Arthur J. Rolnick, Bruce D. Smith, and Warren E. Weber
Federal Reserve Bank of Minnesota

I. Introduction

Monetary unions have been the subject of much debate in recent years. A monetary union is an area in which a single, uniform money is used or exchange rates are fixed between alternative monies. The formation of at least one monetary union is being discussed in Europe, and the dissolution of another is being considered in the former Soviet Union. Yet, despite the recent debates, the gains to be had from a monetary union are still at best vaguely articulated, and the problems of maintaining and enforcing a monetary union seem to be similarly vaguely stated.

While theoretical analyses and empirical explorations using recent data can shed some light on these issues, examination of historical experiences with different monetary systems is also illuminating. In particular, the United States monetary experience during its formative years demonstrates the gains from having a monetary union, illustrates the difficulties in maintaining a monetary union, and suggests why the United States tried to create one in 1787.

Although it may seem odd to think of the United States as having had to create a monetary union, during the United States' formative years each state (colony) had the power to—and many did—issue their own paper currencies. These monies circulated against each other at market determined rates, with the monies of some states holding their values well while the monies of others depreciated markedly. Moreover, this system had been in place in some form since 1690. Thus, the monetary experience of the United States during the colonial, Revolutionary, and Confederation periods seems appropriate for a study of the gains from monetary unions and of the methods that can be used to achieve them.

Our historical analysis is largely based on two theoretical insights. The first is that a flexible exchange rate regime can permit unnecessary and socially wasteful exchange rate variability to occur. The costs of this risk (which may in turn influence other costs of having multiple currencies, like conversion costs) make a monetary union desirable. The second is that if a fixed exchange rate system is achieved while individual countries retain the power of money issue, any member of such a system has the power to levy an inflation tax on its partners in the union. This creates what we term a seigniorage incentive problem. If this problem is sufficiently severe, it can cause a monetary union to break down.

We demonstrate that prior to the Constitutional prohibition of state currency issues, the problem of exchange rate variability was a serious one in the United States. We also demonstrate the existence and importance of the seigniorage incentive problem, and why its resolution probably necessitated eliminating the power of individual states to issue money, as was done in the Constitution. Finally, we review some explanations for why the United States adopted the monetary system that it did and argue that resolving the seigniorage incentive problem must be an important component of any such explanation.

II. A Perspective on Monetary Unions

Some recent theoretical work suggests, and empirical evidence from the contemporary exchange rate regime seems to confirm, that regimes in which exchange rates are not fixed permit exchange rates to fluctuate randomly and unnecessarily. Such fluctuations in exchange rates impose avoidable social costs which render fixed exchange rates or a uniform currency desirable. Monetary unions avoid such costs.

However, monetary unions are difficult to maintain because of what we term a seigniorage incentive problem. If countries in a monetary union have the individual ability to issue money, then they also have the ability to use money creation to collect seigniorage from residents of other countries. In an equilibrium with unchanging exchange rates between monies, countries with money stocks growing faster than the average will collect seigniorage from residents of countries with money stocks growing less rapidly than the average. If this ability is exercised, then those countries bearing the tax may choose to retaliate in any one of several ways. They could increase their own rate of money creation as a way of “collecting back” seigniorage income. The result might be high inflation, diluting the benefits of a monetary union. Or, alternatively, controls on the use of “foreign” currency might be imposed to limit its use, and thereby to limit how much seigniorage revenue other countries can raise at the expense of the domestic country. Such controls, however, work against a monetary union; they reduce the substitutability of currencies, but substitutability is the essence of a monetary union. Thus, a fixed exchange rate regime may be difficult to maintain unless institutional arrangements are made to mitigate the seigniorage incentive problem.

III. Exchange Rate Variability in the United States During its Formative Years

In this section we set the historical stage for our discussion of the cost of exchange rate variability and the seigniorage incentive problems in the United States during the colonial, Revolution, and Confederation periods. To do so, we present an overview of the monetary arrangements in these periods and describe the fluctuations in exchange rates among that various monies that circulated.

IV. The Cost of Exchange Rate Variability

During its formative years, the United States experienced a proliferation of currencies accompanied by apparently unnecessary exchange rate variability. We now present some examples of how such variability increased the cost of trade.

A. Example 1: Virginia 175564

Colonial monetary affairs were subject to British oversight. Over time the monetary relations between Britain and the colonies became an increasing source of frictions. Such frictions came to a head between Britain and Virginia between 1755 and 1764.

Virginia was the last colony to issue paper money. When it first issued paper money in 1755, the colony was desperately short of specie. (See Ernst, 1973 and Brock, 1975) Colonists who borrowed from English merchants, which was a widespread practice, had incurred sterling denominated debts. These debts were routinely (and of necessity) repaid in local currency, which was in fact a legal tender. But, of course, the rate of exchange between Virginia currency and sterling was subject to some fluctuations.

Given the legal tender status of Virginia's currency, British creditors could not avoid repayment in this form. However, British creditors objected strenuously to being subjected to exchange rate risk. In 1758 British merchants petitioned the crown demanding “absolute protection against any fluctuations in the rate of exchange. Such risks were to be born by the Virginians alone” (Ernst 1973, p. 52). Thus the allocation of exchange rate risk became a subject of heated political discussion.

In response to British pressure, Virginia law was amended in 1755 “to allow courts of record to settle all executions for sterling debts in local currency—paper as well as coin—at a `just' rate of exchange. A just rate was taken to be the actual rate at the time of court judgment” (Ernst 1973, p. 54). However, even this was viewed as inadequate protection against exchange rate variation by British creditors. British merchants wanted the option of consenting “to accept paper money in amounts they deemed necessary for the purchase of sterling bills of exchange to the original and full value of sterling debts” (Ernst 1973, p. 52).

This was unacceptable to Virginia. In addition to forcing Virginians to bear all exchange rate risk in exchange with Britain, it would give British merchants bargaining power over Virginians who had only local currency as a means of payment. Thus an impasse was reached, which was resolved in favor of the British merchants by the Currency Act of 1764. This Act prohibited the colonies outside New England from making their currencies a legal tender for public or private debts.

B. Example 2: Confederation

During the Revolution, and continuing into the period of Confederation, interstate commerce was of growing importance. The use of the state currencies in interstate transactions (as well as at home) was plagued by exchange rate uncertainty.

Consider the problems with Pennsylvania's currency, a currency which maintained its value far better than that of some other states. According to Bezanson (1951, p. 326), “in the spring of 1789 James Cox explained `the very fluctuating state that our paper money has always been in, makes it difficult to ascertain the value of it at different periods.” An illustration of the perceived costs of this exchange rate variability is the fact that the Pennsylvania assembly refused to be paid in Pennsylvania currency, which was a legal tender for public, but not private, debts. (Kaminski 1972, p. 70.)

An even more dramatic illustration is offered by the attitude of the Bank of North America toward Pennsylvania currency. The Bank of North America kept accounts in Pennsylvania currency completely distinct from specie accounts, even though the state's currency did not initially depreciate. The Bank actually did receive a substantial quantity of state paper money, and the keeping of separate accounts led to a “considerable extra expense to the Bank.” (Kaminski 1972, p. 67.) Apparently this was a cost the Bank was willing to absorb in order to avoid exchange rate risk.

New Jersey and South Carolina faced similar problems with their currencies.

V. The Seigniorage Incentive Problem and the Maintenance of a Monetary Union

Suppose that some form of uniform currency is achieved in which the individual states retain the power of currency issue. The result is a seigniorage incentive problem: governments can impose inflationary taxation on citizens outside their jurisdiction, and thereby redistribute revenue to their own citizens. With this power being exercised, neighboring states (colonies) have an incentive to retaliate, however. One possible form of retaliation is to impose legal restrictions limiting the use of the offending state's currency. But such restrictions undermine the uniformity of the currency.

This situation actually arose in New England during the colonial period. We now discuss this episode. We then turn to a discussion of how a variation of the seigniorage incentive problem emerged under Confederation.

A. Colonial New England

By 1710 all the New England colonies had issued their own currencies. The respective currency issues easily crossed colonial borders. Interestingly, even though there was no government attempt to enforce or sanction the practice, “the bills of the several New England colonies customarily, although not always, passed current in all the rest at a uniform value” (Brock 1975, p. 35). In other words, the exchange rate among the currencies of the New England colonies were constant at a rate of onetoone. This situation lasted until 1751 when this system broke down.

The potential for one colony to levy an inflation tax on its neighbors did not go unexploited. Here Rhode Island was the culprit: “the fact that Rhode Island bills circulated widely in other colonies permitted her to levy tribute on her neighbors” (Brock 1975, p. 39). Between 1710 and 1744 the New England money supply grew at any average rate of almost 8 percent per year; over the same period the supply of Rhode Island bills of credit grew at an average rate of almost 14.5 percent per year. Most of this increase went into circulation in other colonies: “it was estimated that as many as fivesixths of the Rhode Island bills were absorbed by Massachusetts” (Brock 1975, p. 41). By 1744, 43 percent of the New England money supply had been issued by Rhode Island, which had only about 10 percent of New England's population.

Given the high rate of inflationary taxation levied by Rhode Island on its neighbors, it is not surprising that constant exchange rates did not persist (although it may be a testament to the costs of exchange rate uncertainty that they lasted as long as they did). In 1749 Massachusetts passed a law prohibiting the circulation of other New England currencies within its border, with a fine of 50 pounds for a violation.

B. Confederation

The seigniorage incentive problem also recurred during the Confederation period, although its form was somewhat different than during the colonial period. The seigniorage incentive problem is not limited to regimes in which exchange rates are constant or fixed. Countries with money stocks growing faster than the average can collect seigniorage from residents of countries with money stocks growing less rapidly than average under any exchange rate regime. The only way a country that wants slow money growth can prevent this from happening is to impose currency controls to require the use of the “domestic money” or to prohibit the use of the “foreign money” within its borders. Such restrictions were imposed during the Confederation period.

One place that the seigniorage incentive problem recurred was in New England, and as during the colonial period, a primary culprit in this respect was Rhode Island. It was the only one of the New England states to create a paper currency, and its currency depreciated rapidly. In opposing Rhode Island's paper money emission, some of the merchants of Newport and Providence argued that “a paper money law ... would ruin Rhode Island's commerce with other states, which would not accept payment for their goods in rag money.” (Nevins 1927, p. 228.) Not only did Rhode Island nonetheless emit a paper currency; it made it impossible for creditors to its citizens to insist on payment in any other form. The result was a legal retaliation by Massachusetts and Connecticut, who “passed laws enabling their citizens to pay all debts owed to people of a papertender state in just the same manner as the latter paid their debts to the citizens of Massachusetts and Connecticut. That is, Rhode Island creditors were virtually outlawed in the neighboring states ....” (Nevins 1927, p. 571.) Thus credit transactions between citizens of Rhode Island and citizens of Connecticut or Massachusetts were impaired due to Rhode Island's attempt to force the use of its currency in debt payments.

A similar situation arose in North Carolina. There some merchants who were withinstate creditors might have been expected to oppose the creation of a depreciating currency which was legal tender for private debts. However, a “factor that qualified the [negative] attitude of some merchants [to the state's currency] was that while they were creditors to many persons, they themselves were debtors to other merchants of the state and, more often, to mercantile interests outside North Carolina. A number of North Carolina merchants, in fact, came to dare their outofstate creditors to sue for recovery in the postwar fiat currency.” (Morrill 1969, pp. 6465.) And indeed, so eager was North Carolina to force circulation of its currency by making it a legal tender for private debts “that judges would not allow the nominal value of the currency to be altered even with the consent of the debtor and creditor involved in the case ....” (Morrill 1969, p. 86.) Similarly, “with Virginia merchants particularly in mind, legislators at the 1786 Assembly introduced a bill that would have made it a misdemeanor to demand specie payment for merchandise, to refuse to accept paper money in payment, or to accept paper money at less than nominal value.” (Morrill 1969, p. 89.)

VI. “In Order to Form a More Perfect Union...”

The formative years of the United States help illustrate why a monetary union is desirable, yet difficult to maintain. Exchange rate variability was significant and costly during these years. But the experience with different colonies and states trying to export inflation and tax their neighbors suggests why a monetary union might not be easy to maintain.

Our position is that the newly formed country's desire to create a lasting monetary union was at least partly responsible for the constitutional prohibition on the power of states to issue currency (bills of credit). That is, we think that the country's desire to eliminate exchange rate variability was at least partly responsible for Article I, Section 10 of the U.S. Constitution which states that

No State shall ... coin money; emit bills of credit; make anything but gold and silver coin a tender in payment of debts; pass any bill of attainder, ex post facto law, or law impairing the obligation of contracts....

We are not the first to try to explain the willingness of states to give up the power to issue money. Nevertheless, we feel other explanations are either unconvincing or cannot stand on their own. In this section we briefly review and critique other explanations that have been proposed for the Constitutional prohibition of state currency issues.

VII. Conclusion

In conclusion, we view there as being obvious gains from having a monetary union. However, a monetary union in which the currency can be issued at the discretion of the individual members gives rise to an incentive problem: members of the union can gain seigniorage income at the expense of other members. The only way to address this problem is either to find an acceptable way to share seigniorage among the members or to prevent members from issuing currency. The United States chose to take this latter approach when in 1789 it adopted a constitution that prohibited states from issuing their own currency.

The United States' attempt at creating a monetary union raises an obvious question that we do not address in this paper: Did the actions taken by the United States in 1789 result in a monetary union in the sense that exchange rates were constant among the various monies circulating in the country? In the sequel to this paper we argue that the answer is no. Seigniorage incentive problems plus a loophole in the system regarding bank creation and regulation prevented the United States from achieving a monetary union at least until the late 1800s. In the sequel, we also identify the mechanisms by which monetary union was finally achieved.