Getting over overdraft

Person using an ATM machine.
Editor's note:

This report is based on an article originally published by the Milken Institute Review on October 31, 2022.

Bank account overdraft fees have not always been with us. Overdraft privileges as a paid service became common in the 1990s, when it was introduced by banks as a convenience for account holders who ran out of funds with checks outstanding and would rather have them honored than returned. What began as a modest add-on service to select customers quickly morphed into a profit center for banks (and credit unions), with estimates of total fees paid ranging up to $30 billion a year.

Now, $30 billion is real money even for a banking system as large as America’s: The biggest banks were making over $1 billion a year on overdraft fees, while overdraft income grew to an astonishing 20% or more of earnings for smaller ones. Overdraft fees, effectively interest on loans, are extremely high cost given the small amount of money loaned via an overdraft, the short term of the loan, and the minimal chance of default. As a result, overdraft fees result in nearly pure profit for the bank (or credit union). No wonder one bank CEO named his yacht “Overdraft.”

Every overdraft by definition turns money from someone who has run out of it to revenue for a bank (or credit union). The good news is this reverse Robin Hood is slowing down. After decades of racking up major profits off overdrafts, many banks, including most of the largest banks, have announced sweeping changes that will sharply reduce costs for their customers—by my calculations, the combined savings already announced add up to about $5 billion a year, changes so large that even President Biden noticed and tried to take some credit. But in reality, this turnabout came without new legislation or regulation.

Why? Congress and regulators did put pressure on banks to change their ways. Sen. Chris Van Hollen (D-MD) prodded the Comptroller of the Currency, the agency that regulates national banks about overdrafts. Sen. Elizabeth Warren (D-MA) confronted JP Morgan Chase CEO Jamie Dimon, pointedly asking why his institution earns seven times as much in overdraft revenue as comparably sized Citibank. Rep. Caroline Maloney (D-NY) repeatedly introduced legislation that would force sweeping changes to overdraft policy, although it never came close to enactment. Meanwhile, the Consumer Financial Protection Bureau published research highlighting the overdraft bonanza’s magnitude and who’s paying for it. New financial technology (fintech) firms began offering and marketing products providing consumers banking accounts without overdraft and sometimes with different forms of less expensive small dollar credit. But these entrants are tiny compared to the U.S. banking system.

It’s hard to say whether banks feared new regulation, new legislation, bad publicity, competition, or had bigger fish to fry with their overseers. Whatever the reason, the dam burst. The largest banks are planning to cut overdrafts by about half from 2019 levels. This is not the end of the story, though: some banks’ changes are more meaningful than others. Here, I offer a closer look at what’s happening, in particular how it affects lower-income households, and suggest ways Congress and financial regulators could and still should intervene.

Understanding Overdrafts

One key take-home is that 80% of overdraft fees come from just 9% of account holders. Heavy overdrafters are highly profitable customers, often producing more income for banks than more affluent customers who may use other paid bank services but always maintain positive balances. Oliver Wyman consultants estimated that heavy overdrafters on average generated $720 a year in profit for their basic bank accounts while non-overdrafters yielded a measly $57.

To compare banks of widely disparate size, I analyzed overdraft revenue per consumer account excluding retirement accounts. The numbers are striking. Some banks generate overdraft income at a rate more than seven times those of others, which seems unlikely to reflect differences in the care taken by account holders to remain liquid. The explanation is typically buried in the fine print that hardly anyone reads or back-office practices by banks that only regulators know about (if they bother to look). For example, some banks post debits before credits, triggering overdraft fees, while some allow overdrafts at ATMs rather than simply denying withdrawals exceeding account balances.

Bankers Healing Themselves

The good news, of course, is that all the giant banks and many smaller ones have pulled back on overdraft fees. Consumers’ savings from this subset of 14 banks—which includes the eleven large banks included in the above graph plus three others that have announced changes (USAA, Ally Bank, and Frost Bank)—should be about $5 billion a year (see endnote 1 for information on calculations in this section). This estimate, incidentally, is more inclusive than the Consumer Finance Protection Board’s estimate of the savings from banks eliminating non-sufficient funds fees, which the agency estimated will save consumers $1 billion annually. It also includes institutions that have made announcements since an earlier estimate by the Pew Charitable Trusts, which found potential savings of $2 billion a year based on changes from only the five largest banks.

A breakdown by individual bank explains the large differences in the impact of the changes. Note that Citibank, Capital One, and Ally collected relatively little in overdraft fees before the announced changes. So it isn’t all that surprising that they chose to finish the job, flat-out eliminating overdraft fees. Bank of America similarly stands out in that even though it was making over $1 billion a year in overdraft fees prior to changes, they have so many accounts that on a per account basis they were on the smaller end. Their decision to decrease the fee per transgression from $35 to $10 coupled with other changes will eliminate around 90% of their overdraft revenue.

At the other end of the spectrum Regions Bank and USAA appear to have done the least among big banks, with projected revenue declines on the order of 20 to 25%. This is even more concerning given that Regions had among the highest overdraft revenue per account in 2019 and was just fined $191 million by regulators for illegal, surprise overdrafts.

The details of how banks are reducing overdrafts sheds light on the factors that drove the overdraft bonanza in the first place. Here, I break it down into four “buckets.”

Reducing fees per incident

Overdrafts had generally been priced at about $35 each, with institutions setting a maximum number of daily overdrafts (often between four and eight) as they covered cascading shortfalls for a stiff price. Charging penalty fees for overdrafts may have been designed at one point to reduce their frequency but given the illiquidity of many of their customers at the moment they temporarily run out of money, it became an easy way for banks to turn small fish (small-balance accounts) into big bucks (small balance accounts that generate big profits). Given the high cost and lack of time to cover account shortfalls by other means, the relative standard practices across the industry, a lack of alternative products, and consumers’ frequent lack of awareness that they were even overdrafting, overdraft fees didn’t seem to move customers to leave their bank to find a better deal.

Many banks also charged a non-sufficient funds fee (NSF) for some accounts, refusing payment when an account was overdrawn rather than covering the gap with an expensive automatic overdraft loan. NSF fees tended to be around the same size as overdraft fees. Most of the largest financial institutions have now eliminated NSF fees entirely, while others have chopped them. Some banks have also reduced the maximum number of overdraft fees charged per day, limiting a consumer’s total exposure in cases in which a cascade of small checks bounce because the account holder miscalculated. These changes are straightforward and reduce costs borne by consumers.

Changing timing

Overdrafting is more about running out of time than out of money—people are often minutes or hours away from having the money necessary to cover the overage. Some customers have positive balances when they make a purchase, but because of the time delay in clearing a deposit, the balance turns negative when the purchase clears. This results in a “positive when made, negative when settled” scenario that—no surprise—enrages consumers.

This timing problem is exacerbated by America’s antiquated payments clearing system, which runs on decades-old technology. For one, payments are often credited and debited in batches rather than individually when they occur.

A batch system is analogous to a washing machine in which all the clothes go in together regardless of when they were soiled and come out clean at the same time. The person doing the laundry then decides when to fold and return the clean clothes, much the way a bank has some discretion on which order to post the various debits and credits that come through the payment cycle. And the debits have a habit of being folded and shelved before the credits.

Some banks have now created grace periods in which consumers who cover an overdraft within a day or two are not charged a fee (PNC, Wells Fargo). In addition, many banks have put electronic deposits of wages on the fast track, crediting direct deposits up to two days earlier (Capital One, Regions).

Direct deposits do not clear instantly. Typically, a direct deposit paycheck written on an employer’s account on Tuesday does not become available to the worker until Friday. But banks with direct deposit relationships often know the amount of money their customer will receive and, if they choose, are able to safely provide access to those funds earlier. And some banks have eliminated overdraft fees incurred if a charge was made when the account still had funds but settled negative (JPMC).

PNC, which was among the first banks to change overdraft fees, has been able to collect some data from their changes which they term Low Cash Mode. Some 63% of PNC customers who end the day with a negative balance are able to fix the problem and avoid an overdraft. The average time to “cure” is only 13 hours, evidence that the majority of their customers’ problems are very short-term mismatches between payments and deposits. From PNC’s experience, 75% of their reduction fee income was the result of extra time and the change on the limit on total overdrafts. The remaining 25% came from the elimination of NSF fees.

This helps explain the popularity of early wage access and other faster payment options spreading through the banking and fintech systems. It also makes clear the incredibly high cost of our nation’s slow payment system that weighs heavily on families living paycheck to paycheck. The failure of the Federal Reserve to speed up transaction clearing has taken billions out of the pockets of working families and stuffed it in the bottom line of banks, credit unions, check cashers, and payday lenders.

Small dollar liquidity credits

In economic terms, an overdraft is a form of small-dollar credit. Charging a fixed price (a fee) instead of interest does not change that basic fact. But the courts and the regulators have deemed overdrafts to be fees instead of loans, thereby short-circuiting legal requirements like Truth in Lending that requires disclosures, including the annual percentage interest rate (APR). APRs for overdrafts may or may not be a useful concept. But they would appear astronomical in cases of small overdrafts: One story in the Dallas Morning News reported a $100 fee for covering an overdraft of two cents.

Most banks that have become more consumer-friendly have increased the amount a consumer can go negative without incurring a fee. Many have raised their limits from $5 to $50 (US Bank, Huntington, TD, and JPMC) while some have gone as high as $100 (Truist and Frost Bank).

Another common remediation has been to automatically convert negative balances into installment loans rather than charging a penalty fee. These loans typically still have a fixed charge for the amount borrowed. U.S. Bank offered a similar product (Simple Loan) some time ago for which the banks now charge $6 per $100 borrowed. The loans typically last a few months and are paid back in even, amortizing (i.e., self-liquidating) payments. Institutions typically make repayment automatic but say they will not take such payment from the account if it triggers yet another overdraft.

Changing from a fee-per-transaction when a customer’s balance is negative into a loan where costs are based on amount borrowed rather than the number of transactions is a win for consumers. It is a more honest and transparent product for the lender as well, as the costs/risks of default are related to the total amount borrowed, not the number of transactions.

Consider, too, that separating the cost of automatic installment credit from the time horizon of the loan is simpler for consumers to understand than an interest schedule. Fees on the order of 5% of amount borrowed are substantially lower than most alternatives available to heavy overdrafters for small-dollar credit.

Forgiving temporary negative balances is different than converting the negative balance to a loan. And it’s worth noting that savings to consumers from changes to overdraft fees will be somewhat offset by the costs of small dollar lending. So a full accounting of total savings from overdraft fee changes should include the corresponding costs associated with small dollar credit products that are being rolled out as alternatives to overdraft fees.

Consumer empowerment

Giving consumers advance knowledge of low balances as well as flexibility to stop or delay an automatic payment that puts them in the red would empower consumers to decide whether paying an overdraft was the better alternative. And to their credit, many banks have developed sophisticated systems to alert consumers of low balances in time to stop payments (PNC, TD). Some of these systems, for example, alert to customers when their balances reach a threshold (as in “$50 left in your account”), while others indicate an automatic payment is coming that would force an overdraft.

Consumers can then use this information to decide how to manage their finances and potentially avoid an overdraft. Note, however, that the decision may be more complicated than it first appears. Cancelling an automatic payment may itself result in fees from, say, a credit card company or a car lender. Banks making changes to their policies cannot be responsible for how a third party will respond to overdue payments.

While consumer empowerment sounds good, it may not have much impact. PNC estimates that only about 1% of payments were cancelled or delayed by customers receiving low-balance warnings (see endnote 2). This may be evidence that customers want these payments to move forward regardless of overdraft consequences. Or that they know they will have enough money to cover the payment, given that PNC now allows extra time to cure an overdraft.

Consider, too, that information without the ability to fix a problem is of limited use. The problem people living on the financial edge face with overdrafts is more a combination of temporal mismatches of money and the high cost of small dollar credit than it is about knowing that they are near the edge. Data from the Financial Diaries Project indicates that people living paycheck to paycheck may be more likely to budget and be aware of their finances than those who are comfortably upper middle class. The lack of a real-time payment system further complicates the value of information: when you do not know the precise moment your paycheck will be credited to your bank account or when a payment will be debited, it is impossible to budget or plan in a way to avoid fees.

What’s Government’s Job Here?

The banks who have tempered the “gotcha” aspect of low-balance banking without orders from lawmakers or regulators should be commended. It is not easy for a company to change in a way that reduces its immediate profits but improves the lives of its customers.

Doing the right thing is wonderful, but the reality is that not every bank will or even can. The more a bank depends on overdraft revenue the less likely it is to give it up without a push. Even today, a handful of banks and credit unions operate on business models that require a lot of overdraft revenue for their viability.

First National Bank of Texas, to take one example, has made more than 100%of its profits from overdraft fees in each of the last seven years—and that’s as long as overdraft data have been separately reported. For two other banks, Woodforest and Gate City, that has been true for six of the last seven years. Armed Forces Bank, a private bank exclusively serving current and past military, has made more than 75% of its profit on overdraft fees for each of the last seven years (and over 100%for three of the seven). Academy Bank made more than 100% of its profit in overdraft fees for four straight years from 2017-2020.

I group Armed Forces and Academy Bank together because they are owned by the same holding company, Dickenson Financial Company. The Federal Reserve regulates the holding company, while the Office of the Comptroller of the Currency (OCC) regulates their banks, which are nationally chartered.

Regulators have been asleep at the switch in allowing these banks to operate with what are clearly unsound business models, as they have been losing money every year on all aspects of banking other than overdrafts. And there may well be more overdraft addicts, as banks under $1 billion in assets and all credit unions are exempt from publicly disclosing their overdraft revenue.

The explosive growth and popularity of overdrafts as a profit center reveals deeper structural problems with America’s basic banking system. Slow payments, limited options for small dollar liquidity and fees designed to be punitive rather than linked to actual costs are core reasons why overdrafts became so widely used. The biggest losers are the working poor who can least afford to lose. Consider, too, that low-balance woes also drive people out of the banking system entirely, greatly adding to how expensive it is to be poor.

The solutions are relatively straightforward. I think any financial institution that relies on overdraft fees for the bulk of it profits for multiple consecutive years should be given failing regulatory grades—a position the Washington Post has echoed. Credit unions should publicly disclose overdraft revenue, just as banks do.

Wait, there’s more. America’s payment system needs to move in real time. The Federal Reserve, for example, has the legal authority to require the first $5,000 of every check deposited be available immediately. If the Fed won’t (and trust me it won’t), Congress must. As we’ve seen, a series of tricks allow some banks and credit unions to increase overdraft revenue in part by taking advantage of the slow payment system. Two of these can be ended through joint regulation: posting debits before credits and reordering payment flows from largest to smallest.

Finally, all financial institutions should be required to offer a no-overdraft, low-cost, basic bank account. These accounts have proven popular when properly marketed—Citibank reports that one in five new customers is opening one—and it can be done in a way that is profitable for the financial institution. Interestingly, the bank lobby also likes these types of accounts: The American Bankers Association calls it a best practice for all banks to offer this type of account.

Overdraft fees may be on a downward arc, but they remain a serious drain on millions of Americans living on the financial edge. We know that the financial system works well for the affluent. But we need to redesign the system to discourage practices that have turned the poor and near-poor into a profit center.

The Brookings Institution is a nonprofit organization devoted to independent research and policy solutions. Its mission is to conduct high-quality, independent research and, based on that research, to provide innovative, practical recommendations for policymakers and the public. The conclusions and recommendations of any Brookings publication are solely those of its author(s), and do not reflect the views of the Institution, its management, or its other scholars.

JP Morgan Chase & Co., Wells Fargo & Co., Bank of America, Citi, TD Bank, and Capital One Financial Corporation provide support to The Brookings Institution. The findings, interpretations, and conclusions in this report are not influenced by any donation. Brookings recognizes that the value it provides is in its absolute commitment to quality, independence, and impact. Activities supported by its donors reflect this commitment.


  • Footnotes
    1. The estimates of decline in overdraft revenue collected are based on the authors calculations using a number of data sources, including data from American Banker, bank announcements, and personal correspondence. For more information on the data and methodology, contact the author at [email protected].
    2. Data provided to author, available upon request.
    3. Data on banks’ overdraft revenues, profits, and number of accounts is based on call report data generated on December 31 of a given year from the Federal Financial Institutions Examination Council Central Data Repository’s Public Data Distribution.
    4. Data from author’s coorespondence with Citibank.