Semester Long Journal (50%)

Pe0512

FuentesFEB2016OriginsofInstitutionsandInequalityinLatinAmerica1.pdf

Home >Political Science homework help >Semester Long Journal (50%)

Origins of Institutions and Inequality in Latin

America

Stefania Paredes Fuentes∗

[email protected] Department of Economics

University of Warwick

February 2016

Abstract

Colonial institutions are considered to be an important determi-

nant of post-colonial development with many scholars concluding that

former British colonies inherited better institutions. This paper pro-

vides an empirical analysis of the origins of Latin American institutions

and shows that colonial features have only indirect effects on institu-

tions through their effects on economic and political redistribution. It

also explains the role of British intervention in the region. After in-

dependence, Britain exerted an important economic influence in Latin

America which contributed significantly to the consolidation of the

economic and political institutions in the region. Another aspect of

the region that affect the institutions is the economic dependence on

natural resources which has a negative effect on institutions. Peru,

Bolivia, Uruguay and Costa Rica are used as examples to explain how

colonial inheritance, natural resources, inequality, institutions interact

in the Latin American context. (JEL F54, O17, O54, P48)

∗Author details: [email protected]; +44(0) 2476150045, Social Sciences

Building S2.121

1 Introduction

From the 16th to the mid-20th century, several European powers established

colonies in Asia, Africa and America. Europeans settled and took control

of these territories. The impact of colonisation is considered to be immense

and pervasive. Colonial practices spur the spread of colonial languages,

religion, culture and territorial boundaries and scholars have proposed sev-

eral theories on the colonial origins of institutions in developing countries.

This literature is based on two hypotheses. One is that a series of colonial

aspects shaped the character of those institutions created after the indepen-

dence (i.e. early institutions). The second one is that current institutions

strongly reflect early ones due to institutional persistence (Section 2 reviews

this literature).

Most of these empirical studies use large samples composed by all former

colonies (for which data is available). While this is helpful to maximise the

number of observations in cross-country regressions, it neglects some basic

differences among regions. The discovery and exploration of the Americas

was mainly an Iberian (Spanish and Portughese) project. However, after the

Napoleonic wars and the raise of Britain as coloniser power, Iberian crowns

lost their colonies. Most of the colonised territories in Latin America de-

clared their independence in the 1820s. This is over a century earlier than

the decolonisation process of British and French colonies in other regions in

the 1960s. So, even if the colonial past may have shaped the region’s early

institutions, the greater passage of time has given greater scope for institu-

tional change meaning that current institutions may not just reflect the ones

created at the time of independence as hypothesised by previous literature

(second hypothesis), but other factors and post-colonial events may have

also played a role in influencing the character of current institutions.

The empirical analysis in Section 3 shows that Latin American institutions

are the outcome of a more complex interaction between colonial history, in-

herited colonial inequality and post-colonial events. These interactions help

to explain the current poor institutional setting and high levels of inequality

in the region.

This work proceeds by testing the most popular theories of colonial origins

of institutions based on four colonial aspects: coloniser identity, size of Euro-

pean settlements, organisation of pre-colonial populations, and colonial re-

source endowment. Many scholars have suggested that British did better, i.e.

British former colonies inherited better institutions than non-British ones.

However, these scholars do not elucidate the specific mechanisms through

which British colonial rule exerts a positive effect. This work investigates

these mechanisms. Although Latin American countries were mainly Iberian

colonies, Britain did show interest in Latin America and, after independence,

it exerted a sort of indirect colonial rule through foreign investments and

trade with the region. This intervention has two effects on institutions. One

negative direct effect derives from the problems of economic management

and rent seeking opportunities created by the inflow of British capital. A

second positive indirect effect goes through redistribution. In those areas

where British capital was invested in productive activities, this may have

contributed to the development of a middle class (through creation of job

opportunities) improving thus redistribution and the quality of institutions.

The net effect of British intervention in Latin America depends on which of

these effects prevailed.

Finally, Latin American institutions cannot be explained without consider-

ing the role of natural resources. The colonisation of Latin America was

driven by the discovery of silver and gold. The results show that resource-

rich areas during colonial times inherited higher levels of inequality and

therefore worse institutions. Moreover, after independence, countries with

larger resource endowments attracted British investments increasing the

rent-seeking opportunities and increasing the negative direct effect of British

intervention on institutions. There is no evidence that European settlements

and native population have any impact on current institutions (as hypothe-

sised by previous literature) but British intervention and natural resources

seem to play role.

The results also shed light on the role of natural resource endowments.

Resource-rich colonial territories inherited higher levels of inequality and

therefore worse institutions. After the independence, British investments in

countries with a highly skewed distribution of resources increase the rent-

seeking opportunities for the ruling elite; in these countries a negative direct

effect of British intervention on institutions prevails. In countries with lit-

tle/no colonial resources, British investments provided the needed resources

for economic development but without a strong dominant political and eco-

nomic elite, these investments seem to have benefited a larger share of the

population.

Finally, most Latin American economies were resource dependent even after

the independence. Post-colonial oil and gas discoveries have a negative direct

impact on institutions.

Section 4 presents some historical evidence for four Latin American countries

that supports the empirical findings. The good performance of Costa Rica

and Uruguay – backwater territories during the colonial period due to the

lack of attractive natural resources –contrast to the poor institutions and

high levels of inequality of Peru and Bolivia, both mineral based economies

before and after independence. Section 5 summarises and concludes.

2 On the Origins of Institutions

Colonialism is perhaps the single most salient historical event of the modern

era and as such, it has influenced several studies investigating the origins

of institutions in the developing world. The most common colonial aspects

considered to affect institutions of former colonies are: the identity of the

coloniser power, size of European settlements, native population, and min-

eral and agricultural resources.

Most of the studies looking at how the identity of the coloniser power (e.g.

Spanish, French, British) affects institutions in colonised territories con-

clude that British did better. For instance, La Porta et al. (1998, 1999)

argue that former British colonies have better institutions due to the inheri-

tance of common law legal systems that are linked to protection of property

rights1. The supremacy of British legacy has also been related to tax pol-

1This is because common law developed in England as a defence of Parliament and property owners against the attempts by the sovereign to regulate and expropriate them;

icy (Thirsk, 1997), labour market regulation (Stotsky and WoldeMariam,

1997; Botero et al., 2004), contract enforcement (Djankov et al., 2003), in-

vestments on education and school enrolment (Grier, 1999; Bertocchi and

Canova, 2002).

In their seminal contribution, Acemoglu, Johnson and Robinson (2001) ar-

gue that European colonisers created good institutions in those territories

where they settled down. Good institutions are defined as those ones that

protect property rights from possible government expropriation and are op-

posed to extractive institutions which are considered bad for development

because they did not provide protection for private property and checks and

balances against government expropriation. The aim of extractive institu-

tions was to transfer as much of the colony’s resources to the coloniser crown

and were created in territories where Europeans did not find favourable

conditions to settle due to adverse climatic conditions. Therefore, large Eu-

ropean settlements are related to good institutions (similar conclusion in

Easterly and Levine (2012)) .

An implicit assumption in Acemoglu, Johnson, and Robinson’s works, is that

all European colonisers had the same institutions. Mahoney (2010) argues

that this is not the case and identifies two type of European colonisers: mer-

cantilist (e.g. Spanish, Portuguese and French) and liberals (e.g. British).

While during the colonial period, European institutions were mostly based

on mercantilist principles, this was also the period when liberalism started

as major doctrine and intellectual endeavour in Europe and some countries

approached this doctrine earlier than others, Britain being the first. These

differences may have impacted the type of institutions that colonisers set in

their colonies. According to Mahoney, mercantilists settled and implanted

their institutions in territories with large populations due to the labour ex-

ploitation opportunities. Mercantilist institutions were bad for economic

development while liberal ones promoted the creation of good institutions.

This is because liberal colonisers preferred less complex societies where they

sought possibilities for capitalist accumulation. The organisation of pre-

colonial societies has thus also been considered relevant to understand the

civil law developed more as an instrument used by the sovereign for State building and controlling economic life.

type of institutions created in a territory (Acemoglu et al., 2002; Baker

et al., 2008; Mahoney, 2010). Larger pre-colonial populations required a

more complex organisation and are related to worse institutions than those

created in scarcely populated areas.

Finally, the exploration and further colonisation of Latin America was driven

by the search of mineral wealth and other commodities. The search of gold

and silver carried the Spaniards far and wide across the Americas, contribut-

ing much to the amazing rapidity to which they explored and settled the

continent: on the promise of gold Spaniards settled the Caribbean; finding

little in the islands, they moved to the Isthmus, then to New Spain (Mex-

ico), then to Peru. The presence of mineral resources was in fact the key

determinant of the level of colonisation, at least in a first phase with Mexico,

Peru and Bolivia (where large deposits of silver were found) as key centres

of the colonial power. Following the increased demand of tropical crops in

Europe, agriculture plantation for export became very profitable. In fact,

it was in the context of plantation agriculture and sugar that the Brazilian

colonial society was formed and sugar production emerges in the Caribbean

as an alternative to the rapidly depleted mining industry.

Different natural resources (mining or plantation agriculture) required dif-

ferent ways of organisation and this is reflected in the labour organisation

and land distribution. Mineral resources were generally found in areas with

large native populations (such as Aztecs in Mexico and Incas in Peru), but

in less populated areas American natives were transferred under different

systems (as in Potosi-Bolivia). The discovery of gold or silver in a territory

translated into several direct state interventions in favour of the mining sec-

tor. Due to the initial high investments required in tropical plantations

Crops such as sugar, coffee and cocoa are most efficiently produced on large

estates2 and require a high initial investment and these productions were

mainly supported by the import of African slaves.

Institutions put in place by colonisers in rich-endowed territories aimed to

systematically extract the surplus from the (human and natural) resources

2Scale-neutral food crops such as wheat, rice and maize are historically produced on modest size plots (Kawagoe et al., 1985)

to enrich the coloniser power and prevented the labour force from partaking

the benefits and possibilities from the economic wealth they produced (En-

german and Sokoloff, 1997, 2002). The wealth allocation that arises from the

exploitation of these resources facilitates the creation of different endowed

and motivated groups that could steer the mode of accumulation within

the society as a whole. Landowning elites developed coercive labour mar-

ket institutions such as serfdom, slavery, or permanent debt peonage (Do-

mar, 1970), while trade restrictions and monopolistic structures favoured

the creation of a wealthy elite of merchants. These merchants were often

tied through investments to landowners and together they trapped capital,

stifled investment and entrepreneurial activity, and thus blocked develop-

ment.

Independently of which colonial factor is found relevant for explaining insti-

tutions, there are two implicit assumptions in these theories. The first is that

distinctive colonial features shaped early institutions i.e. the institutions put

in place after the independence of these territories. The second (more or less

implicit) assumption is that current institutions still reflect early ones. This

is the due to a specific characteristic of institutions: their persistence over

time. Institutions tend to change slowly in response to social, political and

economic individuals’ behaviours and interactions.

Figures 1 to 4 show the relationship between the colonial features and two

measures of early institutions in Latin America (democracy and political

constraints - definitions in Appendix). Given the conclusion that British

did better, Figure 1 compares institutions at the time of independence in

British and non-British colonies and shows that, in average, British former

colonies have better institutions than non-British ones.

Figure 2 shows the relationship between the two measures of institutions

and the size of European settlements in 1800. Unlike Acemoglu, Johnson

and Robinson argue, the relationship between European settlements and

institutions seem to be negative. A large number of European settlers is not

enough to secure good institutions in a colonised territory and other factors

may affect the creation of these institutions.

Figure 1: Coloniser Identity

Figure 2: European Settlements

(a) Index of Democracy (b) Political Constraints

Figure 3 shows the relationship between the pre-colonial population density

(in 1500) and early institutions. Unlike argued by Mahoney, we can observe

that there is a (weak) positive correlation between institutions and native

populations.

Figure 4 shows that in average, those Latin American and Caribbean coun-

tries that did not exploit mineral resources during the colonial times had

better early institutions. and that those areas with more suitable lands for

the production of sugar are correlated with worse early institutions.

Figure 3: Pre-colonial populations

(a) Index of Democracy (b) Political Constraints

Figure 4: Colonial Natural Resources

(a) Minerals (b) Sugar crops

Figure 5: Current vs Early Institutions in Latin America

(a) Index of Democracy (b) Political Constraints

Finally, Figure 5 plots the correlation between early and current Latin Amer-

ican institutions. We can observe that there is a weak correlation between

early (at the time of independence) and current institutions in this region.

This result should not surprise much: as shown in Table 1, most Latin

American countries obtained independence in the first half of the 19th cen-

tury. This is over a century earlier than the majority of the colonised world

which acquired independence only during the decolonisation process that

took place in the second half of the 20th century. The greater passage of

time in Latin America has given, in effect, greater scope for institutional

change, and although important, colonial factors cannot be considered as

the only factors that shaped institutions. Post-colonial events may have

also played a role in influencing the character of current institutions.

Table 1: Independence of Latin America and the Caribbean

Period of Independence

Independence Early Late Overseas territories from ( before1830) (after 1960) (no independent)

Spain/ Argentina Portugal Bolivia

Brazil Chile

Colombia Costa Rica

Cubaa

Dominican Rep. Ecuador

El Salvador Guatemala Honduras Mexico

Nicaragua Panama

Paraguay Peru

Puerto Rico Uruguay

Venezuela

France Haiti French Guiana Guadeloupe Martinique St. Martin

St. Barthlemy

Netherlands Netherlands Antillesb Bonaire Aruba Sint Eustatius

Curacao Saba Sint Maarten

Suriname

Britain Antigua & Barbuda Anguilla Bahamas British Virgin Islands Barbados Cayman Islands

Belize Monserrat Dominica Turks and Caicos Is. Grenada Guyana Jamaica

St. Kitts and Nevis St. Vincent and the Gren.

Trinidad and Tobago

Notes: a. Cuba obtained its independence in 1902 b. Netherlands Antilles dissolved in 2010. After dissolution, Bonaire, Sint Eustatius

and Saba became special municipalities of the Netherlands, while Curacao and

Sint Maarten became constituent countries within the Kingdom of the Netherlands,

along the lines of Aruba, which separated from the Netherlands Antilles in 1986.

2.1 Post-colonial Latin America: British Intervention

As previously explained, there is a large consensus that British did better;

that is, former British colonies inherited better institutions compared to

non-British ones3. Few considerations have to be done for the Latin Amer-

ican case. First, British were late-comers in the process of colonisation of

American territories that were discovered, explored and colonised by Iberian

crowns since late-15th century. This means that, unlike argued by Mahoney,

British colonisers could not choose which territories to colonise, but they

were left with the remnants of the Iberian ones. Second, Table 1 shows that

most of the former British colonies acquired their independence much later

compared to Iberian colonies which makes difficult to establish whether the

positive effect of being a British colony is the result of a late independence

rather than being a former British colony. In fact, another question that has

arisen when attempting to assess the impact of colonial rule is whether a

longer period of colonial rule was better or worse for economic development;

nonetheless, for Latin America, it is difficult to disentangle the relationship

between Britain colonies and late independence.

Furthermore, the British intervention in the region, did not stop at the colo-

nial period. A broad literature in economic history explains how British

intervention in the post-independence period played an key role in the

economies of newly independent countries. After the independence, Latin

American countries were freed from the obligation to sell their primary prod-

ucts through Iberian outlets which favoured an increase of trade with other

European powers. The 1870s and 1880s saw a sharp increase in trade be-

tween Britain and Latin America, as the region became fully incorporated

into the Atlantic economy (Platt, 1972).

Moreover, in the late 19th century, the centre of dynamism within the British

economy moved from the manufacturing areas towards the commercial and

financial interests, Latin American governments tapped the London bond

markets and the flow of portfolio capital was followed by direct foreign in-

3This is also the implicit assumption in Mahoney’s argument: given the more liberal background of the British system, they established institutions that promoted a more capitalist accumulation that is linked to higher development.

vestment (Victor Bulmer-Thomas, 1998). First, Latin American economies

approached the international capital markets in order to finance indepen-

dence wars and finance public debt. Following the high political instability

during the post-independence period a wave of defaults ensued, with all

bond issues in default by 1827 (Rippy, 1959). Most countries remained in

default for decades, and new flows of capital started to circulate only during

1850s4. The macroeconomic and financial crisis produced a second wave

of defaults that spread over the region in the 1870s. With the recovery of

trade in the 1880s a new and bigger borrowing boom began. Capital inflows

were mostly concentrated in favour of those countries with new booming

trade sectors (della Paolera and Taylor, 2012). This once again ended with

an economic crisis in 1890s (which affected mainly the greater economies of

Brazil and Argentina).

Therefore, the British impact in Latin America went through two channels.

First, Britain was one of the main destiny markets of Latin American goods

in the nineteenth century. Second, the massive growth in British investment

which occurred after the independence wars totally redefined the nature of

Britain’s relations with Latin America. On the eve of World War I, British

investments in Argentina were the second largest group of investments made

by British investors in a foreign country (US being the largest).

2.2 Other Factors that May Affect Institutions: Natural Re-

sources and Inequality

There is of course more than colonial history. With almost two centuries

since the region’s independence, we need to consider other factors that may

have shaped Latin American institutions. This paper will analyse two (per-

haps the most common) distinctive features of Latin America: natural re-

sources and levels of inequality.

Resource wealth has played a key role in the participation of Latin American

economies on the international market before and after the colonisation.

4First, Brazil, Argentina, Paraguay, Uruguay, Chile accessed to new loans, followed by Costa Rica, Guatemala, Honduras, Bolivia, Peru.

In fact, Engerman and Sokoloff (1997, 2002) link colonial resource wealth

to poor initial institutions, but Latin America still possesses large natural

resource endowments and the export sector in many of the economies in the

region is concentrated on natural resources. In the past decade, commodities

accounted for 52 percent of the region’s exports according to the World

Bank. This is down from 86% in the 1970s, but over the same period

the figure in East Asia and the Pacific fell from 94 to 30% (Sinnot et al.,

2010).

A number of scholars argue that the exploitation and management of natural

resources have negative effects on the institutions of the producer countries.

These rents tend to be large, volatile, geographically concentrated, and con-

trolled by the government. Together these features have important conse-

quences on basic functions of the government. For instance, rentier effects

are associated with a high proportion of government revenue originating

from resource rents. The consequent fiscal volatility may create an unfor-

tunate political dynamic that ratchets up expenditures in booms to levels

that cannot be efficiently absorbed or sustained over time, with a stop-go

pattern of public expenditure that reduces the quality of public investment

and services and thus limits growth potential. In addition, much of a gov-

ernment’s fiscal strength comes from its capacity to extract taxes from the

population, a capacity that often takes decades to develop. A government

that fails to develop this ability may also be unable to establish the type

of bureaucracy that can provide effective public goods, and ameliorate so-

cial conflicts (Mahdavy, 1970; Beblawi, 1987; Ross, 2003; Fearon and Laitin,

2003).

Some scholars offer a more nuanced view of the role of natural resources on

political regimes. Perhaps, the best attempt to address the possibility of

conditional effects of resources on political regimes (rather than just con-

sidering one direct negative effect) is offered by Dunning (2008) who claims

that natural resources may have both democratic and authoritarian effects

and the key task is to understand variables or structural factors that tend

to privilege one of these effects. The conflict over the redistribution of the

resource rents does foster authoritarian desires, but this is only one way

how resources may affect the political regime. In societies with substantial

inequality of assets (different from natural resources), a resource boom may

help to mitigate the negative impact of inequality and therefore strengthens

democracy.

There are also indirect effects on institutions through redistribution. A

windfall of resource rents can generate conflicts over redistribution which

may provide incentives for politicians and/or ruling elites to suppress democ-

racy in order to take possession of these rents. This will thus affect the foun-

dations of political regimes in favour of more authoritarian ones (Sachs and

Warner, 1995; Ross, 2001; Boix, 2003; Jensen and Wantchekon, 2004).

In fact, inequality is another feature of these region’s economies. Despite the

decrease in inequality levels since 2000, the region is still the most unequal

in the world. The richest one-tenth of the population in Latin America

earns 50% of total income, while the poorest tenth earns only 2.5 percent.

Using the Gini index, the inequality in the region measures 50 percent in

the period 2000-09, this is higher than the rest of developed and developing

areas of the world (World Bank, 2011).

There is a large consensus in literature on the relationship between inequal-

ity and institutions: countries with poor institutions tend to have large levels

of inequality. Hoff and Stiglitz (2004) suggest that an equal distribution of

income is a more fertile ground for good institutions; while Easterly (2001)

and Keefer and Knack (2002) empirically show that social polarisation neg-

atively affects institutional quality. Lipset (1959); Rubinson and Quinlan

(1977); Muller (1988) and Boix (2003) argues that wealth distribution has

a positive role on the democratisation of a country.

In particular, Boix shows that, in societies with high levels of asset specificity

(e.g. with big landowners), the demand for redistribution increases and the

potential level of transfers becomes larger (which would make the elites worse

off), this fosters the authoritarian inclinations of the wealthy and decreases

thus the probability of democratisation. Therefore, if the political power is

in the hands of few, the small wealthy elite refuses the implementation of

any change that redistributes economic power. Moreover, high inequality

will also affect the survival of democracy. Democracies tend to survive

only if the regime contributes to narrow the gap between rich and poor

(Friedman, 2002); in fact, the greater challenge for unequal countries is to

sustain democracy once it is established (Houle, 2009).

While it is plausible that inequality plays a part in blocking the adoption

of good institutions, the reverse holds as well, so that poor institutional

quality results in higher degree of inequality. The analysis in this paper will

suggest in fact that inequality and poor institutional quality may reinforce

each other, indicating double feedback between the two.

Institutions, inequality and natural resources are all considered to influence

economic development. However, none of these features on their own can

explain the problems of Latin America. The social structures, the distri-

bution of power and wealth, the role and strength of its elites, and the

complex, often painful process of state-building, in combination with the

legacy of colonial times and the economic and political difficulties that the

newly independent states have in positioning themselves on the world stage,

have all been decisive factors and all have something to do with the suc-

cesses and failures of Latin American economies. Understanding how these

variables interact may benefit the explanation on how institutions evolve in

Latin America and the rest of the paper aims to contribute towards this

explanation.

3 Empirical Analysis

This section offers an empirical analysis of how colonial features, post-

colonial British intervention features along with inequality levels and re-

source endowments interact and affect the creation and further development

of institutions in Latin America. Institutions and inequality are considered

to be simultaneously determined; we can expect that those countries with

better institutions would also have lower levels of inequality, and that at the

same time a better distribution of resources will be translated into better

division of power and therefore better political institutions.

3.1 Data

Data definitions and sources are provided in the appendix. The depen-

dent variables are institutions and distribution of resources. Institutions is

measured using variables for Democracy, Autocracy and Constraints on the

Executive Power from the Polity IV Project. Distribution is measured by

the percentage of family farms from Vanhanen (2003) defined as the area of

family farms as a percentage of the total area of holdings – a family farm em-

ploys no more than four people and the family owns and cultivates the land.

More family farms thus represent a better distribution of these resources and

therefore higher values of this variable are related to lower inequality5.

Four variables capture the colonial inheritance of these countries. First, fol-

lowing the work of Acemoglu et al. (2001), European settlements is measured

by the percentage of Europeans in colonised territories in 18006. Second,

there is a measure of the population density in 1500 as a proxy of pre-colonial

native population. Finally, two variables are used to analyse the effect of

colonial resource endowment. One captures those countries endowed with

gold and silver mines by using a dummy variable that assumes the value of

1 if the country’s main economic activity at the time of the colonial period

was based on the exploitation of either gold or silver and 0 otherwise. The

second variable considers those countries that were specialised in the pro-

duction of cash-crops. For this purpose a measure of the land suitability for

the production of sugar is used7 (from the FAO database).

The impact of British intervention in Latin America after the independence

is measured by the level of trade of these countries with Britain, and the

5Several measures of inequality have been proposed, however the share of family farms is the only measure of inequality that is consistently available through time for a large sample of the countries considered and in the period analysed. Moreover, given the key role that land distribution have had since colonial times in Latin America, this is the most adequate measure to study how inequality affects institutions in the region.

6Acemoglu, Johnson and Robinson (2001) consider European settlements in 1900, how- ever Latin American countries acquired independence in 1820s which makes European settlements in 1800 a better measure for this region.

7sugar was one of the main cash-crops of interest for European markets and the pro- duction of sugar dictates the distribution of land and labour organisation in most of the producer territories.

British investments at the beginning of the century (Paish, 1909). Trade

with Britain is given by the average exports of Latin American economies to

Britain weighted for the total exports for the period 1898-1906. This variable

is constructed using data from the Annual Statement of Trade of the United

Kingdom with Foreign Countries and British Possessions for various years

(Statistical Office, 1906). British investments in Latin America considers

the average of these investments in the period 1905-1911 (data from Paish

(1909)).

Several variables have been used in literature to measure natural resources

and in particular to investigate whether resource-rich countries are actually

cursed. Based on the work of Sachs and Warner (1995) commonly used

variables are the ratio of resource exports to GDP and the ratio of resource

exports to total exports. Brunnschweiler and Bulte (2008); Ross (2006) and

Dunning (2008) argue that these variables measure dependence rather than

abundance therefore they are not independent of economic policies and in-

stitutions (i.e. they are endogenous) and prefer a measure of oil rents per

capita (based on the work of Hamilton and Clemens (1999)). This measure

provides an estimate of the value of a wide range of natural resources, net

of production costs and a return to capital, giving an approximation of the

size of the rents available for public spending. Although this variable pro-

vides a better measure of resource abundance, it is not without problems.

Extraction costs are based on estimates for a single observation in 1990s,

and costs for other years are obtained using a GDP deflator, when no data

on extraction costs are available for a country, the extraction costs for a

neighbour country are used (Ross, 2006).

For this analysis I use a measure that captures when a resource is discovered.

This is a binary variable for the discoveries of oil and natural gas. It assumes

the value of 1 if there has been a discovery in the 5-year period considered,

0 otherwise (constructed from the data provided by Lujala et al. (2007) on

oil discoveries on- and off-shore). The advantages of this variable are two-

fold, (i) there is a larger number of observations (across country and time),

and (ii) it is not subject to the common endogeneity issues. In addition,

the effects of natural resources on institutions may start well before these

resources start producing rents. The mere discovery of a mineral resource

might be a source of rent-seeking behaviour from the governing elites in

order to guarantee an early appropriation of future rents (e.g. guaranteeing

exploration and extraction rights under the promises of future economic

favours).

3.2 Results

The empirical analysis estimates the following model:

Insti,t = α1 + α2Disti,t + α3BRi + α4Nat Resi,t + α5COLi + �it (1)

Disti,t = β1 + β2Insti,t + β3BRi + β4Nat Resi,t + β5COLi + υit (2)

Equation (1) considers Distribution (Dist) as independent variable for ex-

plaining institutions (Inst) while Equation (2) uses institutions as explana-

tory variable for asset distribution. In addition, both regressions use vari-

ables that capture the post-colonial British influence through trade and in-

vestments (BR), the natural resources (Nat Res) and the different colonial

factors (COL).

Table 2 shows the results for the analysis of the relationship between institu-

tions and inequality. In Panel A, a General Least Squares (GLS) estimation

is used to explain institutions (democracy, autocracy and constraints on the

executive) as function of land redistribution (family farms ). The coefficients

are all statistically significant and with the expected sign. In particular,

higher democracy and constraints on the executive are positively related to

a higher land redistribution, while autocracy has a negative sign. In Panel B

land redistribution is considered as function of institutions. The coefficients

confirm the previous findings. This suggests a bilateral causality between

institutions and inequality.

In order to deal with the endogeneity problems caused by the relation-

ship between inequality and institutions a Hausman-Taylor estimator (a

transformed random effect model with instrument variables) is used. The

Hausman-Taylor estimator deals with endogeneity issues while distinguishes

between time-varying and time-invariant regressors8.

8Another approach could be to consider the lags of the endogenous variables into

Table 2: Institutions and Inequality

Panel A - Institutions as dependent variable

Explanatory Contraints on Variable Democracy Autocracy the Executive

Family Farms .683*** -.598*** .805*** (.131) (.123) (.121)

constant .190*** .401*** .249*** (.026) (.025) (.026)

n 20 20 20 T 20 20 20

Wald χ2 27.34*** 23.79*** 43.88***

Panel B - Family Farms as dependent variable

Explanatory Variables (a) (b) (c)

Democracy .098*** (.011)

Autocracy -.086*** (.014)

Contraints on .110*** the Executive (.012)

constant .138*** .200*** .126*** (.005) (.006) (.006)

n 20 20 20 T 20 20 20

Wald χ2 73.53*** 36.74*** .87.16***

notes: Robust standard errors in parentheses ***, **; significance at 1 and 5% respectively

Table 3 shows the results of the Hausman-Taylor estimator for regression

(1) in which institutions is the dependent variable. In all the specifications,

family farms is statistically significant and has a positive sign. That is,

a better distribution of land has a positive effect on institutions. The oil

and gas discoveries variable has a statistically significant negative effect on

institutions which provides support to the literature on ’resource curse’ i.e.

the exploitation of natural resources in Latin America harms development

(through their effect on institutions).

The coefficient for trade with Britain is positive and statistically significant.

This suggests that those countries with more commercial tights with Britain

have better institutions. However, the coefficient for British investments is

negative. While these results may seem contradictory, we can expect that

trade brings wealth to the economy through the support of productive activ-

ities that promote job creation and improve living standards for workers. On

the other hand, British capital could be intercepted by political elites and

used to support their economic dominance. In addition, none of the colonial

factors seem to have a statistically significant effect on institutions. This

result rejects most of the common theories of colonial origins of institutions

for Latin America.

Finally, institutions are themselves endogenous and tend to improve with in-

come, so that richer countries tend to have/afford better institutions. There-

fore, columns (c), (c’) and (c”) of Table 3 include a variable for the initial

level of GDP. This is statistically significant and with the expected sign,

suggesting that there is statistical evidence that richer countries can afford

better institutions. However, the signs and significance of the other variables

remain unchanged so the main conclusions still hold.

Equations (1) and (2). Nonetheless, the validity of the lag variables as instruments is questionable due to the high persistence of these institutions and inequality. In partic- ular, we need to assume that E[�it|Ineqis] = 0 and E[υit|Instis] = 0 for all t > s (but not otherwise) in order for second- and higher-order lags of the endogenous variables to be good instruments in the estimation of our model. Nonetheless, if our endogenous vari- ables display persistence over time, their lagged levels will be poor instruments. A second approach could be to find strictly exogenous instruments. Our colonial variables could be considered as good instruments because they are not subject to reverse causality, nonethe- less, they suffer from the drawback that they do not vary over time, so these cannot be used in a panel framework.

T a b

le 3 : H a u s m a n -T

a y l o r E s t im

a t o r f o r In

s t it u t io n s

D e p

e n

d e n t

V a ri

a b

le s:

In st

it u

ti o n

s E

x p

la n

a to

ry D

e m

o c ra

c y

A u

to c ra

c y

C o n

st ra

in ts

o n

th e

E x e c .

V a ri

a b

le s

(a )

(b )

(c )

(a ’)

(b ’)

(c ’)

(a ” )

(b ” )

(c ” )

T im

e v a r ia n t E n d o g e n o u s

F a m

il y

1 .3

2 * * *

1 .3

1 * * *

.4 8 3 * *

-. 7 5 0 * * *

-. 7 4 5 * * *

-. 3 3 3 *

1 .3

5 * * *

1 .3

4 * * *

.8 4 9 * * *

F a rm

s (.

1 6 4 )

(. 1 6 5 )

(. 2 2 1 )

(. 1 6 5 )

(. 1 6 6 )

(. 2 3 0 )

(. 1 6 8 )

(. 2 3 9 )

(i n

it ia

l) G

D P

.0 7 0 * * *

-. 0 4 5 * * *

.0 4 9 * * *

p e r

c a p

it a

(. 0 1 1 )

(. 0 1 2 )

T im

e v a r ia n t E x o g e n o u s

O il

a n

d g a s

-. 0 8 7 * *

-. 0 8 9 * *

-. 0 5 1 *

.0 9 9 * *

.0 9 1 * *

-. 0 8 5 * *

-. 0 8 7 * *

-. 0 6 9 *

d is

c o v e ri

e s

(. 0 4 3 )

(. 0 4 4 )

(. 0 4 2 )

(. 0 4 3 )

(. 0 4 5 )

(. 0 4 3 )

(. 0 4 4 )

(. 0 4 7 )

T im

e in v a r ia n t E x o g e n o u s

(l o g )

T ra

d e

w it

h .0

4 0 * *

.0 4 0 * *

.0 5 1 * *

-. 0 3 2 * *

-. 0 4 6 * * *

.0 2 2

.0 2 5

.0 3 6 *

B ri

ta in

la te

1 9 th

(. 0 2 0 )

(. 0 2 3 )

(. 0 1 4 )

(. 0 1 7 )

(. 0 1 8 )

(. 0 1 7 )

(. 0 2 1 )

(l o g )

B ri

ti sh

In v e st

. -.

1 1 0 * * *

-. 1 0 7 * * *

-. 0 7 0 * *

.0 5 9 * * *

.0 3 5 *

-. 0 9 7 * * *

-. 0 9 5 * * *

-. 0 6 9 * *

e a rl

y 2 0 th

(. 2 8 2 )

(. 0 2 8 )

(. 0 3 2 )

(. 0 1 9 )

(. 0 2 3 )

(. 0 2 6 )

(. 0 2 4 )

(. 0 2 9 )

M in

e ra

l .0

1 1

.0 2 1

-. 0 8 9

-. 0 4 5

-. 0 4 6

-. 0 0 6

.0 5 5

.0 3 6

.0 0 4

c o lo

n ia

l c e n tr

e (.

0 7 6 )

(. 0 7 9 )

(. 0 8 6 )

(. 0 5 2 )

(. 0 5 6 )

(. 0 6 4 )

(. 0 6 9 )

(. 0 6 8 )

(. 0 7 8 )

S u

g a r

.1 0 3 * *

.1 0 7 * *

.1 0 8 * *

-. 0 4 9

-. 0 5 3

.0 3 7

.0 6 5

.0 5 1

su it

a b

il it

y (.

0 4 4 )

(. 0 4 8 )

(. 0 4 9 )

(. 0 3 0 )

(. 0 3 4 )

(. 0 3 6 )

(. 0 4 0 )

(. 0 4 1 )

(. 0 4 4 )

E a rl

y E

u ro

p e a n

-. 3 5 5

.0 1 0

-. 7 9 0 *

se tt

le m

e n ts

(. 5 3 2 )

(. 3 7 4 )

(. 4 5 6 )

n a ti

v e

p o p

u la

ti o n

-. 0 2 6

.0 0 2

-. 0 0 1

(. 0 3 5 )

(. 0 2 5 )

(. 0 3 0 )

c o n

st a n t

.0 4 0

.0 0 2

-. 0 4 7

.5 3 1 * * *

.5 2 8 * * *

.5 4 8 * * *

.0 5 8

.2 3 0 *

.0 4 6

(. 0 8 9 )

(. 1 5 1 )

(. 1 0 9 )

(. 0 7 3 )

(. 0 2 5 )

(. 0 8 6 )

(. 0 9 0 )

(. 1 3 3 )

(. 1 0 2 )

n 3 2 0

3 2 0

2 7 6

3 2 0

2 7 6

3 2 0

2 7 6

T 1 6

1 6

W a ld χ 2

9 3 .0

8 * * *

9 5 .0

6 * * *

1 2 2 .8

8 * * *

4 7 .9

9 * * *

4 7 .9

2 * * *

6 3 .5

2 * * *

9 3 .0

1 * * *

1 0 2 .0

7 * * *

1 0 0 .2

3 * * *

n o te

s: R

o b

u st

st a n

d a rd

e rr

o rs

in p

a re

n th

e se

s * * * ,

* * ,

* ;

si g n

ifi c a n

c e

a t

1 ,

5 a n

d 1 0

p e rc

e n t

re sp

e c ti

v e ly

T a b

le 4 : H a u s m a n -T

a y l o r E s t im

a t o r f o r L a n d -R

e d is t r ib u t io n

E x p

la n

a to

ry D

e p

e n

d e n t

V a ri

a b

le :

F a m

il y

F a rm

s V

a ri

a b

le s

(a )

(b )

(c )

(a ’)

(b ’)

(c ’)

(a ” )

(b ” )

(c ” )

T im

e v a r ia n t E n d o g e n o u s

D e m

o c ra

c y

.1 3 5 * * *

.1 3 4 * * *

.0 3 7 * *

(. 0 1 7 )

A u

to c ra

c y

-. 0 8 6 * *

-. 0 2 3

(. 0 1 9 )

(. 0 1 7 )

C o n

st ra

in ts

o n

th e

. 1 3 3 * * *

.1 3 2 * * *

.0 5 4 * * *

E x e c u

ti v e

(. 0 1 6 )

(i n

it ia

l) G

D P

.0 2 9 * * *

.0 3 1 * * *

.0 2 8 * * *

p e r

c a p

it a

(. 0 0 3 )

T im

e v a r ia n t E x o g e n o u s

O il

a n

d g a s

-. 0 2 1

-. 0 1 0

-. 0 2 6 *

-. 0 2 5 *

-. 0 1 0

-. 0 1 9

-. 0 1 8

-. 0 0 8

d is

c o v e ri

e s

(. 0 1 4 )

(. 0 1 2 )

(. 0 1 5 )

(. 0 1 3 )

(. 0 1 4 )

(. 0 1 2 )

T im

e in v a r ia n t E x o g e n o u s

(l o g )

T ra

d e

w it

h .0

0 4

.0 0 5

.0 1 1

-. 0 0 8

-. 0 0 9

.0 1 2

.0 0 6

.0 0 7

.0 1 0

B ri

ta in

la te

1 9 th

(. 0 0 7 )

(. 0 0 6 )

(. 0 1 1 )

(. 0 0 7 )

(. 0 0 6 )

(. 0 1 1 )

(. 0 0 7 )

(. 0 0 6 )

(. 0 1 1 )

(l o g )

B ri

ti sh

In v e st

. .0

2 1 * *

.0 2 6 *

.0 2 0 * *

.0 1 2

.0 1 1

.0 2 5

.0 1 9 * *

.0 1 8 * *

.0 2 7 *

e a rl

y 2 0 th

(. 0 1 0 )

(. 0 0 9 )

(. 0 1 5 )

(. 0 1 0 )

(. 0 0 9 )

(. 0 1 6 )

(. 0 0 9 )

(. 0 1 5 )

M in

e ra

l -.

0 3 7

-. 0 5 2 * *

-. 0 7 3

* *

-. 0 4 4 *

-. 0 6 0 * *

-. 0 7 7 *

-. 0 4 3 *

-. 0 5 4 * *

-. 0 7 4 *

c o lo

n ia

l c e n tr

e (.

0 7 6 )

(. 0 2 6 )

(. 0 4 1 )

(. 0 2 5 )

(. 0 4 2 )

(. 0 2 5 )

(. 0 4 0 )

S u

g a r

-. 0 3 7 * *

-. 0 2 7 *

-0 .1

4 -.

0 3 1 * *

-. 0 1 8

-. 0 1 2

-. 0 2 8 * *

-. 0 2 2 * *

-. 0 1 3

su it

a b

il it

y (.

0 1 5 )

(. 0 1 6 )

(. 0 2 4 )

(. 0 1 5 )

(. 0 2 4 )

(. 0 1 4 )

(. 0 1 5 )

(. 0 2 3 )

E a rl

y E

u ro

p e a n

-. 1 3 9

-. 2 1 1

-. 0 8 2

se tt

le m

e n ts

(. 1 7 3 )

(. 1 6 7 )

(. 1 7 0 )

n a ti

v e

p o p

u la

ti o n

-. 0 1 7

.0 1 5

.0 1 3

(. 0 1 1 )

c o n

st a n t

.2 2 4 * * *

.2 5 6 * * *

.1 7 9 * * *

.2 9 0 * * *

.3 3 7 * * *

.1 9 1 * * *

.2 0 6 * * *

.2 5 6 * * *

.1 6 9 * * *

(. 0 3 0 )

(. 0 4 7 )

(. 0 4 8 )

(. 0 3 0 )

(. 0 4 5 )

(. 0 5 0 )

(. 0 2 9 )

(. 0 4 7 )

n 3 2 0

3 2 0

2 7 6

3 2 0

2 7 6

3 2 0

2 7 6

T 1 6

1 6

W a ld χ 2

8 6 .6

4 * * *

9 2 .1

0 * * *

1 8 5 .9

7 * * *

4 2 .5

7 * * *

4 9 .6

0 * * *

1 8 1 .3

1 * * *

8 9 .0

3 * * *

9 3 .9

7 * * *

1 9 9 .0

1 * * *

n o te

s: R

o b

u st

st a n

d a rd

e rr

o rs

in p

a re

n th

e se

s * * * ,

* * ,

* ;

si g n

ifi c a n

c e

a t

1 ,

5 a n

d 1 0

p e rc

e n t

re sp

e c ti

v e ly

The results for redistribution (Equation 2) are shown in Table 4. Institutions

help to explain redistribution: countries with higher levels of democracy and

constraint on the executive and lowers level of autocracy have more family

farms i.e. a more equal distribution of land. The discovery of oil has a

statistically significant effect only if we consider autocracy as institutional

variable (regressions (a’), (b’), and (c’)). A possible explanation is that the

resource shock affects redistribution only under autocratic regimes. Under

more democratic rulers, the effects of a resource on inequality is not signifi-

cant.

The coefficient of Trade with Britain is not statistically significant, but

British investments have a positive and statistically significant effect. When

democracy or constraints on the executive are used as institutional variables,

British investments increase the percentage of family farms (i.e. inequality

decreases) showing a positive indirect effect of British investments on insti-

tutions through inequality. This contrasts with the direct negative effect on

institutions observed in Table 3 and the net effect of British investments on

institutions will depend on which of these two effects prevails.

While colonial factors seem not to matter for institutions, colonial resource

endowments have affected redistribution. Colonial mineral centres and plan-

tation based countries have less family farms (i.e. higher land inequality)

than those territories with no resources. This result supports Engerman

and Sokoloff’s hypothesis that colonial past matters for initial inequality.

These results hold when controlling for colonial European settlements and

native populations (columns (b), (b’) and (b”)) which are not statistically

significant.

Two are the main contributions of this analysis. One is that, given the

early independence of most Latin American countries, institutions are not

just a colonial outcome. Second, The effect of British intervention after

independence goes through different channels. While trade with Britain has

a positive direct effect on institutions, the direct effect of British investments

is negative. This may be explained by the political instability caused by the

series of defaults that followed the entry of these capitals in Latin American

financial markets. These investments have nonetheless an indirect positive

effect through redistribution.

To understand which effect dominates we may need to look at the country’s

individual characteristics. It is likely that in those countries that invested

these capitals in more efficient projects saw an increase of their employ-

ment opportunities which in turn reduces inequality. However, in strongly

autocratic regimes, this indirect effect may be reduced to a minimum. Ac-

cording to Miller (1993), external loans and direct investments in Latin

America contributed to put off taxation reforms and undermined local po-

litical institutions as more politicians became beholden to their links with

the British companies. This affected the adoption of redistribution policies,

and therefore the negative direct effect of British investments on institutions

dominated.

To better understand how the different variables analysed interact and to es-

tablish how post-colonial British intervention may have affected these coun-

tries, the next section offers some historical narratives for four Latin Amer-

ican countries. The experiences of Costa Rica and Uruguay which are con-

sidered consolidated democracies in the region, will be contrasted with the

realities of Peru and Bolivia which history is marked by political instability

with continuous break-downs of democracy.

4 Historical Evidence: Some Case Studies

Costa Rica and Uruguay good development experiences contrast with the

poor economic, social and political scenarios observed in Bolivia and Peru.

The history of these four countries helps to illustrate the specific mechanisms

through which the variables considered in the empirical analysis work and

how their interactions affected these countries’ institutions.

4.1 Costa Rica

Costa Rica is one of the most stable, prosperous, and progressive nations

in Latin America. Nonetheless, it was a poor, isolated, backwater territory

during colonial experience. Costa Rica had no gold or silver and few op-

portunities to promote sugar plantations which made this territory of little

attraction to colonial settlement (Monge Alfaro, 1974; Quirós Vargas, 1990).

According to Quirós Vargas (1990), one of the factors behind Costa Rica’s

colonial poverty was the lack of a significant indigenous population available

for forced labour. For this reason Costa Rican settlers were forced to work

their own land and this prevented the establishment of large latifundios. So

the lack of natural and human resources has been considered to constitute

the basis for a successful rural democracy in Costa Rica (Thorning, 1945).

At the time of independence, this country had in fact the highest level of

land redistribution compared to the rest of the region (based on data of land

distribution from Vanhanen, 2003).

After its independence in 1821, and with the introduction of coffee, there

were clear attempts to stimulate export agriculture. By late 1830s coffee

exports began to reach important levels and the main destiny was Great

Britain. In fact, British merchants played a key role in financing the coffee

expansion9 (Gudmundson, 1986). This specialisation and export-dependence

on coffee was accompanied by high political instability; the 1860s were

marred by power struggles among the coffee elite. As the Costa Rican

economy moved to monoculture and declining returns10, the coffee-based

peasantry and the growing urban middle-class increasingly protested for a

greater wealth distribution and taxation of the coffee-oligarchy. This cul-

minated in the 1948’s Revolution which is seen as the beginning of the new

process of democratisation in Costa Rica11. Social and economic progress

since 1948 helped the return of the country to stability, and though post-

civil war politics reflected the play of old loyalties and antagonisms, elections

have been free and fair since then.

9As matter of fact, the first bank founded in Costa Rica was the Banco Anglo- Costarricense in 1862.

10The declining returns on coffee production were due to the ageing of the groves, soil exhaustion, and the infrequent use of fertilisers (as explained in Hall [1976] cited in Gudmundson (1986, p. 5)).

11See Gudmundson (1984) for a review of the literature on the Costa Rican revolution and civil war in 1948.

4.2 Uruguay

As Costa Rica, the colonial history of Uruguay is also characterised by no

gold, silver, and sugar plantations, making this territory unattractive for

colonisation (especially in early times). In fact, the current Uruguayan ter-

ritories were little inhabited during the colonial times, at least until the es-

tablishment of Colonia del Sacramento by the Portuguese in 1680 (Bértola,

2003). Unlike Costa Rica, Uruguay had a quite unstable transition to inde-

pendence. Uruguay was on the border between the Spanish and Portuguese

empires, and was the subject of several disputes between the two crowns;

this was decisive for the creation, with strong British involvement, of an

independent state in 1828-1830 (Bértola, 2003).

The mid-19th century was characterised by the growth of the Uruguayan

agricultural sector based on the production of meat and livestock production

in general. The main destiny of Uruguayan production was Britain which

attracted British investments in the country (Winn, 1976). Despite episodes

of political unrest and economic stagnation in 1930s for most of the past 180

years, Uruguay has been a model democracy with one of the lowest rates of

income inequality in the region12.

4.3 Bolivia

The history of Bolivia contrasts with the development experiences of Costa

Rica and Uruguay. Bolivia is one of the less developed and more unequal

countries in the region (and in the world). Despite recent improvements in

the Gini index (from 60.1 in 2002 to 56.3 in 2008), the differences in income

still remains: in 2007, the 10% of the population earned 45% of the popu-

lation’s total income, while the poorest earned merely 0.5% (World Bank,

2011). It can hardly be argued that the origins of Bolivian underdevelop-

ment and inequality are found in colonial times. After the discovery of large

silver deposits in Potośı in 1545, Bolivia became a key mining centre and

an important source of revenue for the Spanish Empire and virtually every

12However, Uruguay did not escape the wave of military dictatorships that swept through South America in the 1970s.

aspect of Bolivia’s economic, political but also cultural and social develop-

ment responded to the mining monocultures of silver first and then tin. The

labour force was organised around the exploitation of minerals and based

on forced labour (Cunningham and Jacobsen, 2003).

After independence, the white Creole elite, took control of the State, and

although servitude and slavery were abolished, indigenous people were pre-

vented from participating in the political life through the introduction of

the ‘qualified vote’ i.e. only alphabetised people with a minimum income

could vote at the elections and new forms of forced labour were introduced13.

Universal vote was introduced only after the Bolivian National Revolution

of 1952. However, this was of little help for the redistribution of power

which was in the hands of the wealthy elite (Albro, 2005). One of the main

limitations for political inclusion was the skewed distribution of land that

strongly favours small elite groups. The numerous land reforms introduced

after 1952 implemented only temporary and minor changes and had little

effect on wealth distribution (Medina, 2010). In fact, in the 1980s, over 66%

of land was still controlled by 0.22% of landowners with an average of more

than 16,000 hectares per owner (Weisbrot and Sandoval, 2008).

The last decades have been characterised by political instability and a con-

tinuos economic fluctuations arising from the unstable commodity mar-

ket. A succession of militaristic dictators repressed labour-based organisa-

tions and continued the social discrimination of the indigenous populations

(Madrid, 2012). In current times, Bolivia is still a mining country with the

second largest natural gas reserve in South America14. The economy’s re-

liance on mining has reinforced regional tensions and determined political

power in Bolivia (Morales, 2010). Of all the oil and gas significant producers

in the world, Bolivia is perhaps the only country where sub-national gov-

ernments share these resources revenues according to where they happen

to be underground. This creates further divisions and limits redistribution

(Weisbrot and Sandoval, 2008).

13The most common was ponguaje, through which indigenous population had to provide cheap or unpaid labour in exchange for access to subsistence parcels of land (Bueno, 2011)

14Natural gas and oil and other minerals replaced tin in its role on Bolivian economy, after the collapse of the world tin market in 1980s

4.4 Peru

Peru was also a mining centre during the colonial period. Peru was in fact

described as the “Spain’s great treasure house in South America” (Pike,

1967). Labour was organised following the needs of the mining sector under

different forms of forced labour creating the same social inequalities be-

tween indigenous and colonisers described in the Bolivian case. Colonisers

monopolised control over land and gradually the land tenure system became

polarised between large haciendas and subsistence-based indigenous commu-

nities (Hunefeldt, 2004). After independence, the elite class that inherited

the power from colonisers aimed to preserve and enhance their privileged

economic status15. The new-independent country experienced severe po-

litical instability lasting until the advent of the guano boom in mid-19th

century16.

The Guano Era in Peru represents a period of economic prosperity. Demand

for guano increased with the industrial revolution in the United Kingdom

first and the increase of demand in the rest of Europe and US afterwards.

Although the revenues of guano were used to accomplish some social projects

such as the end of slavery and the Indian tribute (1854), Peru failed to be-

come a modern state. Much of the guano wealth went into the support of

state bureaucracy and some infrastructure projects that were never com-

pleted (Hunefeldt, 2004). The guano revenues were distributed between

British and Peruvian bondholders who held long-standing claims on the

government. According to Quiroz (1987) two-thirds of the total bond value

was held by only 126 people, mostly land-owners and state bureaucrats. In

addition, guano financial windfalls made it easy to get loans on the inter-

national financial markets which eventually led to a deep financial crisis17.

15At the time of independence and for several decades after, Peru had a racially defined occupational structure. Artisans were black, peasants were Indians, smaller merchants were mestizos, and elites were white (Hunefeldt, 2004).

16Guano is created by seabird droppings deposited for thousands of years and sedi- mented n coastal islands. The benefits of guano as fertilising were known by pre-Columbian societies, by it was Alexander von Humboldt who alerted Europeans to the value of guano.

17The Peruvian debt crisis had his origins in the independence wars. In 1822 and 1824 two loans were contracted in London, and by 1848 the principal and interest had increased Peru’s debts of three times the initial loan. In addition, the government acknowledged

The discovery of synthetic fertilisers and the collapse of the guano price in

the international market were devastating to the Peruvian economy. The

discovery of nitrate mines could have replaced the role of guano in Peru-

vian economy, but the conflict between Chile and Peru for the control of

the mines ended in the War of the Pacific where Peru lost its nitrate-rich

provinces (Greenhill and Miller, 1973).

After guano and nitrate, Peru experienced several booms in its primary sec-

tor. Rubber, coffee, sugar cane, cotton, rice and other natural resources

were crucial for Peruvian development. Nonetheless, these resources were

in the hands of a Peru’s oligarchy (estimated as 40 to 200 families) that

retained much influence until late 1960s. In 1980s some attempts to address

the problems of rural communities were made, and although the levels of in-

equality still remain very high, the political participation of the marginalised

communities has increased, fostering a greater redistribution and the new

wave of democracy in the country.

The experiences of these countries contribute to the previous explanation

on how the different variables interact. We can see how the effect of British

indirect rule in the region depends on the levels of inequality and quality

of institutions inherited from the colonial times which in turn depend on

the colonial resource endowment. Due to the lack of gold, silver and sugar

plantations in Costa Rica and Uruguay, these were not considered relevant

for the colonial activities of the European powers, so these countries did

not inherited a strong colonial economic elite. When British investments

arrived into these markets, these went to finance agricultural production

which fosters the development of the peasantry and middle class in these

countries. These investments did have a negative impact on these coun-

tries institutions (shown by a series of revolts in 1948 in Costa Rica and

in 1930s in Uruguay), however, the indirect effect (through the increase in

redistribution) dominate.

internal debt to those citizens who had supplied funds for the patriot armies during the independence wars in order to foster the development of an entrepreneurial middle class. During the guano boom, British bondholders pressured the Peruvian government for re- payment and in 1849 they obtained new bonds to be issued backed by future sales of guano. However, new loans were used to repay old loans and accumulated interest.

In Peru and Bolivia, the negative effect on institutions was dominant. Both

these countries inherited a strong colonial elites that controlled the wealth

of this region and occupied main charges in the governments. Labour was

mainly organised under force labour creating huge social inequalities. The

revenues from mining (in Bolivia) and guano (in Peru) were managed by

these elites which also took control over the British capital. The limited

participation of the working class to the political activities makes the indirect

effect of British investments being only marginal.

5 Concluding Remarks

The discovery and colonisation of new territories has been one of the most

salient events in the World history. This has brought several scholars to

argue that colonial past exerted a strong influence on the economic and

political institutions created in former colonies. Two are the hypotheses

of these studies, (i) various colonial aspects affect the creation of early in-

stitutions, (ii) given that institutions tend to be highly persistent, current

institutions still reflect early ones. This paper shows that Latin American

institutions cannot be explained just by their colonial past due to the re-

gion’s early decolonisation process that leaves scope for other factors and

post-colonial events to change the character of these institutions so that cur-

rent institutions do not just reflect early ones. Two additional distinctive

characteristics of the region are considered: inequality and natural resource

dependence.

Latin America is the most unequal region in the world. High levels of in-

equality and poor institutions are correlated and the relationship tends to

be highly persistent due to the feedback mechanisms put in place. As ex-

plained by Boix (2003), in societies with high level of inequality and land-

concentration, the cost of taxation and redistribution becomes high enough

for the elites to prefer an authoritarian regime which in turn will put in

place policies that allow these elites to keep their economic and political

benefits reinforcing the current levels of inequality and asset-concentration.

As matter of fact, we can observe a long-standing political discrimination

of indigenous populations in Peru and Bolivia after independence. The au-

thoritarian regimes in these countries delayed the required policies to abolish

forced labour and allowed the political participation of the poorer parts of

the society. However, whereas Boix (2003) explains how inequality and

institutions interact, he fails to explain the impact of the colonial (and non-

colonial) past on current institutions. The understanding of the origins of

institutions and inequality can be used to explain the observed persistence

of the relationship between institutions and inequality.

The analysis of the colonial factors shows that colonial resource endowment

has a negative effect on the redistribution of land in these countries having

thus a negative (indirect) effect on institutions. Latin America territories

with larger European settlements and smaller pre-colonial populations do

not seem to have better institutions as discussed by Acemoglu et al. (2001);

Easterly and Levine (2012) and Acemoglu et al. (2002); Baker et al. (2008);

Mahoney (2010) respectively.

This paper also goes more in depth analysing how coloniser’s identity affects

on institutions. Authors tend to conclude that British did better, but they

fail to elucidate the mechanisms through which British left behind better in-

stitutions in colonised territories. This paper explains the channels through

which British intervention affected institutions. While most Latin American

countries were Iberian colonies, Britain showed a great interest on these ter-

ritories after they acquired independence. Latin American countries become

an important provider of natural resources for the booming British economy

and a potential market for British exports. Moreover with the development

of British financial sector, Latin American governments tapped the London

market.

The results show that while trade with Britain played a positive role in im-

proving institutions, the channels through which investment affected institu-

tions are more complex. There is a negative direct effect on institutions that

can be explained by the political instability created once the governments

in these countries could not make front to British investors and decided to

default creating economic and financial instability. However, in those coun-

tries where these investments were efficiently utilised to finance productive

sectors, we can observe a positive indirect effect that goes through the im-

provement in redistribution. Countries with a better initial redistribution of

resources may have used these investments in productive enterprises that,

in the long-run, benefited the economic development and the institutional

setting of these economies (see for instance the improvement of transport

infrastructure in Costa Rica during the coffee era).

In addition, the dependence of some of these economies on natural resources

even after the independence promoted further authoritarianism in order to

gain control over these resources. Discoveries of natural resources have in

fact a negative effect on institutions. Dunning’s claim (2008) that natu-

ral resources may have both authoritarian and democratic effects based on

the redistribution of resources is thus not entirely consistent with the Latin

American experience. There is no evidence that in societies with substantial

inequality of assets (not related to the natural resource sector) a resource

boom helps to mitigate the negative impact of inequality on institutions

through an increase in redistribution policies. On the contrary, there is a

tendency for these resources to promote authoritarian regimes – the pres-

ence of resource rents increases the payoff of controlling power in order to

control the distribution of rents. Even in those countries with a more egal-

itarian wealth redistribution, these resources generated some conflict over

power (see for example the Costa Rican political instability marred by power

struggles among the coffee elite in 1860s).

While the colonial period contributed to the struggles over power control and

redistribution of income and wealth in Latin America, post-independence

discoveries of natural resources provided the elites with further incentives

to maintain authoritarian regimes in order to control the rents generated

by these resources. Future research on what facilitates the redistribution

of power within a country is needed. After independence, the several rev-

olutions and democratisation processes that took place in Latin America

allowed these countries to reach important milestones: in the 1980s-1990s

democracy and universal suffrage became a reality in the region, however

sound political practices have not always kept pace. This may be due to the

persistence of unequal distribution of power in some countries that, even

during periods of transition (that may translate into a temporary loss of

power), allows elites to find alternative ways to influence the distribution

of resources in order to maintain their privileges. How to contrast the elite

behaviour would help us to provide with clearer policy advices to these

countries.

References

Acemoglu, D., Johnson, S., and Robinson, J. A. (2001). The Colonial Ori- gins of Comparative Development: An Empirical Investigation. American Economic Review, 91:1369–1401.

Acemoglu, D., Johnson, S., and Robinson, J. A. (2002). Reversal of Fortune: Geography and Institutions in the Making of the Modern World Income Distribution. Quarterly Journal of Economics, 117:1231–1294.

Albro, R. (2005). The Indigenous in the Plural in Bolivia Oppositional Politics. Bulletin of Latin American Research, 24:433–54.

Baker, M. J., Brunnschweiler, C. N., and Bulte, E. H. (2008). Did History Breed Inequality? Colonial Factor Endowments and Modern Income Dis- tribution. Economics Working Paper Series 86, Swiss Federal Institute of Technology Zurich.

Banks, A. S. and Wilson, K. A. (2013). Cross-National Time-Series Data Archive. Databanks International. www.databanksinternational.com.

Beblawi, H. (1987). The Rentier State in the Arab World. In Beblawi, H. and Luciani, G., editors, The Rentier State, pages 49–62.

Bertocchi, G. and Canova, F. (2002). Did Colonisation Matter for Growth? An Empirical Exploration into the Historical Causes of Africa’s Underde- velopment. European Economic Review, 46:1851–71.

Bértola, L. (2003). An Overview of the Economic History of Uruguay since the 1870s.

Boix, C. (2003). Democracy and Redistribution. Cambridge University Press, New York.

Botero, J. C., Djankov, S., La Porta, R., Lopez de Silanes, F., and Shleifer, A. (2004). The Regulation of Labour. Quarterly Journal of Economics, 119:1339–1382.

Brunnschweiler, C. N. and Bulte, E. H. (2008). The Resource Curse Re- visited and Revised: A Tale of Paradoxes and Red Herrings. Journal of Environmental Economics and Management, 55:248–64.

Bueno, R. (2011). The Politics of Evo Morales Rise to Power in Bolivia. The Role of Social Movements and Think Tanks. Technical report, Overseas Development Institutite.

Cunningham, W. and Jacobsen, J. (2003). Group-Based Inequalities: The Roles of Race, Ethnicity, and Gender. In Inequality in Latin America and the Caribbean: Breaking with History? The International Bank for Reconstruction and Development / The World Bank.

della Paolera, G. and Taylor, A. M. (2012). Sovereign Debt in Latin Amer- ica, 1820-1913. Department of Economics Working Papers 18, Central European University.

Djankov, S., La-Porta, R., Lopez-de Silanes, F., and Shleifer, A. (2003). Courts. Quarterly Journal of Economics, 118(2):453–517.

Domar, E. D. (1970). The Causes of Slavery or Serfdom: A Hypothesis. The Journal of Economic History, 30(1):18–32.

Dunning, T. (2008). Crude Democracy: Natural Resource Wealth and Po- litical Regimes. Cambridge University Press, New York.

Easterly, W. (2001). Can Institutions Resolve Ethnic Conflict? Economic Development and Cultural Change, 49(4):687–706.

Easterly, W. and Levine, R. (2012). The European Origins of Economic Development. NBER Working Paper 18162, National Bureau of Economic Research.

Engerman, S. and Sokoloff, K. (1997). Factor Endowments, Institutions and Differential Paths of Growth among the New World Economies. In Haber, S., editor, How Latin America Fell Behind. Stanford University Press.

Engerman, S. and Sokoloff, K. (2002). Factor Endowments, Inequality, and Paths of Development Among New World Economies. Working Paper 9529, National Bureau of Economic Research.

FAO (2010). GAEZ Global Agro-Ecological Zones. Available at gaez.fao.org.

Fearon, J. D. and Laitin, D. D. (2003). Ethnicity, Insurgency, and Civil War. American Political Science Review, 97(01):75–90.

Friedman, S. (2002). Democracy, Inequality and the Reconstitution of Poli- tics. In Tulchin with Amelia Brown, J. S., editor, Democratic Governance and Social Inequality. Lynne Rienner Publishers.

Greenhill, R. G. and Miller, R. M. (1973). The Peruvian Government and the Nitrate Trade, 1873–1879. Journal of Latin American Studies, 5(1):107– 31.

Grier, R. M. (1999). Colonial Legacies and Economic Growth. Public Choice, 98:317–35.

Gudmundson, L. (1984). Costa Rica and the 1948 Revolution: Rethink- ing the Social Democratic Paradigm. Latin American Research Review, 19(1):235–42.

Gudmundson, L. (1986). Costa Rica Before Coffee: Society and Economy on the Eve of the Export Boom. Louisiana State University Press.

Gurr, T. R. (1999). Polity II: Political Structures and Regime Change, 1800- 1986. Technical Report 9263, Inter-university Consortium for Political and Social Research.

Hamilton, K. and Clemens, M. A. (1999). Genuine Savings Rates in Devel- oping Countries. World Bank Economic Review, 13(2):333–56.

Hoff, K. and Stiglitz, J. E. (2004). After the Big Bang? Obstacles to the Emergence of the Rule of Law in Post-Communist Societies. American Economic Review, 94(3):753–63.

Houle, C. (2009). Inequality and Democracy: Why Inequality Harms Consol- idation but Does Not Affect Democratization. World Politics, 61(4):589– 623.

Hunefeldt, C. (2004). A Brief History of Peru. Facts on File, Inc, New York.

Jensen, N. and Wantchekon, L. (2004). Resource Wealth and Political Regimes in Africa. Comparative Political Studies, 37:816–41.

Kawagoe, T., Hayami, Y., and Ruttan, V. (1985). The Inter-Country Agri- cultural Production Function and Productive Differences among Coun- tries. Journal of Development Economics, 19(1-2):113–32.

Keefer, P. and Knack, S. (2002). Social polarization, political institutions, and country creditworthiness. Policy Research Working Paper Series 2920, The World Bank.

La Porta, R., Lopez de Silanes, F., Shleifer, A., and Vishny, R. W. (1998). Law and Finance. Journal of Political Economy, 106:1113–1155.

La Porta, R., Lopez de Silanes, F., Shleifer, A., and Vishny, R. W. (1999). The quality of government. The Journal of Law, Economics and Organi- sation, 15(1):223–79.

Lipset, S. M. (1959). Some Social Requisites of Democracy: Economic Devel- opment and Political Legitimacy. The American Political Science Review, 53(1):69–105.

Lujala, P., Rod, J. K., and Thieme, N. (2007). Fighting Over Oil: Introduc- ing a New Dataset. Conflict Management and Pease Science, 24(3):239– 56.

Madrid, R. L. (2012). The Rise of Ethnic Politics in Latin America. Cam- bridge University Press.

Mahdavy, H. (1970). The Patterns and Problems of Economic Development in Rentier States: the Case of Iran. In Cook, M. A., editor, Studies in Economic History of the Middle East, pages 428–67.

Mahoney, J. (2010). Colonialism and Postcolonial Development. Spanish America in Comparative Perspective. Cambridge University Press, New York.

Marshall, M. G. and Gurr, T. R. (2013). Polity IV: Project: Politi- cal Regime Characteristics and Transitions, 1800-2012. Available at http://www.systemicpeace.org/polity/polity4.htm.

McEvedy, C. and Jones, R. (1977). Atlas of World Population History. Penguin Books.

Medina, J. (2010). La comprensión ind́ıgena de la Buena Vida. Editorial Garza Azul, La Paz, Bolivia.

Miller, R. (1993). Britain and Latin America in the Nineteenth and Twen- tieth Centuries. Longman, first edition.

Monge Alfaro, C. (1974). Historia de Costa Rica. Libreria Trejos, Costa Rica.

Morales, W. (2010). A Brief History of Bolivia. Facts on File, Inc, New York.

Muller, E. N. (1988). Democracy, Economic Development, and Income In- equality. American Sociological Review, 53(1):50–68.

Paish, G. (1909). Great Britain’s Capital Investments in Other Lands. Jour- nal of the Royal Statistical Society, 72(3):465–95.

Pike, F. (1967). The Modern History of Peru. Frederick A. Praeger Pub- lishers, New York.

Platt, D. C. S. M. (1972). Latin America and British Trade 1806-1914. Adam & Charle Black.

Quirós Vargas, C. (1990). Historia de Costa Rica: La era de la encomienda. Editorial Universidad de Costa Rica, Costa Rica.

Quiroz, A. (1987). La deuda defraudada: Consolidación y dominio económico en el Perú. Instituto Nacional de Cultura, Lima.

Rippy, F. J. (1959). British Investments in Latin America, 1822-1849: A Case Study in the Operations of Private Enterprise in Retarded Regions. University of Minnesota Press, Minneapolis.

Ross, M. (2001). Extractive Sectors and the Poor. Report, Oxfam America.

Ross, M. (2006). A Closer Look at Oil, Diamonds, and Civil War. Annual Review of Political Science, 9:265–300.

Ross, M. L. (2003). Oil, Drugs, and Diamonds: the Varying Roles of Nat- ural Resources in Civil War. In Ballentine, K. and Sherman, J., editors, The Political Economy of Armed Conflict: beyond Greed and Grievance, chapter 9. Lynne Rienner Publishers.

Rubinson, R. and Quinlan, D. (1977). Democracy and Social Inequality: A Reanalysis. American Sociological Review, 42(4):611–23.

Sachs, J. D. and Warner, A. M. (1995). Natural Resource Abundance and Economic Growth. Working Paper 5398, National Bureau of Economic Research, Cambridge, MA.

Sinnot, E., Nash, J., and de la Torre, A. (2010). Natural Resources in Latin America and the Caribbean. Technical report, World Bank Latin American and Caribbean Studies.

Statistical Office (1906). Annual Statement of Trade of the United Kingdom with Foreign Countries and British Possessions - various years. Great Britain. H. M. Customs and Excise.

Stotsky, J. G. and WoldeMariam, A. (1997). Tax Effort in Sub-Saharan Africa. Working Paper 107, International Monetary Fund.

Thirsk, W. (1997). Tax Reform in Developing Countries. The World Bank, Washington, DC.

Thorning, J. F. (1945). Costa Rica: A Rural Democracy. World Affairs Institute, 108(3):171–80.

Vanhanen, T. (2003). Democratization and Power Resources 1850-2000. computer file FSD1216 version 1.0. Tampere: Finnish Social Science Data Archive.

Victor Bulmer-Thomas (1998). British Trade with Latin America in the Nineteenth and Twentieth Centuries. Occassional Papers 19, Institute of Latin America Studies, London.

Weisbrot, M. and Sandoval, L. (2008). The Distribution of Bolivia’s Most Important Natural Resources and the Autonomy Conflicts. CEPR Re- ports and Issue Briefs 22, Center for Economic and Policy Research.

Winn, P. (1976). British Informal Empire in Uruguay in the Nineteenth Century. Past & Present, 73:100–26.

World Bank (2011). World Development Indicators. Available at http://data.worldbank.org/data-catalog/world-development-indicators.

A Appendix

A.1 Data Sources and Variable Definitions

This appendix offers definitions and sources for the data used for the em- pirical analysis. Table 5 provides the descriptive statistics of the variables used.

The variables for institutions, democracy, autocracy and constraints on the executive power, are from Marshall and Gurr (2013). This data set consists of six component measures that record key qualities of executive recruitment, constraints on executive authority, and political competition. It covers all major, independent states, currently 167 countries over the period 1800- 2012.

As measure of inequality, I use the percentage of family farms from the “Vanhanen Index of Power Resources” (Vanhanen, 2003). This covers the period 1850-2000. Family farms are distinguished from large farms culti- vated mainly by hired workers. However, family farms are not dependent on the actual size of the farm which varies with the type of product and the agricultural technology being used.

The percentage of family farms capture the degree of concentration and therefore inequality in the ownership of land. The variable for the discovery of oil and natural gas is based on the dataset PETRODATA (Lujala et al., 2007). This dataset includes 890 onshore and 383 offshore locations with geographic coordinates and information on the first oil or gas discovery and production year. Based on this dataset, I created the variable for the discovery of natural resources.

The British influence in Latin America is considered using measures of British investments in Latin America in the period 1905-1911 and trade with Britain in the period 1898-1906. Investments are from Paish (1909). These are expressed in British Sterling. These investments were mainly based on government loans. Considering the differences in population and dimensions among countries, I have divided this variable by the average gov- ernment revenue in the same period. The data for the government revenue is from the “Cross-NationalTime-Series DataArchive” (CNTS) (Banks and Wilson, 2013). This data is expressed in US dollars, therefore, British in- vestments are converted in US dollars using US$4.85=GBP£1 as exchange rate (during late 19th and early 20th centuries, many countries adopted the

gold standard, as consequence, conversion rates between different currencies was fixed and determined by the respective gold standard).

Trade is from the “Annual Statement of the Trade of the United Kingdom with Foreign Countries and British Possessions”. This is given by a country’s exports to Britain in a given year divided by the total exports of that country. Total exports are also from CNTS (Banks and Wilson, 2013). Trade is also expressed in US dollars following the same procedure than before. The CNTS contains data for over 200 states from 1815 onwards (excluding the periods 1974-1918 and 1940-1945) for a number of social indicators.

Table 6 summarises variables definitions and sources.

Table 5: Descriptive Statistics

Variable Mean Std. Dev. Min Max Observations

Democracy overall 0.332246 0.328692 0 1 N = 400 between 0.197039 0.115 1 n = 20 within 0.266577 -0.20275 1.017246 T = 20

Autocracy overall 0.334746 0.296249 0 1 N = 400 between 0.13727 0 0.534 n = 20 within 0.264231 -0.19925 0.930663 T = 20

Constraints overall 0.408285 0.324914 0 1 N = 400 on the Executive between 0.184884 0.141667 1 n = 20

within 0.270213 -0.15672 1.149951 T = 20

Family overall 0.17665 0.113929 0.01 0.62 N = 400 Farms between 0.071893 0.062 0.34 n = 20

within 0.089762 -0.07435 0.46965 T = 20

Oil and gas overall 0.1375 0.344806 0 1 N = 400 discoveries between 0.169267 0 0.45 n = 20

within 0.302662 -0.3125 1.0375 T = 20

Trade with overall 0.1573601 .1547945 .0000836 .645317 N = 400 Britain between .1586171 .0000836 .645317 n = 20

within 0 .1573601 .1573601 T = 20

British overall 0.356444 0.24639 0.009175 0.871568 N = 320 Investments in between 0.254073 0.009175 0.871568 n = 16 Latin America within 0 0.356444 0.356444 T = 20

European settlements overall 0.157923 0.102771 0 0.4612 N = 49 in 1800

Native population overall 1.386844 0.995758 0 5.64 N = 48 in 1500

Colonial mineral overall 0.183674 0.39123 0 1 N = 49 centre Sugar suitability overall 1.134926 1.214803 0 5.315139 N = 49

Table 6: Variable definitions and sources

Variable Definition and Source

Democracy An eleven category scale, from 0 to 10, with a higher score indicating more democracy. Points are awarded on three dimensions: competitiveness on political participation, competitiveness of executive recruitment, and con- straints on chief executive. This has been re-scaled to 0-1. Variable de- scribed in Gurr (1999). Source: Marshall and Gurr (2013)

Autocracy An eleven category scale, from 0 to 10, with a higher score indicating more autocracy. This has been re-scaled to 0-1. Variable described in Gurr (1999). Source: Marshall and Gurr (2013)

Constraint on Execu- tive

A seven category scale, from 1 to 7, with a higher score indicating more constraints. This has been re-scaled to 0-1. Variable described in Gurr (1999). Source: Marshall and Gurr (2013)

Family Farms The area of family farms as a percentage of the total area of holdings. A family farm employs no more than four people including family members and the family owns and cultivates the land. The data set is reported in averages for each decade. For this study, we use five-years average, therefore the data has been considered twice (e.g. for the periods 1990- 1995 and 1995-2000, I use the data reported for 1990s. Source: (Vanhanen, 2003)

Oil and Natural gas discoveries

This variable assumes the value of 1 if there was a discovery of oil or natural gas in that period of time, otherwise it is equal to 0. This variable has been created based in PETRODATA, and from several other sources (for the missing years). Source: (Lujala et al., 2007), and author’s elaboration

British Investments in Latin America

Average of British investments in Latin American countries in the period 1905-1911, divided by the country’s average government revenue. The value is expressed in US$. Source: Paish (1909) and Banks and Wilson (2013)

Latin American trade with Britain

Average of British imports from Latin American economies in the period 1898-1906, divided by total country’s exports. The value is expressed in US$. Source Statistical Office (1906) and Banks and Wilson (2013)

European settlements in 1800

Percentage of population that was European or of European descent in 1800. Source: Acemoglu et al. (2001) and McEvedy and Jones (1977)

Pre-colonial popula- tion

Total population in 1500. Source: Acemoglu et al. (2002) and McEvedy and Jones (1977)

Colonial Minerals Dummy variable. It is equal to 1 if the main economic activity during the colonial period was based on the exploitation of gold or silver. 0 otherwise. Source: Author’s elaboration

Sugar suitability Percent of national land area suitable for the production of sugar, tak- ing into account such factors as soil, rainfall, temperature, and elevation. Source: FAO (2010)

Introduction
On the Origins of Institutions

Post-colonial Latin America: British Intervention
Other Factors that May Affect Institutions: Natural Resources and Inequality

Empirical Analysis

Data
Results

Historical Evidence: Some Case Studies

Costa Rica
Uruguay
Bolivia
Peru

Concluding Remarks
Appendix

Data Sources and Variable Definitions