Why Transferring Points Usually Loses Value

You are not holding points. You are holding a choice.

This is the part readers resist, so let me put it as plainly as I can. A transferable point is not valued for what it is. It is valued for what it could still become.

When you hold American Express Membership Rewards points, you have not yet chosen a program. At the moment you want to book, you can look at what every partner is charging for the seat you actually want, and move your points to whichever one prices it best. You are holding the best of all of them, decided later with information you do not have today.

When you transfer, you give that up. The points become one program's points, priced by one award chart, subject to one set of blackout dates and one devaluation risk. You have swapped the best of many for one specific thing, permanently.

So the comparison a 1:1 transfer really makes is not average partner against this partner. It is best available partner, chosen later against this partner, chosen now. The first is worth more than the second by construction — that is what an option is — and the gap is roughly the 54%–69% premium in the table above. The publishers are not overvaluing bank points. They are pricing the choice, which is the thing you are actually giving away.

Informational only, not professional advice. This guide explains arithmetic on published third-party estimates. It does not evaluate award availability or transfer eligibility, and it recommends no credit card or loyalty program.

What that costs, on the two most common transfers

Take the single most common transfer in the market: American Express Membership Rewards to Delta SkyMiles, at 1:1. The useful number is the break-even ratio — how many partner points you would need per bank point to come out level — because it turns the question from a matter of opinion into a comparison of two numbers.

You would need between 1.67 and 2.00 Delta miles per Amex point to break even. You get one. On 50,000 points that is a loss of between $400.00 and $550.00 of value, depending on whose figures you use.

More points does not mean more value

The ratios that look generous are often the worst, and this is where the optionality argument earns its keep. American Express Membership Rewards transfers to Hilton Honors at 1:2 — double the points. It still gives up between 54.5% and 60.0% of the value you were holding, because break-even on that pair is 1:4.40 to 1:5.00.

Doubling a currency worth a fifth as much is still a loss. A ratio in isolation tells you nothing at all — it only means something next to the two valuations it sits between, which is exactly what a bare “1:2” on a marketing page hides.

The exceptions, and why they prove the point

6 of the 132 pairs gain value at 1:1 by at least one publisher. 5 of those involve one program: ALL — Accor Live Limitless. That is not noise, and it is the most useful thing in this guide.

Accor points do not buy award nights off a chart. They convert to a fixed cash-equivalent discount — a euro-denominated amount off any bookable rate. So there is no optionality to lose, because there was never an award chart to be locked into. A fixed-value currency is worth the same whenever and however you use it. Transferring into one is not spending an option; it is exchanging one liquid thing for another.

That is the mechanism confirming itself. The pattern breaks precisely where the explanation says it should — at the one program in the table whose points are not a bet on future award availability. If the premium were just publishers being sentimental about bank points, it would not politely disappear at exactly the fixed-value program.

The exceptions are also a reminder that this is a claim about published valuations, not a law of nature. They are listed here rather than smoothed away because a finding that quietly drops its counterexamples is not a finding.

When the loss on the screen is not a real loss

Here is the objection this guide has been building toward, and it is a good one. I transferred to a partner last year and got 5¢ per point on a business-class seat. Are you telling me I lost value?

No. And understanding why is the difference between using this tool well and being misled by it.

A published valuation is an average-case figure: roughly what the currency returns across the range of ways people actually use it. The transfer partner calculator compares one average-case figure against another, so what it reports is an average-case answer. It says: as a general matter, converting a flexible currency into this fixed one costs you.

It does not say your specific award was bad. If you have found a seat that prices at 5¢ per point, you are not making the average-case trade — you are exercising the option, which is the entire reason the option was worth holding. The premium you paid for optionality is not wasted when you use it. It is what you bought it for.

So the practical rule is simple, and it is the opposite of what the raw percentages suggest:

• Never transfer speculatively. Moving points to a program because a bonus is running, or to “top up” a balance for a trip you have not priced, is the case where the loss on this page is exactly real. You have spent the option and received the average case.
• Transfer against a specific, confirmed award. Price the actual booking first with the cents-per-point calculator. If that number is strong, it is the number that decides the transfer — not the average-case comparison, which has already been superseded by better information.
• Check availability before you move anything. Transfers are irreversible and most take effect in minutes. The option has value right up until the moment you spend it, and no value at all afterwards.

Read that way, the 123-of-132 finding is not an argument against transferring. It is an argument against transferring early. The published numbers are telling you that flexibility is the most valuable property these currencies have, and that you should charge a specific, verified award for it before giving it up.

What this argument does not establish

• It is circular in one respect, and that should be named. Publishers value bank points partly becauseof their optionality, so “bank points are worth more because they are flexible” is in part a restatement of how the figure was built, not independent evidence for it. What the data does establish is that three publishers using different methods all reached that judgement.
• The premium is a judgement, and reasonable people set it lower. If you only ever fly one airline, the option is worth much less to you than to someone who books opportunistically — your personal premium might be near zero, and then most of the loss shown here is not a loss you are actually taking.
• Transfer bonuses cannot move break-even. A bonus raises the ratio you receive, not the ratio you need. It can close the gap and occasionally clear it, but it never changes the target — which is a useful way to see that most bonuses improve a losing transfer without fixing it.
• None of this prices minimums, increments, or timing. Programs impose transfer minimums, round to increments, and some transfers take days. A transfer that arrives after the seat is gone is a total loss no percentage here captures.

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.
• ALL — Accor Live Limitless — Reward programme. Accor points convert to a fixed cash-equivalent discount rather than to an award chart — the published mechanic behind this guide's explanation of why fixed-value programs are the exception to the pattern.

Last reviewed: July 2026