Tag Archives: Engerman and Sokoloff



D. Acemoglu, S. Johnson & J.A. Robinson

The Quarterly Journal of Economics, Vol. 117, No. 4 (Nov., 20020 pg. 1231-1294

This is a very short summary taken from an essay I wrote. It is not intended to be a full exposition but rather a study aide.

AJR take a similar approach [to Engerman and Sokoloff] albeit they illustrate their thesis with statistical findings. They show there is a negative relationship between countries that were relatively rich in 1500 and economic prosperity today; the “reversal of fortune”.[1] They use population density and extent of urbanization as proxies for economic wealth in 1500 and argue that it was the relatively poor areas in which the “institutions of private property” were established whereas the norm in relatively richer areas was “extractive institutions” where power is concentrated in an elite and expropriation risks for the population in general are large. This is because “relative prosperity made extractive institutions more profitable for the colonizers; for example, the native population could be forced to work in mines and plantations”.[2] They posit that societies with good institutions are more able to take advantage of the opportunity to industrialize as private property institutions are “essential for investment incentives and successful economic performance.”[3] For example, they show that the regression analysis of current income against urbanization in 1500 predicts that Uruguay which had no urbanization in 1500 should have a current income 105% greater than Guatemala which in 1500 had an urbanization rate of 9.2%, and this turns out to be pretty close to the truth.[4] In order to prove the effectiveness of their instrument, and to thus prove that institutions cause growth and not vice versa, they have to show that the urbanization/population density in 1500 has no direct effect on current GPD levels other than through the effect it had on early institutions. Once they include the variable in their regression and control for the effect of institutions they cannot reject the null hypothesis that the coefficient of the instrumental variable is equal to 0, in other words that it does not explain any of the variation in GDP other than through its effect on institutions.[5]

Although similar to the ES hypothesis in that AJR are supporting what has come to be known as the “institutional hypothesis”, the instrument they employ is pointedly different to that of factor endowments. Indeed they specifically control for geographic variables including soil type and climate and do not find them statistically significant in explaining variation in GDP today.  They also engage much more directly with other hypotheses such as the “geography hypothesis” as well as the “colonial identity hypothesis”, the “latitude hypothesis” and the “religious hypothesis” and conclude that once they have controlled for geographic variants, the identity of the colonizers, the position of the colony relative to the equator and the religious makeup of the colonial society, all of those variables do not significantly explain variation in current GDP across ex-colonies.

[1] Acemoglu, Johnson and Robinson Reversal of Fortune: Geography and Institutions in the Making of the Modern World Income Distribution The Quarterly Journal of Economics, Vol. 117, No. 4 (Nov 2002) pp. 1231-1294

[2] Ibid.

[3] Ibid.

[4] Ibid

[5] Ibid. 



S.L. Engerman & K.L Sokoloff

NBER Working Paper 11057

A Summary

[The following is summarized within an essay written in the MT. It is not intended to be a complete summary, but rather an aide for future essay writing.]

ES suggest that institutions are a fundamental determinant of growth, and the quality of the institutions in Latin America explain why the colonies “that were the choices of the first Europeans to settle in the Americas, were those that fell behind” relative to those North American colonies of the today’s USA and Canada.[1] In reviewing the history of these colonies they claim the systematic factor that explains the diversity of quality in institutions across the “New World” is “extreme inequality in the distribution of wealth, human capital and political influence” in the early settlements. This led to the development of institutions primarily devised by elites to ensure the persistence of such inequalities and it is this poor institutional base that explains long-run economic underdevelopment. The degree of inequality was itself determined by the “factor endowments” that the colonists faced upon arrival in the New World.

The rich soils and warm climates prevalent in much of the Caribbean Latin America but particularly Brazil, leant itself to agricultural practices such as the cultivation of sugar, that exhibited large economies of scale and were thus most effectively produced on large estates using slave labour imported from Africa.[2] Elsewhere in Spanish America, where land was not so fertile, the primary export products were metals extracted particularly in Mexico, Bolivia and Peru. These territories were characterized by a large native population that survived contact with the European settlers. Land was distributed to a similarly small elite in similarly large quantities, and thus the estate owners had control over a large workforce which despite having better social standing than the slaves of Brazil, were nevertheless coerced into service in the mines.[3]

The slave economy of Brazil and the similarly coercive economies found in Spanish America obviously imply that inequality in political power, human capital and wealth was extremely high. Power was concentrated in the hands of a small European elite, and particularly in Spanish America this situation was maintained externally by “restrictive immigration policies”.[4]

Conversely, the soil and climate coupled with the relative dispersion of the native population (and thus the unavailability of coerced labour), meant that the factor endowments that the colonizers of the Northern American territories faced were “more hospitable to the cultivation of grains” which were most effectively produced on small farm holdings. The absence of economies of scale in agriculture meant that production was based on labourers of European descent so there was drastically less human capital inequality relative to rest of the New World. Furthermore “large landholding unravelled because even men of rather ordinary means could set up independent farms when land was cheap and scale economies were absent.”[5] What ES show is that the factor endowments of the Northern colonies meant land, wealth and thus political power was spread much more evenly over a relatively homogenous European population. In short, extreme inequality was absent.

ES argue that this equality in the Northern colonies set in motion a path of institutional development that focused on democracy, state investment in public goods and infrastructure and equality of opportunities both economic and political. Conversely, “where there was extreme inequality, political institutions were less democratic, investments in public goods were far more limited and the institutions that evolved tended to provide highly imbalanced…access to property rights and economic opportunities.”[6] They explain that in the highly unequal societies the elites were concerned that democratic institutions would place too much power in the hands of the poor majority who would use that power to redistribute land and redress the societal imbalances as regards opportunities. In such societies anti-democratic processes are essential for the elite to maintain their hegemony. Conversely in societies where wealth and human capital are more evenly distributed the demand for and merits of (in the eyes of those with political power) a more broadly participatory system are greater because “institutions are established in a society that has some power relations and they must reflect the distribution of this power”[7]. Furthermore, even had the elites in the Northern colonies perceived an advantage in anti-democratic institutions, they could not have instituted them as they were reliant on attracting European migrant labour which would not have been forthcoming had political and economic inequalities been great. In the Spanish colonies elites were able to act on exactly such perceptions as they were at the same time trying to restrict immigration due to the already extensive cheap labour.[8]  This explains why at the beginning of the 20th century none of the Latin American states had the secret ballot nor “more than a miniscule fraction of the population casting votes”[9] whereas the US and Canada had extended the franchise to a much greater extent, removing restrictions based on wealth and literacy.

Similarly education provision by the state in unequal societies is far less likely to develop in unequal societies as small elites are able to command private education for their offspring. There are be significant collective action problems associated with public school provision in such societies tiny elites are the major source of tax revenue (a cost to them) but the major beneficiaries of such schooling provisions are the poor majority. Therefore the benefits and the costs to the elite are hugely misaligned. In more equal societies where a larger proportion of the population makes up the taxable base, those costs and benefits are more equal vis-à-vis the individual taxpayer so public education provision becomes more feasible. As above, education may also have been used to lure migrant workers from Europe. Thus we see why “the rest of the hemisphere trailed far behind the United States and Canada in primary schooling and literacy”.[10]

The two examples above illustrate the method that ES employ. Although not strictly empirical (there is no data analysis in their work on this topic), the method is consistent with the search for the instrumental variable. The thesis would be testable under statistical conditions, and although perhaps more historical in approach, ES are trying to show that causality moves from institutions to growth as the instrument of “factor endowments” through its effect in inequality in early colonial economies in the New World varied the quality of the institutions but cannot be thought of as affecting current growth trajectories. The method is thus decidedly economic in style.

[1] Engerman and Sokoloff Colonialism, Inequality, and Long-Run Paths of Development National Bureau of Economic Research Working Paper 11057 (2005)

[2] Engerman and Sokoloff Factor Endowments, Inequality, and Paths of Development Among New World Economies National Bureau of Economic Research Working Paper  9259 (2001)

[3] Ibid.

[4] Ibid.

[5] Ibid.

[6] Engerman and Sokoloff Colonialism, Inequality, and Long-Run Paths of Development National Bureau of Economic Research Working Paper 11057 (2005)

[7] A. Przeworksi Institutions Matter? Government and Opposition Vol. 49 No. 4 (2004) pp.  527-540

[8] Coatsworth  Structures, Endowments and Institutions in the Economic History of Latin America Latin American Research Review, Vol. 40, No. 3 (2005) pp.126-144

[9] Engerman and Sokoloff Colonialism, Inequality, and Long-Run Paths of Development National Bureau of Economic Research Working Paper 11057 (2005)

[10] Ibid.