Monopoly and competition in the supply and exchange of money

Antitrust Law Journal

Volumn 66, Issue 2, 1998, Pages 313-360

  Frankel, Alan S ^a  

^a Law and Economics Consulting Group   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0032361645     PISSN: 00036056     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (146)

1. In studying the history of earlier payment systems to inform modern controversies, this article follows the lead of other commentators. See, e.g., William F. Baxter, Bank Interchange of Transactional Paper: Legal and Economic Perspectives, 26 J.L. & ECON. 541 (1983); James J. McAndrews, Commentary, 77 FED. RESERVE BANK OF ST. LOUIS REV. 55 (1995). Recently, Federal Reserve Chairman Alan Greenspan suggested looking at the free banking era of the nineteenth century to gain insights about competition in new electronic forms of money. According to Greenspan, "the earlier period affords certain insights on the way markets behaved when government rules were much less pervasive" and "[t]hese insights . . . should be considered very carefully as we endeavor to understand and engage the new private currency markets of the twenty-first century." Remarks of Chairman Alan Greenspan Before U.S. Treasury Conference on Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). Some monetary economists have looked to history to support their challenge to the traditional rationale for the government's monopoly over the supply of currency. See, e.g., COMPETITION AND CURRENCY: ESSAYS ON FREE BANKING AND MONEY (Lawrence H. White ed., 1989).
2. In studying the history of earlier payment systems to inform modern controversies, this article follows the lead of other commentators. See, e.g., William F. Baxter, Bank Interchange of Transactional Paper: Legal and Economic Perspectives, 26 J.L. & ECON. 541 (1983); James J. McAndrews, Commentary, 77 FED. RESERVE BANK OF ST. LOUIS REV. 55 (1995). Recently, Federal Reserve Chairman Alan Greenspan suggested looking at the free banking era of the nineteenth century to gain insights about competition in new electronic forms of money. According to Greenspan, "the earlier period affords certain insights on the way markets behaved when government rules were much less pervasive" and "[t]hese insights . . . should be considered very carefully as we endeavor to understand and engage the new private currency markets of the twenty-first century." Remarks of Chairman Alan Greenspan Before U.S. Treasury Conference on Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). Some monetary economists have looked to history to support their challenge to the traditional rationale for the government's monopoly over the supply of currency. See, e.g., COMPETITION AND CURRENCY: ESSAYS ON FREE BANKING AND MONEY (Lawrence H. White ed., 1989).
3. In studying the history of earlier payment systems to inform modern controversies, this article follows the lead of other commentators. See, e.g., William F. Baxter, Bank Interchange of Transactional Paper: Legal and Economic Perspectives, 26 J.L. & ECON. 541 (1983); James J. McAndrews, Commentary, 77 FED. RESERVE BANK OF ST. LOUIS REV. 55 (1995). Recently, Federal Reserve Chairman Alan Greenspan suggested looking at the free banking era of the nineteenth century to gain insights about competition in new electronic forms of money. According to Greenspan, "the earlier period affords certain insights on the way markets behaved when government rules were much less pervasive" and "[t]hese insights . . . should be considered very carefully as we endeavor to understand and engage the new private currency markets of the twenty-first century." Remarks of Chairman Alan Greenspan Before U.S. Treasury Conference on Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). Some monetary economists have looked to history to support their challenge to the traditional rationale for the government's monopoly over the supply of currency. See, e.g., COMPETITION AND CURRENCY: ESSAYS ON FREE BANKING AND MONEY (Lawrence H. White ed., 1989).
4. See, e.g., GLYN DAVIES, A HISTORY OF MONEY FROM ANCIENT TIMES TO THE PRESENT DAY (1994); George A. Selgin & Lawrence H. White, The Evolution of a Free Banking System, 25 ECON. INQUIRY (1987). Wampum (Native American stringed beads derived from shells) was used by American colonists for many decades. Milton Friedman provides a fascinating account of one island civilization's use of stone wheels as money. MILTON FRIEDMAN, MONEY MISCHIEF: EPISODES IN MONETARY HISTORY 3 (1994).
5. See, e.g., GLYN DAVIES, A HISTORY OF MONEY FROM ANCIENT TIMES TO THE PRESENT DAY (1994); George A. Selgin & Lawrence H. White, The Evolution of a Free Banking System, 25 ECON. INQUIRY (1987). Wampum (Native American stringed beads derived from shells) was used by American colonists for many decades. Milton Friedman provides a fascinating account of one island civilization's use of stone wheels as money. MILTON FRIEDMAN, MONEY MISCHIEF: EPISODES IN MONETARY HISTORY 3 (1994).
6. See, e.g., GLYN DAVIES, A HISTORY OF MONEY FROM ANCIENT TIMES TO THE PRESENT DAY (1994); George A. Selgin & Lawrence H. White, The Evolution of a Free Banking System, 25 ECON. INQUIRY (1987). Wampum (Native American stringed beads derived from shells) was used by American colonists for many decades. Milton Friedman provides a fascinating account of one island civilization's use of stone wheels as money. MILTON FRIEDMAN, MONEY MISCHIEF: EPISODES IN MONETARY HISTORY 3 (1994).
7. In practice, different forms of money and methods of transferring ownership of money impose different relative costs in different transactions. Thus, few consumers obtain a bank cashier's check to purchase a quart of milk, and few purchase a house with nickels.
8. The term "seignorage" is still used by economists to describe the government's profit on the issuance of money.
9. See, e.g., entry for Seignorage, in PALGRAVE, 3 DICTIONARY OF POLITICAL ECONOMY 372-74 (Macmillan Press ed. 1987) (1899).
10. The Lydians of Asia Minor are recognized as the first to use standardized coins, in the seventh century BCE. See JACK WEATHERFORD, THE HISTORY OF MONEY 30 (1997).
11. Competing mints were typically illegal. Such competition was deemed "counterfeiting" and received harsh penalties - one suspects regardless of whether the coins contained the same amount of pure gold as those issued by the government mint. See ADAM SMITH, AN INQUIRY INTO THE NATURE AND CAUSES OF THE WEALTH OF NATIONS 551 (Liberty Classics ed. 1981) (1776).
12. "Sweating" is accomplished by shaking vigorously a bag containing gold coins, thereby leaving the bag lined with a residue of gold dust while leaving each coin only slightly more worn than before. "Clipping" and "shaving" involve the more direct removal of chunks or slivers of gold from a coin, which was the impetus for the invention of serrated coin edges, still retained today in the United States even on coins made from base metals such as the dime, quarter, and half-dollar.
13. The economic tendency, reinforced by law, for all coins to circulate at par gave the coin issuer an incentive to introduce coins of debased metallic content into the economy. Par circulation meant that bearers of all coins, full content and debased content alike, would bear the cost of the resulting inflation. Weighing or assaying coins would have eliminated the incentive to issue debased coins because the issuer would have received proportionately less for the coins as the precious metal content was reduced. Measuring the intrinsic value of each coin, however, would have negated one of the primary benefits they provided, and likely was done only when the circulating stock of coins had deteriorated significantly.
14. Simple examples of price coherence might include checkout line candy or cigarette racks, or vending machines that could post efficiently only a single price point for any item sold. Sellers might be able to mitigate the effects of price coherence in some cases if they have tools other than price (e.g., shelf space allocations and marketing effort) to influence customers' purchase decisions.
15. The price of the high-cost product increases less than would be the case if the merchant broke the pattern of price coherence, and the price for any low-cost product increases even though its cost has remained constant.
16. Sir Thomas Gresham was royal agent in Antwerp and advisor to Elizabeth I during the middle part of the sixteenth century. DAVIES, supra note 2, at 203. The first recorded recognition of Gresham's Law greatly preceded Sir Thomas. In The Frogs, the Greek poet Aristophanes wrote: I have often noticed that there are good and honest citizens in Athens who are as old gold to new money. The ancient coins are excellent . . . well struck and give a pure ring; everywhere they obtain currency, both in Greece and in strange lands; yet we make no use of them and prefer those bad copper pieces quite recently issued and so wretchedly struck. DAVIES, supra, at 77. An historically recent example of Gresham's Law in operation in the United States occurred in 1965, when the government ceased minting silver dimes and quarters and reduced the silver content of half-dollar coins (later eliminating all silver from half dollars). Within a few years, it was extremely unusual to find a silver coin in circulation, though old pennies and nickels, neither of which contained precious metals before or after 1965, remained easy to find.
17. More generally, the effects of Gresham's Law tend to appear when two forms of money are not accepted in trade at the same ratio or exchange rate as the underlying market values of the respective monies would imply if there were no restrictions or imperfections in the market.
18. In the restatement, I include the words "tends to" because there are examples, both historic and contemporary, in which different currencies or payment methods are accepted at some nonpar exchange rate. These, however, are typically cases in which the value of one or the other of the payment methods is so far from par value, yet popular nonetheless, that it pays for merchants and individuals to engage in routine exchange calculations. For example, in Israel today the exchange rate is significantly different from 1:1 between the shekel and the dollar, and yet dollars are so prevalent that virtually all merchants routinely perform exchange calculations. A prediction of this analysis is that if the exchange rate was close enough to 1:1, then dollars would be simply accepted at par by most merchants without bothering with exchange calculations. A lesson for would-be exploiters of Gresham's Law is not to get too greedy. Small deviations from par on a market value basis will be overlooked in ordinary transactions, but large deviations will result in an exchange calculation or denial of acceptance.
19. The price of a gold coin is the difference between the gold content of the coin and the weight of gold bullion demanded in exchange for a coin. With a low enough coin price, some merchants may dispense with weighing and assaying equipment. When coin prices rise, merchants have a greater incentive to keep such equipment, so their customers do not first have to buy coins with their bullion.
20. In antitrust terms, to hold to the contrary would be to commit the well-known Cellophane fallacy. In United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377 (1956), the Supreme Court mistakenly concluded that the manufacturer of cellophane did not possess market power because there were many substitute flexible wrapping materials, such as waxed paper, that were provided by competitive industries, and therefore that price increases would cause a large number of customers to switch to these alternatives. The Court neglected to recognize that these other materials were only good substitutes because cellophane already was being sold at the monopoly price, which was far above the competitive price. This economic error was repeated in the antitrust analysis of credit card systems performed by the court in National Bancard Corp. v. Visa U.S.A., Inc., 596 F. Supp 1231 (S.D. Fla. 1984), aff'd, 779 F.2d 592 (11th Cir. 1986). In NaBanco the court mistakenly found that the existence of other competing payment systems, such as checks and cash, meant that a credit card network could not possess market power. See Dennis W. Carlton & Alan S. Frankel, The Antitrust Economics of Credit Card Networks, 63 ANTITRUST L.J. 643 (1995).
21. Article I, Section 8 of the U.S. Constitution gives the Federal Government the power to coin and regulate the value of money.
22. Banks provided intermediation services and created liquidity. They induced customers to deposit funds (originally, precious commodities) from which they made loans, keeping enough reserves to meet expected cash outflows. A bank note was a demand liability of the issuing bank; like a checking account balance, it was subject to immediate redemption upon request.
23. The public was willing to accept and hold currency in lieu of gold (in a gold standard economy) only if, for a significant number of transactions, the full cost of using gold exceeded the expected cost of using currency (including the expected cost arising from failures to redeem). If the public strictly preferred gold coins to bank notes, bank notes could still circulate if sold by issuers at a discount for gold and, to prevent arbitrage, were irredeemable or redeemable only at a discount for some period of time. In short, the notes would have to pay interest. Such notes might have been issued to some extent in Scotland prior to the Act of Parliament of 1765 that banned the practice of delayed redemption. In practice, however, the time and trouble associated with transacting in notes not circulating at par (for whatever reason) minimizes their usefulness as money, whatever the legal environment. See Lawrence H. White, Accounting for Non-Interest-Bearing Currency: A Critique of the Legal Restrictions Theory of Money, 19 J. MONEY, CREDIT & BANKING (1987). In the United States, nonpar currency resulted in part from the long distances people could travel from issuing banks, combined with prohibitions on interstate branch banking. Nonpar currency usually was converted by brokers (into another currency that did circulate at par), then returned to the vicinity of the issuing bank for redemption. For a discussion of domestic currency exchange markets, see Ronnie J. Phillips & Harvey Cutler, The Domestic Exchanges and Regional Economic Growth in the U.S., 1899-1908: Evidence from Cointegration Analysis, J. ECON. HIST. (forthcoming); Ronnie J. Phillips & P.A.V.B. Swamy, Par Clearance in the Domestic Exchanges: The Impact of National Banknotes, in RESEARCH IN ECONOMIC HISTORY (Alexander J. Field ed., forthcoming).
24. The public was willing to accept and hold currency in lieu of gold (in a gold standard economy) only if, for a significant number of transactions, the full cost of using gold exceeded the expected cost of using currency (including the expected cost arising from failures to redeem). If the public strictly preferred gold coins to bank notes, bank notes could still circulate if sold by issuers at a discount for gold and, to prevent arbitrage, were irredeemable or redeemable only at a discount for some period of time. In short, the notes would have to pay interest. Such notes might have been issued to some extent in Scotland prior to the Act of Parliament of 1765 that banned the practice of delayed redemption. In practice, however, the time and trouble associated with transacting in notes not circulating at par (for whatever reason) minimizes their usefulness as money, whatever the legal environment. See Lawrence H. White, Accounting for Non-Interest-Bearing Currency: A Critique of the Legal Restrictions Theory of Money, 19 J. MONEY, CREDIT & BANKING (1987). In the United States, nonpar currency resulted in part from the long distances people could travel from issuing banks, combined with prohibitions on interstate branch banking. Nonpar currency usually was converted by brokers (into another currency that did circulate at par), then returned to the vicinity of the issuing bank for redemption. For a discussion of domestic currency exchange markets, see Ronnie J. Phillips & Harvey Cutler, The Domestic Exchanges and Regional Economic Growth in the U.S., 1899-1908: Evidence from Cointegration Analysis, J. ECON. HIST. (forthcoming); Ronnie J. Phillips & P.A.V.B. Swamy, Par Clearance in the Domestic Exchanges: The Impact of National Banknotes, in RESEARCH IN ECONOMIC HISTORY (Alexander J. Field ed., forthcoming).
25. The public was willing to accept and hold currency in lieu of gold (in a gold standard economy) only if, for a significant number of transactions, the full cost of using gold exceeded the expected cost of using currency (including the expected cost arising from failures to redeem). If the public strictly preferred gold coins to bank notes, bank notes could still circulate if sold by issuers at a discount for gold and, to prevent arbitrage, were irredeemable or redeemable only at a discount for some period of time. In short, the notes would have to pay interest. Such notes might have been issued to some extent in Scotland prior to the Act of Parliament of 1765 that banned the practice of delayed redemption. In practice, however, the time and trouble associated with transacting in notes not circulating at par (for whatever reason) minimizes their usefulness as money, whatever the legal environment. See Lawrence H. White, Accounting for Non-Interest-Bearing Currency: A Critique of the Legal Restrictions Theory of Money, 19 J. MONEY, CREDIT & BANKING (1987). In the United States, nonpar currency resulted in part from the long distances people could travel from issuing banks, combined with prohibitions on interstate branch banking. Nonpar currency usually was converted by brokers (into another currency that did circulate at par), then returned to the vicinity of the issuing bank for redemption. For a discussion of domestic currency exchange markets, see Ronnie J. Phillips & Harvey Cutler, The Domestic Exchanges and Regional Economic Growth in the U.S., 1899-1908: Evidence from Cointegration Analysis, J. ECON. HIST. (forthcoming); Ronnie J. Phillips & P.A.V.B. Swamy, Par Clearance in the Domestic Exchanges: The Impact of National Banknotes, in RESEARCH IN ECONOMIC HISTORY (Alexander J. Field ed., forthcoming).
26. The public also might have been willing to hold bank notes in lieu of deposits because notes were senior to deposits in the event of bankruptcy.
27. See MILTON FRIEDMAN & ANNA J. SCHWARTZ, A MONETARY HISTORY OF THE UNITED STATES, 1867-1960 at 18 (Nat'l Bur. Econ. Research 1963). The intention was to replace state banks with national banks created during the war. The conversion of banks to federal charters soon slowed, however, as deposits and checks displaced currency as the most important form of money in the economy.
28. FED. RES. STAT. REL. H.6 Table 5 (Mar. 26, 1998) (Statistical Release). The U.S. Treasury has begun introducing new currency designs that will make counterfeit notes more difficult to produce. The availability of these notes could generate a windfall to the government as foreigners (who hold most of the $100 bills currently outstanding) increase their holdings of U.S. currency. Currently, widespread counterfeiting in foreign countries reduces the demand for U.S. currency. See Richard D. Porter & Ruth A. Judson, The Location of U.S. Currency: How Much Is Abroad?, 82 FED. RES. BULL. 883 (1996); R. Schafrik & S. Church, Protecting the Greenback, 273 SCI. AM. 40 (July 1995).
29. FED. RES. STAT. REL. H.6 Table 5 (Mar. 26, 1998) (Statistical Release). The U.S. Treasury has begun introducing new currency designs that will make counterfeit notes more difficult to produce. The availability of these notes could generate a windfall to the government as foreigners (who hold most of the $100 bills currently outstanding) increase their holdings of U.S. currency. Currently, widespread counterfeiting in foreign countries reduces the demand for U.S. currency. See Richard D. Porter & Ruth A. Judson, The Location of U.S. Currency: How Much Is Abroad?, 82 FED. RES. BULL. 883 (1996); R. Schafrik & S. Church, Protecting the Greenback, 273 SCI. AM. 40 (July 1995).
30. FED. RES. STAT. REL. H.6 Table 5 (Mar. 26, 1998) (Statistical Release). The U.S. Treasury has begun introducing new currency designs that will make counterfeit notes more difficult to produce. The availability of these notes could generate a windfall to the government as foreigners (who hold most of the $100 bills currently outstanding) increase their holdings of U.S. currency. Currently, widespread counterfeiting in foreign countries reduces the demand for U.S. currency. See Richard D. Porter & Ruth A. Judson, The Location of U.S. Currency: How Much Is Abroad?, 82 FED. RES. BULL. 883 (1996); R. Schafrik & S. Church, Protecting the Greenback, 273 SCI. AM. 40 (July 1995).
31. It was projected that 9.8 billion notes would be printed in 1997, 8.8 billion of which would replace worn, mostly low-denomination bills. The total cost of printing and distributing this new currency was estimated to be about $400 million. See BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM, ANNUAL REPORT: BUDGET REVIEW 1996-97.
32. Until the 1960s, U.S. Silver Certificates continued to be redeemable in silver. Other U.S. currency has been irredeemable since the 1930s. Though the U.S. dollar remained on the "gold standard" until 1971, private citizens were prohibited from owning gold coins and bullion (except for numismatic purposes). Economists have viewed the government's seignorage profits in two different ways. Typically, it is seen as the interest saved by the government during the year by keeping currency in circulation instead of issuing an equivalent amount of interest-bearing securities. It alternatively has been viewed as the net increase in the face amount of currency issued during the year less the net cost of producing the additional notes. See, e.g., WHITE, supra note 1, at 12, 76, 86. The distinction is probably unimportant. At least in equilibrium, the two amounts will be the same, because the present discounted value of avoiding interest payments (forever) on $100 in Treasury borrowings saved by a permanent increase in the stock of outstanding Federal Reserve Notes by $100 will itself be precisely $100.
33. As one congressman has conceded: [R]ight at the very outset, the Fed and the Treasury have inherent conflicts of interest with the development of electronic money. Theoretically, they stand to lose $20 billion or so in annual earnings, and a substantial measure of independence, if electronic money were to completely supplant physical notes and coins. Remarks of Michael N. Castle, Chairman, Subcomm. on Domestic and International Monetary Policy, Before the U.S. Treasury Conference Toward Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). There has been an extensive debate in the economics literature over whether competition among banks would provide a stable currency, or instead would result in the overissuance of bank notes leading to a worthless currency and macroeconomic disruption. See, e.g., VERA C. SMITH, THE RATIONALE OF CENTRAL BANKING AND THE FREE BANKING ALTERNATIVE (Indianapolis: Liberty Press ed. 1990) (1936); MILTON FRIEDMAN, A PROGRAM FOR MONETARY STABILITY (1960); Benjamin Klein, The Competitive Supply of Money, 6 J. MONEY, CREDIT & BANKING 423 (1974); Lawrence H. White, Competitive Money, Inside and Out, 3 CATO J. (Spring 1983); Milton Friedman & Anna J. Schwartz, Has Government Any Role in Money?, J. MON. ECON. 17 (Jan. 1986); WHITE, supra note 1; STEVEN HOROWITZ, MONETARY EVOLUTION, FREE BANKING, AND ECONOMIC ORDER (1992); Joseph A. Ritter, The Transition from Barter to Fiat Money, 85 AM. ECON. REV. 134 (1995); Howard Bodenhorn, Entry, Rivalry, and Free Banking in Antebellum America, 72 REV. ECON. & STAT. 682 (1990). Until recently, this debate had little practical relevance to competition in payment systems markets, but technological developments are making possible an effective return to the competitive currency era through electronic "stored value cards" and similar devices.
34. As one congressman has conceded: [R]ight at the very outset, the Fed and the Treasury have inherent conflicts of interest with the development of electronic money. Theoretically, they stand to lose $20 billion or so in annual earnings, and a substantial measure of independence, if electronic money were to completely supplant physical notes and coins. Remarks of Michael N. Castle, Chairman, Subcomm. on Domestic and International Monetary Policy, Before the U.S. Treasury Conference Toward Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). There has been an extensive debate in the economics literature over whether competition among banks would provide a stable currency, or instead would result in the overissuance of bank notes leading to a worthless currency and macroeconomic disruption. See, e.g., VERA C. SMITH, THE RATIONALE OF CENTRAL BANKING AND THE FREE BANKING ALTERNATIVE (Indianapolis: Liberty Press ed. 1990) (1936); MILTON FRIEDMAN, A PROGRAM FOR MONETARY STABILITY (1960); Benjamin Klein, The Competitive Supply of Money, 6 J. MONEY, CREDIT & BANKING 423 (1974); Lawrence H. White, Competitive Money, Inside and Out, 3 CATO J. (Spring 1983); Milton Friedman & Anna J. Schwartz, Has Government Any Role in Money?, J. MON. ECON. 17 (Jan. 1986); WHITE, supra note 1; STEVEN HOROWITZ, MONETARY EVOLUTION, FREE BANKING, AND ECONOMIC ORDER (1992); Joseph A. Ritter, The Transition from Barter to Fiat Money, 85 AM. ECON. REV. 134 (1995); Howard Bodenhorn, Entry, Rivalry, and Free Banking in Antebellum America, 72 REV. ECON. & STAT. 682 (1990). Until recently, this debate had little practical relevance to competition in payment systems markets, but technological developments are making possible an effective return to the competitive currency era through electronic "stored value cards" and similar devices.
35. As one congressman has conceded: [R]ight at the very outset, the Fed and the Treasury have inherent conflicts of interest with the development of electronic money. Theoretically, they stand to lose $20 billion or so in annual earnings, and a substantial measure of independence, if electronic money were to completely supplant physical notes and coins. Remarks of Michael N. Castle, Chairman, Subcomm. on Domestic and International Monetary Policy, Before the U.S. Treasury Conference Toward Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). There has been an extensive debate in the economics literature over whether competition among banks would provide a stable currency, or instead would result in the overissuance of bank notes leading to a worthless currency and macroeconomic disruption. See, e.g., VERA C. SMITH, THE RATIONALE OF CENTRAL BANKING AND THE FREE BANKING ALTERNATIVE (Indianapolis: Liberty Press ed. 1990) (1936); MILTON FRIEDMAN, A PROGRAM FOR MONETARY STABILITY (1960); Benjamin Klein, The Competitive Supply of Money, 6 J. MONEY, CREDIT & BANKING 423 (1974); Lawrence H. White, Competitive Money, Inside and Out, 3 CATO J. (Spring 1983); Milton Friedman & Anna J. Schwartz, Has Government Any Role in Money?, J. MON. ECON. 17 (Jan. 1986); WHITE, supra note 1; STEVEN HOROWITZ, MONETARY EVOLUTION, FREE BANKING, AND ECONOMIC ORDER (1992); Joseph A. Ritter, The Transition from Barter to Fiat Money, 85 AM. ECON. REV. 134 (1995); Howard Bodenhorn, Entry, Rivalry, and Free Banking in Antebellum America, 72 REV. ECON. & STAT. 682 (1990). Until recently, this debate had little practical relevance to competition in payment systems markets, but technological developments are making possible an effective return to the competitive currency era through electronic "stored value cards" and similar devices.
36. As one congressman has conceded: [R]ight at the very outset, the Fed and the Treasury have inherent conflicts of interest with the development of electronic money. Theoretically, they stand to lose $20 billion or so in annual earnings, and a substantial measure of independence, if electronic money were to completely supplant physical notes and coins. Remarks of Michael N. Castle, Chairman, Subcomm. on Domestic and International Monetary Policy, Before the U.S. Treasury Conference Toward Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). There has been an extensive debate in the economics literature over whether competition among banks would provide a stable currency, or instead would result in the overissuance of bank notes leading to a worthless currency and macroeconomic disruption. See, e.g., VERA C. SMITH, THE RATIONALE OF CENTRAL BANKING AND THE FREE BANKING ALTERNATIVE (Indianapolis: Liberty Press ed. 1990) (1936); MILTON FRIEDMAN, A PROGRAM FOR MONETARY STABILITY (1960); Benjamin Klein, The Competitive Supply of Money, 6 J. MONEY, CREDIT & BANKING 423 (1974); Lawrence H. White, Competitive Money, Inside and Out, 3 CATO J. (Spring 1983); Milton Friedman & Anna J. Schwartz, Has Government Any Role in Money?, J. MON. ECON. 17 (Jan. 1986); WHITE, supra note 1; STEVEN HOROWITZ, MONETARY EVOLUTION, FREE BANKING, AND ECONOMIC ORDER (1992); Joseph A. Ritter, The Transition from Barter to Fiat Money, 85 AM. ECON. REV. 134 (1995); Howard Bodenhorn, Entry, Rivalry, and Free Banking in Antebellum America, 72 REV. ECON. & STAT. 682 (1990). Until recently, this debate had little practical relevance to competition in payment systems markets, but technological developments are making possible an effective return to the competitive currency era through electronic "stored value cards" and similar devices.
37. As one congressman has conceded: [R]ight at the very outset, the Fed and the Treasury have inherent conflicts of interest with the development of electronic money. Theoretically, they stand to lose $20 billion or so in annual earnings, and a substantial measure of independence, if electronic money were to completely supplant physical notes and coins. Remarks of Michael N. Castle, Chairman, Subcomm. on Domestic and International Monetary Policy, Before the U.S. Treasury Conference Toward Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). There has been an extensive debate in the economics literature over whether competition among banks would provide a stable currency, or instead would result in the overissuance of bank notes leading to a worthless currency and macroeconomic disruption. See, e.g., VERA C. SMITH, THE RATIONALE OF CENTRAL BANKING AND THE FREE BANKING ALTERNATIVE (Indianapolis: Liberty Press ed. 1990) (1936); MILTON FRIEDMAN, A PROGRAM FOR MONETARY STABILITY (1960); Benjamin Klein, The Competitive Supply of Money, 6 J. MONEY, CREDIT & BANKING 423 (1974); Lawrence H. White, Competitive Money, Inside and Out, 3 CATO J. (Spring 1983); Milton Friedman & Anna J. Schwartz, Has Government Any Role in Money?, J. MON. ECON. 17 (Jan. 1986); WHITE, supra note 1; STEVEN HOROWITZ, MONETARY EVOLUTION, FREE BANKING, AND ECONOMIC ORDER (1992); Joseph A. Ritter, The Transition from Barter to Fiat Money, 85 AM. ECON. REV. 134 (1995); Howard Bodenhorn, Entry, Rivalry, and Free Banking in Antebellum America, 72 REV. ECON. & STAT. 682 (1990). Until recently, this debate had little practical relevance to competition in payment systems markets, but technological developments are making possible an effective return to the competitive currency era through electronic "stored value cards" and similar devices.
38. As one congressman has conceded: [R]ight at the very outset, the Fed and the Treasury have inherent conflicts of interest with the development of electronic money. Theoretically, they stand to lose $20 billion or so in annual earnings, and a substantial measure of independence, if electronic money were to completely supplant physical notes and coins. Remarks of Michael N. Castle, Chairman, Subcomm. on Domestic and International Monetary Policy, Before the U.S. Treasury Conference Toward Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). There has been an extensive debate in the economics literature over whether competition among banks would provide a stable currency, or instead would result in the overissuance of bank notes leading to a worthless currency and macroeconomic disruption. See, e.g., VERA C. SMITH, THE RATIONALE OF CENTRAL BANKING AND THE FREE BANKING ALTERNATIVE (Indianapolis: Liberty Press ed. 1990) (1936); MILTON FRIEDMAN, A PROGRAM FOR MONETARY STABILITY (1960); Benjamin Klein, The Competitive Supply of Money, 6 J. MONEY, CREDIT & BANKING 423 (1974); Lawrence H. White, Competitive Money, Inside and Out, 3 CATO J. (Spring 1983); Milton Friedman & Anna J. Schwartz, Has Government Any Role in Money?, J. MON. ECON. 17 (Jan. 1986); WHITE, supra note 1; STEVEN HOROWITZ, MONETARY EVOLUTION, FREE BANKING, AND ECONOMIC ORDER (1992); Joseph A. Ritter, The Transition from Barter to Fiat Money, 85 AM. ECON. REV. 134 (1995); Howard Bodenhorn, Entry, Rivalry, and Free Banking in Antebellum America, 72 REV. ECON. & STAT. 682 (1990). Until recently, this debate had little practical relevance to competition in payment systems markets, but technological developments are making possible an effective return to the competitive currency era through electronic "stored value cards" and similar devices.
39. As one congressman has conceded: [R]ight at the very outset, the Fed and the Treasury have inherent conflicts of interest with the development of electronic money. Theoretically, they stand to lose $20 billion or so in annual earnings, and a substantial measure of independence, if electronic money were to completely supplant physical notes and coins. Remarks of Michael N. Castle, Chairman, Subcomm. on Domestic and International Monetary Policy, Before the U.S. Treasury Conference Toward Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). There has been an extensive debate in the economics literature over whether competition among banks would provide a stable currency, or instead would result in the overissuance of bank notes leading to a worthless currency and macroeconomic disruption. See, e.g., VERA C. SMITH, THE RATIONALE OF CENTRAL BANKING AND THE FREE BANKING ALTERNATIVE (Indianapolis: Liberty Press ed. 1990) (1936); MILTON FRIEDMAN, A PROGRAM FOR MONETARY STABILITY (1960); Benjamin Klein, The Competitive Supply of Money, 6 J. MONEY, CREDIT & BANKING 423 (1974); Lawrence H. White, Competitive Money, Inside and Out, 3 CATO J. (Spring 1983); Milton Friedman & Anna J. Schwartz, Has Government Any Role in Money?, J. MON. ECON. 17 (Jan. 1986); WHITE, supra note 1; STEVEN HOROWITZ, MONETARY EVOLUTION, FREE BANKING, AND ECONOMIC ORDER (1992); Joseph A. Ritter, The Transition from Barter to Fiat Money, 85 AM. ECON. REV. 134 (1995); Howard Bodenhorn, Entry, Rivalry, and Free Banking in Antebellum America, 72 REV. ECON. & STAT. 682 (1990). Until recently, this debate had little practical relevance to competition in payment systems markets, but technological developments are making possible an effective return to the competitive currency era through electronic "stored value cards" and similar devices.
40. As one congressman has conceded: [R]ight at the very outset, the Fed and the Treasury have inherent conflicts of interest with the development of electronic money. Theoretically, they stand to lose $20 billion or so in annual earnings, and a substantial measure of independence, if electronic money were to completely supplant physical notes and coins. Remarks of Michael N. Castle, Chairman, Subcomm. on Domestic and International Monetary Policy, Before the U.S. Treasury Conference Toward Electronic Money & Banking: The Role of Government (Sept. 19-20, 1996). There has been an extensive debate in the economics literature over whether competition among banks would provide a stable currency, or instead would result in the overissuance of bank notes leading to a worthless currency and macroeconomic disruption. See, e.g., VERA C. SMITH, THE RATIONALE OF CENTRAL BANKING AND THE FREE BANKING ALTERNATIVE (Indianapolis: Liberty Press ed. 1990) (1936); MILTON FRIEDMAN, A PROGRAM FOR MONETARY STABILITY (1960); Benjamin Klein, The Competitive Supply of Money, 6 J. MONEY, CREDIT & BANKING 423 (1974); Lawrence H. White, Competitive Money, Inside and Out, 3 CATO J. (Spring 1983); Milton Friedman & Anna J. Schwartz, Has Government Any Role in Money?, J. MON. ECON. 17 (Jan. 1986); WHITE, supra note 1; STEVEN HOROWITZ, MONETARY EVOLUTION, FREE BANKING, AND ECONOMIC ORDER (1992); Joseph A. Ritter, The Transition from Barter to Fiat Money, 85 AM. ECON. REV. 134 (1995); Howard Bodenhorn, Entry, Rivalry, and Free Banking in Antebellum America, 72 REV. ECON. & STAT. 682 (1990). Until recently, this debate had little practical relevance to competition in payment systems markets, but technological developments are making possible an effective return to the competitive currency era through electronic "stored value cards" and similar devices.
41. See, e.g., CHARLES F. DUNBAR, THE THEORY AND HISTORY OF BANKING (Oliver M.W. Sprague ed., 1917) (1891).
42. See, e.g., James D. Magee, Historical Analogy to the Fight Against Par Check Collection, 31 J. POL. & ECON. 433 (1923); WALTER EARL SPAHR, THE CLEARING AND COLLECTION OF CHECKS (1926). Presumably, merchants themselves tried to keep the notes in circulation so they would not incur exchange charges.
43. See, e.g., James D. Magee, Historical Analogy to the Fight Against Par Check Collection, 31 J. POL. & ECON. 433 (1923); WALTER EARL SPAHR, THE CLEARING AND COLLECTION OF CHECKS (1926). Presumably, merchants themselves tried to keep the notes in circulation so they would not incur exchange charges.
44. See, e.g., Phillips & Swamy, supra note 19.
45. While a bank in one city might have few transactions involving a particular bank in another city, it might have many transactions involving the other banks in that city. A correspondent could present notes over the counter and receive payment at par and send a combined shipment of specie to the presenting bank. In practice, deposits were maintained at correspondent banks and specie shipments occurred only when balances at the correspondent became too small or too large.
46. This is also true because if a bank in Town A had many local competitors, as distant Town B banks learned of Town A bank's supracompetitive exchange charges, they would discount the bank's notes more than they discounted notes of other Town A banks and more merchants would refuse to accept the currency. Local Town A customers who traveled then would have an incentive to obtain other bank notes in Town A before traveling to Town B or shipping currency there to make a purchase.
47. See, e.g., Magee, supra note 27, at 438.
48. Such unexpected, large demands for redemption taxed the specie reserves issuing banks kept on hand, and forced them to keep more specie and issue fewer notes. These demands also forced the country banks to maintain fewer assets, which could have provided a benefit to the public if they were otherwise undercapitalized and excessively risky to an extent unknown by holders of their notes. Two early court cases confirmed the obligation of banks to redeem promptly even large amounts of notes at par when presented physically for collection, and provided for significant interest penalties for any delays in redemption. Suffolk Bank v. Lincoln Bank, 3 Mason, I, Fed. Cas. No. 13,590 (U.S. Circuit Court, Portland, Me. 1821); Suffolk Bank v. Worcester Bank, 22 Mass. 106 (1827). See Magee, supra note 27, at 440-45; SPAHR, supra note 27, at 75-76.
49. Of course, collective imposition of a fee in excess of costs (i.e., the collective exercise of market power) would tend to induce entry of banks where that was possible and expansion of note-issuing efforts until (at least in the long run) at the margin the incremental costs associated with inducing the public to accept another bank note with the collusive fee in place would equal the expected additional profit that banks earned from the fee. This concept will become important in the discussion of credit card systems in Part IV below. Country banks could impose profitable exchange fees for a long period of time presumably because the local markets where they were located were insufficiently large to support efficiently more than one or two banks.
50. The prevalence today of penny trays at the counters of small retailers, to aid in rounding transaction prices to even amounts, is another manifestation of the same phenomenon. For small enough deviations from exact pricing, many consumers and merchants will not go to the trouble of being exact. There have been many proposals to eliminate the penny from circulation. However, as long as the cost of minting the copper is less than $0.01, the government will earn seignorage profits on the huge number of pennies that are withdrawn from circulation and replaced each year.
51. This argument is also advanced in George A. Selgin & Lawrence H. White, The Evolution of a Free Banking System, 25 ECON. INQUIRY 439 (1987).
52. If there were no transaction costs, then country bank notes would not have circulated at par, thus confining the incidence of any market power of those banks to their local areas.
53. Having issued the notes (perhaps exercising some monopoly power over the initial recipients in the process), it repurchased them from distant holders only at a discount greater than any cost that actually would be incurred by the issuer to redeem them.
54. See, e.g., United States v. Grinnell Corp., 384 U.S. 563 (1966).
55. It is generally permissible under the antitrust laws today for banks (or other firms) to form cost-saving joint ventures that do not restrict competition between the venture's members. See, e.g., General Motors Corp., 103 F.T.C. 374 (1984). In some cases (e.g., where the members of the joint venture collectively lack market power) it could be procompetitive for joint venture members to agree to charge each other low prices. See discussion of ATM "surcharges," infra Part IV.B.
56. Indeed, even if a firm acquired or exercised market power through unlawful means, it still enjoys the protection of the antitrust laws concerning the unrelated and independent actions of others. See, e.g., Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, 340 U.S. 211 (1951).
57. Exceptions occurred during periodic liquidity crises, which also affected the redemption of bank notes.
58. Alternatively, an individual lacking an account with check-writing privileges could purchase a draft from the bank in the same way that banks today sell money orders and cashier's checks.
59. The practice of fractional reserve banking has been around for many centuries. In ancient Babylonia and Egypt, grain and other agricultural produce, and later precious metals, were deposited in storehouses controlled by royal or private bankers, who provided both lending and payment services. Deposit receipts and monetary contracts were among the first forms of (and perhaps the inspiration for) the practice of writing, itself. Deposits could be paid to the order of the depositor or a third party. See DAVIES, supra note 2, at ch. 2.
60. As of 1986, 70% of the 40 billion checks written annually involved interbank clearing. See Allen N. Berger & David B. Humphrey, Interstate Banking and the Payments System, 1 J. FIN. SERV. RES. 131 (1988).
61. This is valuable to a bank with local market power because it makes its apparent demand curve for checking (and other banking services) less elastic than would be the case if all fees were collected from the bank's customers directly. In the absence of transaction costs, distant banks would assess additional fees to their merchants to exactly cover the exchange fee, and distant merchants in turn would assess customers a surcharge at the point of sale equal to the exchange fee that the merchant would be assessed by its bank days later, when the customer's check cleared. Thus, transaction costs permit a monopolist local bank to make the demand of its local customers less elastic and increase its profits by exporting some of the incidence of its market power instead of exploiting it fully locally. Its profits likely will be maximized through some combination of supracompetitive local prices and supracompetitive exchange fees.
62. Absorption of exchange became a controversy unto itself. Bank clearinghouses in the early 1900s sometimes demanded that their members pass along exchange charges to their merchant customers. See MELVIN C. MILLER, THE PAR CHECK COLLECTION AND ABSORPTION OF EXCHANGE CONTROVERSIES 11 (1949). Such policies could have two effects. First, by forcing merchants to face the costs associated with their own customer banks' check redemption policies, the merchants were in turn perhaps more likely to alter their own polices towards acceptance of out-of-town checks. In other words, this could have represented an attempt to break the tendencies towards par acceptance that permitted country banks to export their market power. Second, absorption of exchange was perhaps a way in which banks could cheat on banking cartels organized through clearinghouses, in which banks had agreed upon fees and interest rates they would charge merchants. Mandating that banks pass along such charges would then be a way to enforce the cartel price. Whether absorption of exchange ran afoul of later federal controls on deposit interest rates (including a ban on the payment of interest on demand deposit accounts) was hotly debated in the 1940s.
63. While roughly 10% of banks were still nonpar check clearers in the 1960s, they were, on average, far smaller than the average par clearance bank, and in the aggregate accounted for a small fraction of total demand deposits. Therefore, the typical set of checks deposited into a city bank would contain only a very small fraction of nonpar checks. See PAUL F. JESSUP, THE THEORY AND PRACTICE OF NONPAR BANKING 3, 23-25 (1967).
64. There were thousands of different banks issuing bank notes and checking accounts, making it impossible to post separate prices for each possible exchange charge. Even when there are only a few possible payment methods, however, this tendency still is observed. See discussion of credit card surcharges and cash discounts, infra Part IV.A.
65. JAMES GRAHAM CANNON, CLEARING HOUSES 11-15 (Government Printing Office 1910). To be sure, clearinghouses also have many procompetitive routine functions and also helped mitigate the effects of banking panics prior to the introduction of federal deposit insurance.
66. Id. at 16.
67. See Daniel R. Fischel et al., The Regulation of Banks and Bank Holding Companies, 73 VA. L. REV. 301 (1987).
68. See, e.g., CANNON, supra note 49, at 72; SPAHR, supra note 27, at 103; MILLER, supra note 46, at 9; JESSUP, supra note 47, at 7-8, 93; Baxter, supra note 1, at 564; McAndrews, supra note 1, at 56.
69. Such a seemingly "inefficient" response to market power is similar to some types of "bypass competition" in telecommunications markets.
70. In Farmers' & Merchants' Bank of Cattlesburg, Ky. v. Federal Reserve Bank of Cleveland, Ohio, 286 F. 610 (E.D. Ky. 1922), the Federal Reserve's right to present checks for par collection over the counter was upheld, but the Fed was enjoined from using that right to coerce banks (e.g., through deliberate accumulation and sudden presentation of items in the manner of the Suffolk Bank) to join the Federal Reserve System and remit at par. In a subsequent case, the Supreme Court took a similar stance: the Federal Reserve banks had the same rights as any other banks to present items over the counter, but could not use this right to "embarrass" or coerce a nonpar bank by rendering the bank suddenly unable to meet its demand claims, thereby harming its reputation for safety. American Bank & Trust Co. v. Federal Reserve Bank of Atlanta, 262 U.S. 643 (1923).
71. JESSUP, supra note 47, at 23
72. Baxter, supra note 1, at 571.
73. Id. at 563.
74. Otherwise, we would not observe so-called "net acquirers" of ATM transactions still imposing fees on their customers who use terminals owned by other banks.
75. Baxter's claim that par collection is arbitrary also is inconsistent with the natural evolution under the common law of par collection of demand obligations over the counter.
76. Baxter claims that it would be inefficient for each bank, in competition with one another, to charge its own merchants and consumers for costs they impose when depositing or writing checks involving interbank settlement. Baxter, supra note 1, at 561. He bases this conclusion in part on a theoretical, joint-demand model in which merchants and consumers each have a separate demand to "complete transactions," with the number of actual transactions required to be equal for the two groups. Id. at 545. He concludes from this model that the "correct" exchange charge is unlikely to be zero, and therefore suggests that banks be permitted to act collectively to determine and assess appropriate uniform exchange charges. Baxter's theoretical model is confusing and leads to incorrect results. Any willingness to sell (i.e., a supply curve) can perhaps be thought of instead as a demand to complete a transaction at some established price, but it does little to advance an understanding of the phenomenon. For example, a seller of apples must in equilibrium desire to complete as many transactions as his customers, yet this hardly leads to the conclusion that collective setting of apple prices is desirable. A more appropriate economic model would probably be one of derived demand, not joint demand. Merchants' willingness to pay fees to their banks to complete transactions reflects a demand for payment services that is derived from the demand of consumers to buy the goods they sell. Baxter's argument, taken to its logical conclusion, leads to the extreme view that all banks should be permitted to act collectively to impose universal exchange fees on all interbank payment system transactions - even local clearance of bank notes and checks (i.e., the Common Law got it wrong by requiring payment of demand claims at par over the counter). Baxter's view apparently is that left to their own devices, even industry-wide associations of banks always will choose some "correct" and efficient exchange charge, whereas history leads one to the expectation that, where permitted, banks will choose an exchange fee that maximizes their monopoly profits from their collective provision and control of the payment system.
77. Hearings on S. 1642 and H.R. 3956 Before the Senate Comm. on Banking and Currency, 78th Cong., 2d Sess. (1945) (quoted in MILLER, supra note 46, at 119).
78. MILLER, supra note 46, at 120 (citing 1945 Senate Hearings at 80).
79. In practice, even the distinctions between financing methods for check and credit card transactions are becoming blurred. Many checking accounts offer "overdraft protection" lines of credit or are purely checkable lines of credit, and some credit cards are "secured" by customer deposits in the card-issuing bank.
80. In the United States, a "bank card" is a credit card issued by a financial institution member of one of the two bank card associations, MasterCard and Visa. Most bank card issuers belong to both Visa and MasterCard. Other credit cards, such as those offered by American Express, Dean Witter (Discover Card), and Citibank (Diners Club, Carte Blanche), are proprietary brands with a simpler flow of funds (because one financial institution handles both the customer and merchant sides of each transaction).
81. Today, authorizing, clearing, and settling credit card transactions typically happen through electronic data networks.
82. Banks also pay "switch" fees to the network to cover the network's own cost of operation. These fees are much smaller than interchange fees.
83. In recent years, the bank card associations have engaged in explicit price discrimination by enacting different interchange fees for different types of merchants. For example, supermarket transactions (generally a low profit margin business) are assessed a lower than average interchange fee. When the fee was uniform, most supermarkets chose not to accept credit cards. Discover Card, a proprietary card able to reduce its quoted merchant discount rates below the level of bank card interchange fees, helped induce the bank card associations' move towards industry-specific interchange fees. Value-based fees are an indication that interchange fees do not compensate card issuers for real marginal costs associated with the issuing side of the business; otherwise, it would not be profitable for issuers to cut the interchange rate for a particular merchant.
84. Davis S. Evans & Richard Schmalensee, Economic Aspects of Payment Card Systems and Antitrust Policy Toward Joint Ventures, 63 ANTITRUST L.J. 861, 890 (1995). Visa and MasterCard recently announced increases in their interchange fee rates. See Seven-Year Itch: Visa Is Raising Interchange Rates, CREDIT CARD NEWS, Aug. 1, 1997; MasterCard's Interchange Rates Boost Its World Card and Encourage Level 3, CREDIT CARD NEWS, Nov. 1, 1997.
85. Davis S. Evans & Richard Schmalensee, Economic Aspects of Payment Card Systems and Antitrust Policy Toward Joint Ventures, 63 ANTITRUST L.J. 861, 890 (1995). Visa and MasterCard recently announced increases in their interchange fee rates. See Seven-Year Itch: Visa Is Raising Interchange Rates, CREDIT CARD NEWS, Aug. 1, 1997; MasterCard's Interchange Rates Boost Its World Card and Encourage Level 3, CREDIT CARD NEWS, Nov. 1, 1997.
86. Davis S. Evans & Richard Schmalensee, Economic Aspects of Payment Card Systems and Antitrust Policy Toward Joint Ventures, 63 ANTITRUST L.J. 861, 890 (1995). Visa and MasterCard recently announced increases in their interchange fee rates. See Seven-Year Itch: Visa Is Raising Interchange Rates, CREDIT CARD NEWS, Aug. 1, 1997; MasterCard's Interchange Rates Boost Its World Card and Encourage Level 3, CREDIT CARD NEWS, Nov. 1, 1997.
87. If banks were permitted, but not required, to redeem credit card transactions at par, they individually would have little incentive to do so unless perhaps merchants could and would give rebates at the point of sale to that bank's customers - which is unlikely for the reasons giving rise to Gresham's Law and its restatements, discussed above, and because the bank card associations prohibit differential treatment to customers based on the identity of the card issuer.
88. The competitive constraint this imposes on the associations is imperfect because bank card issuers in general cannot expect anything close to 100% on-us transactions. For the remaining transactions handled by the merchant, the bank will have to pay the association's mandated interchange fee to the cardholder's bank.
89. National Bancard Corp. v. VISA U.S.A., Inc., 596 F. Supp 1231 (S.D. Fla. 1984), aff'd, 779 F.2d 592 (11th Cir. 1986). For a critique of the NaBanco decision, see Carlton & Frankel, supra note 16, at 652-53, 655-61.
90. One defense of interchange fees raised by Baxter, NaBanco, and subsequent commentators is that collectively set interchange fees are far more efficient than requiring a multitude of bilateral agreements between each possible pair of banks. See Baxter, supra note 1, at 576; Evans & Schmalensee, supra note 68, at 892; Luc Gyselen, EU Antitrust Law in the Area of Financial Services, in FORDHAM CORP. L. INST. 23D ANNUAL CONF. ON INT'L ANTITRUST LAW AND POLICY 27-28 (Barry Hawk ed., 1997). This superficially appealing argument, however, accepts the premise that the efficient, competitive outcome in payment markets must include some interbank exchange fee. If each financial institution party to the transaction can recover costs directly from customers, then exchange fees probably are not necessary at all for efficiency.
91. Baxter, supra note 1, at 572-82; Evans & Schmalensee, supra note 68, at 887-97; Gyselen, supra note 72, at 27-28.
92. Unlike the situation in Broadcast Music, Inc. v. CBS, Inc., 441 U.S. 1 (1979), in which individual artists could not effectively monitor or negotiate royalty fees with every user of their music, in credit card networks the card issuer enters into contractual agreements with each of its own customers, always knows when they use their accounts, and can recover all privately incurred costs through direct charges on these customers.
93. Baxter states: In four-party payment mechanisms . . . a side payment between [consumer] and [merchant], coupled with payment each by [consumer] and [merchant] to [consumer's bank] and [merchant's bank], respectively, in amounts equal to respective bank costs . . . is theoretically sufficient to attain equilibrium. That in practice side payments between banks occur instead is strong evidence that higher transaction costs characterize side payments that take the form of price adjustments between the principals. Baxter, supra note 1, at 554. See also Carlton & Frankel, supra note 16, at 660-61; Evans & Schmalensee, supra note 68, at 895.
94. For a discussion of the effects of this phenomenon on the adoption of debit card systems by merchants, see John P. Caskey & Gordon H. Sellon, Jr., Is the Debit Card Revolution Finally Here?, FED. RESERVE BANK OF KAN. CITY ECON. REV. 79 (Fourth Quarter 1994).
95. Even without interchange fees, the merchant discount would be borne by all consumers, but it would be a much smaller amount and due entirely to competitively, not collusively, determined costs.
96. For details, see Carlton & Frankel, supra note 16, at 660. Equal pricing for different payment methods has the same effect as would likely result from a contractual requirement that cans of a popular Brand A soft drink not be sold at any greater price than cans of a competing Brand B - a form of variable maximum resale price maintenance. Such a requirement can, at least in some circumstances, permit the owner of Brand A to increase prices and profits. Facing a higher price for Brand A, the merchant faces an all-or-nothing choice: drop Brand A altogether or raise the price of both Brand A and Brand B to reflect the now higher average cost of the two brands. The manufacturer will lose fewer sales due to a price increase with this restriction in place, at least in stores not dropping the popular Brand A altogether. Not all of its price increase will be passed along to its consumers at retail, because consumers buying the competing Brand B fund the cost increase to exactly the same extent per can as consumers purchasing the higher cost Brand A. The relative prices of the brands to consumers remain unchanged, so consumers have no incentive to avoid the now higher cost product. Only if a sufficiently large number of merchants are willing to drop Brand A will the combination of the retail price restriction and wholesale price increase be unprofitable.
97. Contractual restrictions go beyond prohibitions of credit card surcharges. Credit card organizations often include in their agreements (or insist that their member banks include in their agreements with merchants) clauses that prevent the merchant from discriminating in any way against the brand at the point of sale, in favor of other brands.
98. Though interchange fees themselves enjoy some immunity from antitrust challenge in Europe, the combination of interchange fees and vertical price restrictions has been recognized as troubling in connection with European payment systems. Gyselen, supra note 72, at 30-35.
99. The net impact on credit card customers is ambiguous. Credit card customers also pay retail prices inflated by the cost of interchange payments, yet gain not only payment services, but also rebates and other benefits, such as frequent flier awards and rebates on automobiles that can be quite valuable.
100. There are some significant exceptions. The litigation between Dean Witter and Visa involved Visa's refusal to admit a financial institution owned by Dean Witter because Dean Witter also competed with its own Discover Card brand. Conditioning membership on the vigor (or lack thereof) and manner in which potential members compete can have significant anticompetitive effects similar to those resulting from the nineteenth century check clearinghouses' practice of expelling a bank offering high deposit interest rates or assessing low intercity exchange charges. SCFC, ILC, Inc. v. Visa U.S.A., Inc., 819 F. Supp 956 (D. Utah 1993), rev'd in part and aff'd in part, 36 F.3d 958 (10th Cir. 1994). (The author served as a consultant to Dean Witter, parent of SCFC.) Dennis Carlton and I found the evidence compelling that consumers were harmed by Dean Witter's exclusion. Carlton & Frankel, supra note 14, at 662-68; Carlton & Frankel, supra note 71, at 904. Other analysts and the Tenth Circuit disagreed. See, e.g., David A. Balto, Antitrust and Credit Card Joint Ventures, 47 CONSUMER FIN. L.Q. REP. 266 (1993); Evans & Schmalensee, supra note 68, at 861.
101. That is not to say that the market for checking services is not intensely competitive. There is vigorous competition by banks seeking the deposits of individuals and businesses, which often is reflected in extended branch operations and innovative services, such as computer or phone banking. Some might argue that it is high interest earnings, not interchange fees, that create the huge solicitation efforts for credit cards. However, the distinction is unimportant. Any supracompetitive profits from either source will have this effect. In fact, aggregate interchange fee revenue collected by banks are a substantial fraction of aggregate issuer profits, if not revenue. For example, Ausubel reported that for one "typical" large credit card bank, interchange income totaled 3.06%, 3.00%, and 2.92% of outstanding balances in the years 1985, 1986, and 1987, respectively, while in the same years net income in the bank's credit card business overall totaled 7.09%, 6.63%, and 4.80% of outstanding balances. Lawrence M. Ausubel, The Failure of Competition in the Credit Card Market, 81 AM. ECON. REV. 50, 58 (1991). Moreover, despite the conventional wisdom that it is unprofitable to serve "transactors" - those customers who do not revolve a balance but instead pay off their statement in full each month - issuers continue to carry large numbers of them as customers, solicit customers likely to be transactors (e.g., the American Bar Association affinity credit card), and even sometimes make available to them significant rebates based on their purchases, not revolving balances. The dissipation of market power rents through higher induced solicitation costs, if that is what is occurring, could represent a significant deadweight loss to society similar to that which has sometimes occurred in other markets with regulated prices, such as banking (e.g., the once proverbial free toaster with deposits) and airlines (low load factors, excessive quality). See, e.g., GEORGE W. DOUGLAS & JAMES C. MILLER III, ECONOMIC REGULATION OF DOMESTIC AIR TRANSPORT: THEORY AND POLICY (1974).
102. That is not to say that the market for checking services is not intensely competitive. There is vigorous competition by banks seeking the deposits of individuals and businesses, which often is reflected in extended branch operations and innovative services, such as computer or phone banking. Some might argue that it is high interest earnings, not interchange fees, that create the huge solicitation efforts for credit cards. However, the distinction is unimportant. Any supracompetitive profits from either source will have this effect. In fact, aggregate interchange fee revenue collected by banks are a substantial fraction of aggregate issuer profits, if not revenue. For example, Ausubel reported that for one "typical" large credit card bank, interchange income totaled 3.06%, 3.00%, and 2.92% of outstanding balances in the years 1985, 1986, and 1987, respectively, while in the same years net income in the bank's credit card business overall totaled 7.09%, 6.63%, and 4.80% of outstanding balances. Lawrence M. Ausubel, The Failure of Competition in the Credit Card Market, 81 AM. ECON. REV. 50, 58 (1991). Moreover, despite the conventional wisdom that it is unprofitable to serve "transactors" - those customers who do not revolve a balance but instead pay off their statement in full each month - issuers continue to carry large numbers of them as customers, solicit customers likely to be transactors (e.g., the American Bar Association affinity credit card), and even sometimes make available to them significant rebates based on their purchases, not revolving balances. The dissipation of market power rents through higher induced solicitation costs, if that is what is occurring, could represent a significant deadweight loss to society similar to that which has sometimes occurred in other markets with regulated prices, such as banking (e.g., the once proverbial free toaster with deposits) and airlines (low load factors, excessive quality). See, e.g., GEORGE W. DOUGLAS & JAMES C. MILLER III, ECONOMIC REGULATION OF DOMESTIC AIR TRANSPORT: THEORY AND POLICY (1974).
103. See, e.g., Lester Telser, Why Should Manufacturers Want Fair Trade?, 3 J.L. & ECON. 86 (1960).
104. The interchange fee "tax" on cash paying customers that funds the benefits provided to credit card customers is probably regressive, because the poor use cash relatively more and credit cards relatively less than the wealthy. Visa claims the opposite, stating that it is credit card surcharges, not interchange fees, that are regressive: "surcharges represent a tax limited to members of the public who choose to use payment cards. This 'tax' is likely to fall more heavily on low- and moderate-income citizens, because they are the ones most likely to need the flexibility of credit card use when faced with an unexpected bill." VISA FAQs About Surcharges, (visited Apr. 21, 1998) 〈http://www.visa.com/cgi-bin/ vee/fb/merch/biz/govt/surcharge.html?2+0〉. While over 90% of families with annual income exceeding $100,000 have at least one general purpose credit card, however, only 26% of Americans with annual family income less than $10,000 have one. Usage of General Purpose Credit Cards by Families: 1989 and 1992, STATISTICAL ABSTRACT OF THE UNITED STATES, Table 811 (1995).
105. The interchange fee "tax" on cash paying customers that funds the benefits provided to credit card customers is probably regressive, because the poor use cash relatively more and credit cards relatively less than the wealthy. Visa claims the opposite, stating that it is credit card surcharges, not interchange fees, that are regressive: "surcharges represent a tax limited to members of the public who choose to use payment cards. This 'tax' is likely to fall more heavily on low- and moderate-income citizens, because they are the ones most likely to need the flexibility of credit card use when faced with an unexpected bill." VISA FAQs About Surcharges, (visited Apr. 21, 1998) 〈http://www.visa.com/cgi-bin/ vee/fb/merch/biz/govt/surcharge.html?2+0〉. While over 90% of families with annual income exceeding $100,000 have at least one general purpose credit card, however, only 26% of Americans with annual family income less than $10,000 have one. Usage of General Purpose Credit Cards by Families: 1989 and 1992, STATISTICAL ABSTRACT OF THE UNITED STATES, Table 811 (1995).
106. Credit card issuers have attempted to refute charges that cash customers subsidize credit card customers. See, e.g., Statement of American Bankers Ass'n on S. 414 (Cash Discount Act) Before the Subcomm. on Consumer Affairs, Senate Comm. on Banking, Housing and Urban Affairs 34 (Feb. 18, 1981). Visa maintains that credit card surcharges "require payment card users to pay not only for the costs of card acceptance, but also to pick up a portion of the costs of cash and checks, since these costs are not isolated and are charged directly to the user." FAQs About Surcharges, supra note 85. But if the cost of accepting credit cards does not exceed the cost of accepting cash or checks, then merchants will have no interest in surcharging credit instead of cash. Visa's insistence that credit cards do not cost merchants more than cash and checks is inconsistent with Visa's apparent need to forbid credit card surcharges. Ironically, my analysis suggests that embracing the cash-to-credit subsidy effect is perhaps the only economically sensible way Visa could try to defend interchange fees as necessary to correct some market failure.
107. Credit card issuers have attempted to refute charges that cash customers subsidize credit card customers. See, e.g., Statement of American Bankers Ass'n on S. 414 (Cash Discount Act) Before the Subcomm. on Consumer Affairs, Senate Comm. on Banking, Housing and Urban Affairs 34 (Feb. 18, 1981). Visa maintains that credit card surcharges "require payment card users to pay not only for the costs of card acceptance, but also to pick up a portion of the costs of cash and checks, since these costs are not isolated and are charged directly to the user." FAQs About Surcharges, supra note 85. But if the cost of accepting credit cards does not exceed the cost of accepting cash or checks, then merchants will have no interest in surcharging credit instead of cash. Visa's insistence that credit cards do not cost merchants more than cash and checks is inconsistent with Visa's apparent need to forbid credit card surcharges. Ironically, my analysis suggests that embracing the cash-to-credit subsidy effect is perhaps the only economically sensible way Visa could try to defend interchange fees as necessary to correct some market failure.
108. The anticompetitive interpretation of interchange fees presented here is similar to Mark Patterson's discussion of cases in which market power is exercised against customers of another seller's product. Mark Patterson, Coercion, Deception, and Other Demand Increasing Practices in Antitrust Law, 66 ANTITRUST L.J. 1, 34 (1997). Other examples Patterson provides of allegedly anticompetitive "demand increasing" activities are more problematic, insofar as they do not involve shifting costs to customers of another seller's product.
109. In addition to its traditional green, gold, and platinum charge cards, American Express is the issuer of Optima brand credit cards. Morgan Stanley, Dean Witter, Discover & Co. is the owner of the NOVUS network and issuer of several proprietary credit card brands, including Discover Card.
110. Similar reasoning was used by the U.S. Department of Justice to explain why it had agreed to a five-year phase-out of the remaining portions of the 1956 IBM Consent Decree. According to the Department's press release, "By extending the life of the decree's main provisions into the next century, the Department ensures that those who have made important business decisions based on the decree will have time to make adjustments before it expires." Justice Department Agrees to Terminate Last Provisions of IBM Consent Decree in Stages Ending 5 Years from Today, U.S. Department of Justice Press Release 96-324 (July 2, 1996).
111. Even where merchants have the legal and contractual right to charge different prices for cash and credit transactions, they generally do not exercise it. For example, Amoco Oil recently abandoned its two-tier pricing system, and few major merchants still charge less for cash (though many consumers likely have experienced informal "discount for cash" offers from small merchants).
112. See, e.g., David A. Balto, supra note 82, at 266; Gyselen, supra note 72, at 28-30.
113. Banks can sometimes join multiple regional networks, though this, too, has generated antitrust controversies. See, e.g., United States v. Electronic Payment Servs., Inc., No. 94-208 (D. Del. Apr. 21, 1994) (Complaint). ATM terminals usually are connected to a national network in addition to a local or regional network.
114. Technically, ATM networks might not be considered a "payment" technology and these profits not "seignorage" because they do not complete a retail transaction. But they do facilitate those transactions by reducing the cost of completing some of the steps that lead to a complete transaction.
115. A card-issuing bank wishing to provide good customers with inexpensive access to the ATM network could offer to credit their accounts by a certain amount when they use a "foreign" ATM, or, if the network data protocols identified fees separately, could reimburse the customer for actual fees incurred in total or up to some contractual limit. Explicitly identifying the size of the fee separately from the size of the transaction itself could be important because odd-sized withdrawals do occur through machines used in some retail locations, and an increasing number of retail locations now permit direct debit transactions using ATM cards. Depending on how the network is configured, issuers may or may not now be able to identify fees separately from the transaction debit.
116. Some claim that consumer acceptance of ATM fees even depends on whether the price is posted in signage next to the terminal or only on the terminal's screen after the card has been inserted. See Hearing on Automatic Teller Machine Fees and Surcharges Before the Senate Comm. on Banking, Housing and Urban Affairs (June 11, 1997) (prepared testimony of Edmund Mierzwinski). While Mierzwinski uses that claim to criticize surcharge fee notices on ATM screens, he is silent about the effect of interchange fees that show up only as foreign fees much later on bank statements.
117. The effect of a fixed interchange fee, if issuers always simply passed the fees along to customers, would be identical to an explicit horizontal agreement to fix the retail price at that level.
118. This is changing somewhat, as more ATM cards are accepted by retailers in payment for goods and services. In this use, ATM cards function much the same as general-purpose debit cards.
119. A problem arises, however, when networks seek to extend the functionality of their ATM cards by making them (or alternative cards) usable as general-purpose debit cards at retail locations. There is no fundamental economic distinction between ATM and debit transactions. Instead of disbursing currency, debit merchants dispense gasoline, groceries, or other goods. The logic underpinning the ATM network interchange fee works against the banks in this situation. The terminals that, in the ATM networks' paradigm, require collectively determined fee income to survive are in the retail debit environment typically owned by the retailer itself and not a bank. Consistency would require that banks permit retailers to earn interchange fee revenue to subsidize their provision of debit-enabled terminals. Not surprisingly, many bank networks have opted instead to implement interchange fees for retail debit transactions that flow to card issuers, not terminal owners. Some debit card networks have also begun implementing bans on merchant surcharges on debit transactions, paralleling similar contractual restrictions on credit card sales. Such restrictions reinforce the tendencies discussed above for all forms of money to circulate at par at retail. At the very least, this inconsistency calls into question any claim that networks tend to set interchange fees in a manner that maximizes economic efficiency. Indeed, it suggests that the primary purpose of these collectively imposed fees is to transfer revenue to banks. Recently, a coalition of retailers filed suit against Visa, alleging that Visa has violated the antitrust laws by requiring that merchants who accept Visa credit card transactions also accept Visa debit card transactions, without imposing any customer surcharge. See Dirty Business Part Uno, RAM RESEARCH CARDTRAK (Nov. 1996).
120. One reason sometimes given for why networks assessed interchange fees rather than direct terminal charges was that it was important initially that consumers not pay for using ATMs if ATM usage was to gain popularity. Of course, they were paying, through higher account fees, higher minimum balances, lower interest rates, etc., or for many consumers by the late 1980s, through foreign fees. By now, ATM use is widespread and it is difficult to argue that consumers will not rely upon the networks if the fees they pay are not hidden from them.
121. In Canada, for example, the major banks that formed the Interac nationwide ATM network set an interchange fee of $0.75 for interbank cash withdrawals, yet typically assessed foreign fees of $1.00 to customers who cause the bank to incur those fees. U.S. network interchange fees for cash withdrawals reviewed in 1994 ranged between $0.30 and $0.60 (with an additional $0.02-$0.18 "switch fee" assessed to cover the costs of operating the network). See BANK NETWORK NEWS (Mar. 25, 1994). Foreign fees per withdrawal were nonetheless often $1.00 or more. The incremental cost to a card-issuing bank of processing an ATM withdrawal is likely less than the incremental cost of processing a check transaction, and much less than this markup over the interchange and switch fees. Thus, it appears that customers have long paid "surcharges" for the convenience of ATM terminals, but in many cases it is issuing banks that are receiving the additional compensation.
122. Networks sometimes argue that uniform pricing is another type of external benefit achieved with an interchange fee system, but antitrust does not (and probably should not) accept pricing uniformity as an adequate defense for collective rate setting absent compelling evidence, such as the significant transaction efficiencies underlying the uniform collection of royalties in Broadcast Music, Inc., 441 U.S. 1. After all, all effective cartels can boast of uniform rates, as did the check clearinghouse cartels of a century ago.
123. The first major challenge involved the PULSE network of Texas. First Texas Savings Association sought to eliminate the PULSE interchange fee and replace it with an all-surcharge compensation system, to enable it to profitably emplace nearly 1,000 additional terminals. The arbitrator (Thomas Kauper) compromised, reducing the interchange fee and permitting terminal owners to assess surcharges or offer rebates. See In re Arbitration Between First Tex. Sav. Ass'n and Fin. Interchange, Inc., 55 Antitrust & Trade Reg. Rep. (BNA) 340 (Aug. 25, 1988). A second challenge to surcharge prohibitions involved Valley Bank of Nevada. This bank sought to assess surcharges at ATM terminals installed in casinos. See Valley Bank of Nev. v. Plus Sys., Inc., 749 F. Supp. 223 (D. Nev. 1989), aff'd, 914 F.2d 1186 (9th Cir. 1990). While these locations are likely more costly than bank locations, the surcharges probably reflect as well significant locational rents available to the casino and the ATM vendor due to the convenience to customers of on-site ATMs. Nevada passed legislation that required networks to permit members to surcharge, mooting the litigation, and other states soon enacted similar laws. In 1995 another surcharge case was litigated, and this time a decision was rendered in favor of the network (PLUS) and its prohibition on surcharges. See SouthTrust v. Plus Sys., CV-93-P-2291-S (N.D. Ala., 1995). Shortly after its victory, however, PLUS (followed by many other networks) eliminated its surcharge prohibition. For an overview of these cases and some of the arguments typically raised in the surcharge debate, see David A. Balto, ATM Surcharges: Panacea or Pandora's Box?, 12 REV. BANKING & FIN. SERV. 169 (Oct. 9, 1996).
124. The first major challenge involved the PULSE network of Texas. First Texas Savings Association sought to eliminate the PULSE interchange fee and replace it with an all-surcharge compensation system, to enable it to profitably emplace nearly 1,000 additional terminals. The arbitrator (Thomas Kauper) compromised, reducing the interchange fee and permitting terminal owners to assess surcharges or offer rebates. See In re Arbitration Between First Tex. Sav. Ass'n and Fin. Interchange, Inc., 55 Antitrust & Trade Reg. Rep. (BNA) 340 (Aug. 25, 1988). A second challenge to surcharge prohibitions involved Valley Bank of Nevada. This bank sought to assess surcharges at ATM terminals installed in casinos. See Valley Bank of Nev. v. Plus Sys., Inc., 749 F. Supp. 223 (D. Nev. 1989), aff'd, 914 F.2d 1186 (9th Cir. 1990). While these locations are likely more costly than bank locations, the surcharges probably reflect as well significant locational rents available to the casino and the ATM vendor due to the convenience to customers of on-site ATMs. Nevada passed legislation that required networks to permit members to surcharge, mooting the litigation, and other states soon enacted similar laws. In 1995 another surcharge case was litigated, and this time a decision was rendered in favor of the network (PLUS) and its prohibition on surcharges. See SouthTrust v. Plus Sys., CV-93-P-2291-S (N.D. Ala., 1995). Shortly after its victory, however, PLUS (followed by many other networks) eliminated its surcharge prohibition. For an overview of these cases and some of the arguments typically raised in the surcharge debate, see David A. Balto, ATM Surcharges: Panacea or Pandora's Box?, 12 REV. BANKING & FIN. SERV. 169 (Oct. 9, 1996).
125. See The Fair ATM Fees for Consumers Act, S. 1800, 104th Cong., 2d Sess. (May 23, 1996); S. 885, 105th Cong., 1st Sess. (June 11, 1997). A confusing factor in this debate is the terminology. By labeling directly priced ATM services "surcharges," while accepting interchange fees and foreign fees as a normal pricing practice, there is a predictable tendency among the public and their elected representatives to focus their attention and hostility on the former. For example, one Senate opponent of direct ATM terminal fees decries them as an "anti-consumer, anti-competitive, and anti-free market practice - double ATM fees," and claims that "ATM double-charging is a monopolistic practice that eliminates competition and distorts the free market." Hearing on the General Accounting Office Report on ATM Surcharges, Senate Banking Comm. (June 11, 1997) (Prepared Opening Statement of Senator Alfonse M. D'Amato (R-NY)). It is ironic and misleading to suggest that in a market in which consumers are paying both collectively imposed and independently determined prices that the collective price setting represents the "free market" and independent pricing is "monopolistic."
126. See, e.g., DENNIS W. CARLTON & JEFFREY M. PERLOFF, MODERN INDUSTRIAL ORGANIZATION ch. 14 (1994).
127. This was the system advocated by the plaintiff in the PULSE arbitration. See Steven C. Salop, Deregulating Self-Regulated Shared ATM Networks, 1 ECON. INNOVATION & NEW TECH. 85 (1990).
128. A surcharge ban functions much like collusive, industry-wide maximum resale price maintenance, much as would occur if all soft drink companies agreed among themselves to enforce maximum vending machine prices. While such a practice should not be condemned per se under the antitrust laws, as its direct effect is to keep prices lower, there are clearly some circumstances in which inefficient and anticompetitive effects can occur, particularly when firms with collective market power are setting the maximum price collectively. David Balto has pointed out to me a possible similarity between deregulated ATM prices and deregulated pay phone prices. Such a comparison is fair, but only to a point. Deregulated pay phone service suffers an important defect in some cases - lack of adequate price disclosure at the point of sale. This defect is perhaps more easily avoided at ATM terminals.
129. Written Statement of the Electronic Funds Transfer Association, Prepared for the U.S. Senate Comm. on Banking, Housing, and Urban Affairs (June 11, 1997).
130. For a demonstration of how an output test can be used to evaluate the net competitive effects of higher prices that are accompanied with greater terminal deployment, see Dennis W. Carlton & Alan S. Frankel, Antitrust and Payment Technologies, 77 FED. RESERVE BANK OF ST. LOUIS REV. 42 (1995).
131. See, e.g., Balto, supra note 102, at 173-74; Mierzwinski, supra note 95.
132. In Canada's concentrated banking market, the large banks that established the nationwide Interac ATM network did not permit shared deposit or balance inquiry services, thus denying small financial institutions the deposit-gathering potential they otherwise would have enjoyed. Part of the relief obtained by the government in the Interac settlement included a provision intended to at least permit subsets of Interac members to agree with one another to use their Interac software to accept shared deposit services, shared balance inquiry services, or other ATM services. The Interac consent agreement also permitted surcharges and rebates, but forbid members from assessing differential surcharges to customers depending upon which competitor's account was being accessed. See Competition Tribunal Consent Order in The Director of Investigations and Research v. Bank of Montreal et. al. (June 20, 1996). (The author served as consultant to the Director of Investigations and Research.)
133. See U.S. GENERAL ACCOUNTING OFFICE, REPORT TO THE CHAIRMAN, COMMITTEE ON BANKING, HOUSING, AND URBAN AFFAIRS, U.S. SENATE, AUTOMATIC TELLER MACHINES: BANKS REPORTED THAT USE OF SURCHARGE FEES HAS INCREASED, GAO/GGD-97-90 (May 1997).
134. See InterCept Switch Introduced as Surcharge-Free Network, Bus. WIRE (Atlanta) (Aug. 5, 1997); Testimony of Edmund Mierzwinski, supra note 95.
135. See InterCept Switch Introduced as Surcharge-Free Network, Bus. WIRE (Atlanta) (Aug. 5, 1997); Testimony of Edmund Mierzwinski, supra note 95.
136. Individual card-issuing banks could still assess their own fee just as they now often mark up the interchange fee when setting their foreign fee. Perhaps for good measure, the use of the term "surcharge" can be reserved for any foreign fees assessed by banks on their customers despite the absence of an interchange fee.
137. In other words, a network lacking market power would be exempt from the prohibition on collective interchange fee determination, and networks with market power would not be permitted to require membership exclusivity.
138. Even in such sub-networks it would be desirable to require that all fees be disclosed and assessed directly rather than hidden through the interchange fee and foreign fee mechanism. It might also make sense to require that a transaction takes place using the lowest price path where multiple networks are available.
139. The problem is in part to avoid a situation where a subgroup of banks forms with a share of deposits and terminals sufficiently high that the remaining banks cannot form an efficient network of their own. This was part of the problem in Canada.
140. For example, some networks are for-profit enterprises, while others are structured as non-profit joint ventures or associations. My proposal differs in key respects from proposals to maintain "intersystem" competition by refusing to permit network mergers to assemble universal regional networks in the first place (a now mostly moot issue), or to promote competition by prohibiting duality and requiring that each bank belong to only one of the available competing networks. See, e.g., Donald I. Baker, Shared ATM Networks - The Antitrust Dimension, 77 FED. RESERVE BANK OF ST. LOUIS REV. 5 (1995); David A. Balto, Payment Systems and Antitrust: Can the Opportunities for Network Competition Be Recognized?, 77 FED. RESERVE BANK OF ST. LOUIS REV. 19 (1995). Instead, this approach takes full advantage of network duality to generate more competition.
141. For example, some networks are for-profit enterprises, while others are structured as non-profit joint ventures or associations. My proposal differs in key respects from proposals to maintain "intersystem" competition by refusing to permit network mergers to assemble universal regional networks in the first place (a now mostly moot issue), or to promote competition by prohibiting duality and requiring that each bank belong to only one of the available competing networks. See, e.g., Donald I. Baker, Shared ATM Networks - The Antitrust Dimension, 77 FED. RESERVE BANK OF ST. LOUIS REV. 5 (1995); David A. Balto, Payment Systems and Antitrust: Can the Opportunities for Network Competition Be Recognized?, 77 FED. RESERVE BANK OF ST. LOUIS REV. 19 (1995). Instead, this approach takes full advantage of network duality to generate more competition.
142. Price coherence might be uncommon, but it is a key economic characteristic in payment system markets. There are likely to be other markets in which this phenomenon can be observed, and the analysis of antitrust disputes in such markets must recognize the importance of the resulting ability of a seller to shift the incidence of its market power away from its own customers. One recent case that, in part, might be related to price coherence is the FTC's decision to block Rite-Aid's attempted acquisition of Revco. In that case, the institutional constraints imposed by managed health care companies would likely mean that consumers would have little incremental incentive to choose the lowest cost member of an insurer's provider network. The network was faced with an all-or-nothing choice whether to accept a pharmacy chain in its entirety into its network. If it accepted, some of the incidence of any market power would be borne by consumers who, despite patronizing other pharmacy members of the network that charged the network lower prices, nevertheless shouldered the same higher health care premiums and co-payments as did customers who patronized the dominant and more expensive pharmacy chain. The dominant provider thus lost no sales, compared to a hypothetical no-transaction-cost alternative in which consumers paid a surcharge for any higher price incurred by the network for prescriptions filled by the dominant chain.
143. See, e.g., BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM, REPORT TO CONGRESS ON THE APPLICATION OF THE ELECTRONIC FUND TRANSFER ACT TO ELECTRONIC STORED-VALUE PRODUCTS (March 1997); Jeffrey M. Lacker, Stored Value Cards: Costly Private Substitutes for Government Currency, 82 FED. RESERVE BANK OF RICHMOND ECON. Q. 1 (1996); John Wenninger & David Laster, The Electronic Purse, 1 CURRENT ISSUES IN ECONOMICS AND FINANCE, FED. RESERVE BANK OF N.Y. 1 (1995); Adam M. Zeretsky, Will that Be Cash, Check, Charge or Smart Card?, REGIONAL ECON. 5 (1996).
144. See, e.g., BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM, REPORT TO CONGRESS ON THE APPLICATION OF THE ELECTRONIC FUND TRANSFER ACT TO ELECTRONIC STORED-VALUE PRODUCTS (March 1997); Jeffrey M. Lacker, Stored Value Cards: Costly Private Substitutes for Government Currency, 82 FED. RESERVE BANK OF RICHMOND ECON. Q. 1 (1996); John Wenninger & David Laster, The Electronic Purse, 1 CURRENT ISSUES IN ECONOMICS AND FINANCE, FED. RESERVE BANK OF N.Y. 1 (1995); Adam M. Zeretsky, Will that Be Cash, Check, Charge or Smart Card?, REGIONAL ECON. 5 (1996).
145. See, e.g., BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM, REPORT TO CONGRESS ON THE APPLICATION OF THE ELECTRONIC FUND TRANSFER ACT TO ELECTRONIC STORED-VALUE PRODUCTS (March 1997); Jeffrey M. Lacker, Stored Value Cards: Costly Private Substitutes for Government Currency, 82 FED. RESERVE BANK OF RICHMOND ECON. Q. 1 (1996); John Wenninger & David Laster, The Electronic Purse, 1 CURRENT ISSUES IN ECONOMICS AND FINANCE, FED. RESERVE BANK OF N.Y. 1 (1995); Adam M. Zeretsky, Will that Be Cash, Check, Charge or Smart Card?, REGIONAL ECON. 5 (1996).
146. See, e.g., BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM, REPORT TO CONGRESS ON THE APPLICATION OF THE ELECTRONIC FUND TRANSFER ACT TO ELECTRONIC STORED-VALUE PRODUCTS (March 1997); Jeffrey M. Lacker, Stored Value Cards: Costly Private Substitutes for Government Currency, 82 FED. RESERVE BANK OF RICHMOND ECON. Q. 1 (1996); John Wenninger & David Laster, The Electronic Purse, 1 CURRENT ISSUES IN ECONOMICS AND FINANCE, FED. RESERVE BANK OF N.Y. 1 (1995); Adam M. Zeretsky, Will that Be Cash, Check, Charge or Smart Card?, REGIONAL ECON. 5 (1996).