Cash is increasingly being displaced by private forms of payment. Currently the U.S. economy functions with a minimal stock of cash, probably amounting to less than 2.6 percent of its annual gross domestic product (GDP)[1] This figure is markedly less than historical estimates for the United States (for example, about 3.2 percent in 1960) or contemporary estimates for other countries (as high as 4.9 percent for some European countries, according to Humphrey 2002). Roughly three-quarters of all transactions still take place on a cash basis (Committee on the Federal Reserve in the Payments Mechanism 1998), but the average amount of a cash-based transaction is small, probably less than $10.[2] When payment technologies are compared on a value basis, payments based on the transfer of “inside money” (payments by check, payment card, or direct transfer) dominate, accounting for the vast majority of the value of transactions within the United States.[3]
Payment in inside money is, of course, hardly a recent phenomenon. By the fourteenth century, European merchants had discovered the essential advantage of inside money: Exchange using debt ties up fewer resources than does the exchange of costly coin.[4] Since not everyone’s debt is likely to be equally reliable, however, inside-money payment systems have historically singled out the debt of a select group of “strong credits” (banks) as closer proxies for commodity (or outside) money. These privileged forms of debt possess the moneylike property of finality—of being able to extinguish other debts by virtue of their transfer from debtor to creditor.[5] However, the limitation of this privilege to certain strong credits also imposes constraints on those parties whose debt does not qualify as money. Hence, there has been an incentive to extend the reach of inside money with payment devices of limited finality, such as the check. Such instruments can broaden the benefits of inside money but may also increase the risk of default or fraud.
Monetary history is punctuated by innovations deposit banking, checks, banknotes, credit cards—that have expanded the role of inside money. For example, in recent years technology has made it possible for virtually anyone with a credit or debit card to pay for any purchase (from a merchant with an account with a credit card company) anywhere with a relatively high degree of finality. In many situations, card-based payment systems have offered considerable improvements over their paper-based predecessors.[6] A merchant selling a good to an unfamiliar customer can accept a card payment with the confidence that such payment is usually, if not completely, final.[7]. Payment by check would not offer the merchant the same degree of finality, and requiring cash payment could deny customers access to credit.
The finality associated with card payments does not extend to every transaction environment, however. Payment cards, and especially credit cards, are often used in situations—such as mail order, telephone, and Internet transactions—in which the cardholder is not present and cannot sign a receipt. In such cases the risk of fraud is elevated, but little of this risk is borne by credit card holders because (under U.S. law at least) their liability is limited to $50 and in practice is often zero.[8] A credit card holder may also withhold payment if he believes he has been charged for goods or services that were not delivered or were defective. In such circumstances, offering blanket guarantees of payment finality to merchants would create an unmanageable risk for card issuers. Instead, merchants bear most of the fraud risk in the form of liability for chargebacks (debits to a merchant’s account resulting from disputed payments) from the card issuers. This risk allocation has made “cardholder not present” credit card payment more expensive and generally less attractive to merchants unwilling to accept the risk of chargebacks. Internet transactions seem especially at risk, and this riskiness is reflected in fraud rates for on-line transactions. Trade publications have reported rates of credit card fraud as high as 2.1 percent for Web-based transactions, roughly ten times the rate for face to-face transactions.[9]
The past few years have seen the debut of several new types of on-line payment arrangements, at least partly in response to the difficulties associated with card-based payment over the Internet. These arrangements offer the promise of making it possible for anyone to pay anyone on-line, even in situations in which card-based payment would be infeasible or uneconomical. The most innovative arrangements, sometimes referred to as on-line currencies, bypass the traditional, bank-based methods for clearing and settlement of payments in favor of a simple “on-us” funds transfer—that is, a transfer of a claim on the on-line currency issuer (in the form of an account balance) from payor to payee.[10] While the finality of such transfers has thus far been of a limited nature, the most successful on-line currency issuer, PayPal, now offers its users finality guarantees under some circumstances.
What is the future of this type of payment arrangement? To date, industry reviews have been mixed. Most observers concede that on-line currencies have offered a useful service for person-to-person on-line transactions, most typically those associated with on-line auctions. On the other hand, on-line currencies have seen relatively little use in purchases by consumers from businesses, and most of these exchanges have been restricted to small enterprises. This situation has led some analysts to believe that future use of on-line currencies will be, at best, restricted to the person-to-person niche.
This article examines the likely success or failure of on-line currencies by means of a historical analogy. Specifically, the discussion compares the introduction of on-line currencies to the debut of the bearer banknote, the direct predecessor to modern currency, in late-seventeenth-century London. Despite the obvious differences between these on-line currencies and everyday, physical banknotes, the argument presented here will show that they share some conspicuous similarities in the circumstances of their birth. In particular, the article argues that the key innovation of the earliest banknotes was to provide finality under circumstances in which extant payment systems either could not ensure final payment or could do so only at an unacceptable cost. The next section describes how on-line currencies may be able to fill the same role in the context of e-commerce. The discussion concludes with some observations about future prospects for on-line currencies, again using the (clearly successful) introduction of the banknote as a historical model.
Early Forms of Inside-Money Payment
An initial summary of the prebanknote payment system in Europe, which combined deposit banking, orders to transfer deposits, and transfer of those orders by endorsement, is helpful in explaining the innovation offered by banknotes and the potential for on-line currencies. The system began with deposit banking in Italy, where two merchants desiring to transfer funds would together visit a banker and have one account debited and the other credited. Such transfers in banco spared merchants the transportation, protection, assay, and opportunity costs of using coin the outside money of the time. The banker’s ledger formed a permanent record, and payment within the bank was final.
To avoid the need for both parties to visit the bank together, deposit banking developed payment by check or draft. Checks drawn on banks in early modern Europe, including the goldsmith bankers of seventeenth-century London, fulfilled a role similar to that of personal checks drawn on modern deposit banks such checks enhanced decentralized exchange. Then, as now, the convenience of payment by check created a risk of default because payment was not final until the bank honored the check, and then, as now, whether the bank honored the check depended on the adequacy of the check drawer’s account balance or the willingness of the bank to allow an overdraft. This risk was manageable, but only because checks were generally used by prominent personages and for local payments only.
To arrange the payment of funds outside the local banking system, one had to arrange for payment by bill of exchange. Much like a modern traveler’s check, a bill ordered someone in a distant location to pay a fixed sum to a payee at that location. However, a bill was different from a modern traveler’s check in that it was payable only after some fixed amount of time had passed. Bills of exchange were generally payable in the prevalent currency of the distant location. For a bill to work, the person who wrote the bill (the drawer) had to arrange for someone to pay the bill at the other end (the acceptor). This arrangement was most easily made if the drawer had a close relationship with the acceptor. For example, Renaissance Italians established international family networks to act as acceptors. Later, bankers used systems of agents or correspondent banks. Once the bill had been accepted (always indicated in writing on the bill), it became a legally enforceable claim against the acceptor. Or the acceptor could refuse the bill by protesting it (and indicating so in writing on the bill) and returning it to the drawer.
The transfer of checks, drafts, and bills of exchange extended the opportunity to use inside money beyond the immediate range of a deposit bank. Remote transfer of third-party debt had a beneficial netting effect, reducing a chain of obligations to a single obligation between the original obligor and the ultimate creditor. The benefits of remote transfer were especially pronounced for places that outlawed deposit banking, such as London and Antwerp (van der Wee 1997). Instead of checks and ledger entries, inside money in these locales had to take the form of circulating personal obligations.11 A key advance in promoting extensive use of remote transfer was recognition of the legal standing of parties who had been assigned the debt of a third party in payment. The London Mayor’s Court granted such recognition in 1436, and the concept spread to Antwerp (Munro 2000).
Even with legal recognition, the effectiveness of remote transfers without banks was limited because information was needed to assess the credibility of the debt issuer (the acceptor of a bill), and such information was often asymmetric and idiosyncratic. Transfer created an incentive to pass on high-risk or fraudulent debt. In 1507, Antwerp mitigated this problem by creating a legal obligation of contingent liability on anyone who transferred third-party debt (van der Wee 1997, 325). According to the new rule, when a payor paid in the debt of a third party, the payor was also obligated to accept liability for the debt should the original obligor (or previous transferors of the debt) be unable to settle. Contingent liability gave anyone who wanted to circulate debt a strong incentive to screen the quality of the debt he was attempting to circulate. In practice, the simplest way of recording who had transferred a debt was to have each party sign the back of the debt.12 The institution of endorsement (transfer with contingent liability by means of a signature) spread across Europe and was applied to checks and bills of exchange. Combining legal standing with transfer by endorsement gave rise to the concept of a negotiable instrument, essentially a freely transferable debt whose possession automatically confers upon its holder well-understood rights as a creditor.13 Amsterdam became the dominant hub of international finance by buttressing a payment system based on the exchange of negotiable instruments with a municipal exchange bank (Dehing and ’t Hart 1997).
Another distinctive feature of negotiable instruments was the idea that anyone receiving an instrument by means of endorsement became a “holder in due course.”14 Essentially this concept meant that endorsees almost always enjoyed full creditor’s rights, even in cases when the good that was supposed to be delivered against the original obligation was not delivered or was defective (with some exceptions for sham transactions associated with fraud schemes). This feature enhanced the “moneyness” of negotiable debt by ensuring that good-faith transfers of such debt were final, barring default of the original obligor.
A Model of Debt Transfer
Kahn and Roberds (2001) analyze debt transfer and circulation by endorsement in a formal economic model in which payment by transfer of negotiable debt results in a desirable allocation of risks among payor, payee, and outside parties. They consider a stylized example in which party A supplies an intermediate good to merchant B, who uses the intermediate good to produce a durable final good, merchandise. Merchandise is delivered to customer C in return for a promise of future payment (see Figure 1). However, C may default on the promised repayment for one of several reasons (C may change his mind about the value he places on the merchandise or may be subject to an event such as fraud). Of course, knowledge of his own propensity to change his mind is C’s private information. Knowledge of the customer’s susceptibility to fraud risk is also private information, but the merchant may have some better knowledge of this information than the supplier does. All contracts between parties are subject to limited enforcement in the sense that assets held by a party defaulting on an obligation are not always attachable by creditors.
Optimal payment arrangements in this environment have two salient features. First, overly risky customers (those who have decided they do not want the merchandise or those too susceptible to credit events) should not receive merchandise. Second, in cases in which the merchandise is delivered, some portion of the promised payments by the customer should flow directly from the customer to the supplier, bypassing the merchant. In the latter case, an optimal allocation of risks can be implemented by a pair of debt contracts, one from the customer to the merchant and the other from the merchant to the supplier, as long as the merchant can discharge his debt by transferring the customer’s debt to the supplier (see Figure 1B). In other words, the merchant uses the customer’s debt to pay his own.
A potential problem with this type of arrangement is “adverse selection.” That is, in cases when the merchant deals directly with the customer and the supplier does not, the merchant is apt to have better information about the customer’s creditworthiness than is the supplier. The merchant may then have an incentive to pass on the debt of less credit worthy or nonexistent customers to the supplier. To guard against this temptation, the merchant must accept contingent liability for (endorse) the customer’s debt should the customer be unable or unwilling to pay. For this endorsement to be meaningful, the merchant himself must have sufficient wealth at stake. The intuition behind this result is straightforward. Payment by transfer of debt is desirable because it short-circuits the credit chain from customer to merchant to supplier, thereby limiting the possibilities for successive defaults. Transfer, however, creates an adverse selection problem, so adding endorsement gives the merchant an incentive to avoid transactions with overly risky customers. Enter Banknotes The combination of local deposit banking and circulating debt via endorsement created a successful system of inside money for the commercial elite but left out many people. Merchants, nobles, and others with sufficient standing could pay local obligations by means of checks drawn on a local bank, but these checks were useless for trading at a distance. Prominent firms could pay obligations incurred in long-distance trade by drawing bills payable on their overseas branches, but this option was out of the question for smaller firms. Likewise, large players could introduce others’ bills into circulation by endorsing them over to their creditors, but such players had to have sufficient wealth (and sufficient information regarding the creditworthiness of the acceptor) to have their endorsements valued.
0 comments:
Post a Comment