The strange reversal: Cash is disappearing, but governments are bringing it back

For a growing number of American retailers, the checkout counter has become a place where two seemingly contradictory economic trends meet. Customers increasingly expect to tap a card or phone, while governments are passing laws requiring businesses to accept the very banknotes that digital commerce is gradually pushing aside. The contradiction is more significant than a debate about whether a shop should accept a $10 bill. It exposes an unresolved question about who should pay for financial inclusion when the infrastructure of everyday commerce is becoming increasingly digital.

Cash is losing market share, but it has not become irrelevant. Meanwhile, accepting electronic payments is no longer a cost-free alternative for businesses. Card networks, issuing banks and payment processors collect fees from merchants, while cash requires physical handling, security, transportation and banking services. For large retailers, these costs can be absorbed across substantial transaction volumes. For a small restaurant, convenience store or independent retailer operating on thin margins, the economics are considerably less forgiving.

The result is an emerging policy conflict. Legislators are increasingly treating cash acceptance as a consumer protection and financial inclusion issue. Yet the businesses required to preserve that access are not necessarily compensated for doing so. The costs of maintaining cash as a viable payment method are distributed across consumers, merchants, banks and the wider payments system, often in ways that remain invisible to the people making the purchase.

Cash is declining, but its disappearance has been exaggerated

The Federal Reserve's 2026 Diary of Consumer Payment Choice provides an important starting point. The annual study, conducted by Federal Reserve Financial Services, found that Americans made an average of 47 payments per month in 2025. Cash accounted for approximately six payments per consumer per month, or 14% of all payments. Credit and debit cards together continued to account for roughly two-thirds of transactions.

The more interesting finding is the stability of cash use. Despite the expansion of contactless cards, mobile wallets and instant payment technologies, cash's share of payments has remained around 14% for three consecutive years. The Federal Reserve's May 11, 2026 announcement described this as a period of relative consistency in consumer payment behaviour. Approximately four out of five consumers had used cash during the preceding 30 days.

That distinction between declining market share and persistent absolute demand matters. A payment method can become less prominent without becoming dispensable. The number of transactions conducted electronically can increase while the number of cash transactions remains relatively stable. Indeed, the 2025 figures suggest that the American economy is adding digital payment activity rather than simply replacing every cash transaction with a digital equivalent.

The generational and income differences are equally important. The Federal Reserve's earlier Diary findings showed that adults aged 55 and older used cash for 22% of their payments in 2023, compared with 12% among younger adults. The 2025 research also identified households earning less than $25,000 and older consumers as groups with comparatively greater reliance on cash.

Cash also serves a purpose that transaction statistics can overlook: it is a backup. The Federal Reserve's 2025 Diary findings showed that nearly two-thirds of cash payments were made by consumers who preferred another payment method. In other words, people who normally use cards or mobile devices still turn to cash when circumstances make it useful. It can provide resilience during technical outages, help consumers manage spending, preserve privacy and allow transactions without access to a functioning bank account.

This creates an important distinction between the decline of cash as a preferred payment method and the decline of cash as essential economic infrastructure. The two are not necessarily moving at the same speed.

The unbanked problem is smaller than before, but far from solved

The Federal Deposit Insurance Corporation's 2023 National Survey of Unbanked and Underbanked Households provides a second perspective. Published in November 2024, the survey found that 4.2% of American households, representing approximately 5.6 million households, had no checking or savings account at a bank or credit union. Although this was the lowest unbanked rate recorded since the survey began in 2009, the number of affected households remained substantial.

More revealingly, 66.2% of unbanked households relied entirely on cash to conduct their transactions. The remaining 33.8% used a combination of prepaid cards and nonbank online payment services. For millions of households, therefore, refusing cash is not simply an inconvenience. It can mean refusing a payment method that the customer actually possesses and can use.

The FDIC's findings also complicate the assumption that financial exclusion is principally a problem affecting younger people. Households headed by someone aged 65 or older had the lowest unbanked rate among age groups, but older households represented 19.4% of all unbanked households in 2023, compared with 10.3% in 2013. This reflects, among other factors, the changing age composition of the unbanked population.

Geography creates another layer of disadvantage. Rural communities have experienced bank branch closures, leaving some residents with fewer convenient opportunities to withdraw or deposit money. The FDIC found that in-person banking remained relatively more important for lower-income, older and non-metropolitan households. Although ATMs and mobile banking have expanded access, they do not eliminate the practical limitations faced by people who have difficulty travelling, using digital interfaces or maintaining reliable internet connectivity.

Cash is therefore not merely an alternative chosen by people who dislike technology. It is part of the economic infrastructure that allows individuals with different levels of banking access, digital capability and financial resources to participate in ordinary commerce.

When the bank disappears, the retailer becomes the cash machine

One of the less obvious developments in the cash debate concerns the changing role of retailers. A grocery store or dollar store can increasingly function as an informal cash-access point, particularly in communities where traditional banking services have become less convenient.

The Consumer Financial Protection Bureau examined this development in its August 26, 2024 report, Issue Spotlight: Cash-back Fees, prepared by its Offices of Consumer Populations and Markets. The report examined how consumers obtain cash back when making purchases using debit and prepaid cards, and how retailers have begun charging for this service.

The CFPB estimated that three major retail companies in its sample collected more than $90 million annually from cash-back fees. Dollar General, Dollar Tree and Kroger were among the companies identified. At the two largest dollar-store corporations, fees for smaller cash withdrawals were at least $1 for amounts below $50. Kroger's Harris Teeter brand charged 75 cents for cash back of $100 or less.

Yet the report highlighted a striking difference between the fees charged to consumers and the incremental processing costs incurred by merchants. The CFPB estimated that the marginal cost of processing a cash-back transaction could amount to only a few cents. The report also observed that cash-back services can reduce the cash held by retailers, potentially lowering some cash-handling expenses.

There is an important economic contradiction here. Retailers are filling a gap created partly by the withdrawal of traditional banking infrastructure, but consumers can end up paying additional fees to access their own money. The CFPB specifically identified lower-income communities, rural towns and communities of colour as places where these fees may be particularly burdensome because banking alternatives are less readily available.

The issue is not that retailers should necessarily provide unlimited cash withdrawals without compensation. Rather, it is that the cost of maintaining access to cash is gradually being transferred from financial institutions to retailers and consumers. When bank branches disappear, the economic responsibility does not disappear with them. It moves elsewhere.

The merchant's dilemma: Cash is expensive, but cards are not free

Cash-acceptance mandates are often discussed from the consumer's perspective. Their effect on small businesses deserves equal attention.

Accepting cash requires a retailer to maintain a till, keep sufficient change, reconcile transactions, secure physical money, train employees and arrange deposits. Businesses may also need to purchase insurance, pay for cash transportation or spend staff time travelling to a bank. The costs become more significant when nearby branches close or when cash deposits are inconvenient.

However, cashless businesses do not necessarily operate with a cheaper payment system. Card acceptance carries its own costs, including interchange fees, network assessments, processor charges and other acquiring expenses. Some processors also impose fixed transaction charges, monthly fees and equipment costs. These expenses can be especially noticeable for businesses processing a large number of low-value purchases.

Federal Reserve research is particularly useful for understanding this imbalance. In her Kansas City Federal Reserve Bank research, Debit Card Interchange Fees Charged to Small Merchants After Regulation II, Fumiko Hayashi examines how interchange costs have evolved for small merchants that do not qualify for volume discounts.

The research distinguishes between different debit-card networks and shows why headline interchange averages can obscure the costs faced by smaller retailers. For exempt debit transactions, the average interchange fee on single-message networks fell from $0.31 in 2011 to $0.27 in 2023. On dual-message networks, however, the average rose from $0.51 to $0.62 over the same period.

The percentage of transaction value represented by those fees moved differently. For dual-message transactions, it declined from 1.44% to 1.41%, partly because the average transaction value increased from $36 to $44. These figures illustrate why comparing only the dollar amount or percentage rate can produce misleading conclusions.

Hayashi also identifies a structural disadvantage for smaller merchants. Large retailers can negotiate volume-based discounts and may have greater ability to direct transactions towards cheaper networks. Smaller merchants generally have less negotiating power and fewer resources to identify and exploit lower-cost routing options. Consequently, the interchange fees appearing in broad industry averages may not reflect the actual economics of a small shop.

The Kansas City Fed's August 2026 interchange-fee update continues to distinguish between small merchants that do not qualify for volume discounts and large merchants that receive the most favourable pricing. The distinction is important because a payment policy that looks inexpensive when measured across the entire retail sector can impose a disproportionate burden on independent businesses.

Consider a hypothetical independent convenience store processing a $5 purchase. A percentage-based card fee combined with a fixed processing charge can consume a noticeable portion of the transaction's gross margin. The same business accepting cash must pay for handling and banking arrangements, but does not incur the same per-transaction card-network charges. Neither payment method is free; the relative cost depends on transaction size, volume, location, banking access and the merchant's contractual arrangements.

This is where the financial-inclusion debate becomes complicated. Requiring cash acceptance protects consumers who cannot easily use electronic payments, but it also requires businesses to maintain an additional payment channel. Meanwhile, abandoning cash may reduce operational complexity but expose the business to card-processing costs and exclude some customers.

How cities and states have responded

The United States has no general federal requirement that private businesses accept cash. The Federal Reserve's explanation of the legal-tender statute makes an important distinction: under 31 U.S.C. Section 5103, US currency is legal tender for debts, public charges, taxes and dues, but that provision does not independently compel every retailer to accept cash when selling goods or services. State and local governments have therefore developed their own rules.

Massachusetts is an early example. Its General Laws, Chapter 255D, Section 10A, enacted in 1978, requires retail establishments to accept legal tender and prohibits discrimination against cash buyers. The law predates the modern smartphone economy by decades, illustrating that the concern about excluding cash customers did not originate with recent cashless retail experiments.

New Jersey and Rhode Island introduced broader cash-acceptance requirements in 2019. New Jersey's Public Law 2019, Chapter 50, prohibits retailers from requiring customers to pay by credit or refusing cash. It provides for civil penalties of up to $2,500 for a first violation and $5,000 for a second violation, subject to statutory exceptions. Rhode Island's General Laws, Section 6-13.1-30, similarly classifies discrimination against cash-paying retail customers as a deceptive trade practice, while excluding online purchases.

Municipal governments have also taken action. Philadelphia's cashless retail prohibition, effective July 1, 2019, prohibits covered retailers from refusing cash, displaying signs stating that cash is not accepted or charging cash-paying customers more. Its rules contain exclusions for certain transactions, including online purchases, parking and particular membership-based arrangements.

San Francisco adopted Article 55 of its Police Code in 2019. It requires covered brick-and-mortar businesses to accept cash for qualifying in-person transactions and prohibits charging fees or imposing additional conditions on cash acceptance. The ordinance includes exemptions, including for mobile businesses, transactions exceeding $5,000 and bills larger than $20. Importantly, a proposal introduced in January 2026 to repeal the requirement remained a legislative proposal rather than an enacted repeal in the available city legislative record.

New York City introduced its own cashless-establishment prohibition in 2020. Covered food stores and retail establishments must accept cash, although businesses can use qualifying cash-to-prepaid-card conversion machines as an alternative under the city's rules.

The geographical scope of the debate expanded significantly in 2025 when Washington, DC's cashless retail prohibition took effect on January 1. DC Code, Chapter 28-54, prohibits covered retailers from refusing cash, displaying signs rejecting it or charging higher prices to cash-paying customers.

Then came an important development in 2026. New York extended the cash-acceptance requirement beyond the five boroughs through a statewide law effective March 21, 2026. Attorney General Letitia James announced that food stores and retail establishments throughout New York State must accept cash. Covered businesses cannot charge customers more for paying in cash, and the law provides maximum civil penalties of $1,000 for a first violation and $1,500 for subsequent violations. Exceptions include bills exceeding $20, certain remote transactions and qualifying cash-to-prepaid-card conversion arrangements.

The differences between these laws are significant. Some jurisdictions require straightforward acceptance of legal tender. Others prohibit additional charges, require equal pricing or allow cash-to-card conversion machines. Exemptions for online transactions, parking, temporary businesses and membership-based retailers also differ. A national retailer operating across several states therefore faces a patchwork of requirements rather than a single uniform payment policy.

The hidden distribution of costs

There is a tendency to describe cashless commerce as a technological improvement and cash-acceptance mandates as a restriction on business freedom. Neither description captures the full economic picture.

Consider the distribution of costs across four groups.

First, older consumers and people with limited digital access. For some, cash avoids the need to manage passwords, banking applications, authentication procedures or digital security risks. Their reliance on cash can also reflect personal preference rather than technological inability. When businesses refuse it, these customers must either adapt or find another retailer.

Second, unbanked and underbanked households. The FDIC's 5.6 million unbanked households represent a substantial population that cannot be assumed to possess the conventional checking-account infrastructure required for straightforward electronic payments. Prepaid cards and nonbank payment applications offer alternatives, but they can introduce fees, restrictions and additional complexity.

Third, rural communities. The closure of bank branches and limited availability of ATMs can increase the importance of retailers as cash-access points. The CFPB's research shows how the decline in traditional banking access can create opportunities for retailers to charge fees for services that were previously available more affordably through banks.

Fourth, small merchants. Independent businesses must maintain payment flexibility while absorbing the costs associated with multiple payment channels. They may face card fees that large competitors can negotiate down, alongside the fixed and operational costs of cash handling. Cash mandates can therefore have a different economic impact on a neighbourhood shop than on a national supermarket chain.

The distributional problem is that the benefits and costs are not necessarily borne by the same parties. Consumers gain access to an inclusive payment system, while merchants shoulder much of the operational burden. Banks can benefit from shifting more transactions towards electronic networks, while the costs of maintaining physical cash access become increasingly concentrated among the remaining participants.

Why cash is also an infrastructure and resilience issue

There is another reason why the debate cannot be reduced to consumer preference. Cash is a form of payment infrastructure that functions independently of many of the systems required for digital transactions.

Card payments depend on networks, processors, telecommunications, electricity and the availability of functioning devices. Mobile wallets add further dependencies, including smartphones, battery power, software and authentication systems. During a network outage, cyber incident or major infrastructure disruption, cash can provide an alternative means of conducting transactions.

The Federal Reserve's Diary research captures part of this role through the continued use of cash by consumers who otherwise prefer electronic payments. Its persistence suggests that cash has a value beyond its ordinary share of retail transactions. It is a form of payment insurance whose usefulness becomes particularly visible when digital alternatives are unavailable.

However, preserving that insurance requires more than instructing retailers to accept banknotes. Currency must remain available through banks, ATMs, cash-handling services and transportation networks. Retailers must have access to affordable deposit services and sufficient change. Consumers must have practical opportunities to withdraw and replenish cash.

A legal requirement to accept cash does not solve the problem if obtaining cash becomes expensive or physically difficult. Similarly, a business can technically comply with a cash-acceptance law while facing rising costs because its nearest bank branch has closed. The availability of cash at the checkout counter depends on an entire infrastructure that extends far beyond the retailer.

Could payment inclusion become another small-business subsidy?

The deeper policy question is whether governments are effectively asking small businesses to subsidise a public-interest objective.

Cash-acceptance laws generally do not compensate merchants for maintaining tills, handling deposits or arranging security. Nor do they address the pricing structures that determine what a small business pays to accept debit and credit cards. They establish a consumer entitlement without necessarily addressing the underlying economics of payment infrastructure.

One possible response would be to distinguish between the obligation to accept cash and the obligation to provide free access to every cash-related service. A retailer could be required to accept reasonable cash payments without automatically being expected to operate as an ATM or provide unlimited cash-back services.

Another approach would involve improving access to affordable cash-handling and deposit services, particularly in rural communities. Banks and financial institutions could be encouraged to maintain shared cash-access facilities or collaborate with retailers where traditional branches are no longer commercially viable.

Payment competition is also relevant. Greater transparency around card-processing charges, improved merchant access to lower-cost debit routing and more competitive acquiring arrangements could help reduce the burden on smaller businesses. The Kansas City Fed's research illustrates why the ability to negotiate and select payment networks matters to merchants of different sizes.

Finally, governments could assess cash-acceptance legislation against measurable outcomes rather than simply counting the number of businesses covered. Relevant indicators would include the availability of free cash access, the geographical distribution of bank branches and ATMs, cash-back fees, merchant payment costs and the number of consumers who experience exclusion at the point of sale.

These measures would help establish whether cash-acceptance mandates are actually improving financial inclusion or simply transferring the expense of preserving cash from one part of the economy to another.

A payment system with two competing definitions of progress

The decline of cash and the expansion of cash-acceptance mandates are not necessarily contradictory trends. They reflect two different definitions of progress.

For technology companies, banks and many consumers, progress means faster payments, fewer physical transactions, improved convenience and greater integration with digital financial services. For legislators concerned about exclusion, progress means ensuring that the ability to purchase food, clothing or everyday necessities does not depend on possessing a bank account, a smartphone or a functioning electronic payment connection.

Both objectives have economic merit. The Federal Reserve's research demonstrates that cash continues to occupy a meaningful position in household payment behaviour. The FDIC shows that millions of households remain outside the conventional banking system. The CFPB reveals that access to physical money can become more expensive when traditional banking infrastructure contracts. Meanwhile, Federal Reserve interchange research demonstrates that electronic payments impose costs that are not evenly distributed across merchants.

The emerging regulatory challenge is therefore not simply to decide whether cash should survive. It is to decide who should bear the cost of keeping it available, and how that responsibility should be shared between financial institutions, payment networks, retailers and government.

A payment system should not be judged solely by how quickly the average consumer can complete a transaction. It should also be judged by whether an older person can buy groceries without a smartphone, whether a rural household can access its money without paying repeated withdrawal fees, and whether an independent retailer can serve both cash and card customers without sacrificing an unreasonable share of its margin.

The quiet collision between digital payments and cash-acceptance laws is ultimately a debate about the distribution of economic convenience. Digitalisation may reduce costs for some participants while increasing them for others. Cash mandates may protect vulnerable consumers while imposing new obligations on small businesses. Unless policymakers address both sides of that equation, the United States risks creating a payment system that is technologically advanced but economically less inclusive than it appears.