Macroeconomics in an Open Economy

8/9/2019 Macroeconomics in an Open Economy

1/6

Macroeconomics in an Open EconomyAuthor(s): Richard N. CooperSource: Science, New Series, Vol. 233, No. 4769 (Sep. 12, 1986), pp. 1155-1159Published by: American Association for the Advancement of ScienceStable URL: http://www.jstor.org/stable/1697417

Accessed: 17/08/2009 03:20

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=aaas.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the

scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that

promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].

American Association for the Advancement of Science is collaborating with JSTOR to digitize, preserve and

extend access to Science.
http://www.jstor.org/stable/1697417?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=aaashttp://www.jstor.org/action/showPublisher?publisherCode=aaashttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/1697417?origin=JSTOR-pdf


2/6

Macroeconomcsn a n O p e n EconomyRICHARD . COOPER

The customary treatment of national economies as cloand self-contained must be substantially modifiedallow for those economies that typically trade gocservices, and securities with other countries in increasvolume. Open economy macroeconomics is essentialunderstanding the major events of the U.S. economy othe past half dozen years. Both the sharp rise in the doand the unprecedentedly large U.S. trade deficitlinked to the U.S. budget deficit, as is the drop in the tof inflation.

M ACROECONOMICS DEALS WITH THE ENTIRE NATIOeconomy, rather than with markets for particularpiucts, and has been principally concerned with explairfluctuations over time in output, employment, and the general 1of prices. Most theoretical work in macroeconomics is modeledclosed economy, that is, one that has no economic interactionthe rest of the world. Open economy macroeconomics, thereforthe subbranchof macroeconomics that allows for international tiin goods and services and for movements of capital across natiPboundaries in response to economic incentives. Theoretical maleconomics differs from the main body of economics by charactcing the economy as a whole on the basis of empirical observawithin a framework of consistent national economic accounts.

contrasts with microeconomic theory, which derives behavirelationshipson the assumption that decision-making units, famor firms, are maximizing some entity of value, such as profits.

A Simple Model for a Closed EconomyA widely used framework for exploring macroeconomics fociattention on two macroeconomic variables, the level of nati(income, Y, which in a closed economy equals national output,the rate of interest, r. The framework proceeds from two marclearing conditions:

E(Y, r)= YMIP = L(Y, r)Here E is total expenditure on goods and services in a given perof time, M is the supply of money, and P is the national price leEquation I shows that in equilibrium total expenditure on goand services must equal total income and that income andinterest rate will adjust so as to bring this about. Expenditurassumed to be negatively related to interest rates but positixrelated to income. To maintain the equality, higher interest r;requirelower levels of income. This balance in the market for goand services is maintained along a downward sloping schedule sas GG in Fig. 1.The supply of money (economists do not agree on the t

I2 SEPTEMBER I986

definition of this term) is assumed to be determined exogenously byosed the monetary authorities. L is the public's demand for the sameto money, depending positively on income (more money is needed for)ds, a higher level of transactions) and negatively on the rate of interest?ing (because higher r, which is the interest on bonds or other financialI to assets, leads households to conserve their lower yielding money inver order to hold less of their wealth in the form of "money").lilar Equilibrium in the money market thus leads to an upward-slopingare schedule such as LL in Fig. 1. The intersection of the two schedulesrate determines the interest rate and the level of income that will clearboth the goods and financialmarkets.Employment is determined bythe level of output.This framework has been widely used for examining various

NAL changes introduced into an economy. For example, an increase inrod- government expenditurenot offset by a tax increasewill shift the GGning schedule to the right, say, to G'G', and this will lead to a rise inlevel national income and the rate of interest, from K to K' in Fig. 1. Anon a autonomous investment boom (or, in an open economy, an exportwith boom) would cause a similar shift in the GG schedule. An increasein*e, s the supply of money by the central bank will shift the LL schedule torade the right and thus lead to a rise in national income and a fall in theonal interest rate.cro- This formulation does not account for the price level. It iseriz- Keynesian in flavor and arose from an attempt by the Britishtion economist and Nobel laureate John Hicks to characterize thethis macroeconomic equilibrium underlying the system put forward by[oral J. M. Keynes in 1936 (1). Keynes' theory was written at a time whenilies it was assumed that prices (more accurately, money wages) wouldnot respond either to an increase or to a decline in output in theshort run. That assumption has been one of the most controversialones in macroeconomics, especially after the rapid inflation of the1970's. It is possible, however, to adapt this framework to circum-stances when prices are variable and output is fixed at "full-uses employment" in the short run, but output is variable in the long runDnal in response to new investment. The transition dynamics fromand "short" to "long" run are, however, complicated and not well*ket- developed (2).This general framework of analysispermits comparison of short-run macroeconomic equilibria before and after some change to the(1) economy-an exercise in comparativestatics-but it lacksdynamics,(2) that is, the path of key variables moving from one equilibrium toanother over time. One way to deal with the dynamics is to assumeriod that all prices are fully flexible and demand equals supply continu-vel. ously, with the implication that there is no such thing as involuntaryods unemployment. Many unemployed workers would be astonished atthe such a conclusion. Moreover, the assumption of instantaneous ore is rapid clearing of markets makes it difficult to understand manyvely features of actualeconomies, such as the liquidation of inventories inates an economic downturn or the granting of noncontractual payods increases in a recession, when no new hiring is taking place (3).uch

best The author is the Maurits C. Boas Professor of International Economics at HarvardUniversity, Cambridge, MA 02138.ARTICLES I155

=~~~~~9 ! -


3/6

Fig. 1. Simultaneous eterminationof income,Y,and he nterest ate, .

Y

The formalization offered above pertains to a self-containedeconomy-one that does not trade with the rest of the world.Indeed, economics is generally taught in the United States as if theUnited States were a closed economy, with only brief mention givento the importance of recognizing that the United States trades ingoods, services, and assets with the rest of the world. The opennessis not usually integrated into the analysis.Moreover, the widespreaduse of American texts around the world suggests that a closedeconomy framework is taught in many other countries as well.

Adaptingthe Framework o an Open EconomyAnalysis of macroeconomic events must be adapted to take theincreasing openness of national economies into account, and thisanalysisinvolves greater complexity. The framework must incorpo-rate foreign buyers as a source of demand for domestic goods andforeign sellers as a source of supply. There may be net inflows oroutflows of funds, which create the possibility that total expenditureon goods and services may exceed or fall short of total income andoutput, as residents (including government) borrow from or lend tothe rest of the world. If exchange rates are flexible, that is, free torespond to market pressures, the effects of fluctuations on demandfor both goods and financialassetsmust be considered. With flexibleexchange rates, it is also necessary to specify expectations about

future values of the exchange rate, since prospective movements inthe exchange rate may influence current behavior.How should the macroeconomic analysisbe amended? First, theGG schedule must include net foreign trade, and its position can beinfluenced by the exchange rate, since a more depreciated currencywill generally enlarge the foreign demand for our country's goods.Equation I becomesY = E(Y, r, e) + X(Y* e) - I(Y, e) (3)

Here X is exports, I is imports, e is the exchange rate (units of homecurrency per unit of foreign currency), and an asterisksignifies therest of the world (4). Any excess of imports over exports of goodsand services must be "financed"by the sale of securities, such asstocks, bonds, and interest-bearing bank accounts, to foreigners.Indeed, the attractionof foreign securities because of their safety oryield may play a decisive role in determining the exchange rate,which in turn influences tradeflows and production. The growth ofinternationalcapitalmobility over the past two decadeshas been oneof the most notable changes in the international economy, and itplayed a major role in the move by large countries from fixed toflexibleexchange rates in the early 1970's. These developments haveled to new lines of economic theorizing.Second, the interest rate r* on foreign bonds and on domesticbonds will influence the demand for money. Opening the economyintroduces two further considerations bearing on the demand formoney. (i) The overall price level is no longer the same as the price

level for domestic output, since the former includes imported goodsand their prices in domestic currency may be changed by movementsin the exchange rate. (ii) Total expenditure need not be equal tototal income, as it is in a closed economy, and it is plausible thatexpenditurerather than income influences the demand for domesticmoney. The problem is compounded further for the United Statesbecauseforeignershold hundreds of billions in interest-bearingU.S.dollar balances, many of them outside the United States, andalthough exact information is not available,foreigners also probablyhold more than $20 billion in U.S. currency outside the UnitedStates. Should such holdings be counted as part of the U.S. moneysupply or not? (The former are not included in various measures ofthe U.S. money supply; the latterare.) But if they are not in the U.S.money supply, in whose money supply are they?Third, it is in principle necessary to "close" the model byspecifying behavioral relations for the rest of the world to allow forthe determination of Y*, P*, and r* within the economic system.The usual practicefor classroomor textbook exposition is to assumethat our country is so small that its influence on the restof the worldcan plausibly be neglected, so Y*, P*, and r* can be taken asexogenous factors of the economy. Changes in these variables are asource of disturbance to our economy. The small-country assump-tion is not plausible for the United States, however, which stillaccounts for about one-quarterof the gross world product. Analysisof the U.S. economy should take into account its impact oneconomic developments in the rest of the world and the feedback ofthose developments on the U.S. economy.Finally, if the exchange rate is an important, rapidly changingvariable,some allowance must be made for the influence on today'sbehavior of expected future changes in the exchange rate. Threeassumptionscompete in this question. (i) The present exchange rateis expected to last indefinitely. (ii) The present exchange rate isexpected to converge exponentially at some rate on the long-termequilibrium exchange rate, as determined by the model [usuallyonthe basis of "purchasing power parity" (PPP), a comparison of thedomestic price level with that prevailing in the rest of the world].(iii) The exchange rate is expected to be what the model willdetermine it to be in every period; this last assumption is sometimescalled "rationalexpectations."Under a systemof freely floating exchange rates,the exchangerateis a flexible price that is influenced by what happens in the marketsfor financial assets as well as the markets for goods. As such, it isresponsive to changes in marketviews about assets denominated inone currency relative to those denominated in other currencies. Itcan thus jump substantially, like stock prices, in response to newinformation that bears on future movement of exchange rates.Suppose, for example, the centralbank takesan action that leads to afall in domestic interest rates. Then holders of domestic bonds willfind foreign bonds more attractive (assuming that the yield offoreign bonds has not changed). Their attempt to purchaseforeignbonds will lead to an immediate depreciation of the home currency,overshooting its new medium-run equilibriumvalue to the point atwhich expected future appreciation of the home currency compen-sates for the difference in domestic and foreign interest rates. Thissituation is portrayed in Fig. 2, when the bond purchase occurs attime to.The exchange ratee, measured in units of domestic currencyper unit of foreign currency, immediately rises (that is, the homecurrency depreciates), and thereafter it falls gradually as the newlycreateddesire for higher yielding foreign bonds is satisfiedthroughan emerging trade surplus. (In terms of Fig. 1, the currencydepreciation after a rightward shift of LL will also shift GG to theright.) If the economy is near full utilization of its productivecapacity,domestic prices P will rise over time, so that by tl the realexchange rate (the exchange rate adjusted for changes in the price

SCIENCE, VOL. 233

L'


4/6

e, P Fig. 2. Time, t, profileof the ex-change ate, , andprices,P, afteranopenmarket urchase y the centralp bank.e

to ti

level) is restored to what it was before to. In terms of Fig. 1, both LLand GG shift gradually back to the left. [It is assumed here thatforeign prices and interest rates remained unchanged (5).]An implication of such "overshooting" is that there will be muchvariability n realexchange rates, andhence in internationalcompeti-tiveness, owing to developments in financial markets. If wages andother costs are not fully flexible, competitiveness as reflected inprofit margins will vary. This variation in profitability for reasonsarisingoutside each industry may reduce investment or encourage aworldwide diversificationof the activities of business firms, so as toreduce the exchange risk. Moreover, differing sectors of the econo-my are subject to differing degrees of foreign competition, somovements in exchange rates affectrelativeprices within the econo-my as well. If a country's currency appreciates, its "tradable"sector-those firms subject to foreign competition either in exportmarkets,such as U.S. grain farmers,or in the home market, such asU.S. textiles-will be put under competitive pressureand may sufferaccordingly. In contrast, its nontradable sector, for example, healthcareor the defense industry, will not feel such competitive pressureand indeed may benefit from the same factors, such as expansionistfiscal policy, which encouraged the currency to appreciate. Overtime, resources will shift from the tradable to the nontradablesectors.As the complexity of the economy grows, especiallyregardingthenumber of ways in which people can hold assets and the substitut-ability among those assets, stronger (and more limiting) assump-tions must be made about the structureof the economy in order todetermine even the sign of the impact on output, prices, and theexchange rate in response to specified changes in one of theexogenous variables (6). Examples of these variables include weath-er, strikes, technical innovation, or policy actions under the controlof governments. Since complexity greatly increases the number oftheoretically possible outcomes in response to outside shocks orpolicy actions, the likely outcome must be determined empirically.This process has only recently begun for open economies underflexibleexchange rates,partly because the length of experienceis notlong and because serious theorizing about this kind of regime isrelativelyrecent. Economists, like astronomers and meteorologists,must closely observe the environment as it exists, and then makecareful inferences. They cannot, like most scientists, set up experi-ments in which many of the potentially influential variables areclosely controlled. Yet they are expected to make judgments aboutconsequences of policy changes. Under these circumstances, econo-mists have resorted increasingly to numerical simulations on large-scale economic models. To provide the detailed quantitative struc-ture to such models it is necessary to estimate the coefficientswithout the benefit of controlled experiments. That is the task ofeconometricians, who operate on the assumption that the economicstructure is unchanged during the period of estimation, a problem-atic assumption for economies that are constantly undergoingevolutionary change. An alternative approach is to make informedguesses about the values of the coefficients and to test the sensitivityof the results to these guesses through simulation of alternativevalues (7).12 SEPTEMBER 1986

Two of the key empirical questions for open economy macroeco-nomics concern the degree of international mobility of financialcapital and the relation of the exchange rate to national prices. (Ofcourse, the flexibility of real wages, the responsiveness of prices toweakness in overall demand, and the sensitivity of investments,savings, and the demand for money to changes in interest rates arealso important in open economies.)The mobility of financial capital determines how closely interestrates and other financial variables are connected between countries,and therefore how much changes in monetary conditions will alterthe exchange rate and influence economic activity through theexchange rate channel. Various factors affect the internationalmobility of financial capital. The first is the presence or absence ofexchangecontrols, government-imposed impediments to the inwardor outward movement of funds. Broadly speaking, controls oncapital movements are absent for the United States, Canada, WestGermany, Great Britain (since 1979), Japan (since 1980), Switzer-land, and the Netherlands, but most developing countries and manyindustrializedcountries still prohibit or sharplylimit the outflow ofcapital in important ways.The second is the extent to which similar assets-for example,fixed interest securities, such as short-term time deposits or govern-ment bonds-are seen by investors as highly substitutable for oneanother, despite their different jurisdictions of issue (hence legalframework) and despite the fact that assets are denominated indifferentcurrencies. Empirical work has focused on the relation

r= r* + 4e + p (4)where for roughly comparable financialclaims, r is the interest ratein one country, r* is the interest rate in the second country, be s theexpected change in the exchange rate between the two currenciesover the maturity of the asset, and p is a risk premium. Tests ofr = r* do well for comparable securities traded in the majormarkets, such as New York and London, that are denominated inthe same currency, but very poorly when denominated in differentcurrencies. These results suggest that jurisdiction (at least amongmajormarkets) is relativelyunimportant but currencyof denomina-tion is important in reducing substitutability.Expected changes in exchange rates arenot directly observable. Itis appealing to assume, on average, ee= e; that is, market partici-pants estimate correctly the change in exchange rates that will takeplace. Tests of this proposition on the assumption that p = 0 do notperformat allwell. Efforts to introduce a well-behaved riskpremium(well-behaved in the sense that it is determined in some plausible,simple way) on the assumption that be= e also have not succeeded(8). One must conclude that denomination in different currenciessharply differentiates securities in the eyes of investors, but theprecise nature of the influence has not yet been discovered. Thelimited amount of direct survey data on expected changes inexchange rates does not help predict changes. Financial specialistssaid that they expected a drop in the dollar against other majorcurrencies over the period 1981-1984, yet the dollar, influenced bythe collective behavior of these and other financial operators,actuallyappreciatedduring much of this period (9).It is possible that participantsin financial markets have in mind along-run equilibrium exchange rate (e) and that they expect themarket rate to converge gradually on that equilibrium rate overtime, as shown by

e = ot(e- e) (5)where at indicates the rate of convergence.But what determines the long-term equilibrium rate?In principle,the answer is all variablesexogenous to the economic system, alongwith the structure of relations among variables. If there were some

ARTICLES 1157


5/6

simple binding constraint on this complex outcome, the exchangerate could be set by only a few relevantvariables in the system. Theleading candidate for this role is the purchasing power parityof thetwo currencies in terms of goods and services. Usually PPP in theperiod under scrutiny is compared with that of some base periodassumed to have been in equilibrium. Unfortunately, PPP does notexplain movements in exchange rates among the major currenciesvery well, either in the short run of 1 or 2 years or in the mediumrun of 5 to 10 years (10).Despite the weak empirical support for some of the detailedpropositions of open economy macroeconomics, these theories offersome insights into the functioning of modern economies, whichmay be overlooked or neglected when the model of a closedeconomy plus a foreign trade sector is used. In particular, hey offersome interpretation of the macroeconomic events in the UnitedStates over the past 5 or 6 years.Comparedwith the closed economy approach, fiscalexpansion inan economy with high international capital mobility and flexibleexchange rates may be less effective in stimulating overall demandbecause it leads to an appreciation of the currency. It is also lessinflationaryin the short run and might even lead to a reduction ofthe overall price level even while domestic output is expanding.Expansionist fiscal action is less likely to "crowd out" investmentwhen the economy is operating close to its capacity because itinduces inflows of foreign funds that help to sustain both domesticinvestment and the budget deficit. Exports arecrowded out instead.In Fig. 1, a given fiscal expansion in a regime of floating exchangerates will shift GG only as far as G"G"because of the currencyappreciation, and the consequential drop in prices will shift LL toL"L", eading to a new equilibrium at K" (11).Monetary policy too will be affected by the openness of theeconomy and flexible exchange rates. A given increase in the moneysupply may be more effective at expanding demand because it leadsto currency depreciation and a decline in interest rates. However,price increases from this source of demand expansionwill also showup sooner because of depreciation of the currency. In addition, thesectoral impact of monetary policy will differ. In a relativelyclosedeconomy, the impact falls on the interest or credit-sensitive sectors,

such as housing. In an open economy, it affects all tradablegoods,because of the exchange rate.

Interpreting Developments in theU.S. EconomyThese stylized results apply also to other disturbances that havestructuralsimilaritiesto changes in fiscal or monetary policy. Theydo not hold up under all degrees of complexity; but they arereasonably robust, and they help to interpret economic develop-ments in the United States during the past half dozen years, whenthe U.S. economy demonstrated a markeddegree of openness (12).The FederalReserve System adopted a tighter monetary policy inlate 1978 and changed its mode of operation in November 1979 toconcentrate on money magnitudes rather than on other indicatorsofmonetary conditions, especiallyinterest rates. One consequence wasan appreciationof the dollar from the low values of 1978. With thetax reduction and defense buildup of 1981, the United Statesadopted an expansionaryfiscal policy. Slightly earlier, in 1979 and1980, Great Britain, West Germany, and Japan adopted policies offiscal contraction. The result of this configuration of policies was asharp relative increase in interest rates on U.S. securities. Theattractionof U.S. securities was enhanced further by the relaxationof exchange controls in Great Britain and Japan and by politicaljitters in Europe arising from extensive nationalizations by the new

socialist government in France and by political turmoil in Polandduring 1981. The combination of high U.S. interest rates andchanges in asset preferencesled to a sharp appreciationof the dollaragainstother majorcurrencies as foreigners attempted to buy dollar-denominated assets. The tight U.S. monetary policy at first led theUnited States-and the world-into a deep recession in 1982. Theexpansionist fiscal policy, accompanied in late 1982 by somemonetary easing, led to a marked economic recovery in the UnitedStates in 1983-84.The U.S. dollar appreciated about 20% against other majorcurrencies (weighted by their importance in U.S. trade) between1980 and 1982. This appreciation contributed to a sharp drop inU.S. exports, which exceeded the drop in U.S. production duringthis period. The dollar appreciatedfurther until March 1985, to atotal (U.S. trade-weighted) appreciationof about 50% since 1980,partly as a consequence of the monetary-fiscal mix of policiesadopted in the United States and abroad. By year-endthe dollarhadreceded from its peak to a point 28% above its 1980 value, asmonetary policy eased and fiscal policy tightened marginally.High U.S. interest rates, the worldwide recession of 1982, andthe strong dollar were majorcontributing factors to the debt crisis indeveloping countries (and among U.S. farmers).The strong dollarmeant weakened dollar prices for many commodities in the worldmarket. Yet most of the external debt was denominated in dollars,so, while high interest ratesincreaseddebt-servicerequirements,therecession and the strong dollar weakened export earnings of manycountries and made it more difficult to service the debt.Once the United States began its rapidrecovery,the strong dollarreinforced growing U.S. demand for imports and that, by itself,eased the debt-servicing burden of many countries and also spreadthe U.S. recoveryto the rest of the world. During 1983 and partof1984, the U.S. economy was the majorsource of economic growthin the world.During this period of economic growth, the United Statesexperienced a sharp drop in the rate of inflation from more than12% in 1980 (heavily influenced by oil price increases) to less than4% in 1985. Some of this drop was due to the disciplining effect ofhigh unemployment and excess productive capacity on wages and

pricesduring the period, but perhapshalf of the drop was due to thesharp appreciationof the dollar. To the extent that the strong dollarhas led to an unsustainably large trade deficit, a portion of theseinflation gains were "borrowed from the future," so to speak, andwill have to be repaid as the dollar depreciates.In the meantime, the United States reduced its own foreign assetsand incurred new debts to the rest of the world by about $270billion between 1982 and 1985, and a further $300 billion to $400billion will probably be added before external balanceis restored.Asa result, the United States will have smaller net earnings on itsforeign investments in the future and will require a larger tradesurplus to compensate for the decline in net foreign earnings.Much of this development was foreseeable, but little of it was infact foreseen. The framework outlined above helps in the interpreta-tion of recent events, but it still falls short of being able to forecastwith confidence the detailed consequences of changes in policy inthe open U.S. economy.REFERENCES AND NOTES

1. J. M. Keynes, GeneralTheoryof Employment,nterest,andMoney (Harcourt, Brace,New York,1936).2. J.Tobin and W. Buiter, n MonetarismNorth-Holland, msterdam,976), pp.273-309. These authorsdiscuss the transitorynatureof the "equilibrium"postulatednthis rameworkf analysis, hichchanges sstocks f assets, ssumedconstant ere,change romperiod o period.3. A. Okun, Economics fPolicy-Making MIT Press, Cambridge, 1983), p. 137.4. Forfull development f an open economymacroeconomic odelwith adaptiveexpectations,see R. Marston [in Handbook n InternationalEconomics,R. W. JonesSCIENCE, VOL. 233


6/6

and P. B. Kenen,Eds.(Elsevier,Amsterdam, 985),vol. 2, pp. 859-916]. Foramodel with rationalexpectations, ee M. Obstfeld["Floating xchangerates:Experienceand prospects," BrookingsPapers n EconomicActivity,no. 2 (BrookingsInstitution,Washington,DC, 1985)].5. In technicalerms, he set of (linear)differentialquationsused to describe hemovements f the exchange ateandothereconomicvariables asa saddlepointsolution,wherebyone pathto equilibriums stablewhile the othersarenot. Inforward-looking arketshe exchange atewilljumponto the stablepath.Thisprocess ivesrise o theexchangeate"overshooting"ts new ong-run quilibriumvalue in order to maintainmomentary quilibriumby compensatingor thesluggishness r inflexibility f other prices n the economy.This phenomenonillustrates e Chatelier'srinciplen physicsasapplied o an economic ystem. nFig.2, thecumulativeurplus chieved etween oand , willresult n new claimsonthe restof theworld,hencenewearnings n thoseclaims.Torestore situationin whichnet claims reno longerchanginghereforeequiresomeappreciationfthe home currencyn realterms,for example,a tradedeficitoffsetby the netearnings romabroad.Thus in the long run, the e scheduledropsbelowthe Pschedule Fig.2).6. J. Tobin and J. B. de Macedo, in Flexible Exchange Rates and the Balance ofPayments,. S. Chipman nd C. P. Kindleberger, ds. (North-Holland, mster-dam,1980), pp. 5-28.

and P. B. Kenen,Eds.(Elsevier,Amsterdam, 985),vol. 2, pp. 859-916]. Foramodel with rationalexpectations, ee M. Obstfeld["Floating xchangerates:Experienceand prospects," BrookingsPapers n EconomicActivity,no. 2 (BrookingsInstitution,Washington,DC, 1985)].5. In technicalerms, he set of (linear)differentialquationsused to describe hemovements f the exchange ateandothereconomicvariables asa saddlepointsolution,wherebyone pathto equilibriums stablewhile the othersarenot. Inforward-looking arketshe exchange atewilljumponto the stablepath.Thisprocess ivesrise o theexchangeate"overshooting"ts new ong-run quilibriumvalue in order to maintainmomentary quilibriumby compensatingor thesluggishness r inflexibility f other prices n the economy.This phenomenonillustrates e Chatelier'srinciplen physicsasapplied o an economic ystem. nFig.2, thecumulativeurplus chieved etween oand , willresult n new claimsonthe restof theworld,hencenewearnings n thoseclaims.Torestore situationin whichnet claims reno longerchanginghereforeequiresomeappreciationfthe home currencyn realterms,for example,a tradedeficitoffsetby the netearnings romabroad.Thus in the long run, the e scheduledropsbelowthe Pschedule Fig.2).6. J. Tobin and J. B. de Macedo, in Flexible Exchange Rates and the Balance ofPayments,. S. Chipman nd C. P. Kindleberger, ds. (North-Holland, mster-dam,1980), pp. 5-28.

7. For a summary f empiricalwork on open economies, ee J. Helliwelland T.Padmore [in Handbook n InternationalEconomics,R. W. Jones and P. B. Kenen,Eds. (Elsevier,Amsterdam, 985), vol. 2, pp. 1107-1151; and the referencestherein].For a quantitativemodel of the United States hat allowsexplicitlyordisequilibriumin some markets, see R. Fair [Specification,stimation,andAnalysis ofMacroeconomicodels HarvardUniv.Press,Cambridge, 984)].8. For summaries f and referenceso the work in this area,see K. Rogoff [J.Monetary con.14, 133 (1984); and(9)].9. J. Frankel,in How Openis the U.S. Economy?,R. W. Hafer, Ed. (Lexington Books,Lexington,MA, 1986), pp. 33-67.10. R. Dornbusch,"Purchasingower parity,"n TheNew Palgraves ictionary fEconomicsMacmillan, ondon,1986).11. Theslopeof the GGschedulewouldbe altered yamove romaclosed o anopeneconomy,with the GG schedulebecoming steeperbecausesome part of anyadditionalxpenditure ill be usedfor thepurchase f importsandthuswill notincrease domestic output.12. The ratio of imports o the U.S. gross nationalproducthas more thandoubledsince he mid-1960's.The shareof foreignownership f securitiesn the UnitedStates nd he ratioofforeign o total oansbyU.S. banks averoughlyrebled. eeR. Cooper, n HowOpens theU.S.Economy?,. W.Hafer,Ed. (LexingtonBooks,Lexington,MA, 1986), pp. 3-24.

7. For a summary f empiricalwork on open economies, ee J. Helliwelland T.Padmore [in Handbook n InternationalEconomics,R. W. Jones and P. B. Kenen,Eds. (Elsevier,Amsterdam, 985), vol. 2, pp. 1107-1151; and the referencestherein].For a quantitativemodel of the United States hat allowsexplicitlyordisequilibriumin some markets, see R. Fair [Specification,stimation,andAnalysis ofMacroeconomicodels HarvardUniv.Press,Cambridge, 984)].8. For summaries f and referenceso the work in this area,see K. Rogoff [J.Monetary con.14, 133 (1984); and(9)].9. J. Frankel,in How Openis the U.S. Economy?,R. W. Hafer, Ed. (Lexington Books,Lexington,MA, 1986), pp. 33-67.10. R. Dornbusch,"Purchasingower parity,"n TheNew Palgraves ictionary fEconomicsMacmillan, ondon,1986).11. Theslopeof the GGschedulewouldbe altered yamove romaclosed o anopeneconomy,with the GG schedulebecoming steeperbecausesome part of anyadditionalxpenditure ill be usedfor thepurchase f importsandthuswill notincrease domestic output.12. The ratio of imports o the U.S. gross nationalproducthas more thandoubledsince he mid-1960's.The shareof foreignownership f securitiesn the UnitedStates nd he ratioofforeign o total oansbyU.S. banks averoughlyrebled. eeR. Cooper, n HowOpens theU.S.Economy?,. W.Hafer,Ed. (LexingtonBooks,Lexington,MA, 1986), pp. 3-24.

Molecular Transformationsn S i n g l e Cry s t a lM e ta l Surfaces

R. J. MADIX

Molecular Transformationsn S i n g l e Cry s t a lM e ta l Surfaces

R. J. MADIXOne of the primary objectives of modern surface chemis-try of transition metals is the synthesis of surface com-pounds and complexes and the understanding of theirreactivity, structure, and bonding. Such considerationsare paramount for advancing understanding of catalysis,adhesion, organic thin-film growth, and electrocatalysis.On selected metals, particularly copper, silver, and gold,selective scission of X-H bonds (where X is oxygen,carbon, nitrogen, or sulfur) by surface-bound atomicoxygen occurs to form moderately stable species that canbe isolated for further study. Selective oxidation reactionsmay occur heterogeneously by means of this novel oxy-gen-activated route. Furthermore, this selective chemistryoffers a paradigm for synthesis of a wide variety of surfaceorganometallic complexes, whose formation can be pre-dicted from acid-base principles. These subjects are dis-cussed in this article with emphasis on their role incatalytic oxidation cycles.

DEALLY,A CHEMICAL CATALYSTOR CATALYTICAGENT IN-creases the rate of a chemical reaction without itself beingconsumed in the reaction. For example, without chemicalintervention carbon monoxide (CO) and nitric oxide (NO), formedin the combustion process, exhaust from an automobile withoutreacting further and contribute to atmospheric pollution. In thepresence of a catalyticconverter, these gases react to form harmlesscarbon dioxide (CO2) and N2. The relativerate at which a reactiontranspireson a given catalyst is known as the catalyst activity; theI2 SEPTEMBER I986

One of the primary objectives of modern surface chemis-try of transition metals is the synthesis of surface com-pounds and complexes and the understanding of theirreactivity, structure, and bonding. Such considerationsare paramount for advancing understanding of catalysis,adhesion, organic thin-film growth, and electrocatalysis.On selected metals, particularly copper, silver, and gold,selective scission of X-H bonds (where X is oxygen,carbon, nitrogen, or sulfur) by surface-bound atomicoxygen occurs to form moderately stable species that canbe isolated for further study. Selective oxidation reactionsmay occur heterogeneously by means of this novel oxy-gen-activated route. Furthermore, this selective chemistryoffers a paradigm for synthesis of a wide variety of surfaceorganometallic complexes, whose formation can be pre-dicted from acid-base principles. These subjects are dis-cussed in this article with emphasis on their role incatalytic oxidation cycles.

DEALLY,A CHEMICAL CATALYSTOR CATALYTICAGENT IN-creases the rate of a chemical reaction without itself beingconsumed in the reaction. For example, without chemicalintervention carbon monoxide (CO) and nitric oxide (NO), formedin the combustion process, exhaust from an automobile withoutreacting further and contribute to atmospheric pollution. In thepresence of a catalyticconverter, these gases react to form harmlesscarbon dioxide (CO2) and N2. The relativerate at which a reactiontranspireson a given catalyst is known as the catalyst activity; theI2 SEPTEMBER I986

degree to which the catalyst directs the course of the chemicalconversion toward one product is known as its selectivity. Normally,one desires a maximum rate along with a high selectivity, since thiscombination maximizes production. For specialty products, howev-er, selectivity considerations may dominate.Often, catalytic materials interact with the reactant molecules toform intermediate compounds or complexes that undergo furtherreaction easily. For example, in the conversion of N2 and H2 toammonia (NH3), if the catalystis to be of any utility, it must providea more facile path than that available in the gas phase. Thisconversion is accomplished catalytically by N-N and H-H bondscission on the surface of the metal catalyst to produce atoms ofnitrogen and hydrogen bound to the metal (1). A reaction sequencethen occurs on the surface with energetic barriersfor each step thatare lower than the barrierwould be in the absence of the catalyst.This mechanistic principle of catalysis also applies to complexcycles of reactions in living systems (2). However, even for simplereactionschemes, a number of reaction paths can be accessible to thereactants; he selective formation of a single product depends on therelative rates of the possible reaction channels. Thus, an understand-ing of the kinetics of these conversions is important for progress inthe field, and manipulation of the state of the surface may enhanceor suppress certain reactions. Furthermore, specific identification ofthe intermediates formed and a detailed understanding of theavailablereaction channels is crucial. It may be possible to guide thereaction in more desirable directions by intercepting it at specificstages. The methods of surface science offer the opportunity todissect such processes on the molecular scale and to predict reaction

degree to which the catalyst directs the course of the chemicalconversion toward one product is known as its selectivity. Normally,one desires a maximum rate along with a high selectivity, since thiscombination maximizes production. For specialty products, howev-er, selectivity considerations may dominate.Often, catalytic materials interact with the reactant molecules toform intermediate compounds or complexes that undergo furtherreaction easily. For example, in the conversion of N2 and H2 toammonia (NH3), if the catalystis to be of any utility, it must providea more facile path than that available in the gas phase. Thisconversion is accomplished catalytically by N-N and H-H bondscission on the surface of the metal catalyst to produce atoms ofnitrogen and hydrogen bound to the metal (1). A reaction sequencethen occurs on the surface with energetic barriersfor each step thatare lower than the barrierwould be in the absence of the catalyst.This mechanistic principle of catalysis also applies to complexcycles of reactions in living systems (2). However, even for simplereactionschemes, a number of reaction paths can be accessible to thereactants; he selective formation of a single product depends on therelative rates of the possible reaction channels. Thus, an understand-ing of the kinetics of these conversions is important for progress inthe field, and manipulation of the state of the surface may enhanceor suppress certain reactions. Furthermore, specific identification ofthe intermediates formed and a detailed understanding of theavailablereaction channels is crucial. It may be possible to guide thereaction in more desirable directions by intercepting it at specificstages. The methods of surface science offer the opportunity todissect such processes on the molecular scale and to predict reactionThe author s professor ndchairman f theDepartmentf Chemical ngineering ndprofessor f chemistry bycourtesy) t StanfordUniversity, tanford,CA 94305.The author s professor ndchairman f theDepartmentf Chemical ngineering ndprofessor f chemistry bycourtesy) t StanfordUniversity, tanford,CA 94305.

ARTICLES 1159RTICLES 1159

- ~ ~ _ I I~ ~ _ I I

Documents

Macroeconomics in an Open Economy