Contents

1 Introduction 1
2 Dutch Disease and intersectoral effects of resource booms 4
　2.1 Resource booms retard growth 4
　2.2 When resource booms do not cause a problem 6
3 Political economy considerations 8
　3.1 The importance of institutions 8
　3.2 Institutional malfunctioning 9
　3.3 Natural resources and conflict 10
　3.4 Institutional determination 11
4 A dynamic model of growth collapse combined with rent-seeking 14
5 Empirical investigation 19
　5.1 Data 19
　5.2 Descriptive statistics 19
　5.3 Regression results 21
　　5.3.1 The regression model 21
　　5.3.2 The random effects model 22
　　5.3.3 The feasible generalised least squares (FGLS) estimates 23
　　5.3.4 Cross-sectional estimate 24
6 Conclusions 26
7 References 28

Appendix 1 32
Appendix 2 32

Figure 1 Growth collapse 34
Figure 2 The probability of remaining at a given level of democracy 46
Table 1 Countries with growth failure 35
Table 2 Export classification, GDP per capita and export growth for selected developing
countries 36
Table 3 Governance indicators for selected developing countries 1997-98 and 2000-01 39
Table 4 Distribution of major exports 46
Table 5 Distribution of democracy by region 47
Table 6 Joint distribution of democracy and major exports 48
Table 7 Major exports by region 49
Table 8 Summary of mean democracy by region and major exports 49
Table 9 Summary GDP growth by region and major exports 49
Table 10a Countries with a mean democracy index =<5 for 1950-2000 50
Table 10b Countries with a mean democracy index >=5 for 1950-2000 51
Table 11 Empirical results: dependent variable democracy 52
Table 12 Comparisons of fixed and random effect coefficients 52
Table 13 Empirical results: dependent variable natural logarithm of per capita GDP 53
Table 14 Cross-sectional estimates of democracy and GDP growth 54

ABSTRACT

This paper looks at the relationship between natural resource endowment, particularly the type associated with minerals and plantations, and economic development. It may not be natural resource endowment per se but its type that matters, when it comes to growth and development. Certain types of natural resources such as oil and minerals have a tendency to lead to production and revenue patterns that are concentrated, while revenue flows from other types of resources such as agriculture are more diffused throughout the economy. The former category is dubbed a point-source economies, while the latter type is referred to as diffuse. Most countries in the first group have been prone to growth failure in recent times, with notable exceptions such as Botswana and Indonesia. The paper reviews two sets of models, the first set outlining the onset of the resource curse, and the second category sketching conditions where resource booms actually aid growth, or policies to avoid the resource curse. The vast majority of these models stress intersectoral linkages. Political economy considerations associated with resource rents are also reviewed. The focus is on institutions that determine the policy superstructure. The importance of institutions is highlighted, followed by a sketch of institutional malfunctioning and an overview of the empirical models of institutional determination. An explicit model of growth collapse with micro-foundations in rent-seeking behaviour and contests is also presented. The empirical analysis put forward is one of the few econometric
investigations into the resource curse that includes analysis over time, as it is a panel data estimation. Due to this time series property our proxy for institutional quality is the degree of democracy. Our findings suggest that a point-source type natural resource endowment does retard democratic and institutional development, which in turn hampers economic growth. Institutions and institutional functioning are the crucial link between resource endowments, geography and policies, on the one hand and economic outcomes on the other hand.