Junk Fees Cost You $650 a Year (Here's the Math)

The hotel is $129 a night until checkout, when it becomes $161 because of a resort fee for a pool you will not use. The concert ticket is $65 until the service fee, the facility fee, and the order processing fee turn it into $94. Junk fees work by separating the number that wins your business from the number that leaves your account, and the distance between those two numbers has become an industry.

The Council of Economic Advisers tried to size it. More than $90 billion a year, about $650 per household. Treat that as an estimate over a wide category rather than a measured total; the Cato Institute and others have argued the definition sweeps in pricing that is ordinary rather than deceptive. Even discounted, the number is large, and the mechanism behind it is not in dispute.

What counts as a junk fee?

A mandatory charge that is not in the advertised price.

That definition does real work. A fee for something optional is a price. A fee you cannot avoid while still completing the purchase is part of the price, disclosed late. The resort fee is not optional. The ticketing service fee is not optional. The apartment "administrative fee" charged alongside an application fee is not optional. None of them appear in the number that made you click.

Categories drawn from the White House Council of Economic Advisers' junk fee analysis and CFPB fee research.

The last two rows are different from the rest, and worse. A resort fee is charged to everyone who books the room. An overdraft fee is charged only to people who ran out of money, which makes it a surcharge on being poor, collected at the exact moment the customer has the least.

How much do junk fees cost a household each year?

The financial-sector slice is the best documented, because the CFPB required banks to report it.

That drop from $15 billion to under $6 billion is the one encouraging number in this article, and it is worth understanding why it happened. No regulation forced it. The CFPB published the figures, named the banks, and proposed rules. Banks read the room and cut or dropped the fees before anything became law. Disclosure alone moved nine billion dollars.

Why doesn't competition get rid of them?

Because hidden fees beat honest pricing in a search ranking, and every seller knows it.

Picture two hotels. One advertises $161 all-in. The other advertises $129 plus a $32 resort fee at checkout. Sort by price and the second hotel appears first and wins the booking, even though it costs the same. The honest lister loses. Within a season, everyone charges the resort fee, because the alternative is invisibility.

Economists call this a shrouded-attribute equilibrium, and the important feature is that no single company can exit it alone. A hotel that unilaterally adopts all-in pricing does not get rewarded for candor. It gets sorted below its competitors. That is why the fees persist in markets with dozens of sellers, and why fixing it requires a rule that binds everyone at once rather than a consumer who shops harder.

The same dynamic drives the gap between an advertised phone plan and the actual bill, where regulatory recovery charges and line access fees live below the number in the commercial.

What did the government fix?

One thing, narrowly.

The FTC's Rule on Unfair or Deceptive Fees took effect May 12, 2025. Sellers must disclose the total price upfront, including every mandatory charge they know about and can calculate, and the total has to be the most prominent number. Taxes and shipping can come later, but before the consumer enters payment information. The FTC estimated the rule would save consumers up to 53 million hours a year of time spent hunting for the real price.

It applies to live-event tickets and short-term lodging. That is the entire scope. Concert tickets and hotel rooms are cleaner now. Apartment application fees, airline ancillaries, delivery service charges, and restaurant surcharges are untouched by it.

Which rules died?

Two, within three months of each other, both on procedural grounds rather than on the merits of the pricing.

The CFPB finalized a rule in March 2024 capping most credit card late fees at $8, down from a typical $32, which the agency estimated would save households about $10 billion a year. On April 15, 2025, a federal court in the Northern District of Texas vacated it after the CFPB itself agreed with the challengers that the cap exceeded what the CARD Act allowed. The prior framework returned, and so did the $30-plus fee.

Then on July 8, 2025, the Eighth Circuit vacated the FTC's click-to-cancel rule in its entirety, days before compliance was due, finding the rulemaking process procedurally defective. The FTC opened a fresh rulemaking in January 2026. Until it finishes, canceling a subscription is legally allowed to stay harder than starting one.

So the scoreboard for the household: all-in pricing for tickets and hotels, nothing for late fees, nothing for cancellation friction, and a patchwork of state auto-renewal and fee-disclosure laws that depends on where you live.

Who pays the most?

The people who can least absorb it, by a wide margin, because the largest fee categories are triggered by shortfall rather than by consumption.

Overdraft fees fall on accounts that hit zero. Late fees fall on people who were short on the fifteenth. A household with a $4,000 cushion never pays either one, so the entire category functions as a fee charged for not having savings. The Federal Reserve's survey work has long found that a large share of American adults could not cover a modest emergency expense with cash. Those are the accounts generating the fee revenue.

A $32 late fee is 4.4 hours of work at the federal minimum of $7.25 an hour, a floor unchanged since 2009. It is eleven minutes for someone earning $175,000. Identical fee, identical bank, and a gap in consequence so wide the two customers are not in the same economy. That is the split running through what corporate pricing power has extracted since 2020.

The price of the price

Fees are not dishonest by nature. Businesses have real costs, and unbundling them can be legitimate. Nobody is owed a free bag on a discount airline.

What the $90 billion estimate describes is something narrower and harder to defend: revenue that depends on the customer not seeing the number until after the decision. That is not a market outcome. It is a design, it produces the most revenue from the people with the least, and it survives because no individual seller can abandon it without losing.

Two of the three federal attempts to change it are gone, vacated on process rather than substance, while the fees they targeted went right back to where they were. Households absorbed the difference at a wage floor that has not moved since 2009, which is the pattern underneath every price that outran the paycheck and one more line in an American dream that stopped adding up.