SciELO - Scientific Electronic Library Online

Home Pagelista alfabética de revistas  

Cuadernos de economía

versión On-line ISSN 0717-6821

Cuad. econ. v.37 n.112 Santiago dic. 2000 




This paper shows how the analytical framework developed in Lal-Myint (1996) and Lal (1998) can be used to think rigorously about the role of culture and politics in addition to traditional economic factors like factor-proportions in explaining divergent economic performance can be applied to Latin America. It is in particular used to explain the cycles of repression and reform and to judge the long-term sustainability of the current period of economic reform in Latin America.

Key words: Culture, political cycles of liberalization and repression.
JEL Classification: O1, P00, N00, Z1


The purpose of this paper is to show how the analytical framework I have developed over the last two decades, to allow economists to think rigorously about the role of culture and politics in addition to their traditional concerns with purely economic factors like factor proportions in explaining differing economic performance can be applied to Latin America. In this work. I have tried to answer two questions.

The first is: "why do countries repress their economies at the cost of efficient growth, and when and why do they reform?". This is the subject matter of Lal-Myint (1996) which through a comparison based on the economic histories of 21 developing countries in the post war period and a novel classification scheme for different polities found that rather than the type of polity, the initial resource endowment, in particular the availability or lack of natural resources, was a major determinant of the policies which impinged on the efficiency of investment and thereby the rate of growth. Section III provides a model of this process which we found useful and shows how it can be used to explain the cycles of economic reform and repression in Latin America.

The second question I sought to answer in part in my Ohlin lectures (Lal (1998) was: what determines the polity in different countries/regions and could this in turn explain the divergent economic performance of countries in the same natural resource endowment group? The basic answer is that political systems tend to exhibit the cosmological as opposed to material beliefs of different cultures (see section II ). There is considerable hysterisis in these cosmological beliefs, which in turn are largely determined by the broad ecological conditions (including factor endowments) faced by the relevant cultures/civilizations during their formative period. Given the hysterisis in cosmological beliefs it is therefore useful to think of the political habits of different countries/ regions as both limiting the political forms that are consistent with these habits as well as conditioning the political, and thence changing policy responses of different countries even in the same natural resource endowment group. Section II provides some cross-cultural comparisons of these political habits and their origins, with particular reference to Latin America.

In the light of these sections, the last section speculates on the future of economic reforms in Latin America.


(1) On Culture

In Lal (1998) I used a definition of culture due to ecologists which is particularly useful for economists1. They emphasize that, unlike other animals, the human one is unique because its intelligence gives it the ability to change its environment by learning. It does not have to mutate into a new species to adapt to the changed environment. It learns new ways of surviving in the new environment and then fixes them by social custom. These social customs form the culture of the relevant group, which are transmitted to new members of the group (mainly children) who do not then have to invent these new ways de novo for themselves.

This definition of culture fits in well with the economists notion of equilibrium. Frank Hahn2 describes an equilibrium state as one where selfseeking agents learn nothing new so that their behavior is routinized. It represents an adaptation by agents to the economic environment in which the economy "generates messages which do not cause agents to change the theories which they hold or the policies which they pursue." This routinized behavior is clearly close to the ecologists notion of social custom which fixes a particular human niche. On this view, the equilibrium will be disturbed if the environment changes, and so, in the subsequent process of adjustment, the human agents will have to abandon their past theories, which would now be systematically falsified. To survive, they must learn to adapt to their new environment through a process of trial and error. There will then be a new social equilibrium, which relates to a state of society and economy in which "agents have adapted themselves to their economic environment and where their expectations in the widest sense are in the proper meaning not falsified".

It is useful to distinguish between two major sorts of beliefs relating to different aspects of the environment: namely the material and cosmological beliefs of a particular culture. The former relate to ways of making a living and concerns beliefs about the material world, in particular about the economy. The latter are related to understanding the world around us and mankind's place in it which determine how people view their lives -its purpose, meaning and relationship to others. There is considerable cross-cultural evidence that material beliefs are more malleable than cosmological ones. Material beliefs can alter rapidly with changes in the material environment. There is greater hysterisis in cosmological beliefs, on how, in Plato's words, "one should live". Moreover the cross-cultural evidence shows that rather than the environment it is the language group which influences these world-views 3.

These cosmological beliefs are, moreover, important for the polity. For no matter how tyrannical and predatory, any State must be based on some general acceptance by the populace of its legitimacy- some general acceptance of its right to rule. These political habits of different cultures have been formed as much by the geography of the territory where the relevant culture was formed than any ideology.

(2) China and India

Thus, China in its origins in the relatively compact Yellow river valley, constantly threatened by the nomadic barbarians from the steppes to its north, developed a tightly controlled bureaucratic authoritarianism as its distinctive polity which has continued for millennia to our day. Despite its Communist trappings the modern Chinese State retains all the lineages of the traditional pattern (see Jenner).

By contrast Hindu civilisation developed in the vast Indo-Gangetic plain, protected to a greater extent by the Himalayas from the predation of barbarians to the North. As I argued in The Hindu Equilibrium, (Lal 1988) this geographical feature (together with the need to tie down the then scarce labour to land) accounts for the traditional Indian polity which was notable for its endemic political instability amongst numerous feuding monarchies, and its distinctive social system embodied in the institution of caste. The latter by making war the trade of professionals saved the mass of the population from being inducted into the deadly disputes of its changing rulers. Whilst the tradition of paying a certain customary share of the village output as revenue to the current overlord, meant that any victor had little incentive to disturb the daily business of its newly acquired subjects. The democratic practices gradually introduced by the British have fit these ancient habits like a glove. The ballot box has replaced the battelefield for the hurly-burly of continuing "aristocratic" conflict, while the populace accepts with a weary resignation that, through various forms of `rent-seeking', its rulers will take a certain share of output to feather their own nests.

But what of the New World in particular Latin America?

(3) The Americas

One obvious divide in Latin America is between Brazil —the ex-Portuguese colony— and the rest of Latin America (excluding the Caribbean) which was part of New Spain. But given the common religion and the similarity of the patrimonial states4 established in the late 15th century in Iberia, the similarities of the two colonial heritages are more important for our purposes than the differences.

Another divide is geographical and ecological. This is more significant. There are three broad geographical and ecological regions in Latin America. The first is formed by its mountain backbone which extends from Mexico south to Bolivia and Paraguay. This is where the Indian populations conquered by the conquistadors lived, and where the social structure is still partly based on a tripartite ethnic division between Indians, mestizos and creoles. Here the Europeans took over and adapted an existing form of production. The second region is the Caribbean- Atlantic coastal zone from Mexico south to northern Brazil. Here (and in the coastal valleys of Peru and the hot lands of Mexico) the Europeans organized a system of intensive tropical or semi-tropical agriculture which depended on the importation of African slaves. In these regions the Africans, mulattos and whites formed the ethnic mix and the social divide. The third region covers the Southern cone countries where the Europeans found few Indians, and those they did were quickly exterminated. These became countries, by and large, of white immigration- primarily from the Iberian peninsula, until this century. Temperate agriculture was the major form of production. In most parts of Latin America the treasures below the soil (minerals) —which provided the primary motive for the conquests— were as important as those to be garnered through agriculture.

The pattern of colonial landholding was determined both by these ecological conditions and the policies of the Iberian patrimonial states. It is notable that these policies diverged markedly from those in another region of land abundance- North America. Whereas much of development economics is concerned with the development of labour surplus economies of relevance for Asia, it is the economics of land abundant, labour scarce economies which is relevant for the New World. A seminal essay by Domar (1970) provides the necessary theoretical framework. He cogently argues that in a land abundant economy, free labour, free land and a non-working upper class cannot coexist. Any two can but not all three.

Thus consider the case where land and labor are the only two factors of production. Land is so abundant that there are no diminishing returns to labor, whose marginal and average product are the same (mpl=apl). If employers seek to hire labor, they will have to pay a wage equal to this common marginal and average product of labor, leaving no surplus rents from land for the employer. Hence the agrarian form that will emerge is family labor based farms, as any form of hired labor or tenancy will be unprofitable and landlords -who have to depend on one or the other- cannot exist. A government, by taxing this independent peasantry through direct or indirect taxes, could support a non-working class of retainers; but the latter or an independent nobility of landlords could not support themselves from land rents- as none would be available. Economic expansion based on an independent yeomanry was the form that North American development and its agrarian structure took in the colonial period.

Next, suppose the government wants to create an independent class of landowners and grants the chosen few sole rights of ownership to land. If the peasants are free to move, competition amongst landlords will drive the rural wage up to the marginal product of labor which is close or equal to its average product because of the abundance of land. There will be little or no surplus left for the landlords. In order to provide this surplus some means will have to be found to restrict or abolish the peasant's freedom to move. Various forms of tying labor down to land- serfdom, slavery and the caste-system- emerged in the great agrarian civilizations. They created a landowning class which derived a rent not from land but from the peasants by expropriating a large part of their income above a subsistence level5.

Finally, as the labor force expands from natural increase and/or migration and land becomes scarce relative to labor, diminishing returns to labor appear with the mpl < apl. This allows landlords to obtain the rents from land and an assured labor supply to work it through hired labour paid its mpl, or else through various forms of tenancy.

The land abundant US sub-continent was gradually tamed by the gradual westward spread of the family farm. In Latin America "a society of small farmers failed to take shape. Spain could not export many, and in America the lure of mines, the possibilities of large scale, preemptive acquisition of land, and the opportunities for exploiting Indian and African labour militated against such a design" (Morse (1964), p.128).

Equally important were the differing ecological conditions for agriculture in North and South America. In the North, unlike the tropical parts of the Americas, grains were the most suitable crops for cultivation. These have constant returns to scale in their production, unlike plantation crops like sugar which have increasing returns to scale. The same is true, to a lesser extent, of tobacco and coffee6. Where climatic conditions in the Americas were suitable for cultivating tropical crops, the use of coerced labor had enormous cost advantages over free labor, which led to great social and economic differentiation in society with large inequalities of income and wealth.

By contrast, given its factor endowments (including the climate), in most of the U.S. (except for the South) the family farm became the backbone of the colonial economy and a society with fairly egalitarian mores could develop. That these factor endowments (including the climate) rather than the cultural differences between the Protestant North and Catholic South were responsible for the development of these different types of societies in the Americas is illustrated by the case of the Puritan colony of Providence Island, which developed the Caribbean and Latin American pattern of land ownership and settlement rather than the North American one of its co-religionists7.

These cultural differences were however vital in the different polities that were established in the areas of Iberian and Anglo-Saxon colonisation. Morse (1964) argues that, Spain after the reconquest (from the Moors) was a patrimonial state in which feudalism never developed fully. It was a centralising state without the decentralisation of rights of the manorial system. Thus when the New World was conquered, the soil and what was beneath belonged to the Spanish crown. Instead of the manor, in New Spain, the Crown granted an encomienda which "was a distribution and entrusting of a community of Indians to a Spanish colonist who collected from them various forms of tribute and was obliged to protect them and assist in their assimilation into Christian civilization" (p. 146). It was not a grant of land and only permitted limited rights to acquire holdings as long as these did not encroach on Indian community lands. But by the end of the 18th century this institution was decaying, partly as in 1718-21, the Crown decreed that encomiendas would revert to it on the death of their holders.

By the 19th century the major form of land-holding was the hacienda and the plantation. The hacienda differed from the encomienda in that "(1) the proprietor held title to land. (2) The Indian workers had little or no land and were bound to the hacienda through debt peonage. (3) The Indians enjoyed no tutelage from the State and in effect were under the jurisdiction, and at the mercy of the hacienda" (Morse, op cit., p. 149). The hacienda was to be found mainly in the livestock and cereal zones.

By contrast, in the tropical and sub-tropical regions, the characteristic agrarian institution was the plantation, which specialized in single export crops for world markets. They tended to be "highly capitalised and to rely on slave or immigrant labor, utilised more or less efficiently to minimise labour costs. Mechanisation, as it becomes available and economical, tends to be introduced on plantations, not haciendas" (p. 150).

Despite the continuing rhetoric of the desirability of land reform in Latin America, attempts to replace haciendas by peasant farms have foundered (a) as in Mexico and Bolivia because of the failure to educate the peasantry in more productive methods (see Lal-Myint (1996), Chp. 8), and (b) on the plantations as in Cuba where they were replaced State enterprises or co-operatives (see Morse). These collective forms are manifestly inefficient compared with the capitalist enterprises they replace because of the perverse system of incentives associated with most forms of state enterprise.

The patrimonial rather than feudal states that Latin America inherited were further distinguished by their Catholic lineage. Whereas in the Protestant colonies- as Luther succinctly expressed in his "Open Letter to the Christian Nobility"- the duty of Christians who found themselves in a land populated by pagans "was not to convert the pagans but to elect their own religious leaders. The American Indians ..were be tamed or exterminated. Moreover, the idea of salvation of one's neighbor never enters the Calvinist ethic because only divine grace, not human action, can save man..As the intermediaries between the individual conscience and God are suppressed, the evangelizing mission of Christianity disappears." (Paz (1988) p. 27).

By contrast evangelism was the public justification given for the Conquest and for the Spanish and Portuguese domination of Latin America. New Spain even more than its parent state adopted the neo-Thomism developed by Suarez and his disciples as part of the Catholic church's revitalisation during the Counter-Reformation. This provided an ideological justification for the patrimonial state. Society is considered to be a hierarchical system in which every person and group "serves the purpose of a general and universal order that transcends them" (Morse, op cit.). This hierarchy is part of a universal and natural order and not the product of any social contract. The sovereign is responsible to God not to society even though his authority originates in the people. "Neothomism was a philosophy destined to offer a logical and rational justification of the Christian revelation. In turn the teaching and defense of the Christian revelation formed the basis of the Spanish empire. Religious orthodoxy was the foundation of the political system" (Paz p. 30). Its economic correlate was corporatism.

This political and economic system was par excellence an 'enterprise' association as delineated by Oakeshott (1993)8. By contrast, the Protestant colonies were relatively indifferent to religious orthodoxy. In the previously cited work, Luther maintained that, in the colonies, if a group of Christians had no priest or bishop amongst them they should elect one of themselves as a priest and this election would not only legitimize their authority but also consecrate it (see Paz, Morse, op. cit.). As Paz notes "nothing similar exists in all of Catholic tradition" (p. 27).

Thus in the Protestant North a pluralist society developed, with the view that "the world is composed not of one highly differentiated society for which certain common forms, acts, and ceremonies are a needed binding force, but of a multitude of unrelated societies, each of them a congregation of similar persons which in finite time and place and ordered by the declarative terms of a compact rather than by common symbolic observances" (Morse, p. 152). This allowed the notion of the State as a civil association in Oakeshott's terms to develop, with the State as the umpire between many competing interests. This difference in cosmological beliefs explains the observation by political scientists that: "politically, North Americans confine their feuds primarily to selecting officials and debating public policies, but in Latin America feuds are more fundamental...democrats, authoritarians, and communists ..all insist they know what is best for themselves and their neighbors" (Wynia, p. 3). This "universalism" of the neoThomist tradition was further strengthened by the attempt of the Jesuits in Latin America (and in other parts of the world) to promote a religious syncretism which would lead to a "unification of diverse civilizations and cultures..under the sign of Rome" (Paz. p. 39)9.

This fundamentalist universalism also provides, in my view, an explanation for the continent wide swings in political and economic fashions over the last two hundred years10. In the post war era, the pronouncements of the Economic Commission of Latin America (ECLA) have been accorded the status of gospel truth. When it advocated dirigisme, that became the policy for most of Latin America, when it finally endorsed economic liberalism in the early 1990s that has become the new gospel. More than other parts of the world, therefore, a universalist ideology matters in Latin America. Instead, therefore, of searching as political economists do in other Western societies for the changing equilibrium of interest groups, in Latin America one needs to explain how these intellectual swings of "fashion" take place, as they are rather like religious conversions _Menem in Argentina and Cardoso in Brazil being outstanding examples.

This penchant for universalist ideological beliefs has also meant that there is a continuing dissonance between the Latin American social reality of the extreme inequalities which are the result of its ecological and political heritage and its Christian cosmological beliefs emphasising equality -which it of course shares with the North. There is no such Northern dissonance as both for ecological and political reasons a uniquely egalitarian social and political society developed there.

In this context it is worth noting the important difference between the cosmological beliefs of what became the Christian West and the other ancient agrarian civilisations of Eurasia. Nearly all of these believed in some form of hierarchical social order, which for instance in Hindu India -with its belief in reincarnation- was rationalised as resulting from the system of just deserts for one's deeds in the past life. By contrast, alone among the Eurasian civilisations, the Semitic ones (though least so the Jewish) emphasised the equality of men's souls in the eyes of their monotheistic Deities. Dumont has rightly characterised the resulting profound divide between the societies of Homo Aequalis which believe all men are born equal (as the philosophes, and the American constitution proclaim) and those of Homo Hierarchicus which believe no such thing.

This matters for the polity. With the rise of Demos, those societies infected by egalitarianism have a greater propensity for the populism which damages economic performance than the hierarchical societies. If as in Europe the granting of democratic rights can be phased in with the growing economic and social equality that modern growth helps to promote, then the political effects of the dissonance between an unequal social reality and egalitarian cosmological beliefs can be avoided. In the colonial and 19th century patrimonial states of Latin America this dissonance was avoided by restricting the polity- in effect to the property owning classes. But if as in this century, while still in the early stages of modern growth, the polity is expanded by incorporating the "dangerous classes", through an extension of democratic rights to the whole populace, then this dissonance can, as it has, lead to political cycles of democratic populism followed by authoritarian repression as the distributional consequences of the populist phase are found unacceptable by the Haves. By contrast, hierarchical societies can more easily maintain majoritarian democracies, however corrupt and economically inefficient- as the notable example of India shows- despite continuing social and economic inequalities. Thus, as many Latin American commentators11 have noted, the historic and continuing inequalities of Latin America make democracy insecure, largely- I would argue- because of the dissonance between `society' and `cosmology' noted above.

Two other points need to be made about the transition from the colonial polity and economy. The first is that, though couched in the universalist language of the French Revolution, the wars of Independence- including the American revolution- were nationalist creole revolts. They were prompted by the policy of the European powers of barring the entry of the creole elite to higher official and political office in the metropole, even as "penisulars" had access to high positions in both the colonies and the metropole. This led to resentment amongst the creole elites. The accident of birth in the Americas seemed to condemn the "creole" to an inferior status, even though in every other respect —language, descent, customs, religion, manners— he was indistinguishable from the "peninsular". "There was nothing to be done about it: he was irremediably a creole. Yet how irrational his exclusion must have seemed! Nonetheless, hidden inside the irrationality was this logic: born in the Americas, he could not be a true Spaniard; ergo, born in Spain, the peninsular could not be a true American". (Anderson (1991) p. 58).

But if independence was to be declared in the name of a "nation" whose inhabitants were distinguished by being born in the New World, it would have to include all the people in the territorial area formerly controlled by the metropolitan power. Recognizing this, Bolivar's fellow-liberator, San Martin decreed that "in the future the aborigines shall not be called Indians or natives; they are children and citizens of Peru and they shall be known as Peruvians", (Lynch (1973) p. 276). The ideas of the Enlightenment which had spread from the metropole also meant that these new "nations" were opposed to dynastic rule, and were pervaded by republicanism. The Nation serving Demos which has been a defining characteristic of the modern age was born.

But despite the similarities in the political forms adopted by the newly independent Latin American with the Anglo-French states, the reality was that the "collapse of the supreme authority activated the latent forces of local oligarchies, municipalities, and extended-family systems in a struggle for power and prestige in the new, arbitrarily defined republics. The "caudillo" of the independence period, controlling a clan-like or an improvised retinue through charismatic appeal, was the latter-day version of the conquistador..[T]he contest to seize a patrimonial state apparatus, fragmented from the original imperial one, became the driving force of public life in each country" (Morse, p. 162)

Though they took on the trappings of Western constitutional democracy, the polity was limited. The two party systems which emerged, "reflected an alignment of "conservative" landed and monied interests, high clergy, and former monarchists against "liberal" professionals, intellectuals, merchants, and those with a creole, anticlerical, and anticaste outlook" (Morse, p. 163). This period of oligarchic rule ended with the expansion of the polity, with fresh immigration and the emergence of an industrial working class -with the import- substituting industrialisation induced by the breakdown of the world economy in the interwar period. The need arose to co-opt these dangerous classes into the political system. A number of political entrepreneurs -Peron in Argentina, and Vargas in Brazil- demonstrated how a populist coalition could be put together. But the colonial heritage still lingered in the minds of the rulers and the ruled.

Morse has summarised what he believes are "a number of enduring premises that underlie Latin-American political life" (p. 172). These are (1) "now as in the past the sense that man makes and is responsible for his own world is less deep or prevalent than in many other lands" (p. 172). (2) "The Latin American peoples still appear willing to alienate, rather than delegate, power to their chosen or accepted leaders." (p. 173) (3) There has been a continual quest for legitimate government once the pinnacle of the colonial patrimonial system the Crown was removed from the polity. In a patrimonial state "to which command and decree are so fundamental, the legitimacy of the command is determined by the legitimacy of the authority which issues it. Hence the importance of sheer legalism in Latin-American administration as constant certification for the legitimacy, not of the act, but of him who executes it". (p. 175). (4) "The innate sense of the Latin American people for natural law is matched by a more casual attitude toward man-made law" (ibid). (5) As in the 16th century "the larger society is perceived .... as composed of parts which relate through a patrimonial and symbolic center rather than directly to one another" (p. 176). Hence the persistent appeal of corporatism in the region, of which the politico-economic system established by the Partido Revolucionario Institucional, PRI, after the Mexican revolution, is the most notable example.

Second, note that, in the 19th century, Latin America was successfully and profitably integrated into the liberal international world economic order (LIEO) created by Pax Britannica. Its economic policies were those of economic liberalism, and its prosperity was unprecedented. For instance, Argentina was considered at the turn of the century to be on a par with the US. The current century has by and large been a period when Latin America has repudiated the LIEO. It has provided its own distinctive continent wide justification of dirigisme through ECLA and its gurus of "inwardness" like Raul Prebisch, and the current President of Brazil F.H. Cardoso, who in an earlier incarnation was a peddler of "dependencia" theories. Its relative economic performance has nevertheless been more than respectable (see Lal and Maxfield). But with the debt crisis of the 1980's and a muddled period where the old dirigiste panaceas were tried one last time (except most notably in Pinochet's Chile) Latin America has reverted to the 19th century policies of economic liberalism12. Is this most recent conversion likely to last any longer than that of the 19th century?

(4) The Political Economy of Post-war Latin American Development

In answering these questions it is useful to ask: why in the 1960's, Latin America did not adopt the East Asian mode of development —outward orientation— which is now generally agreed to be necessary for promoting stable and equitable growth in the long run (see Lal-Myint (1996))?

Sachs (1985) and Hirschman (1968, 1981) have provided various political economy explanations which are not persuasive (see Lal (1998a)). Instead, the typology, which I developed for a multi country comparative study of the political economy of poverty and growth (see Lal (1984)(1986), Lal-Myint (1996)) is more useful.

The first of the types of State, rarely observed but implicitly assumed by much of the technocratic economic policy literature, is the "benevolent State" - run by selfless Platonic Guardians or a benevolent dictator, maximising the social welfare of its citizens.

The second, I label the "predatory State". This is a state run by a self-serving absolute ruler, such as a monarch, dictator or charismatic leader. This sovereign, too, is autonomous, in so far as the constellation of domestic interest groups has little direct effect on his/her policies. For analytical purposes the selfish predatory sovereign is assumed to either maximise net revenue (treasure) or courtiers (bureacrats).

The third type of State I call "factional". This is a State which subserves the interests of the coalition of pressure groups which succeeds in its capture. The interests served are narrowly defined (again for analytical simplicity) to be the economic self-interests of the constituents of the government. The method of capturing the State need not be majoritarian democracy, even though such a form of government would be compatible with this type of State13.

I would contend that, the Latin American State is typically a factional State, though the other two types are also observed. Thus, controversially, it can be argued that Pinochet's Chile fits the model of the "benevolent State". It set itself above the hurly burly of pressure group activity and sought to subserve the commonweal - as it saw it. All the evidence shows that Chile stands apart, as the one state which has met the adjustment needs of the debt crisis by preserving targetted social welfare programmes, while cutting back on other inefficient public expenditures and in liberalising the economy - much as mainstream economic policy advisors would recommend14 (see Castaneda (1992)).

Of our second ideal type, Mexico under the PRI, I contend, provides an example of the "bureaucratic maximising" variety of "predatory State". But most of the other Latin American States are factional. In a few of these, where there is a political consensus, as in Costa-Rica, a relatively stable majoritarian democracy is the form of the factional State. In most of the others, the lack of political consensus has meant a constant shuffling of different co-alitions of interest groups capturing the State by fair means or foul.

In all these models the controllers of the State are maximisers -though the maximand clearly differs from one type of State to the other.

It is in this context that, the implicit explanation provided by Veliz (1980) for the failure of Latin America to adopt the required outwardly oriented industrialisation policies is of some interest. Veliz contrasts the centralist Iberian tradition —which he argues has molded the actions of all the incumbents of Latin American States— with the more decentralized tradition based on Lockean rights and the ideas of the Scottish enlightenment, which provided the ideological ballast for the states of Anglo-Saxon extraction during their period of industrialization15.

Veliz considers the liberalisation of trade in the nineteenth century open economies of Latin America as merely "a liberal pause" in the deep seated centralizing tendencies inherited by Latin America from its Iberian conquerors. The Great Depression and its aftermath greatly enhanced the role of the central state, which became the main financier of private industrial ventures, the "arbiter in the process of income redistribution through the implementation of a variety of social policies; and it assigned a dynamic role to the public sector" (Veliz (1980) p. 259).

The role of the intelligentsia in the subsequent decades, however was to aid and abet the centralising state because of the "redistributive bent of [the] populism and social democracy many of them supported". "By the early sixties, the populist and social democratic cornucopia had consolidated huge pressure groups that regarded their better interests as identical with those of a prosperous central State" (ibid p. 290).

As the means of funding these growing entitlements declined, "the social democratic regimes had to face growing discontent, especially among the middle sectors whose expectations had risen highest during the decades of largess and were consequently more vulnerable to rising inflation and declining economic fortunes. Their appeals however, were not for a revolutionary change of the social and economic system, but for more state intervention along the same lines as before" (ibid, p. 292). Meanwhile the intelligentsia, Veliz contends, adopted the fashionable view of the Cuban Revolution current amongst the radical intelligentsia in the West at the time, namely that, "Latin America [had] become the revolutionary frontier" (p. 295). The terrorist campaigns they launched changed the attitudes "of the urban middle sectors.... and opened the door to the current process of authoritarian recentralisation", (ibid p. 293). It is particularly pertinent in this context that, "in Latin America the military have traditionally been drawn not from the upper strata of society, but from the middle and, increasingly, from the lower middle sectors.... To the violent revolutionary actions of the extreme left wing terrorists, the middle sectors responded first with legal repressive measures administered by social democratic regimes. When these appeared insufficient to stem the tide of violence or when the parallel economic deterioration threatened general institutional collapse, the middle sectors appealed to the military for help" (ibid p. 297).

But Veliz notes two paradoxes in the economic policies of economic liberalization that the subsequent military governments espoused (though not always followed)16. The first is that, "it was precisely those [failed] policies that greatly enlarged and consolidated the vast urban clientele that called them to power, to dismantle the public sector built over the past three or four decades would risk their disaffection". (ibid p. 299). The second paradox is that "absolute central power is exercised absolutely in the name of liberal economic theories and schemes whose obvious pre-requisites are diversity and freedom of choice", (ibid p. 301).

Veliz seems to me to provide a more cogent picture of the intellectual and social forces underlying the continuing Latin American crisis than that presented for instance by Hirschman. It is these forces which I seek to examine more analytically in the next section, to show why even though the technocratic remedies for long term growth in Latin America are by now uncontroversial, it maybe difficult for Latin American governments to follow these policies consistently17.


(1) A Model

The starting point for any political economy of the "factional state" in Latin America, has to recognize an important but paradoxical difference between Latin America and East Asia (for instance), This is the former's relative "embarrassment of riches". An excellent picture of this is provided in Fig. 1, which reproduces Leamer (1987)'s recent depiction of the factor endowments of a sample of countries in a triangular endowment simplex which has three factors of production - land, labour and capital19 as its three vortices.

In this diagram, as one moves rightwards along the labour-capital axis, the countries depicted have higher and higher capital-labour ratios, and as one moves upwards from the labour vortex along the labour-land axis, the countries have higher and higher land to labour ratios. A point within the triangle gives the factor endowments of a country with a mix of the three factors of production. Thus the point TOT which represents the average factor endowment of the world economy has a capital-labour ratio given by the intersection of the ray drawn from the land vortex with the labour- capital axis (a ratio between 10 and 20 as shown in Fig.1); a land to labour ratio given by the intersection of the ray starting from the capital vortex with the labour-land axis (just less than 1 in Fig.1); and a capital to land ratio given by the intersection of the ray starting from the labour vortex and the land-capital axis. This reference point (TOT) allows he triangle to be divided into a number of regions where the factor endowments of a country can be ranked relative to the average "world" endowment. Thus in region A the countries shown (which include Mexico and Brazil) have factor endowments which are relatively land and labour abundant and capital poor compared to the world endowment. While in region C they are labour abundant and both capital and land poor (like Hong Kong in the diagram) and so on for the other regions shown.

Thus we can succinctly characterize the differences in factor endowments between countries by reference to their positions in the Leamer triangle. Furthermore, in the same triangle we can also depict the relative factor intensities of the different commodities produced and traded in an open world economy as we do in Fig. 2. So we have a geometrical representation of the a three factor many (n) commodity open economy model in which a country's factor endowments tell us in which of the 7 (assumed) regions of "diversification" it will be placed, and thence the commodities that the country will produce under free trade (see below).

One striking fact emerging from Fig. 1 is that, whereas most of the East Asian Gang of Four are on, or close to, the "labour-capital" axis, most Latin American countries - being relatively "land and natural resource abundant" (both with respect to labour and capital) - are positioned towards the land vortex. This is their "embarrassment of riches" which in a 3 factor-n good model can lead to some surprising and, from the view point of political economy, difficult development paths, as compared with the land scarce East Asian countries. In her Graham lecture, Krueger (1977) emphasised the importance of explicitly accounting for land as a separate factor of production in models of trade and growth. Leamers' work (1980, 1987) has provided further empirical and analytical support for her insights20.

It is the political economy of industrialization that we consider in this 3 factor - n good model, and show how "natural resource" riches can turn out to be "a precious bane". We follow Leamer (1987) and assume for simplicity that all goods are produced through fixed proportions techniques, so that there is a single input vector corresponding to each commodity. The production functions are the same all over the world, in keeping with the Hecksher-Ohlin tradition. Free trade is assumed. For simplicity we also assume that all goods are traded21.

Let us assume that in the World economy there are five manufactured goods indexed from 1-5, produced with only labour and capital. They are of increasing capital intensity and their input vectors are shown as M1 ....M5 along the "labour-capital" edge of the Leamer endowment triangle if Fig 2. In addition there are two agricultural goods. The first A1 is produced with only labour and land, and lies on the Labour (L) and land (T) edge of the endowment triangle. The second A2 uses all three factors of production, but it is more land - intensive than the agricultural good A1.

The seven points representing the input vectors and the three axis co-ordinates are connected by line segments, to divide the endowment triangle into seven "regions of diversification"22, for a given set of commodity prices for the seven commodities. The factor endowment vectors of different countries can also be represented by points in "land-labour-capital" space in the endowment triangle. Then it can be shown that, countries with endowment points in the same region of diversification will have the same factor prices and produce the same commodities with the vector inputs given by the vertices of the regions of diversification. The commodities produced in the seven regions are listed in the note to Fig. 2. Given commodity prices, relative factor intensities determine factor prices in each of these regions.

Now consider two possible paths of development in this model. The first is that of the typical East Asian Gang of Four, whose endowment point EA is on, or close, to the "labour -capital" axis. With capital accumulating faster than the growth of the labour force, the country moves from region I, to III to 1V. In this process it moves up the ladder of comparative advantage with respect to manufactured goods, with rising capital intensity. Hence on this development path the wage rises and the rental on capital falls.

The second path is for a land abundant Latin American country whose endowment point EL lies in the region of diversification II, where it produces both the relatively labour intensive agricultural good A1 and the land-cum-capital intensive good A2. We concentrate on one possible path of development, which seems plausible for a number of Latin American countries. We assume that, as in the previous case, both capital and labour are growing. Over time the economy's land to labour ratio will be falling, and its capital-labour ratio rising. Suppose this path of the economy's changing endowments is given by the dashed line from EL. Over time, the economy will then move from region II to VII to VI to IV. In this process it will begin to industrialize as soon as it moves into region VII, but in the most capital-intensive manufacture. Over time it will move into regions of specialization in which specialization in increasingly more labour intensive goods is required. The factor price consequence of this development path, in line with the emerging and changing comparative advantage of the economy, is that the wage rate will fall, and from the time the economy moves into region VII, the rental rates on capital and land will keep rising. Clearly, as compared with the stylized East Asian case (with rising wages), the distributional implications of this stylized Latin American development path (with a falling or constant wage) for an open economy would seem to be politically hazardous. But note that even though the wage might be falling on the Latin American development path, it will still be higher than on the East Asian path till they converge on the region of specialization in region IV23 24.

It is time to introduce the political features of the factional state, I do this following Lal (1986), using certain ideas developed by Mayer (1984) in endogenising tariff formation for the political economy of trade policy in developed countries.

Suppose, initially that there are only two factors of production, capital (K) and labour (L). All individuals in the economy can be described by their respective capital- labour (k1 = K1/L1) endowments. The mean of the individual k1 endowments will be the aggregate capital-labour endowment (K/L = k) of the economy.

Next we define the set of individuals who are decisive, in the sense that they can compete for the capture of the State and thus the determinants of economic policies subserving their interests. Suppose initially that, all economic agents in the population form part of the decisive set of the polity, and the political mechanism is democratic - with "one man one vote", and the majority capturing the State. All voters vote their economic interests. Then from the well known median voter theorem, the median voter's capital/labour endowment (km) will determine the interests that will be served by the coalition of majoritarian interest groups who capture the State. If the distribution of individual factor endowments is symmetric so that its median and the mean are the same, the median endowment will be identical to the average for the economy as a whole (k = km). Then, from the law of comparative advantage, we know that the income of the median individual will be maximized by free trade. If, however, the median individual endowment is more (less) capital intensive than the average, the median voters income generating interests will be in a tariff (subsidy) on capital intensive imports or a subsidy (tariff) on labour intensive imports. Thus, in this form of the pressure group model, what we need to know is the mean - the national factor endowment- and median of the distribution of the income generating factor endowments of the set of decisive individuals.

Initially, in our typical Latin American polity with factor endowments given by EL in Fig 2 the decisive individuals are the landlords. This is the oligarchic phase of Latin American development. The oligarchs' median endowments of land to labour is likely to be greater than the average endowment for the economy as a whole (in region II). This implies that, (as in the 19th century), the interest of the median `decisive' individual in the oligarchic polity will be served by free trade. As the economy grows, with capital accumulating and population expanding, but without any expansion of the "polity", the median endowment of the land and capital of the landed oligarchy will still be higher than the average of the economy, and so the oligarchy will still wish to maintain free trade.

However, suppose over time, with the introduction of manufacturing and the growth of the labour force, pressures grow for enlarging the polity and a co-alition (populist) dominated by individuals with more labour, relative to capital or land, comes to capture the factional state. Suppose this happens during the transition of the economy from region VI (where the wage was higher than in region II) to region V. During this phase, as relatively more labour intensive manufacturing comes into the "region of specialization" there will be inexorable downward pressures on the real wage. To prevent this outcome is in the interest of the median member of the enlarged polity (whose endowments are dominated by labour). This end maybe achieved by placing a tariff on the more labour intensive of the two goods produced in region VI, namely M4. This will, ceteris paribus, raise the equilibrium capital labour ratio in M5, lower that in M4, and raise the land-labour ratio in the production of the agricultural good A2 25. Thus the A2 point will shift north eastward (towards the T and K axes), the M5 point will shift towards the K vortex, and M4 towards the L vortex. This will enlarge the region of specialization VI, and allow the new endowment point which would otherwise have fallen in region V to remain in region VI, with a higher wage, and lower rental rates on capital and land 26.

Over time, however, for well known reasons, the inefficiencies associated with protection will lower the efficiency of the economy. If this also leads to a lowering of the rate of capital formation relative to the labour supply then the future transition to a lower real wage economy will be accelerated.

Similar pressures for protection can be expected to arise in a factional state, where the median individual in the ruling coalition of interests controlling the state has an endowment dominated by labour relative to capital or land, when there is a terms of trade improvement. For in a country whose exports are more capital intensive than its imports, the rise in relative export price will put downward pressure on the real wage, as the expanding capital intensive sector creates incipient excess supply of labour, and excess demand for capital and land (see Lal(1986) for the details)27. It is not surprising therefore that, the populist phase in Latin America, by and large, co-incided with the terms of trade improvements for primary product producers associated with the Korean War boom of the early 1950s. Mutatis mutandis, periods of terms of trade losses would also be periods where the interest of the median member of the enlarged polity is no longer at odds with a reduction in protection. For the terms of trade decline would mean that the relative profitability of the more labour intensive goods would rise, and this would enlarge the region of specialization in Fig. 2, where the more labour intensive good would be viable at the continuing high wage (previously validated by protection) without protection. Apart, therefore, from the swings in Latin American politics based on conjunctural factors (primarily of a macro economic nature)28, these considerations based on our model of the long run development of the Latin American factional state, provide some further reasons why (because of "deeper" political economy factors) periods of marked terms of trade improvements should be associated with pressures for protection and those with terms of trade losses with moves towards liberalization.

There are a number of other features of Latin American political economy (briefly touched upon in the preceding section) which I hope the model of the Latin American factional state can illumine. First, as we have noted, the development path charted by the dashed arrow in Fig. 2, can be deflected towards the (K) vortex, the higher the rate of capital accumulation relative to the economy's rate of population growth. The greater this deflection, the more likely it is that, the development path will follow the "normal" course (as in East Asia) with wages rising with capital accumulation and growth. One way to raise the rate of capital accumulation is to supplement domestic savings (which in any case are low in Latin America by East Asian standards - see Lal and Wolf (1986), Edwards (1997)) through foreign borrowing. Most Latin American countries have attempted to do so. However imprudent this course may seem, with hindsight, its temptations for the harried factional states of Latin America are obvious. As are the consequences of the closing of these borrowing opportunities. The worsening in the terms of borrowing being partly due to the development policies which were part and parcel of the same imperatives which compelled over-borrowing.

A second illustration of the uses of the model concerns the political economy of Latin American inflation. Even if it is true as Hirschman argues that, the "infla-tion cum-over valued exchange rate" policies were initially adopted as an expedient for taxing the agricultural sector, it is less plausible that the wild inflationary component of these policies still serves the same purpose. For many developing countries have maintained overvalued exchange rates and a host of other micro economic distortions which have the net effect of indirectly taxing the agricultural sector. But these policies have not been accompanied by Latin American rates of inflation. Thus the inflation tax is not likely to be an integral part of the arsenal to tax the agricultural sector. In a high inflation environment, the inflation tax will ultimately be borne by those who cannot take pre-emptive action. These are invariably the poorest, who rely purely on their labour, and do not have sufficient financial assets to use the various ingenious devices that are increasingly available to richer citizens, to allow them to substitute some indirect or direct form of foreign currency based assets for domestic money as a store of value (including, of course, capital flight!). The inflation tax is therefore better viewed as an indirect method of legislating the cut in the real wage that may be periodically required on some of the "land and natural resource abundant" economy's paths of development.

In this context it is interesting to consider the consequences for our Latin American factional state if it could not periodically -and at least for a brief moment- reconcile the dynamic distributional conflict it has inherited from factor endowments increasingly at odds with the polity 29. There is one example where the country's geography in effect imposes a constraint on the levying of the inflationary tax - Mexico. This case is also of interest because it provides an example of an exception to the story of the typical Latin American factional state we have been trying to develop in this section. Because of its contiguous border with the US, -a country with a strong currency and no exchange controls- it is virtually impossible for Mexico to levy the inflation - tax for any considerable period. For currency smuggling across contiguous borders is likely to be easier than smuggling goods and people (if for no other reason than bulk). High inflation in Mexico sustained over a number of years would lead to the virtual dollarisation of the economy. The non - inflationary policies of postwar Mexico, the absence of exchange control and the relative stability of the peso are not therefore completely inexplicable in the Latin American context.

The Mexicans, however, like all the other land and natural resource rich Latin American polities have to face the incipient and actual social conflict which is inherent in the efficient and growth promoting development path charted in fig. 2. The Mexicans have succeeded, partly because of historical accident, in bucking the Latin American trend by establishing and maintaining, what I like to call, a "bureaucrat maximizing" predatory state. The PRI provides an elaborate spoils system to buy out, co-opt, and coerce those who might seek to undermine its hegemony. The secret lies in the public sector using various direct (through the expansion of parastatal enterprises) and indirect methods (though creating microeconomic distortions), to garner a large share of the rents which accrue in a land and natural resource rich economy. These are then used to buy off discontent and create a political consensus legitimizing PRI hegemony, whose disruption might lead to the conversion of the polity in to, perhaps, a populist factional state.

The reduction of the resources available to the State to finance the enhanced entitlements created by Lopez-Portillo following the oil boom, put this method of mediating the possibly inherent conflict between "growth and equity" present in most land rich economies, under considerable strain. With the recent financial crisis accompanying the political opening, and the gradual erosion of support for the PRI it is unlikely that the Mexican predatory states' method of suppressing this conflict is viable over the longer run.

(2) Why Reforms?

What then explains the recent move to economic liberalism in Latin America? In answering this question it is useful to note the similarities between the motives for mercantilism in post Renaissance Europe and the reasons for its collapse and what has happened in the neo-mercantilist states of our own time.

Economic nationalism is the root cause of the widespread dirigiste impulse in the Third World. This echoes the nationalist impulse of the absolutist monarchies after the Renaissance, which also sought to build cohesive nation states through dirigiste policies (see Hecksher (1955)). Their mercantilist system of controls and regulation, set up to pursue the objective of nation building, has obvious resonances in the neomercantilist policies pursued by most Third World states. But these absolute monarchies found that, after some initial success, far from building nations, their dirigiste policies were becoming counterproductive, and leading to national disintegration. The consequences of their regulations and controls , particularly of internal trade and industry, were similar to those observed in many developing countriescorruption, rent seeking, tax evasion and the growth of illegal activities in growing underground economies.

As has been observed for many developing countries (see Lal (1987), Lal-Myint- (1996)), one paradoxical dynamic effect of attempts to exercise political control over everincreasing areas of economic life has often been that, after a certain stage, there is a diminution of the government `s effective areas of controlan unMarxian withering away of the State as private agents find numerous ways of avoiding them. The most important sign of this for any state is loss of control over its fiscal affairs. With the expansion of politically determined entitlements as part of the mercantilist system of a politicized economy, the accompanying tax burden to finance them leads at some stage to generalized tax resistance. Faced with inelastic or declining revenues but burgeoning expenditure commitments incipient or actual fiscal deficits become chronic. A joint balance of payments, fiscal and inflationary crisis results, and if it is extreme enough it leads to reform. This is borne out by Rajapatirna et al. (1997, Table 2) who show how the attempts to deal with these crises by dirigiste means failed in a number of Latin American countries, until the exploding crises in the late 1980's led to reforms. These reforms are aimed to regain government control over increasingly ungovernable economies. Economic liberalization was undertaken in the great Age of Reform in the 19th century for similar reasons. Paradoxically, as Hecksher notes, then the new economic liberalism achieved the goal sought by mercantilism:

"Great power for the state, the perpetual and fruitless goal of mercantilist endeavor, was translated into fact in the 19th century...The result was attained primarily by limiting the functions of the state, which task laissezfaire carried through radically. The maladjustment between ends and means was one of the typical features of mercantilism , but it disappeared once the aims were considerably limited. Disobedience and arbitrariness, unpunished infringements of the law, smuggling and embezzlement flourish particularly under a very extensive state administration and in periods of continually changing ordinances and interferences with the course of economic life. It was because the regime de l'ordre bore this impress that disorder was one of its characteristic features" (Hecksher, p. 325).

The same process occurred in the neomercantilist states of all three worlds, which led to a new worldwide Age of Reform in the 1980s.


But will these reforms be sustained, particularly in Latin America? As Edwards (1995, Table 1.3) has shown in those countries where the reforms are the deepest, the benefits are clearly evident. Particularly in Chile, where they have survived a turn in the political wheel, it is fair to say that they are now irreversible. This cannot yet be said of most other Latin American countries, though there are some favorable auguries.

The most important, given the "universalist" mindset of the Continent's medieval Catholic heritage is the conversion of the agency which has been the keeper of the economic conscience of the continent- ECLA. Its conversion means that economic orthodoxy is now in favour of reform and economic liberalism in the Continent. But, as in the earlier period when the LIEO was undermined, there is no assurance that populism might not in the future rear its ugly head and undermine this consensus. The continuing inequalities in the Continent _as much the result of its colonial heritage than of liberal economic policies_ combined with its cosmological beliefs emphasizing equality, could (as argued in section I) once again be used by populist political entrepreneurs for dirigiste ends, if the open economy does not deliver rapid growth.

The recent travails of Mexico and the continuing worries in Argentina are related to the management of exchange rates. Latin America is likely to have more volatile real exchange rates than East Asia. This is because the supply elasticity of the bulk of its exports based on natural resource is likely to be lower than for manufactured exports (as from East Asia) -as land is fixed but manufacturing capital is not. This means that fixed nominal exchange rates can lead to volatility in domestic wages and prices. Whatever the merits of using the nominal exchange rate as a nominal anchor to break hyperinflationary expectations (as in Argentina's currency board system), given the need for real exchange rate flexibility, a fixed nominal rate combined with rigidities in domestic money wages can lead to unemployment and social unrest- as in Argentina after the Mexican crisis. The latter in turn was due to the combination of loose fiscal policy and a fixed exchange rate. It is difficult to judge if Argentina will be able to either achieve sufficient domestic wage-price flexibility to make its currency board sustainable in the long run, or be able to adopt a floating nominal rate without igniting hyperinflationary expectations.

Mexico, by contrast seems to be moving to a new equilibrium with the adoption of a floating exchange rate, conservative fiscal and monetary policies, and the political opening by President Zedillo which is likely to bring an end to the PRI's near century of political domination. Whether the move to a factional democratic state will lead to populism and the dissipation of Mexico's natural resource rents remains to be see. In Brazil, whose problem remains primarily a fiscal one, the recent acceptance by the legislature of two important reforms - of social security and the civil service- should help in dealing with this thorny problem. It is worth noting that after being stalled for three years in Parliament, President Cardoso finally succeeded in obtaining these reforms only when Brazil seemed to teeter from the backwash effect of the S.E. Asian financial crisis of late 1997.

For many countries in the region spanned by the continent's mountain backbone there is a graver danger -of the very destruction of the State- posed by the US inspired war on drugs. This is criminalising and corrupting many of the polities in this region. It might be time for Latin America to take a united stand against what should be a US not Latin American problem. Theory suggests that even for trade in "bads", there is no case for protection or the equivalent voluntary export restraints- which is what enlisting Latin American suppliers in this American war amounts to30. Thus if the US bans drugs, it is a domestic US problem to police this within its own borders. As long as there is a demand for the "bad" in the US, welfare in the US and the world is best served by the most efficient suppliers providing the "bad". Attempting to control supply world wide is ineffectual given the huge profits available from the trade31. These currently illegal profits not only corrupt the producing countries polity but increasingly the world's financial system through which they are laundered. It maybe time for Latin Americans jointly to refuse to participate any longer in the US war on drugs.

Most seriously, for the future of reform in Latin America (as we noted in Section I) its polities, despite their current embrace of economic and political liberalism, have not given up their attachment to the State seen as an enterprise association. The current movements in many countries to strengthen domestic legal systems and to reinvigorate civil society are laudatory, but given the long standing Catholic tradition of natural law and the corporatist patrimonial State which embodies it, it is not evident that the resulting view of the State seen as a civil association has much resonance yet in Latin America. One hopeful sign that some see for a change in cosmological beliefs is the growing strength of the Protestant evangelical churches in the Continent (see Veliz (1994)). But without a change in cultural attitudes there is a continuing danger that the current period of economic liberalism will, as in the 19th century, be a mere interregnum before Latin America veers back to its old dirigiste, corporatist ways- influenced as much by its Iberian cultural and political heritage as its factor proportions.

* James S. Coleman, Professor of International Development Studies, University of California, Los Angeles. This paper is based on Lal-Myint (1996), Lal (1998) and Lal (1998a).

1 See P. Colinvaux (1983), The Fates of Nations, Penguin, London. This was the definition adopted in Lal (1988) and in Lal (1998).

2 F. Hahn (1973), On the Notion of equilibrium in economics, Cambridge University Press, Cambridge.

3 See Lal (1998) for details.

4 A patrimonial state is defined and distinguished from feudalism by Max Weber, as a form of traditional rule where government is an extension of the royal house. It is the rule of one aided by trusted advisers. Whilst feudalism consists of rule by a privileged few. Paz notes: "In a patrimonial regime the central authority must strive to prevent the excessive growth of an independent aristocracy of landowners who enjoy hereditary privileges. In New Spain, a distrustful central authority assiduously monitored the activities of the criollos" (p. 20).

5 On this interpretation of the origins and persistence of the caste system in India see Lal (1988). For Russia, Domar notes that the use of serfdom to deal with the scarcity of labour to provide a surplus for the nobility is attested by two facts: "the first ..[is] the replacement of the basic land tax by a household tax in the 17th century and a poll tax under Peter the Great. The second is ...[a] cultural trait..: as late as in the first half of the 19th century, the social position of a Russian landowner...depended less on his landholding.than on the number of souls (registered male peasants) he owned". (Domar (1970), p. 232).

6 See Engerman and Sokoloff (1994), Barbara l. Solow (1991). For evidence on the substantial economies in producing certain crops on large slave plantations see Fogel (1989, Engerman (1983) and Deer (1949).

7 See Kupperman (1993).

8 Oakeshott makes an important distinction between two major views of the State in Western thought going back to the Greeks and the Judaeo-Christian tradition. The former viewed the State as a civil association which was the custodian of laws which do not seek to impose any preferred pattern of ends, but which merely facilitate individuals to pursue their own ends. The latter viewed the State as an enterprise association, which was the manager of an enterprise using the law for its own substantive purposes, and in particular to legislate morality.

9 But there was also a more positive aspect of the common Catholic culture of Latin America. As compared with North America, as Hugh Thomas has rightly pointed out to me, the Latin's succeeded "socially" where the US `failed'. This was in part due to the planned cities, the skilful appraoch to racial matters enabling manumission of slaves, and of course a common Catholic culture which left their personal and social mores closer to the "communalist" ones of the other great ancient civilisations than those based on the new fangled "individualism" of their Protestant brethren (see Lal (1998)).

10 Veliz (1994) contrasts this universalism of Latin with the greater tolerance of diversity in beliefs in North America as between the Baroque hedgehogs of the South and the Gothic foxes of the North in the New World, reflecting Archilochus's maxim: "the fox knows many things, but the hedgehog knows only one thing" (quoted in Berlin (1978).

11 See for instance Castañeda (1995).

12 See Edwards (1995) for an excellent account of the post debt crisis developments in Latin America. For Chile's move to the market see Edwards and Edwards.

13 But note that, because of Arrow's impossibility theorem, even the majoritarian democratic version of the factional State will not become a benevolent State, because of the lack of a majoritarian social welfare function!

14 It has made various technical errors in its economic policy. An excellent and balanced account of the Chilean economy, post-Allende, is in Edwards and Edwards (1987).

15 This contrast is also noted and used to explain the different outcomes in the Northern and Southern parts of the land abundant Americas by North (1989).

16 Veliz (1980) argues that the rhetorical espousal of economic liberalisation by the military, was the result of their desire to curry favour with their social superiors - the upper class business and land owning groups - "their new upper-class friends who hold the key to their social advancement (their wives! and families as well as their own)" (p. 299). He finds the upper-classes in Latin America as being "culturally dependant" on the West. The constituents of these upper classes are the intellectual and old "oligarchs". He writes "the bastions of Latin American cultural dependence are well-manned by the radical intellegensia and the upper class business and land owning groups. The loyalty that the former have demonstrated to the doctrines of violent revolutionary action, the latter have shown to the tenets of free - trading liberalism their forefathers embraced during the prosperous days of the liberal pause" (Veliz, ibid, n. 22. p. 299).

In Lal (1978) I have also argued (in explaining the North-South confrontation of the 1970's) that, the similarities between Latin American and other Third World elites in seeking self respect, lie in the continuing sense of inferiority felt by "white" Latin American elites towards the dominant power and life styles of "white Anglo-Saxons".

17 On the political economy of liberalisation see Lal (1987).

18 This section is based on Lal (1989) and Lal-Myint (1996).

19 The endowment triangle "represents graphically the relative endowments of three factors (physical capital, labour, arable land). In the three-dimensional factor space, straight lines emanating from the origin contain all endowment vectors with the same ratio of factors. These lines in three dimensions can be represented by points in two dimensions intersecting the positive orthant with a plane to form the endowment triangle. The three co-ordinates in the three-dimensional factor space are represented by the corners of an endowment triangle, and the endowment vectors of each of the countries are represented by points. The main point to keep in mind is that every endowment point on a straight line emanating from one corner of the triangle has the same ratio of the other two factors. For this reason the scales of the three factor ratios can be placed on the edges of the triangle" (Leamer (1987) p. 963-4).

20 Also see Deardoff (1984) for a geometric presentation of Krueger's model.

21 For a model with non traded goods see leal (1986).

22 These regions of specialisation will depend upon commodity prices, and will alter as commodity prices change.

23 It might seem paradoxical that, whilst the economy's capital - labour ratio is rising it is falling in manufacturing. But remember that the rate of growth of labour for the economy is not the same as in manufacturing. Then it is possible for the agricultural labour force to grow more slowly (because of fixed land) than for the economy as a whole, thereby allowing and requiring the labour force in the manufacturing sector to grow more rapidly than for the economy as a whole. Thus a rising capital labour ratio for the economy as a whole can be associated with a falling capital labour ratio in manufacturing. Of course there will be some rate of capital growth at which the capital labour ratio for manufacturing will also be rising along with that for the economy as a whole, and this paradoxical development path would not occur.

24 This is the basis for Krueger's statement that: "First, the distinction between poor and underdeveloped countries emerges clearly from the model, A poor country is one with an unfavourable land-man endowment. An underdeveloped country is one with a relatively small endowment of capital per person. An underdeveloped country, however, could conceivably have a higher per capita income and real wage than a more developed but poor country. Second, a country abundantly endowed with land and therefore with a relatively high wage would not necessarily have a comparative advantage in labour-intensive manufactures even in its early stages of capital accumulation: the real wage at which persons would leave agriculture might be too high. In such an instance, the capital-labour ratio in manufacturing would be higher in the early stages of development in a poorer country, whilst the output per unit of capital and the rate of return on capital would be lower than in the low-wage country" (Krueger (1977) p. 15).

25 This follows from the simple observation that with the tariff on M4, producers will attempt to expand the output of M4. As it is the most labour intensive good being produced, an expansion in its output will create excess demand for labour, and excess supply of the land and capital used in producing the other two goods. Full employment of the three factors will only be maintained as producers switch to using more land and capital intensive techniques, which save on labour producing the other two goods M5 and M..

26 See Deardoff (1984) p. 740.

27 This can also be seen in terms of Fig. 2, but this exercise is left to the reader! In an earlier paper, see Lal (1986a) I have used a similar model to explain the long term decline in real wages in the Philippines - another "land abundant" country, whose political economy is closer to that of Latin America than its East Asian neighbours. Lal and Maxfield (1993) apply the model to Brazil and show how its various "big pushes" have been motivated by an incipient decline in real wages.

28 A model of these for the factional state is provided in Lal (1986).

29 This dynamic distributional conflict is modelled more explicitly in Lal (1986), which provides an explicit role for the real exchange rate, and the non-traded good sector -an aspect I have ignored in this paper.

30 See Krueger (1998). The central theorem of the modern theory of trade and welfare can be applied to show that the "bad" in drugs is a external diseconomy in consumption, which being a domestic distortion requires a domestic tax (if necessary of 100%) as a corrective but no interference in trade or production.

31 Krueger (1998) provides estimates that it is currently worth $400 billion which is about 8 per cent of the value of legal world trade. In Latin America in Bolivia, Colombia and Peru between 600 thousand to 1.5 million are directly employed in the drug trade. The labour force in each of the above countries is 5, 28 and 17 million respectively. Some idea of the immense profits to be made in this illegal trade can be gleaned from the following data in Krueger (1998). 1 kg. of Bolivian coca base costs between $650-1000, when processed into cocaine it sells for $900- 1200. Sold wholesale in the US it fetches $13,000-40,000, and then retails at $17,000- 172,000.


Anderson, B. (1991), Imagined Communities, Verso, London.         [ Links ]

Balassa, B. (1985), Change and Challenge in the World Economy, Macmillan, London,1985.         [ Links ]

Baumol, W.; J.C. Panzar, R.D. Willig (1982), Contestable Markets and the Theory of Industry Structure. Harcourt Bruce Jovanovich, San Diego.         [ Links ]

Berlin, I. (1978), Russian Thinkers, London.         [ Links ]

Castaneda, J.G. (1995), The Mexican Shock, New Press, New York.         [ Links ]

Castaneda T. (1992), Combating Poverty, ICEG, ICS PREss, San Francisco.         [ Links ]

Collier, D. (ed.) (1979), The New Authoritarianism in Latin America, Princeton, New Jersey.         [ Links ]

Corbo, V. y J. de Melo (1985), "Overview and Summary: Liberalization and Stabilization in the Southern Cone of Latin America", World Development 13, Nº 8, 1986.         [ Links ]

Deardoff, A. (1984), "An Exposition and Exploration of Krueger's Trade Model", Canadian Journal of Economics, Vol. 17, November.         [ Links ]

Deer, N. (1949), The History of Sugar, Chapman and Hall, London.         [ Links ]

Domar, E. (1970), "The causes of slavery or serfdom: A hypothesis", The Journal of Economic History, vol. XXX, March, pp 18-32, reprinted in his Capitalism, Socialism and Serfdom, Cambridge University Press, Cambridge.         [ Links ]

Edwards, S. (1995), Crisis and Reform in Latin America, Oxford Univ. Press, New York.         [ Links ]

Edwards, S. y A. C. Edwards (1987), Monetarism and Liberalization - The Chilean Experiment, Ballinger, Cambridge, Mass.         [ Links ]

Edwards, S. y S. Teitel (1986), "Introduction to Growth, Reform and Adjustment: Latin America's Trade and Macroeconomic Policies in the 1970's and 1980's Economic Development and Cultural Change, vol. 34, Nº 3 April.         [ Links ]

Engerman, S.L. (1983), "Contract Labor, Sugar, and Technology in the 19th century", Journal of Economic History, vol. 43, pp.635-59.         [ Links ]

Engerman, S.L. y K.L. Sokoloff (1994), "Factor Endowments, Institutions and Differential Paths of Growth Among the New World Economies: A view from Economic historians of the United States", NBER Working Paper, Historical Paper Nº 66, Cambridge Mass.         [ Links ]

Fogel, F.W. (1989), Without Consent or Contract, Norton, New York.         [ Links ]

Hirschman. A. (1968), "The Political Economy of Import-Substituting Industrialization in Latin America". Quarterly Journal of Economics, Vol. LSSS11, Nº 1. February.         [ Links ]

Hirschman, A. (1979), "The Turn to Authoritarianism in Latin America and the Search for Its Economic Determinants", in D. Collier (ed.).         [ Links ]

Hirschman, A. (1981), "Policy making in Latin America", in his Essays in Trespassing, Cambridge University Press, Cambridge.         [ Links ]

Kindleberger, C. (1967), Foreign Trade and the National Economy, Yale, New Haven.         [ Links ]

Krueger, A. (1977), "Growth, Distortions, and Patterns of Trade among Many Countries", Princeton Studies in International Finance, Nº 40, International Finance Section, Department of Economics, Princeton University, Princeton, New Jersey.         [ Links ]

Krueger, A. (1998), "International trade in `bads'" in H. Giersch (ed): The Merits of Markets, J.C.B. Mohr, Tubingen.         [ Links ]

Kupperman, K. O. (1993), Providence Island, 1630-1641: the Other Puritan Colony, Cambridge University Press, Cambridge.         [ Links ]

Lal, D. (1978), Poverty, Power and Prejudice, Fabian Research Series 340, Fabian Society, London, reprinted in Lal (1994).         [ Links ]

Lal, D. (1983, 1985, 1997), "The Poverty of Development Economics" Hobart Paperback Nº 16. Institute of Economic Affairs, London 1983, 2nd rev. edtn. American edition, Harvard University Press, Cambridge, Mass.         [ Links ]

Lal, D. (1984), "The Political Economy of the Predatory State" Development Research Department, Discussion Paper 105, Washington D.C. Processed. (Revised April 1988) University College, London).         [ Links ]

Lal, D. (1986), "The Political Economy of Industrialization in Primary Product Export Economies", in N. Islam (ed): The Balance between Industry and Agriculture in Economic Development, vol. 5. IEA, Macmillan, London (1989), reprinted in Lal (1993).         [ Links ]

Lal, D. (1986a), "Stolper-Samuelson-Rybczynski in the Pacific-Real Wages and Real Exchange Rates in the Philippines, 1956-1978", Journal of Development Economics, vol. 21, April 1986, reprinted in Lal (1993).         [ Links ]

Lal, D. (1987), "The Political Economy of Economic Liberalization", World Bank Economic Review, vol. 1, Nº 2. January, reprinted in Lal (1993).         [ Links ]

Lal, D. (1988), The Hindu Equilibrium, Clarendon Press, Oxford.         [ Links ]

Lal, D. (1989), "After the Debt Crisis: Modes of development for the longer run in Latin America" in S. Edwards and F. Larraín (eds) (1993), Debt, Adjustment and Recovery, Blackwell, Oxford, reprinted in Lal (1993).         [ Links ]

Lal, D. (1993), The Repressed Economy, Economists of the 20th century series, Edward Elgar, Aldershot.         [ Links ]

Lal, D. (1994), Against Dirigisme, ICEG, ICS Press, San Francisco.         [ Links ]

Lal, D. (1998), Unintended Consequences: the impact of factor endowments, culture and politics on long run economic performance, the Ohlin lectures, MIT Press, Cambridge Mass.         [ Links ]

Lal, D. (1998a), "The Political Economy of Reform in Latin America", R. Skidelsky (ed): The Politics of Economic Reform, Profile Books, London.         [ Links ]

Lal, D. y S. Maxfield (1993), "The political economy of stabilisation in Brazil", in Robert H. Bates y Anne O. Krueger (eds): Political and Economic Interactions in Economic Policy Reform, Blackwell, Oxford, reprinted in Lal (1993).         [ Links ]

Lal, D. y M. Wolf (eds.) (1986), Stagflation Savings and the State -Perspectives on the Global Economy, Oxford University Press, New York.         [ Links ]

Lal, D. y H. Myint (1996), The Political Economy of Poverty Equity and Growth, Clarendon Press, Oxford.         [ Links ]

Leamer, E. (1984), Sources of International Comparative Advantage, MIT Press, Cambridge, Mass.         [ Links ]

Leamer, E. (1987), "Paths of Development in the Three-Factor, n-Good General Equilibrium Model", Journal of Political Economy, Vol. 95, Nº 5, October.         [ Links ]

Lipton, M. (1977), Why Poor People Stay Poor, Temple Smith, London.         [ Links ]

Little, I. (1982), Economic Development, Basic Books, New York.         [ Links ]

Lynch, J. (1973), The Spanish American Revolutions , 1808-1826, Norton, New York.         [ Links ]

Mayer, W. (1984), "Endogenous Tariff Formation", American Economic Review, December.         [ Links ]

Mitra, P. (1986), "A Description of Adjustment to External Shocks: Country Groups" in D. Lal and M. Wolf (eds.).         [ Links ]

Morse, R. M. (1964), "The Heritage of Latin America" in L. Hartz (ed): The Founding of New Societies, Harcourt, Brace and World, New York.         [ Links ]

North, D. (1989), "Institutions and Economic Growth- An Historical Introduction", World Development, 17/9, Sept., pp. 1319-32.         [ Links ]

O'Donnell, G. (1973), Modernization and Bureaucratic-Authoritarianism: Studies in South American Politics, Politics of Modernization Series Nº 9, Institute of International Studies, University of California, Berkeley.         [ Links ]

Oakeshott, M. (1993), Morality and Politics in Modern Europe, Yale, New Haven.         [ Links ]

Olson, M. (1971), The Logic of Collective Action, Yale, New Haven.         [ Links ]

Olson, M. (1982), The Rise and Decline of Nations, Yale, New Haven.         [ Links ]

Paz, O. (1988), Sor Juana, Harvard, Cambridge Mass.         [ Links ]

Sachs, J. (1985), "External Debt and Macroeconomic Performance in Latin America and East Asia", Brookings Papers on Economic Activity Nº 2.         [ Links ]

Rajapatirana, S., L. M. de la Mora y R. A. Yatawara (1997), "Trade Policy Reforms in Latin America", The World Economy, vol. 20, Nº 3, p. 307-38.        [ Links ]