World Bank Reprint Series: Number 228 Guy P. Pfeffermann Latin America and the Caribbean: Econonlic Peformance and Policies Excerpted with permission from Thle Soittlhwestern Review of Manogemiienit and Econiomtiics, vol. 2, no. 1 (Winter 1982), pp. 129-59. Latin America and the Caribbean: Economic Performance and Policies Guy P. Pfeffermann EIntroduction This article reviews the econiomic performance of Latin American and Caribbean nations and relates it to underlying policies in order to forecast likely growth rates for the 1980s. The article is intended mainly for those who are not familiar with the Latin Americanl economi2 scene and is thus not a detailed scholarly analysis of the inany complex relationships involved. The views expressed are my own. My judg- ment is, of necessity, colored by the thirteen years I have worked as an economist in the Latin American and Caribbean Regional Office of the World Bank. The first section provides background on the natural and human resource base of the Region-its strengths and weaknesses. These fundamental factors continue to influence the Region's economic performance and its potential. The second section ie'.iews econonmic performance as well as related aspects such as population expan- sion, the urban/rural balance and social progress. The stress is largely on differences between countries. The third sectioi deials with a few critical domestic policy areas- agriculture, finance and externial Lebt --and relates them to performance. Throughout, the work "Region'" refers to the Western Hemisphere nations south of the United Staies, with the -\centrion of Cuba for which comparable data are not available. Natural and Hulmaii Resources This section. looks at the principal elements of growth-natural and human resoiurces-over time zind stresses the rural,/urban cleavage which has persisted oilirouehorl the Region's modern histOry . Geography and ,Vatural Asses The fortwlu!es of Latini \nicrica l. c beeni -pro lrindly influenced by its hmmen- skv and its minleral wealth. The c i rt is 8 ast. The Indian s'Ubcontinenrt \ %otld fit t ' r^ i>. 1' : fir~lt :: i. ( F; t(i st efor the atin Armrica and ( ril'--. .u Region at the World Banik. 129 130 . HE SOUT H VI S I I 51 N REVIEW Winter 1982 more than six timies into Latin Aiimcrica, and Brazil aloic is almost as large as China. TheC continent eXtcind from south of the Untiited States to Ushuaia, the southernmost town in the world. There are vast stretches of desert, the world's largest river basin, and the Anides, separating the Pacific coast frotmi the rest of South Anerica, which are .>ecoud onlv to the Himalayas in height. The Atlantic coast of Brazil i.s the mirror image of West Africa's, very similar in climiate and vegetation. The fe-tility of the panmpas is inatched on a similar scale onlv in the Midwest of the United States and the black lands of Russia. Table 1 -ho'A > some figures on the natural resource wealth of the Region today. rhe Region produices a signiificant share of the wvorld's output of bauxite, copper, lead, tin, ziinc, sil. er, tunigsteni, and crude oil. Its reser% es of bauxite, copper, iron ore and nickel reprcsent a larger share of world reser%es than the Region's share of otutput, which uzUggets a potential for increasedl production. The Recion's total resouirces (including those that are not profitably exploitable today) represent an even larger share of world resoiurces in bauxite arid tin. These sketchy figures convey a pieture of a resource-rich continent. This is truie also for the Region's hydroelectric as well as for its agricultural, forestrv, and i'. e"tock potenitial. Human Resources As late as 1800, three centuries after the Portuwiese first landed, only about three million people lived in Brazil, and, according to Humboldt, 17 million in Spanish America. The scant population was d\.s8rfed by the immense natural resource wealth of the continient. WVhile the North American colonists had to make a living from agriculture, ofteni in difficult ecological conditions and without the help of an Indian labor force, Latin America was developed 'rom the start with the idea of exploiting natural re.ources and the labor of the Indiws. In this sense the existeince of natural wealth has been a curse as well as a blessing. The Spanish historian Salvador de Maldariaua noted: "That a man of quality, by fighting, acqtmires wealth more honorably and quickly than a meaner man by work has been called a basic belief of the Spaniards as they emerged from Reconquest" (Hirschinan, 1977, p. 58). Or: "Spanish America was born of a gold rush, just as were, in later years, Australia, California, South Africa, and the Klondike. Men sold all they had, left their families, and journeyed to lands of treasure" (Herring, 1962, p. 198). As it turnecd out mining was a distinct success. By 1800, Spanish Amnerie.a and Portuguese Brazil were producing about 90 percent of the world supply of iirecioti metals. Agriculture was based on slas cry or, what amounted to much the same, the encomnienda s%iteni of Spanish America. And even though manufacturing made surprising progress in the face of opposilioni by Spanish monopolists, the ganmbling spirit asociated with mineral exploitation, the Spanish tradition that manual labor is demeaning, and the association between slavery and agricultural toil all help explain the fitful economic performance of the Region during the first four centuries after colonization. I'litorically, lack of and iniappropriateness of cducation is a recurring theme: "'he provision for Llducating the people was meager and haplhazard. Learning was Lhletly a pri ilege accorded the sonw, of the more proqperomms Spaniards, Creoles and n-we-ti/os (di0eihtcr,, of course, were not rcuarded as educable).... As a result, ,l.unish Anie-kca '.as largely illiterate at the enid of the colonial period'' (Herring, l962, p. 209). Table 1. Natural Resources Relative lanporlanec of Latin America and the Caribbean A A( as a perceiit of World Indiex (LAC 10=) Olutpult 197, 9 Output -per catpita (19'9) Rcources per ipita Other Indiustrial Other Indtistrial 196(1 19-) Reserl-esa Resourcest, ID(\ (Countries 1t1) Cs Countries liiti\ilc 47 2 6 28 64 18 31 8 2 Copper 19 20 34 na 23 47 14ka '20 1Lon )Ore 8 9 13 12 31 146 20 106 Lead 17 1 1 9 1( 22 115 33 129 Niam-eanlese Ore 15 8 2 4 89 117 100 303 > ill 4 5 2 130 245 134 607 , Tin 12 17 - 33 63 24 23 111 Zinic 15 14 12 1 1 17 84 43 99 Z lPlhospha[L Rock 1 1 na na 450 909 na na D Sil\ er 37 41 26 27 2 23 - 46 Tungsten 6 11 3 4 22 113 5(1 411 > (rude Oil 18 8 na na 103 87 13, ' 45c Iniduistrial Wood 3 5 - - 45 257 26c 39c *' Rc, :' aJ derpoir' that can he profiablx recoxered under presenlt techi nological and econiomic conditions. h"ResourceN" are identified mlineral deposits whichn may or may not be prol iiahlk recoxerable with existing technology and econotniic condition,. Reservcs, not resource,. na not a,;aiklble Source: World Bank 132 THE SO;FFHVST'ERN REVIEW Winter 1982 Rathler than assimilating through edLication the Spaniards and Portuguese vir- tually de,troyedl the local cult Lires. To this day the Indians have not been fully inte- grated into the modern cultural mold. In the Spani.-di-speaking countries the number of illiterates roughly coincides with that of mostly rural indigenous groups whose command of Spanish is imperfect or lackingi. D)espite enormous progress in educa- tion that number has remainied almost constant corresponiding to historical attempts at assimiiilation that were ruthless, sweeping but inmpe-rfect. While many Indians were aNinilated indi%iduallv through intermarriage, as a cultural group they were con- .inered, not assinilae d, and this remiain,s true today in the countries where large Indian groups live. The major cleavage between the poorer pop0uLlationl and the somewhat better off coinicides rather closely with ruralVurban differences. There are historical, cultural, andl geographical reasolls for this. The Spaniards developed cities first; they were their pride and glory, as can be seeni by any tourist visiting M\Iexico, Lima, Arequipa, QuitO or Cartaaeena. These cities were first and foremost administrative and religious centers. Usually they were financed out of mineral rather than agricultural surpluses. The earlier, pre-Colombian cities that they often sLucceeded, in contrast, had been .sustained nmainly through agriculture, as was the rule throughout Europe, North America, anid much of the de\ eloping world. Except in the temperate coun- tries of the Southern Cone-Chile, Argentina, Uruguay, and Southern Brazil- dexeloplnent of agtriculture came later, an inIVwrsion of the historical process ob- oer\ edl in most of the world. Thus, \%ltile in most of today's industrialized countries the rural and the urban sectors d\cxeloped pari pa.ssu, in nmuch of Latin America enclave mining finianrced the de\elopment of the cities without a concomitant development of agriculture. Therefore the benefits of growth largely by-passed the bulk of the population, which was rural. Today an even greater proportion of the native population lives in the rural areas, and this contributes to explaining rural neglect. Distances and peographlic obstacle. also contribtute to rural neglect, Two of Brazil's smallest states, Santa Catarina and Pernamnbuco, are each about the size of South Korea. Even Guyana, one of the smallest countries, is more than twice the size of Soutli Korea. Many of the poorest people live in remote, mountainous areas. These e-onomic, cultural, and geographlie factors not only tend to depress incomes; they also make it extremely hard to build effective administrative systems able to delixer reasonably good public education, health and other services in these areas. Naturally, the sharp ditference in conditions between rural and urban areas is the reality underlying the massive migration to the cities that is taking place everywhere in the Region, and the tendency may be reinforced with time. As more people cluster in the cities, it becomes cheaper to provide public services for them, and as fewer people remain in the rural areas, their political power becomes weaker. In spite of considerable progress achieved in improving rural social and economic public services, the gap between rural and urban standards of living remains enormous in 1nlo5t countries. Ihnnligrants, who later contributed to forming the Region's middle class provided an essential stiniLulus to d.veloprIo ent. To quote Herring: Thr..'iughOUt the; clonial period a few migrants drifted in from Northern tiliol e, legallv or illcually, and m1a1detI thieir lhomes in tthe New World. V'ol. 2, No. 1 i:(ONO(M)\tIC PERO)RNI \N( I. IN L.ATIN AMlERI(A 133 After indepenidenice, and particularlv after the 1850s, immigration in- creased. Spaniarid*, Poritguewe, and Italianis came in large nunmbers, as well as Frenich, English, Japanese, Germans, and Slavs. Over the length aind breadth of Latln Antierica., we find people who speak no work of the hilgLMUaC suggested bv their English, French, G1erman, or Dutch names. Writh few e\ceptioni these migraints becamiie . . . iationials of the lands of their adoption settled do%% n. anid intermarried (Herrinig, 1962, p. 16). Any tourist will corroborate this. The main street in Punta Arenas, Chile is Calle Jugoslavia. There are large numibers of Japanese in Brazil, and a sizeable Chinese commlll l itl X inl Peru. A series of papers was issued by the Pan Anmerican Union in 1950 on the middle cla£Sen in Latin A4merica (Crevenna, 1950-5 1). In Argentina immigrants made up 30 percent of the population in 1885 and 20 perceint in 1935. By then 45 percent of indit'.trial establi'rnlenlts were owned by foreign-born persons. In 1914, 49 percent of corninercial eniployees, 18 percent of public administration employees, and 45.5 penren of professionals in Argeintina were foreign-born. As late as 1942, 33 percent of employees in prixate cstablihimcnts in Buenos Aires were foreign-born (Germani, 1950). A recent book on population in Brazil cites the results of a survey of domestic nonforein2i industrial firims in the greater Sao Paulo area, the largest industrial area in Latin America. The 1962 sur%cv shows that appro\imnately one-half of the firms %xerc owlned and directed by first-genieratio n immigrants and if second generation ininmiirants are included the proportion rises to 73 percent ('.\ler;ick and Graham, 1979). Nineteerllh anid twentieth century European, Japanese and North American immigration not onlu provided a very important economic stimulus, but also helped create effect ie demanid for reaLsonably good public social institutions in education, health, and so ovn. The .1x\ es of middle-class immigrants during the late nineteenth century and the twentiellt century (notably those who came to Latin America because of the secoind World War and the attending political turmoil in Europe) played a decisi\ c role in consolidating the middle classes in Latin America. The mid- dle class had an importanit function in generating savings, investing in productive facilities both in induistry and in agriculture, and cons[titulted a new dynamic ec(tlolnic force in the traditional society. The remarkable acceleration in growth of the Region owes much to the coiitribution of immigrants. Le-'.onomic Growth Tihi ecti!on corniders the Region's growth record. It review the underlying dLelnourapllic treird, with Pncial stress on the rural7 urbani balance, and then turns to (utput growth, concluding vn iih a few remarks on social trends, inconmc distribution an1d enipl t, clltcn ialation Growth and Lrhanzgation After Ccinttries of' 10\ porplalion growth and labor s,horlagse the population of the Rk!i i ncrtc;i,.d very rapidly duiring the twentieth century. Around 1950 the p ptt hit ln hd rcair.d about 190 imillioll, r0itghlA thlazt of hle I'lnited States during the sorc eir;d. T-d;la. the i(MlIklat i0n is in the order of 365 miiillion-, and exceeds that otf thle Unitc:d.1 States by 65 ncrcent . Fi-iL' m te or B3razil illPt,t11 tt this dr!rtnatiL in- 134 IHE S;U1 tS\O 1PsI IRN RLV1-\W' Winter 1982 crease, The population of 3r-azil pa ,sed the 10t million mlark in the 1870s, reached 17 ;nifliion in 1900, 52 millioni in 1950 and 119 million in 1980. There are considerable \ ariationm anmong cotuntries xNith the lowest rates of population chanige in Uruguay, Areentina and the Enulish-spcakinq, Caribbean, but thlesc areas represent a -mall shalre of the Region's total populiI-ation. The rate of increase in populatioll has, hox\%eer, beiun to abate, In Brazil, the annuial rate of poppulation growthi increa-ed first frorm 2 percenit in the late nine- teenth cent UQ\, 2.3 percent dlur1ingL the first fifty vears of this century to a peak of 3 perceint durinig the 1950,. It the.. amie down xerv gradtially to 2.9 percent in the 1960s anid 2.5 percenit Lnrlingi the 1970s. Slighlt as the slowdo%. ni may appear, it has favorable implications, f-or social c.\penditnres today and for future employment nceds. Brazil's trenid is fairly typical for mlost of the Region. The total fertilitv rate for Latin Ameriica as a whole has fallen s.readlx since 1955-60 and has been estimated recenitly to be lower than rates in Africa and inost of Asia. Many canises uniderlie the decline in fertility rates. Rural. urban differeneCs in fertility are commonly found; MiLraltionT to the cities tenids to depress fertility. Furthermore, statistics suggest that fertilitv varies exen more withi education and income. In Brazil the fertility rates of .A omen1cl] with less thani three vears' schtooliru are rcarl\ twice those of women with six years' Jclo1im_ul and mnore (Nlerr ick and Nloruln, 1979). It seemus clear that social and ecOnl)nli-tI dcx elopl)mnn generallv and eduLcation andcl uirbainization particularly have coitriblnted powerfully to reduce fertility tlitc-. The tollo\\ ing figures indiczite the ratio of clilidrvii per uariHed x\%onian with 'occondarv vs. primary ediucatioln: Buenos Aires (1.7), Rio de .Janeiro (1.6), Bogota (1.2), San Jose (1.6) and Caracas (1.6). In &;Finrancianicnto de la diUcacion en America Latiraa" (Brodersohnl, 1978). These 1rond . can be expected to contini.u in niost1 ot the Regioni. Ncx ertheless, the popula- ion of Latin America cointinuies to increase at high rates: around 2.4 percent per \ ear. eFrtilihemnore, the ftill impact of tthe popuilation spurt of tthe 1960Os is yet to be fully felt oni the labor marl, ctt. 'ome p.. t~ .of Latin America l haxe reachled a stage Nhere the rural population is aleclim:ine s a result of out-miligrationi. This is an iImpOrtant turning point, because incrcai.niu sc(areity of rural labor is a precoinditioin for-and often a caLuse of-rising real x'. Lze- in the poorest-paid jobs u%chlm as rurdl lahorers and unsTkilled construction 'x'.r kc .. TI lilmilelnom n is relativelI recenit, anid b)ecaue few of the 1980 popuila- in censu,, rctilt, have ,ll r,epblished vet, thle LLant (ita;iv cx ideice is limiited. The &-aniplL of B3razil is klznificani. Between 197(0 an:d 1980 Brazil's population in- crctcd by 28 !rcent. 2 hile the urban population inckreCUaseLd by 55 per ent. the rural popull;ationi decline. l bN 6 pL'rc1nt. This. comripares to increases oft 63 pe-rcenti anid 8 percent reslpec ixclx in the 1960-t() Oecaiel. Absoluite rural potpul!aion declines were k,bs"er' .-l in :il the *.t ates of the 11)01C ePrIsPeIM1ou Soin ibast anld South. )eclines , .ere particularly s harp in Piarana (-29 pe[rcent) and inl tile StIaes of Sao Pa;ulo and Minas (jerais (- 19 percent . (Soa'e (if thie decHlne are am ihii able to niLration to ouit-of- 'I rtural area1s lal iLlar I> to lIta/zil \ v e2,terri frontier. All these figures are from ;! Ai` iIa.;,i ceinsusir results as;ed On a -naple. TFinal;l results may tuirni otut to be dif- fe. cut.) Buit eceni in the poorer Northeast tlle ruiral pt)pulalion of Crear;. Rio C(mande do Nonrc a and I salr iha dcLIi aed b-met x I1() 19a0 ltd)t), xx hi theat of thile Ntrtiheastt- rIm Rcui,' it: as a whole iiit iteasied by 6 pel-vo - doxIx frotmr nearly 10 pieccmt il tiring Iw 1 1,16; A. i m if tunrnin,_ poin)t is likely in \in Vene.'ucla, ( lAm 1i'Ia alnld i erl L.ps Vol. 2, No. I I( ONONCl' PERI ORMIANCE IN LATIN ANlMlRI(CA 135 some of the couintries in Central America. However, in most couintries including NMexico populatioii continlues to increase in the rural areas. The folloH iig quote is from the 1981 Atitninal Report of the World Bank: Latin America and the Caribbeain is the developing region with the lhighest level of urbanization. In 1950, less than half the population of the region lived in towIIs and cities, in 1980, the region's urban popula- tion of some 200 millionl made up almost 60 percenit of the total.. By the end of the century, there may be nearly 400 million urban dwellers, and the level of urbaniizatioii may rise to more than three quarters. In the 1950)- and 1960s, urban growth was for the most part a function of migration from rural areas, as national economic policies frequently dis- cotirm2ed the growth of atriCUlt ural otutput and employment and failed to provide adequ;ate social cer%ice|, to the rural population. In the 1970s, it was primarily a result of the natural increase of an already urbanized population1; no,cthcl&s the rate of urbanization in the 1970s showed little Ijiininlltioll trom earlier years, and it exceeded 3 percent in most countrie', and 4 percent in several. Within its general frar,. xork, there are important contrasts. The preenct level of urbanization, for example, exceeds 80 percent in Argen- tiina, Chile, and Uruguay, whereas it is still less than a third in such coun- tries as Bolivia, Haiti, and Trinidad and Tobago. The degree to which the urban poppulation is concentrated in the largest city also varies-from more than half in Co,st.a Rica, the Donminican Republic, Jamaica, Panama, and Urtuguay to less than 20 percent in Brazil. Some of these cities are now a1Mone the world's largc,l1; NMexico Citv, which has growni from a city of 3.6 million people in 1950 to one with an estimated popula- tion of 14 million in 1980, and Sao Paulo, whose estimated 1980 popula- tion of 12.6 million represemrt more than a threefold increase since 1950, are outs,t anding examples (pp. 62-63). Output Growth EconomIlic growth in the major countries of the Region can be viewed as having undelergionie three ph&lc,-the opc,ni trading phas.e between independence and the depresNioi of the late 1920s; the phase of acti%e import-substitution of the 1930s, 1940)( and 1950s; and the phase of gradual return to sonme\%hat more outward- orienlted policie\ in the 196(ts and 1970s. The exolution of the Region's economies is des,cribed oticci nctlv in Latin Atnerica-a Brde(lcr rforld Role Vasena and Pasos, 1973). During the open phase several major countries experienced rapid groc%th not only in auriciltuire. hut also in industry, spurred by massive foreigni investment. "In Brazil the r ncil\\ l et v orvk expanded Ifrom 3,400 to 21,300 kilometers between 1880 allnd 1910, anld beme en theni and the beginnim6ng of the First Wm Id WVar coffee exports miore t1han t:ii:iJiripled. In Argenllan thle P.Opu1lardll dolubled betweein 1890 and 1914, x%hilc cereal exports increased ti efold aTnd e'ort of refrigermted meat rose troiii 27-,00(1 imetric ion.- to 37.0()00 Iln'- (op. cit., p. 24). By 1899 the GiDP per -1pri l ofAryentima clLeedeLCl 50 percen I of that in the t iii ted StIl.es. "Indu-,triafi,ia- tion xas ul'.O t.b.iML, place in the inedin-rie.dl couintries of the regioii, such as 136 THE SOUTHWESTERN REVIEW Winter 1982 Colombia and Peru, even if to a lesser extent. In Chile, for example, the rate of miianufacturing expansion during the years 1914 is calculated to have run at an anrnual cumulative rate of sone 4.8 percent" (ibid, p. 65). The collapse of world trade caused by the (Great Depression had a devastating effect on the Region. GovernlMents adopted "a varied range of defensive policies aimed at maintaining employmnent le-el s . . . Governments increased public expen- diture, issuinlt money to do so, anid economies were closed to imports, through ex- change controls, quantitative restrictionis, or tariff increases . . . The process of "import-substitution" . . . was reinforced during the Second World War by the dif- ficulty of obtaining supplies from the industrialized countries" (ibid, p. 32). In the early 1960s "it began to be apparent that too much had been expected from policies of exaggerated protection. Industrial development was not continuing to accelerate as had been hoped, and the state of external trade was deteriorating (ibid, p. 33). The section on economic policies below discusses some of the pros and cons of reverting to a more "open" economic system. During the third phase of deLVelopnment growth of the Region as a whole has been reasonably rapid until the "oil shock' of 1974 as shown in Table 2. Comparable GDP growth figures for non-communist industrialized countries are 5.1 percent per year (1960-70) and 3.2 percent (1970-78). The GDP of the United States-to which about one-third of the Region's exports go-grew by 4.3 percent per year and 3.1 percent respectively. Cirowth in the Region thus exceeded that in the industrialized countries. Per capita growth rates also accelerated between the two decades but per capita growth in all middle income oil-exporting countries of the world (includinig Latin America and the Caribbean) exceeded that in the Region, especially during the 1960s (3.5 percent per year 1960-1970 and 3.2 percent per year 1970-1978). Naturally, there have been many differences between countries. The following tables focus on the eVoluItion of per capita income and on trends in crop production per capita in different Iatin American and Caribbean countries. Agriculture- essentially crops rather than livestock-remains a major source of employment in many countries oi' the Region. For these, agricultural progress may well be a sine qua tnon for long-term sustained economic and social development without major tensions. The growth performance of some of the Region's countries can usefully be compared with that in the United States. Table 3 shows the pLrcentage by xkhich per capita GDP chanipged during the last 20 years. Perhaps the most interesting findirig is not the very ide divergence in per calitia growth performance Table 2. (DP Gro,wth Rates Per Year 1960 )7)0 19-0 78 1970-74 1974-78 Region: C,DP Growth 5. 5.9 7.5 4.4 P'opllation Grkm iih 2.8 2.7 2.7 2.7 Per Capita GDP 2.9 3.1 4.7 1.6 Sourc(e: N'trld Bailk Vol. 2, No. I ECONOMIC PERFORNIANCE IN LATIN AMERICA 137 Table 3. Per Capita GDP Growth, 1960-79 (percent) Countries W'hich Countries Which Narrowed the Gap Percent Change Fell Behind Percent Change Relative to the U.S. 1960-1979a The U.S. Trend 1960-1979a Barbados 135 Bolivia 50 Brazil 122 Argentina 48 IBelize 94 Grenada 42 Surinam 94 Chile 41 Paraguay 79 El Salvador 39 Costa Rica 75 Ecuador 39 Dominican Republic 73 Peru 32 Panama 70 Uruguay 24 Colombia 70 Honduras 23 Guatemala 67 Jamaica 10 Trinidad and Tobago 66 Nicaragua 7 Mexico 64 Bahamas 6 Venezuela 63 Haiti 5 Guyana I United States (1979-80 average) 62 Netherlands Antilles -2 aAt 1977-79 prices and exchange rates, Source: World Bank Atlas between each country as much as the fact that 73.5 percent of the Region's popula- tion live in countries that have done as well or better than the United States in improving their standard of living since 1960. Brazil's performance stands out as a most remarkable achievement; per capita income there increased at nearly twice the United States rate over the last 20 years. Since Brazil's population represents over one-third of the Region's, its performance greatly influences the regional total. Some of the factors underlying the divergent performances are dealt with later. National income information based on international price comparisons has recently been developed and can be used to give an idea of the relationship between the average standard of living in the Region and in the United States. Table 4 shows these relationships for 1960 and 1979. A sutbstantial proportion of the population still depends on agriculture for its liv- ing. Table 5 shows changes in crop production per capita (i.e. per head of the popu- lation-not per person employed in agriculture) during the 1970s. While the countries which have experienced positive per capita crop production trends during the 1970s include nearly 60 percent of the Region's population, this owes much to the weight of Brazil. In 15 out of 24 countries crop production per capita stagnated or declined. Performance was particularly poor in Peru, Barbados, Trinidad, Jamaica, Nicaragua, Mexico, Guyana and Haiti. The unweighted average crop output per capita of the Region's countries declined by one percent during the period. This performance is in contrast to the good overall growth record of the Region and is contrasted in Table 6 to the overall growth record of the United States and regional groups. 138 THE S0UT'H%VL-S ELRN REVIEW Winter 1982 Table 4. Per Capita GDP Expressed as Percent of U.S. Per Capita GI)P Interniational Ilrice Comparisonis Nlemorandtum 1960 1979 Change Item: 19791 Argentina 32 31 -I 21 Barbados 25 39 14 22 Bolivia 16 15 -1 5 Brazil 20 28 8 15 Colombia 22 23 1 9 Costa Rica 24 26 2 17 I)ominican Republic 18 20 2 9 Eeuador 17 18 1 10 El Salvador 15 13 -2 6 Guatemala 17 18 1 9 (iuyana 25 15 -10 5 Honduras 16 13 -3 5 Jamaica 25 17 -8 1 1 Me.xico 29 30 1 15 Panama 22 25 3 12 Paraguay 18 20 2 10 Peru 22 19 -3 7 Trinidad and Tobago 50 48 -2 31 Urutuav 53 41 -12 19 Vcne,'iiela 44 42 -2 29 Chile na na na 16 Haiti na na na 2 NiCarIaLLia na na na 6 'Wordl Bank Atlas method (current cwchallgc rate consersion based on 1977-79 averages) rather than interrnationial price ;nrpdri(;on1. na not available Source- \\ Lrld Bank N1(ost of Latin American industry has grown rapidly durinig the last two decacdes. The most rapid uross tli rates were achieved in Central America during the 1960s and earlx 1970s when the Central American Conmnmon MlarkLet was making its effects felt, as well as in Brazil and in a few countries at the earliest stages of industrialization (H,aiti during the 1970s, the Donminiican RepLublic. Ecuador and Paraguay). The more "open" countries have expanded iniport-sLuhstihLttion as well as exports fairly rationallv wlhile the "closed' and particularly the smaller countries drew their dvnamism mostly frorm, (frequently uneconomic) import-substitution (see the section oni economlic policies belos). A number of countries have found it desirable to reduce protection graduall in order to sustaini the growth nmomentum into the future. Some of the countries which did not follow that path hase experienced declinina industrial growlh rates (Professor Balassa concludes that exports and effi- cient import-substitution often go together [WN'orld Bank, 1981, notably p. 21]). A physical indicator of the Region's growth is the onutput Of alutonlsvi% e velhicles after imany decades of supply restriction, Of vNalr1i,us kinds. In 1970 the Region pro- duced 950,000 vehicles; in 1979, 2.1 million, of which 1.1 million were produced in Bra1il, 450,0(10 in Mexico, 250,000 in Arcrnlina and the rernmindcr in \eneiuela, Vol.2, No. I LCO)N)N'lIC PFRFORNIANCE IN LATlN ANIERICA 139 Table 5. Percent Change since 1969-71: Per Capita Crop Production (1978-80) P a ra u als v45 Argertinia 39 Colombia 28 Guatemala 23 Siirinarn 1 5 Brazilb 14 Venezuela 9 Dominican Republic 2 FLIaldon 0 Bolivia -I Uruguay -3 Chilec -3 Costa Rica -3 Honduras -4 Panama -7 ElS Salk ador -9 Haiti -11 Iu VI I- 1 2 Mexico -13 Nicarauwa -18 Jamaica -19 Trinidad and Tobago -30 Barbadwo -31 Peru -31 d(Crop trends are presented over a decade rather than two because of kcihglitinV problems with the longer- term series. Apriuldr:nl statistics are particularly deficient in most of thle Region. Livestock is excluded because of its relatively small contribuitioni to .uilos il . blihese figures based on 1970 *eights fail to captuire fully Brail' s SpeciaLilar perftormance in soy expan- sion during the late 1970s. A rate based on 1980 weights woould showk substantially better pc,rforrnancee during the 197(s. iThe %%cilht problem also dcpresses Chile's rate because prodtiuction shifted to high-value crops during hc dccadv. Thits is n ot id cLiti Li I Cflel tdL inI thle able. T1 able 6. T he N%eiglited .A eragpe for Food and lotal Agricultural Ouipul (;ro [ii Per Capita for the l'niied Smaiie, and Various Regional Groups Plereenit Changes (195 8(1 o%cr 1969 71t) per Capita 'rotal Aericidr tiral I.ood Pouto uIi tcLLi States 13 12 Ilndtist riali/edoLt nlit ries 1() 9 T)De'loping cotuntries 2 I l atin America and Cm ibbean 9 7 East Asia (excl. .trantl 17 15 South iAsia I - I West Asia 12 10 Africa fc.';l. South Af'rica) l13 -14 Soturce: U .S. D)epartmcnt ofA Aricill im u 1981. 140 THE SOUTHWESTERN REVIEW Winter 1982 Colonibia, Chile and Peru. Mexico cxported 25,000 vehicles in 1979 (besides export- ing components), and Brazil's exports reached 106,000 vehicles. Brazil's vehicle ex- ports increased from 2.2 percent of output in 1972 to 13.8 percent in 1980. The modernization and expansion of the services sector in the more advanced countries of the Region deserve mention. As will be noted, labor absorption in ser- vices can be (and has been in the Region over the past 20 years) associated with in- creases in producti%ity. This sector descrr\es much more research on productivity trends, policies and prospects. The Region's share of world trade declined fronm about 8 percent in 1960 to 5 percent in 1979. Table 7 shows the Region's export performance compared to the trend in world trade and that for industrialized countries. While the Region's share of world trade declined, performance during the 1970s improved considerably over that in the 1960s, and manufactured goods (SITC 5 through 9 minus 68) became significant, their share in regional exports rose from 3.3 percent in 1960 to 17 percent in 1978. On the whole, export performance has not been outstanding by world standards but this has not kept the Region's income from growinig at rapid rates. Social Developments Economic growth has been accompanied by social improvements. Between 1960 and the late 1970s life expectancy at birth in the Region increased from 55.7 to 63.7 years compared to 69 years in European middle-income countries; access to potable water increased from under 40 percent to 66 perceni; the population per physician declined from 2,400 to 1,760; access to electricity increased from 46 to over 60 per- cent. The share of the labor force employed in agriculture declined from one-half in 1960 to about one-third in 1980. Finally women in the labor force increased from 18.6 percent in 1960 to about 23 percent in the late 1970s. As the population census results for 1980 become available many of these figures can be improved. But there is no doubt that the share of the population deprived of safe water, medical atten- tion, access to electricity and other essentials declined ssharply during the past two decades in spite of very rapid population growth. Urbanization, which facilitates access to basic service., has been a major cause of improvement. A very importanit apect of Latin Americain development is the educational effort of the past few decades. At the turn of the century, as noted earlier, the over- %lhelniing ma1-ijirity of the population of Latin America was illiterate. Wealthy people frequently sent tlheir childreni onersea, for Celducation, and public and pri%ate schoolS, catered to a small minoritv. By 1950 the primary school pLpulation had risen tc, a little uinder half- of all children aced 7 to 14. By the late l970)^ most children atteldedl priniar\ s,chool, andl Necondar\ schiool curollment lhad reached about 35 percenit in spite of a very rapid I)OpuLk;tion growth. The increase in unik ersity enroll- imierit is trulv dramatic. Between 1960 and 1975 the niumber of students increased by over 300 pe rcent in Central Amnerica a-ld the Caribbean (incluJinig Mexico) and by over 600 percenit in South Aimierica (UNESCO Statistial Yearbooks). While the uirban aireas are relad-i,c]\ well covered by educational institutions, rural areas con- tinue to be weeicted. 'vith the axerte rural chiilu iN1c ndin- less than three years at .thtl ia iVSt Citriet;. lui theL ilrnore, educatinail pHrorsc lias, naturally, Vol. 2, No. 1 ECONOMIC PERFORMANCE IN LATIN AMERICA 141 Table 7. Regional Export Trends itn Perspective (percent share of world exports) Items (with 1978 share of total regional exports) 1955 1960 1970 1979 All Exports (100) LAC Region 9.8 7.7 5.4 5.0 Other LDCs 15,0 13.1 11.5 19.0 Centrally Planned 11.9 14.1 13.5 11.8 Industrialized -63.3 6.5.1-1 69.6- 64.2- Total 100.0 100.0 100.0 100.0 Food Items (38) LAG Region 20.7 17.3 15.6 15.6 Other LDCs 22.3 19.6 15.8 14.9 Centrally Planned 7.7 11.1 9.5 5.7 Industrialized 49.3 52.0 59.1 63.8 Total 100.0 100.0 100.0 100.0 Fuels & Related Minerals (31) LAG Region 26.6 23.9 13.6 8.4a Other LDCs 29.0 32.7 45.5 57.8 Centrally Planned 13.6 18.7 16.1 17.2 Industrialized 30.8 24.7 24.8 16.6 Total 100.0 100.0 100.0 100.0 Non-Ferrous Metals (4) LAC Region 14.2 11.4 10.8 8.5 Other LDCs 19.4 17.2 16.9 12.3 Centrally Planned 8.3 10.6 11.7 12.0 Industrialized 58.1 60.8 60.6 67.2 Total 100.0 100.0 100.0 100.0 Manufactured Goods (7) (only SITC 6 + 8 minus 67 and 68) LAC Region 0.9 0.6 1.6 1.9 Other LDCs 7.7 7.9 9.2 14.2 CentrallvyPlanned 10.9 11.2 8.6 6.7 Industrialized 80.5 80.3 80.6 77.2 Total 100.0 100.0 100.0 100.0 Machinery and Transportation Equipmncii (5) LAG Region 0.1 0.1 0.4 0.9 Other LDCs 0.6 0.6 1.1 3.2 Centrally Planned 15.6 16.9 13.4 12.2 Ind u,trialived 83.7 82.4 85.1 83.7 Tlotal 100.0 100.0 100.0 100.0 'Mtexico's petroleum exports increases are likely to have raised the Region's share after 1978. Note: LAC refers to Latin America and the ('3rihbcan. LDC refers to Less Developed Countries. SITC means Standard International rrade C la..,.'ticanr io. Source: United Nations Confterence on Trade and Development, 1955, 1960, 1970, 1979. HandbooksV of Initerniationial Trade and Development Statistics 142 TIHE SOUTHW'E:*WSTERN REVIEW Winter 1982 benief'ited the newer generation while illiteracy a-monmg older people corntinues to be hiLrh in many countries. In Brazil, for example, 40 percent of the agricultural labor force had nio formial slcooling in 1973; 1 I percent of those employed outside agricul- ture were in that situiation (IBGF, 1973). Nc ertlhless, progress in Lhis area ha.& been very impressive, and human capital ormnationi on a vast scale has been one of the main fad ors uinderlyinig economiiic noderni/atioln. Nlticli has been1 written about the d%elopineiil of a middle class in Latin America. One indicator of this phieomeniilon is passeinger car ownelrhip. It can safely be a',wnied that miainily imiiddle and upper class families own passenger cars. Therefore the level and trenid in the shar-e of families ox% niniLg cars gives a rough idea of the iimiportanice of the miiddle cla>s. Clearly this is a very crtide measure: it ignores the % at' in ilmportance of multiple car o%x nership and of rented cars (which may dis- tort parliculariv the statistics of ,oine of the Caribbean tourist islands); it also ir2niores differences in family size betmcen counitries. Nevertlhles.s, for wlhat the fiulres are worth, thev are shownv in Table 8. For the Region as a whole, the "''niddle anid upper classes" so definied liae grown fromLi uLnder 6 perceint in 1960 to over 20 percent in the late 1970s (the exact years of the latter observanices vary by conintries). TIhe total numil ber of cars in the Revioii increased from 2.3 million in 1960 to over 14 millioin. The mos(t91 CpectcLular inlcrease occurrdl in Brazil from 500,000 cars in 1960 to 6.2 miillion in the late 1 970s. Ihe theme of poverty and ince11C distribtution in Latin America has been much dLebaLted (Fischlow, 1972; Webb, 1974; Ber's!man, 1980). While tlhere is little doubt that thle distrihut ion of income in miost Latin Americani coLuntries is verv skewed- allt hotlh perhap tO io more so thain in otther dex%Ceopillg countries-considerable debate has sirr'wumLiideJ the questioI ol' trenids: lhas the distributioni of income im- proved or worsened over t i n me? I he eniipimical :aois for statemints aboui in.come distribution trends is so weak even in the most aLx inwd. cmoutiries in the Region that comnparisons in time lack statitlical \ lidit\. So, for \ainiple, the B3razilian poptulation celnsses of 1960 and 197(0 captured only 58 andcl 56 percnt. repect ivelv, of personal inicome as estinmated in thle nlalionll accounts. Nat iomial household surveys of Birazil also failed to captulre a ,ili'startial share of personmal inicoIIIe (see ''The I)istribution of Income in Brazil," W Odrld 13ank Statl \'Wo kin Paper No. 356, 1979, by (iuy Pfeffermann and Richard Wlth, pp. 15 and t'f.). Co1mpa)l1ra.ble data in other coutntries of the Region are w,eaker even thiani in B3razil. \N hcAt cani be said oin the basis of recenl ainalyses is that in at least three major co01t lies ( Hr,i i, Ie\%ico and (Colomibdia) econotIIic growth has been accomiupiaiiied by sibV.anitiial i:nprox.ciev for the poorer segmiient-' ol' the popiulation. In Brazil, ''The twvo imiaini 1½t uie e of thie changl in en1 plo%nirtuti patterns bemxx een 1960 and 1976 are a suibst lul move otit of r..ricuhltrc, aniid an umpipradlim! of the tirban ralJl1ineiu .ric nrc (Pt'effermiainn anid NWebb, 1979). By iiio% inig otit of agricul- tuire lanldless labitrers cranu in.reasc their incomles. "In all regions (of Brazil) the land- les: Il irr 'A "imle. hiis income byN noinc to turban iiianuial employment withlini his xx n n c m t . A1lo ;ic11 or) r urbaIl-ruiral os of living di t't'ercices v oild still leave in- re,eCs,,of x'- :,, oxc- cv St 1, . - The ru a!cl-Nototheast to uirban-Rio move roUghly '.l' ;;t' , (1 h - i:.o lx ii\ ioincom of a Sao Paulo maimnual xx or.cr is 4.7 timies tha!t 0 a `i 'tc-aitst f:il 1t Lil'110t1rc;' jP1`ItCtL1l} r I marand 1979). At the amine tinlc t, ' 't'rem;l e2tx'l --.:.10 11ihilral aLi116 Fl01Tmu-ariumltrall mean i in(omes in Brazil Vol. 2, No. 1 ECONOMIC PERFORMANCE IN LATIN AMERICA 143 Table 8. Percent of Families Owning Carsa 1960 Late 1970s Bahamas 35.5 86.5 Trinidad and Tobago 22.0 52.9 Barbados 17.5 50.4 Argentina 12.0 40.5 Venezuela 18.5 37.8 Brazil 3.5 26.8 Jamaica 1.0 22.5 Uruguay 19.5 22.3 Mexico 7.0 21.2 Panama 7.5 19.9 Guyana 9.5 17.5 Costa Rica 7.0 16.1 Chile 3.5 14.1 Peru 4.0 9.3 Colombia 3.5 9.1 Dominican Republic 2.0 8.0 Nicaragua 2.5 8.0 Guatemala 3.5 6.6 El Salvador 4.0 5.5 Ecuador 1.0 3.7 Paraguiav 1.0 3.5 Bolivia 1.5 3.4 Honduras 1.5 3.2 Haiti 1.0 1.8 TOTAL 5.8 21.6 aAssumes average fanmily ot 5. Source: W'orld Bank data has remained fairly constant since 1960, suggesting that the enormous labor absorp- tion by urban areas occurred without a flooding of the lower income categories. Per- haps most important, real wage rates for casual rural farm laborers in Brazil went up by about 60 percent between 1970 and 1977 and have remained at that real level since then. Thus, an important group of workers who are among the poorest in the coun- try experienced a substantial absolute improvement in their daily earnings. The experience in Colombia is very similar. A recent study (Montoya, 1981) con- cludes after an exhaustive review of evidence focusing largely on wage trends. The empirical analysis of income trends in Colombia during the last two decades is a difficult task, but it appears that absolute poverty was re- duced, since the incomes of poor families rose during the 1970s at a rate matclhing or exceeding that of per capita income . . . There were in- crease.> in real agricultural day wages in the 1970s, especially during the latter half of the decade, There were also increases in the real wages of oinsirslrtion Nworklers and other unskilled urban workers." The study nlotes that middlc-class incomes rose more slowlv. "Because the incomes of the N\calhhiest families also rose more rapidly than national per capita income, it is clear that the relative position of the so-called middle-class \N%orsened. In the years prior to 1964, in conitrast, that class benefited most from the economnic development process. 144 THE SOUTHWESTERN REVIEW Winter 1982 Finially, in the case of Mexico a recent study (Gregory, 1981) concludes: The rapid and sustained growtth of the economy has been accompanied by a very substantial shift of the labor force from the agricultural sector, where productivity is relatively low, to the secondary and tertiary sectors, where productivities are much higher. Furthermiore, this large shift has been achieved without depressing the productivity or wages of labor in the expallding sectors. Our analysis of the course of remunerations and productivities in those parts of the nonagricultural sectors that are likely to reflect the uninhibited forces of supply and demand in the labor market, recorded substantial improvements in both of these measures. Finally, I have set forth reasons for believing that the degree of under- utilization of labor resources may be considerably smaller than most estinmates have held. The study covers the period 1940 to 1980, with particular focus on the last decade. Esen fewher reliable statemenits can be made about poverty trends in the Region as a whole. However, a recent paper bv Peter Gregory (1980) which covers eighteen COLu]itries over at least one decade substantiates the following points: (1) agricultural employment lhas become less and less important over time; (2) industry has gained relative to total employment; (3) productivity has risen in the services sector, sug- estiing that the rapid increase in employment is not due solely to the expansion of marginal service acti% it ies. Gregory concludes: In none of the sets of data examined ", which could be presumed to reflect the direction of qualitative changes in labor market conditions . . . have I been able to discern unambiguous evidence of deteri- oration." Gregory's findings are conisLtent with micro studies on Bogota, Lima, Rio de Janeiro, Santiago and other metropolitan centers, W. all suggest that migrants do move up the scale; that their employment and earnings situations do not differ much from those of comnparable city d ellers; and that migration is, indeed, ihe main mechanism through which rapid economic growth is translated into life- time aind inter-generational upward mobility. Economic 7olicies Having re%iev.ed the elceincts of growth and performance of the Region, this -;ection turns to economic polpicies aind relates them to success or failure in achieving sus,tained growvth. ThIe sectioin deals with agriculture, financial policies-exchange alnd interest rate milanagement in inflationary conditions-and external indebted- ness. The seLtionl Thows- how donmstic p)olikics have beeni largely responsible for eCommome 5per!fornman,e. InWlee,wl lanazd J Z<'(romw l,' vje7';.Otjnielt7 Smi;all psx'chological a in po;id-al ifeartre,s help e\plain the nmany inlstaelle of p'icie . 1eemii-vl v at oddl. with '0ln.1)1iCi ratiunalitV. 4- s in most deo eloping regions ianixN \.:t e of an arlie,,r muinatit:i: ha..e surviNCil in I atin America. Thus, pr rrons clOS. to tihe L!\1 rim;nenzt mrliiii he aiven ectl'i\ e Franchise to imiport a partictlar br;-tid of .:t' -. Others biL: he k Len di{1 inmort ban on l'. hater their firms pro- diuc'..6 l'i; llili iigh1t bt., placedl at the lhelm 1t1 bank*s ad it- ol!. * hile tlhese ;>Iiz, t.. .. ei ^1,1 Ll; theQ -.A Il li j1!d 110nt on11y' the Je' cotfi ut rl di. Vol. 2, No. I Et C)NUMI( PERF(0)R.M.ANCIl IN LATIN AMERICA 145 the "closed ecoinomiiies" promided a particularly fertile ground for their prolifera- tion. And efforts at dismantling barriers sometiines failed because of the highly per- sonal nature of the political arrangements underlying them. The more open econ- omies afford far fewer opportunities for this sort of "neo-feudal" nexus. These '"neo-tfeudald" relations continue to exist in many parts of the Region and form an obstacle to sustained econonmic proeress. Seconid, with the rise of the middle classes during the past decades economic gi o%\ th1 has becomie increasingly important in the political arena. While thirty years ago it was still possible-in some countries common-for a regime to remaii. in po%. cr for a long time while the economy was doing badly, this has become more dif- ficult and rarer today (the eight-month presidency of General Viola in Argentina (1981) is a case in point). Middle classes expect governments "to deliver." Indeed, rapid growth has been used increas,ingly as a legitimizing force by non-democratic governments. The corollary is that governments "take economics more seriously" than they used to, and this in turn has favored economic progress. Ironically, the cl ire to remain in po-wer has thus led to undermining the "neo-feudal" arrange- menlts when they stood in the wav of "deliverinig the goods." At times the neo-feudal" and the "imeeting of expectations" combine on a large s;cale, when gomernrTeI1ts try to co-opt a whole segnment of society-usually among the middle class--bv delivering economic (and more often social) services. Generally they will try to do this without financing them through taxes. This has come to be knowni as ppilimni. It may coiisist of attempts to deliver increases in consumption exceeding the productive sectors' ability to respond, usually through wage increases, price control,, or improvement and expansion in social security systems. It often . n>ist'. as well in the expansion of public sector employment beyond reasonable limits. All these policies have in common the fact that they put a burden on the pro- ductti%e -ctom-s-pa rticularly agrictilture-which depresses incomes and output therebv creating increa.ing stresses. One particularly common burden is the taxation of agricultural exports either directly or through overvalued exchange rates. These policies are unsustainable in the long run because the erosion of the productive base eventuallv undermines the very delivery of "additional welfare." Perhaps Latin Amuierica is more pironie to this sort of policy than other regions because the spirit behind "milking the cow to death" is not so diflerent from the drive that fired early W0lllists Who sOLiliI', to maximize mineral extractioni. Unsnurprisin!tlv, frequent episodes of excess demand ratified by governments have mleant, e.pedall in South America, frequent and often prolonged periods of highl inflat ion. Inflationi has inadeed been "rationalized" on thle groutnd.sl that it in- crease2d investible *:.h ings a% ailable to pot ernumleits, without explicit additional taxa- tioii. But thte scope for such "intlation tax" is rather small and the public's defense jncchltisz..tl -n.iahiv aln l d%voidance of nione% in favor of substitut.is-cause harnm whichi far ouli w cich the limited (lnenefits of the 'In flartion tax." Economl0ists of the -Wi h Aool'' e ti.\. aliid (r.inii k aldti in Ilirschiimali. 1981') t, without advocat- ,iu iiflanoito, '-'ic' e, thad Qit inft'latioTn was., Limavoidablc becau'.e the suipply of food, :1 eu 2.ch:nc nuJ 0!her . :.di i pCT11,e ' .al'L rCqiisi Lt for de'dCCiO\nl'lt ep )oUld not meet ,hC o-- i) ni.i 's,:1; : thle P0pl]LI i iii. i .I, ef rr. their aneiunent went ;soniewhat dIi .icl + t. '..'iul" ii~ f tI' thr'. nghdlt 2n 'LI cu1rbs was futile anid N %,uld c'aus,e oilly t:j;i E. * ' Iv l71 ii '11 and Iuicin-lo1j il . Ma. ie is said belot iibowt Ii i e Xil : witlh in- tt.!' 1 }1. In ; 1 'it to taitx lhere eat en i i. A c t id l F i c for 36 dil t c!klpiw_- ,. %A11t I1 tie 146 THE SOUTHWESTERN REVIEW Winter 1982 (includinig 18 in the Region) by Harberger and Edwards (1980) over a period of 15 to 20 years strongly suggesth the existence of "a tendency toward a negative relation- ship between the rate of real growth and the rate of inflation" (pp. 35-36). Indeed the study comes to the surprising conclusion that only few countries suffered output losses (slowed down GDP growth) during periods when inflation was coming down, and that in most cases a reduction in inflation was indeed associated with accelerated growth. It is regrettable that this empirical evidence only surfaced more than a decade after the argument between "monetarists" and "structuralists." This review of the intellectual and political environment touches on a policy debate that continues to divide the Region: "openness" versus intervention. Open- ness in this context refers to the role of the price mechanism in an economy. Advo- cates of closed economies see the market mechanism as prejudicial to welfare. This view rose naturally from the catastrophic effects on welfare of the drop in primary commodity prices of the 1930s. By contrast, economies like Japan, Korea, Taiwan, Hong Kong and Singapore never have been affected significantly by fluctuations in export prices. Following the early lead of the United Nations Economic commissiorl for Latin America many Latin American countries have attempted to shield them- selves from the influence of external markctls t;eT,!g import-substitution, often at a very high cost to consumers and to the eco nony. Especially in small economies this has fa%ored the development of domestic monopolies producing at high costs. In "closed economies" inflation is often suppressed through government price con- trols or sales by government enterprises at artificially low prices which generate fiscal deficits and in turn fuel inflation. Prices of food and other necessities in par- ticular tend to be controlled at low levels to "protect urban consumers." As a result, scarcities and some form of rationing usually ensue. Farmers are not encouraged to produce, and food has to be imported. Likewise, financial markets are frequently repressed in an attempt to keep interest rates artificially low, often below the rate of inflation. The public is not encouraged to put their savings in the banking system. The available "cheap" credit is rationed to a small number of privileged recipients. These price distortions indeed contributed to bring about the very supply rigidities which the "structuralists" considered systenmic obstacles to development. Con- versely, in "open economies" there is less scope for price interventions. Even thought the record of open economies has been better world-wide than that of the c,oheed eocnonmies both in terms of growth and in flexibiliky to external shocks (e.g. the oil price increases of the 1970s) many of the Region's academic economists con- tinue to support "Closed economy" policies. Development Policies A- ricullure. Table 9 sLIggesl, that good agricultural performance may be a prere- quisite to rapid per capita income growth. The mediocre growth performance of Argentina coupled with the second-best growth in crop output per capita in the Region largely reflects the response to sharply improved internal producer prices for auricLIlture after 1976 and the adjust- ment problenii of an inefficient inuciwtrial sector. Out of the five countries which conmbined uood GDCP per capita growth with poor performance in crops, three are oil or gab counitriec (Ecuador, Bolivia and Trinidad and TobaPo); Barbados had to di%ersif; a%NaN from auriculture ])eCaLPe. of an absolute Thortage of arable land; and CoNsta Rica's growth was based largely on the expansion of public services which Vol.2, No. I ECONC)NMIC PERFORMANCE IN LATIN AMERICA 147 Table 9. Crop Outptit and Income per Capita Trends (1970S)a Positive Trend in Negative Trend or per C(apita Crop Output Stagnation in Crop Output Above-average Paraguay Ecuador growth of per Colombia Bolivia capita GDP Gluatemala Costa Rica Surinam Trinidad and Tobago Brazil Barbados Venezuela D)ominican Republic Below-a.erage Argentina Uruguay growth of per Chile capita GDP Honduras Panama El Salvador Haiti Guyania Mexico Nicaragua Jamaica Peru aThe fact that most countries are in the Nortti'.\eC and Southeast segments is not due to the large sihare or crops in ( DIl. That share is too small to cauise "auto correlation.' Source: Wsorld Bank statistics proved unsustainable after the end of the 1970s. Of the countries on the bottrnm right-hand quarter of the table, several have slhifted into the fast-growing category toward the end of the decade: Nle\ico (owing to its oil and gas), Chile and Uruguay (owing to suceCsful restructuring of the incentives environment), Because of the continued inportance of the rural population in Mexico, continued overall growth witlhout rapid agricultural prouress may spell increased social tensions in that coun- try. In Uruguay and Chile agr-iculttire represents a much smaller share of employ- ment than in Mexico. Table 9 may point to a need to improve agricultural per- formance in Honduras, Panama, El Salvador, Haiti, Guyana, Nicaragua, Jamaica and Peru lest overall growth remaini mediocre during the 1980s. The agricultural policies of Brazil and Peru during the 1910s can be contrasted in sy yoptic form. Peru's land resources a,e far poorer than those of Brazil, but this is not enough to account for the .sustained decline in agricultural output per capita dur- ing the 1970s. Table 10 suggests that production-oriented policies have contributed to Brazil's impressive agricultural growth. The question arises why Peru and other countries followed policies which led to agricultural stagnationi. The rationale for the constellationi of policies that retarded agricultural development in Peru and several other countries in the Region was a combination of self-defeating beliefs: that farmers do not respond to price incentikes but rather that agricultural produc- tioin is guided mainly by structural factors such as the pattern of land o%% nership and that therefore low agricultUral prices benefit the nopulation insofar as they raise the urban li'ing standard -without diLouraging the supply of food-this often goes tiethetier with a belief that the most acute poverty probllen is found in uirban slums ratlher ihan in the rural area.; that %idesprcad and substantial industrial protection 148 THE SOUTHWESTERN REVIEW Winter 1982 Table 10. Agricultural Policies of Brazil and Peru Policies Peru Brazil PRICES Generally held much below Price controls eased world level considerably after 1964 EXTENSION SERVICES Virtually discontinued Well-established during 1970s expanding service AGRARIAN REFORMa Caused great uncertainty, Virtually none diverted government support services EXCHANGE RATE :iid Heavily biased against Flexible exchange rate PROTECTION agriculture especially stimulating export crops during fixed exchange and strong bias of export rate period of early 1970s promotion policies and subsequent lagging favoring processing rate period industries aThe examples of Japan and South Korea suggest that agrarian reform can contribute to output growth. Until now, however, the experience of Latin America has been that for a host of administrative and political reasons and because of the farmers' very poor educational level, land reforms have not con- tributed to agricultural growth. stimulates industrialization; and, finally, that state intervention in property relations can occur without reducing investment. Most of these attitudes are symptoms of an opposition to an outward-oriented, open economy that can be found, world-wide, much more frequently among countries that failed to grow rapidly than among the better performers (Balassa, 1981; Behrman, 1968). Policies with negative effects on growth and often distribution-as is the case of cheap credit-were invariably pur- sued in the belief that they would generate more growth, more employment, and a better income distribution than the open economy. Living with Inflation. As noted, inflation has plagued South America for generations-and more recently also Central America and some Caribbean coun- tries as well. The growth of food production and exports over the past decade has been rapid in several countries which suffered high inflation rates, This casts doubt on the structuralist proposition that inflation is due largely to the rigidity of aglicul- tural output and of exports. Rather the sources of inflation must be sought in ex- cessively expansive monetary and fiscal policies-including widespread subsidies- usually meant to mask more explicit conflicts between social groups over limited real resources. Whatever the political and sociological roots of inflationary policies the paper by Harberger and Edwards (1980) leaves little doubt (on the strength of obser- vatiorns in 33 developing countries of which 14 are in the Region) that the rate of credit expansion, in particular, credit to the public sector, is strongly associated with the rate of inflatlion. This is not to sa that external events do not cause infla- tion-tlhe increase in the price of imported petrolcum undoubledly exacerbated inflationi in 1974 and 1979 in several of the region's counitries (Galbis, 1981). But domnestic policies do seem to play the decisive role. For an illuminating discussion of Vol. 2, N(i. I IC('NOMII( P) KI OlRl *\N I; IN I Al IN A%MJRIK( A 149 the jiolitic, anid so)cioIotc\ of inflation see Ih liichian;, 1981; while I Iirschniian cOWi- 'siiders a inn] iit tidle of, fwcts of, the in flilat Ii issue, oneC saH ice SUMS uip succinict l thie thrust of' the u-i iii immen mointed here: "tlie state hands over to inflhation the diL- ;11.' c iceIIIC jo of) o a INi%' nI, Io." 'Ihei issuie III this sect iml is nlot hlow iniimtitio can be best avoided but rather hiow WelII ditTeren ClIIt'tI IIIIICi I iiits, hIave II laIIiil'ued to 'live with inlatbion." Th'is description touchles onlyi onl moie of thie effects of jii1flation on theC iiNiMiICHI'iun of' theC balan,11c of, pavylents. J lie issuie of, li% idw withi iitlat ioi i is oIf cou r se illii IIch h,ICrde anid has spaM%%Ined a debate oni iiidIc\ation11 1I ncAt ion1 iS theC 111i 1I oitic (and usullydI leIslahted increase of cntmcIvin](111 pviliieiitoh Ohutionlikls SuICh as vaesoi- debt s'i i'cc ill1 linie with anl indexC of irI'itialoi, uiM 11iull a pilc a]ICI pr'ice i!Ilde\. PILuVInciiIs dueI dtil'iiii'. Onle Aicklr LIixe is to) eflini k int 'o at least lese ie des effects of' iii tlaitioii onl [mu uacial I nohl Ki it i esourlce allocation, andiL the dist ibuimitol of' i nconie. I at in American count'C1 ne par'tWILIarlyH u,il, C hile, Argent ina, and( U ruguav) have espcmi iien ted with inidexationl for long, peroiMIN* FIIese exper iCnCC- suggeJ~(St that iliidc\at ion, Whlile p s~hvunavoidable onice iniiltimtio i czches hiigh hex els f'or lonig pImiods, is itmperf'ect and L'I ea,te, ')I 0ihiCiiis Of its own. IPz tiil indexation mlay leadL to svr d ist t ionsI f'or CVUnipleC thle pi ix ilCf'd indexL'inl', ofl sax in1's anld hoaiis ONblIgutioii has led in sonic coil ii itic to) exces-xe t'lii, adbeill Cig i iu nel ele inItocos C01Mct ion at thle eXpenlse Of other iiei'stleiilts. Hull iiidcxution by insmiitutionaui/ing infiilationi has bieei hield r-csJI'urisile 'or pici petiiatiliw it (eveni :hoiili' iiiflatOil has Cl.ceiith\ conlic downl to U. S. I 1l es in) ( 'huie 1i iiderI N% iLdL pi cald i IILde\aI lionI). Whenl inl flat i0t \ x aries from11 pCi i oI to pci WoI1, pi\ Iiieiis to past infilationl iiia1 II C\ cllt niiiz Lets f'romi beinig cleured; f'or example, Whlenl %%.igeS are ilde\Cd to pas,t hiiilluioii a diecliningir raite of' inflation caulses i'Cal x esto liNie, anld t le Same mnax occuir withi inei ci et utes. iciierul in- dkeCes O1 Ili'Iithaolri. IimI thici iiiore, are Very rOUgh N allrdst ickl s es peciall Whll ii Halhut ioil is hligh and xliu,as is tile Cacas\1e x~le iiii oiIs ar-e Iihei :li'cd, the composition of coii- liiiipt h in'l cliNiiiIpes 0% Li timeI. Iheref ore iidx gmav miss the niiirk andt create (his- 0I ti tail, Of its own oni that at:ctuiit, Also, as iifitltiii iiiciceuccs, SO doeC.s thie dcci ec of1 dlisper ,it Ii be1t%\LTcil difficut'C I pr*iCes, t hierebh LIIIiderrnliIiIIig tile ability of1 aniy onie iiid.'-\ to ,.;iilcel out the efftects of' inffalitioni or particular ecoionhj. agents. Fiiialhy, 0oi iOftlie pa Ptp ses ol 1iide\ At itm is ito imiai at ai i the existence ofa priate mar11ket Ibr Itiag Ilili securities inl tile face of, Intilc est rate 1iicertaiivt ; tIle e\pel ieiicc inl I a liii Anl I a is that, wihile publi,1c debt int ncl a es iiait iid thle piteSee- tor hias icnmiiicd ti.luctua to aeni.11 fullyI iiidIC\Cd bugterml deb11 kt obhliviiions at pi IiC'rel licit te II\lCia '. lt)(0.", CI i 1e,iai, 1974; ( ;iuiIbhc ic i and \ Sohoda, 1 97 4), Ine lia!!di1i-iieii O1 two) Crucial %m i,iIvihl die e\%.fiaut- raite and the itilci est rate. I~ cit n tell Of pertI Miillice. It bcCameI e6Li11 aI b0ug1 time ,iti tht a fi\ed c-x chiamu'e ratie - xQiii,ii\ ususainabIle MIC di 0 ltuic'tic i11ii ii01CLdd ioneLeel iner'il tionail \i itt b x de uitaw.I -'iii d or toit a c0II,i dcI,abl ju.- I \cii MeXiCo, Ira1.1mi1 1nitIx a loss intli etiC0t)n0In\ , had ito dc\,ihii in lQ,1 zlti adiL i iin 1976 c Ii t had ou ri iA II p t had b"comeik too cicalld and eJ( t Is 0 tooe ;~- xcat u df -.. i' ii I: idup 0 lsesiiijuti ht i 'ItI eeres ndevenl- u~al\ t Ilihuc i.iiriLx aci. ( alt Hi ;il :uu I oiuiii j''R uedthle i 'I u L' c us' L I, ' I Ia a Ic ~t c lier I c IIs t Il l nIi I l' aIta i'i.''iieip t Ita u:i it e cvl W\ l c c1iatioiis v'I j ht ,. i' i fs l ' 150 THE SOUTM'l PWESiTRN RhEVIE\V Winter 1982 effects of "maxi-devaluation." This technique proved to be successfuil, but the vir- tual indlexing of foreign exchange was thought by some governments to influence inflationary expectations. Hence they were tempted to yse the exchange rate as an anti-intlationary tool by lagging it behinid inflation, thereby undermining or defeat- ing altogether the very rationale for a crawlinig rate. This was often offset by export subsidies but these usually benefited manufacturing firms rather than farmers and ai ricultural output and exports often suffered as a result as did food output, becauLse lagging the exchange rate makes food imports artificially cheap. Of course, success or failure of exchange rate policies depends on the quality and consistency of o\ erall ecoiiomic policies including monetary, fiscal and income policies. Because it is difficul to maintain tiglht fiscal and monetary policies, there is often a trade-off between maintaining a realistic exchange rate and curbing inflation. The matter is complicated further by the interest rate issue. For a long time the problem of short-ternm capital flows ("hot money") was associated mainly withl the indus- trialized non-Communist economies. It has become very clear during the past decade, however, that developing countries are far from immune from the problem. Sam ers have a choice between investing their money domestically or (legally or il- legally) sending it abroad. Therefore, unless donmestic interest rates for savers (the passive rate) are close enough to foreign rates (say in the London market), it is highly likely that there will be capital outflow. Since developing countries need to be by definition, net capital importers, it inmplies tf-t external debt is usually stepped up in periods of capital flight to finance not only the normal current account defici' but also to try to make up for capita flight. After some time such a situation can become unstistainable (i.e., force corrective action). The resistance to maintaining realistic interest rates arises out of the closecd economy belief that cheap credit is socially desirable; cheap credit usually involves low passive interest rates, partic- ularly in countries where the banking systemn is inzfficient and the cost of in- termiiiediatioln high. Furthermore, because savings invested abroad may not be taxed much (or escape taxatioil easily) while .a% ings in domestic banks are often subject to substantial taxa- tioin, capital flight can only be discouraged at the cost of a fairly high "real" pre-tax domnestic lending rate. This will be oppos,ed by borrowers-industrialists and farmiers-and also x%rongly by government officials who believe that high interest rates are thernmel es inflationary in a cost-push sense. Finally, some countries, notably Argentina (between late 1978 and early 1981), not only have lagged the exchange rate but announiced in advance that the exchange rate would remain at a level generally judged belo'. the likely difference between in- ternational and donmestic inflation. With free capital movements, this policy assured high y ields in foreign currency to foreign investors. Such policies indeed managed to attract substantial capital inflows. But while short-term capital can undoubtedly balanice the books there is a danger that the exchanige rate xalidated by these finan- cial inftlows will convey wrong signals to the importing and exporting sectors. If, in the futuire, capital llo%. s should ebb for xx. hat ever reason the country Could find itself "addled with a weakened export infrastructure and possibly a severe lo,;s of oCUtput dnld i1lCOIlLC. Foreigln e\clianee crises in ILatin America almIIost invariably occur when both the exLchangle rate and the interest rate are lagging. Figuire I illuhstrates this for eight colllltries. \W hen both lines swerve to the left, exchange rates are becoming more Vol. 2, No. I I LCOINOMW( PI.RlMRNMANCE IN LATIN AMERIC(A 151 over alued and real interest rates more negative. This constellation of policies has been associatcd with a later foreign exchange crisis in Argentina (early 1970s), Chile (early 1970s), Costa Rica (1974), Peru (early 1970s), Bolivia (late 1970s), Mexico (mid 1970s) aiid Venezuela (late 1970s). I am indebted to Dr. Luis Landau of the World BaInk for this analysis. The index of real exchange rates was computed as follows: (Pesos/US$) t DPt (Pesos/'US$) o IPt where: DPt index of donmestic inflation; iP, index of "international" inflation. International inflation is the CIF (Cost, Insurance, Freight) index of US$ prices of inanula.citiired exports (SITC 5-8) of industrialized countries to developing coun- tries. The index of real interest rates w as defined as: 100 + Rm -~ where: Rm is the nmoney rate of interest on saving:s or time bank deposits; P: currcni rate of inflaLion as measurec by the Consumer Price Index Economies can stand a lagging exchange rate for some time, so long as capital in- flowxs rou-ghly offset the Nxideni;gi current account deficit; they can also stand in- creasingly negative passlive interest rates for some time so long as the current account balance of paymennts remains fairly healthy; but they cannot sustain a deterioration in both prices withlout running into a foreign exchange crisis. Such a crisis and the ensu inL co ixale>cence almost al % ays represent a cost in foregone economic growth. The corollary of this analvsis is that in most if not all cases, market-determined interest and exchange rates can indeed minimize foregone OLutpUt. Economic managenment of these crucial variables has improved over the past decade, owing much, unfortunlately, to un.satislfactory expericnces (notably Chile, Peru, and Argentina during the early 1970s). The most acute problems remain in small economnies (Costa Rica, Dominican Republic, and Jamaica, for example) while ten years ago some of the larger countric in the Region were suffering from them. Thi.s improvemient aLLurs *well for the decade of the 1980s. [:`orL'gnl Debt. While external borrowing has contribLItcd only some 10 to 15 per- cent ot the Region's fixed capital formation, the size of the Region's major econ- omiks is so large that they figure pronmiinently among the world's borrowers. Mlost of the expansion in the Region's debt has been from commercial banks rather than at cznccs,iona terms. Brazil, Nlexico, Argentina, Chilc, and Colombia alone account for nearly one-hlalf of the dec cloping world's \ azriable interest debt outstandinig. e\ ,r the least tei years maniy of the Region's couintries have had debt problems (severe airrears anid/or unplanned Iel ehdul1ines). This section looks at the extent to whllichl the dLebt problem can be trIcLd to inideqatil. -ccooionc managemient. Txwo aspects of -aerital debt have a dlccisiNke bearing on \%letler or not there is a debt probleim. One i the relationlthip bet %%. cen do;vie.t iic n% e,stmient anid niet e:oernal borrowing (the proporion otl inme,tnient that is financed out of borro\\ing rather thani c.(lne,tic savings), hlence tlihe adeqla.Lv Of thle d101siCSOC > inlus effort. The other is the use to which hmoi nmed fUnds are put. \\ hietlier- the returns on invest- 152 THE SOUTHWESTERN REVIEW Winter 1982 Argentina 100 1970- 1971 - 1972 -- 1973 - Real Interest t---~',-- 1974Rae 1975 t974 Rate -r;-. - --.------------------ 1976 -Real Exchange 1977 - Rate, Pesos 1978-Pe 1979 F et 35.20 5512 75.04 94 96 114.88 134.80 Index Chile 100 1970- 1971 - f Real Exchange 1972 -Rates 1973 --- Pesos per $ 1975 Real Interest---------------------------------- 1976Rae 1977 - - - - - - - - 1978 , 1979- 22 70 67 63 112 56 157 49 20242 24735 Index Costa Rica 100 1970 - 1971 Real Interest 1972 - Rates 1973 - Real Exchange 1974 'Rate 1975 - Colomes 1976 - $ 19 77 - 1978 1979 8050 9016 9983 10949 119 16 128.82 Index Peru 100 1970 - 1971 - 1972 - Real Interest 1973 - Rdtes 1974-t 1974 Real Exchange 197s - /Rates 97 - Pesos Per S 107 ----------------------------------------------------------- 1978 -- ~-- 1979 7440 92 54 InIea 1111168 12882 14696 165 10 Figure 1. Trends in Real Interest Rates and Exchange Rates in Eight Latin American Countries Vol. 2, No. I ECONOMIIC PERFORNMANCE IN LATIN AMERICA 153 Bolivia 100 1970 - 1971- 1972- 19'73 197| ------ Real Exchange 1975 Real Interest Rate 1976 "ares > JPesos 1977 ~1/ tlPer $ 1977 1979 67 60 85.41 103 21 121.02 138.83 15663 Index Colombia 100 1970 - 1971- 1972 FReal Exchange 1973 - Real Interest Rate 1974 Rates Pesos - 1975 Per $ 1976 - 1976--------- - ------ -- 1978 1979 ---- . I I I. 8760 94,71 10183 10894 11605 123.17 Index Mexico 100 1970 4 1971 1972 - Real Inter--t 1973 - Rates j Real Exchange 1974 - -------- Rates 1975 , --- . Pesos 1976 - ---Per $ 1977 - ------------------ 1978 1979 - 3090 90 12 9934 lU856 117 78 12700 Index Venezuela 1011 1970 - = 1971 1972 - tieal nterest - - -- 1973 - Rates Re3l Exchange 1974 Rates l97? R oblv;Uares - 1 976 Per $ 19 7 7 - 9~~~- 1977 9440 100 46 106 59 1t15 1 1t, 13.1 124t69 '. si I ant . F1 ,*,* 171)T B1a,ed 10 0 VVVrld Bdne 23723 154 THE SOUfIW'ESTI:RN REVIEW Winter 1982 ment-measured at world prices-exceed the interest costs of borrowed funds ultimately determinies whetlher a country is gaining or losing by borrowing. This issue is of course partictularly relevant at the present time when interest rates are high . The following countries have experienced debt servicing problems during the las. decade: Costa Rica, Argentina, Bolivia, Chile, Guyana, Jamaica, Nicaragua, Peru, and Uruguay. Brazil experienced debt servicing difficulties in the mid 1960s. There was some concern in 1980 with the size of thle Country's debt, but in the absence of arrears or unplanned rescheduling that courntry :s not considered to have had debt problems during the 1970s for comparative purposes in Table 11. Weak domestic savings el orts are often associated with debt difficulties, in cases where investment itself continues at high levels. The graphs in Figure 2 show savings (gross domestic savings) as a percent of GDP over the last decade (values are calculated at current prices). Even though Brazil has had no servicing problems the decline in its savings performance paral- leled an increase in borlro\sing costs, sLuggesting growing concern on the part of creditors. The graphs for Bolivia, Chile (1973-75), Guyana, Jamaica, Nicaragua, and Peru (mid 1970s) all suggest associationi between faltering domestic savings per- formance and debt servicing difficulties. Conversely, the graphs for Colombia, Ecuador, El Salvador, Guatemala (until recently), Mlexico and Paraguay suggest im- proved savings performance over time; these countries did not run into debt servic- ing difficulties. The avings perfolnance as showrn in the -rnlphm in Figure 2 reflect.1 mainly the follov. inlg factors: (a) llan6.ges in international terms of trade (i.e., windfall gains, as in the case of the Domiinican Republic in 1975 when the country comlmanded high sugar prices or losses, as in the case of nmost oil-importing- countries after 1973), (b) changes in a countr\'s own savings effort; and (c) accidental events such as a bad agricultural crop or political disruiption whielh can also depress the savings ratio but are unlikely to persist long enough to account for trends in the ratio over several Xe.ars. The second of these factors is of decisive importance. Changes in the interna- tional terms of trade have caused peaks and troughs; they do not explain trends sus- tained over a number of years. Therefore, barring political disruptions the savings perfornmance of a country is largely within the bounds of economic policy. Public savings depend directly on the management of current governiment revenues and ex- penditures, decisions about the efficiency of public enterprises, taxes, power and other public service rates. Private savings depend on pricing in the econiomy at large and also generally respond to financial incentives maniaged by the government. To a large e\tent, thierefore, the prhesence of debt servicing problems is a reflection of poor econonic management. Historically the variation of real interest rates over time has beenl too narrow to provide e% idence in support of a correlation het cen in- terest rates on sa inugs deposits and the volume of such savings. The recelnt ex- perience with substantiall\ increased positi\ -eal rates in Argenitina and Chile should provide the possibilily to carry out empirical explorations of this question soon. Wyhile national savin;]s may be inldependent of interest rate poflicie, the slhare of such ,aingu that is chanleled into the finaneial ,s\stenm (as opposed to real estate for e\aI-mpl1) does var xithli interest rates. Ihe higher the share of savings chan- nelled into the financial svs ltei, the mnore etli cient ov erall resource allocation should Vol. 2, No. I E((C)0NONI(' PERFORMANCE IN LATIN AMERICA 155 Table I 1. ICOR and Debt Difficultiesa ICORhl Debt Problems Jamaica Guyana - - Argentina Nicaragua - - Chile Honduras - + Peru Ur LIg IaW Venezuela + Nle%ico + + Costa Rica + Colombia + + Brazil + + El Sal\ ador + + Bolivia + Dominican Republic + + Guatemala + + Haiti t + Ecuador 4 -F Pa-aguay 1 + dNegative .i-jrso devote I(OR1 above the average for the Region; the IK OR,, reflect the relationship be- tween 1965 and 1975 insestment and 1970 and 1979 output growth. The absence of debt service problems .s denotecd by positive .iun>. iPin.a:n:i is nlot inicluded in the table. Since its currency is fully convertible, external debt and internal debt caninot be incallingfully separated out, hIC)Rs generally bear a strong inverse (but non-linear) relation to the GiDP growth rate. T o the extent that this relation obtains rapid growth itself, if sustained over many years, can be vicwed as enhancing be. Thus, interest rates policy does have a bearing on the efficiency of resource allocation even if one cannot relate it to the level of aggregate saviuns. 'tirniing to the efficiency of capital, we find that it is a very difficult thing to measure. O(ne very imiperfect and partial way to measure the efficiency of an economy- ihence of investnment-is by the incr-emental capital/output ratio (ICOR). The ICOR nmeasures the amounc of investment associated with a particular increase in output. The higher the ICOR, ceterisparihus, the less efficient is resource use in terms of growth. There are many methodological problems with using the ICOR in this manner. For exanmple, the measure attributes to investment changes in output that may be causecl by labor, technology or other factors contiributing to output. It may be worth finiding out, however, whether or not there is a relationship between uinfaa% orable ICORs (x% hicki are usually the result of poor economic policy, especially a poor u.se of' public funds or an excessi.e proporiion of social investment) and debt servicing problents. Table 11 .uytgests that suT,h a relationship may indeed exist. CoLIntries are ranked in order of (lecreasing ICOR. Nine out of the twVenty countries have e.xperienlced (or are experiencinig) debt servicing problems. Out of these, seven have uinfavotable 1CORs. Out of the eleven 156 THE SOUTHWESTERN REVIEW Winter 1982 Argentina Bolivia Brazil 269 F A21 - 258 260 - 211- 25 0 2590 20 1 F24 2 241 -191 23 4 232 - 182 r 226 223 17 2 21 8 214 16 2 210 205 - 152 202 19 5 14 2 19.4 -, _,_,_, _,_,_,_, Chile Colombia Costa Rica 1 7 4 268 18 6 - 15 6 25 7 F17 7- 13 0 24 6 F169 1 20 F23.1- 16 1- 1 01h 22 4 15 3 8 3 -21 3 -14 4 6B5 F 203 136 471- 192 128 1 ' 9J 18 1 119L. Dominican Republic Ecuador El Salvador 238 270 238 - 22 24 L 220 - 20 6-38 20 2- 189 h 22 2 184 - 173 V 207 - 166 - 157 191 148 - 141F 175 130 124 159 112 108 144 L , ,,,,,,,,,, | 94 _,,,,_._._,_._*_.- Guatemala Guyana Haiti 179 31 9 - 7 7 17 2 292 - 7 7 164 264 - 67 156\ 237 6.2 A 14 8 20 9 8S 0 5. 7 141 / 181 4. 133- 1 5,4 4 1 5 126 - 4j 11 7 38 3 '5969 't 73 75 77 79 1969 71 73 75 7? 79 t9t- 71 73 75 )7 79 6fI 72 74 76 78 89 72 74 3 6 78 8C. 71) 72 74 76 78 8 0 VorBd Bilnk 23724 Vol. 2, No. I ECONOMIC PERFORMANCE IN LATIN AMERICA 157 Honduras Jamaica Mexico 221 7 1 25 8- 210 24 9 29 iq L 31 23 9- 1A6 aF 23 0 - ' / 18 6. 220 142 16 4 11 i I1 - 2 1 201 10 8 I9- 92 943 18 2 Nicaragua Panama1 Paraguay ;iJ 261 ; \ /99 86 190 48- 1 88 19) t L , , , ,4 8 Is 8- I 1 123 9 17 73 71 22i9 16. 7 163 2ifln 14 4 14 4 ,0 I 1:2 181 _ 1L-4- 18 11- 114 Peru Uruguay Venezuela ,4 3f 3 -j 476f I liP 4 4~ Fa 43 1 40 9j 38 61 163 106VI 694k 9'i 34 2k VW,,il 8a.1 23725 Figure 2. Gross Domestic Savings as a Percent of GDP 158 'THE S(l I SO 1 IS IS H RN REVIFW :Winter 1982 countries with fii% ot able I'CORs only two (13(flivia aind Costa Rica) have experienced (or are cx.\prielicing) debt servicing lproblems. Bolivia's ICOR may be misleading because the growth pericd under cONsideralioi includes a once-and-for-all growth spurt due to petroleumn exports dluring the miid 1970s; Bolivia no longer is a signifi- cant oil exprlet, and its growtth has res,umcd a more normal lower rate. It is there- fore likely that Bolivia's normnal ICOR is higher than stigge,,ted by the table. The debt problems of Costa Rica % crc brought about by poor financial management rather than by struictural N%ealnc',s: the overwhelmin g problem is one of excessive puLlic e\pnC\ linintre. Out of ninie coutitries with miil.'morahle ICORs, seven are debt problem coun- tries. Of the others, one is a petroleurim e.xpolter (Vencela). The other (Honiduras) CoInMiLine,s an tin tfa orabhl ICOR and no debt prol eiws mnainly because most of the countr- 's debt during the 1970s was at concessionarv teri-ws. The ICOR thus enielmrC, as a reasonalbl]v robu.st iiidicator of long-teirn credit- \oriliine'.. Thle foreg0ing suggests that the quality of overall economic policies- notabiy the savings eftort an(d the qualiy[ of investmeniert-- is a basic deterninant of crecdit v orthiiiuc s.. This seiCtsin 'IIl,oii, then,, that agricultural policies, the nmanagenient of ex- change antdI interesl r.ate., sa'; nieLcnn'.. and the choice of inivestment are all crciul;:d in atl-iine -wist !itidLl i'proeentil'no in the stanidard ot living. REFERENCES il.,i., 13.. 1981. A4In'nt to Lxtertal .Shock.s in [) wlopuiw, Fconotie.5. Ihe World l3ank, \1 1'1!1111 1-III, 1).(.. Behrtian. 1. R.. 1968. Supplv Reyonw in Undler(evelolped . eriulfurc' North Hololand Publishing c(o., Aumsterdanm. Bergs-nan, .J., I ISii* Incoic di'aributioii and ps Ci t in Nlexico. W\orld Bank Stafit \Wolrinp Paper No. 395, June. Washington, 1).C. Bruderssohn, N., ed., I978. Financianilento de la educaci(n en Amnerica Latin. I.D.B. Cr.' .rma. T. R., ed., I --II- 51. I)ata tLor thC 'IUdl `t middle-class in Latin Aimerica. \Iiimeouiraph. Vk4w,h}im!'ion'l, D).C.. -ehxi, 1)., 1961. In I a/ tI titAerican Ix11sSlt-'.v and( 'ominenis, ed. A. 0. Hirschmarn. TIhe Tw eli iciti i Fnturi lund, Ness \York. E scLhlOm. A_, I'J--. Ilr.i,liul: vi/k dLkriknrirb u t in)iicome. .A inrican lconoinic Review, May. I riednian, NI . 19-4. U ,inu escalatoi s to help fi til inflation. / ortune, Jul. I 1 .. 1981. 1' ffults ss iich alternate mionetary policies hliae had o)n intlation anii growth in Latin Amiericani coutri ie .i. Dliscussioni IPa.i. Fourth D)eselopmnent IVikilik n Round T'able, Interanierican D)es iopnrent lHank, March 2-4. (,ernitaiii, (i., I9'0JA( , \r.ew:.l.l NtI;,noc,)LIIIIi '.iilJl, L 'i'. i o l l ). inwber", 1., I-and A. S obtoda. I i174. 1l I al:,i and indeatiOnl, it sUimmary ot issues. P'roLeedings of the ;Sastaobaten ('ont"rence on tle ,,it, of Indv%atioll. 1 aihI oi, l l.. , 1 iskilada 13ankeri. Stockholm. _ i: P.. 19 9t". I iN I;.io . uIi.'n j'it' iw t'l!. anid z inlA ci L il, i I iciit in Latin Atnerica. .Staitistiual d *': ofthe I S, ( Jcto a ` I)ecember,vol, 2. no. 4. - , I 'i''1. E tnfiW I.tecltiotni'nta th lMcLabor U,,n!. !in Ae xico. I atin American Intitute at the t n~ker,mt ot Nev M Cpil. inuald .1e,il Me > x^.1f!ic.E pub.ta, \t, p ioc%,jl Grumald, J.. in lt an Amnerican I.S---s-4ssavs anld C'a'm-, ed. &. 0. Hirschman The T\%% cntfictlh (Ceurv F tnnd, NeW ' orTk Harberger, A., and S. Edwards, 1980. International evidence on the sourees of inflation. Paper presented at the Conference oni inflation, Getulio Vargas Foundation, Rio de Janeiro, Mimeograph. Herring, H., 1962, A History of Latin America, Alfred Knopf, New York. Hirschman, A. O., ed., 1961. Latin American Issues-Essa,vs and Comments. The Twentieth Century Fund, New York. Hirschman, A. O., 1977. Thte Passions and the Interests. Princeton University Press, Princeton, New Jersey. 1981. Essavs in Trespassing-Economics to Politics and Bevond. Cambridge University Press, Cambridge. IBGE, 1973. Pesquisa nacionalpor aunostra de domicilios. Rio de Janeiro. Keesing, D. B., 1981. Exports and policy in Latin American countries: prospects for the world economy and for Latin American exports, 1980-90. Quarterly Review of Economics andBusiness, vol. 21, no. 2. Merrick, T. W., and D. H. Graham, 1979. Population and Economic Development in Brazil 1800 to the Present. Johns Hopkins University Press, Baltimore. Merrick, T. W., and R. Moran, 1979. Popuilation. Brazil-Human Resources Special Report. The World ilank, Washington, D.C. Montoya, M. U., 1981. 1i innerv and Losers in Colombia's Recent Growth Experience. FEDESAROLLO, Bogota. Morag, A., 1962. For an inflation proof economy. A merican Economic Review, vol. 52, p). 177-185. Pfeffermann, (C. P., and R. Webb, 1979. The Distribuiton o,f Income in Brazil, World Bank Staff Work- ing Paper No. 356, Washington, D.C. UNCTAD, 1955, 1960, 1970, 1979. Handbooks of International Trade and Development Statistics. New York. UNESCO, 1981. Statistical Yearbook. New York. U.S. Department of Agriculture, 1981. Wforld Indices of Agriculture andFood Production. U.S. GoNvrn- ment Printing Office, Washington, D.C. Vasena, A. K., and J. Pasos, 1973. Latin A merica-A Broader Rorld Role. Ernest Benn, Ltd., Lor,don Tcirnbridpe. von Hrumboldi, A., 1962. In A History ofLatin America. Alfred Knopf, New York, p. 192. Webb, R., 1974, Government policy and the distribution of income in Peru, 1963-73. Harvard University Press, Cambridge, Massachusetts. The Worid Bank, '981. Atlas. The World Bank, Washington, D.C. The World Bank, 1981. Social Indicators. Unpublished. The World Bank, Washington, D.C. The World Bank, 1981. Annual Report. The World Bank, Washington, D.C. World Bank H1eadqlulart ers: 1818 H Street, N.W. U Washington, D.C. 20433, U.S.A. Telephone: (202) 477-1234 Telex: RCA 248423 WORLDBK WUI 64145 WORLDBANK Cable address: INTBAFRAD WASHINGTONDC Eutropeani Office: 66, avenue d'Iena 75116 Paris, France Telephone: 723.54.21 Telex: 842-620628 -f Toky,i Off-ice: Kokusai Building 1-1, Marunouchi 3-chome Chiyoda-ku, Tokyo 100, Japan Telephone: 214-5001 Telex: 781-26838 The full range of World Bank publications, both free and for sale, is described in the World Bantk Catalog of Puiblicationis, and of the continuing research program of the World Bank, in World Banlk Researchl Program: Albstracts of Current Studies. The most recent edition of each is available without charge from: PUBLICATIONS UNIT THE WORLD BANK 1818 H STREET, N.W. WASHINGTON, D.C. 20433 U.S.A. ISSN 0253-2131