Mr. Chairman, I wish to commend you and the Committee for holding these timely hearings on a very important subject. I am honored to be asked to testify.
In fact, I have been around so long that my professional career spans the entire history of ATMs in America. When they began in the early 1970s, I was Deputy Assistant Attorney General for Regulation in the Antitrust Division of the Justice Department. In the mid-1970s, when I was Assistant Attorney General in charge of the Antitrust Division, I debated with other public officials and industry leaders about competitive policies for ATM networks. In the late 1970s, I was a professor at the Cornell Law School and taught a novel course on electronic banking and securities market. In 1980, 1 worked with my Cornell colleague the late Norman Penney in producing the first edition of a treatise on electronic banking systems. During the 1980s, I helped various joint venturers establish new ATM networks regionally and nationally. In the 1990s, I have worked on substantial variety of electronic network problems for various clients. Today I testify for myself as a partner of Baker & Miller PLLC, and not on behalf of any particular client. Indeed, if I can escape today without having some former or present client very annoyed with me over something I might have said, I will have had a good day!
The Committee Staff has asked me to provide a broad overview of the history of ATM systems and to offer a general perspective as an antitrust lawyer on some of the competitive issues that seem most pressing and important.
The subjects that I have been asked to address are covered very extensively (and I would like to think well) in the treatise that my old friend Roland Brandel and I have produced over the years since the tragic death of my friend Norman Penney in 1982: it is entitled "The Law of Electronic Funds Transfer Systems" and the most recent edition was published by Warren, Gorham & Lamont in 1996. Accordingly, I have provided the Committee with a set of mercifully brief selected readings from this great work. They cover:
These materials should provide the Committee and the Staff more useful detail than I possibly could provide in the very few minutes that I have this morning.
In the beginning, a few innovative banks deployed what we now call ATMs for the purpose of gaining competitive advantage over their local competitors in the retail banking era, by offering consumers round the clock, convenient access to their funds. This occurred in the late 1960s and early 1970s, and the machines were still few in number and unfamiliar to most consumers. By the time we reached the early 1970s, there began to be substantial discussions about the implications of these machines for the consumer banking business in which convenient local access was regarded as very important. People even began to dream of hooking these machines together into networks covering ATMs at several banks in order to increase consumer convenience and generate transaction volumes that would reduce unit costs.
This talk of networks generated a populist reaction that led to enactment of what came to be known as "compulsory sharing"statutes in a good many states; these laws were generally inspired by smaller banks and their associations which feared that networks would exclude smaller banks. At the Justice Department, we felt compulsory sharing statutes were a terrible idea: by assuring that anyone could joint (at some late date) a successful network, the statutes deprived the would-be network innovators of any first mover advantage. The National Commission of Electronic Funds Transfers, on which I sat at times and kibitzed at others, came out against compulsory sharing laws, if I recall it correctly.
In the late 1970s, there was a great deal more policy noise than market place action. It may well be that all the talk from the antitrust agencies, the banking agencies, and Capitol Hill -- not to mention state capitals -- simply discouraged would be-network innovators from ever getting started, while stand-alone ATMs doing only "on us" transactions were unlikely to generate sufficient volumes to be economic in most locations.
Finally, in the early 1980s, ATM networks really took off. Somehow the business equation changed and network pioneers were willing to undertake the cost of hiring consultants, lawyers, logo designers, system engineers, and all the others it took to make the system work. I personally worked on the creation of the national Plus network, the Pulse network in Texas, the Yankee 24 network in New England, and the NYCE network in New York.
We started off in the early 1980s with a good many regional ATM networks and now we have only relatively few. In many places (such as Texas and Pennsylvania), two strong regional networks competed with each other to attract member institutions and to promote their service marks and offerings to the consuming public. It was believed at the time that belonging to a good network would enable the institution to attract business away from less dynamic competitors. Network participation was seen as a way of getting ahead.
Gradually, however, in the late 1980s and the 1990s, competition at the "network" level disappeared in most parts of the country. This occurred because of about three different factors. The first was that those who wanted to use the network as a way of getting ahead competitively were probably in a minority in the banking industry: the larger number of institutions probably feared being left behind and thus tended to favor policies (such as compulsory sharing statutes) to protect themselves from the winds of competition. Secondly, de facto mergers were produced when a member of one system would threaten the competing system with an antitrust boycott suit unless it allowed it to join the competing network (a situation that occurred most notably in Texas). Thirdly, the Justice Department and the Federal Reserve Board seemed passive or confused about competition issues at the "network" level, so they tended to rubber stamp any network merger that came along.
The heart of the analytical problem is created by confusion over what are generally called 64 network externalities" -- or less accurately "the economics of ubiquity." Either way, what this means is that, all things being equal, a network is more valuable to each user if it offers more outlets. This was seen as justifying network merger to monopoly. The trouble here is that all things are not always equal; and network participants may be better off with two networks (with adequate card bases and ATM coverage) competing against each other as opposed to having no choice but a monopoly network with universal coverage.
In any event, the breakdown of competitive oversight by the Fed and Justice was most
notable in the case of the 1988 acquisition by Electronic Payments Systems (which ran the MAC
network in Pennsylvania) of the CashStream network run by Mellon. The result was a powerful
network monopoly that was extended to surrounding states by sale of EPS equity to leading
members of nearby networks (such as MoneyStation). Indeed, this monopoly was so abusive in
so many ways that the Justice Department finally was compelled to do something in United
States v. Electronic Payment Services. The consent decree loosened up the defendant's rules that
had required smaller banks to have their ATMs driven by the network switch. The decree also
weakened the exclusive routing rule.
The MAC story is but the tip of the proverbial iceberg. As long as participating banks
and consumers have real "network" alternatives, any network's ability to harm consumer welfare
by enacting anti-competitive rules is really quite limited. Consumers will, at the margin, avoid
banks that belong to the competitively abusive networks, while banks will know that, in the long
run, they have the ability to switch to a more consumer friendly network. This reality would
keep networks honest.
Conversely, where there is only one network in a region -- especially where it is owned
by a few dominant banks -- the network can become a source of many competitive abusive and
rent-seeking opportunities. It is almost inevitable in that environment that Congress, the
regulators, or the state legislators will feel obliged to step in and protect interests that are not
being protected by the non-existent competitive process.
An ATM network -- and particularly a monopoly network -- involves a subtle blend of
competitive interests. At one level, the members (or the owners) are cooperating in trying to
make the network efficient and reliable. At another level, the members are competing with each
other to issue ATM access cards (or more accurately for consumer accounts utilizing ATM
access cards) and/or to deploy ATMs at potentially profitable locations. It is here that
competitive interests diverge and competitive problems may be created for networks. A major
issuer of cards, with relatively few ATMs, would rely on other members' ATMs to satisfy its
customers' convenience; and a major deployer of ATMs with relatively few (or even no) cards
issued may be relying on other members' cardholders to generate revenues on "foreign"
transactions at its machines. Any member which is a substantial issuer of cards and deployer of
machines will seek to use its position in ATMs to attract deposits (a) by favoring "on us" versus
"foreign" transactions at its machines and (b) by making "foreign" transactions by its cardholders
more expensive than "on us" transactions.
The big card issuer will wish for a low interchange fee payed to ATM owners, so long as
the fee is not so low that insufficient numbers of ATMs are deployed. A major deployer of
ATMs will want high interchange fees to support its expansive deployment strategy and will
want the right to charge surcharges. A bank that is a substantial issuer of cards and deployer of
ATMs may be less concerned by the absolute levels of network interchange fees. Such a bank,
however, may want the right to levy surcharges as a way of making "on us" transactions for its
own customers seem even more favorable than "foreign" transactions -- thus, possibly
encouraging consumers to switch accounts to it.
This, of course, brings us to the issue of surcharges in which, Mr. Chairman, you are so
interested. The practice of levying surcharges on "foreign" transactions but not "on us"
transactions is highly disadvantageous to a network because it encourages bank's customers to
travel farther and accept more inconvenience in order to do "on us" transactions which deprive
the network of volume. So-called "foreign fees" levied by the cardholder's bank on network
transactions have the same potential effect, but they do not seem to be quite as visible and
aggravating to consumers.
On the other hand, surcharges may encourage the deployment of more ATMs in more
remote, inconvenient, or expensive locations and, to that extent, they might be useful to both
consumers and the network. The problem that I have always seen is that the core ATM product
is "no hassle cash." Surcharges turn the whole thing into a "seek and search" operation.
As a market minded person, I see monopoly at the regional network level as generally
preventing competitive correction of abusive network practices. Networks that wish to prohibit
surcharges face antitrust suits. Those that wish to allow surcharges may find themselves faced
with consumer discontent, a diminished network volume, and a less valued service mark (or
alternatively, they may find enhanced network volume generated by substantially more ATM
deployment at high-priced sites, such as airports.) At this point, I see some reasonable arguments
on both sides of this issue, which probably counsels for legislative caution.
I appreciate the opportunity to be here. I would be more than glad to answer questions,
either now or later, from the Committee or the Staff on the issues I have raised in my oral
presentation or in my written materials. Home | Menu | Links | Info | Chairman's Page
Conclusion