Make your own free website on Tripod.com
Strategies for  Currency Unification
The Economics of Currency Competition
and the Case for a European Parallel Currency
Dr. Roland Vaubel
wrote the 13 page summary of his 400 plus page academic beaut
which was scanned, corrected
and is now within a few pagedown keytaps reach below, but first:
Those of you unable to obtain this ponderous tome but interested enough in the subject matter may just contact the IEA in London
and order the 32 page pamphlet (transcript of a '79 lecture)
of which you see the cover page and table of content next:
Choice in European Monetary Union
Ninth Wincott Memorial Lecture
delivered at the Institute of Bankers, London,
on Wednesday, 6 December, 1978
ROLAND VAUBEL
University of Kiel
The Trustees of the IEA have agreed that any surplus over costs arising from the safe of this paper should be donated to the Wincott Foundation.
INTRODUCTION
GLOSSARY OF TECHNICAL TERMS
THE AUTHOR
Published by
THE INSTITUTE OF ECONOMIC AFFAIRS for THE WINCOTT EOUNDATION
1979
Contents
Intro by Lord Robbins
I THE NEW EUROPEAN MONETARY SYSTEM
II AN EVALUATION OP THE SYSTEM
(i) Pooling of monetary reserves
(ii) Extension of international  credit facilities
(iii) Foreign exchange interventions
(iv) Return to an adjustable-peg exchange system
(v) Difficulties of operating a 'snake'-type system in the EEC

III EUROPEAN MONETARY UNION A WORTHWHILE GOAL ?
Benefits of monetary union and costs flexible exchange rates v. monetary union
Dangers of European money cartel/monopoly

IV ALTERNATIVE STRATEGIES
(i) Monetary-harmonisation approach
(ii) Combined monetary` end exchange-rate agreements
(iii) Tke common weakness of all co-ordination strategies
(iv) The common-currency strategy

V TOWARDS A EUROPEAN COMMON CURRENCY
The 'big-leap' approach
Advantages of gradual currency substitution
Currency substitution: market or governmental process

VI SUMMARY: EIGHT ADVANTAGES OF THE PARALLEL
CURRENCY APPROACH
***************************************************************************
**************************************************
This file was created near end of nov 98; last minor changes in dec 98
Tripod tool Nedstat installed near the end of nov 98 but not correctly till july 99
Nedstat Counter
   poetpiet@hotbot.com
Go see what else poetpiet can puzzle people with here
....................................or here..................................
or check the intro to this and other files in this first batch of guest appearances
concerning all sorts of currency issues
*********************************
Back to the big book; it was published as:
Kieler Studien 156
Institut fuer Weltwirtschaft an der Universitat Kiel
Herausgegeben von Herbert Giersch
J. C. B. MOHR (PAUL SIEBECK)-TUEBINGEN
ISSN Q340-6989

                (some comments by yours faithful added in fiery red
and all emboldenments are your digitalizer's also)
402    Summary
This is written for the busy reader. It is notably written for the practical political economist, for it concentrates on the policy implications rather than on the theoretical innovations of this study. More specifically still, it is written for those practical economists who are prepared to look beyond their day-to-day business and to think in terms of years and decades rather than weeks or months. Finally to add a further limitation, it is written for those interested practical economists whose approach to policy consulting is not merely to forecast the probability of each possible policy action being taken and then to come out in favour of the most probable one, but to recommand what they have realized to be desirable and necessary and to try in a personal effort to change the probabilities of the possible policy actions in favour of the optimal solution.

       A policy action cannot be a solution, let alone the optimal one, unless it is possible. however, what is possible is difficult to judge. To many present observers, the aim of monetary union in Europe may seem as unattainable as the creation of the Common Market must have appeared to those who thought about it in the 1930s and '40s. However economic circumstances and political constellations change, and with them the (defunct?) economists who somehow manage to influence gover nulent policy.
         With respect to European monetary unification, we observe not only polltical charges but also an important economie development the market reduction in the effectiveness of monetary expansion, of inflation and of the associated external currency depreciation in alleviating unemployment. High constant rates of inflation no longer have a recognizable effect on employment, and high variable rates of inflation, by increasing uncertainty and inefficiency, seem to reduce employment: the temporary increase of employment in the phase of accelerating inflation is smaller than the temporary decrease of employment in the phase of decelerating inflation. The reason for this change ln the economy's response to monetary-policy impulses is the adaptatlon of expectations to policy instability, the so-called erosion of money-(including exchange-rate) illusion. The implication of this catching-up process is a need for preannounced (how about project specific? Sound good to you too?) rather than impulsive monetary expansion, for price-level stability rather than inflation-rate variability and for a reduction or elimination, rather than an increase of exchange-rate charges. Thus, the chances for monetary union are likely to become better rather than worse...............
The theoretically-minded reader is referred to the survey which is given in the introductory outline and to the partial summaries on pp. 98-99, 153-154, 289-290, 322-323 and 346.
403   Starting from this basis, this study has analyzed and compared various strategies for monetary unification:
on the one hand, the coordination approach with its variants:
(i) exchange-rate unification,
(ii) monetary-policy harmonization and
(iii) a combination of the two,
on the other hand, currency unification with its variants:
(i) "big-leap strategy",
(ii) free currency competition and
(iii) parallel-currency approach.
          Exchange-rate unification is defined here as the fixing of exchange rates without ex ante harmonization of monetary policies, let alone cantralization of monetary policy in the hands of a European central bank. Exchange-rate unification is rejected because it involves foreign-exchange interventions and requires a hegemonial currency. Foreign-exchange interventions are considered undesirable on general welfare-economic grounds becall se they involve avoidable internationa externalities, i.e. interference with the monetary policy of foreigr countries. Moreover, the present system of intervention under which weak-currency countries obtain external seigniorage through subsidized credits or a reserve-currency role, is criticized for its moral hazard: its incentive to inflate and, ultimately, its self-defeating nature. If foreign-exchange interventions are ruled out, exchange-rate fixing can only be maintained if domestic monetary policies are continually adjusted to attain external equilibrium. But who is to adjust to whom?. If at all it will be the smallest countries which adjust to the larges countries (country) because for the smaller countries, being more open, the benefit of exchange-rate constancy has a-larger weight as compared with domestic objectives (price-level stability) than for the larger countries. However, the satellites will only adjust as long as the hegemonial-currency country offers an acceptable performance Since the benefits of exchange-rate constancy are not fully internalized by the hegemonial-currency country, it does not have a sufficient incentive to conduct a monetary policy acceptable to the satellites. Thus central bank charters with an informal price leader tend to be unstable and short-lived. Moreover, it has been shown that, also on economic grounds, the deutsche mark is a particularly inappropriate pivot currency for the European Community.
    Monetary-policy harmonization without fixed exchange rates can only lead to monetary union if rates of domestic monetary expansion can be found and agreed upon ex ante which make free exchange rates con 404  stant exchange rates. Quite apart from the very high political friction and the past failure ofthis strategy (1973), it has been shown that the income elasticities of the demand for moneyend the real exchange rate charges within the Europe an Community do not seem to be sufficiently stable to permit the attainment of exchange-rate constancy through mere ex ante harmonization of domestic rates of monetary expansion.
  Exchange-rate agreements and monetary-policy agreements may be combined in a simultaneous ex ante commitment. The easiest way would be to agree on the rate of money-supply growth for one member country and to oblige all other member countries to maintain fixed exchange rates vis-à-vis the currency of that country by adjusting their domestic monetary policies. The obvious (political) disadvantage of this strategy is that it still appears to involve a currency hegemony. The other variant of the simultaneous commitments approach is to agree in advance on each member country' s rate of domestic monetary expansion, to choose these rates in such a way that they are expected to beconsistent with exchange-rate constancy and to correct for any errors that may subsequently becorne apparent by rnaintaining the agreed exchange rates through symmetrical foreign-exchange interventions. This solution has the advantage of not involvingexternalities or morel hazard and inflationary bies (because monetary end exchange-ratepolicy would be entirely predetermined by consent). But although it would be a considerable improvement on present arrangements (the "snake"), it still suffers from many drawbacks of both approaches which it combines. Since it maximizes political friction, the required agreements are unlikely to be concluded, and if they are concluded, they probably will not be kept (or only at the cost of restrictions of capital movements end even of trade).To suggest that exhange rates between the member currencies could remain permanenty fixed as a result of regular negotiations and complete coordination mong nine governments or central banks is to presuppose a "brave new world of which we have no experience and which seems inconsisent with our observations of politica! end bureaucratie behaviour.
   It is the common and crucial defect of all variants of the coordination trategy that they rely on discretion instead of automaticity. This lack of automaticity manifeste itself in four ways: There is no way of ensuring automatie ex ante consistency of price targets (exchangerates) and quantity targets (money stocks).
   Since the participants retain their power of discretion and the instrurnents to exercise it,there is no automatic mechanism which prevents them from violating the coordination agreement.   405
~ 1  A fortiori, there is no way of ensuring that the coordination agreement(s) will automatically be renewed.
   A fortiori, there is no automatic process by which discretion (the possibility of opting out) is reduced over time.
    Since the coordination approach lacks automaticity, it fails to make exchange ratespredictable. If it involves the fixing of parities, it may indeed lead to more errors and uncertainty than exchange-rate flexibility and the pre announcement of definite rates of national money supply growth which exchange-rate flexibility makes possible.
     The failure to give guidance to expectations tends to perpetuate itself in a vicious circle. Any strategy which makes it easy for member governments to opt out tends to be self-defeating because it leads trede unions end entrepreneurs to assume that the national authorities will yield to their pressure. Wage and price setting then becomes inconsistent with the government's international commitments, national monetary policy accommodates to (try to) avoid a rise in unemployment, and the exchange-rate target has to be abandoned.This, in turn, strengthens the public' s expectations of government compliance even further,etc. Since, in terms of risk and transaction costs, the maintenance of national currencies has many obvious economic disadvantages as compared with currency unification, the public's expectations can even be regarded as entirely rational; for it is difficult to see why exchange-rate unification could ever be preferred to the adoption of a unique common currency ("currency unification") by the member governments, if the latter were unconditionally committed to exchange-rate fixity. They do not dare to undertake unconditional commitrnents because they have experienced a "need" for accommodation in the past, not realizing that this "need" was the result of their failure to cornmit themselves unconditionally in the first place. The circle cannot be broken unless the authorities renounce their discretion. Currency unification, being as irreversible and credible as any unification can be, tends to be self-fulfilling: economie agents know what to expect and what to adjust to, and they will find it in their interest to avoid a collision.
     The problem of expectation adjustment is also the key to another defect of the coordination approach. The coordination approach requires an assimilation of inflation rates. Since the member countries with the lower rates of inflation are unlikely to accept an acceleration of inflation for the sake of exchange-rate unification, the more inflationprone members will have to (continue to) follow strongly disinflationary policies. The downward harrmonization of monetary rates of expansion and inflation is liable to produce temporary reductions in the 406 level of employment because inflation expectations are slow to adapt. No such cost of transition arises if the inflating national currencies - instead of being stabilized - are replaced by a new currency which especially if it is subject to value guarantees and issued by a new and independent institution, does not suffer from a record and expectations of inflation. This is why many times in history governments have preferred currency reform to currency stabilization. If the new currency is the same for all member countries, currency reform coincides with currency unification: instead of nine currency reforms there would be only one.
    The most obvious approach to currency unification is the "big-leap strategy, " i. e., the uno actu displacement of the national member currencies by a unique Community currency. From a political point of view, this strategy has the disadvantage that the currency unification process would start later, if at all, and involve more political friction, than a gradual approach. From an economie point of view a gradual transition to currency union would facilitate the adjustment of expectations, contracts and accounting and, hence, minimize mis-allocation of resources and government interference with contract denominations. However, since there are also costs to prolonged currency coexistence, (and methinks that's not the only nor best reason why) it is important that the speed and pattern of currency substitution should be left to the market.
     Free currency competition among the national member currencies is likely to lead to currency union and to permit both automaticity and gradualism in the process, if the use and production of money is subject to gradually but permanently increasing returns to scale (money production as a natural monopoly). Moreover, during the transition to currency union, free currency competition would serve to fight inflation and tering about dynamic optimum currency domains in the European Community. These three main effects of free currency competition have been analyzed in great detail, including the possibility of a free and competitive private supply of (base) money and the theoretical objections that have been advanced against it. It has notably been argued that, in the presence of a government-issued legal (but not forced) tender, competition between private suppliers of base money will result in value-maintenance provisions guaranteeing a constant or even increasing purchasing power for the private "parallel" currencies. The objection that the use of money produces Pareto-relevant real-income externalities or that money is even of a public-good character has been refuted. Gresham's Law has been shown to be inapplicable in the absence of a fixed forced (or legal) disequilibrium exchange rate between the concurrent monies. If free currency competition and, hence, exchange-rate flexibility is permitted, "good rmoney chases bad money" (the Anti-Gresham Law) instead of the reverse.
407   The only valid objections to currency competition between the nine member currencies seem to be that it would involve the stabilization of at least the surviving national member currency (and, therefore unnecessary stabilization unemployment) and that the less competitive issuing countries, threatened by a loss of seigniorage and prestige to the prospective winner, would not admit the competition. By contrast national member central banks might have an economic incentive to accept competition from, and displacement by, a Community currency if they shared in the seigniorage accruing from its issuance and if it could be expected to make inroads into the currency domains of non-member currencies (notably the dollar). Political as well as economic considerations thus indicate that competition and substitution should take place not between the national currencies, but between each national currency and a new Community currency.
         This is the case for the parallel-currellcy approach to European currency unification. It implies that the European parallel currency (EPC) should be so stable that it need not be stabilized. This rules out all proposals which view the EPC as a pivot or average (basket) of the inflating member currencies or which provide for a link between the EPC and the ab initio "strongest" member currency. Moreover, it is very doubtful whether a less than stable EPC would be sufficiently attractive to be able to overcome the barriers to entry, which an infant currency has to face, and to outcompete the established national member currencies. The examination of a large number of potential valuation formules and the detailed discussion of the nature of domestic inflation risk, real exchange-rate risk and nominal exchange-rate risk and of ways to obtain protection against them (through indexation, currency diversification and hedging/forward cover, respectively) have yielded the result that the EPC would most appropriately be defined as a basket of weighted amounts of the national member currencies that would each be increased whenever, and by the same percentage by which, the respective member currency lost purchasing power as measured by the national consumer price index. While the indexed basket formule minimizes inflation and real exchange-rate risk, it does not minimize nominal exchange-rate risk (a bilateral or multi currency fixed-amount basket would) or exchange-rate information cost (par-valuation would) or the deviation from the money-holding optimum (a basket whose purchasing power increased at the real rate of interest would). However, since exchange-rate information cost is relatively srnall, since norminal exchange-rate risk is relevant om: in the short run in which old national-currency contracts have not yet matured, and since seigniorage can be eliminated through interes payments on all types of money except currency in circulation, an EPC 408 defined as an indexed basket of the member currencies seems to offer a maximum of advantages.
       Since the EPC is to be subject to a value guarantee that would be validated instantaneously and exactly through exchange-rate adjustments vis-à-vis the basket of national member currencies, the EPC Bank would have no discretion as to the quantity of EPC, Supply would have to adjust to demand so that the automatically changing guaranteed exchange rate vis - à -vis the member currencies would be the equilibrium exchange rate. This means that, for the purpose of "controlling" the EPC supply, there is no need to impose reserve requirements for EPC deposits with central banks: if, for example, the voluntary reserve ratio fell and the effective deposit multiplier increased, the EPC Bank would ceteris paribus be forced to validate its value guarantee by withdrawing EPC base money so as to keep the EPC money supply constant. It also means that no ceiling could be put on EPC expansion if the EPC were to be both standerd and store of value.
          Interference with the monetary policies of the me mber states is minimized if the EPC is not issued to finance Community expenditure and loans or to purchase financial assets but issued in exchange for the member currencies at the guaranteed rate. In this way, any shift of demand from the national member currencies to the EPC could be matched by an equivalent substitution of EPC supply for member-currency supply without effects on total money supply in the European Community and on the national inflation rates and exchange rates. lf the national currencies were not withdrawn pari passu with the issue of EPC, inflation would reduce the value of the increased nominal money balances until the equilibrium value of real money balances were restored. Since an EPC of constant purchasing power cannot, by definition, be subject to inflation, the adjustment would ceteris paribus have to be brought about through inflation and depreciation of the national member currencies. This means also that the national central banks should not, in principle, reissue the amounts of national currency which the public sells to the EPC Bank and which the EPC Bank withdraws from circulation. The target which each national central bank may - hopefully - have and announce with regard to its national monetary base should not be confined to national currency in circulation and national-currency reserves of domestic banks but it should also include the central bank' s liabilities to the EPC Bank. By doing so, each national central bank would also allow tor the fact that its target rate of increase of high-powered national money would partly be supplied already as a result of the nominal increase of its national currency liabilities to the EPC Bank which the EPC purchasing-power guarantee and the payment of interest on the EPC reserves of com 409 mercial banks imply. For if the balance sheet of the EPC Bank is to remain in balance, its claims un the national cantral banks (the balances of national member currencies) must be indexed and beer interest just as the outstanding volume of EPC.
     The national central banks would thus lose seigniorage on the monetary base underlying that part of the national money supply which has been rejected by the public in favour of EPC. Their loss would be the same if the EPC Bank invested its national currency proceeds in the private capital and Euro-currency markets, for if the national cantral banks still wanted to attain their price-level targets, they would have to withdraw an equivalent amount of their national monetary base from circulation. This would notably hold if the EPC were also issued in exchange for non-member currencies (say, the dollar) and if the Federal Reserve System did not agree to cede the corresponding part of its seigniorage to the EPC flank (and, ultimately, to EPC holders).
          The principle that each addition to the EPC supply should be matched by an equivalent reduction of the national currency supply on a one to-one basis has to be modified only to the extent that the introduction of a stable and interest-bearing European currency raises the demand for real balances or to the extent that the effective multiplier (reserve ratio) for high-powered EPC differs from the effective multiplier (reserve ratio) for the high-powered national currency supply which it replaces. If the EPC were only issued in exchange for national membe r currencies, the demand for the mermber currencies may also increase because original owners of non-Community currencies wish to temporarily hold member currencies for conversion into EPC. Most conveniently, the national high-powered money-supply targets might be adjusted ex ante to allow for these effects.
    It has been argued that in order to minimize short-term intra-Community cross-rate effects and conversion costs, the EPC should be issued to EC residents in exchange for the current basket equivalent of their national currency, while with regard to non-EC residents strict basket conversion is preferable both from a cross-rate and an external seigniorage point of view.
      Proposals to confer the functions of money only gradually upon the EPC have been rejected becall se only a fully-fledged new currency is likely to be able to overcome the barriers to the entry of an infant currency. Proposals to restrict the use of the EPC for specific purposes have been rejected for the same reason and becall se such restrictions are inconsistent with the exploitation of the comparative advantages of the different monies.
410     The analysis of the prospective speed of EPC penetration has shown that the process is likely to be gradual and that, for various reasons, gradual acceleration will be followed by gradual deceleration. With regard to the pattern of EPC expansion, it has been shown that as a standard of value it will initially be in greatest demand for large longterm contracts between non-financial institutions in countries where inflation risk and real exchange-rate risk are high and money illusion small.
      As a store of value the EPC is most attractive for internationally oriented companies and large holders of notes and coins in high-inflation cantral member countries. As a means of payment the EPC is least unlikely to be used in open central member countries between residents who happen to hold store-of-value EPC. The partial criteria which are identified offset each other to some extent. Whether the market's choice of currency will lead to EPC expansion mainly in the central or the peripheral countries depends on whether, in the present inflation-ridden European Community, the theory of optimum currency areas or the theory of optimum stabilization domains has more weight.
      In theory, the national central banks could be left free to issue national money ad infinitum. However, after some time, such sovereignty could become purely formal; for if all newly-issued national money were immediately exchanged for EPC at the EPC Bank, the national central banks would no longer earn seigniorage or conduct a monetary policy. It is advisable to stop new issues of national money and, perhaps to transfer the status of legal tender from the national currencies to the EPC, once (and as long as) the EPC accounts for an overwhelming proportion (say 60 percent) of total real balances from Community sources. The precise percantage should be agreed and announced in advance. The EPC Bank would then become a genuine cantral bank and take over control over the total money supply in the European Community, hopefully by announcing a. rate of EPC money supply growth that is consistent with continued purchasing power stability. This "limping union-currency standard" would be succeeded by full currency union when the last holder of national member currency chose to convert it for what would now become the unique Community currency. Perfect currency competition would no longer be simulated by a duopolist but by a monopolist.

To summarize, the parallel-currency approach possesses at least nine characteristics of an optimal currency unification process:
     1. It works without foreign-exchange interventions and, hence, with out international externalities and morel hazard or ex ante agree 411 ments on national rates of money-supply growth.
    2. It triggers an àutomatic mechanism which works without politica! discretion and which leads to currency union in a predictable and self-fulfilling process.
     3. payment at a very early stage, thus facilitating market integration while still leaving control over national monetary policies with the member governments.
     4. It has all the political and economie advantages of gradualism.
     5. It permits the speed and pattern of currency unification to be determined by the nceds of the market and the degree of money disillusionment.
    6. It avoids the temporary unemployment that would be created by the downward harmonization of inflation rates of the national member currencies.
    7. It does not encoura~e competitive inflation, but through competition tands to discourage it.
    8. It avoids nationalist rivalries and hegemony.
    9 It gives the monetary all thorities in the European Community a positive incantive to desire or accept EC currency unification.
     By contrast, the traditional coordination strategy does not qualify on any of these counts except one (4. ).
The standard objections against the parallel-currency approach are that it leads to
(i) inflation,
(ii) exchange-rate instability and
(iii) monetary inefficiency, or
(iv) nowhere.
It has been shown in this study that inflation will not result if the issue of EPC is offset by withdrawals of national member currency in an automatie conversion mechanism. Exchange-rate stability should be attained not by restricting the choice of currency, but by requiring the adjustment of supply to demand as in other lines of production. Consumer sovereignty in the field of money will not be inefficient, for the production and use of money does not give rise to Pareto-relevant real-income externalities. After all, everybody will be free to reject the EPC if he prefers the national currencies. The parallel-currency approach would lead nowhere if it were an attempt to coax European politicians into an action which they did not really wish to undertake. The trick would be discovered and, ultimately, rejected by them. However, the parallel-currency approach is not a ruse designed to dodge the political process.
412    It minimizes political friction, but it is not conceived as a substitute for political agreement on the desirability of currency union. There can be no such substitute. The reason why it has not yet been tried by the European Community (or a group of its members) is not that it is not the optimum currency unification process, but precisely that it would lead to currency union and that several, if not most, member governments do not seem to desire currency union. The parallel-currency approach to European monetary unification cannot become operative before the destruction of money illusion in the markets has been followed by a destruction of Phillips-Curve illusion on the part of (a sufficient number of) member governments.
               Unlike the first part of this study the second is empirical in nature. Its first three chapters contain a quantitative analysis of several issues which have been addressed in the first part. At the and of each chapter the main findings are summarized so that there is no need to repeat them here (see pp. 229-230, 322-323 and 346). For this reason, the following paragraphs will merely focus on those results which are of direct relevance to the arguments presented in this summary.
    As has been explained, the traditional coordination approach to European monetary unification is inferior to the parallel-currency approach. The European Community's failure to make progress towards monetary union may, therefore, be due to the choice of a suboptimal strategy. A competing explanation is that the European Community is an undesirable currency area or that it has become one after 1971.
    The analysis of internal real exchange-rate charges in Chapter I has shown that the European Community is probably indeed a less desirable currency area than, for example, West Germany, Italy and the United States but that only a small part of the failure can be attributed to structural economic causes: two thirds of the nominal exchange rate charges are due to the member governments' unwillingness and/or inability to coordinate.
     The official excuse that the 1971 strategy for EC monetary union, while being justified and realistic at the time it was adopted, was largely thwarted by unforseeable external disturbances (such as the dollar "crises," the increase in the price of oil and other raw materials, and the recant world depression) seems to be disconfirmed by the evidence. If an attempt is made to allow for the increase in roal exchange-rate variability which the transition to floating is lilmly to have brought about in the short and medium term, the need for real exchange-rate 413 charges within the European Community seems to have decreased instead of increased since 1971. What has increased is not the structural differences and divergencies among the member countries but the unwillingness and/or inability of their governrments to coordinate monetary policies.
     These results imply that the economic case for European currency unions is at least as strong now as it was in 1971. Moreover, the real exchange-rate criterion has been shown to permit the identification of a pioneer group of rmertiber countries that would be most suited from an economie point of view for a first start with currency unification (through the parallel-currency approach). The group consists of the present five EC members of the "snake" plus France.
    The analysis of the foreign-currency preferences of depositors, official reserve holders and bond issuers which has been presented in Chapter II derives only a small part of its importance from the implications which it has for the parallel-currency approach. Nevertheless, these implications are worth ermphasizing.
     First of all , there is abundant evidence that non-bank depositors and bond issuers (and even central banks) are averse to exchange risk. While depositors and central banks reduce their holdings of foreign currencies when unexpected exchange-rate charges have occurred, bond issuers reduce their issue of bonds denominated in a currency that is foreign to them not in response to exchange-rate variations but in response to the abandonment of parities. Moreover, they react by somewhat increasing the proportion of unit-of-account bonds in their issue. Thus exchange-rate flexibility has raised the demand for a composite unit like the EPC.
    Secondly, the analysis has shown that non-bank depositors (unlike cantral banks) evaluate exchange risk not in terms of (past) nominal exchange-rate charges but in terms of real exchange-rate charges (and of deviations of the spot rate from the past forward rate). This means that an indexed-basket EPC which protects primarily against real exchange-rate risk will be a more attractive and useful money than a fixed-basket EPC defined on the lines of the new European Unit of Account.
     Thirdly, depositors (including cantral banks) and bond issuers prefer currencies which, in the long run, are subject to least inflation (or which, for other reasons, are expected to appreciate). In the money market, this preference for "strong" currencies is largely indepandent of observable covered arbitrage margins. A currency of guaranteed purchasing power like the EPC is thus likely to be in great demand as a standard and store of value
     414 Fourthly, the evidence indicates that yield and risk conaiderationa are of less importance ior the choice of a deposit currency than the size of the transaction domain. A new currency like the EPC which initially lacks transaction economies of scale is therefore likely to spread relatively slowly and gradually as a deposit money.
   Index clall ses relating to the national consumer price index may be viewed as a parallel standerd of value just like foreign currencies or multi-currency units of account. Thus the last of the three quantitative chapters contains an analysis of the determinants of the extent of ("demand for") indexation in four countries and with regard to four types of contract. The evidence indicates that the demand for an indexed standard of value like the EPC will be stronger, the larger the rate, and the increase (but not the decrease) of the rate, of inflation. It reacts with a lag of up to one year. Just as there remains a large number of new non-indexed contracts even in galloping inflation, many indexed contracts are concluded even in conditions of price-level stability. The inflation-autonomous part of indexation demand is larger, the longer the term of the contract. Thus even at low rates of inflation, the EPC will be demanded as a long-term standerd of value.
The last chapter of this study has presented the history of the parallel-currency approach and of currency competition.
With regard to private currency competition, the analysis shows that the competitive supply of private (base) money has worked welf where, like in Scotland and Switzerland, the state did not interfere with the policies of the issuing banks but merely rigorously enforced the law against fraud, and that inflation and bankruptcies were the result where, like in France (during the 18th cantury) and the United States (during the period of "wild-cat banking"), the state(s) meddled with the banking business and/or confounded freedom with lawlessness.
More important, a look at history reveals several cases in which parallel or dual currencies have been issued by governments with considerable success, either to get rid of inflation without a stabilization crisis or to tering about currency unification. The best examples for the anti-inflationary parallel-currency approach are the Massachusetts "equity bills" (1747-1749) and "depreciation notes" (1780-1786), the notes issued by the (Congressional) Bank of North America (17811783), the Soviet chervonets (1922-1924), the German rentenmark (1923-1924) and the Hungarian tax pengö (1946). On the other hand, the parallel-currency approach for currency unification (or a dual currency approach) has notably been applied in France (13th-18th century),the United States (1791-1811, 1816-1836, 1863-1865) and Italy
(1866-1927), and, as a limping union-currency standard, in Britain (1844-1921), Germany (1875-1935)and Japan (until 1882 ).
If empirical tests and historical precedents are capable of supporting a theory, then the case for a European parallel currency should benefit from this support.