Levels of economic development vary widely across countries. For example, in a sample of sixteen countries in the Americas, the richest country (the US) has fifteen times the GDP per capita of the poorest country (Honduras). Several recent papers have argued that these large differences in economic development have their roots in history, particularly in the colonial era (See Acemoglu et al., 2005 and Nunn, 2009 for reviews of the literature). This paper provides new empirical evidence for a relationship between colonial activities and labor supply at the beginning of the colonization period and current day economic development using within country data.
In particular, we empirically test, at the sub national level, two arguments present in the literature: the claim that different types of economic activities that colonizers engaged in led to different growth paths (Engerman and Sokoloff, 1997, 2002), and the claim that the colonizers often used the native population as an exploitable resource, which also led to different development paths for regions with different precolonial population density (Engerman and Sokoloff, 1997, 2002, Acemoglu et al., 2002).
Engerman and Sokoloff argue that depending on factor endowments, such as climate, geography, and precolonial population density, colonizers engaged either in \bad' or \good' colonial activities. Bad activities are plantation agriculture involving slavery and other forms of coerced labor (sugar, cotton, rice, and tobacco) and mining. These activities tended to create extractive institutions and encouraged fewer Europeans to settle in the area due to the fact that the production technology was inherently repressive.
Good activities, which included many other economic activities that did not rely on coerced labor, created inclusive institutions and encouraged more Europeans to settle since most individuals performing these activities stood on an equal footing. Good colonial activities thus created a stronger institutional environment which implies that, in the long run, areas with good colonial activities are more economically developed than areas with bad colonial activities.
Engerman and Sokoloff illustrate with a number of examples and summary statistics that the differences in colonial activities across countries translated into very different patterns of suffrage, public land, schooling policies, financial policies, and innovation policies. We add to this evidence by showing empirically that colonial activities are correlated with current day measures of economic development (log GDP per capita and poverty rates).
Based on history books, we collect data on economic activities performed during the colonial period in 332 different regions in 16 countries in the Americas. Each region is assigned three dummy variables summarizing whether it had predominantly good, bad, or no colonial activities.
We collect this data at the sub national level since the distinction between colonial activities is much sharper across regions within a country than it is across countries. Easterly and Levine (2003) examine at the country level whether the presence of certain crops and minerals is correlated with economic development. Whenever a country produced a certain crop or mineral in 1998-1999, they code a dummy to be equal to one for that crop/mineral.
This method does not allow for distinguishing between good and bad activities since most countries produce a range of crops and minerals (falling under good and bad). Moreover, good crops, such as maize are produced in almost all countries (68 out of 72 for maize).Within the countries in our sample, however, it is typically the case that the types of crops and minerals produced vary from region to region. Moreover, our sub national analysis is also made possible by the fact that levels of economic development vary almost as widely across states or regions within country as they do across countries (Table 1)".
We also collect data on precolonial population density at the sub national level. Acemoglu et al. and Engerman and Sokoloff point out that, in areas with high precolonial population density, the colonizers often used the native population as an exploitable resource, which also led to low long run development. We label this exploitation of labor an \ugly' activity. Using data for a sample of former colonies, Acemoglu et al. (2002) show that there is a negative correlation between pre colonization population density and current economic development at the country level.
Our results show that precolonial population density is also negatively correlated with current economic development at the region level within countries. Going from the 25th percentile in precolonial population density (00.97) to the 75th percentile (2.10) is associated with 22 percent lower GDP and 14 percentage points higher poverty rates. Our evidence also supports the notion that bad colonial activities did worse than areas with no colonial activities: these areas have 15 percent lower PPP GDP per capita and 15 percent higher poverty rates than areas with no colonial activities. Similarly, conditional on being colonized areas with bad colonial activities have 17 percent lower PPP GDP per capita than areas with good colonial activities.
This paper then extends Engerman and Sokoloffs argument by showing that having good colonial activities did not always lead to a better development path than having bad colonial activities. Instead, we argue and provide some historical examples that the technologies used in different areas with good activities were endogenous to the availability of a local labor force. In areas with high precolonial population density, good colonial activities tended to be performed with technologies that featured exploitation of labor, thus turning into an ugly activity.
Using regressions with interaction terms, we show that areas with good activities only have higher current levels of development than areas with bad activities if they coincide with low precolonial population density (values below the 60th percentile of the distribution of population density). Having good colonial activities together with high precolonial population density had approximately the same impact on long run economic development as having bad colonial activities.
In addition, these same regressions show that population density did not have a statistically significant effect in areas with bad economic activities. We argue that this is due to the fact that activities such as mining or sugar production had less flexibility in terms of technology adoption since the technologies depended heavily on economies of scale. Moreover, they depended less heavily on the availability of local labor because the profitability of these activities made it possible to import labor from other areas, using slavery, personal service, or the mita system.
Next, we examine the link between colonial activities and current day economic outcomes. Engerman and Sokoloff state that different colonial activities led to a different set of 'institutions,' which in turn led to different development paths. Their definition of institutions is quite broad and includes variables such as inequality, suffrage, schooling policies, financial policies, and innovation policies. Acemoglu et al (2001, 2002) argue that, depending on the local conditions, colonizers either set up \extractive states,' or \neo0European' states with strong emphasis on private property and checks against government power, in a given country.
The institutions associated with each of these types of states persisted over time and influence economic outcomes today. Even though they stress the role of property rights and checks against government power in their empirical analysis, Acemoglu et al argue that these variables are related with a cluster of institutions \including constraints on government expropriation, independent judiciary, property rights enforcement, and institutions providing equal access to education and ensuring civil liberties, that are important to encourage investment and growth' (Acemoglu et al. 2001, footnote 3). On the other hand, Glaeser et al. (2004) point out that the colonizers brought with them many other things, such as human capital, which could also explain the effect of history on current levels of development.
Given data availability, we investigate four different possible mechanisms that could link colonial activities to current development: economic inequality, education, political representation, and the ethnic composition of the population. Our results indicate that income inequality and human capital are not correlated with colonial activities and thus cannot explain the link between colonial activities and current levels of development.
In contrast, the results for political representation are mostly consistent with our reduced forms in terms of the signs, sizes, and significance of the effects, suggesting that political institutions may be the inter mediating factor between colonial activities and current economic development. The current ethnic composition of the population is not correlated with pre0colonial population density, but areas with bad colonial activities have a higher share of minorities today than areas with no activities and areas with good activities. Minorities may face discrimination, which could prevent them from achieving higher levels of productivity. The ethnic composition of the population could thus be an additional factor linking bad colonial activities to current levels of economic development.
The paper is organized as follows. Section 2 discusses the theoretical background. Section 3 gives historical examples for the theory. Section 4 describes the data. Section 5 analyzes the relationship between colonial activities and development. Section 6 investigates the mediating factors between colonial activities and development today and Section 7 concludes.
Download
PDF Ebook Good, Bad, and Ugly Colonial Activities: Do They Matter for Economic Development?
