Household money leaks in 2026: nine spots federal data flags

Most lists like this one open with coffee. The federal data says the money is somewhere else entirely. The Federal Reserve put the rate on credit card accounts assessed interest at 22.15 percent in the second quarter of 2026. Ordinary savings deposits paid 0.38 percent nationally on the FDIC’s August reading. The BLS put twelve-month inflation at 3.4 percent through July, with energy up 14.7 percent over the year.

Those four numbers point at where household money actually leaks in 2026, and none of them is discretionary spending. What follows is ranked by the size of the lever rather than by how easy it is to write about, with the arithmetic shown so you can run it against your own figures. No total is promised, because the total depends entirely on which of these apply to you.

1. The revolving balance, if there is one

Nothing else on this list is in the same weight class. The Fed reports two credit card rates and the one that applies to anyone carrying a balance is the higher one: 22.15 percent on accounts assessed interest in the second quarter of 2026, against 20.94 percent across all accounts.

Round illustrative numbers chosen to show the mechanism, not market averages. A $5,000 balance held for a year at 22.15 percent costs roughly $1,100 in interest. Every other item below would have to work very hard to match that single line, which is why the order here is not negotiable: this one comes first or the rest is decoration.

2. Repricing that balance rather than just paying it

Commercial banks charged an average 11.86 percent on two-year personal loans in the same quarter, according to the same Fed release. Set beside the card figure, that is close to half the carrying cost on identical money, and it arrives with a scheduled payoff date rather than an open-ended minimum.

Approval and pricing depend on the borrower and are not guaranteed, so treat the gap as the size of the prize rather than as a quote. The related move, a balance transfer, works on different arithmetic that turns on a fee and a deadline; we covered where that breaks even separately.

3. Late fees, which are a scheduling problem wearing a cost

Late fees came to $17.0 billion across the market in 2024, on the CFPB’s count, and they do not land evenly. Two fifths of that total was generated by the deep subprime tier, which holds roughly a seventh of all accounts. At the opposite end, the 59 percent of accounts belonging to superprime and prime plus cardholders accounted for 7.8 percent of the fees.

Read the unevenness as good news for anyone on the wrong side of it, because the fix is administrative rather than financial. Autopay for the minimum on every account, set for the day after payday, converts the entire category to zero without changing what anyone spends. Pay more manually on top; the autopay exists only to make the missed-date outcome impossible.

4. Subscriptions, and the number that proves the point

Here is the most revealing statistic in the CFPB’s 2024 dispute data. Cardholders disputed $9.8 billion in charges that year, and for general purpose cards the single most common reason was a cancelled recurring transaction, meaning subscriptions, membership fees and utility bills. That category made up 40 percent of all disputes.

Forty percent of every credit card dispute in the country is somebody being billed for something they thought they had ended. Forgetfulness explains very little of that. Cancellation flows are built to be harder to finish than signup flows, and the dispute volume is what that design produces at national scale. It also means an audit of recurring charges returns more per minute than anything else here.

Pull three months of statements, list every repeating charge, and cancel by contacting the merchant directly rather than by deleting an app. Then confirm the cancellation lands on the next statement, because the dispute data says that step is where it fails.

5. The insurance renewal nobody reads

The NAIC found average homeowners premium per policy rising in every region between 2018 and 2024, with inflation-adjusted increases of 18.3 to 43.3 percent. Auto policies reprice on a six-month cycle in much of the market. Both are contracted costs that renew silently and are quoted competitively only to people who ask.

The mechanism is worth naming plainly, because it is the same one that governs item six. Firms price to the customer who does not shop, and the discount is available to anyone who does. This is the closest thing to free money on the list, and it costs a phone call per policy per year.

6. The cash reserve earning 0.38 percent

Savings paid 0.38 percent nationally in the FDIC’s August 2026 figure, a number that has not moved since April. Measured against 3.4 percent inflation, a balance parked there gave up around three points of purchasing power across the year.

Hold $20,000 in cash and the spread between that national figure and the better insured offers available is worth several hundred dollars over twelve months. The full version of this calculation, including the tax adjustment that most articles skip, is in our note on savings rates against inflation.

7. Annual fees on cards nobody is using

Roughly one general purpose account in six came with an annual fee during 2024. The average was $127, having been $105 two years earlier, and the CFPB puts the increase down to premium products built around travel perks and monthly service credits rather than any broad repricing.

A statement credit is worth its face value only to someone who was buying that service anyway, and nothing at all to anyone else. Run the fee against last year’s actual redemptions rather than this year’s intentions. Where it does not clear, ask the issuer about a product change to a no-fee card in the same family, which generally preserves the account age that closing it would destroy.

8. Energy, the category that actually moved

The BLS energy index rose 14.7 percent over the twelve months ending July 2026, against 3.4 percent for all items and 2.5 percent for all items less food and energy. Nothing else in the basket moved like that.

Which means the household budget line worth attacking this year is the one the averages hide. Where the state permits supplier choice, the supply rate is separable from the delivery charge and comparable. Where it does not, utility efficiency programmes and budget billing are the available levers. Either way this is a category where a 14.7 percent move justifies an hour that a 2.5 percent move would not.

9. Groceries, ranked last for a reason

The food index rose 3.0 percent over the year to July 2026, roughly in line with the headline. Food is where almost every article like this starts, and on the federal numbers it is neither the fastest-moving category nor the one with the largest structural inefficiency.

It is worth attention because it is large and because substitution genuinely works there. It is ranked ninth because an hour spent on items one through four is worth more, and because grocery advice is the part of personal finance writing that is most about appearing frugal and least about arithmetic.

What is deliberately not on this list

Three staples of the genre were left off because the numbers do not support their usual position.

Small daily purchases are the most written-about and among the least consequential; the arithmetic that makes them look dramatic works by annualising a habit while never annualising anything else. Cashback and rewards optimisation belongs to households that already clear the statement balance every month, which the CFPB puts at 43 percent of cardholders, and for everyone else the interest rate swamps whatever the card pays back. And side income, whatever its merits, is a different category of decision from any of the nine above, since every item here is recoverable without changing how anyone spends their week.

None of those is bad advice. They are just misordered in most of the writing on this subject, and ordering is the entire value of a list like this one.

The order matters more than the list

Real average hourly earnings fell 0.2 percent between July 2025 and July 2026, which is the context for all nine. When pay is losing ground in real terms, the levers that do not require earning or spending differently carry more weight than usual.

Work down the list in order and stop when the remaining items stop being worth the hour. For most households that is somewhere around item five. Anyone whose situation starts at item one should read our order of operations on debt first, because sequencing there is worth more than any single tactic here.

Written by

Keith Guirao

Founder & Editor, ConsumersWeek

18+ years in consumer marketing and lead generation across insurance, personal finance, and home services. ConsumersWeek explains how these products are priced and sold so you can evaluate them with the same information the industry has.

About the editorLinkedIn