ATM Fees: Does Size Matter?
Stavins, Joanna, New England Economic Review
The debate on automated teller machine (ATM) fees heated up recently, when voters in San Francisco and Santa Monica approved bans on banks' charging non-customers for using their ATMs. A federal court ruling has temporarily barred the two California cities from enforcing the bans, however. Connecticut and Iowa used existing laws to ban such ATM fees, but the Connecticut Supreme Court rejected the ban in that state. The Pentagon has said it would consider a ban on ATM fees on U.S. military bases. So far, the Congress has rejected legislation that would eliminate non-customer ATM fees nationwide.
ATM networks have allowed banks to charge non-customers for withdrawing money from their ATMs since 1996, but ATM fees have been criticized repeatedly by consumer advocates and politicians and the California bans may soon be copied in New York and elsewhere. Large banks  have been especially targeted, because they are more likely to impose the fees and their fees tend to be higher than those charged by small institutions. Critics of ATM fees call large banks greedy. Supporters argue that the fees represent the cost of convenience, and that consumers are willing to pay for being able to withdraw money anywhere, and not just at their own institutions. Large banks' ATMs are more convenient, because they allow access to cash at more locations.
Surveys comparing ATM fees across financial institutions have shown that large banks' fees exceed those charged by small banks (Board of Governors of the Federal Reserve System 1998 and 1999; U.S. General Accounting Office 1998). However, most surveys do not control for differences in quality among banks of various sizes. ATM fees are prices charged for a service--typically a cash withdrawal. The more machines a bank has, the more convenient it is for cardholders to withdraw cash. This article analyzes differences in ATM fees among banks in order to test whether large banks impose higher ATM fees than do small banks, controlling for some quality and cost differences associated with more ATMs.
The article is organized as follows. The next section describes ATMs and the costs banks incur to operate them. Section II analyzes network externalities and their relevance for the ATM market. The following section discusses the sample used in the study. Section IV uses regression analysis to test whether larger banks charge higher ATM fees, controlling for their greater convenience and costs, and Section V concludes.
Although most banks still allow cash to be withdrawn from a bank teller, ATMs have become an increasingly important way to access cash. The number of ATMs in the United States has risen steadily since the early 1970s, although exact estimates vary depending on the source. Table 1 shows American Bankers Association (ABA) statistics indicating that the number of ATMs in the U.S. reached 227,000 in 1999. Some of the growth in the number of machines, especially since 1996 when two major networks lifted their ban on ATM surcharges, has occurred in ATMs located off bank premises, in places such as airports and convenience stores. After a period of rapid increase, the number of ATM transactions increased at a declining rate, and the number actually dropped in 1999. The number of transactions per machine has declined every year since 1995. The recent public opposition to surcharges, combined with high rental costs for off-premise machines, has led analysts to believe that the ATM market is saturated (Keenan 1998).
ATMs have been widely recognized as a convenient way to obtain cash. With the majority of ATMs connected to regional or national networks, cardholders can withdraw cash from most institutions in the country. At the same time, banks have regarded ATMs as a way to lower their costs, as customers substitute ATM transactions for costly live teller use. To induce customers to use ATMs instead of live tellers, some financial institutions impose fees for teller use or reduce monthly charges to depositors who use only ATMs (Stavins 1999). …