The development of terminal-based EFT delivery systems in the eighties
Richard R. Dart
Editor’s Note: The author is product manager of Trans Data Corporation of Cambridge, Maryland. As product manager, he is responsible for assessing the impact of various retail EFT technologies on the financial services industry. Also, he has overall responsibility for product development and industry relations within his firm’s retail EFT program.
The views expressed in this article are those of the author, and do not necessarily reflect the views of the Federal Home Loan Bank Board.
Although the fundamental business of banking continues to revolve around deposits and loans, the manner in which banking services are being delivered is undergoing unparalleled change. The development of extensive brick and mortar branch networks throughout the Sixties and Seventies is now being augmented and, in some cases, displaced by another, more revolutionary concept: the development of terminal-based electronic funds transfer (EFT) networks through which a host of consumer financial products can be offered. Consequently, just as branching provided the vehicle for effectively reaching consumers and facilitating banking transactions, one of the primary factors that will determine an institution’s competitive stature in the technology-oriented Eighties will be the ability to offer a wide distribution of conveniently accessed terminal services. Savings and loans must rethink their business in light of this changing technology and the opportunities inherent in the age of electronic banking.
Even though numerous opportunities exist, the S&L industry is faced with one of its more critical challenges in recent years. Along this line, deregulation of the financial services industry has created a highly competitive environment in which both the number of players and services offered has increased substantially. Banking is no longer a closed profession in which chartered members are confined by geographic boundaries and insulated by regulation. Deregulation has effectively redefined the products financial institutions can offer, how they can package and promote their services, and against whom they must compete. At the same time, changing social conditions and consumer preferences have created a demand for convenient services. Evidence of this demand can be seen in the striking trend toward self-service in everything from pumping gasoline to executing financial transactions. The combination of these conditions has forced financial institutions to invest in electronic banking services and delivery systems in an attempt to defend their market position and compete in an increasingly volatile business.
To meet these challenges, S&Ls must develop cohesive EFT strategies that will position them to compete successfully in a rapidly changing financial services environment. Although EFT has had an impact on a number of areas of banking, its most substantial growth within the past five years has been in the retail sector. Within this sector, S&Ls are essentially faced with four service options from which to develop EFT strategies: automated teller machines, point-of-sale (POS), telephone bill payment (TBP), and videotex/home banking services. Automated Teller Machines
The demand for convenient banking facilities that are accessible on a 24-hour basis has made automated teller machines (ATMs) the major focal point of EFT develpment. The response from consumers has generally been overwhelming, particularly in areas where conventional banking facilities cannot always be accessed, such as hospitals or work areas. Despite the importance of ATMs, however, some segments of the financial industry continue to lag substantially in terms of development, and the competitive gap that now exists can be expected to grow even wider. Along these lines, Trans Data Corporation’s (TDC’s) latest ATM survey of financial institutions, conducted in the second quarter of 1983, reveals that approximately 35 percent of the large S&Ls (deposits greater than $500 million) and 15 percent of the smaller S&Ls (deposits between $50 million and $500 million) currently support ATMs. An equal share of these institutions reported they were planning to offer these services, which suggests that close to 40 percent of the entire savings and loan industry with deposits greater than $50 million may be expected to have ATMs by the end of 1984.
Nevertheless, the growth of ATMs is far more prevalent among commercial banks. TDC survey results reveal that the vast majority of commercial banks (88 percent) with deposits greater than $500 million have installed ATMs, while 60 percent with deposits betweer $50 million and $500 million have done so. Taking into consideration those commercial banks that plan to offer these services, over three-quarters (78 percent) of this industry segment are expected to support ATMs by the end of 1984. In short, while commercial banks are quickly approaching saturation levels with regard to ATM development, the S&L industry generally lags far behind in terms of the overall number of current offerers.
Clearly, the S&L industry will have to follow a steep ATM growth and development curve to close the competitive gap that currently exists. For many S&Ls, the question is not whether to offer ATMs but, rather, how to best participate in the ATM environment and what avenues or options are available. The manner in which S&Ls participate in the ATM arena is particularly critical today because the ATM industry is in the midst of a secondary phase of development–a phase that promises to radically change the competitive relationship among those offering the service. This phase, of course, is the rapid movement from proprietary ATM systems that have dominated the past decade of development to participation in shared ATM networks.
Currently, about half of the S&Ls that support ATMs are members of a shared network, while some one-third of the remaining institutions already have plans to join a shared system. Moreover, almost 65 percent of the S&Ls hoping to introduce ATMs for the first time plan to do so through membership within a shared network. Taking into consideration the substantial development of shared systems on the part of commercial banks, there is an emerging concensus among the financial industry that the shared networking concept is the direction to take in developing competitive ATM strategies.
Although there is little debate as to the viability of shared ATM networks, S&Ls must decide how they will participate in this environment. Essentially, participation in shared networking has been based on two levels of involvement. On one hand, some of the largest institutions have started and sponsor their own ATM systems. For these institutions, owning and operating a system potentially provides lucrative opportunities for generating revenues from transaction processing fees, etc. Moreover, these institutions are in many cases positioned to control the financial and membership terms associated with joining a shared network. As a result, the fate or success of the network resides with the network sponsor, and not with a third party or another institution providing network services.
Despite these potential advantages, developing and sponsoring a new network require a substantial capital investment, particularly in the processing hardware necessary to support a viable system. In addition, most networks begun by financial institutions have initially operated with a substantial base of terminals, the majority of which has been provided by the network sponsor. Only the largest S&Ls have the internal resources to support a competitive shared network and, even then, the substantial lead-time that established networks enjoy might prove to be insurmountable to any new network. Most S&Ls therefore, will play a participating role in the shared networking game, rather than owning or controlling the system.
Institutions participating in a shared network enjoy a number of advantages not found in a proprietary systme. The combined terminal base of a shared network almost always is more extensive and better positioned than that of a single institution. Shared network also are often promoted by all the participants with a collective marketing thrust focused on a single delivery system. Most importantly, the established switching capabilities and the internal capacity among participant within a network to interface on-line with other financial institutions provide the basis for quickly moving into other EFT areas, such as videotex or point-of-sale.
Clearly, few S&Ls can afford to weather the competitive pressures of shared networking. Moreover, several institutions report moving toward shared networks, in part, because of their concern with the growing number of nonbank competition. Safeway supermarkets, for instance, has developed its own ATM network. Although the network is open to any financial institution that wishes to participate, Safeway’s primary market remains smaller banks and thrifts that do not have the capital resources to support their own terminals or develop extensive ATM networks.
Faced with these options, choosing the right network will become one of the more critical decisions S&Ls must make. Along this line, one of the most important points to consider in selecting a shared network is terminal placement within the network. This is important for two reasons. If an institution’s ATM system in relatively new, it most likely will want a network with a good distribution of off-premise terminals to increase ATM access for its customers. Institutions can expect greater off-premise terminal access because off-premise sites generally produce more transactions than on-premise sites. TDC’s latest survey, in fact, reveals that on-premise terminals average almost 2,500 transactions a month, including balance inquiry, compared with an average of over 3,600 a month for off-premise sites. On the other hand, for those with a more mature system with high yield off-premise ATM sites (such as retail centers), it is more beneficial to join a network whose terminals are concentrated in on-premise locations. This is because off-premise terminals most likely will generate disproportionately high transaction volumes. AS a result, network interchange fees will be charged back to other network participants. In any event, financial institutions must look closely at whether a network allows participants to designate some of their terminals as open access ATMs, while prohibiting network access to others. Today, over 10 percent of the networks limit access to certain terminals, primarily on-premise ATMs.
Probably, the most important consideration in selecting a network is whether it will remain competitive with other networks within a particular region. Along this line, a good indicator of competitiveness is the current growth rate of the network. While growth rates should be compared on the basis of how long a network has been in operation (among other factors), the average number of institutions per network has increased from 47.6 to 51.9 since this time last year. Such an increase amounts to an average annual growth rate of 9 percent.
Finally, network fees, which vary considerably, are an important variable to consider in selecting an ATM network. The fee structure may determine whether institutions incur an operating deficit or generate a profit, depending on whether they pay or collect switching fees, processing fees, or terminal access charges. Beyond the flat fees charged to all participants then, the net dollar cost of joining a network depends on current terminal and card base, position within the network (in front or behind the switch), and a projecting of the transactions that will be generated through each institution’s ATMs versus others.
For example, at least one large supermarket that plans to support its own terminals will charge financial institutions $0.70 per ATM withdrawal. Taking into consideration TDC survey findings revealing that supermarket ATMs average above 3,000 transactions a month and that withdrawals generally represent 60 percent of all ATM transactions, some of these supermarkets will be in a position to realize $15,000 per ATM in annual revenue from financial institutions in withdrawals alone. Clearly, S&Ls must assess membership fees closely to determine the effect shared ATM membership will have on their balance sheets. The Growth of Point-of-SAle
While ATMs are the most visible component of the EFT terminal revolution, S&Ls must also look at their investment in other technologies, such as point-of-sale (POS), telephone bill payment, or videotex. Despite early banking industry expectations that POS would be the first EFT service to attain broad popularity, point-of-sale has been slow to develop. One of the primary factors that has hindered POS development has been the inability of retailers and financial institutions to overcome their philosophical differences over the basic features of the system and the expected results. Simply stated, retailers are interested in accepting most forms of payment, including checks and credit cards, to provide convenient payment options for customers. Financial institutions, on the other hand, are more interested in potentially reducing the number of checks that are written and have focused primarily on the acceptance of debit cards at the point-of-sale. Adding to this problem is the fact that shared networks did not exist, for the most part, when these initial ventures took place. As a result, only a minority of a store’s customers could access the POS system, because the terminals often displaced valuable floor or counter space that could otherwise be used by the merchant to sell goods and services to all customers.
Today, the POS environment has changed dramatically. They explosion of shared networking has provided the potential for universal access at the point-of-sale. At the same time, positive consumer acclimatization to ATM services has led many industry analysts to expect that the same acceptance can be carried over into the POS environment. As a result, renewed interest in POS as a viable part of retail EFT is emerging. For S&Ls, opportunities whithin the POS environment will again most likely be realized through membership within shared networks. Shared networking provides the groundwork for quickly establishing additional EFT services, such as POS. In fact, TDC’s most recent survey of 150 existing shared networks reveals that 19 of these systems currently support POS services, while and additional ten networks currently have plans to offer POS.
The expected growth in terminal-based delivery services at the point-of-sale indicates that now is the time for S&Ls to start assessing the POS market. As a starting point for assessing POS, Table 2 identifies the various service
options available and the extent to which financial institutions ofter these options. Clearly, the focus of POS activity among financial institutions centers around credit authorization services. Through terminals linked to on-line authorization files where negative account files are stored, these authorization networks are designed to ensure that most credit card transactions made at the merchant’s point-of-sale are valid and will not result in charge-backs. In addition to offering credit authorization services, however, one-half of current POS offerers surveyed by TDC also support check guarantee services. More importantly, the terminal-based credit authorization networks now being developed provide the basis for a “true” point-of-sale service in which debit cards are also accepted as a means of payment. Essentially, all that is needed to enhance a credit authorization terminal for acceptance of debit cards is a PIN pad reader or other device that identifies the customer’s card and a receipt printer to validate completion of the transaction. Although there are operational problems to overcome in a direct debit environment, such as developing PIN algorithms for the network and handling transaction reversals, the technology is currently available and is being tested in some POS networks on a limited basis. More importantly, the terminal technology, along with the growth of shared networks, provides the basis for universal acceptance and electronic authorization of several types of payments at the merchant point-of-sale. Eventually, such a development may satisfy most of the retailers’ requirements and drive enough electronic payments through POS networks to substantially reduce the per item cost associated with processing POS transactions. At this point, the large capital investments associated with the development of terminal-based EFT networks will hopefully begin to pay large dividends.
Sensing the potential opportunities associated with POS, major third parties and bank-sponsored networks are aggressively expanding their base of POS terminals. According to TDC surveys, at least one major bank-sponsored network has plans to install almost 4,000 credit authorization terminals by the end of 1984. For the POS industry as a whole, TDC survey results suggest that the number of terminals installed by financial institutions offering POS could double in less than a two-year period. Moreover, large retailers are getting involved by providing authorization services directly to smaller retailers. J.C. Penney’s, for instance, has announced plans to support a nationwide credit authorization and processing network and has already signed service contracts with Gulf and Shell Oil to authorize credit cards at the point-of-sale. In short, even though widespread access to direct debit POS is probably from three to five years away, major players are rapidly positioning themselves to capitalize on the opportunities as they arise. Consequently, just as shared ATM networks are necessary to ensure that S&Ls maintain a solid ATM market position, membership within a viable POS network will be of major competitive importance to S&Ls meeting the demands of consumers and their growing merchant customer base. Telephone Bill Payment
In contrast to ATMs and POS, telephone bill payment (TBP) is an area in which thrift institutions were the innovators and have taken the lead. For the most part, TBP was introduced by thrift institutions to circumvent the prohibitions on their issuing transaction accounts. Responding to TDC’s latest TBP survey, 60 percent of the thrifts with TBP said they introduced the service “to increase market share” and 70 percent responded “to acquire an innovative image.”
Even though the overall number of TBP offerers grew from 1975 through 1982, the service never penetrated more than 420 depository institutions nationwide. Moreover, in 1983 alone, there was a startling 30 percent decline in the number of TBP programs offered. Although this drop was most apparent among commercial banks (almost half of the banks that offered TBP in 1982 terminated the service in 1983), close to one-quarter of the S&Ls with TBP also terminated TBP operations. The decline in TBP is attributable to two interrelated factors. First, TBP is not generally cost-effective. Only a handful of the institutions (less than 5 percent) that have offered TBP reported that their programs generated sufficient revenues to cover costs. To be profitable, TBP requires a large number of transactions, which few institutions have realized. Overall, consumers simply do not use TBP enough to support the service. Second, institutions may not need to offer TBP to maintain their competitive standing. It is not surprising, therefore, that many institutions have elected to abandon the service.
While TBP is not about to disappear, further contraction is expected. In contrast to ATMs and POS, TBP involves the use of a single-service terminal. This feature renders TBP inflexible and, at least in part, explains the failure of TBP systems to generate volumes of transactions. Nevertheless, TBP has always been consisdered the precursor to videotex/home banking, and this may explain some of its continuing support from the financial community. For S&Ls that are currently offering TBP, the knowledge and experience generated from maintaining direct service links to the customer’s home may provide at least some advantages in the event that home banking services take hold. Videotex/Home Banking
In contrast to TBP, videotax/home banking offers the potential to provide consumers with a full spectrum of financial services. Home banking, in fact, is but one part of videotax, which is an interactive audio-visual communications system used to deliver a wide range of information. The emergence of videotax has been predicated upon the booming personal computer and communications industry, with banking capabilities as more of an afterthough than initial stimulus.
Since the initial development of videotax, however home banking has become an integral part of these home delivery systems. This has changed rapidly, such that 92 percent of the videotax projects currently underway have incorporated home banking services. Simultaneously, the number of financial institutions participating in videotex/home banking projects has increased dramatically. By end of 1983, more than 100 financial institutions will have been involved in some type of videotex project.
As with early ATM developments, the large commercial banks overhelmingly dominate the videotex/home banking projects. Only three S&Ls (Home Federal, San Diego; Great American, San Diego; and Commercial Federal S&L, Omaha) are represented in any of the home banking projects. Home banking still is in the prototype stage, with first commercial service not expected before the fourth quarter of 1984.
A number of financial institutions has completed pilots and decided against commerical implementation as this time. Some announced projects also have been temporarily discontinued or “put on the back burner” for the time being. There are some technological, control, and cost problems with current systems. Videotex/home banking, however, is expected to hit the retail market hard in the latter half of the 1980s as the premier EFT facility.
Once commercialized, videotex/home banking systems will give rise to shared networks affording all financial institutions the opportunity to participate in the expected home banking revolution. The position of S&Ls in this area is analogous to that in ATMs: the largest institutions will be in on the ground floor, committing large amounts of capital and human resources. Once commercialized and viable, multi-institutional networks will enable S&Ls to offer home banking services without a major capital investment or protracted start-up costs. S&Ls will be able to enter the home banking market simultaneously with regional and community banks, most of which also will enter home banking via a network of some type.
In sum, S&Ls should be positioning themselves as “outlets” for financial services. S&Ls need not be direct suppliers of ATMs and other retail electronic financial services to remain competitive. Rather, S&Ls should rely upon the developmental capabilities of other institutions, while focusing primarily on the marketing and promotion of these services. As S&Ls enter the electronic age, the combination of effective marketing campaigns and associations with thriving EFT networks may provide the ability to compete with both banks and nonbanks vying for shares of the financial services marketplace.
COPYRIGHT 1984 U.S. Government Printing Office
COPYRIGHT 2004 Gale Group