"Earn points and they are yours." Most people who carry a rewards card have heard that line. It is not true. A rewards point is not an asset you own. It is a liability the issuer owes you, and the issuer holds sole power to reprice it.

In the past few weeks, Chase changed its Hyatt transfer ratio from 1:1 to 4:3. A Sapphire Preferred cardholder who sent 40,000 Ultimate Rewards points saw 30,000 come back on the confirmation screen. Ten thousand points gone. No fraud. No glitch. The issuer repriced its own liability.

The evidence sits in a public document.

JPMorgan Chase filed its 2025 Form 10-K with the SEC. One line item tells the story. The company carries $16.0 billion in unredeemed rewards on its balance sheet. Not as an asset. As a liability. That figure was $14.4 billion one year earlier. Points earned outpaced points redeemed all year long.

Every point a cardholder "earned" sits on Chase's books as a debt the company owes. Not a gift. Not a perk. A debt.

The same filing carries a note most people skip. A small shift in how many points get redeemed and what each point costs would move that liability by roughly $512 million. Half a billion dollars from a tiny rate change. That is the pressure behind every ratio cut and every program tweak. The issuer is not being greedy. The issuer is managing a growing obligation on its own balance sheet, and the transfer ratio is one of the levers it can pull.

The data do not lie, even when we would prefer they did.

Chase's Hyatt ratio cut did not happen alone.

For years, Chase set itself apart with 1:1 transfers to every travel partner, no matter which card you held. That era ended. Since mid-2025, Capital One, Citi, American Express, and Chase all cut or ended their transfers to Emirates Skywards. Four issuers. One airline partner. Same direction.

Hyatt itself repriced award categories in the same period. Some went up by as much as 67 percent. Category 8 hotels moved from 45,000 points per night at peak to 75,000. The ratio shrinks on one side. The cost per night rises on the other. The cardholder gets squeezed from both ends, and neither move required his consent.

The Consumer Financial Protection Bureau released its credit card market report in December 2025. The numbers confirm the pattern. Consumers forfeit roughly $500 million in rewards each year. Complaints tied to rewards programs jumped 70 percent above pre-pandemic levels. The system does not have a flaw. The system is working the way the balance-sheet math predicts.

The $16.0 billion keeps rising. The issuer needs tools to bring it down.

The problem is not that Chase changed the ratio. The problem is that years of marketing trained the cardholder to treat points as savings. Points were never savings. They were someone else's debt, priced at someone else's will. Most people treat this as a trust problem. It is a system problem.

The better principle is plain: price every point at its cash floor at the moment you decide. Not at the best-case ceiling the issuer's marketing suggests.

Once that is clear, three moves follow from it.

The guaranteed cash-back rate on your card is your baseline. If the card pays 1.5 cents per point in straight cash, that is the floor. Anything above it is a bonus the issuer can revoke without notice.

That means letting go of the story about "getting 2 cents per point on Hyatt stays."

Hoarding points is lending the bank money at a rate the bank sets. When a specific booking beats your cash floor by a margin you define in advance, redeem. A point redeemed above your floor is a win. A point held for a better day is a bet on terms you do not set.

Chase kept the 1:1 Hyatt ratio for the $795 Sapphire Reserve card. That is not a perk. It is the price of entry to terms that used to come standard. If the card does not pay for itself at worst-case value per point, it does not earn its spot.

Running this for one quarter does something the old advice never did:

You see which cards earn their fee at the floor, not just in the best case. You see how much value sat on the issuer's books while you waited for the right trip. You see the real cost of the "save up and redeem big" approach. And you stop treating a number that shrinks at someone else's will as one that grows at yours.

At the end of the quarter, ask three things.

→ What did I redeem, and what was the cents-per-point value at the time?

→ What looked like a growing balance but lost ground while I held it?

→ Where did the issuer's terms shift without my consent?

That is the difference between advice that sounds right and a system that proves itself.

The confirmation screen showed 40,000 going in and 30,000 coming out. Now you know what it was telling you: not a rewards balance, but a line on someone else's ledger, priced at someone else's discretion. The mechanism will not change. What you do with that knowledge can.