published third-party estimates. It recommends no credit card, ranks no product, and carries no affiliate links.

Nobody spends entirely in one category

The table above is a set of pure cases, and real spending is a mixture. Here is $24,000.00 of annual spend split between a 3x category and everything else at 1x, with the $95.00 fee charged once against the combined result:

The card that looked indefensible at 1x becomes comfortably worth keeping once a meaningful share of spend hits the bonus category — and the crossover is not at some exotic level of spending. This is why “is this card worth the fee?” has no general answer, and why anyone who gives you one without asking what you buy is not doing the arithmetic.

Credits are worth what you would have bought anyway

Cards with larger fees usually offset them with statement credits, and the marketing totals them at face value. The honest number is not face value. A credit is worth what you would have spent on that thing regardless.

A monthly credit at a merchant you already use every month is worth close to its full amount. The same credit at a service you would never otherwise buy is worth zero, and it is worth lessthan zero if it nudges you into spending money to “capture” it. The difference between those two readings is often larger than the fee being argued about.

This is the same discipline as the forgone-rewards term in the welcome offer calculator and the earn-rate term in the award vs cash calculator: count what you would have done anyway as the baseline, and only the difference is real. Applied here it means subtracting credits you would genuinely have purchased from the fee before running any of the arithmetic above — and counting the rest at zero, however prominently it is printed.

Running the year-two calculation on your own card

1. Work out what the card you'd otherwise use returns. A flat cash-back rate is the easiest honest baseline. This is the number everything else is measured against, and setting it to zero is the single most common way to talk yourself into keeping a card.
2. Compute the edge, not the rate. Points per dollar × what you think the currency is worth, minus that baseline. Use a figure from the valuation reference — and try the low end as well as the high, since the answer can flip.
3. Subtract only the credits you'd have bought anyway from the fee.
4. Divide. That is the spend required to break even. If it exceeds what you will actually put on the card, the card is costing you money — regardless of how good the first year was.
5. Sanity-check the currency assumption. Every figure above assumes you eventually redeem at the published rate. Points sitting unredeemed are worth nothing at all, and the cents-per-point calculator is what tells you the rate you are actually achieving rather than the one you are assuming.

What this calculation leaves out

• Non-points benefits. Lounge access, insurance, elite status and purchase protections are frequently the larger part of a card's value and none are here. They are worth what you would otherwise have paid for them, which is a number only you can set — the same test as credits.
• Closing a card is not free. It can affect the average age of your accounts and your available credit, and in some programs it can put an accumulated points balance at risk. Downgrading to a no-fee card in the same family often preserves both; that option is outside this arithmetic entirely.
• Retention offers exist. The fee is not always the fee. This calculation tells you what the card is worth to you, which is exactly the information you need before deciding whether a retention offer changes the answer.
• It assumes your spending is fixed. If holding a card changes what you buy or how much, the baseline comparison stops being clean — and the direction of that error always favours the card.

Sources

• The Points Guy — points and miles valuations. Figures as of 1 July 2026. Refreshed monthly.
• CardRatings — points and miles valuations. Figures as of 10 June 2026. Updated periodically; no published schedule.
• Upgraded Points — points and miles valuations. Figures as of 1 July 2026. Refreshed monthly.
• No card or offer data is cited, by design. Every figure describing a card in this guide — fee, earn rate, spend requirement — is an illustrative structure, not a quotation from any product. PointCents holds no offer database and recommends no card; see the welcome offer calculator for why.

Last reviewed: July 2026