Resource Dependence, Economic Performance, …

1 OF CONFLICT RESOLUTIOND unning / Resource dependence , Economic PERFORMANCER esource dependence , EconomicPerformance, and political StabilityTHAD DUNNINGD epartment of political ScienceUniversity of California, BerkeleyIn many Resource -dependent states, elites may face an important trade-off between the Economic bene-fits of diversification and the possibility for future political competition that diversification may , distinctive features of global Resource markets and national political economies may make diversi-fication more or less attractive to political elites. The author argues that in three cases which illustrate theequilibrium paths of the game-theoretic model developed here postindependence Bostwana, Mobutu sZaire, and Suharto s Indonesia three variables influenced elites incentives for diversification and therebyshaped outcomes along the dimensions of political stability and economicperformance: the world marketstructure for the Resource , the degree of societal opposition to elites, and the prior development of thenonresource private sector.

2 These countries varied paths from resourcewealth to political and economicoutcomes suggest the need forconditionaltheories of the Resource : Resource curse; natural resources; political losers; Economic development; Botswana;Indonesia; ZaireTheories of the relationship between natural Resource wealth and political instabilityface important explanatory challenges. On one hand, recent research provides someaggregate evidence linking Resource rents to coups and the incidence and duration ofarmed the other hand, analysts are tasked with explaining the large varia-tion in outcomes among natural Resource exporters: one only need contrastthe politi-cal stability enjoyed by the house of Saud in Saudi Arabia with the historyofpostindependence Nigeria to have an idea of the difficulties faced by theories that givecausal priority to natural Resource S NOTE.

3 This article was presented at a conference on Natural Resources and Conflict con-vened by the Research Group in Human Rights and War, McGill University, Montreal, Canada, September13-14, 2003. I am grateful to all of the participants in the McGill conference, especially to James Ron,Michael Ross, and Richard Snyder for their detailed suggestions. I would also like to thank Chris Ansell,Jennifer Bussell, Robert Powell, James Robinson, three anonymous reviewers for their very helpful com-ments, as well as Reilly O Neal for her editorial assistance. The Office ofthe Chancellor at the University ofCalifornia, Berkeley, provided a research grant in support of a brief field research trip to OF CONFLICT RESOLUTION, Vol.

4 49 No. 4, August 2005 451-482 DOI: 2005 Sage Publications1. Collier and Hoeffler (inter alia, 1998, 2000); see Fearon (2005 [this issue]) for a discussion; andalso Humphreys (2005 [this issue]); Lujala, Gleditsch, and Gilmore (2005[this issue]); Collier, Hoeffler,and S derbom (2004); Le Billon (2001); and Ross (2003, 2004).This variation in outcomes along a range of dimensions is one reason, perhaps, thatresearch on the Resource curse2appears to have entered a phase in which analysts areincreasingly concerned with specifying the conditional impact of natural resourcewealth. In this spirit, Snyder (2003) and Snyder and Bhavnani (2005 [this issue]) notethat while lootable resources (that is, high-value natural resources with low eco-nomic barriers to entry) may provide the means and the motive for rebellionand thusengender political disorder, under other circumstances such wealth may contributeto the consolidation of political control and, perforce, political stability .

5 Ross (2001,2003) and Le Billon (2001) provide evidence that various types of natural resourceshave varied widely in their association with internal conflict, and Fearon (2005 [thisissue]) and Humphreys (2005 [this issue]) test various mechanisms that might linkresources to conflict. Analysts have also moved toward a more nuanced understandingof the relationship between natural resources and political institutions ( , Englebertand Ron 2004).This article attempts to contribute to such conditional theories of the Resource Snyder and Bhavnani (2005), I adopt a state-centered and revenue-centeredapproach, focusing on the incentives that Resource wealth may pose to incumbentpolitical elites.

6 I concentrate here, however, on exploring the political causes and con-sequences of Resource dependence . I argue that political elites in control of manyresource-dependent states face an important trade-off: while they mightlike to pro-mote the diversification of the economy, thereby reducing fiscal volatility and poten-tially improving aggregate Economic performance , diversification may create societalbases of power outside of the control of political elites. These independent bases ofpower may then facilitate future challenges to the political power of state incumbents,especially during the Economic downturns and fiscal crises that typically characterizeresource-reliant countries.

7 Thus, while diversification may be economically reward-ing, it can also be politically , distinctive features of global Resource markets and national politicaleconomies may make diversification of the economy more or less attractiveto postindependence Bostwana, Mobutu Sese Seko s Zaire, and Mohamed Suharto sIndonesia, cases that illustrate the equilibria of the game-theoretic model I developbelow, three key factors shaped the incentives of political elites to encourage diversifi-cation: the volatility of Resource revenues, the degree of societal opposition to incum-bent elites, and the prior development of the nonresource private Botswana,the unusual structure of the world diamond market and Botswana s uncommonrela-tionship to its chief multinational investor reduced the volatility of Resource revenues,significantly decreasing the Economic imperative for diversification relative to otherresource-dependent countries.

8 By contrast, in the more typical cases presented byMobutu s Zaire and Suharto s Indonesia, the volatility of Resource revenues did createimportant Economic incentives for reducing Resource dependence . However, in theselatter two cases, the degree of societal opposition to incumbent elites shaped the politi-452 JOURNAL OF CONFLICT RESOLUTION2. The term Resource curse has described both the tendency of Resource -rich countries to performworse on various Economic indicators than comparable Resource -poor countries (Sachs and Warner 1995;Karl 1997) and the apparent link between natural resources and other outcomes, including civil war and thepolitical regime incentives for diversification.

9 The high degree of societal opposition to Mobutu inZaire led him to believe that investments in infrastructure and other public goodswould pose a threat to his grip on political power. In Indonesia, on the other hand,Suharto was able to reduce the political risk of diversification by promoting the privateactivities of economically powerful but politically weak groups of ethnic Chineseentrepreneurs, allowing substantial diversification over the course ofhis tenure inpower. The prior development of the nonresource private sector also influenced theattractiveness of diversification in both cases: in Zaire, the potentialeconomic benefitsof investments in diversification were minimal, while they were substantial influencing the incentives of elites to promote the diversification ofthe eco-nomic structure, these factors shaped diverse outcomes along the dimensions of eco-nomic performance and political stability .

10 Botswana, with its unusuallystable flow ofdiamond income, experienced low fiscal volatility and good Economic Performance, notwithstanding substantial Resource reliance. Continued Resource dependence andde-diversification of the economy in Zaire led to poor Economic performance but bol-stered Mobutu s hold on political power, while Indonesia s diversification underSuharto brought Economic benefits but also brought increased risk of politicalcompetition. I develop this argument in more detail focus on the political causes and consequences of Resource dependence has sev-eral merits. First, although a large literature suggests that oil or mineral developmentcan cripple other sectors of an economy, leading to monoexport of the leading com-modity, the mechanism posited for this is generally macroeconomic: rising realexchange rates associated with Resource booms hurt other exports and drawproductiveresources away from these sectors ( , the Dutch Disease ).