a fundamental symmetry in international payment patterns

7/29/2019 A Fundamental Symmetry in International Payment Patterns

1/12

Journal of InternationalEconomics 3 (1973) l&-l 16.8 North-Hollandhblishing Company

A FUNDAMENTAL SYMMETRY NINTERNATIONALPAYMENTPATTERNSSven GRASSMANUniversity f Stockholm

1. IntroductionDiscussions on international monetary issues widely presuppose thatthe U.S. dollar is the dominant means of payments in internationaltransactions. In vol. 2, no 4 of this journal Richard Cooper, in discuss-

ing hat he calls the fundamental asymmetries of the world intema-tional monetary system, repeatedly refers to the extensive private useof the U.S. dollar as a medium of exchange . . (Cooper (1972), pp.327, 330, 33 1, 336). In fact, Coopers argument largely depends on thealleged dominance of the U.S. dollar as a vehicle currency. The assump-tion is only expressed in qualitative terms, and the only countries dealtwith explicitly as to actual currency distributions are Hong Kong andJamaica. Since the assumption as to the role of the U.S. dollar as ameans of payments seems crucial not only to Coopers argument but tomany aspects of international monetary analysis, quantitative data frommore important countries are needed.Another examp?e of the important role attributed to the dollar as ameans of payments is found in a recent essay by Robert Mundell 197 1,p. 3).(6. companies engaged in internationaltrade and those located abroad need a currency, aworld currency, They are using the dollar now and if the central banks of Europe move outof the market and let the excfiangerates up, the multinationalcorporationsand the interna-tional tradersstill need a world currerncy nd they can only US the dollar because it is theone that has the required nternationalmonetary properties.The belief that most international paymerIts are settled ;n dollars is sowidespread in the literature and in present and past debates that furtherexamples should hardly be needed. The pattern of international pay-ments, therefore, is a topic of general interest, which is not confinedonly to Coopers and Mundells articles referred to above.


2/12

106 S. Grassman, Fundamental symmetry in international payment pattemsContrary to the widespread assumption that most transactions aresettled in U.S. dollars, empirical evidence for countries where data on

actual currency distributions of foreign transactions are available indi-cate, in fact, that most transactions are settled in tha sd l ds 01 th epurchaser s currency, As shown below, this is the fundamental patternin international payments. The US. dollar, it turns out, is ua somewhat greater extent than would be indicated by the U.S. share ofworld trade alone.The limited role of the U.S. dollar in current international paymentswas revealed in a recent survey of payments habits in Sweden - thefirst. survey, I believe, to give a picture of the financial arrangements offoreign trade on a total, country-level basis (Grassman, 1971$ch. 3). Weshall present the evidence for Sweden - and a few other countries aswell - and briefly discuss the factors determining the choice of cur-rency in international payments. Some implications of the limited roleof the dollar as a means of payments are touched upon in the conclud-ing section.

2. The atnbiguous vehicle conceptThe dollar derives its capacity as a reserve and key currency pwtlyfrom the fact that it serves as vehicle currency or means of paymentsfor a fairly large part of total international transactions. Since the func-tions of the dollar are thus closely interwoven, it is interesting to assessthe role of the dollar as a vehicle currency in world payments. But,first, we should scrutinize more closely the meaning of the concept of avehicle currency, or that of mearls of payment or medium of exchange,which are all frequently used interchangeably. The meaning of these

three terms is not wholly clear, neither in Coopers article nor withmost other writers. The only authors who have, as far as I know, madea clear distinction between the various possible meanings of the vehicleconcept are Swoboda ( 1969) and Cohen ( 197 1).Swoboda is most careful to limit his discussion of th role of thedolhr to what he calls foreign-exchange market transactions in thenarrower sense as distinguished from the role of the dollar in intional economic relations in general* (Swoboda, 1969, p. 3 1). InSwobodas discussion the foreign-exchange market concept is qualifiedas to cover capital and inter-bank transactions only, and in this respect


3/12

the dollar no doubt is more dominant. Coopers article, however, dealswith the use of dollars in L much broader sense, not only by banks butby firms and the private public in gen ral. This is obvious from hisr&%VSlces o tile doll of accounts in tradin contracts,which are effluent of the medium of exchangeimportune of currency denomination insks for lo s and bankruptcies by fiisper, 1972, p* 332).2rate roles of international money: the

) a unit of account and a store4s - private transactions and officialfiis point of view the medium of exchange9 for ail practical purposes, inseparable:the currency denominations in contracts and offers have to be used inactual payments (unless exchange rate clau s are stipulated or, other-wise, the seller agrees to receive payment other currency than thecontract currency).In focusing the vehicle function it is useful, in addition, to have therather clearcut and operational distinction & terms of tmwctor, i.e.to distinguish between the dollar use of banks and the dollar use offirms This would roughly mean that inter-bank transactions are con-ceived of as the very transfer (or mutual clearing and set-off) mecha-nism, while fitms transactions n foreign exchange consist of quoting,price-furing n trade contracts, invoicing, and making and receiving pay-ments through the banks. The traditional vehicle concept readily en-compasses both these elements of the intem&ional payment me?ha-

ctions are predominantly in bollars, and the failurend firm tran ctions msry be one reason forhe U.S. dollar glays such 8 signifiesnt role inies of the general public. As is evident from

8 nnd aipitali : in 197122% of redptrtinated in U.S. ddlilf b,ts were 6ensnlnatcd in U.S. do&us (DanmaIks

3 Cohen, 1971, p. 17.


4/12

PO8 S. Grassman, undamental ymmetry n nternati onal payment p& ternsthe earlier quotations, the analysis of Cooper, Mundell and others refersto the role of the dollar for the public in general - such as the choice ofinvoice currency in foreign trade payments and the various activitiesrelated to this choice. It is this aspect of the vehicle currency which isrePevant when we are discussing e.g. uncertainty and disruptions intrade (Lanyi, 1969), the choice of a pivot-currency (Williamson,1971), the interrelation between the vehicle and reserve functions ofthe dollar (McKinnon, 1969; Cohen, 1971; Mundell, 1971) and manyother aspects of the international tilonetary system.In the following we shall, therefore, concentrate on the general pub-lics foreign-exchange transactions and the role played by the dollar onthis market. We thus broaden the perspective as compared to e.g.Swoboda b:y supplying empirical evidence as to the vehicle-aspect of thedollar as it stands out in the article by Cooper and others.

3. Actual currency dktribution: a Swedish case studyThere is a vast literature describing payment and financing,practicesin international trade, but almost all evidence has had the commonlimitation of being piecemeal and lacking generality. Usually it iseither based on practices in individual firms, banks etc. or on aggregatedadministrative materials which only casually happen to shed light onthe relevant issues, Thus the results have been either of a too specific ora too general nature. A few attempts have been made to map directlythe payments and credit patterns on a larger scale. But the resultsobtained have been of a poor quality - mainly due to the diffic&iesfirms experience in answering the questions posed (Hansen, 196 1, p.119). The solution to this dilemma was to get down to the individual

commercial transactions, to obtain a representative sample, and to posethe pertinent questions to the companies.The survey was carried out by the Swedish Central Bureau of Statis-tics, where about ten thousand items in the Swedish foreign trade sta-tistics for 1968 were chosen in a random sample. For the items choseneach company was questioned in detail about, for example, the formand size of payment for the specific export or import item, creditperiods, payment conditions, currency etc. The survey thus gave us acomprehensive picture of the financial structure of Swedish foreigntrade, including methods of payment, currency denominations, credit


5/12

S. Grassman, Fundamental symmetry in nternati onal payment patterns 109

Table 1The distributionof currenciesused in foreign trade payments, Sweden 1968a.currency Exports

Payments in per centimports

Swedish kronorU.S. dollarsPounds sterlingD-marksDanish kronerFrench francsNorwegiankronerSwiss francsLireOtherTotal

66.1 -12.3 ( 7.7)b11.2 ( 14.8)3.8 ( 11.6)1.8 ( 9.5)0.8 ( 4.7)0.7 ( 10.5)0.5 ( 2.4)0.3 ( 3.1)2.4 ( 35.7)

100 (100)

25.8 -22.0 ( 9.3)b17.3 ( 13.6)17.4 ( 18.7)3.9 ( 7.2)2.5 ( 4.5)2.2 ( 5.8)2.4 ( 2.6)1.8 ( 3.6)4.7 ( 34.7)

100 (lOO$a Sources: Grassman 1971) and SverigesRiksbank: Yearbook, 1971.b The share of Swedens trade due to the countries whose currenciesare shown in the tab!e isgiven within brackets.

periods offered and actually used etc. The information obtained shouldin principle be representative for the total Swedish exports and importsin the year 1968.The data for payments arrangements in the total Swedish foreigntrade were distributed and cross-tabulated according to several interest-ing variables. These were, for example, export and import currencies fortrade with different countries, commodity groups, size of payment,contractual form of payment and credit periods. It suffices here tc statethe actual distribution of exports and imports payments by differentcurrencies (table l), and to give a brief summary of the factors affectingthe choice of currency in foreign trade payments.4. The choice of currency

As is obvious from table 1, the general rule for both exports andimports is that they should be invoiced in the currency of the sellers orthe buyers country. According to the condensed table 2, the sellerscurrency is used in 66% and 59% of export and import payments re-spectively, while the purchasers currency is used in 25% and 26% of


6/12

110 s. Grassman, Fundamental symmetry in nternati onal payment patterns

Table 2Currency denomination in foreign trade payments, Sweden 1968a.Currency Exports Impot ts

(%I (%%)--Selling countrys 66b 59Purchasingcountrys 25 26bThird countrys 9 15Total 100 100a Source: Grassman (1971).b i.e. Swedish crowns.

export and import transactions respectively. There are, however, varia-tions and exceptions; in trade with some countries we even have theopposite relation. The general payment pattern is displayed in the detailed fig. 1.In Swedish trade with the United States and the United Kingdom thedollar and pound sterling respectively dominate both in exports and inimports. In Swedish trade with Eastern countries convertible currenciesare almost exclusively used, i.e. Swedish crowns or third-country cur-rencies. About 62% of the value of Swedish exports to the developingcountries was settled in Swedish crowns, while imports from Easterncountries are settled to the extent of 24.3% in third-country currenciesand only to the extent of 0.1% in the currency of the sellerscountry.(Trade with the LDCs and Easterti Europe amounts only to approxi-mately 15% of Swedens total exports and imports.) For both buyerand seller it seems to be an important considerdtion that the transactionshould be conducted in a currency with which each is well acquainted.This expiains the preferences for the sellers or the buyers own cur-rencies a.nd the fact that, when third-country currencies are used, this ismostly done in dollars (75%) or pounds sterling.Traditions in certain industries al.ldprice quotations for certain com-modities or in certain markets affect the choice of currency. Thus muchof the claw-material imports including oil is invoiced in dollars orpounds (see fig. 2). The pound plays an important role in, for example,exports of forestry products, but its role in shipping seems to have beentaken over by the dollar. Incidentally, it could also be mentioned thatthere is a posi,tive correlation between the size of transaction and theuse of dollar payment.


7/12

s. Grassnuzn. Fundamental symmetry in international payment pattems 111EXPORtS

U.K. , . . . . . " I t- _t EMt- USA Latl. RelIt All. . . . . of I. ... ea.da AMrl . of 00UIItri.1aacI E..... E.. , . . .lei

Per . . h . . . . . . . ' l i iN -- . , ' , , . . .. . .1" ......v,'. ow,..., Per . ._ iah . . . . . . .

_ 'hirel .-..try' reno' Selling --" ' . our,..."

Soandin- v.tavia Ger-"1u.c. France $wit - R .t Eaat- USA Latt.. h.' Alller- of V... caA...ica of oountrt..lancl EUNpe Europe world

Fig. 1. Sweden: Buyer's, seller'sand third country's currencies in tradewith various countries in1968.Source: see text.


8/12

112 S. G r a ~ s m a n , Fundamental symmetry in international payment patterns

' .-o.n\&, Sdi.h krclnorCPa "119.11an_..... ,. .ion o.r......I. .I II ......IcEIII.

Food- F....l . ehe.. TI.er Tex- Met. MI- 1"v_ eva Shipe Ot... All.tuffa prod- prod- t i l ala o h l ~ e q u l p - eto. goode ~uot. uot. r.r ...t

Food- F ~ I . Che.. T l ~ e r Tex. Met. MI- Eleotr. Car. Ship. Other Allstuff. prod- prod. tU. . ala ohi ....eqvlp- eto. p d e goodeuot. uot. ry ntFig. 2. Sweden: currency distributed overcommodities, 1968. Source: sue text.


9/12

S. Grassman, Fundament al symmetry i n nt empt bnal payment put t erns 113

The choice of currency of settlement can also be affected, it seems,by the availability of credit and other banking services, which vary fortransactions in different currencies. The relatively strong position of thedollar and the pound in international payments probably partly de-pends upon the fact that banks in New York and London have for along time played an important role in the f.inancing of internationaltrade.4 For some currencies no functioning forward market exists, andso when forward transactions have to be effected it is necessary to usesome of the generally accepted currencies.Apart from the general considerations of choice of currency dis-cussed here9 the choice is of course also affected by the specific andtemporary demands made on the portfolio composition of the cashbalances of the firm by the fymsdeliberate foreign exchange manage-ment. Two main considerations in this management are expectations asto exchange rate alterations and international interest rate differentials.Finally, the question as to which currency should be chosen in aparticular case is a matter of bargaining between the seller and thebuyer. Together with the various constraints mentioned above, it is theinterest in, and knowledge of, the question and the general bargainingposition of the parties that will be decisive. The overall pattern of allthese factors is, however, that # of foreign trade transactions are paidin the sellers own currency and * are paid in the purchasers currency- leaving for third country currencies only a marginal role.

5. Currency distribution in a global perspectiveUntil recently little was known concerning the representativeness ofthe Swedish data: does the general pattern of currency distributions

found in Sweden hold in other countries too? Evidence from othercountries where data are athered, such as Denmark and West Germany,indicate a p yments pattern very similar to that in Sweden.In Denmark forei xchange control records give a complete pictureof the currency di ution in foreign payments, and the data onDenmarks forei n trade payments (Danmarks Nationalbank, 197 1, p.31) conform to the basic pattern that transactions predominantly aresettled in the seller% or the buyers currency and that the U.S. dollar

4 Cf. the Danish case, discussed in section 5.


10/12

lP4 S.Grassman, Furuiamen tal symmetry in international payment patternsTable 3The distribution of currenciesused in foreign trade payments, Denmark1971 a.

Currency Exports imports(payments in per cent)

DanishkronerU.S. dollars PoundssterlingD-marksSwedishkronorSwiss francsOtherTotd

41 -22 ( 7,Q)b15 ( 19.3)8 ( 12.4)6 ( 16.2)2 ( 3.0)6 ( 41.2)

100 (100)

19 -25 ( 8.4)b14 ( 13.5)1!5 ( 18.5)11 ( 16.5)4 ( 2.4)12 ( 40.7)

100 WOBa Sources:DanmarksNationalbank,Report and accounts for the year 1971, and OECD,Over-all tradeby countries, October 1972.b The share of Denmarks rade due to the countrieswhose currenciesate shown in the tableis given within brackets.

and pound sterling play a marginal role in commercial transactions. Thedomestic currency, however, is used to a smaller extent than in Swedenand accordingly the U.S. dollar is used somewhat more than in Sweden(see table 3).One reason why the domestic currency s relatively ess significant nDenmark than in Sweden may be that the Danish krona has repeatedlybeen under strain during recent years. Another possible explanationwhy the U.S. dollar is somewhat more frequently used in Denmark isthe fact that Denmark heavily relies on capital imports and that manyof the cu,rrent transactions are refinanced by Euro-dollar Ioar&In West Germany the pattern of using the sellers or the buyerscurrency in foreign trade payments (leaving to the dollar and sterlingvery marginal roles) is still more accentuated than in Sweden andDenmark. Since the figures supplied to the writer by Deutsche Bundes-bank are approximate, and as they have not yet been released forpublication, I have been asked to confine myself in the West C&mancase to t:lis general qualitative statement.Until data become available for other countries, there are no a pi-iorireasons to assume that Sweden, Renmark and West Germany should

This point was suggestedby Mr. F. Hollensen, Vice Presidentof DanmarksNation&auk.


11/12

S. Grassman. Eundamental symmetry in ntemational payment patterns 115

differ from other industrialized countries as to the use of dollars inforeign trade.A calculation of the dollars share in woyld trade payments in 1970 -under the assumption that the pattern of Swedens foreign trade pay-ments is ~p~~nt~t~~ for payment arrangements in all other countriesthat US, dollars are used in 2 1If in the extrapolation the payments patte of world trade payments.6for Sweden is split up intothree grou transactions with Western industriahzed countries, theLDCs and Eastern countries, the percentage of total world trade pay-ments settled in U.S. dollars increases to 26%. The figures change onlyibly if the weighted avera of distributions in Denmark, WestGermany and Sweden is used as the blow-up standard instead of theSwedish figures alone.6. Conclusion

The purpose of this comment has been to assess quantitatively thesignificance of the U.S. dollar in world payments. The empirical evi-dence from Sweden, Denmark and West Germany suggests that the roleof the dollar as a vehicle currency has been overestimated. Approxi-mately one-fourth of global foreign trade payments are settled in dol-lars - reflectingmainly the United States own share in world trade.The different functions of the dollar in the international monetarysystem are closely interrelated, as are the functions traditionally as-signed to money as such in a closed economy. Thus the actual signifi-cance and use of the U.S. dollar as a vehicle currency should have somebearing on how to regard &hedollar also as a key and reserve currencyin a future international monetae system. The belief that most interna-tional payments are ttled in dollars has been generally held, .lnd it isimportant to point out that a change of assumptions on this pointmakes a difference in the reasonin on vital issues.ing and pinpointin the nature of the relationship between the

6 The oxtrapobtion is calculated an the assumption that the distribution between sellersand put&&r currency as well as the use of third-country dollars is the same in all countries,and that the U.S. share of world exports is 13. (GATT, 197 1, pp. and 109).

Hare it irrassumed that the Western industrialized countries account for 70% of worldexports (GATT, 1971, p. 6) and that the rest is settled to 38% in U.S. dollars (the actualpercentage in Swedens import payments to the non-western countries).


12/12

116 S. Grassman, undament al syrztnetry n nternational ayment patt erns

vehicle function and the other functionsof the dollarnowemerges s achallenging esearch task (cf. Swoboda, 1969). A preliminary eflectionis that the dollarsstatus as a reservecurrency is highlyconventional,nthe sense that for historical reasons, and due to certain propertiesofthis particular currency, the practice graduallyemerged o use the U.S.dollar as a reserve asset in most countries. One characteristic of thedollar which has contributed to making hiscurrency a reserve currencyis, no doubt, that after all it is the singlemost important vehiclecur-rency, since it is used in about 25% of internationalpayments. Thischaracteristicis, however, neither a necessarynor a sufficientconditionfor a currency to serve as an internationalreserveasset.* The vehicleaspect is only one of several factors explaining he positionof the U.S.dollar in the monetary system. To say that the dollar has derived tscapacity as a reservecurrency because it is the normal or dominantcurrency in internationalpayments is thus an error in both assumptionand inference.

ReferencesCohen, Benjamin J., 1971, The Future of Sterling as an International Currency Londonand

Ba%gstoke).Cooper, Richard N., 1972, Eurodollars, reserve dollars, and asymmetries in the internationalmoneta? system, Journal of International Economics 2,4,325-344.Danmarks Nationalbank, 1971, Report and Accounts for the Year 1971.GATT, 197 1, lnterrrtitional Trade 1970.Grassman, Sven, 197 1, Valu tareserven och utrikeshandelns fiiansiella struktur (Stockholm).Hansen, Bent, 1961, Foreign Trade Credits and Exchange ResenresAmsterdam).Lanyi, Anthony, 1969, The case for floating exchange rates reconsidered, ssays n Interna-tional Finance, No. 72 (Princeton).McKinnon, Ronald I., 1969, Private and official international money: the case for the dollar,Essays in International Finance, No, 74 (Princeton).Mundell, Robert A., 197 1, in Monte Dei Paschi di Siena, Note Economiche, Supplement to no.4, July-August.Swoboda, A., 1969, Vehicle currencies and the foreign exchange market: the case of the dollar,in: Aliber, R.Z., ed., The International Market for Foreign Exchange (NewYork).Swoboda, A., 1968, The Eurodollar market. An interpretation,Essays in InternationalFinance, No. 64 (Princeton).Wiliiamson, John, 197 1, The choice of a pivot for parities, Essays in International Finance, No.90 (Princeton).

a For a discussion of the interrelation between the different functions of an international cufirency and dn analysis of necessary and sufficient conditions for a currency to fulfill the reservefunction, see Cohen (1971), pp. 13-33, and Swoboda (1969), pas&z.

a fundamental symmetry in international payment patterns

Documents