Inequality in the New World. Discussing the Institutionalist Approach

Term Paper, 2017

20 Pages, Grade: 1,3



1. Introduction

2. The Institutionalist Approach - A Literature Overview
2.1. Engermann and Sokoloff - Economic and Political Inequality
2.2. Acemoglu - Property Rights

3. Empirical Work

4. Conclusion

5. References

6. Appendix

Abstract: Current differences in development between North and Latin American countries led to different academic approaches explaining this phenomenon. One thesis is the institutionalist approach with two popular theories brought up by Daron Acemoglu and Engermann and Sokoloff differing in their definition on institutions. Further research confirms these theories in parts depending on their empirical methodology. My empirical work focusses on Engermann and Sokoloff, finding soft evidence for their claim of colonial activities influencing growth paths until today. I do not find evidence for colonial activities and slavery influencing modern income inequality.

Key Words: Latin America, Inequality, Slavery, Colonial Activity

1. Introduction

Nowadays, countries in the Americas offer a wide range of economic development. While Northern American countries (Canada and the U.S.) are highly developed, Latin American countries account for poorer economic performances at present. Furthermore, countries in Latin America such as Brazil and Chile are among the most unequal societies in the world (Bértola and Williamson, 2017). The discovery and colonization of the New World by Europeans in 1492 made countries in the Americas be part of a great natural experiment, thereby leading academic research to come up with various approaches explaining the different growth path of former colonies in the Americas like the geography hypothesis (Diamond, 1997; Sachs, 2000) or the cultural approach (North, 1988). This paper focusses on the institutionalist approach which links differences in growth paths with colonial institutions installed by European settlers in the Americas. Chapter 2 discusses the two most popular versions of this approach which differ in their definition of institutions and their empirical examination by further academic research. In Chapter 3 I shall adopt data and regressions from Bruhn and Gallego (2012) to develop deeper insights on the link between the different empirical results presented in Chapter 2. Chapter 4 gives a conclusion of the main findings. All tables I refer to can be found in the Appendix.

2. The Institutionalist Approach - A Literature Overview

The current differences in the developmental situations of countries in the Americas, particularly between North America and Latin America, led to the emergence of various approaches to address this phenomenon. For instance, one might ask why countries that were among the richest in pre-colonial and colonial times (such as Bermuda and Cuba) are now underdeveloped compared to previously poorer countries, such as the US and Canada. One of the most popular approaches focusses on the development of institutions in American countries over the course of their national history, linking colonial activities with contemporary economic outcomes. The following discussion gives an overview of the two main institutionalist approaches, which differ in their definition of institutions: Engermann and Sokoloff (henceforth ES) focus on the importance of economic and political inequality for prosperity, whereas Acemoglu focusses on secure property rights.

2.1. Engermann and Sokoloff - Economic and Political Inequality

The institutionalist approach by ES links the different growth paths of North and Latin America to the colonial activities the European conquerors engaged in when settling in the New World. In a series of papers (Engermann and Sokoloff, 1997, 2002, 2006) it is argued that different growth trajectories of American countries can be explained by different initial factor endowments, which in turn led to differences in production. To a large degree based on slavery or the exploitation of the indigenous population by the European colonizers, these differences caused enormous inequality, according to ES. Those regions in the New World endowed with climate and soil enabling the plantation of sugar and highly valued crops relied on a vast population of African slaves as large scale plantation and economies of scale demanded for high amounts of manual labor. For example, Brazil was the main recipient of African slaves as large scale plantation was particularly wide-spread in South America (Fujiwara et al., 2017). The intrinsic inequality of slavery enabled the creation of exclusive institutions that protected the privileges of the small elite restricting the rest of the society from fully participating in the economy. Due to these institutions, inequality became persistent over time even after the abolishment of slavery. The importance of factor endowments can also be seen in ES’ second type of colonies, which were rich on mineral resources and a large indigenous population. These colonies (mainly today’s Mexico and Peru) relied on the exploitation of their resources through mining and forcing the indigenous population to labor. Furthermore, colonizers adopted the pre-conquest social structures of tax and tribute. ES identify a couple of channels through which inequality became harmful for long-term growth. Firstly, they see economic inequality in dimensions such as wealth, income and commercial participation as a key institution hindering prosperity in comparison with Northern American colonies, where inclusive financial systems and relatively equal policies on land ownership enabled the extension of a franchise and diverse entrepreneurship. The second aspect focusses on inequality in human capital, under-provision of public goods restricting innovation, education and competition in Latin American colonies. Thirdly, restriction on voting rights depending on wealth or literacy made it almost impossible for the mass population to politically participate and to abandon the political power of the elites protecting their advantages. Against the backdrop of this line of argumentation, ES see a turning point in development in the late eighteenth century, when Northern American countries started industrializing while former rich colonies in Latin America missed their chance of growth due to the reasons mentioned above.

Further academic research investigates the ES thesis checking the influence of slavery on inequality and the resulting impacts on growth and development. For instance, Bruhn and Gallego use data on colonial activities for 345 regions in the Americas at the subnational level (Bruhn and Gallego, 2012). They distinguish between three types of colonial activities. Bad colonial activities are associated with economies of scale and the large-scale exploitation of labor force, which ES expect to have a bad influence on long-term development. The second category focusses on regions with a large native population with forced labor and economies of scale. They are called ugly colonial activities as they rely on the exploitation of native populations. The third category is called good colonial activities as they do not rely on economies of scale and were focused on regions with low colonial population density like North America (Bruhn and Gallego, 2012). Table I shows the main determinants of colonial activities that Bruhn and Gallego identify. Consistent with ES’ thesis on factor endowments, they find that several of the climate and geography variables have a significant influence on the existence of bad colonial activities. The same holds for ugly colonial activities, which are positively and significantly correlated with pre-colonial population density, thus confirming ES’ argumentation on the exploitation of the native population in crowded regions. Beyond that, Bruhn and Gallego find evidence for colonial activities influencing current GDP per capita (Table II), which they interpret as an instrument of development. They find that regions with bad colonial activities have a lower GDP per capita today than regions with no or good colonial activities. The same holds for ugly colonial activities at a slightly lower level. Since the predictions are confirmed by the empirics so far, Bruhn and Gallego also test which of the inequality channels proposed by ES account for the negative impact of colonial activities on the current level of development. Consequently, Table III plots colonial activities against log income inequality (as a measure of economic inequality), log schools per child and log literacy rate (as a measure of inequality in human capital) as well as the log seats in the lower house per voter (as a measure of political inequality). In contrast to ES, only inequality in political participation seems to be a significant channel linking colonial activities with today’s level of development.

Summing up, Bruhn and Gallego confirm the standard institutionalist approach by ES with inequality in political participation being the main channel that links the use of slavery with current economic underdevelopment.

Another paper by Nathan Nunn that partly approves ES uses data - in contrast to Bruhn and Gallego - at the cross-country level (Nunn, 2007). Nunn tests the chain of causality proposed by ES which can be summarized in this context as:

Source: Nunn, 2007

Using a sample of 29 former New World countries for which data on slavery and income data are available, Nunn first examines the impact of slavery on current development. He uses the natural log of GDP per capita in 2000 to measure current development, and the share of non-free persons of the total population as a measure of an economy's reliance on slavery. Table IV shows the results of the OLS estimation indicating a positive and significant correlation between slavery and modern economic development. At this point, the results are consistent with the predicted negative correlation by ES. But in contrast to ES and Bruhn and Gallego (2012), Nunn finds no evidence for large-scale plantation slavery being more detrimental to contemporary development levels than other forms of slavery. Next, Nunn examines the first part of ES’ chain of causality, according to which plantation slavery leads to economic inequality. Relying on the 1860 U.S. census, Nunn shows a strong historical persistence of economic inequality in the U.S., with slavery causing initial land inequality resulting in modern income inequality in terms of Gini-coefficients. This again contradicts Bruhn and Gallego's findings on the subnational level for a considerably larger sample. Nunn’s results suggest that slavery at the U.S. level not only caused contemporary underdevelopment, but also it increased income inequality. Scrutinizing the second implication of ES, Nunn finds no evidence for economic inequality affecting modern development. Income in 2000 is not correlated with either economic inequality, or land inequality in 1860, which repudiates the second part of ES' chain of causality (Table V).

In sum, Nunn’s estimations at the country and state level suggest that the positive relationship between slavery and economic inequality cannot explain slavery's negative impact on development as these two effects appear to be unrelated.

Another, entirely novel approach to the relation between slavery, inequality and development at the subnational level in the Americas was recently proposed by Fujiwara et al. (2017). The authors focus on Brazil, as it was the main recipient of African slaves and the last country in the Americas to abolish this institution in 1888. To identify the impact of slavery on income inequality and the channels leading to it, they introduce a region's distance to the so-called Tordesillas line as an assignment variable. The Tordesillas line, dating back to pre-colonial times, separated what is today Brazil into a Spanish and a Portuguese part, with the Spanish territory accounting for a smaller share of slaves relative to the total population. Another advantage arises due to the fact that different slavery institutions on both sides of the Tordesillas line can be exploited by holding local factors like geography and climate constant. Table VI shows the regression results for slavery and development outcomes using state fixed effects for Spanish and Portuguese Brazil. The number of slaves in 1872 does not appear to have a significant impact on GDP per capita on both Spanish and Portuguese Brazil. In fact, we see a positive impact of slavery on GDP per capita which is not significant. This contradicts the ES standard approach as well as the negative correlation between slavery and GDP per capita, which Bruhn and Gallego and Nunn find at the subnational level for the Americas as a whole and at the aggregate level, respectively. The influence of slavery on income inequality measured in Gini coefficients is positive and significant for both sides of the Tordesillas line. This result supports ES’ that slavery causes economic inequality and confirms Nunn’s findings at the aggregate level described above. Furthermore, the authors identify income and educational racial imbalances and public institutions as channels through which inequality persisted from colonial times until now. To put it briefly, average black households are poorer than white households today due the degree of slavery. While the selection of academic research considered thus far somewhat confirms the ES thesis in certain aspects and on different data samples, Williamson (2009) does not see evidence for slavery causing extraordinary inequality in Latin America. He uses a sample of estimated income Gini-coefficients for different world regions over a period of almost 2000 years. While he finds that countries' status a former colony is associated with a higher income inequality of almost 13 Gini points and population density is exacerbating inequality as well, his general findings do not match the results of the literature discussed above. The author divides his sample into pre-industrialized and industrialized countries and draws the conclusion that until Latin American countries industrialized in the so-called belleépoque around 1880, inequality was not extraordinarily high compared to other pre-industrialized countries around the world, with a short exception in the year 1790. Williamson sees the reasons for high inequality in Latin America in improving terms of trade over this period. The first globalization boom until the early 20th century made Latin American countries expand their primary goods export sectors, benefitting the countries’ elite that owned land, natural resources and infrastructure at that time. This point of view argues against the common thesis of persistent economic inequality due to colonial activities as discussed above, since it is a relatively recent phenomenon in Latin America compared to the rest of the world. Based on this line of argumentation, Pablo Astorga (2015) expands the view on the entire 20th century, not stopping with the belleépoque. He finds evidence for Latin American inequality being a relatively new phenomenon as it did not reach its peak in the belleépoque, but in the 1990s.

2.2. Acemoglu - Property Rights

Acemoglu et al. (2001) argue that European colonizers were engaged in two different colonial strategies. First, they settled in large numbers in regions with either a low settler mortality due to climatic conditions that prevented diseases, or small indigenous populations that were less prone to engage in warfare. These regions (Canada and the Northern U.S.) accommodated neo-European and inclusive institutions which allowed for secure property rights and limited governmental power. Furthermore, this settling strategy included relatively equal economic participation opportunities for the vast majority of the population, including access to finance, the right to acquire land and the ability to engage in self-employment or entrepreneurship.


Excerpt out of 20 pages


Inequality in the New World. Discussing the Institutionalist Approach
University of Bonn
Catalog Number
ISBN (eBook)
ISBN (Book)
File size
1784 KB
inequality, world, discussing, institutionalist, approach
Quote paper
Thomas Craemer (Author), 2017, Inequality in the New World. Discussing the Institutionalist Approach, Munich, GRIN Verlag,


  • No comments yet.
Read the ebook
Title: Inequality in the New World. Discussing the Institutionalist Approach

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free