Resource Wealth, Democracy and Mobilisation

Abstract

The impact of resource wealth on society remains highly contested in the literature. Some perspectives suggest that resource wealth is associated with political apathy, while others indicate that resource wealth fosters violent civil conflicts. We leverage these seemingly contradictory impacts of natural resources on society by expanding the scope of inquiry to explore different types of resource wealth (oil and minerals) as well as other dimensions of political life (protest). Utilising a global dataset for the period 1950–2006, we test the impact of resource wealth on mobilisation. We find that while oil wealth demobilises citizens in all regime types, mineral wealth strongly correlates with higher levels of mobilisation in democracies, though not in autocracies. In addition, using survey data, we examine individual-level attitudes toward protest participation in two resource-rich states, Peru and Ecuador. The results indicate that an individual living in a mineral-rich country like Peru is more likely to participate in a protest compared to an individual living in an oil-rich country like Ecuador. Our findings highlight the contributions of the resource activism framework for understanding the connection between natural resources and mobilisation.

Acknowledgments

We would like to thank Jessica Steinberg, Todd Eisenstadt, and reviewers for their constructive feedback and helpful comments. The authors will make the data and code available for download through the personal website of Chris Patane (https://christopherpatane.com/).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. In this paper, the terms mobilisation and protest are used interchangeably. We use these terms to refer to ‘the act of challenging, resisting, or making demands upon authorities, powerholders, and/or cultural beliefs and practices by some individual or group’ (Goodwin & Jasper, 2003, p. 3).

2. We use oil wealth to refer to the value of a country’s oil production on a per capita basis. We use mineral wealth to refer to the value of a country’s mineral production on a per capita basis.

3. The causal mechanisms linking natural resources and civil conflict, however, remain poorly understood. See Basedau, Mähler, and Shabafrouz (2014).

4. The distinction between rights- and service-based demands follows Arce (2014a). This classification also presupposes the abundance of natural resources and the presence of windfalls from commodity booms in the context of ‘new extractivism’ (Bebbington, 2009).

5. Rights- and service-based mobilisations are not mutually exclusive categories, rather they aim to provide a citizen-centred explanation on how extraction affects local communities. In several mining conflicts, some groups continue to oppose extraction due the adverse impact of mining on the environment (rights-oriented mobilisations), but other people who are benefiting from mining operations are more likely to raise claims about financial improvement (service-oriented mobilisations).

6. Ecuador has also comparatively fewer protests than Peru. OCMAL (2015) records only seven mining-related conflicts since 1995.

7. Generally, newspapers tend to over-report protest events when protests are high and, conversely, to underreport protest events when protests are low. Similarly, the geographic proximity of the newspaper to the protest event tends to inflate the reporting of some protests in comparison to others. The reporting of protest events can also be slated depending on the ideological orientation or ‘political culture’ of the newspapers (see Barranco & Wisler, 1999).

8. This measure includes antimony, bauxite, chromium, copper, gold, iron ore, lead, manganese, mercury molybdenum, nickel, silver, tin, tungsten, and zinc. The variable Mineral Income excludes oil.

9. These data end in 2006, and it is a limiting factor in our time-series analysis despite the longer time-frame of Banks’s data. Despite recent changes in the political landscape of Latin America since 2006, including the stability of populist, leftist leaders such as Evo Morales of Bolivia, there is still a dominance of neoliberalism in the region (see Mazzuca, 2013).

10. Due to issues of stationarity in the data, we log this variable in our models. As is common practice when logging values with distributions of this type, we add a 1 to the measure of mineral wealth prior to logging it.

11. The vertical axis indicates the expected number (count) of protests in the same country year for democracies at different levels of mineral wealth, holding all other control variables at their means. Despite the continuous shape of the curve, the vertical axis represents a whole number given the count nature of our dependent variable. In other words, it is a comparison between countries in the same year rather than a comparison within a single country over time. The lower graph in Figure 1 depicts a histogram of our mineral wealth variable.

12. The Latinobarometer is a nationally representative, random, and stratified survey reaching both urban and rural inhabitants in 18 Latin American countries. The Latinobarometer (2013) is available at www.latinobarometero.org.

13. The geographic area (or ‘city’) variable refers to the smallest political or administrative division of a state as recorded by the Latinobarometer (2013). In Peru and Ecuador, the political or administrative units are also known as provinces. While the Latinobarometer surveys do not reach the populations who are directly impacted by extraction, it would be safe to assume that citizens residing in nearby cities or provinces are familiar with the problems created by resource extraction.

14. Table A1 in the Appendix displays the ranges of 95 per cent confidence intervals for these probabilities.

15. By looking to natural resources per capita, we employ a measure that is comparable across developed and developing countries. Other measures, such as those that look to exports as a proportion of the state’s gross domestic product, would bias the importance of extraction towards developing states (Ross, 2009). However, opposition in extractive regions occurs throughout both the developed and developing worlds (Özkaynak et al., 2015).