A small-business owner in Australia pays ten employees from two accounts on CommBiz, Commonwealth Bank's business banking platform. The fee schedule says one charge per payment: 25 cents, plus tax. The statement shows twelve charges. Two of them are for moving money between the owner's own accounts. No line in the terms covers them.
The schedule was not wrong. It was not complete. And the economics literature has a name for why.
The Case and the Mechanism
Most operators who run payroll through a business bank have seen a fee schedule. It lists the cost per transaction. It looks finished. You sign, you pay, you move on.
Commonwealth Bank told its CommBiz customers they would pay one fee per direct debit or direct credit. That was the schedule. The charges that showed up were different.
Ten fees for the payments. Two more, without notice, for the money moving between the owner's own accounts. CBA called these "contra leg" fees. The terms did not.
Australia's corporate regulator, ASIC, filed civil proceedings against the bank. The suit covers roughly 42,000 business customers, most of them small operations. The total: $8.5 million in charges that did not match the published terms. Those fees had been running since the platform launched in 2006. Commissioner Alan Kirkland put it in one line: "Small businesses are busy enough. They should not have to scrutinise every transaction to make sure that their bank is charging them the right fees."
CBA's own 2016 internal audit rated the controls around these fees "red," not fit for purpose. The audit required corrective action. The bank did not take it. The charges kept running for years after that audit landed. When pressed, CBA said the issue arose from changes "intended to simplify" the terms. The simpler terms hid the fee. That is what a shrouded cost looks like when the institution describes it in its own words.
In 2006, economists Xavier Gabaix and David Laibson published "Shrouded Attributes" in the Quarterly Journal of Economics. Their finding: when enough customers do not check, firms have no market reason to disclose add-on costs. Even in a head-to-head competitive market. They named bank fees as a primary case. The model predicts that if enough people skip the per-line review, the hidden price stays in place. Not because anyone is careless. Because each charge is too small to trigger a call.
The median payout in CBA's remediation program was about $85. That is not a rounding error. That is a price point built to sit below the threshold where you would pick up the phone. CBA booked a record $11 billion profit last financial year. The $8.5 million did not build that number. But the structure that made those fees stable, a schedule that looks complete but is not, runs in every bank that serves small business.
The Structural Flaw
This is not just an Australian pattern. A 2025 study by the Centre for Economics and Business Research found that U.S. small businesses lose $153 billion a year to fees they did not expect. That is $4,400 per business per year. The equilibrium Gabaix and Laibson described runs wherever a fee schedule is built to look finished without being finished.
I have looked at my own bank statements and skipped the per-line detail more times than I can count. Most operators do the same. The monthly total looks close enough. You move on. That is the bet the fee structure is built around.
Most people treat this as a trust problem. It is a system flaw. A schedule built to look whole is not the same as one that is whole. The gap between the two is where the equilibrium lives. It holds because checking every line costs more time than any single hidden fee is worth. Gabaix and Laibson proved that this math keeps the hidden price in place even when banks compete for your business. The incentive to disclose does not exist as long as enough customers skip the review.
The Audit the Equilibrium Predicts You Will Not Run
The fix is not outrage. It is a manual count, the one the equilibrium predicts you will never run. Once that is clear, three moves follow from it.
Move 1: Pull 90 days of statements and count every fee line
Not the monthly total. Every single line. Mark each charge with a pen. Most operators have not done this in years, if ever.
That count will cost you an hour. It will also show you whether you have been paying for a schedule or paying past it.
Move 2: Match each fee to a named line in your current terms
Open the terms that are live right now. Not the ones you signed when you opened the account. If a fee on your statement has no match in the current terms, mark it. That is the gap.
Move 3: Add up the yearly cost of every fee with no match
Take the 90-day total and multiply by four. If that number tops $100, bring it to your banker. That is not a complaint. It is a data point.
What the Audit Shows
Running this once does something the fee schedule never did.
It shows you which charges are real costs and which are drift. It tells you whether your bank updated its terms to match its billing, or moved the billing without telling you. It puts a number on the gap between what you agreed to and what you pay. It turns the next call with your bank into one backed by a figure on paper.
Three Questions at the End of One Quarter
→ Which fees matched the terms exactly?
→ Which fees showed up with no match and no prior notice?
→ Which charges repeated every cycle without you catching them?
That is the gap between trusting a schedule and testing one.
Where You Stand
The owner looking at twelve fees on ten paychecks was not sloppy. The schedule was built to look finished. It was not. You now know the name of the equilibrium that keeps it that way, and you know the one-hour audit that breaks it.
