(for reading only; not to be reproduced)



Obstacles to Economic Growth in

Nineteenth-Century Mexico



JOHN H. COATSWORTH



The American Historical Review (83:1:80-100) (Feb. 1978)



THE ECONOMIC HISTORY OF MEXICO HAS ADVANCED dramatically in recent years. The most important revisionist work has concentrated on the eighteenth and nineteenth centuries, although major works have appeared on the sixteenth, seventeenth, and twentieth centuries as well.(1) Much of the new research has been quantitative or comparative, linking the study of Mexico's past to methodological currents in the United States and Western Europe.(2) It is now possible to revise some of the principal hypotheses about Mexico's relative backwardness since the end of the colonial era. Comparative estimates of Mexican gross national product in the nineteenth century can be used to assess the impact of revisionist work on standard interpretations of Mexico's economic history from the Bourbons to the Revolution of 1910

By way of introduction, two caveats are worth stressing. First, spurious precision infects all, or nearly all, quantitative historical work. Margins of error are necessarily large in estimates of national product based on fragmentary data, the guesswork of interested contemporaries, and the imagination of the modern researcher. This is even truer of estimates produced to test hypotheses about what might have been, as the "new economic history" has amply demonstrated during the past decade or so in the United States. I take it as a duty, therefore, to issue a clear warning: all of the numbers in this article are, without exception, inaccurate; however, that is not a valid argument against their use. Literary estimates typically contain fewer errors (we all know that Mexico was "poorer" than its northern neighbor), but only because they specify a ridiculously wide range of values. Numerical estimates have the virtue of inspiring controversy and thus may lead to the compilation of more accurate statistics. As Charles Gibson reminds us, even failures can prove fruitful for the advancement of historical knowledge.

Second, the precision of historiographical essays can be equally spurious. Precision requires restatement, the deliberate partitioning of what can be rejected from what was really intended. To set aside a well-defined error is easier than to modify a complex interpretation that rests on many legs at once. Separating what makes sense from what does not inevitably does violence to interpretations that contain elements of both or that rest on analytical distinctions the reviewer does not find important.

ESTIMATES OF MEXICAN NATIONAL INCOME reveal expected trends. Per capita as well as total income fell until sometime after 1860. A recovery began during the Restored Republic (1867-76), but Mexico did not surpass colonial income levels until well into the Porfiriato (1877-1910). Between 1877 and 1910 national income per capita grew at an annual average rate Of 2-3 percent extremely rapid growth by world standards, so fast indeed that per capita income more than doubled in thirty-three years.

In Table I, I have assembled estimates of national income for Mexico, Brazil, Great Britain, and the United States at selected years from 1800 to 1910 (These are the only countries for which such estimates are available for the entire nineteenth century.) The comparative data show that national income per capita in Mexico was closer to that of Great Britain and the United States in 1800 than at any point thereafter. In that year Mexico produced more than a third of British income per head and nearly half that of the United States. The gap in productivity between the Mexican economy and that of the advanced countries of the North Atlantic has never been so small. By 1877, Mexico's per capita income had fallen to a little over one-tenth that of the industrial nations. It has fluctuated between 10 and 15 percent of U.S. per capita income ever since. The comparison with Brazil shows a different pattern. At the beginning of the century, Mexico's per capita income was nearly 20 percent above that of Brazil. While Mexican productivity fell during most of the century, Brazil's increased and surpassed that of Mexico. During the Porfirian period, however, the Mexican economy grew very rapidly and by 1910. Mexico's per capita income was 40 percent higher than Brazil's. Today, the difference between these two countries stands about where it was in 1800.(3)

TABLE 1. NATIONAL INCOMES, 1800-1910

A. PER CAPITA INCOME

YEAR Mexico Brazil Great Britain United States

1800 73 62 (118) 196 (37) 165 (44)

1845 56 72 ( 78) 323 (17) 274 (20)

1860 49 77 ( 64) 370 (13) 359 (14)

1877 62 83 ( 75) 497 (12) 430 (14)

1895 91 89 (102) 745 (12) 735 (12)

1910 132 94 (140) 807 (16) 1035 (13)

B. TOTAL INCOME (MILLIONS)

YEAR Mexico Brazil Great Britain United States

1800 438 198 (221) 2094 (21) 858 (51)

1845 420 510 ( 82) 6293 ( 7) 5493 ( 8)

1860 392 778 ( 50) 8510 ( 5) 10900 ( 4)

1877 613 1115 ( 55) 16690 ( 4) 21629 ( 3)

1895 1146 1633 ( 70) 27930 ( 4) 50754 ( 2)

1910 2006 2129 ( 94) 36556 ( 5) 95201 ( 2)

NOTE: Parenthetical numbers show Mexican income as a percent of each estimate. The estimates have all been computed in U.S. dollars of 1950

SOURCES: Data and estimating procedures for Mexican national income available from author on request. Note that the 1845 and i 860 figures for Mexico represent the lower of two alternative estimates constructed for these years. In both cases, the alternative estimates are 5 percent higher than those in the table. For Brazil, see Nathaniel H. Leff, "A Technique for Estimating Income Trends from Currency Data and an Application to Brazil," Review of Income and Wealth, 5 (1972): 355-68. Leff calculated that Brazil's national income grew at an annual average rate Of 0.4 percent between 1822 and 1913. Values in the table are extrapolated back from an estimated U.S. $98 in 1920-25 at this rate. For Great Britain, see Phyllis Deane and W. A. Cole, British Economic Growth, 1688-1.959: Trends and Structure (Cambridge, 1962), 282, 329-30. The initial estimate for 18oo employs the Deane and Cole figure Of L232 for 18oi, converted to 1950 dollars by using the Warren Pearson and Bureau of Labor Statistics (WP/BLS) wholesale price index in U.S. Bureau of the Census, Historical Statistics of the United States from Colonial Times to 1957 (Washington, 1958), 115-17. Subsequent British estimates employ the Deane and Cole growth rates applied to the initial figure. The estimates for 1800 1845, and 1860 are for Great Britain alone; the rest are for the United Kingdom (which includes Ireland). The U.S. estimates for 1800, 1845, and 1860 are based on Paul A. David, "The Growth of Real Product in the United States before 1840: New Evidence, Controlled Conjectures," journal of Economic History, 27 (1967): 151-97. David's estimates of real growth rates between 1800 and 186o were used to extrapolate his estimate of U.S. gross national product in 1840, adjusted to 1950 dollars with the WP/BLS index. For 1877, 1895, and 19 1 0 a similar adjustment to 1950 dollars was made in the national income estimates in U.S. Bureau of the Census, Historical Statistics, 139. Note that the 1877 figure is an average estimated for 1877-81 and the 1895 estimate is an average for 1892-96.

In terms of total size, the contrast between Mexico and the industrial nations, especially the United States, is even sharper. In 1800, Mexico produced more than half as many goods and services as the United States. By 1877, Mexico produced only 2 percent of the output that flowed from the factories, farms, and warehouses of the northern colossus and only 5 percent of total British output. Population growth accounts for a large portion of the difference between Mexico and the two industrial economies. In 1800, Mexico's population of six million was larger than that of the United States (just over five million) and over half that of Great Britain (almost eleven million, excluding Ireland). In 1910, Mexico's population stood at fifteen million, the United Kingdom at forty-five million, and the United States at ninety-two million. Even Brazil, whose population in 1800 was only somewhat more than three million, passed Mexico during the nineteenth century and reached over twenty-two million by 1910. Both the United States and Brazil received large numbers of immigrants, especially after 1890. Mexico attracted very few immigrants, while high death rates kept down the natural increase in population until the 1930s.(4)

Although the estimates in Table I may contain errors, the relative position of the four countries and the magnitude of the gaps among them are not likely to be affected by better data. For an analysis of Mexico's relative backwardness in the nineteenth century, two aspects of the comparison stand out. The first is the substantial difference between the economies of Mexico and the United States at the beginning of the century. In 1800 the United States was still a predominantly agrarian country, decades away from its industrial revolution. Mexico in the same year counted as Spain's richest colony in the New World, with an advanced mining industry exporting vast quantities of processed metals. Indeed, the values of U.S. and Mexican exports at the end of the eighteenth century were quite similar.(5)

Why did the Mexican economy begin the nineteenth century less than half as productive as that of the United States? The second striking aspect of the data is the extent to which the gap between Mexico and the industrializing countries widened between 1800 and the last quarter of the century. Had Mexico's economy kept pace with the growth of the United States for the entire century, Mexico would have reached its 1950 level of per capita income before the Revolution of 1910. Had the gap between Mexico and the United States remained the same from 1800 to the present, Mexico would now rank among the world's industrial powers. From the standpoint of the twentieth century, the question might be reversed. Why did the Mexican economy fall so far behind the industrializing giants of the North Atlantic during the nineteenth century?

THREE MAIN OBSTACLES TO ECONOMIC GROWTH have been postulated to explain Mexico's relative backwardness at the end of the colonial period: Spanish colonial rule, the system of land tenure, and the Roman Catholic Church. On the basis of evidence of direct economic effects, regardless of more general problems, all must be rejected.

First, since Spain managed, or mismanaged, its New World colonies for three full centuries, it cannot be denied that Spain was responsible for whatever the colonies achieved or failed to achieve--a truism that has been repeated by scholars and historians ever since Baron Alexander von Humboldt made it popular in his monumental studies of the Spanish colonies in the first decade of the nineteenth century. But emancipation from Spain promised few benefits. For purposes of measurement, the costs of Spanish colonialism are defined as those economic constraints that independence actually eliminated: (1) mercantilist restrictions on direct trade with foreign countries and (2) the uncompensated export of gold and silver extracted by the colonial government as net fiscal revenues. (This definition leaves for later discussion those effects of Spanish colonialism that lingered on after independence.) There are two questions for which precise answers may be sought. How much would the Mexican economy have gained if independence from Spain had been achieved by the end of the eighteenth century? And how much of the gap in productivity between the U.S. and Mexican economies would have been eliminated by these gains?

TABLE 2. COMPARISON OF THE COSTS OF BRITISH AND SPANISH MERCANTILISM



Colony Costs

(millions of pesos)

Costs Per Capita

(pesos)

Costs as

% of Colonial Income

BRITISH NORTH AMERICA (1775) 0.5 0.26 0.3
NEW SPAIN (1797-1820)
Trade Burden 7.2 1.2 3.0
Fiscal Burden 10.1 1.68 4.2
Total 17.3 2.88 7.2

NOTE: Colonial income for the thirteen British colonies is calculated as the 1800 per capita estimate (U.S. $go) multiplied by the 1775 population In current pesos of 1800 this comes to a total income of 173 million pesos. Note, however, that Thomas estimated the costs of British mercantilism in current dollars of 1775, before the inflation that began in the 1780s Thus, the magnitude of the British burden as a percent of total and per capita income is somewhat overstated. For New Spain, the estimates are annual averages for 1797 to 1820.

SOURCES: For British North America, see Robert Paul Thomas, "A Quantitative Approach to the Study of the Effects of British Imperial Policy upon Colonial Welfare," Journal of Economic History, 25 (1965): 615-38. For Mexican estimates, see text and notes 7-8.

Table 2 presents estimates of the answers. It compares the annual average cost of Spanish colonialism to Mexico (then called New Spain) between 1797 and 1820 with the cost of British colonialism to the thirteen North American colonies in 1775. The North American figures are taken from Robert Paul Thomas' well-known estimates,(6) which calculate the loss of colonial income due to British restrictions on direct trade (mainly with northern Europe and the foreign West Indies) and subtract from this loss certain benefits (bounties for the production of strategic materials, naval and frontier protection, and the like). The Mexican estimates have two components. The first is an estimate of colonial income lost through Spanish restrictions on direct foreign trade, similar to Thomas' estimates for the British colonies.(7) The second component is the fiscal burden-tax revenues collected by crown officials in Mexico and exported to subsidize Spanish administration of other colonies or to fill the king's coffers at Madrid.(8) Since the British raised no internal taxes, no such burden existed in the thirteen colonies.

As Table 2 indicates, the total cost of Spanish sovereignty came to a little more than seventeen million pesos per year. By contrast, the cost of British colonialism was only half a million pesos. The Spanish burden was, therefore, thirty-five times greater than that of the British. Table 2 also shows, however, that the Spanish burden amounted to less than three pesos per capita at the beginning of the nineteenth century. It is arguable, of course, whether the psychic splendors of Spanish citizenship were worth that much. In economic terms, fully 7.2 percent of the colony's 1800 income was lost, an amount by no means negligible. But even if this estimate were five or ten times too low, independence would not have eliminated the gap in productivity between the Mexican and U.S. economies in 1800. Adding three pesos to Mexico's per capita income would still have left it at less than half that of the United States.

Many Mexicans in the nineteenth century, and not a few historians thereafter, have wondered why independence failed to stimulate the Mexican economy. Many have sought the answer to this question in the turmoil of the struggle for independence and the political instability that followed. Certainly these phenomena helped to depress the economy after 1810, but Mexico's income did not increase dramatically because independence had direct adverse economic results, which more than offset the benefits. Indeed, the continued depression in the crucial mining industry that persisted well into the nineteenth century was largely the result of an unanticipated cost of independence: the loss of assured supplies of mercury--indispensable for processing low-grade ores--that Spain had provided at low, fixed prices from the huge state-owned mercury mine at Almaden.(9)

Second, a major obstacle to economic advance cited in the conventional accounts is the system of land tenure or, more precisely, the organization of Mexican agricultural production into large estates called haciendas This general explanation can be divided into a number of distinct assertions, of which at least two can now be rejected: (1) that the large estates of both the colonial period and the nineteenth century were inefficiently organized and badly managed and (2) that concentration of land ownership per se caused waste and misallocation of resources. Full discussion of the system of land tenure is impossible in so short a space. Fortunately, it is now possible to rely on a large body of hacienda studies, many of them completed within the past five years. Most of the estates studied were located in the Central Plateau from Oaxaca in the south to San Luis Postoí in the north. The records --including account books, correspondence, and property transfer-more than fifty large haciendas have been intensively studied, while twice that number have been examined in less detail. These records often covered more than a century of activity. The colonial era (especially the first and last half centuries of Spanish rule) have fascinated historians most, but recent work has concentrated on nineteenth-century estates.(10) In addition, a number of regional studies have been added to Charles Gibson's pathbreaking work on colonial Tlaxcala and the Valley of Mexico. These often contain the best information available on nonestate agriculture as well as data on labor conditions and the status of tenants and sharecroppers.(11)

Collectively, the hacienda and regional studies have transformed the traditional view of Mexican agriculture and estate management. Although primarily narrative descriptions of estate operations and organization, their impact has been as dramatic for Mexican history as the generation of econometric work on slave agriculture has been for the history of the southern United States. Not one estate owner has been found who might qualify as the sort of aristocratic, prestige-oriented, economic nincompoop once thought by many to be typical of Spanish American hacendados. Each was greedy in the ordinary way--even the managers of Church estates, for the income went for good works after all. Every one of them demonstrated a primordial desire to maximize income and to minimize production costs. During periods of prosperity, estate owners invested in their operations, experimented with new crops and new methods, and sought new markets. In periods of economic decline, they shifted from crops to livestock, reorganized their estates into tenancies, sold out to cut losses, or abandoned their holdings altogether. Given the relative costs of labor, capital, and especially management and supervision, their economic rationality was comparable to that of modern entrepreneurs. No evidence has yet been found to sustain the hypothesis that the estate sector of Mexico's agricultural economy wasted resources that might have been put to more productive use under different land-tenure conditions.(12)

The picture of colonial and nineteenth-century agriculture in Mexico that emerges from the evidence now available suggests two main conclusions. Estate agriculture enjoyed advantages not available to Indian villagers, small landowners, or tenant farmers: economies of scale, access to outside credit, information about new technologies and distant markets, a measure of protection from predatory officials, and greater security of tenure. But these advantages, important as they were, did not eliminate small-scale production because they were not sufficient to offset the high cost of recruiting and supervising labor. The large estates held a competitive advantage in the production of cattle, sheep, wool, food grains, pulque, sugar, and sisal. In other products that required very close supervision (or highly motivated workers) either to produce or to transport without great losses, the villages and small-scale producers held the advantage: fruits, garden products like tomatoes and chiles, silk, cochineal, small animals including pigs, poultry, eggs, and the like. Even cotton, tobacco, and wine were commonly produced by villagers and small landholders. Product specialization among units of varying size, location, and organization made Mexican agriculture more efficient than it would otherwise have been. The discipline of local and regional markets, moreover, acted to push the size of production units toward what a modern economist would describe as an "optimal mix." Far from distorting resource allocation in agrarian Mexico, concentration of landholding functioned to allow more efficient production of crops suitable for large units and did so without sacrificing the advantages of small-unit output for other produce.(13)

In the late nineteenth century the increasing availability of new transport and production technologies made the historic division of labor between estate and nonestate agriculture inefficient. The economic balance tipped dramatically in favor of the large haciendas. Agricultural productivity, stagnant during the half century after independence, increased rapidly as the estates expanded at the expense of the free villages and small landowners.(14) Additional evidence may be adduced from comparative data. In 1800, between 70 and 80 percent of the Mexican labor force worked in agriculture to produce approximately 40 percent of the colony's gross product. Almost identical are the best estimates for the United States in the same year: slightly more than 80 percent of the labor force in agriculture produced approximately 40 percent of the nation's income. The gap in agricultural productivity between the two regions was exactly equal to the gap in nonagricultural productivity between the two economies. Mexico was only half as productive as the United States in both agricultural and nonagricultural production.(15) In comparative terms, the agricultural sector of the Mexican economy did not, therefore, act as a drag on the nation's growth. Although this evidence is not decisive, it does at least suggest that historians should, in searching for obstacles that hindered Mexico's economy in the colonial era and the nineteenth century, look less at a single sector, important as it was, and more at conditions affecting the economy as a whole.

Third, the Roman Catholic Church purportedly retarded Mexico's economic growth during the colonial period and for at least three decades after independence, until the liberal revolution succeeded in expropriating the Church's wealth and reducing its role in the nation's political life. The anticlerical argument runs through volumes of highly charged prose and involves not only the Church's strictly economic activities but its political, social, and cultural influence as well. The portion of the argument that can now be rejected is that which assigns significance to the Church's strictly economic activities: (1) the tithe, (2) mortgage lending, and (3) ownership of real property.

The tithe was a 10 percent tax on gross output charged mainly on the agricultural and livestock production of the private estates. Like any direct tax today, the tithe reduced the profitability of agricultural enterprise. But by how much did this tax reduce Mexico's gross national product? It has often been implicitly assumed that agricultural production (and thus GNP) was reduced by the amount of the tax--that is, by 10 percent--as though the tithe collectors made off with a tenth of every harvest and burned it as a sacrifice. Even if this entirely inappropriate measure were adopted, the per capita drain of the tithe in 1800 would have amounted to less than half a peso--a little more than 1 percent of national income. By the time of independence (until 1833 when the tithe was abolished as a legal obligation of the citizenry), Church revenues from this source had already dropped to negligible sums.(16)

An appropriate measure of the impact of the tithe would have to take into account its effect on the profitability of private agricultural enterprise. By reducing profits in private agriculture, the tithe reduced employment and investment in that sector of the economy and pushed labor and capital into other, less productive activities. The negative effect on GNP is thus the difference between what the labor and capital pushed out of private agriculture actually produced elsewhere and what they would have produced (in the absence of the tithe) had they remained on the farms and estates. The amount of labor and capital pushed out of private agriculture was negligible; thus, the amount of this difference was close to zero, even when tithe collections reached a maximum at the end of the eighteenth century. Not only did the Church itself and the Indian villages produce a major portion of the colony's farm products and livestock (thus removing a sizable portion of agriculture from the full effect of the tithe), but differences in productivity between private agriculture and the rest of the economy indicate that nonagricultural pursuits were already more productive than agriculture. Therefore, the most important effect of the tithe was distributional: a larger portion of land remained in the hands of the Church and the Indian villages than would have been the case if the tithe had not been levied.(17)

Even when historians have realized that tithe collection did not directly reduce the gross national product, they have often suggested that the Church used this income unproductively: instead of investing its revenues in new industries and enterprise, the Church dissipated a tenth of the economy's output on new religious construction or on the wages of priests. There are two problems with this contention. First, the multiplier effect of "unproductive" expenditure in modern times has proved reasonably effective in stimulating, rather than depressing, economic activity; and there is no reason to assume, a priori, that this effect was absent in colonial Mexico. Second, the Church did invest a sizable portion of its revenues in mortgage loans to private entrepreneurs. Indeed, the Church probably raised the rate of investment in the economy above what it would have been had the tithe revenues remained in private hands.

The Church's role as the country's chief banking institution has also been misunderstood. The Church earned a net income from the tithe, private donations, and its various properties. In addition, it acted as fiduciary agent for trust funds left in its care. The Church invested a large portion of its net income and all of the trust capital it managed, usually at 6 percent interest on the security of real property. Because it charged a low, nonmarket interest rate, the Church dominated the mortgage-lending market. What effect did this have on economic activity? Practically none at all. Once again, the main effect was distributional. The Church lost money when it lent funds below the market rate of interest, while the recipients of Church credit gained. It performed like a modern development bank, charging taxpayers to subsidize the accumulation of private capital. The Church imposed no legal or practical obstacles to prevent recipients from investing in factories rather than haciendas or high living.(18) If factories were not built, reasons other than interference by the Church were involved.

Finally, the Church was a property owner. Studies of Church-owned estates suggest they were at least as well managed as those in private hands. The larger Church estates enjoyed considerable advantages, including a longdistance communication network which made it possible to plan both sales and purchases to take advantage of prevailing market conditions in widely scattered areas.(19) Most Church estates after independence were rented to individuals, so the efficiency of these properties did not depend on Church management at all.(20) The major difference between the ecclesiastical and the private sectors lay in the Church's exemption from most of the taxes levied on private enterprise by colonial and, later, national governments. There is no evidence, however, that public authority would have put revenues gained from taxing the Church to better use; and much evidence, in fact, suggests the contrary. After expropriation, for example, a large number of Church-supported charitable activities disappeared, so the short-run effect of moving Church properties into the taxable private sector seems to have been a drop in the welfare of poor people.

THERE WERE TWO MAIN OBSTACLES TO ECONOMIC GROWTH in colonial Mexico which together explain most of the difference in productivity between the Mexican and U.S. economies in 1800: inadequate transport and inefficient economic organization--geography and "feudalism."

Mexico's population and economic activity have always been concentrated in highland valleys and plateaus far from the sea. Inland waterways did not exist, nor could they have been constructed. The entire economy, therefore, depended on costly overland transportation to move goods and people. One example will suffice to show how geography affected transport costs and thus both the development of markets and the growth of productivity. In the late eighteenth century the Mining Deputation of Guanajuato estimated that this city received its supply of maize from estates located within a radius of ten leagues (fifty-five kilometers).(21) At prevailing rates in this area, producers had to pay between one and one and one-third reales (0. 125 to 0. 166 pesos) per ton kilometer to ship goods commercially.(22) The price of maize fluctuated widely, but for purposes of illustration it may be put at thirty pesos per ton.(23) Estates at the margin of this radius had to pay more than eight pesos (40 percent of the sale price of maize) just to carry it to Guanajuato. If maize producers had been able to ship by water to Guanajuato, at the same rates paid for shipments by canoe on the lakes surrounding Mexico City, Guanajuato's supply radius would have increased to between 485 and 725 kilometers.(24) The effects of such cheap transport can only be imagined, but the impact on national income would have been very large: increased regional specialization and division of labor, new centers of production previously undeveloped because of distance from centers of population and markets, greater reliance on markets to exchange products, greater mobility for both capital and labor, external economies due to better communication, and the like. The United States, of course, already possessed these advantages. Had Mexico shared them, the difference in productivity between the Mexican and U.S. economies would have been reduced, all other things being equal, by at least one-third.

This conclusion is based on what actually happened when railroads were constructed during the Porfiriato. Freight-transport costs fell to less than one-tenth of their prerailroad levels. Social savings by 1910 amounted to at least 10.8 percent of gross domestic product, equivalent to one-third of the productivity gains of the Mexican economy between 1895 and 1910.(25) If Mexico been endowed with a system of rivers, like those along the eastern seaboard of the United States, a major portion of the U.S. advantage at the beginning of the nineteenth century would have been eliminated.

All other things were not, of course, equal. The viceregal government might well have decided to raise internal customs duties (alcabalas) to match reductions in transport costs. Or Madrid could have ordered the colonial government to deny licenses to entrepreneurs clamoring to take advantage of lowered transport costs and ready to invest in new productive enterprise. Or the crown could have decided to make transportation a royal monopoly (estanco, from which estancar, to stagnate) and control the supply of shipping and charge inflated rates. Or the Council of the Indies might have urged the king to protect the indigenous population by forbidding its employment in the production of goods sold in faraway places. Or, after multiple lawsuits lasting several decades, the Audiencia might have decided in favor of the petitions of muleteers, wagonmasters, and hotel keepers and ordered that all boats, rafts, and canoes in the colony had to be owned exclusively by former muleteers, wagonmasters, and hotel keepers who registered with the authorities and agreed to lend the king ten thousand pesos.

The very plausibility of measures such as these suggests a seriously deficient institutional environment for entrepreneurial activity. Indeed, the second of the two main obstacles to Mexican economic growth was inefficient economic organization. This term does not mean, for example, that private estates or mining companies wasted resources, given the environment they faced. It refers instead to an ensemble of policies, laws, and institutions that magnified, instead of reduced, the gap between the private and the social benefits of economic activity. During the colonial period and most of the nineteenth century, activities which could have contributed to economic growth were never undertaken because they promised too small a return to potential owners and producers. Either existing law and practice discouraged more productive enterprise or new laws and practices needed to protect and stimulate more productive activity never developed.(26)

In the colonial period, legal constraints on the mobility of capital and labor inhibited the development of factor markets.(27) Minute public regulation of economic activity for fiscal and other purposes raised start-up costs and discouraged enterprise.(28) The judicial system increased the risks of entrepreneurial activity by failing to enforce a well-defined set of property rights.(29) Fiscal policy made transactions more costly, discouraged use of markets as a means for exchanging products, and contributed to the geographical isolation of those regional and local markets which did develop.(30) Royal monopolies on the production and distribution of many commodities distorted prices and reduced production.(31) Investment by public authority or voluntary agencies in infrastructure or human capital was negligible. No general legislation existed to promote the realization of economies of scale through joint stock companies or corporations. Innovation was discouraged by a system of privileges that did not guarantee a return to inventors or investors in the application of new processes.(32) Corporate exemptions from a portion of the risks and constraints imposed on the rest of society were distributed by the crown to favored groups and individuals, with the net effect of increasing the burden on others. The special courts established to hear cases involving members of each corporate group compounded the ordinary chaos of the judicial system with interminable litigation over which court was appropriate to hear each case and thus increased the uncertainty that plagued the legal environment.(33) The property of Indian villages, town councils, certain public bodies, entailed estates, and the trusts administered by the Church were defined by law as inalienable. The land thus held could not be used as collateral for mortgage loans, rented for extended periods of time to a single tenant, or sold to anyone.(34) However much evaded or even violated in practice, these constraints helped immobilize resources or divert them to less productive uses.

At the apex of this system of government sat the crown, whose power was constitutionally absolute. No rights of citizens and no law, regulation, or settled custom bound the king's freedom of action in the colonies. All legislative and judicial acts derived their authority from the crown. The king could, and often did, grant individual exemptions from the application of his own laws or issue judicial decisions on appeal that controverted his own decrees in order to take into account the personal merits of the litigants. Not infrequently, the king's ministers, viceroys, and appellate courts acted in the same way.

The interventionist and pervasively arbitrary nature of the institutional environment forced every enterprise, urban or rural, to operate in a highly politicized manner, using kinship networks, political influence, and family prestige to gain privileged access to subsidized credit, to aid various stratagems for recruiting labor, to collect debts or enforce contracts, to evade taxes or circumvent the courts, and to defend or assert titles to land. Success or failure in the economic arena always depended on the relations of the producer with political authorities--local officials for arranging matters close at hand, the central government of the colony for sympathetic interpretations of the law and intervention at the local level when conditions required it. Small enterprise, excluded from the system of corporate privilege and political favors, was forced to operate in a permanent state of semiclandestinity, always at the margin of the law, at the mercy of petty officials, never secure from arbitrary acts and never protected against the rights of those more powerful.

This system of government made "free" enterprise impossible. It was not merely a matter of specific policies, laws, and institutions or their collective impact at a particular point in time that discouraged enterprise. The chief obstacle was the nature of the state itself, its operating principles, the basis for all of its acts. Mexico's economic organization could not have been made more efficient without a revolution in the relationship between the state and economic activity.

Unfortunately, the cost-benefit approach so helpful in identifying the impact of economic organization in theoretical terms cannot be applied quantitatively. It is possible to know the direction (positive or negative) of the effects of specific laws, policies, or institutions but not the magnitude of their impact. Direct measurement is impossible in practice, even to distinguish the relative importance of the various components of the system. Some advances may come from the aggregation of individual case studies at the enterprise level, but these may not be of much help in estimating the loss from economic' activities that never got started in the first place. Treating the impact of economic organization as a residual is scarcely satisfactory, especially when the size of the residual is suggested by an international comparison rather than an estimate of Mexico's own potential for growth. Nonetheless, a starting point for future work might well be the hypothesis that most of the gap between the Mexican and U.S. economies in 1800 was due to differences in economic organization.

THE TWO MAIN OBSTACLES TO ECONOMIC GROWTH in colonial Mexico-inadequate transport and inefficient economic organization--could have been eliminated early in the nineteenth century. Independence in 1821 emancipated the country from the source of the policies, laws, and institutions that inhibited enterprise. And railroad technology developed in the 1830s and could easily have been imported by 1840.(35) By the time of independence, the liberal Spanish Cortes, established to resist the Napoleonic invasion and later revived under popular pressure, had already eliminated many important constraints on economic activity. Ethnic distinctions between citizens in employment, taxation, and justice were abolished; corporate property rights were restricted to the Church and the Indian villages and town councils; the number of royal monopolies was reduced and their activities were curtailed; the corporate privileges of certain groups, including most of the guilds, were eliminated; efforts were made to streamline the judicial system; and revision of the antiquated law codes was begun.(36) But then Mexico plunged into a half century of political, social, and international warfare. The collapse of stable government nullified the potentially positive effects of the changes that accompanied independence and deprived both the new government and the private sector of the resources needed to improve transportation.

Mexican independence came through a virtual coup d'état by the colony's Creole elite, carried out largely to separate Mexico from the liberalizing process under way in the mother country. For the next half century, repeated efforts were made to recreate the arbitrary centralism of the colonial state. The principal proponent of these conservative efforts was a limited social group of major landowners and industrialists in the center of the country (often residents of Mexico City), who had been the principal beneficiaries in the colony of the crown's interventionism or who, like the large merchant houses of the capital, sought to regain privileges the crown itself had abolished in the reforms of the late Bourbon era. Allied with this group were the Church, seeking to preserve its privileged status and to reverse the anticlericalism of the later Bourbon kings, and the new professional army, endowed with privileges in the early constitutions, whose chief raíson d'etre was to be found as enforcer of the new centralism against regional demands for greater autonomy and liberal clamorings for reform.(37)

Traditional accounts of Mexico after independence have attributed the conservative orientation of the new country's early governments to the influence of a "feudal" or "semifeudal" landowning class. Early national governments were weak. The nation disintegrated into a multiplicity of regional satrapies controlled by local caudillos. The parallel to Europe's Middle Ages seemed very apt. The chief problem with this interpretation has been and remains a lack of systematic data on the social composition of the political forces that fought for control over the new nation's destiny. Aside from the small group of magnates in the capital, there is little evidence to suggest widespread support for conservative centralism among Mexico's landowners. Many-including important regional chieftains-supported the liberal cause, because it promised less interference by the national government in local affairs. This support, in turn, aided the Church in its efforts to convince the Indian population to support the capital against the local liberals.

Feudalism, in the broad conception of Marxist historians, is usually linked to serfdom. Serfs were peasants permanently attached to the persons or properties of aristocrats. They labored with their own implements and were obligated to yield up a portion of their product or to contribute a portion of their labor to the lord. In Mexico, however, legal serfdom did not exist and recent research has demonstrated that debt peonage (often assumed to be its New World equivalent) was effectively practiced only in parts of the sparsely populated geographical extremities of the country. Indeed, Mexico's landowners enjoyed none of the privileges of the old European nobility, either before independence or after. Spanish raison d'etat-principally the fear of an American nobility rising to claim sovereignty over New World populations prevented any such development. Indian access to land came from the crown in the form of corporate, and thus inalienable, land grants to villages. The crown never countenanced legal obligations between Indian villagers and landowners, save those regulated by royal decree and administered by royal officials. The royal courts and the tenacious resistance of the villagers themselves prevented the incorporation of villagers and their lands into the haciendas. Preservation of nonexistent servile obligations or other privileges played no role in determining the social composition of the warring parties in independent Mexico.(38)

If Mexico's landowners were not privileged, neither were they powerless. By designating the villages as corporate entities, the crown had virtually tied the bulk of the Indian population to specific pieces of real estate. Geography, culture, and lack of communication restricted Indian mobility still further. Most Indian villagers worked as seasonal labor on the great estates in the vicinity of their homes. Often, they did so out of need-village land grants in the sixteenth and seventeenth centuries were inadequate for the rising population in later periods. In the eighteenth century the royal courts became less protective of village land titles as the large estates expanded to profit from the colony's prosperity. In many areas seasonal labor on the estates acquired the force of habit over many years or continued as a form of rent charged for access to estate water, salt, or woodlands.

Coercion, however, was widespread and pervasive. Usually it involved various kinds of pressures applied by landowners in cooperation with local civil and ecclesiastical authorities. The stratagems employed varied from place to place, as did the success of the landowners who employed them. The colonial authorities resisted these informal attempts to obligate local villagers but their efforts were never entirely effective, for the landowner's capacity to manipulate the local environment was always decisive. Far from desiring a restoration of Bourbon centralism, most landowners wanted to be allowed a free hand to control their immediate surroundings.

The social composition of what became the liberal movement has not received adequate attention from researchers. The only significant study of this question--a survey of the delegates to the Constitutional Convention of 1856-57--has revealed that most of the participants were either lawyers or generals in the liberal forces.(39) In the port cities on the Gulf of Mexico, support for liberal governments was apparently quite strong, which suggests ties to merchants and tradesmen in addition to local landowners and caudillos. The liberals also appealed to the "professional classes, lawyers, doctors, small property owners, merchants, the middle ranks of the clergy and military. . . . "(40) While the thrust of liberal demands for change-at least those emanating from liberal political leaders who assumed command of the movement at the national level-was for institutional change modeled on the example of the United States and Western Europe, Mexico's bourgeoisie under any definition of the term-constituted a small, weak, and highly fragmented social grouping. Neither landowners nor capitalists can be said to have formed a national governing class in independent Mexico. With more than fifty changes of government in a half century, no group effectively dominated national government.(41)

Because it fought to prevent institutional change, the Church did constitute an important obstacle to economic growth, even though its strictly economic activities did not directly impede progress. The lack of evidence demonstrating systematic support for the conservative cause by Mexico's landowners, however, makes it impossible-even in this general sense-to describe the system of land tenure as a significant obstacle to economic advancement. The social base of Mexico's conservative movement was not determined by the nature of rural social relations but by the pattern of relations between a narrow stratum of the economic elite-however their fortunes were made and the central government. In both colonial and independent Mexico, Enrique Semo's observation was brilliantly precise. "Feudalism in Mexico," he wrote, "was strongest at the level of superstructure."(42)

FOR A MULTITUDE OF REASONS that need not detain us here, liberalism emerged triumphant from Mexico's postindependence turmoil. The last hope for restoration of the colonial pattern of government ended when the short-lived regime of Emperor Maximilian (1862-67) embraced an essentially liberal program and moved rapidly to remove the same obstacles to capitalist development that his liberal enemies hoped to abolish. Not only did Maximilian promulgate the nation's first modern commercial code to replace the Ordenanzas of Bilbao, but his government used French aid to push construction of the nation's first railroad as well.(43)

When the apparatus of national government fell to the liberal regime of Benito Juárez in 1867, the first and most important step in the transformation of property rights proclaimed in the Constitution of 1857 was already an accomplished fact. Most of the wealth of the Church was now in private hands. Little more was accomplished during the decade of the Restored Republic. The liberals did manage to produce a new civil code (1870) to recognize the new status of Church-state relations, but the commercial code was revoked and nothing was enacted to replace it. Tariff reform, which affected the largest source of federal government revenues, was enacted by Congress in 1872, but modernization of the tax system and reform of the antiquated and inefficient treasury were postponed. The judiciary was reorganized and purged, but the main qualification for appointment was loyalty to the new regime and the judicial system remained as chaotic as ever. The Juárez and Lerdo regimes lacked the resources to repair roads, subsidize railroad construction, build schools, or lower taxes. The Veracruz-Mexico City rail line was finally completed in 1873, but the debt-ridden company that owned it charged rates only slightly below the cost of shipments by wagon or mule.(44)

When Porfirio Diaz seized power in 1877, nothing had been done to reform the colonial mining code since the 1820s, when Congress abolished the mining guild and relaxed the prohibition on foreign investment. No legislation existed to encourage the formation of corporations with limited liability. No banking laws were passed, except for concessions to particular banks. No mortgage credit law existed to protect long-term investment and replace the spiritual sanctions on which the Church had relied. A modern patent law did not exist. Despite constitutional provisions that specifically outlawed them, colonial fiscal measures like the internal customs still provided most of the revenue for state and municipal governments. Economic activity of all kinds still required special permits and licenses for which special taxes and fees were charged. Though Church wealth had been expropriated, the corporate holdings of the Indian villages remained unaffected throughout most of the country. The liberal movement had destroyed the political power of the Church, seized the apparatus of government, and changed the constitution. But a new superstructure of laws and institutions for a capitalist society had yet to emerge.

The Porfirian military coup occurred at a fortuitous moment. In a short time, the Díaz regime issued major railway concessions for lines running across the Central Plateau and northward to the U.S. border. Railroad concessions raised land values along projected routes and precipitated widespread usurpations of Indian village lands by estate owners and land companies. Needless to say, the railroad companies reported no difficulties in recruiting thousands of propertyless wage laborers for the massive construction projects that got under way in late 1880 In the next three years, nearly five thousand kilometers of track were built by tens of thousands of Indian workers, many of whom had only recently been driven from their lands. As cheap transport and the nation's new prosperity revived the profitability of the estates, expropriation of village lands progressed apace. The dispossessed villagers swelled the ranks of the nation's landless rural and urban proletariat.(45) Capitalist modernization had begun.

As railroad lines spread throughout the countryside and the first signs of massive foreign interest in Mexican resources appeared, a series of major legislative reforms was enacted. In 1884, Congress passed a new commercial code, the single most important piece of economic legislation since independence. (The new code had to be reformed--ironically, the revisions were modeled after Spain's 1885 code--in 1889, largely because it did not make adequate provision for limited-liability corporations.) In 1887, a new mining code followed. Banking, first included in the commercial codes, became the object of special legislation in 1897 and 1908. Reform of the fiscal system, begun in 1881 with reorganization of the treasury, continued in stages for the rest of the decade until new tax laws, tariff schedules, and public-debt reorganization had been achieved. After more than a decade of virtual isolation, the Mexican government signed commercial treaties, first with the United States and, later, after successful renegotiation of the foreign debt, with all of the European powers.(46)

THE SIMULTANEOUS DEVELOPMENT OF TRANSPORTATION and a more efficient economic organization made possible the economic growth of the Porfirian era. That growth had characteristics that made Mexico's advances in this period markedly different from the economic and institutional development of the industrial economies of the North Atlantic. While Mexico was resolving its internal conflicts, exacerbated as they were by foreign wars and invasions, the Industrial Revolution brought Europe and the United States to new heights of productivity. The main significance of Mexico's greater relative backwardness lay in the vast comparative advantages of foreign technology and resources for the development of Mexico's own economy. No Mexican government, whatever the social composition of its leaders and supporters, could have long resisted the benefits offered by foreign participation in Mexico's economy. What Mexico might have done on its own when it first won independence could no longer be done without a far greater sacrifice of palpable immediate gains by the 1870s.

The participation of foreign capital in Mexico's first period of sustained capitalist growth had five main consequences. First, the construction of railroads and subsequent development of major export industries interrupted the gradual disintegration of the large estates into smaller units, which had begun in 1810 with the outbreak of the independence movement. The new prosperity also ensured that the properties expropriated from the Church passed into the hands of large-scale operators. Second, the new industries developed by foreign capital and by Mexican entrepreneurs with foreign financing also operated on a large scale. In both agriculture and industry, large-scale units ensured more rapid growth. They also defined the nature of the institutional changes Mexico adopted to facilitate capitalist development.

Third, Mexico's numerous class of petty tradesmen and small-scale producers were therefore assigned a position in Mexico's new society similar to that which it had occupied in the colonial era. Since it was clearly marginal to the country's economic progress, it suffered from the same arbitrary treatment and operated in the same netherworld of semiclandestinity as before. Nor did that numerous class of independent farmers envisioned by some liberal leaders ever emerge. Without either force, Mexican governments reverted to authoritarian models from the colonial past, despite the new institutions of the liberal era, and spent their increasing resources on more important projects than the development of the nation's human resources.

Fourth, the Revolution of 1910 was produced by the discontent of precisely those elements in Mexican society whose importance in the nation's political and social life would have increased, rather than diminished, in the absence of foreign resources: the dispossessed villagers who never became family farmers and the "middle sectors" that continued to suffer exclusion from both political and economic opportunity. Fifth, foreign resources established a pattern of economic activity that fixed Mexico's position in the world economy at the most unfavorable moment in the nation's economic history. Foreign resources attracted their Mexican counterparts to produce, labor, and invest in activities made profitable by Mexico's short-term comparative advantage in the production of raw materials and agricultural products for export. It may be doubted, of course, that Mexico's present condition would be more favorable had the nation followed a different course. But it is clear enough, in any case, that the path Mexico did follow promised no more than a long-term dependence on foreign technology, resources, and markets.

1. For the colonial period, see Charles Gibson's excellent review, "Writings on Colonial Mexico," Hispanic American Historical Review (hereafter HAHR), 55 (1975): 287-323. For a review of the nineteenth- and twentieth-century work, see Enrique Florescano, "Ensayo de Interpretación, " Introduction to the section on Mexico in Roberto Cortés Conde and Stanley J. Stein, eds., Latin America: A Guide to Economic History, 1830-1930 (Berkeley and Los Angeles, 1977), 435-55.

2. See the many references to Mexico in John J. TePaske, "Recent Trends in Quantitative History: Colonial Latin America," Latin American Research Review, 10 (1975): 51-62. TePaske's own work on colonial fiscal data provides an excellent example of the new trends: La real hacienda de Nueva España. La real caja de México ( 157(;-1816), no. 41 of Coleccibn científica: Fuentes, Instituto Nacional de Antropología e Historia (Mexico, 1976). TePaske's Mexican study forms part of a larger project, undertaken in collaboration with Herbert Klein, that involves the reconstruction of treasury office accounts in all of the major colonial centers.

3. The data in Table I may be compared to estimates of per capita income for other Latin American countries cited in William P. McGreevey, "Recent Research on the Economic History of Latin America," Latin American Research Review, 3 (1968): 98-99. Also see Laura Randall, A Comparative Economic History of Latin America, 1,500-1914, 1 (Ann Arbor, 1977): 224, for an attempt to construct "upper limit," estimates of Mexican national income beginning in 1520.

4. The 1910 British estimate includes Ireland. For population data for these countries, see the sources cited in Table I For Mexico, see El Colegio cle México Estadísticas económicas del Porfiriato: Fuerza de trabajo actividad por sectores (Mexico, n.d.), i.

5. The United States exported roughly twenty million pesos (or dollars; the exchange rate was 1 : 1) worth of goods annually just prior to the Napoleonic Wars. American State Papers, 7, Class 4: Commerce and Navigation (Washington, 1832), 1: 34ff. Mexican foreign trade data begin in 1796. While U.S. trade increased markedly during the war, Mexico's did not. Miguel Lerdo de Tejada, Comercio exterior de México desde la Conquista hasta hoy (1853; reprint ed., Mexico, 1967), no pagination.

6. Thomas, "A Quantitative Approach to the Study of the Effects of British Imperial Policy upon Colonial Welfare, "Journal of Economic History, 25 (1965): 615-38. Thomas has been criticized on methodological grounds, but re-estimates employing more refined procedures have yielded the same results. See Peter D. McClelland, "The Cost to America of British Imperial Policy," American Economic Review, 59 (1969): 370- and Joseph D. Reid, Jr., "On Navigating the Navigation Acts with Peter B. McClelland: Comment," American Economic Review 60 (1970): 949-55.

7. A detailed account of the data and estimating procedures is available from the author on request. The income loss is calculated by using tariff and transshipment cost data applied to the value of goods traded legally between 1797 and 1820. Indirect costs are estimated by assuming unit-price elasticity of demand. Like Thomas' estimates for British North America, the Mexican estimates here provide only a crude upperbound measure of the burden of indirect trade.

8. According to Humboldt, the revenues of New Spain in 1803 amounted to SOME 20 million pesos, of which just over half were exported to Spain or to other Spanish colonies. Alexander v. Humboldt, Political Essay on the Kingdom Spain, trans. John Black, 4 vols. (London, 1811; reprint ed., 1966), 4: 224-29.Humboldt's figures do not coincide with those of the Royal Treasury Office in Mexico City. TePaske's reproduction of the accounts of the Real Caja in La real hacienda de Nueva España shows 1803 revenues at 39.1million pesos, nearly double Humboldt's figure. Unfortunately, the Caja's accounts do not show which revenues were exported. Direct subsidies to other colonies, called situados, are shown at 2 million, close toHumboldt's estimate. Net revenues from royal monopolies were remitted to Spain by law; the accounts show gross revenues from this source at 5.8 million. It is likely that a large portion of the difference betweenHumboldt and TePaske lies in war loans not counted by Humboldt, the proceeds from which amounted to10 million pesos in 1803. In the absence of precise export data from the original source, Humboldt's estimate of 1 0 1 million pesos for the fiscal burden is used in Table 2. Exports of bullion from New Spain declined drastically after the beginning of the independence wars in 1811. Thus, even if Humboldt's estimate for 1803 is too low using it to approximate the annual average outflow for the entire period from1797 to 1821 doubtless exaggerates the average fiscal burden over the whole period.

9. Jean Paul Berthe, "El problema del abastecimiento del azogue a las minas de plata en México independiente," Paper presented to 41st International Congress of Americanists, Mexico City, 1974.

10. Juan Felipe Leal and Mario Huacuja produced an excellent bibliography in 1975 that covers most of the nineteenth-century work: "Fuentes para el estudio de la hacienda en México 1856-ig4o," no. io in Avances (le Investigación, Centro de Estudios Latinoamericanos, Universidad Nacional Autónoma de México (Mexico, n.d.). Enrique Semo covers the colonial period as well in his review of recent hacienda studies in "La hacienda mexicana y la transición del feudalismo al capitalismo," Historia y Sociedad, 2d ser., 5 (1975): 63-81. For seven recent studies, see Enrique Florescano, ed., Haciendas, latifundios y plantaciones en Ambrica 1,atina (Mexico, 1975). Also see, for an additional seven studies, Enrique Semo, ed., Siete ensayos sobre la hacienda mexicana, 1780-1880, no. 55 in Colección científica: Historia, Instituto Nacional de Antropología e Historia (Mexico, 1977).

11. See, for example, two oft he most outstanding: William Taylor, Landlord and Peasant in Colonial Oaxaxa (Stanford, 1972); and Claude Morin, "Croissance et disparit~s sociales dans une 6conomic coloniale: Le Centre-ouest Mexicain au XVIIIe siecle" (Ph.D. dissertation, École Practique des Hautes Ludes, University of Paris, 1974). And, of course, see Charles Gibson, Tlaxcala in the Sixteenth Century (New Haven, 1952), and the masterful .~lztecs under Spanish Rule: A History of the Indians of the Valley of Mexico, 1519-1810 (Stanford, 1964).

12. Only one of the hacienda studies directly addresses the question of productivity; and, without doubt, Ward Barrett's work provides an excellent model for historians concerned with the analytical issues posed here: The Sugar Hacienda of the Marqueses del Valle (Minneapolis, 1970).

13. On the production of small units, including villages, see William B. Taylor, "Hacienclas coloniales en el valle cle Oaxaca," Historia Mexicana, 23 (1973): 284-329; Celia Rabel, "San Luis cle la Paz: Estudio de economía y demografía históricas (1645-1810)" (M.A. thesis, Escuela Nacional de Antropologia, Universidad Nacional Autónoma de México 1975); Brian R. Hamnett, Politics and Trade in Southern Vfexico, 17501821 (Cambridge, 1971); and Woodrow Borah, Silk Raising in Colonial Mexico (Berkeley and Los Angeles,1943).

14. John H. Coatsworth, "Anotaciones sobre la producción de alimentos durante el porfiriato," Historia Alexuana, 26 (1976): 167-87

15. For U.S. data, see Stanley Lebergott, Manpower in American Economic Growth: The United States Record since 1800 (New York, 1964), 117; and Paul A. David, "The Growth of Real Product in the United States before 1840: New Evidence Controlled Conjectures," Journal of Economic History, 27 (1967): 151-97. For Mexican estimates, see John H. Coatsworth, From Backwardness to Underdevelopment: The Mexican Economy, 1800-1910 (forthcoming), chap. 2.

16. Michael Costeloe, Church Wealth in Mexico: A Study of the "Juzgado de Capellanias" in the Archbishopric of Mexico, 1800-1856 (Cambridge, 1967), 14-18.

17. This discussion of the impact of the tithe is indebted to Pascual García Alba, "Los liberales y los bienes del clero," Seminar paper, El Colegio de México, 1974

18. Costeloe, Church Wealth in Mexico, chap. 3.

19. Herman Konrad, "Santa Lucia, 1676-1767: A Jesuit Hacienda in Colonial Mexico" (Ph.D. dissertation, University of Chicago, 1973), 167-68.

20. Jan Bazant Alienation of Church Wealth in Mexico.- Social and Econornic Aspects of the Liberal Revolution, 18561875 (Cambridge, 1971), 287.

21. For the supply radius for Guanajuato, see the citation in Eric Wolf, "The Mexican Bajío in the Eighteenth Century: An Analysis of Cultural Integration," Synoptic Studies of Mexican Culture Tulane University Middle American Research Institute, Publication 17 (New Orleans, 1955), 183.

22. Brading and Potash cite identical wagon freight rates when converted to ton kilometers. See Davíd A. Brading, Miners and Merchants in Bourbon Mexico, 1763-1810 (Cambridge, 1971), 16; and Robert Potash, El Banco de Avio de México: El fomento de la industria, 1821-1846 (Mexico, 1959), 17.

23. For the maize price cited, see both Bracing, Miners and Merchants in Bourbon Mexico, 16, and Wolf, "The Mexican Bajío in the Eighteenth Century," 17, where the figure is given as approximately eleven reales per fanega.

24. For canoe rates on the Mexico City lakes, see Humboldt, Political Essay on the Kingdom of New Spain, 2: 174-75.

25. For the railroad data, see John H. Coatsworth, Crecimiento contra desarrollo: El impacto económico de los ferrocarriles en el porfirialo, 2 VOLS. (Mexico, 1976).

26. See Douglass North and Robert Paul Thomas, The Rise of the Western World: A New Economic History (Cambridge, 1973), chap. i.

27. For discussion of restrictions on geographic and occupational mobility, see Eric Wolf, "The Mexican Bajío in the Eighteenth Century," 179-83. Mobility of capital was impeded in numerous ways. Commercial activity, for example, was forbidden to certain occupational groups (e.g., clerics and public officials) and restricted for the majority of the population (Indians and individuals of mixed Indian and European ancestry). Guillermo Hernández Peñalosa, El derecho en India y en su Metrópoli (Bogota, 1969), 408-12; and jacinto Palleres, El derecho mercantil mexicano (Mexico, 18gi), 167-gi.

28. See Wolf, "The Mexican Bajío in the Eighteenth Century," passim., for some examples. For an excellent survey of fiscal and economic policy in the eighteenth century, see Enrique Florescano and Isabel Gil Sanchez, "La época de las reformas borbónicas y el crecimiento económico, 1750-1808," in Alejandra Moreno Toscano el al., Historia General de México, 4 vols. (Mexico, 1976), 2: 183-301. Also see Enrique Semo, Historia del capitalismo en México: Los origenes, 1.521-1763 (Mexico, 1973), 162-65

29. In the eighteenth century, eleven different Spanish law codes, the earliest dating from 693 A.D., were still consulted to resolve disputes. Gustavus Schmidt, The Civil Law of Spain and Mexico (New Orleans, 1851), 152. In addition, mercantile law was partially codified in the Ordenanzas de Bilbao, updated at various points after their initial promulgation in 1459; and a new mining code, the Ordenanzas de Mineria, was issued in 1783. John T. Vance, A Guide to the Law and Legal Literature of Mexico (Washington, 1945), 67, 203. Lesser codes existed and were applied to special areas of activity without ever clearly superseding more general legislation.

30. Internal customs were established throughout the country to tax goods that passed between separate jurisdictions. Required permits (guias) for transporting foreign-made goods or goods destined for export had to indicate the precise route the merchandise would take and specify the amount of time to be consumed in transport. On arrival at the destination, a new document was issued that had to be physically transported to the origin of the shipment and submitted within a specified time to prove that the guia had been complied with. Deviation in time or route made shipments subject to confiscation. For most markets, permits and licenses were required or goods could be confiscated. Enrique Orozco, La evolución de la legislación en la República (Mexico, 19"), 7.

31. For a list of royal monopolies in the late eighteenth century, see Andres Lira, "Aspecto fiscal de la Nueva España en la segunda mitad del siglo XVIII," Historia Mexicana, 17 (1968): 388-89.

32. Juan de la Torre, Legislación de patentes y marcas: Colección completa de todas las disposiciones que ha regido en México sobre esta materia, desde la dominación española hasta la época actual, concordada y explicadas (Mexico, 1903),

33. One of the colony's most distinguished viceroys, Count Revillagigedo, put it most succinctly: "Everyone in his own court believes he will be better treated than in that of others, and thus all efforts are bent toward moving disputes and suits to one's own ground." As quoted in Jacinto Pallares, El poder judicial: o Tratado completo de la organización, competencia y procedimientos de los tribunales de la República Mexicana (Mexico,1874), 35.

34. Jose Maria Ots Capdequi, España en América: El régimen de la tierra en la época colonial (Mexico, 1959), chaps. 2-3.

35. The first concession issued by the Mexican government for construction of a railroad was dated 1837. The line that was envisioned, from Mexico City to Veracruz, was not completed until 1873. John G. Chapman, La construcción del Ferrocarril Mexicano, 1837-1880 (Mexico, 1974), 22-25, 16o.

36. For the principal acts of the Cortes and the major legislative changes enacted in Mexico after independence, see Manuel Dublan and José Maria Lozano, Legislación mexicana o colección completa de las disposiciones legislativas expedidas desde la independencia de la República i (Mexico, 1876).

37. Michael Costeloe, La Primera República Federal de México: 1824-1835 (Mexico, 1975), 438-39; and Jan Bazant, A Concise History of Mexico from Hidalgo to Cárdenas, 1805-1940 (Cambridge, 1977), 74-77

38. On peonage in the far north and in the south of the country, see Friedrich Katz, "Labor Conditions on Haciendas in Porfirian Mexico: Some Trends and Tendencies," HAHR, 54 (1974): passim. For an earlier period, see the hacienda studies reviewed in the literature cited in note 10.

39. Richard N. Sinkin, "The Mexican Constitutional Congress, 1856-57: A Statistical Analysis," HAHR,53 (1973): 1-26.

40. Costeloe, La Primera República Federal de México, 439.

41. The most disastrous consequence of Mexico's relative economic backwardness and prolonged internal strife was the loss of one-half of the national territory to the United States. The magnitude of the loss became apparent almost immediately. The California "Gold Rush" began in 1849, a year after Mexico ceded that territory in the Treaty of Guadalupe Hidalgo. By the beginning of the century, the mineral output alone of the lost territories exceeded the GNP of the Mexican Republic. Mexico did not have the capacity to develop its vast territories, assimilate immigrants, and repel aggression in the mid-nineteenth century. Nor did any of Spain's former colonies at that early date. Had Mexico encountered a less powerful and warlike neighbor than the United States, the vast wealth of the northern half of the country could have altered the whole course of the country's economic history later on.

42. Enrique Semo, "El desarrollo del capitalisrno en la minería y la agricultura de la Nueva España," Historia _y Sociedad, 5 (1969): 5.

43. On railroads during the "second" empire, see Chapman, La construcción del Ferrocarril Mexicano, chaps. 5-7. On policy and law under Maximilian, see José Luis Martinez, "México en busca de su expresión," in Josefina Zoraida Vasquez et al., Historia General de México: 3: 138-62.

44. Francisco Calderón, La República Restaurada: La vida económica. in Daniel Cosio Villegas, ed., Historia moderna de México (Mexico, 1965), contains the most exhaustive account of economic policy and legislation during this period.

45. John H. Coatsworth, "Railroads, Agrarian Protest, and the Concentration of Landholding in the Early Porfiriato, " HAHR, 54 (1974): 48-71.

46. Legislative and administrative changes during the Porfiriato are surveyed in appropriate chapters of Daniel Cosio Villegas, ed., Historia moderna de México: El Porfiriato, La vida económica, 2 VOLS. (Mexico, 1965), i: chap. 3, 2: chaps. 7-9.