EMU and Swedish Trade andrew k. rose

2001

Production: Herlin Widerberg Printing: Tryckmedia Stockholm

EMU andSwedish Tradeandrew k. rose

4

Contents

foreword 5

summary 6

introduction 6

methods for determining the relationship between currency unions and trade 7

estimating the relationship with a gravity equation 9

understanding the relationship 11

impact on sweden 12

broader implications 14

conclusion 15

technical annex 16

5Foreword

Sweden is, by tradition, an open country. It iswhen we have been open to trade, ideas andimpressions from abroad that we have developedfastest. We should encourage this trend. The newmonetary union that comes into effect in earnestin the New Year represents another importantstep forward for open economies. A single cur-rency makes its easier to engage in cross-bordertrade, travel, work and business enterprise. TheConfederation of Swedish Enterprise hopes torestore Sweden to the top of the GDP league,and joining the Euro-area is an important steptowards this goal.

The Confederation of Swedish Enterprise hasasked one of the world’s more prominent resear-chers into trade theory to analyze the effects onforeign trade of Swedish membership of EMU.Professor Andrew K. Rose has previously contri-buted an appendix to the Swedish government’s

report on EMU, which recommended in

that Sweden should adopt a “wait-and-see” stanceon the euro.

This study presents new figures indicatingthat Sweden would experience a dramatic increa-se in foreign trade if it joined the single currency.This would, in turn, give a considerable boost tothe country’s general prosperity.

These new findings, published now for thefirst time for a Swedish audience, show that thearguments in favor of joining EMU have becomeeven more convincing than they were five yearsago when the government report was written.

Stockholm, November

Göran TunhammarConfederation of Swedish Enterprise

Foreword

6 Abstract, Introduction

The effect of a common currency on trade is animportant issue. The increase in trade stemmingfrom a common currency is one of the fewundisputed gains from European Economic andMonetary Union (EMU). Even EMU-skepticssuch as Feldstein () agree that substituting asingle currency for several national currenciesreduces the transactions costs of trade withinthat group of countries. Indeed, this was one ofthe official motivations behind the EMU project(European Commission, ).

Clearly it is cheaper to trade between twocountries that use the same currency than bet-ween countries with their own monies. Thequestion is: How much? Skeptics believe that(intra-EU) trade may only rise a little because ofthe Euro. For instance, the Economic Reportof the President (pp -) states “… There isuncertainty as to how much additional benefitwill be yielded by the permanent fixing ofexchange rates implied by a single currency.”

This seems reasonable: exchange rate volatili-ty was low before EMU, and whatever volatilityremained could be inexpensively hedged throughthe use of forward contracts and other derivati-ves. Europhiles, in contrast, think that sharing acommon currency will lead to an increase in thedepth of trading relations, while precluding the“beggar thy neighbor” competitive devaluations

that can destroy a common market.The primary objective of this paper is to

resolve the argument by estimating the effect ofcurrency union on trade using empirical data.The results are clear. Currency unions do in facthave an effect on trade. And it is large; possiblyas big as the effect of joining a free trade arealike the European single market or NAFTA. Inparticular, I estimate that joining EMU is likely toincrease Sweden’s trade with the Euro-area substan-tially, leading to a significant increase in Swedishwelfare.

If entering EMU does substantially increaseSweden’s trade with the Euro countries, therewill be important repercussions. Perhaps mostimportantly, a big increase in trade will lead tosubstantial extra gains from trade, primarily forconsumers. In my work with Frankel (),I estimate that joining EMU may eventuallyraise Swedish GDP by over %, although thiseffect would be spread over decades. The reasonis that more open economies tend to grow morequickly and consequently enjoy higher standardsof living, even controlling for other factors.

To summarize: Sweden has much to gainfrom entering EMU in terms of increased tradeand consequently higher GDP, gains that havebeen under-stated in the existing academic litera-ture.

Introduction

note 1 This paper, commissioned by theConfederation of SwedishEnterprise, borrows from my rese-arch in this area, including myEconomic Policy 2000 paper“One Money, One Market:Estimating the Effect of CommonCurrencies on Trade”, my paper“EMU’s Potential Effect onSwedish Trade: A QuantitativeAssessment”, my CaliforniaManagement Review 2000 paper“Does A Currency Union BoostInternational Trade?”, and myjoint work with Charles Engel,Jeffrey Frankel and Eric vanWincoop.

I thank Harry Flam and LarsSvensson for advice and AllanÅberg for comments and encou-ragement. My research papersand data set are freely available atmy website.

Abstract

This paper1 estimates the effect of joiningEuropean Economic and Monetary Union(EMU) on Swedish Trade. Sweden is very opento international trade: exports and importscombined together amount to two-thirds ofSwedish GDP. Almost half of this trade is withthe Euro-area.

The empirical evidence presented here showsthat currency unions are associated with substan-

tially higher trade and welfare. I estimate thattrade with the Euro-area would increase conside-rably if Sweden joins EMU. Sweden’s trade withthe Euro-area would probably rise by over fiftypercent and could conceivably triple. A tradeincrease of this magnitude could result in a sub-stantial boost to Swedish output and welfare.

7Methods for Determining the Relationship between Currency Unions and Trade

The relationship between currency union andinternational trade is clearly important. Thus, itis no surprise that economists have worked hardto quantify the effects of reduced exchange ratevolatility on trade. Sadly, there is almost no con-sensus in the area, save that the effect (if any) isdifficult to estimate, even with high-tech time-series econometrics.

Much ink has been spilled on the issue ofinternational trade and the international mone-tary regime; there is a long and inglorious tradi-tion of ambiguous, weak and negative results.For instance, the Calmfors Commission (,p. ) stated:

“Many empirical studies have been done onthe effects of exchange-rate fluctuations on thevolume of foreign trade. The somewhat surpri-sing, but fairly unanimous, conclusion is thatthese fluctuations seem to influence foreign tradevery little, if at all. This conclusion must beregarded as fairly robust, because the various stu-dies have been done with different methods.”

Essentially, researchers have looked at periodsof high and low exchange rate volatility andattempted to map them into trade during thesame periods. Unfortunately, time-varyingexchange rate volatility simply does not seem tohave a strong effect on international trade orinvestment patterns. Basically, exchange ratevolatility for most of the OECD was low in thes, much higher in the s and s,and moderate in the s. The problem, forthis literature, is that trade has risen more or lesscontinuously. Unsurprisingly, it has been diffi-cult to establish a consensual view about thiseffect, or even its sign.

Not only is this literature weak; it is not evenclear that it is asking the right question. Havingeven a very stable exchange rate is not the sameas being a member of a currency union. Sharinga common currency is a much more serious anddurable commitment than a fixed exchange rate.This is manifest empirically in much more inten-se trade inside countries than between countries,a phenomenon known as “home bias” in inter-national trade.

McCallum () quantifies the size of the

intra-national bias at more than twenty to one(although this estimate is much disputed, e.g.,by Anderson and van Wincoop, ). In parti-cular, he finds that trade between two Canadianprovinces is more than times larger than tradebetween a comparable Canadian province/-American state pair. Part of this home bias effectmay stem from the fact that a single currency isused inside a country.

One might imagine that trying to measurethe effects of a common currency on trade is apurely academic (i.e., trivial) exercise. The onlycountries that have adopted a common currencyof late are the EMU-, for whom there arenecessarily few data. True enough. But there isno reason to rely on before and after differencesto estimate the effect of currency unions ontrade, just as one need not use time-series varia-tion to discern the effects of exchange rate volati-lity on trade. This paper exploits cross-sectionalvariation – using evidence across countries – totrace the effects of currency unions on trade.

Is a cross-country approach to investigatingcurrency unions doomed to failure since thereare so few of them? Not at all. Above andbeyond the twelve current members of EMU,some ninety “countries” are currently in somesort of official common currency scheme (thirty-one of these areas are official dependencies orterritories), as shown in the table.2 The empiricalwork in this paper hinges on exploiting theselinkages. In particular, the question is: “Docountries inside currency unions tend to trademore, holding other factors constant?” The otherfactors held constant are dictated by the “gravity”model of international trade, a framework with along track record of success.

Methods for Determining the Relationship between Currency Unions and Trade

note 2Most currency unions occurwhere one of the geographic unitsdoes not issue its own currency,and uses that of another. A fewoccur where there is massive curr-rency substitution (also known as“dollarization”) and two currenci-es exist with a long-term peg at1:1. I do not include currencyboards (such as Hong Kong orArgentina), countries that areinformally or unofficially dollari-zed (such as Brazil or Russia), orevents like German Unification in1990, or the re-integration ofOkinawa with Japan in 1972.

8 Methods for Determining the Relationship between Currency Unions and Trade

table 1: currency unions 1970-1995

australia belgium new zealand

Christmas Island (territory) Luxembourg Cook Islands (self-governing, associated

Cocos (Keeling) Islands (territory) with NZ)

Norfolk Island (territory) CFA franc zone Niue (self-governing, associated with NZ)

Kiribati Benin Pitcairn Islands (territory of UK)

Nauru Burkina Faso Tokelau (territory of NZ)

Tuvalu Cameroon

Tonga (pre ’71) Central African Republic turkey

Chad Northern Cyprus

denmark Comoros

Faroe Islands (part of Denmark) (Republic of) Congo uk

Greenland (part of Denmark) Cote d’Ivoire Falkland Islands (territory)

Equatorial Guinea (post '84) Gibraltar (territory)

east caribbean currency area Gabon Guernsey (dependency)

Anguilla (territory of UK) Guinea-Bissau Jersey (dependency)

Antigua and Barbuda Mali (post '84) Man, Isle of (dependency)

Dominica Niger Saint Helena (territory)

Grenada Senegal Scotland

Montserrat (territory of UK) Togo Ireland (pre '79)

St. Kitts and Nevis

St. Lucia italy usa

St. Vincent and the Grenadines San Marino American Samoa (territory)

Vatican Guam (territory)

france US Virgin Islands (territory)

French Guiana (overseas department) morocco Puerto Rico (commonwealth associated

French Polynesia (overseas territory) Western Sahara with US)

Guadeloupe (OD) Northern Mariana Islands (common-

Martinique (OD) norway wealth in political union with US)

Mayotte (territorial collectivity) Svalbard (territory) Swedish Virgin Islands (territory of UK)

New Caledonia (OT) Turks and Caicos Islands (territory of UK)

Reunion (OD) south africa Bahamas

Saint Pierre and Miquelon (TC) Lesotho Bermuda

Wallis and Futuna Islands (OT) Namibia Liberia

Monaco Swaziland Marshall Islands

Micronesia

france and spain switzerland Palau

Andorra Liechtenstein Panama

Barbados (2:1)

india singapore Belize (2:1)

Bhutan Brunei

9Estimating the Relationship with a Gravity Equation

The strategy of this paper is to link cross-countryvariation in currency arrangements to cross-country variation in international trade. Ofcourse, many things affect trade above andbeyond international monetary relations. Whilethese other factor are not of direct interest, theireffects need to be taken into account so as to beable to see if there is any remaining role for cur-rency unions. Ordinarily, this would be difficultin applied economics. But in this context, thereis a simple and persuasive model in which onecan embed the objects of interest to me: the“gravity” model of international trade.

The gravity model is a very simple empiricalmodel that explains the size of internationaltrade between countries. It models the flow ofinternational trade between a pair of countries asbeing proportional to their economic “mass”(read “national income”) and inversely propor-tional to the distance between them (literallyinterpreted). The gravity equation acquired itsname since a similar function describes the forceof gravity in Newtonian physics.

The gravity model of international trade hasa remarkably consistent (and thus, for econo-mics, unusual) history of success as an empiricaltool. The responses of trade to both income anddistance are consistently signed correctly, econo-mically large, and statistically significant in anequation that explains a reasonable proportion ofthe cross-country variation in trade.

The technical details – presentation of theprecise model, methodology and data set – arepresented briefly in the appendix. The appendixalso presents the actual estimates of the model.There are six different sets of estimates todemonstrate that the results do not dependstrongly on the exact specification of the econo-metric model.

Unsurprisingly, the standard features of thegravity model of international trade work well.For instance, both higher GDP and higher GDPper capita (for the country pairing) increasetrade. The coefficients are statistically significantand economically reasonable; both higher incomeper capita and larger country size increase tradeless than proportionately. The greater the distance

between two countries, the lower their trade. Allthree of these traditional “gravity” effects areintuitively reasonable, similar in magnitude toexisting estimates, and very statistically signifi-cant.

Sharing a land border, a language, or a regio-nal trade agreement also increase trade by econo-mically and statistically significant amounts. Ex-colonies and their colonizers, countries with thesame colonizer, and geographically disparateareas of the same state (for instance France andits overseas departments) all have disproportiona-tely intense trade, consistent with intuition andreceived wisdom. Landlocked countries and geo-graphically large countries trade less; islandstrade more. The equations fit the data well,explaining almost two-thirds of the variation inbilateral trade linkages. All this is well and good.

Above and beyond all of these real – andconventional – factors, there is compelling evi-dence that the international monetary regimematters. Countries that use the same currencytend to trade disproportionately, even holdingup to eleven other real factors constant. Theeffect (which is measured as the exponential ofthe coefficient on the currency union dummy) iseconomically large. A reasonable estimate is thatcountries with the same currency trade over threetimes as much with each other as countries withdifferent currencies!

Without taking the precise estimates too lite-rally, it seems clear that trade is substantially hig-her for countries that use the same currency, hol-ding other things equal. This positive resultstands in contrast to received wisdom. Forinstance, the European Commission (, p ) wrote: “Since the empirical research hasnot found any robust relationship betweenexchange rate variability and trade it is not possi-ble to estimate the increase in intra-EC tradethat might derive from the irrevocable fixing ofexchange rates.” The mistake the EC made wasin identifying currency union with the elimina-tion of exchange rate volatility, when belongingto a currency union is clearly very different fromsimply stabilizing exchange rates.

Extensive sensitivity analysis has been perfor-

Estimating the Relationship with a Gravity Equation

10 Estimating the Relationship with a Gravity Equation

med to check the robustness of these results;skeptical readers can check it out in Rose(b). In particular, the results do not dependsensitively on the exact way that the equation isspecified or estimated, or the exact way that thevariables are measured. An extensive search foromitted variables – which might lead one toconclude incorrectly that currency unions affecttrade when it is really some third factor thatmatters – turned up nothing. Reverse causalityalso does not explain away the findings. In all,some fifty different perturbations of the basicmodel yield no smoking gun. The effect of cur-rency unions on trade remains large and signifi-cant throughout.

Of course one should remember that nocurrency union of the size and scope and EMUhas been attempted before. Most currency uni-ons involve countries that are either small orpoor (or both). The enormous impact that cur-rency unions seem to have on these nations maythus be much bigger than the effect of EMU onEuropean trade. (Then again, it may not; “homebias” in trade indicates that a large expansion ofintra-European trade is plausible.)

Only time will tell for sure, so it is best notto take the estimation results too literally. Still,one can adjust the estimates to take into accountexisting trade patterns. Sweden’s external trade is

already integrated with that of the Euro area.But even taking these patterns into account, Iestimate in the appendix that joining EMU willlead Swedish trade to increase by more than%.

To summarize, the gravity model of interna-tional trade works well in a variety of differentdimensions. This bolsters confidence in my mainfinding: there is a large positive effect of a commoncurrency on trade.

11Understanding the Relationship

It is clear that a common currency shouldencourage trade. The puzzle is that the effectseems to be so enormous. Why does sharing acurrency have such an enormous effect on trade?There are many possible explanations. A com-mon currency represents a serious governmentcommitment to long-term integration. Thiscommitment could, in turn, induce the privatesector to engage in much more internationaltrade. Perhaps hedging exchange rate risk ismuch more difficult than commonly believed, asbusiness managers often state. Alternatively, acommon currency could induce greater financialintegration, which then leads to stronger trade ingoods and services.

More generally, money facilitates trade in itsroles as both unit of account and as medium ofexchange. Fewer, more widely accepted moneysfacilitate more trade, as has been recognizedsince at least Mundell (). Still, it is wisest toconclude that we simply don’t know why acommon currency seems to facilitate trade somuch. The most obvious benefit – foregoing thecost of hedging exchange rate risk – appears tobe low, especially for a sophisticated open coun-try like Sweden.

Nevertheless, even if we don't know why acommon currency makes a big difference, it isplausible that it does. The evidence presentedhere has separated the common currency compo-

nent from the other characteristics that differen-tiate within-country intranational trade fromcross-country international trade.

The evidence of intranational bias is disputedbut substantial; trade within countries appears tobe large compared to trade between countries,even for countries within well-integrated areaslike the European Union. Countries have a num-ber of important aspects for commercial trade,including a common currency, common culturalnorms, common legal system, common history,common norms, and so forth. A common cur-rency is a piece of this package; and it seems tobe an important piece. One need not take myprecise estimates too literally to agree with thisreasoning.

Understanding the Relationship

Sweden is an open economy. Data from both theIMF’s International Financial Statistics and theOECD’s Main Economic Indicators is tabulatedin Table 2. The IMF’s data indicates that trade ingoods and services as a proportion of GDPexceeded % in ; the OECD figures fortrade in goods alone are somewhat lower. It isclear that Sweden is a very open economy.

table 2: swedish openness

Sweden currently does approximately just lessthan half of its external trade with the twelvecurrent EMU countries. The exact figures are inTable , and vary slightly depending on whetherone uses data on trade in goods and servicesfrom Statistics Sweden or trade in goods alonefrom the OECD’s “Monthly Statistics of ForeignTrade.” But over % of Sweden’s exports go tocountries within the Euro area, and around halfof Swedish imports come from those countries.Thus EMU accounts for almost half of Swedishtrade.

This combination of strong ties to the Euroarea and openness means that Sweden has poten-tially a lot to gain from the trade boost whichjoining EMU may provide. In my work with

Frankel (), I estimate that Swedish entryinto EMU could result eventually in a tripling ofSwedish trade with EMU, raising total trade forSweden conceivably to almost % of GDP. Inthat work, we also find that every one percentincrease in trade (relative to GDP) eventuallyraises income per capita by roughly / of a per-cent over the long run.

If our estimates are accurate, the total increa-se in Swedish GDP that results from the tradeexpansion spurred by GDP could be substantial.While our estimates lack precision, we estimatethe eventual boost to Swedish GDP to be asmuch as %. Of course, there is considerableuncertainty about these results, and one shouldnot take the precise estimates too seriously. Theeffects will probably take decades to appear fully.

Further, most countries in currency unionsare considerably smaller than Sweden, so that theexpansion of Swedish trade may be overstated; Ishow in the appendix that adjusting for theseeffects reduces the impact of EMU on Swedishwelfare to a still large %. But even an effectthat is half as large would still be of enormousconsequence.

12 Impact on Sweden

Impact on Sweden

Billions of SEK. Source: OECD Main Economic Indicators.

Billions of SEK. Data from IMF International Financial Statistics CD-ROM.

A1995 1996 1997 1998 1999

Exports 694 686 774 828 863Imports 576 569 645 709 754GDP 1713 1756 1813 1890 1972X/Y 41% 39% 43% 44% 44%M/Y 34% 32% 36% 38% 38%(X+M)/Y 74% 71% 78% 81% 82%

B1996 1997 1998 1999

Exports, monthly 47.4 52.7 56.3 58.4Imports, monthly 37.4 47.8 45.4 47.2GDP 1756.4 1823.8 1905.3 1994.9X/Y 32% 35% 35% 35%M/Y 26% 31% 29% 28%(X+M)/Y 58% 66% 64% 64%

13Impact on Sweden

table 3: swedish bilateral trade pattern

PANEL A TradeExports Imports X Shares M Shares Intensity

country 2000 1999 2000 1999 2000 1999 2000 1999 2000 1999Germany 72224 63908 95530 83099 11% 11% 17% 18% 14% 14%UK 62213 54585 52061 48263 9% 10% 10% 10% 9% 10%USA 62038 52304 37043 27444 9% 9% 7% 6% 8% 8%Norway 49552 45039 44675 33688 8% 8% 8% 7% 8% 8%Denmark 38117 34903 41537 33642 6% 6% 8% 7% 7% 7%Finland 37081 32240 30340 25543 6% 6% 6% 6% 6% 6%France 34328 30506 31983 29831 5% 5% 6% 6% 6% 6%Netherlands 33155 35022 41613 38326 5% 6% 8% 8% 6% 7%Belgium 27372 25149 19997 17753 4% 4% 4% 4% 4% 4%Italy 25434 21188 17735 16244 4% 4% 3% 4% 4% 4%Spain 19364 20235 7541 8351 3% 4% 1% 2% 2% 3%Japan 17981 13475 16934 12954 3% 2% 3% 3% 3% 3%China 14184 10750 6854 5130 2% 2% 1% 1% 2% 2%Poland 11181 10437 7293 5159 2% 2% 1% 1% 2% 2%Turkey 8586 8473 2249 1335 1% 1% 0% 0% 1% 1%Switzerland 7695 8562 7297 7836 1% 1% 1% 2% 1% 2%Canada 7613 5476 2016 1876 1% 1% 0% 0% 1% 1%Mexico 7560 3681 1% 1% 0% 0% 1% 0%Austria 6584 6122 5363 4812 1% 1% 1% 1% 1% 1%Australia 6578 6956 1% 1% 0% 0% 1% 1%Hong Kong 5242 3579 7846 6645 1% 1% 1% 1% 1% 1%Taiwan 5090 4161 4493 3612 1% 1% 1% 1% 1% 1%Brazil 5060 5605 2128 1812 1% 1% 0% 0% 1% 1%Greece 4963 3942 1% 1% 0% 0% 0% 0%Saudi Arabia 4631 2392 1% 0% 0% 0% 0% 0%Ireland 4469 3791 9965 6987 1% 1% 2% 2% 1% 1%Malaysia 4442 1951 2953 1136 1% 0% 1% 0% 1% 0%Russia 3999 3302 4487 2961 1% 1% 1% 1% 1% 1%Singapore 3977 3203 1% 1% 0% 0% 0% 0%Portugal 3913 3772 3168 3229 1% 1% 1% 1% 1% 1%Estonia 6684 4176 0% 0% 1% 1% 1% 0%South Korea 3120 2142 0% 0% 1% 0% 0% 0%Latvia 2907 2872 0% 0% 1% 1% 0% 0%Czech Rep. 2664 2150 0% 0% 0% 0% 0% 0%Thailand 2341 1680 0% 0% 0% 0% 0% 0%All 7E+05 6E+05 5E+05 5E+05EU-15 4E+05 3E+05 4E+05 3E+05 56% 59% 65% 69% 60% 63%EMU-12 3E+05 2E+05 3E+05 2E+05 41% 43% 48% 51% 44% 46%

Values, Jan-Oct. Source: Statistics Sweden

PANEL BImports Exports Export Share Import Share

TradeIntencity

1998 1999 1998 1999 1998 1999 1998 1999 1998 1999World 5388 5414 6895 6910EU-15 3695 3636 3924 3972 57% 57% 69% 67% 62% 62%EMU-12 2778 2704 2862 2920 42% 42% 52% 50% 46% 46%

Millions US $, monthly ratesSource: OECD Monthly Statistics of International Trade

14 Broader Implications

The findings presented in this paper imply thatEMU may lead to an expansion of trade insideEurope. The rise in trade could be enormous;my estimate is that intra-European trade mayeventually triple. It will also be unexpected.

As a result, there will be great benefits forconsumers. The most important consequence ofincreased trade is increased gains from trade. Asthe deadweight loss of using different currenciesvanishes, competitive pressures increase, pricesfall and consumers gain. The size of these gainsmay be large; Frankel and Romer () estima-te that increasing the ratio of trade to GDP byone percentage point raises income per personby between one-half and two percent. Givenpotential gains of this magnitude, trade need nottriple for a common currency to induce largewelfare gains! There may also be dynamic gains if growth rates increase.

Even more visible consequences of an increa-se in trade caused by EMU may take place outsi-de EMU. If EMU causes radically increasedintra-European trade and its benefits, othercountries may well take the plunge, spreadingcurrency unions even further. Many countriesboth inside Europe and elsewhere are toeing thewater at present. Above and beyond Sweden, theUK and future EU-entrants are contemplatingjoining EMU. Argentina, Mexico, and Canadaare considering adopting the American dollar,while Ecuador and El Salvador have recentlydone so. If the benefits of a common currencyhave been underestimated, more will considerrelinquishing monetary sovereignty.

A large increase in trade precipitated forwhatever reason (such as the introduction of a

common currency) brings benefits but maybring also tensions. For instance, there may bean increase in trade disputes as a result of theincrease in trade. A common currency may crea-te much trade, but it may also divert trade fromlow-cost non-European producers to less effici-ent European producers who benefit from beingin EMU (though in my research I find no evi-dence of harmful trade diversion in the data).

An increase in trade also affects the verysustainability of the currency union. As tradeincreases, business cycles can in principle moveeither more asynchronously (as countries specia-lize to take advantage of comparative advantage)or more closely together (if most shocks aremonetary or most trade is intra-industry trade).The relationship between trade and businesscycle synchronization depends on the nature ofbusiness cycle shocks and the evolving economicstructure of the countries. Historically, closerinternational trade between countries has beenassociated with more synchronized businesscycles. Thus, an increase in intra-European tradeprecipitated by EMU, could make EMU itselfmore sustainable by increasing the synchroniza-tion of European business cycles.

Broader Implications

15Conclusion

The decision to enter a currency union is basedon many economic and political criteria. Thispaper has ignored nearly all of them. Still, cur-rency union-skeptics are skeptical in part becausethey perceive few advantages from a commoncurrency. One of the few undisputed benefits ofjoining a currency union is the encouragementof trade. That effect has not been quantifieduntil recently. Instead, economists have used thenegligible effect on trade of eliminating exchangerate volatility. As a result, the current consensusis that currency unions have hardly any effect ontrade. The case for a common currency is accor-dingly perceived as being weak.

I contend that such skepticism is unwarran-ted, so that a potent argument in favor of cur-rency unions has been under-stated in the litera-ture. Data for the many countries that sharecurrencies in the real world point to an unambi-guous conclusion. Even after taking a host ofother considerations into account, countries thatshare a common currency engage in substantiallyhigher international trade. And more traderesults in higher income. My estimate is thatSwedish trade with the Euro area could conceiva-bly rise by more than % as a result of Swedishentry into EMU, resulting in a boost to Swedishwelfare of over ten percent in the long run.

The case for currency unions is stronger thancommonly considered. The cost of foregoingindependent monetary policy may be low. Evenperfectly effective monetary policy has a smalleffect if the welfare costs of business cycles aresmall. Frankel and Rose () argue that busi-ness cycles may become more synchronizedacross countries because of currency union, fur-ther lowering the opportunity cost of nationalmonetary policy. Further, currency union may bean efficient institutional arrangement to handlecredibility problems, as Alesina and Barro ()discuss.

More importantly, I have tried to show inthis paper that currency union reduces tradebarriers associated with national borders, leadingto substantial increases in both trade and welfare.That is, a national currency seems to be a signifi-cant barrier to trade. Reducing these barriers

through joining EMU will result in increasedinternational trade for Sweden. The data indica-tes that this effect may be large, in excess of %for EMU. It will be unexpected. And it will bebeneficial; my estimate is that Swedish welfarewill rise by more than % as a result of EMU.

Sovereign monies are important (though per-haps inadvertent) national barriers to trade. Themonetary barriers are now falling across Europe.Sweden should seriously consider whether itwishes to forgo this historic opportunity for abeneficial expansion of its European trade.

Conclusion

16 Technical Annex

This appendix describes the model, methodologyand data set used to estimate the effect of com-mon currencies and exchange on trade.

the gravity methodologyAn augmented gravity model is used to estimatethe effects of currency unions and exchange ratevolatility on trade. The model is “augmented” inthat the standard gravity model only includesincome and distance variables. In order toaccount for as many other factors as possible, theequation adds a host of extra conditioning varia-bles as well as the all-important monetary varia-bles

ln(Xijt) = β0 + β1ln(YiYj)t + β2ln(YiYj/PopiPopj)t

+ β3lnDij + β4Contij + β5Langij + β6FTAijt

+ β7ComNatij + β8ComColij + β9Colonyij +β10Landij + β11log(Areai+Areaj)+ β12log(AreaiAreaj) + β13Islandij + γCUijt +ε ijt

where i and j denotes countries, t denotes time,and the variables are defined as:

• Xij denotes the value of bilateral trade betweeni and j,

• Y is real GDP,• Pop is population,• Dij is the distance between i and j,• Contij is a binary variable which is unity if i

and j share a land border,• Langij is a binary variable which is unity if i

and j have a common official language,• FTAij is a binary variable which is unity if i

and j belong to the same regional trade agree-ment,

• ComNatij is a binary variable which is unity ifi and j are part of the same nation (e.g.,France and its overseas departments),

• ComColij is a binary variable which is unity ifi and j were colonies after with the samecolonizer,

• Colonyij is a binary variable which is unity if icolonized j or vice versa,

• Landij is if both i and j are land-locked, isone of them is, and otherwise,

• Areai is the area of country i,

• Islandij is if both i and j are islands, is oneof them is, and otherwise,

• CUijt is a binary variable which is unity if iand j use the same currency at time t,

• β is a vector of nuisance coefficients, and• εij represents the myriad other influences on

bilateral exports, assumed to be well behaved.

The coefficient of interest is γ, the effect of acurrency union on trade flows. This is a coeffici-ent that has not been estimated by others in theliterature to my knowledge.

This equation is estimated with ordinaryleast squares, though the exact estimation techni-que turns out not to matter very much. I esti-mate various specifications of pooled regressionwith year controls (individual year results can befound in Rose, b); the last column alsoincludes country controls. To test the signifi-cance of individual coefficients, standard errorsare reported which are robust to heteroskedasti-city and clustering.

Substantial sensitivity analysis can be foundin Rose (b). In that paper I show that myresults are robust to: the exact measurement ofCU, the exact measure of distance, the inclusionof extra controls, sub-sampling, and differentestimation techniques.

the data setThe model is estimated using a data set with, bilateral trade observations spanning sixdifferent years (, , , , ,

and ). Observations are missing for some ofthe regressors so the usable sample is smaller. All countries, dependencies, territories, overseasdepartments, colonies, and so forth for whichthe United Nations Statistical Office collectsinternational trade data are included in the dataset. For convenience, all of these geographicalunits are referred to as “countries.” In this sam-ple, there are observations where two coun-tries trade and use the same currency.

The trade data are taken from the WorldTrade Database, a consistent recompilation of theUN trade data presented in Feenstra, Lipsey and

Technical Annex

17Technical Annex

Bowen (). This data set is estimated to cover% of all trade. Further description of the dataset can be found in my Rose (b).

standard gravity resultsThe results are tabulated in Table A1. The pointestimates of the currency union effect indicatethat two countries that use the same currencytrade more. Lots more. Since exp(.) ≈ ., theestimate without country-fixed effects indicatesthat currency union is associated with an increa-se in trade of around three hundred and fiftypercent. The effect is statistically significant,with a high robust t-statistic. This despite the

presence of eleven other controls (the least signi-ficant of which has a t-statistic of .)! Addingcountry effects reduces both the economic andstatistical impact of the currency union effect,but it remains economically large (a trade effectof over %) and statistically significant (the t-statistic is .).

table a1: impact of currency union on international trade, 1970-1995

Currency Union 2.11 1.53 1.22 1.25 1.37 .86(.13) (.13) (.13) (.13) (.13) (.19)

(Log) Distance -1.22 -1.09 -1.09 -1.04 -1.06 -1.31(.01) (.02) (.02) (.02) (.02) (.03)

(Log Produkt) .66 .64 .66 .56 .49 1.06Real GDP per capita (.01) (.01) (.01) (.01) (.01) (.04)(Log Product) .78 .79 .80 .88 .94Real GDP (.01) (.01) (.01) (.01) (.01)Regional 1.31 1.25 1.08 1.17 .46Trade Agreement (.07) (.07) (.07) (.07) (.12)Common .73 .44 .57 .53 .48Language (.03) (.04) (.04) (.03) (.06)Common .37 .43 .62 .63 .30Land Border (.07) (.07) (.07) (.07) (.13)Common .65 .47 .45 .68Colonizer (.05) (.05) (.05) (.08)Same 1.08 .97 .99 .81Nation (.28) (.28) (.29)( .32)Colonial 2.19 1.99 1.99 1.74Relationship (.07) (.07) (.07) (.13)Number of Land- -.39locked Countries (.03)(Log of) Sum -.22of Land Area (.01)(Log av) Product -.15of Land Area (.01)Number of .04Island Countries (.02)R2 .61 .62 .63 .64 .64 .72RMSE 2.05 2.03 2.00 1.98 1.98 1.74

Time Time Time Time Time Time Country Effects Effects Effects Effects Effects Effects

OLS estimation. Robust standard errors recorded in parentheses. Sample size = 31,101. Regressand is log of bilateral trade in real American dollars at 5-year intervals, 1970-1975.

18 Technical Annex

Since most currency union members are small,poor or remote, only about % of the observa-tions are members of currency unions. Still, thepaucity of observations does not appear to pre-vent them from having a strong and identifiableeffect. National money seems to be a significanttrade barrier.

more sophisticated resultsA currency union should stimulate trade some-what, since one money is more efficient thantwo as both unit of account and medium ofexchange. The real question is: why is the impactso large? In a world with derivative markets (atleast for developed countries), it is hard to belie-ve that lower transactions costs could lead tradeto rise so much. Perhaps the straightforward anddirect interpretation of the gravity model is mis-leading or inappropriate. How to proceed?

Anderson and van Wincoop (), here-after “AvW”, derive a simple theoretical gravityequation that easily lends itself to interpretationand estimation. There are four advantages tousing their structural approach. First, one canuse the model to investigate the impact of acurrency union among any set of countries, eventhose that have never been in a currency union.This is critical; without a structural model onemay question the relevance of pre-EMU curren-cy unions (which consist of small or poor coun-tries) when considering the impact of EMU.Second, it provides an estimate of the tariff-equi-valent of the national monetary barrier. Third,the model provides an explicit welfare metric.Finally, it may lead to a more accurate estimateof the impact of currency unions on trade.

Adopting the assumptions of complete speci-alization and identical constant elasticity of sub-stitution (CES) preferences that are central tothe previous theoretical gravity literature, AvWobtain a simple and intuitive equation:

xij = (yiyj/yW)(tij/PiPj)1-σ

where: xij is the nominal value of exports from ito j, yi is the nominal GDP of country i, yW is

the nominal value of world output,; σ is theelasticity of substitution between the countries’goods, tij is the gross price-markup due to tradecosts, and Pi is i’s “multilateral trade resistance,”a price index that depends positively on tradebarriers between i and all of its trading partners(not just j). Multilateral resistance can be solvedas a function of all bilateral trade barriers, {tij}.

In the model, trade between a pair of coun-tries depends on their bilateral trade barrier rela-tive to average trade barriers with all trade part-ners. According to the theory, each region pro-duces a fixed quantity of goods which have to besold somewhere in the world (analogous to theassumption of fixed factor supplies commonlymade in trade theory). More goods will be soldto a region with which the exporter has a relati-vely low trade barrier.

The theory has an intuitive implication forthe impact of currency unions on trade flows.The stronger the level of pre-union trade amongthe members of a currency union, the smallerthe percentage increase in trade among currencyunion members. If trade barriers are reducedamong a set of countries that already trade a lotwith each other, multilateral trade resistance willdrop a lot and relative trade resistance will falllittle. The drop in multilateral resistance ofmember countries reduces the impact on trade.

Pre-union trade levels can be high eitherbecause the countries have relatively low pre-union bilateral barriers (e.g. due to proximity ora regional trade agreement), or because the over-all size of the union is large. These considera-tions imply a smaller effect of EMU on bilateraltrade flows than most other currency unions.Existing currency unions, such as the EastCaribbean Currency Area, are small and there-fore imply a large effect on trade flows. Weexpect a smaller percentage increase in tradewhen Mexico or Canada dollarizes than whenArgentina dollarizes, as Argentina trades less withthe US than Canada or Mexico.

A rise in trade among members of the cur-rency union implies a corresponding drop intrade with other countries and within membercountries. That is, the model implies trade diver-

19Technical Annex

sion as well as trade creation. But there is a posi-tive welfare effect because fewer resources arewasted on trade costs. This is reflected in lowermultilateral resistance; the price index Pi falls.Welfare, as measured by the CES consumptionindex, can be shown to be approximately pro-portional to (1/Pi)

2. The more countries tradewith each other before joining the union, the lar-ger is the welfare benefit from joining the cur-rency union, but the smaller the percentageincrease in trade among union members. That is,welfare rises the most in currency unions wheretrade rises the least.

I estimate the AvW model using a linearcombination of the controls in Table A1 (otherthan land area and the GDP controls) for thebilateral trade barrier tij; details are availableonline. The model is estimated with country-fixed effects in place of the country-specificmultilateral resistance terms. and

data are used for a set of countries for whichthere is complete bilateral data, which is neces-sary to solve for the impact of currency unionson multilateral resistance and trade. The currencyunion coefficient remains large and significant at., with a robust standard error of .1. Thetheory tells us that this is an estimate of [(σ-1)ln m], where (m-1) is the tariff equivalent of thenational monetary barrier. If we use DavidHummels’ () estimate of σ_=5, the tariff-equivalent of the monetary barrier to trade isestimated to be %! While larger values of σreduce this estimate, for almost any value of σthe monetary barrier accounts for a little overhalf of the AvW estimate of the total nationalborder barrier.

The . estimate implies that the currencyunion is estimated to raise bilateral trade byaround % (exp(.) ≈ .), ignoring theeffect on multilateral trade resistance. But this iswarranted only in the unlikely case when there isa negligible amount of pre-union trade inside thecurrency union. To estimate the effect of currencyunions on trade more realistically, we need toincorporate multilateral resistance effects. That isdone in the Table A2 for a number of actual andhypothetical unions.

The theory allows one to estimate the effectsof currency union for any set of countries, evenif they have never been in one. The only assump-tion made is that the reduction in bilateral tradebarriers for union members is the same as thatfor existing currency unions. I tabulate the averagepercentage change of trade among countries inthe union, along with its standard error.

The key results are in the top two rows,which portray the experiment of Sweden joiningthe EMU- inside the Euro area. The first rowportrays the average effects across all thirteencountries; the second row portrays the effect onSweden. Other rows depict different currencyunions, and are tabulated to provide a means ofcomparison.

20 Technical Annex

table a2: impact of currency unions on trade and welfare

% Trade Increase % Welfare Increase

EMU-12 + Sweden 52 12.1(11) (3.3)

Effect on Sweden of joining EMU-12 53 11.8(11) (3.1)

EMU for original 11 members 58 11.1(12) (3.9)

EMU-11 + Grecce 59 11.1(12) (3.0)

EMU + UK 44 13.8(9) (3.6)

EMU for all (15) EU members 40 14.4(8) 3.8)

Argentina dollarizes 132 1.7(37) (0.5)

Ecuador dollarizes 106 4.5(26) (1.4)

El Salvador dollarizes 89 6.6(20) (2.0)

Mexico dollarizes 53 12.4(13) (3.8)

Canada dollarizes 38 15.3(9) (4.3)

Mexico and Canada dollarizes 27 18.4(8) (5.3)

Existing currency unions 91 5.0(22) (1.2)

World monetary union 10 21.3(2) (5.1)

Averages across countries except second row. Standard errors recorded in parentheses.

The trade-creating effects of currency union werelarge in Table A1; the effects are smaller in TableA2. Instead of EMU causing Swedish trade torise with the Euro area by %, it is estimatedto rise by % (the average increase in trade forthe Euro area plus Sweden is %), with a stan-dard error of %. Evidently taking multilateralresistance into account makes the effects appreci-ably smaller.

The trade-creating effects of currency unionsare smaller in Table 2; but the effects are large.(They are large even after dividing by two.)These large effects also characterize the otherEMU and dollarization scenarios.

The last column of Table A2 reports the effect ofcurrency unions on the welfare of their mem-bers, measured by the average percentage increa-se in the consumption index, (assuming σ=5)_.The welfare increases are large.

21References

Alesina, Alberto and Robert J. Barro ()“Currency Unions” NBER WP. No. .

Anderson, James E. and Eric van Wincoop() “Gravity with Gravitas: A Solution tothe Border Puzzle”.

Calmforsutredningen () Sverige och EMU(Statens offentliga utredningar :).

EU-kommissionen () “One Market, OneMoney” European Economy .

Feenstra, Robert C., Robert E. Lipsey and HarryP. Bowen (1997) “World Trade Flows, -

, with Production and Tariff Data” NBERWP. No. .

Feldstein, Martin () “The PoliticalEconomy of the European Economic andMonetary Union” Journal of EconomicPerspectives -, -.

Frankel, Jeffrey A. and David Romer ()“Does Trade Cause Growth?” American EconomicReview -, -.

Frankel, Jeffrey A. and Andrew K. Rose “TheEndogeneity of the Optimum Currency AreaCriteria” Economic Journal , (), pp.-.

Frankel, Jeffrey A. and Andrew K. Rose ()“An Estimate of the Effect of Currency Unionson Trade and Growth” NBER WP. No .

Hummels, David () “Toward a Geographyof Trade Costs” mimeo.

International Monetary Fund (IMF) ()International Financial Statistics CD-ROM.

McCallum, John () “National BordersMatter: Canada-U.S. Regional Trade Patterns”American Economic Review -, -.

Mundell, Robert A. () “A Theory ofOptimum Currency Areas” American EconomicReview , -.

OECD () Monthly Statistics of InternationalTrade December.

OECD () Main Economic IndicatorsJanuary.

Rose, Andrew K. () “Does a CurrencyUnion Boost International Trade?” CaliforniaManagement Review.

Rose, Andrew K. (b) “One Money, OneMarket: Estimating the Effect of CommonCurrencies on Trade” Economic Policy.

Rose, Andrew K. and Charles M. Engel ()“Currency Unions and International Integration”NBER WP. No. .

Rose, Andrew K. and Eric van Wincoop ()“National Money as a Barrier to Trade: The RealCase for Currency Union” forthcomingAmerican Economic Review.

Statistics Sweden () Export and Importdata from http://www.scb.se/snabb/expeng.htmand http://www.scb.se/snabb/impeng.htm

the author

Andrew K. Rose is B.T. Rocca Jr. Professor ofInternational Business, Economic Analysis andPolicy at Haas School of Business at theUniversity of California, Berkeley. He is alsoResearch Associate at the National Bureau ofEconomic Research (NBER), and ResearchFellow at the Centre for Economic PolicyResearch (CEPR).

References

This document is included in the idea series published by theConfederation of Swedish Enterprise. The series’ purpose is to reflectand stimulate debate around urgent questions that will influence thefuture of enterprise in Sweden. As an organisation, the Confederationof Swedish Enterprise does not necessarily support the content of all thedocuments in the series; rather it is the authors themselves who areaccountable for their contributions.

This document can be ordered from:Svenskt Näringsliv Förlagsservice- SandvikenPhone: +

Fax: + -

storgatan 19, se-114 82 stockholm, phone +46 8-553 430 00

www.swedishenterprise.se