Resource Extraction and Inclusive Development
Resource Extraction and Inclusive Development
Extending the Bases of the Political Settlements Approach
Abstract and Keywords
This chapter develops a conceptual framework for understanding the politics of extractive industry governance. Building from the work of Karl, Ross, Watts, and others, and their efforts to understand the political drivers and consequences of the resource curse, the chapter proposes an approach that also engages with political settlements theory, addressing the political implications of the materiality of natural resources and the politics of ideas surrounding resource governance. The chapter then introduces a programme of cross-country, comparative research designed to address the relationships among political settlements, extractive industry, and patterns of development; describes the questions that guided this research; and presents the methods used.
The large-scale extraction of minerals, hydrocarbons, and other natural resources1 has long been central to national economies and the global economy. This has been the case in colonial and post-colonial economies alike, and the harnessing of such resources continues to feature prominently in the economic development strategies of many countries. The economic significance of resource extraction can be quantified through data on percentage contribution of oil, gas, and mining to GDP, exports, foreign direct investment, and the public budget (Mosley, 2017). In contrast, the contributions of extractive industries to ‘development’ and ‘inclusive development’ are more uncertain, not least because definitions of these terms differ based on the degree to which analysts emphasize goals of poverty and income inequality, environmental justice, gender equity, or human and citizenship rights. Even those advocating for the role of extractives in economic development and poverty reduction recognize that results have often been ambiguous (Davis, 2009). These observations have inspired much discussion of the factors explaining such disappointments, a series of policy recommendations regarding how to improve the contributions of resource extraction to development, and significant advocacy and social conflict surrounding the extractive economy across the globe (see Ross, 2015).
Much of the discussion regarding how to manage resource extraction more effectively has emphasized the importance of institutions and governance (Karl, 1997, 2007; Humphreys et al., 2007). Disappointing development outcomes in economies with substantial extractive activity have been explained in terms of the ‘poor quality’ or ‘weakness’ of institutions existing prior to and (p.2) during the growth of investment in mining and hydrocarbons, as well as the adverse impacts of a growing extractive economy on institutional quality (Weber-Fahr, 2002; Robinson et al., 2006; Humphreys et al., 2007; Davis, 2009; Ross, 2015). Such centring of institutions in explanations of poor performance has also meant that many proposals for change have emphasized the need for new, improved, and strengthened institutions. The sorts of institutions proposed are diverse, often depending on the most pressing concerns of the commentator. These can include: institutions and processes for free, prior and informed consent (FPIC); sovereign wealth funds to manage resource rents without distorting incentives; transparency institutions ranging from subnational government oversight to the global Extractive Industries Transparency Initiative (EITI); and tax codes to reallocate benefits along the value chain and among stakeholders.2 While such proposals each have their logic, they tend to say less about the conditions that would need to exist for such institutional changes to come into being, be sustained over time, and be effective.
Of course, there is only so much that any particular act of analysis or advocacy can achieve, but it remains important to explore the conditions under which institutions affecting natural resource governance have emerged. The literature on institutional change suggests that political, ideational, and bureaucratic factors are all likely to matter (Mahoney and Thelen, 2010). While much of this literature invokes exogenous drivers of change, drivers can also be endogenous to the institution and the geographical and social units that it governs (Bebbington et al., 2008; Mahoney and Thelen, 2010; Berdegué et al., 2015; Ospina et al., 2015). Regardless of whether drivers are primarily exogenous or endogenous, historical precursors are likely to affect the possibility of, and form taken by, such processes of emergence (Mahoney and Thelen, 2010; Thorp et al., 2012; Bebbington, 2013b).
This book synthesizes lessons from research enquiring into the drivers of institutional change in extractive industry governance over time and in different geographical contexts, and the implications of these governance arrangements for patterns of social, political, and economic inclusion. The project paid special attention to the political drivers of institutional change. We are certainly not the first to step into this terrain, and there is already a strong scholarly literature addressing the political dimensions of extractive industry governance (e.g. Karl, 1997; Leith, 2003; Le Billon, 2005; Robinson et al., 2006; Rajan, 2011; Mitchell, 2012; Ross, 2012; D’Argent et al., 2017). Our intended contribution is to take insights from such authors and explore what might be gained by reading the politics of resource extraction through a theoretical framework that grows out of the literature on political settlements and national developmental trajectories (Hickey, 2013; Hickey et al., 2015). In the remainder of this introduction, we discuss how we made use of these (p.3) literatures, introduce the questions that the book addresses, describe the methodology underlying our research, and briefly introduce the remaining chapters.
1 Mapping the Conceptual Terrain
1.1 The ‘Resource Curse’ and the Politics of Extractive Industry Governance
In the words of Ross (2015), ‘The resource curse might be defined as the adverse effects of a country’s natural resource wealth on its economic, social, or political well-being’ (2015: 240). Since the early 1990s, debates over the nature and existence of this ‘curse’ have been the dominant frame for discussions of extractive industry governance. These debates emerged around the claim that growth indicators for countries that are dependent on natural resource extraction (especially minerals, oil, gas, and timber) do not compare well with those of countries whose economies are more diversified and demonstrate more substantial manufacturing and service sectors (Gelb, 1988; Auty, 1993; Sachs and Warner, 1995). It was also argued that extractive industry-dependent economies were less effective in reducing poverty than more diversified economies (Ross, 2001b) and that ‘natural resource wealth tends to adversely affect a country’s governance’ (Ross, 2015: 239). Claims regarding the existence of a ‘resource curse’ quickly stimulated a counter-literature arguing the converse, or at least asserting that there was no clear pattern to the relationship between resource dependence and national economic and social performance (Davis, 1995; Davis and Tilton, 2005; ICMM, 2007; Brunnschweiler and Bulte, 2008; Davis, 2009). At stake in much of this debate are arguments about statistical methods, choice of indicators, time horizons, and more. Discussions have made clear just how difficult it is to define and then measure appropriate indicators of ‘resource dependence’ or ‘institutional quality’, which then makes it challenging to build causal arguments about the effects of one of these on the other. Similar problems of indicator definition and measurement apply to other terms that circulate in discussions of the resource curse.
As debates progressed, they moved away from making strong statements on whether hydrocarbons and minerals were ‘good’ or ‘bad’ for development, towards recognition of the diverse outcomes associated with extractive industry wealth, and efforts to account for this diversity (Bebbington et al., 2008; Thorp et al., 2012; Ross, 2015). Most frequently, institutional quality became identified as the primary determinant of these diverse outcomes, and ‘getting the institutions right’ became a central objective in efforts at ‘escaping the resource curse’ (Humphreys et al., 2007). As authors, we take it as a given (p.4) that the relationship between extractive industries and the quality of development is indeterminate and has much to do with the nature of dominant institutions. However, we would also argue that more often than not, resource-dependent development has delivered disappointing results and been plagued by a range of pathologies, and that the balance of political power has a significant influence on the qualities of the institutions that influence the effects of natural resources on development. Our primary concern is to understand how the interactions among politics, natural resources, and institutional form play out over time and across countries with a view to identifying convergences that might help us theorize these relationships.
There is already a literature on the political drivers of resource-dependent development on which we build; some of this precedes the ‘resource curse’ literature, while some of it has emerged as part of that literature. Of particular importance is the early work of Terry Karl (1987, 1997) that used a political economy lens to understand how oil exporting affects the relationships between politics, democracy, and development. In The Paradox of Plenty, Karl drew inspiration from her long-term empirical work on Venezuela, as well as the experiences of other oil and mineral-dependent nations, to move beyond generic assertions that ‘politics matter’. Instead, she argued that the pacts and bargains made by elites in one phase of history, including the decisions they make about whether and how to sow revenue from resource extraction, deeply affect the space and options for development thereafter. She examined the decision to exploit oil revenues in the second half of the twentieth century in Venezuela, and showed how it affected the country’s ability to create democratic institutions over time. In particular, she demonstrated the ‘paradoxical’ relationship between the abundance of oil and the existence of institutions that were deeply corrupt and fragile. This work showed the weight of external factors and actors in the politics of oil exporters, and how bonanza rents from such resources can ‘grease the wheel’ of domestic pact-making, keeping dissidents quiet by buying them off. When that no longer works, however, the revenues are often used to build up militaries that use force to exclude and suppress opponents. Yet her analysis also showed how such pact-making can unravel, as more people become excluded from the original deals and are negatively affected by the dependency of the state on extraction. When the structural inequalities are no longer masked by redistributive politics, things fall apart. In many regards, Karl’s work was not just analytical—it was sadly prophetic, certainly for Venezuela.
The political significance of oil has also been at the core of the contributions of Michael Ross (2001a, 2012) and Timothy Mitchell (2012). Ross (2001a, 2012, 2015) argues that resource curse-like outcomes are much more systematically associated with oil dependence than with the extraction (p.5) of other mineral resources. There are various reasons for this, including the price volatility of oil. However, Ross places particular explanatory emphasis on whether oil resources are in private or state hands. He suggests that oil began to have more systematically adverse impacts on the depth and quality of democracy following the creation of state-owned oil companies in the 1970s. State ownership offered elites privileged instruments with which to politicize oil—whether through using revenue for purposes of political patronage, pact-making, or private gain, or through instrumentalizing the control of oil as a metaphor for talking about imperial dependence and resource sovereignty. In either instance, the developmental effects are adverse, he argues. For Mitchell (2012), whose focus was more broadly on the relationships between the control of carbon and democracy, oil has also had nefarious political effects. These have derived from the capital intensity of oil extraction and transportation, which means that oil tends to strengthen economic and political elites rather than organized labour. In addition, oil has attracted the attention of powerful transnational companies with close ties to their host states, who together undermine national sovereignty and regulatory institutions in the countries whose oil deposits they have sought out. The adverse political consequences of oil also figure prominently in Michael Watts’s sustained work on the Niger Delta (Watts, 2004a, 2004b; Watts with Kashi, 2008). Watts insists that any effort to understand the relationships between oil, politics, violence, and development must attend ‘to the ways in which oil becomes an idiom for doing politics as it is inserted into an already existing political landscape of forces, identities and forms of power’ (Watts, 2004b: 76). He demonstrates that this political landscape is one characterized by longer histories of colonialism, post-independence state-making, subnational power, and identity formation around locality and indigeneity. The politics driving oil governance thus have deep historical and cultural roots.
While also pointing to the centrality of politics in determining the relationships between natural resource wealth and development, Robinson et al. (2006) pursue a completely distinct analytical line from that of Watts, one that attends to political incentives rather than history. They do this through formal modelling of the ways in which leaders calculate how they will use the revenues from extractive industries if their goal is to stay in power. They argue that these calculations lead political authorities to over-extract resources in order to finance strategies for retaining power. While the incentive to over-extract declines in contexts of resource boom, the tendency to use resources to purchase political support increases, leading to further inefficiencies in how public resources are allocated across the economy. In either instance, the development consequences are adverse, as financial resources are not invested in those sectors where they would have the most economic (p.6) and social impact. The authors argue, however, that the extent to which this is so depends on the quality of institutions and the extent to which they constrain leaders’ proclivity to misallocate state revenue. What their framework does not offer is a means of accounting for the quality of such institutions in the first place.
Other parts of the literature on the politics of natural resource extraction explore the political strategies of activist groups and social movements (Kirsch, 2012; Bebbington and Bury, 2013; Li, 2015), the emergence of civil conflict (Le Billon, 2005; Collier et al., 2009), and the politics of scholarly engagement (Coumans, 2011), among others. The precursors laid out by Karl, Ross, Watts, and Robinson et al., however, help define the base from which we work in this study. Together, their insights make clear that elite strategies of political pact-making are deeply affected by the presence of natural resource revenues in ways that then structure subsequent room for manoeuvre in politics and policy. Watts and Karl show in different ways how history matters, and Watts draws attention to the importance of ideas and the cultural realm in affecting political action around resource revenues. Each of them—though particularly Ross and Robinson et al.—also emphasize that whoever owns resources (the state or private actors) influences the impact of resource extraction on politics. Like much of the broader literature in which their work is situated, each author shows that institutions matter to the relationships between extractive industry and development. Karl and Watts also imply (albeit with quite distinct languages) that the nature of such institutions must be explained from within the same concepts that are used to explain the broader relationships between politics and natural resources. Put another way, one cannot build a framework to analyse the relationships between politics, extractive industry, and development, and then introduce ‘institutions’ as an independent mediating variable. Such institutions are themselves a product of the same relationships that they mediate and have to be accounted for historically.
In the following sections, we develop a framework for addressing these relationships by drawing upon the literature on political settlements. We suggest that this framework extends earlier work on the politics of natural resource governance. It centres the focus on power and elite relationships, analyses institutions in terms of these political relationships, and accounts for drivers of institutional and political change that can derive from within national elite pacts, as well as from the strategies of subaltern and transnational actors. In this way, the framework converges with other recent work that seeks more explicitly to go beyond the focus on ‘institutions’ within the resource curse literature to focus on the political coalitions and ideas that shape the emergence of institutions and how they actually function in practice (e.g. Poteete, 2009; Hickey and Izama, 2017; Mohan et al., 2017).
Khan (2010) defines a political settlement as ‘…a combination of power and institutions that is mutually compatible and also sustainable in terms of economic and political viability’ (2010: 4). The concept describes the agreements among powerful actors that enable social order in a given country (North et al., 2009; Khan, 2010). Framed slightly more broadly, for di John and Putzel (2009), a political settlement ‘refers to the balance or distribution of power between contending social groups and social classes, on which any state is based’ (2009: 4).
In these overlapping formulations, a focus on political settlements draws attention to the ways in which governing relations are negotiated among economic and political elites. Such negotiations produce distributions of economic and political opportunity that reflect the bargaining power of these different elites. Elite status is not necessarily understood in terms of long-standing historical privilege based on race, ethnicity, kinship, or inherited wealth, but more specifically refers to that minority of individuals or groups that hold resources which allow them to exercise control over the state and the overall distribution of assets and rights in society. Such resources can be economic (land and financial or industrial capital, held nationally or transnationally), intellectual (technocrats, think tanks, mainstream media), or coercive (in the case of military elites, control over the organized use of force) (di John and Putzel, 2009: 15; Hickey, 2013: 3, 4).
In maintaining these orders and distributions, elites who are party to the settlement—who are ‘on the inside’, if you will—have to manage relationships with actors who are not ‘on the inside’—actors often referred to as ‘excluded factions’. In the sometimes dichotomizing language of political settlements, these factions are referred to as ‘horizontally’ excluded and ‘vertically’ excluded. Horizontally excluded factions are those who are opposed to the coalition of elites controlling the state and social order; vertically excluded factions are those groups who are supportive of controlling elites but who are not themselves party to any of the negotiations through which mechanisms of control and order are determined. In a strict sense, these vertically ‘lower’ factions are not ‘excluded’ because they feel an affinity with governing elites, but they are in no way ‘on the inside’. Elites have to manage relationships with these different factions in order to ensure that they do not threaten the settlement. Such relationships can be managed through the exercise or credible threat of force, through the framing of legitimate discourses (e.g. ideas around what constitutes legitimate forms of power and credible models of national development), or through clientelism and the manipulation of incentives (e.g. through distribution of benefits, entitlements, and subsidies).
(p.8) How these relationships among elites, and between elites and excluded factions, are managed depends, in Khan’s terms, on the distribution of ‘holding power’ among different actors (i.e. their power to exercise a hold over others). The settlement is what holds these different relationships of power together in stable form. In this sense, the nature of the settlement (the balances of power and how they are managed) drives the nature of institutions and policies. These institutions and policies are the formalization of the practices and rules governing the distributions that lead all parties (included and excluded) to stay on board with the settlement rather than to openly contest and seek to overturn it.
Various authors have taken these ideas and converted them into typologies that seek to relate ‘types’ of settlements to ‘types’ of policy regimes or forms of development (Khan, 2010; Hickey, 2013; Levy and Walton, 2013; Levy, 2014; Booth, 2015; Levy and Kelsall, 2016). Levy and Walton (2013) and Levy (2014) argue that how elites within a dominant coalition manage these relationships and approach development depends on whether the broader political system is characterized by competition among leaders and parties or by the dominance of a particular leader or party. Political competition would lead elites to recruit clients into the coalition as a means of holding on to power (a form of political settlement they refer to as ‘competitive clientelism’), and would therefore tend to operate with shorter-term time horizons oriented towards recruitment of clients ahead of electoral cycles. The notion of clientelism here is not in the traditional sense of patron–client relationships of dominance and dependence, but rather of elites who, in competing for political power, need to recruit and sustain supporters both to assume power and then to remain in power. Conversely, dominant leaders and parties worry less about client recruitment and so use resources for longer-term political projects based on particular ideas of development or personal gain, a form of settlement referred to as a dominant leader/dominant party settlement. Dominant party/leader settlements can range from those underlying the authoritarian modernization project of a developmental state through to personalized forms of dominance, with or without corruption and theft. Forms of leadership may be autocratic or elected, but in either instance are dominant.
1.3 Change in Political Settlements
Political settlements frameworks are helpful in different ways: in thinking about the dynamics underlying bargains; in conceptualizing the relationships among states, actors within ruling coalitions, and other actors; and in elaborating typologies for classifying forms of rule. However, they are less powerful in other regards. Perhaps most importantly, they have no obvious theory for why and how groups emerge, decline, and become more or less powerful. (p.9) Second, and in some regard related to the first point, the frameworks have no clear theory of the relationship between time, power, and the nature of settlements. While political settlements frameworks clearly presume time, talking of different settlements as preceding or following each other, they say little about how this movement happens, and, more importantly, about the ways in which time ‘t’ affects time ‘t+1’. Yet to the extent that human agents learn, that memories affect action, that formative discourses are framed and then continue to circulate and affect future action, and that prior institutional forms affect the nature and emergence of subsequent institutions, then these questions of time are critical (cf. Watts, 2004b; Hall, 2010; Mahoney and Thelen, 2010).
Such processes of change in political settlements demand theorization and empirical demonstration. Within the language of the settlements framework, such change would primarily be explained as a result of excluded actors becoming more powerful, which requires theorizing how such power might accrue. However, it might also be explained in terms of changing ideas regarding what are deemed to be legitimate and acceptable distributions of political and economic opportunity. As such ideas change, the strategies of incorporation, clientelism, and/or repression that have been used to sustain a settlement may become less effective at quelling unrest and resistance. Finally, change may also occur because factions within elites begin to shift their calculations of the settlements that suit them best and so begin to realign alliances.
The sorts of empirical phenomena that might drive such change can also vary widely.
One driver of change may be the emergence of national or subnational actors who steadily develop the capacity to challenge the dominant settlement. Such capacity may develop for different reasons: the increasing value of the income streams controlled by excluded factions; the increasing ability of these factions to mobilize a broad base of citizens; changes in their proclivity to use violence; or their ability to reframe debates and challenge ideas that have historically legitimated particular distributions of opportunity and power. In this sense, conflict can be a mechanism of change in the national settlement. Across much of Latin America, the example of indigenous movements comes to mind as a clear case of such emergence of new actors. For while indigenous people are still disadvantaged, their presence within national debates and within the nature of national settlements has expanded enormously over a fifty year period in ways that have shifted the terms of national debates (Yashar, 2005; van Cott, 2007). Another source of change may be the arrival or increasing leverage of transnational actors who for one or another reason have the capacity and legitimacy to affect arrangements in any given country. Such transnational actors might drive changes in dominant (p.10) ideas, contribute to the increase in ‘holding power’ of certain actors, or present new opportunities and incentives that lead elite parties to the settlement to reconsider alliances. Of many possible examples, the joint arrival of Newmont Mining Corporation and the International Finance Corporation to Peru in the early 1990s would be a case in point, as it increased the leverage of national mining elites, helped craft the idea that Peru could be a ‘mining country’ once again, and helped endorse the idea that opportunities should be distributed in Peru so as to favour mining investment (Bury, 2005).
1.4 Natural Resource Governance and Political Settlements
The logic of the political settlements literature implies that the ways in which natural resources (or in our case, extractive industries) are governed would be a function of the political settlement. This approach to extractive industry governance is important and useful in that it keeps hold of a focus on the relationships between institutions and development, while decentring the primacy of institutions as causal factors and instead placing power and politics at the core of any analysis. This framing suggests that efforts to seek significant change in sector governance without important changes in the balance of power within a settlement are unlikely to prosper. A related implication would be that a starting point for reform efforts should be to analyse the existing settlement and identify room for manoeuvre, or ‘cracks’ (Holloway, 2010; Bornschlegl, 2017) within that settlement and work from there to identify politically feasible changes. This would be quite distinct from classic reform strategies that identify technocratically desirable changes, try to implement them, and then end up banging up against the nature of existing power relations and inter-elite pacts. Similarly, it would be distinct from those classic oppositional approaches to resource extraction that focus on direct resistance and advocacy but frequently lack a language and a strategy for operating among, and negotiating with, elites, or for shifting the overall balance of political power in any way that might be sustained.
A political settlements approach to extractive industry governance is also helpful because it provides a set of concepts and typologies that enable comparative discussion of the politics surrounding diverse forms of resource extraction. It is one thing to argue that institutional change is central to any improvement in the relationships between extractive industry, environment, and development, and to recognize that the drivers of institutional change are primarily political (Karl, 2007). It is quite another thing to theorize those political drivers in a way that allows them to be analysed systematically and comparatively and on the basis of shared typological frameworks. The political settlements framing offers such a comparative analytical framework. Importantly, in contrast to the bulk of research on extractive industries that (p.11) treats large-scale mining, artisanal and small-scale mining (ASM), and hydrocarbon extraction as distinct objects of study, a political settlements approach permits discussion of these different sectors all within the same framework. It could also allow for incorporating ancillary sectors such as hydropower and large-scale infrastructure into the same analysis.3 This is because the primary organizing language is not one that is related to the sector, but rather one that describes the relationships between power, institutions, and order.
Just as the political settlements lens is helpful for analysing extractive industry governance, we would also argue that engaging with the question of natural resource extraction and its implications for development brings new insights to the political settlements literature. In particular, a focus on natural resources raises issues of place, scalar relations, and time in ways that also help explain how new actors emerge and accrue holding power, while other actors wane and cease to be parts of governing coalitions. We will elaborate these ideas in more detail in Chapter 6 in the light of the four country studies. Here we simply outline why the weakness (or in many instances, complete absence) of a sensibility to place/space, scale and time/memory is a serious limitation of the political settlements literature.
Our starting point for this part of our argument is the concept of ‘territorializing projects’ (Wainwright and Robertson, 2003; Wilson, 2004; Humphreys Bebbington and Bebbington, 2010). Territorial projects are efforts to construct and govern space in ways that accord with a given actor’s or coalition’s political and ideological concerns. Though rarely framed as such, political settlements are themselves territorial projects. If a political settlement is about securing social order through an elite agreement over the state and institutions, this order always refers to how a particular geographical space should be governed. By extension, if a political settlement involves negotiations among elites and the management of relationships with excluded factions, then the territorial project underlying the settlement reflects an accommodation among the distinct territorial projects of the parties to the settlement as well as coexistence with (and likely domination over) the territorial projects of excluded factions. In this sense, the organization of space is an outcome of political settlements. But the converse is also the case: conditions set by the contradictions and synergies among these diverse territorial projects set limits on the settlement. This is reflected in the different ways in which the tensions between centralization/decentralization, unitary state/federal state, single nation/plurination, and centralized control/territorial autonomy are resolved across different country contexts.
The spatiality of natural resources is often a factor in the constitution of actors and their territorializing projects. For example, the emergence of cooperative miners as a political actor that has pursued its own projects of territorial control in the highlands of Bolivia is itself a consequence of prior (p.12) geographies of state-led mining and of existing mineral deposits (because many cooperatives emerged from the ashes of these mines). In Zambia, the constitution of miners as an organized group emerged as part of the same process that produced the Copperbelt as a mining area, and that group continues a (more implicit than explicit) territorial project that revolves around the Copperbelt. Meanwhile in Ghana, the inherited spatial–administrative structure of chieftaincies, stools, and local authorities influences how actors organize politically and make spatial and identity-based claims around mining and oil development.
In each of these examples, we have subnational actors pursuing projects of territorial control that are in some way related to the geography of natural resources. Also in each instance, these subnational actors have made their presence felt in national politics, and have been involved in ongoing negotiations with national elites over who should have access to subsoil resources and/or how the rents from extractive industry should be shared between the territory and ‘the centre’. Such examples help illustrate the notion that any given settlement is also a settlement over the distribution of powers and opportunities between the centre and the subnational, and that this distribution (coupled with the strength of ideas about the relative importance of unity and devolved autonomy) is an important factor in the constitution of political actors. For example, in the Peruvian mining sector since the 1990s, some of the most important players have been subnational political leaders who have emerged as powerful in part because of the ways in which the geography of subsoil resources, ideas about decentralization, and the structure of subnational government intersect with each other (cf. Watts, 2004b). This does not mean that the political and territorial projects of these subnational leaders are of a kind. For instance, in the north of the country, one regional government’s president (himself a product of social mobilization in the region) was a key player in stopping the expansion of mineral investment. In contrast, a different regional government president in the south of the country played a critical role in securing greater protections and benefits for the region as a means of brokering conflicts over resource extraction in such a way that allowed a large copper mining project to proceed. In both instances, these leaders became important because of the existence of subnational administrative powers (otherwise they would not have become regional presidents) and because of the political resonance of the territorial projects that they articulated and the ways in which these have intersected with the geography of mining.
Examples such as these are also significant because they point to individual and collective actors emerging subnationally through the contentious politics surrounding natural resource extraction, and then becoming ‘national’ actors by virtue of being able to exercise power on the national stage. This in turn is a (p.13) reminder that there is nothing inherently ‘national’ about ‘national’ elites. Such elites are not born ‘national’, nor are they necessarily defined by being based in the principal urban centres of the country. Rather, they are defined by an ability to exercise power over decisions that affect much or all of a national territory. They become ‘national’ elites because they have accrued resources which allow them to exercise influence over the state and the overall distribution of assets and rights in society. Put another way, all actors are place-based but differ in their ability to exercise influences over other actors in other places. Certain places are more apt to produce actors with such national projection, and capital cities are indeed one such place. What these two Peruvian examples illustrate is that territories with significant mineral resources and the potential to produce very substantial rents are also apt to produce such actors. The types of political project these actors pursue, and thus their implications for the political settlement, are influenced by these and other qualities of the places from which they emerge.
This does not mean that all place-based actors who become able to influence—or force themselves into the core of—national settlements emerge on the back of natural resources, but many have. In this sense, a natural resource lens draws particular attention to the importance in any political analysis of understanding the place- and resource-based processes through which parties to (or strong excluded factions from) the national settlement emerge. Something similar occurs with transnational actors. Indeed, just as the ‘national’ character of certain actors should be understood historically, so too should the transnational character of other actors. That is, actors become transnational just as actors become national. That certain actors are ‘transnational’ reflects the fact that they have acquired the power to exercise influence beyond the borders of the national spaces from which they originated and in which they have their primary locations. In turn, this acquisition of global power grows out of prior processes of becoming powerful within the domains of their original national states. Once again, the geography of natural resources becomes important here. Mining and hydrocarbon companies like Vale, Petrobras (Brazil), or Chinalco (China), for instance, only became influential in Africa or Latin America because of their prior emergence as nationally important companies and political actors within their countries of origin. Newmont (the principal owners of Yanacocha gold mine in Peru) grew out of mining in the western United States; Broken Hill Proprietary (BHP) grew out of Broken Hill, Australia; and so on.
While these issues are not unique to the natural resources sector, a focus on extractive industries does help bring themes of place, territory, and scalar relations into the forefront of thinking about how political settlements are contested and stabilized, and how certain elites have emerged to become party to elite settlements. A focus on natural resources also brings one more (p.14) important contribution to political settlements discussions. The history of political contention surrounding extractive industries makes palpable that it is not just abstract ideas (of efficiency, national unity, development, etc.) that help constitute settlements (Hickey et al., 2015), but also memories. Of course, much is remembered in national politics; but minerals, and over the last century oil, are often remembered in particularly resonant ways. It is not accidental that in the opening of Eduardo Galeano’s Open Veins of Latin America (Galeano, 1971), the book chronicling centuries of extraction in the region that Hugo Chávez gave to Barack Obama, Spanish mining in the Andes features prominently. Myrna Santiago (2006) has argued that oil is central to Mexican identity, and memories of its nationalization still loom large in debates over how the sector should be governed. Memories of copper in Zambia, and its extraction by colonial powers, still impinge on political alliances and proposals for mining governance. In Ghana, contestations between chiefs and traditional authorities over the distribution of mineral rents are often based on historical memories that go back to the precolonial period. In Bolivia, silver and gold speak both of pre-Hispanic greatness as well as colonial and imperial subjugation and dependency, while tin invokes memories of a small number of early twentieth-century families who become phenomenally rich and powerful while delivering very little by way of development. The politics surrounding these resources can never be separated from those meanings. More generally, ideas of resource nationalism—that help drive the extension of state control over mining and hydrocarbons—are more often than not imbued with memories of other powers which have taken advantage of national resources in times past.
1.5 Extractive Industries and Inclusive Development
Extractive industry faces particular challenges around inclusion, in large measure because—except in the case of ASM—the activity of extracting minerals, oil, and gas depends upon the existence of significant asymmetries of power, resources, and technical expertise. The capital intensity of extraction requires elites with control over or access to investment funds measured in the billions, or at least hundreds of millions of dollars. The technologies associated with the investment generate fewer and fewer jobs—and show trends towards increased use of robotics to operate mine sites in the spirit of increasing occupational safety (by reducing the number of human jobs where there is risk). The environmental and social impacts of extraction are concentrated in space, leading to spatial inequalities in exposure to these impacts. The rents generated by extractive industry can be very large, and are controlled in the first instance by just two sets of actors: the company, and government tax authorities. For these different reasons, any model of development based on (p.15) resource extraction is a compensatory model (Gudynas, 2012; Bebbington et al., 2014). That is, it is a model that compensates for these original, inherent inequalities, and that uses compensation in the form of transfers to offset these inequalities and asymmetries and to seek development. These transfers can take the form of corporate social responsibility (CSR) programmes, public social protection programmes, payments for losses, indigenous and local development funds, fiscal transfers to subnational governments or chiefs in areas affected by extraction, and so on. A large part of inclusion linked to resource extraction, if there is inclusion, therefore occurs through compensation (Humphreys Bebbington and Bebbington, 2010).
It merits note that the very idea of inclusion itself implies another asymmetry: namely the incorporation of other worlds into the world defined by development. For most people working in development this would be a non-issue, on the presumption that, of course, the rural poor would want greater inclusion in a world defined by the desirability of increased income, consumption, and access to basic services, as well as of enhanced participation in the political systems of the dominant order. This is not a position shared by all. Indeed, some authors writing critically about the extractive economy would argue that this notion of inclusion is another form of violence, one that refuses to acknowledge—or at least grant validity to—alternative ontologies (Blaser, 2010; Yampara, 2011; Gudynas, 2014; de la Cadena, 2015). Such authors might argue that any notion of inclusive development should be considered through, and superseded by, concepts such as buen vivir, vivir bien, and sumak kawsay, which emphasize the importance of living well, in harmony with other humans and with nature, rather than the inherent desirability of income, consumption, and economic growth. In the Andean countries, at least, such authors would argue that extractivism and any development based upon it should be viewed as destructively colonial because it requires the despoliation and disappearing of worlds in order to gain access to the subsoil (whether these are mountain worlds, or forest worlds: Descola, 1996; de la Cadena, 2015). In these readings, enhanced inclusion is less the goal than the search for modes of joint existence between different worlds that connect but simultaneously diverge (de la Cadena, 2015).
This is not the approach we take here. This is not because we necessarily reject the arguments of such authors. Rather it is a matter of consistency: the language of political settlements is an inherently modernist one, and so it is only consistent on our part to proceed with a notion of inclusive development informed by the commitments of modernism. As we seek to assess the ways in which political settlements interact with extractive industry governance and the extent to which this interaction fosters development that enhances or diminishes inclusion, we have focused on process indicators of this inclusion. That is, our primary approach has not been to measure the impacts of (p.16) extractive industry on inclusion through evidence on increased incomes and consumption, or increased access to services, for instance (though we include some of this data). Instead, we have assessed inclusion through the extent to which extractive industry has affected opportunities for political participation, has been used to increase budgets for social spending, has led to increased budget for subnational authorities, has created new labour markets that expand quality employment, and has enhanced the status of poor peoples’ rights. In keeping with the focus on institutions, then, our emphasis has been on the extent to which extractive industry has enhanced or weakened institutions that foster social, economic, and political inclusion. In this sense, we work from what might be termed a ‘critical modernist’ view on inclusion that values notions of emancipation, progress, and rights, while also allowing for popular reinterpretations of the forms that such aspirations should take (c.f. Peet and Watts, 1996; Peet and Hartwick, 2015).
To summarize, then, the basic conceptual framework running through our argument takes its core ideas from political settlements theory, as well as from earlier work on the politics of extractive industries, namely: the importance of pacts, bargains, and elites; the significance of relationships among elites, excluded horizontal factions, and lower level vertical factions; the political drivers of institutional form and performance; and the sources of stability and instability in settlements. To these, we add an emphasis on: the significance of relationships across time; the importance of political and cultural ideas tied to natural resources; the interacting influences of transnational, national, and subnational actors, especially those living in areas where natural resources become rapidly valorized; and the factors driving the emergence and decline of powerful actors. In particular, we associate much of this emergence and decline with the ways in which national and local political economies of minerals and hydrocarbons interact with transnational (f)actors. New actors emerge and decline as part of changes in the economic structure (cf. Boix, 2008), which are themselves often related to articulation with external markets. In addition, actors can emerge because of articulations with transnational ideas and activists that give new resources and legitimacy to those local and national actors in ways that are themselves often linked to natural resources.
The framework pursued here also suggests the value of making natural resources central to political settlements thinking. This is because in many countries access to and control over resources and resource rents are central to elite bargains; the transnational valorization of resources serves to bring political actors into being and into demise; and the economic and cultural values apportioned to natural resources become critical elements of state and nation-building.
2.1 Our Research Concerns
Overall, our intent is to use a political settlements approach in a way that captures the key insights of and builds on existing literature on the politics and institutions of natural resource governance, while proposing a more explicit typological and analytical framework that helps establish power and politics, rather than institutions, as the central focus. At the same time, the book seeks to deepen the political settlements framework in several ways. First, as just noted, we believe that the task at hand should not be one of only applying a political settlements lens to understand natural resources, but also exploring the ways in which the very qualities of these natural resources have implications for how they are governed and the forms of politics that surround and drive this governance. In this sense, we bring a natural resources lens to political settlements as much as the other way around, and work from the presumption that the materiality of mineral and hydrocarbon resources matters (Bakker and Bridge, 2006; Bebbington, 2013b). This materiality is important both because the qualities of deposits affect their potential for commodification and profitability, and because these resources have geographically specific locations. These locations interact with other geographies and histories (of populations, territorial claims, drainage basins, land use, urban–rural and centre–periphery relations, etc.) in ways that affect incentives, ideas, and interests surrounding resource extraction and the emergence of political actors.
Second, the book brings an explicit attention to scale and space in understanding political settlements. While we are in no doubt that the sorts of national-level relationships that dominate political settlements analysis are indeed key to understanding the nature of these settlements and the ways in which they interact with natural resources, we will argue that political settlements need to be understood as a product of scaled processes, running from the subnational to the transnational. We will argue that the materiality and political economy of natural resources make subnational and transnational actors particularly powerful players in the constitution of settlements.
Third, the book recovers the importance of history and historical sensibility in political approaches to resource governance. Most work on political settlements—especially that which is also concerned to speak to policy and/or questions of inclusion and development—has tended to focus on recent time periods.4 This stands in some contrast to some of the formative texts underlying political settlements thinking, which address processes over longer historical periods (North et al., 2009; Khan, 2010; Acemoglu and Robinson, 2012),5 as also does Karl’s work on comparative oil states and regime change (1997). The reason for addressing history is not merely to give a ‘fuller (p.18) account’ of processes, but to show how this history is important in understanding contemporary arrangements and policy options. A longer-term historical analysis helps understand more fully: the conditions under which resource governance arrangements change, or do not change; the nature and origins of contemporary elites and excluded factions; the relative resilience of their holding power; and how history is remembered (Bebbington, 2013b).
2.2 Our Approach to the Research
There were two points of departure in building a methodology to speak to these research concerns. First was a deeply held belief that Latin American and African experience have much to contribute to each other, and that comparative work across the two continents is both valuable and valid. Team members had fruitful prior experience with such cross-regional comparison. When similarities emerge in the context of apparently large differences in political economic and post-colonial context, then this implies something fundamental about the phenomenon in question. When there are differences, then the challenge to the analyst is to use systematic comparison to tease out the causal relations at work in the relationships between political economy and the issue being studied and why these relations work out differently across space. Given that the Effective States and Inclusive Development (ESID) initiative, of which this book is a part, has been funded by the UK Department for International Development, the overwhelming tendency has been for the programme to focus on African and Asian countries. However, we assert that Latin America’s long-standing and troubled history with natural resource governance make it an important case for the research topic at hand. Further, we believe that comparative analysis between it and Africa yields crucial insights for both regions and the topics of natural resource governance and political settlements more broadly. We began this project committed to the value of learning across regions, and ended only further convinced of the validity of this conviction.
The second point of departure was a focus on countries as a unit of analysis. As we have discussed in Subsection 2.1, the project was motivated to go beyond the methodological nationalism of political settlements research and to recognize the causal importance of other scales of political action and bargaining. However, if we were to challenge methodological nationalism, it would be important to explore the significance of these other scales from within a focus on the country level. Furthermore, notwithstanding our multi-scalar approach to the relationship between settlements and the governance of resource extraction, we remained convinced of the importance of the national scale. Country-level analysis was, therefore, clearly the appropriate entry point for this enterprise.
(p.19) The challenge was then to identify countries that it made some sense to compare. This brought us to a focus on Bolivia, Ghana, Peru, and Zambia. These four countries were selected because they share certain overall similarities with respect to politics and natural resources, while manifesting specific differences that allow for pairwise comparisons. All four are low- to middle-income countries, though their relative status has changed over time. None of the four countries has experienced extreme levels of autocratic rule characterized by sustained forms of kleptocratic natural resource extraction.6 Each has a long history of hard-rock mining, and three of them have more recent histories of hydrocarbon extraction (Zambia is the exception). In each case, these histories are both colonial and post-colonial, and have meant that each of these economies has long been substantially dependent on commodity extraction, although the degree of dependence varies among the countries and also over time. In addition, each country has experienced shifts over time in the ways in which the extractive sector is governed, allowing for an analysis of how far these shifts can be explained by global trends and ideas, international commodity prices, or national and subnational political relationships. Likewise, each of the four has experienced similar national political changes, including periods when single leaders, institutions, or parties have been dominant and others that have been characterized by greater degrees of inter-elite and democratic competition. This allowed us to address the effects of these political changes on the governance of extractive industries, and to compare forms of extractive industry governance under dominant party/leader regimes and under regimes characterized by competition among contending elites.
In designing our approach to these country studies, our belief in the importance of history in political settlements also had implications for how we defined our primary unit of analysis. We opted for what we called a concern for the longue durée, and so decided that in each of Bolivia, Ghana, Peru, and Zambia, we would begin our analysis at the end of the nineteenth century or beginning of the twentieth.7 While our time horizon is not as expansive as that often adopted by the French Annales School, from which the term longue durée originates, it is motivated by the same conviction that many ideas, structures, and relationships change only very slowly, and that it is important to capture these slow histories if one is to understand the governance of extractive industries in four countries with centuries-long histories of mining but disappointing development outcomes. This concern for the slowness of change also seemed to us consistent with the very notion of political settlements. Indeed, as will become clear in the chapters, we ended up identifying elements of these settlements that have proven highly resistant to change over the long term.
In this sense, then, our unit of analysis was a century or so of national history. In order to analyse the relationship between political settlements and (p.20) extractive industries over this time frame, and to keep the country analyses comparative, our approach combined the following elements. Literature review was critical as a mode of building the historical narrative on political settlements. In this sense, our country level research builds upon, and from, other research literature on social, political, and economic history in each of the four countries. Second, we conducted rounds of interviews with key informants. These informants were a combination of participants in and observers of the relationships between politics and extractive industry governance that we were studying. Interviewees ranged from former ministers and elected officials to social movement leaders and miners.8 In each country, we also made use of recent field research in sites of resource extraction in which we had been involved in the context of other studies. Third, workshops were conducted in each country with a mix of researchers, members of civil society, mining company representatives, and public sector actors working on extractive industry governance.9 These workshops were important in refining our interpretations and arguments. Research was also presented for feedback at conferences and seminars. Fourth, drafts of the country analyses were shared widely with experts10 for critical commentary, and these reviews helped us to refine our arguments and address crucial themes we may otherwise have missed or misunderstood. Fifth, the five of us were spread across three continents during the conduct of the research. Keeping the research comparative and allowing for each of us to feed into the analysis in each country required fluid communication. At different stages of the project, we convened to workshop versions of the country studies and attempts at synthesis of emerging findings. Workshops were held among all five of us on two occasions, and among subgroups of us on three other occasions. In the final chapter, we return to these methodological choices and reflect on some of their implications for the book as a whole.
3 Outline of Book
Following this introduction, the core of our book is made up of four chapters, each based on one of the four countries in which we worked. The chapters follow similar—though not identical—forms. They first trace the history of national political settlements, beginning from the late nineteenth century through to the second decade of the twenty-first century. They then relate these settlements to the changing ways in which mining—and in the case of Bolivia, mining and natural gas—have been governed and regulated over a period of approximately 125 years. They trace relationships between these forms of natural resource governance and the nature of political settlements and political conflict. The chapters also relate this changing governance of (p.21) natural resources to the changing policies for, and patterns of, socio-economic and political inclusion. While history weighs heavily in each chapter, the bulk of the analysis focuses on the period since 1990.
At one point, we debated whether, rather than country analyses, the book would be better organized around key elements in the relationships between political settlements, resource extraction, and inclusion, marshalling evidence from all four countries to explore those elements in turn. Ultimately, however, we concluded that country level dynamics and the influence of national history and memory are central to understanding the relationships between politics and resource extraction. A thematic organization of our data would displace the centrality of both country and history in the relationships we are exploring. Opting for the focus on the country, however, makes for some quite long chapters: in some sense, each chapter is a mini-monograph. Moreover, while the country chapters each engage with the political settlements literature, and have been in conversation with each other over the course of this research, we have opted to write each of them as a relatively free-standing piece, recognizing that some readers will only dip into those country analyses of most interest to them. As a consequence, each country chapter opens with a short engagement with the political settlements framework, and ends with conclusions regarding that framework from the perspective of that particular country experience.
The risk of this extended country-based approach is that the reader may get immersed in country details, but lose a sense of convergences and differences across Bolivia, Ghana, Peru, and Zambia. Our final chapter therefore offers this comparative and synthetic reading of the four country essays, relating that comparison back to our conceptual framework and our guiding research concerns. In that chapter, we lay out the ways in which we think a focus on extractive industries and natural resource governance makes specific, conceptual contributions to political settlements theory around issues of space, scale, time, and memory.
(2.) For just a few examples of such proposals, see the websites and policy documents of organizations such as the Natural Resource Governance Institute, Oxfam, the International Council of Mining and Metals, or the World Bank.
(3.) Investment in hydropower is often closely related to large-scale mining as a source of energy for mine and processing sites, while infrastructure such as waterways, (p.22) railways, and electricity transmission is linked for energy supply and commodity transport. National and transnational elites involved in investments in these different sectors are also frequently linked.
(4.) See, for instance, the long list of Working Papers in the Effective States and Inclusive Development Research Consortium, here: http://www.effective-states.org/publications/, accessed 3 February 2018. These are almost entirely focused on periods of the last two to three decades.
(5.) While North et al. and Acemoglu and Robinson do not directly employ the concept of political settlement, their work is very much in line with political settlement thinking.
(6.) Of course, there have been periods of particularly nefarious rule, such as that of the Fujimori government in Peru and the governments of Banzer and Garcia-Mesa in Bolivia. Nonetheless, these still seem substantially different from the long and sustained commodity-funded machineries of centralized, repressive power that have characterized some other mining and hydrocarbon-dependent economies and societies.
(7.) Precise periods varied depending on the periodization of settlements in each country.
(8.) Because of potential risks to interviewees in some of the countries analysed in this study, and for consistency across the book as a whole, in general, we decided to cite information derived from interviews by referring only to the professional title and/or affiliation of the informant. The only exceptions are of visible public officials.
(9.) Of these different types of invitees, more researchers and civil society actors tended to accept invitations. That said, government and private sector representatives did also attend, and in one instance, after the workshop in Peru, one company representative asked for a follow-up meeting and gave us several more hours of his time and views.
(10.) These are noted in the Preface and Acknowledgements.