WPS5742 Policy Research Working Paper 5742 The “Resource Curse� in MENA? Political Transitions, Resource Wealth, Economic Shocks, and Conflict Risk Michael Ross Kai Kaiser Nimah Mazaheri The World Bank Poverty Reduction and Economic Management Network Public Sector Governance Unit July 2011 Policy Research Working Paper 5742 Abstract The recent political upheavals in the Middle East and endowments, political stability, and conflict risk; and North Africa region have exposed growing concerns third, to assess how prospective political transitions about conflict risk, political stability, and reform have implications for the World Bank Group’s work in prospects across its societies. Given the prevalence of the region on public sector management and private oil and gas resource endowments in the region, which sector development. The authors find that resources and a voluminous literature suggests can be associated regimes have intersected to provide stability and limited with adverse development consequences, this paper violent conflict in the region, but that these development examines the interplay between their associated rents patterns have yielded a set of policy choices and and political economy trajectories. The contribution of development patterns that are proving increasingly brittle the paper is threefold: first, to examine the quantitative and unsustainable. A major institutional challenge for evidence of violent conflict in the region since 1960; reforms will be to consolidate a requisite degree of inter- second, to provide a nuanced review of the regional case temporal credibility and stability in these regimes, while study literature on the relationship between resource expanding inclusiveness in state-society relations. This paper is a product of the Public Sector Governance Unit, Poverty Reduction and Economic Management Network. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank. org. The author may be contacted at kkaiser@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team The “Resource Curse� in MENA? Political Transitions, Resource Wealth, Economic Shocks, and Conflict Risk Michael Ross Kai Kaiser Nimah Mazaheri* JEL Classifications: O57 - Comparative Studies of Countries, O53 - Asia including Middle East, P5 - Comparative Economic Systems, P48 - Political Economy; Legal Institutions; Property Rights; Natural Resources; Energy; Environment; Regional Studies * UCLA, World Bank, and Univ of Washington/WBI. This paper was commissioned as a background paper to the World Bank‘s MENA Social Development Regional Study Reducing Conflict Risk and Supporting Peaceful Transition. Special thanks goes to Jorge Araujo and Alan Gelb for comments on previous versions of this paper, and participants comments for an initial version presented at a MENA Region presentation and Economist Forum at World Bank in Washington, DC March and April 2011. Particular thanks goes to Lorena Viñuela for her significant role in helping shape the paper and comparative insights from Latin Americas democratic transition. All views and opinions expressed in this paper are those of the authors. All remaining errors are our own. i Table of Contents 1. Introduction ............................................................................................................................. 1 2. Analytical Framework ............................................................................................................. 3 Development Trajectories ........................................................................................................... 3 Resource Endowments and Dependence in MENA ................................................................... 4 Autocracy to Democracy?........................................................................................................... 5 Political Typologies and Prospective Transitions ....................................................................... 8 3. Violent Conflict and Oil .......................................................................................................... 1 4. Violent Conflict Trends in the MENA Region, 1960-2009 .................................................... 9 5. Oil, Political Stability, and Conflict: MENA Case Study Evidence ..................................... 12 6. Political Transitions and PSM and PSD Priorities in MENA ............................................... 15 Fiscal Management for Sustainability and Conflict Mitigation ................................................ 17 Promoting Economic Diversification and Private Sector Development................................... 22 Promoting Female Labor Force Participation ........................................................................... 24 7. Conclusions ........................................................................................................................... 25 References ..................................................................................................................................... 28 ii Tables, Figures and Boxes Table 1: Resource Endowments in MENA ..................................................................................... 5 Table 2. Political-Economy Typologies ....................................................................................... 10 Table 3: Institutional Characteristics of MENA ........................................................................... 12 Table 4. Civil Wars, 1960-2006 ...................................................................................................... 2 Table 5. Separatist Conflicts in Oil-Producing Regions ................................................................. 4 Table 6. Policy Management Domains, Prospective Instruments/Interventions, PE Drivers ...... 17 Table 7. Medium Term Fiscal Management, by Regions ............................................................. 19 Table 8. Medium Term Fiscal Management in MENA ................................................................ 20 Figure 1. Stylized Development Trajectories ................................................................................. 3 Figure 2. Democracy trends by region, 1960-2005 ........................................................................ 6 Figure 3. Income and Democracy in MENA and non-MENA countries, 2000-09 ........................ 7 Figure 4. Overall Intertemporal Credibility and Political Inclusiveness ...................................... 11 Figure 5. Annual Conflict Rates in Oil and Non-Oil Producers ..................................................... 2 Figure 6. Oil Income and Conflict rates, 1990-2006 ...................................................................... 3 Figure 7. Oil Prices and Conflicts in MENA, 1960-2009............................................................... 6 Figure 8. Oil and Female Labor Force Participation in the Middle East ........................................ 7 Figure 9. Income and Fertility Rates in Oil and Non-Oil Producers, 2003 .................................... 8 Figure 10. Conflict by Type, 1946-2009 ........................................................................................ 9 Figure 11. Conflict by Intensity, 1946-2009 ................................................................................. 10 Figure 12. Conflicts by Region, 1946-2009 ................................................................................. 11 Figure 13. Conflicts in MENA Region, 1960-2009 ...................................................................... 11 Figure 14. MENA Conflicts by Country, 1960-2009 ................................................................... 12 Figure 15. World Bank ―Doing Business‖ Rankings, Regional Averages in 2010 (By Region, 2005-2011) .................................................................................................................................... 23 Box 1: Lessons from Latin America‘s Authoritarian Transition? ............................................... 13 Box 2: A Revolution in Public Distribution Systems? ................................................................ 21 iii Acronyms and Abbreviations ACD Armed Conflict Dataset AFD French Development Agency ARE United Arab Emirates BHR Bahrain BP British Petroleum DJI Djibouti DZA Algeria EAP East Asia and the Pacific ECA Europe and Central Asia EE/FSU Eastern Europe/Former Soviet Union EGY Egypt GDP Gross Domestic Product GNI Gross National Income IDB Inter-American Development Bank IMF International Monetary Fund IRN Iran IRQ Iraq JOR Jordan KWT Kuwait LAC Latin America and the Caribbean LBN Lebanon LBY Libya MAR Morocco MENA Middle East and North Africa MTBF Medium-Term Budget Framework MTEF Medium-Term Expenditure Framework MTFF Medium-Term Fiscal Framework MTPF Medium-Term Policy Framework NRM Natural Resource Management OECD Organisation for Economic Co-operation and Development OGM Oil, Gas, and Minerals OMN Oman PE Political Economy PEA Political Economy Analysis PFM Public Financial Management PPA Public Policy Attributes PREM Poverty Reduction and Economic Management Network PRIO Peace Research Institute Oslo PRMPS PREM Public Sector Governance Anchor PSD Private Sector Development PSM Public Sector Management QAT Qatar SAU Saudi Arabia SME Small or Medium-Sized Enterprise SSA Sub-soil Assets SYR Syria TUN Tunisia UAE United Arab Emirates WDI World Development Indicators YEM Yemen iv ―A single spark can light a prairie fire‖ Mao Tse-tung, Chinese proverb 1. Introduction 1. The recent ―Arab Spring‖ political upheavals in the Middle East and North Africa (MENA) have brought heightened attention to the question of whether there are shared characteristics to the region that pose particular challenges for political transitions, development and poverty reduction. The purpose of this paper is to assess the extent to which dependence on resource rents, notably from oil and gas, interacts with political economy factors to produce violent conflict. We focus on documenting the particular empirical patterns for MENA concerning historical patterns of overt conflict, oil-dependence, and political regime typologies, and assess the implications this might have for international development partners such as the World Bank. 2. International observers have noted the prevalence of authoritarian and paternalistic regimes in the region, whose governments are often bolstered by natural resource rents from oil and gas (cf Benn Eifert et al. 2002). While these regimes may originally derive their legitimacy through military, religious, tribal, or other forms of authority, as resource rents have contributed more and more to the government‘s purse this has enhanced public living standards and further increased governmental legitimacy. This prevalence of ―rentier state‖ dynamics has been studied by a number of scholars of the MENA region, who note the shared features and common political economy trajectories among many countries. 3. A ―resource curse‖ literature also suggests that natural resource endowments will tend to produce a number of adverse economic outcomes, and a related literature argues that there will be adverse political outcomes. While the existence of an economic or political resource curse has increasingly been questioned in the literature, there is a broad consensus on the importance of the interactions between resource endowments and institutional quality in shaping outcomes (Barma et al. 2010). Equally, observed resource dependence needs to be appreciated as both an absolute and a relative measure (Brunnschweiler and Bulte 2008). Further, resource dependence is likely to be endogenous to a country‘s overall political and economic development trajectory. Careful attention therefore needs to be paid to how prevailing institutional arrangements interact with resource endowments, and how this might affect prospective reform trajectories. 4. From a global perspective, the MENA region presents a number of ―paradoxes.‖ While frequently perceived as a region of high conflict owing to a number of prominent conflicts (Israel-Palestine, Iran-Iraq), most countries in the region have actually been spared significant violence during the past two generations. Stability has co-existed with limited democracy, with resource rents frequently helping buttress prevailing state-society 1 interactions. Stability has also co-existed with an absence of reforms and an absence of sufficient economic dynamism, which has in turn bred a number of prevailing challenges in terms of economic opportunities and participation. Yet, the recent upheavals in the MENA region appear to not be driven by a clear set of economic shocks (e.g., food, fuel, or finance), but rather are the result of a contagion effect, accelerated by the growing prominence of new digital media. This being said, there are very real economic challenges that persist in many of these countries which have certainly fueled anti-government mobilization. 5. Our paper sets out to review potential frameworks for thinking about the linkages between resource wealth, violent conflict, and development prospects in the region, and presents salient quantitative and qualitative evidence. As the outset, we argue that it is critical to define notions of conflict in both positive and normative terms. Indeed, the prevalence of political contestation will in many instances be an integral part of development processes and prospects. Contestation is part of any healthy and evolving state-society relationship. However, prolonged violence and intense conflict over long periods of time will affect overall growth and poverty reduction prospects, in addition to the victims it directly creates. Further, prolonged conflict may itself undermine the functioning of a credible and legitimate state that can be critical in order for the government to pursue development over the long term. 6. The operational objectives of this paper are to discuss how the World Bank might more effectively project virtuous versus vicious development trajectories in resource-rich countries of the MENA region. Vicious development trajectories are cast particularly in terms of the risk of prolonged and violent conflict, whereas measures to enhance resilience against violent conflict are understood in conjunction with efforts toward broad-based, inclusive, and sustainable development. The 2011 World Development Report highlights some broad factors, combined with external and internal stresses, that make countries more susceptible to violent conflict. We seek to identify key aspects of public sector management (PSM), including its links to private sector development (PSD), that provide important and promising entry points set against the prevailing political economy. We cast this discussion of political economy not so much in terms of how to make reforms happen from a tactical perspective, but rather how particular PSM and PSD reform entry points should be seen to contribute to more virtuous rather than vicious developmental trajectories. 7. We proceed in seven sections. Section 2 provides a analytical framework that can be useful for thinking about development trajectories and political typologies in the MENA region. Section 3 provides an analysis of the relationship between violent conflict and oil using quantitative evidence and data on violent conflict incidence globally and in MENA countries. Section 4 provides a more in-depth overview of conflict trends in the MENA region. Section 5 provides a summary of the MENA regional literature on the relationship between resource rents, political instability, and conflict. Section 6 then identifies a series of public sector management (PSM) and private sector development (PSD) priorities that have typically been associated with policy challenges in resource dependent settings, and presents these through a lens designed to help understanding and help mitigate associated conflict risk in the MENA region and promote virtuous rather than vicious developmental trajectories. It also highlights how engagement around these priorities could be specifically catered towards resource-dependent contexts given potential shifts in political regime (using the framework set out in Section 2). Section 7 concludes. 2 2. Analytical Framework Development Trajectories 8. We begin by contextualizing the risk of violent conflict in a resource-dependent setting in terms of vicious versus virtuous development trajectories (see Figure 1).1 This stylized view suggests that countries may be on very different developmental trajectories. On the one hand, the ―prosperity‖ path emerges after the harnessing of resource rents to develop robust economies and effective states. Here, the risk of prolonged violent conflict is reduced. However, abundant resource rents may lead countries to ―muddle along‖ in economic and political terms, stable but failing to innovate or diversify. The lower trajectory is where economic decline in the form of the ―resource curse‖ is matched with violent conflict, especially as ―greed or grievance‖ begins to erode stability (Collier 2009). Of course, it should be recognized that economic decline can exist without violent conflict or conflict can exist in spite of prosperity. Figure 1. Stylized Development Trajectories Performance “Prosperity� and Low Levels of Violent Conflict Low-level equilibrium History / Time Economic “Resource Curse� and High Levels of Violent Conflict Source: Barma, Kaiser, Le and Vinuela (2010) 9. There has been growing emphasis on the extent to which oil, gas, and mining-rich countries pursue sustainable and effective policies for development and diversification, coupled with a recognition that political economy factors are central to understanding observed outcomes and reform prospects (Benn Eifert et al. 2002). Various forms of public sector distribution – energy and food subsidies, public sector employment, housing, sub- national allocations – often represent integral elements of the state-society compact in resource rich settings. The MENA region suggests that these types of programs can often be integral to stability, but over time also pose particular challenges for prospective political and 1 This section draws from the forthcoming book, Rents to Riches? The Political Economy of Natural Resource- led Group (Barma et al. 2010). 3 economic transitions and trajectories/scenarios. Cross-country predictions of civil war onset and their subsequent implications for societies are hard to predict. Resources and ethnic fissure, as well as low income levels, have been typically associated with higher conflict propensity (Blattman and Miguel 2008). There is also a growing agreement that institutions will significantly condition civil war onset and the duration of conflict, but this concept of ―institutions‖ needs to be disaggregated. The ability of ruling elites to credibly coordinate vis-à-vis themselves and society to remain in power will also affect the likelihood of violence, and the prospects for broader inclusiveness (North et al. 2009). Resource Endowments and Dependence in MENA 10. The terms resource-rich and resource-dependent are frequently used quite inter- changeably in popular discourse. A major point of the recent ―resource curse‖ literature has been to underscore that these terms need to be examined more specifically in both absolute and relative terms, as well as in terms of both flows and stocks (―proven reserves‖) (Brunnschweiler and Bulte 2008; Brunnschweiler and Bulte 2009). A country may be resource dependent because this is truly the only game in town, or because it generates a high level of income from resource production relative to other economic activities. While both Chile and the Democratic Republic of Congo are often described as resource rich, the former produced 90 times as much copper per capita in 2008 as the latter. In short, measures of resource dependence may be endogenous. It is not clear whether high export shares for oil are due to ―god given‖ resource endowments, or failure to develop alternatives.2 Table 1 presents selected measures of resource endowment for North Africa and the Middle East.3 While the usual suspects lead in terms of per capita oil income, notably only the Palestinian Territories, Lebanon, and Morocco could be described as truly resource scarce based on prevailing oil/extractive incomes.4 A major question is which phase of the resource cycle a country is in (e.g., early, mature, or ―sun setting‖). For example, Tunisia‘s oil production peaked in 1980, but it has recently increased gas production, and Israel is now in the process of developing significant off-shore gas deposits. 2 The recent literature also suggests that discovery of resource stocks and their exploitation will itself be contingent on a country‘s ―above ground‖ institutional quality (Gelb et al. 2011). 3 The literature uses a variety of quantitative thresholds for what constitutes resources dependence/richness. Our set of definitions focuses on a country exceeding 25 percent of exports, GDP, or government revenues coming for oil, gas, and/or mining (cf Barma et al. 2011). Other measures focus on the absolute value of income from oil, gas, and mining, normalized by population. By this measure, almost all MENA economies exceeded oil incomes of over USD 100 per capita in 2008 (while mining plays a more limited role in the region). 4 The Palestinian Territories and Lebanon are each in their own right dependent on particular international income transfers, which in some ways could be considered as rental dependence. 4 Table 1: Resource Endowments in MENA Economy & Population Resource Dependence Resource Endowments Per Capita Per Capita OGM Proven Sub-Soil PC Income Pop Resource Rent Rents/GDP Reserves Assets/Total (2008) (millions) (USD, 2008) Rank (%) (USD, 2005) (%, 2005) (i) (vii) (ii) (iii) (iv) (v) (vi) N. Africa Algeria* 4,974 34.4 2,112 10 42.5 13,293 43.9 Dijbuti 1,157 0.8 … 19 … … … Egypt** 1,997 81.5 450 12 22.5 1,989 9.3 Libya* 14,802 6.3 9,232 7 62.4 … … Morocco 2,769 31.6 3 18 0.1 71 0.2 Sudan* 1,404 41.3 371 15 26.4 1,554 12.8 Tunisia** 3,955 10.3 397 13 10.0 1,051 2.2 Other Bahrain* 28,240 0.8 11,408 5 40.4 82,923 41.1 Iran* 4,700 72.0 2,351 9 50.0 13,987 41.8 Iraq* 2,817 30.7 2,468 8 87.6 … … Israel 27,652 7.3 105 17 0.4 253 0.1 Jordan 3,910 5.8 287 16 7.3 74 0.1 Kuwait* 54,260 2.7 35,227 2 64.9 212,013 65.0 Lebanon 7,138 4.2 … 19 … … … Oman* 21,649 2.8 11,122 6 51.4 71,631 48.5 Palestinian Territory … - … 19 … … … Qatar* 86,436 1.3 43,485 1 50.3 … … Saudi Arabia* 19,152 24.8 13,445 4 70.2 86,620 59.3 Syria* 2,649 20.6 660 11 24.9 4,657 22.9 UAE* 58,272 4.5 21,717 3 37.3 118,111 33.8 Yemen* 1,175 22.9 375 14 31.9 … … MENA 17,455 406.7 8,623 na 37.8 43,445 27.2 Global Average 15,673 6,672.0 1,842 na 10.2 7,782 9 Source: World Development Indicators (WDI) (i, vii), World Bank Wealth of Nations (WoN) (ii, v, vii) Notes: * Net oil & gas exporters, ** Significant oil and gas exports. Global averages are unweighted. ~ UAE incorporates Abu Dhabi, Ajman, Dubai, Fujairah, Ras al-Khaimah, Sharjah and Umm al-Quwain. Autocracy to Democracy? 11. The MENA countries have long been distinguished by the absence of political inclusivity (Diamond 2010). Figure 2 displays global democracy trends by region since 1960; values on the x axis indicate the mean Polity scores of countries in the region. Polity is the most widely-used measure of democratic accountability in the field of political science. Notice that MENA has been either the most or second most undemocratic region in the world since 1960. Despite a slight democratizing trend since the late 1970s, the gap between MENA and all other world regions has grown larger. The lack of democracy, but more critically the extent to which MENA lags behind other regions, represent a special feature of the region in thinking about prospective transition dynamics moving forward. The way oil has interacted with political regimes in the past, and the way in which it may do so across particular country settings in the future, are important factors for the region meriting further scrutinity. While other regions have in the past experienced more of less rapid transitions to democracy, few would probably have predicted the on-set and speed of these transitions. 5 Figure 2. Democracy trends by region, 1960-2005 10 OECD Latin America 5 EE/FSU E/S Asia Africa 0 -5 MENA 1960 1970 1980 1990 2000 2010 year Source: calculated from Marshall and Jaggers (2007) 12. Figure 3 plots MENA country scores for democracy against the log of GDP per capita. While democracy scores, like many other governance indicators, typically show a positive relationship with income, MENA observations clearly underperform. Elsewhere, Gelb and Grassman (2010) has shown than MENA countries, and oil producers in general, underperform on the Kaufman-Kraay Accountability and Voice Indicators.5 Thus, we may not find the recent emergence of popular democratic movements across the region so surprising given that public opinion surveys demonstrate strong support in the MENA region for more democratic governance (Tessler and Jamal 2008). 5 Haber and Menaldo (2011), using a long-run panel of 162 countries from 1800-2006, contest the notion that natural resource reliance mechanistically leads to a higher propensity for authoritarianism. We concur with the premise that resource endowments do not mechanistically lead to particular development outcomes, but are contingent on the particular institutional preconditions and state-society make-ups of a given country‘s development trajectory. We do, however, contend that resources will condition the particular strategies and policies by which governments leverage these resources to rule, and in turn the implications these have for longer run development trajectories, including the depth of civil society and private sector development. 6 Figure 3. Income and Democracy in MENA and non-MENA countries, 2000-09 Source: Polity IV 13. The timing of anti-regime protests in undemocratic countries, however, is notoriously difficult to predict. One possible reason for this is because citizens living under repressive governments are more likely to conceal their true preferences until they perceive a moment of government weakness (Kuran 1989). This inability to predict anti-regime activity can be witnessed by a feature article entitled ―Next Year‘s Wars‖ in the December 28, 2010 edition of the journal Foreign Policy. Sixteen countries were listed as where the ―next crises…might erupt in 2011.‖ The only MENA countries on the list were Lebanon and Iraq. None of the government crises that occurred between January and March 2011 in the Arab world were predicted. 14. Other pro-democracy movements have emerged in other regions over the past three decades, spreading through a similar contagion process in Latin America and East Asia in the 1980s (see Box 1), Eastern Europe and the former Soviet Union in the late 1980s and early 1990s, Sub-Saharan Africa in the early 1990s, and the ‗color revolutions‘ in post-communist Eurasia between 1996 and 2005. 15. We expect that those countries with greater petroleum wealth are less likely to make successful transitions to democracy. Many studies suggest that oil-funded autocracies are unusually durable, especially outside of Latin America (Aslaksen 2010; Dunning 2008; M. Ross 2001). Since resource rents typically accrue to the central government, an executive who controls their deployment – even if that executive was democratically-elected – 7 commonly gains exceptional political influence, which they can use to entrench themselves in office. Oil-rich monarchs have been overthrown in the past – including in Libya and Iraq in the 1960s, and Iran in the 1970s – but these transitions have failed to produce democracies. This does not mean that democracy activists in Libya, Bahrain, and other oil exporters will fail, only that they face a more difficult challenge than their counterparts in Egypt and Tunisia. 16. The qualitative literature on democratic transitions suggests that countries with denser networks of civil society organizations are likely to have more successful, and more peaceful, transitions from authoritarian to democratic rule. Having a denser network of civil society groups – e.g., Latin America during the democratic transitions of the 1980s and 1990s; Indonesia in 1998-99 – seems to help resolve or deter conflict among opposition groups, and helps them build stable coalitions that can run the government after a dictator falls. 17. In the Middle East, the oil-rich countries in North Africa and the Persian Gulf tend to have weaker civil societies than the oil-poor ones, like Tunisia, Egypt, and Jordan. Oil wealth, in undemocratic states, seems to crowd out independent civil society: studies of Algeria (Moore 1976) and Libya (First 1980; Vandewalle 1998b), and Angola (Messiant 2001) have suggested that oil-funded rulers have deliberately used their patronage to prevent independent social groups from forming. One reason that Iraq fell into civil war after the departure of Saddam Hussein was that his regime had systematically crushed the independent civil society organizations that populated Iraqi society in the 1950s and 1960s. 18. The prevailing governance indicators in MENA suggest the need to unbundle governance dimensions in the region, with a view to better assessing potential political transitions and conflict potential. A key question is to provide a dynamic interpretation of the interplay between institutions characterizing state-society relationships, rents, and conflict potential. Political Typologies and Prospective Transitions 19. A variety of contributions have been made concerning political economy typologies in the MENA region and their relationship to observed policy outcomes in oil-dependent settings, including differentiating various types of autocracies and democracies.6 In thinking about the prevailing political economies, and prospective transition challenges of MENA resource-rent dependent countries in terms of development and diversification policies, we argue that it is useful to arrive at analytical typologies through dimensions that both capture the orientation and ability of political economies to pursue more ―ideal‖ development and diversification policies leveraging resource rents. The Rents to Riches? (―R2R‖) framework argues that political economy contexts and corresponding operational implications can be grouped according to two dimensions (Barma et al. 2010). 7 While aspects of discovery and 6 Eifet, Gelb, and Tallrott (2002) present a typology of five political economies (mature democracies, fractional democracies, paternalistic autocracies, reformist autocracies, predatory autocracies). 7 The Rents to Riches typology builds in particular on Evans (1989, 1995), Olson (1993), Lal and Myint (1996), Eifert, Gelb, and Tallroth (2002), and Kohli (2004). With respect to Eifert, Gelb, and Tallroth (2002), we argue 8 depletion are key upstream challenges for resource rich settings, we focus in this paper on prospective reforms around development and diversification policies, which in turn will have implications for longer term conflict resilience. 8 In order to help country counterparts and development practitioners diagnose the political economy trajectory a resource-dependent country embarks upon, the R2R framework advances two dimensions:  The credibility of intertemporal commitment—or the degree to which policy stability and bargains over time can be enforced and deviations from such agreements are subject to sanction; and  The overall political inclusiveness of the prevailing state-society compact—or the extent to which diverse social, economic, and political viewpoints are incorporated into decision-making, and a sense of either collectivist or clientelist welfare is privileged over purely elite interests. 20. Although these dimensions are interdependent to some extent, positioning them against each other yields a typology of four distinct country settings (Table 2), each with distinctive implications for natural resource rent generation and allocation. In sum, a country‘s positioning along the two key dimensions captured in the typology—the credibility of intertemporal commitment and degree of political inclusiveness—determines the manner in which stakeholder incentives and the institutional landscape interact with the structural characteristics of natural resources and hence how a country actually experiences the resource paradox. In non-inclusive settings where the intertemporal credibility of commitment is low, rent generation will be weak since the state will find it difficult to make beneficial extractive bargains with resource developers; and rent allocation will be biased toward consumption by political-economic elites and away from saving and investment for society. Factors that make intertemporal commitments more credible—by lengthening time horizons and strengthening institutionalization and the enforcement of property rights—will tend to improve a country‘s performance in terms of rent generation by enabling governments to strike better deals, at a lower risk premium, with developers. Factors that increase political inclusiveness— incorporating more political, social, and economic groups into decision-making—will make the state more accountable to society and orient rent allocation toward collective welfare through the provision of public goods and investment for sustainable development. that focusing on inter-temporal credibility, as opposed to a more static view of bureaucratic capacity, provides some advantages in particular when thinking about political transitions. 8 Viewed through the disaggregated lens of the NRM value chain, two key issues emerge in characterizing how a government manages its natural resources: (i) How effectively does a government generate and capture rents from the extractive industries?; and (ii) How does the government spend resource wealth and to what extent is it invested in a sustainable, pro-development manner? In essence, outcomes across the NRM value chain can be reduced to two core rent arenas: generating rents through extraction and taxation; and distributing rents through spending and investment. Many different domestic and international stakeholders are involved in natural resource policy-making and extraction, and the relationships among these actors are constantly shifting as one moves across the value chain. 9 Table 2. Political-Economy Typologies Credibility of intertemporal commitment Political inclusiveness Less credible / weaker enforcement More credible / stronger enforcement Patrimonial Rule Hegemonic Government Individualized political authority, built on a Institutionalized one-party regime; either Less inclusive / hierarchy of cronyism; emphasis on private predatory or benevolent; emphasis on less collectively oriented (elite) goods; exploitation of public resources private (elite) goods with some particularist for private gain and public goods Clientelist Pluralism Programmatic Pluralism More inclusive / Political competition based on extensive use Electoral competition based on programs more collectively of clientelism; provision of particularist geared toward collective welfare oriented goods; low horizontal accountability enhancement; provision of public goods; democratic accountability Source: Barma, Kaiser, Le and Vinuela (2011) 21. We draw on proxy indicators to broadly situate the MENA countries at the on-set of the Arab Spring by this typology in Figure 4. This raises questions about what kind of shifts one might want to look out for, and in turn what implications these will have for inter- temporal credibility and inclusiveness. Ideally political transitions would see movements over time to programmatic pluralistic democracies (i.e., to the North East quadrant), and the policies associated with them. But as democratic transitions from other regions demonstrate (see Box 1), major regional political transition episodes can yield a significant diversity of associated policies and institutional evolutions. These may also have implications for future violent conflict resilience, and well as PSM/PSD reform prospects and priorities. As with any governance indicators, these are subject to significant measurement error and issues of interpretation. However, we argue they provide a useful benchmark to assess potential forward looking trajectories. The highly oil endowed and institutionalized MENA countries of Bahrain, Oman, and Saudi Arabia are based squarely in the hegemonic quadrant. This in part is also a reflection of their monarchical institutions. By these measures, Kuwait is actually just placed under programmatic pluralism, suggesting that potential ―beacons‖ do exist in the region. The fragility of Yemen, particularly around inter-temporal credibility, places it in the patrimonial government setting, although it rates relatively highly on inclusiveness. 10 Figure 4. Overall Intertemporal Credibility and Political Inclusiveness Programmatic Pluralism Clientelistic Pluralism Patrimonial Rule Hegemonic Governments Source: Barma et. al. (2011) using IDB Public Policy Attributes (PPA) Policy Coordination (Stein et al. 2008) for inter-temporal credibility and AfD IPD 2009 Political Inclusiveness (Crombrugghe et al. 2009) (see Table 3) 22. A casual review of the political economy prevailing in MENA suggests a number of initial observations, many of them already echoed in the literature. The first is the prevalence of paternalistic authoritarian regimes in the region. These are taken to involve quite managed political space, often with strong elements of repression. This would position many of MENA‘s regimes in the hegemonic government and patrimonial rule quadrant. A key distinction between the former and the latter is whether authority is invested in an institution (which, paradoxically, can include monarchies), or in a single, often charismatic, personality. The recent political upheavals are certainly raising questions as to whether most of MENA‘s regimes fall closer to the potential patrimonial instability risks than hegemonic resilience. An important related question is whether greater pressures to popular inclusion will move regimes south to the clientelistic or programmatic pluralism quadrants, with all its implications for the political environment for public sector management and associated bureaucratic institutions. A key message of this paper is that such transitions may bring opportunities for reform, but some of the relatively functional and well performing parts of the state bureaucracy may also be at risk. Control and contestation over resources will remain an important dynamic of regime stability and conflict risks in the region. This may be most overt in polities such as Iraq, where the ultimate political settlement between the major political groupings will depend on access to its oil resources and associated benefit streams. 11 This is also true for settings like Egypt that depend on a mix of resources, remittances, canal fees, and aid rents. 23. Much of the MENA region continues to be characterized by a limited degree of transparency, for example as captured by indices of press freedom and budget transparency presented in Table 3. The region‘s regimes at the end of 2009 were also characterized by a high degree of continuity. In part of buoyant commodity prices, basic rates of growth have been respectable. These indicators have however served to mask a number of the strains and figures that are now increasingly coming to light as part of the ―social media‖ contagions in the region, including around accommodating the aspirations of a new generation driven by relatively high rates of population growth, education, but limited jobs opportunities in the region. As we show, resource rents are likely to continue to condition the political economy trajectories of the region. Greater transparency are likely to be an important ingredient for reforms in the region, but a deeper challenge will be how to balance pressures for greater inclusiveness with the ability to strike new credible political bargains across groups and time. Table 3: Institutional Characteristics of MENA Leadership & Institutional Quality Opacity Vulnerability Policy Political Dev. Pop Ch GDP GDP GDP Years in Capacity Coordination Inclusiveness Orientation OBI 2008 FH Media % p.a Grow Grow Grow Power (IPD) (1-4) (PPA) (0-4) (IPD) (1-4) (IPD) (1-4) (0-100) (1-100*) Unemp (%) 2000-9 2000-9 2005-9 2009 (i) (ii) (iii) (iv) (v) (vi) (vii) (xiii) (ix) (x) (xi) (xii) N. Africa Algeria* 11 2.3 1 0.7 2.3 1 39 13.1 1.5 3.6 2.9 2.1 Dijbuti 11 … … … … 2.0 3.6 4.8 5.0 Egypt** 29 2.3 1.6 1.3 2 43 39 9.4 1.9 4.9 6.0 4.7 Libya* 41 2.6 1.1 2.7 1.5 1 2.0 4.3 5.5 2.1 Morocco 11 2.6 1.7 1.7 2.7 27 39 9.6 1.2 4.8 4.8 5.0 Sudan* 21 2.6 0.5 1.7 2.2 17 2.1 7.1 7.8 4.5 Tunisia** 23 2.6 2.5 1.7 2.7 1 1.0 4.7 4.8 3.1 Other Bahrain* 11 3.6 2.2 1 2.7 28 2.2 6.3 7.3 Iran* 5 2 0.8 2.3 1.7 16 10.5 1.5 5.1 4.5 1.8 Iraq* 4 … 0.6 … … 29 2.6 0.7 4.1 4.2 Israel 4 4 2 3 3.5 72 7.3 2.0 3.6 4.2 0.8 Jordan 11 2.6 2 2 3 52 39 12.7 2.4 6.3 6.9 2.3 Kuwait* 4 2.4 1.6 2.7 3.7 40 2.8 7.0 6.7 Lebanon 2 2.6 1.6 2.3 2.2 32 40 1.3 4.7 5.5 9.0 Oman* 40 3 2.5 1.3 3.2 30 1.9 5.4 7.3 Palestinian Territory 5 … … … … 23.7 3.3 -2.0 6.3 Qatar* 15 2.9 1 3.2 39 8.7 13.5 17.4 8.6 Saudi Arabia* 5 1.6 1.1 0.7 3 1 21 6.0 2.3 3.4 3.0 0.2 Syria* 10 2.3 1 1 1.5 16 2.7 4.2 4.6 4.0 UAE* 6 3.6 2.3 1.7 3.3 35 4.1 5.8 5.5 -0.7 Yemen* 32 1.9 1.3 2 1.7 19 2.9 4.0 3.9 3.8 MENA 14.3 2.6 1.5 1.7 2.6 26.0 29.5 11.5 2.5 4.8 5.9 3.6 Global Average 7.4 2.5 1.6 2.2 2.8 39.7 50.9 8.0 1.5 4.2 4.5 0.1 Source: WB Database of Political Institutions (2010) (i), AfD Instutional Profiles Database (ii, iv, v), Insititute (vi), Freedom House (vii), WDI (viii) IaDB Public Policy Attributes (PPA) Database (iii), Open Bugdet Indicators (vi) Notes: Unemployment for 2007-2008, as available. Freedom House Scores presented as 1-100, with 100 free (based on original 1-4 scale) * Net oil & gas exporters, ** Significant oil and gas exports. Global averages are unweighted. 12 Box 1: Lessons from Latin America‘s Authoritarian Transition? After a prolonged period of authoritarian rule, democratization in Latin America or, redemocratization in some cases, took place in a relatively short period of time. From 1979 to 1986 seven countries democratized in South America, five in Central America and six more in the Caribbean. After democratization in Argentina, Brazil and Uruguay quickly transitioned to electoral governments. Brazilian and Uruguayan demands for direct elections were closely tied to the events in Argentina. Later, the three countries pressed together for democratization in Paraguay and Chile. And, although the 1989 coup in which Stroessner was overthrown by his son-in-law General Andrés Rodríguez was the product of infighting, the new president soon faced considerable international pressure to call for elections (Zagorski, 2003). In 1990, Chile, Panama and Paraguay transitioned to democratically elected regimes. El Salvador and Guatemala became fully democratic after the settlement of their civil conflicts in 1992 and 1996 respectively. This pattern of accelerated regime change has been described as a domino effect or a diffusion process. Przeworski, Limongi et al. (1996) explain that economic factors are not the only ones that matter for the durability of democracy and that the larger the proportion of democracies in a region, the more likely democracy is to survive in a country that is part of it. They highlight that contagion operates independently of the direct influence of Western governments and international institutions. Similarly, Brinks and Coppedge (2006), using worldwide data on transitions to democracy from 1972 to 1996, show that region matters to predict the likelihood of democratization in a given country. In their work, diffusion is a determinant of the magnitude and direction of regime change since countries tend to follow the direction in which the majority of countries are moving and change to match the average degree of democracy of the contiguous countries, or ‗neighbor emulation.‘ The size of the gap in the level of democracy between a country and its neighbors explains the direction and strength of the pressure that exists for convergence. Thus, democratic neighbors have an impact on authoritarian regimes and this impact spreads across authoritarian neighborhoods until they all stabilize at an average level of democracy. Large, populous and comparatively more prosperous countries have the largest impact on neighboring regimes. Some of the mechanisms used by neighbors to pressure undemocratic regimes include domestic actors in one country urging their governments to demand better human rights protections in the other, military interventions, funding insurgents, supporting a coup, and the promotion of regional organizations interventions in the adjacent country. Furthermore, Pevehouse (2002) and Briks and Coppedge (2006) indicate that regimes that belong to common networks or regional organizations are able to exert more force on each other. In the same vein, Smith (2005) underscores that the process of diffusion is possible because elites are highly aware of regional phenomena. This allows opposition groups to draw moral and material support from the democratization of a neighboring country, expanding in this way their support base. As well, the first transitions in a region can demonstrate to elites that democracy does not necessarily lead to changes in the economic policies, or the reduction of national security. In Latin America, military regimes and elites were very conscious of the fate of their regional counterparts and drew lessons from the events in nearby countries to protect themselves once the transitions began in their own countries. In addition, the trials faced by the Argentine Junta leaders in 1984 was a particularly threatening development for neighboring military dictatorships, which meant that the costs of prolonged authoritarianism for them soon surpassed the costs of democratization. Source: Viñuela (2008) 13 3. Violent Conflict and Oil 24. A large number of quantitative studies have examined the role of mineral wealth in general, and petroleum wealth in particular, on the onset, duration, and intensity of violent conflict.9 A country‘s petroleum wealth – measured as the value of oil and gas produced per capita – is strongly correlated with the likelihood that a conflict will begin. The relationship between oil and conflict risk is non-linear, as Collier and Hoeffler (2004) have suggested: while modest sums of petroleum wealth can increase a country‘s conflict risk, exceptional oil wealth – enough to raise per capita incomes above about $5000 – makes countries sufficiently rich that their conflict risk begins to drop. To put it differently, oil wealth only raises the risk of conflict in countries with relatively low incomes; if they are rich enough – even if they became rich through their oil – insurgencies become less likely.10 Fjelde (2009) notes that political corruption may not necessarily be associated with higher conflict risks in oil-rich states, because this is inherent to the stability of the regimes themselves. But political corruption in turn generates particular symptoms and outcomes that pose challenges for more ―virtuous transitions.‖ 25. To describe conflict trends globally and in the MENA region (Section 4), we use the Armed Conflict Dataset (ACD), originally produced by Gleditsch et al. (2002) and updated on the website of the Peace Research Institute of Oslo; it is the most transparent and complete list of armed conflicts around the world.11 It records both ―minor‖ conflicts (which cause from 25 to 1000 battle-related deaths in a calendar year) and ―major‖ conflicts (which cause more than 1000 battle-related deaths). Table 4 displays some simple cross-tabulations, showing the annual conflict rate among countries in different categories.12 Figure 5 shows the same results in a bar chart. Among countries at all income levels since 1960, the conflict rate in the oil states has been about 40 percent higher; since the Cold War ended in 1992, it has been almost 50 percent higher; among lower-income countries since 1960, it has been roughly 75 percent higher; and among lower-income countries since 1992, it has been more than twice as high.13 9 For surveys see Ross (2004b; 2006); McNeish (2010). 10 Basedau and Lay (2009) come to similar conclusion but through a different logic. 11 The most recent version (v4 2009) can be downloaded at: http://www.prio.no/CSCW/Datasets/Armed- Conflict. For conflicts in 2009, we use Harbom and Wallensteen (2010). 12 We define ―oil producers‖ as states that generate at least $100 per capita in oil and gas wealth (using constant 2000 dollars) in a given year. 13 In defining ―low-income countries‖ as those below the $5000 per capita threshold, we also excluded states like Libya, Kuwait and Saudi Arabia, whose per capita income from oil alone exceeds $5000. 1 Table 4. Civil Wars, 1960-2006 These figures show the percentage of countries that had a new civil war in a given year. Non oil Oil Difference producers producers Overall: All States and Periods 2.8 3.9 1.0** By Income: Low Income (below $5000) 3.8 6.8 3.0*** High Income (above $5000) 1.2 1.4 0.2 By Period: 1960-1989 2.4 2.7 0.3 1990-2006 3.6 5.3 1.7** *** significant at 1%, ** significant at 5%, * significant at 10%, in a one-tailed t-test. Source: calculated from data in Gleditsch et al. (2002). Notes: Oil producers are here defined as countries with over USD 100 dollars per capita in oil income for 2008 (using 2000 general prices). Data are drawn from US Energy Information Administration (EIA). Price data from BP statistical yearbooks. Figure 5. Annual Conflict Rates in Oil and Non-Oil Producers 10 9.1 9 8 7 6.8 6 5.3 5 4.2 3.9 3.8 4 3.6 3 2.8 2 1 0 All Post Cold War Low-income Post-Cold War & Low- Income Non-oil states Oil states Source: calculated from conflict data in Gleditsch et al. (2002). Notes: The figures represent the annual conflict rate for oil-producing states (black bars) and non-oil states (gray bars). The data are for 1960-2006, except the ‗Post-Cold War‘ period, which is 1992-2006. Low income countries are those with incomes below $5000 per capita. All of the differences between oil and non-oil states are statistically significant in t-tests. 2 26. Figure 6 shows the non-linear relationship between oil income per capita (on the horizontal axis) and the annual conflict rate from 1990 to 2006 (on the vertical axis). As a country‘s oil income rises from zero to about $1000 – Russia‘s average level during this period – its conflict rate increases. Yet above this level, it drops back towards zero. Strikingly, Libya‘s oil income is well above the previous cut-off point for conflict in oil producers; its civil war is an exceptionally rare event. Figure 6. Oil Income and Conflict rates, 1990-2006 Source: calculated from Gleditsch et al. (2002), and Ross oil dataset. 27. There are many arguments about why oil wealth makes low-income countries more conflict-prone. The evidence is relatively strong for two arguments; two other arguments are more speculative but are worth considering for their potential significance in the MENA region. 28. The first argument is that when oil is extracted from a region populated by an ethnic or religious minority, it increases the benefits of secession and hence the likelihood of a secessionist rebellion. Table 5 lists 16 separatist conflicts that broke out in petroleum-rich territories between 1960 and 2006.14 These are the separatist conflicts that broke out between 14 Eight of these conflicts began in ‗oil states,‘ meaning they produced at least $100 per capita in oil and gas income in the conflict‘s first year. In six of these eight cases, petroleum extraction was already under way in the secessionist region (Angola, Iran-Kurdistan, Iran-Arabistan, Iraq, Nigeria-Niger Delta, and Russia); in two cases, extraction was occurring in other parts of the country, and about to begin in the region where the rebellion took 3 1960 and 2010, in which armed groups in an oil-producing region fought for independence. Country Income is for the year the conflict began, or the closest year for which data are available. Figures are in constant 2000 dollars per capita. Notice that the danger of resource- fueled separatism is far greater for poor countries than rich ones: all 16 conflicts broke out in countries with incomes below $2100 per capita; 11 of the 16 occurred in countries with less than $1000 per capita. Countries with incomes above $2100 per capita – about the level of Jordan or El Salvador – seem to face a much smaller danger of petroleum-related separatist wars. Table 5. Separatist Conflicts in Oil-Producing Regions Country Conflict Years Country Income Region Angola 1975-2007 $1073 Cabinda Bangladesh 1974-92 $243 Chittagong Hill Tracts China 1991- $422 Xinjiang India 1990- $317 Assam Indonesia 1975-2005 $303 Aceh Iran 1966- $1053 Kurdistan Iran 1979-80 $1747 Arabistan Iraq 1961- $2961 Kurdistan Nigeria 1967-1970 $267 Biafra Nigeria 2004- $438 Niger Delta Pakistan 1971 $275 Bangladesh Pakistan 1974-77 $280 Baluchistan Russia 1999-2001 $1613 Chechnya Sudan 1983-2005 $293 South Turkey 1984- $2091 Kurdistan Yemen 1994 $443 South Source: conflict data are drawn from Gleditsch et al. (2002); income data are from World Bank (various), with missing figures taken from Maddison (2009). place (Indonesia, Yemen). The other eight conflicts broke out in petroleum-rich regions of countries that did not cross the $100 per capita threshold – either because oil had been discovered but not yet extracted (in Pakistan- Bangladesh, Bangladesh-Chittagong Hills, Nigeria-Biafra, and Sudan), or because oil was relatively scarce at the national level, even though it was abundant in the secessionist region (in China, India, Pakistan-Baluchistan, and Turkey). Interestingly, none of these conflicts occurred in Latin America. In Latin America, oil is statistically associated with governmental conflicts but not separatist ones. This is not because Latin American petroleum has unusual properties, but because the region is uniquely secession-proof: there have been no separatist conflicts in Latin America for over a century. For more on the Latin American anomaly see Ross (2010). Other types of mineral wealth, including copper and gold, have sometimes been linked to secessionist movements in exceptionally poor regions – for example, on the Papua New Guinea island of Bougainville, and in Indonesia‘s West Papua province (M. Ross 2004a). 4 29. It is also noteworthy that 15 of the 16 conflicts broke out in regions traditionally populated by ethnic or religious minorities.15 This is consistent with the idea that oil wealth is only likely to encourage secessionist rebellions when it is found in regions with pre-existing ethnic or religious grievances. 30. The second mechanism that ties oil wealth to violent conflict is looting: petroleum facilities, with their high sunk costs and sometimes-remote locations, are ideal targets for looting and extortion by criminal gangs or political movements intent on raising money. One innovative study of 900 Colombian municipalities between 1988 and 2005, by Oeindrila Dube and Juan Vargas (2009), found that oil-producing municipalities were more frequently subject to paramilitary violence, especially when oil prices rose. During the civil violence in Iraq between 2004 and about 2007, opposition forces appeared to raise large sums from black market oil. According to one estimate, in 2006 insurgents earned between $25 million and $100 million from oil smuggling and related activities.16 31. Both of these mechanisms are consistent with a bounty of new research that finds a strong correlation within countries between the location of oil wealth and the location of armed conflict: when oil and gas are found in a conflict zone, conflicts are more likely to flare up, especially if the region is relatively poor17; casualties tend to be higher18; and the fighting tends to last longer.19 32. A third possible mechanism – for which evidence is admittedly more fragile – is that civil conflicts can be triggered by macroeconomic shocks; since oil prices are unusually volatile, this would help account for their higher conflict risk.20 The most convincing study of conflict and economic shocks is Miguel, Satyanath, and Sergenti (2004), which used variations in rainfall to identify the relationship between economic growth and conflict in 41 African states between 1981 and 1999: it suggested a five percentage point drop in GDP was associated with a 50 percent rise in the likelihood of conflict the following year. Still, extrapolating these findings to other settings may be hazardous (Blattman and Miguel 2008). 33. One hint of a possible link between oil shocks and violent conflict in the Middle East comes from comparing the region‘s conflict trend with trends in oil prices (Figure 7). The subsequent section will show that MENA‘s conflict trend closely mirrored the global trend, except for the regional jump in violence from 1978 to about 1984. This period also coincided with a pair of large price shocks – one positive (1978-79), the other negative (1979-1985). 15 The only exception was the 1994 civil war in Yemen, when the Yemen Socialist Party fought to re-establish independence for South Yemen, which had become part of the unified Republic of Yemen in 1990. The territory claimed by the separatists included recently-discovered petroleum reserves in the Hadramout region. 16 John F. Burns and Kirk Semple, ―US Finds Iraq Insurgency Has Funds to Sustain Itself,‖ New York Times, November 26, 2006. 17 Dube and Vargas (2009); Østby, Nordås, and Rød (2009). 18 Lujala (2009). 19 Buhaug, Gates, and Lujala (2010); Lujala (2010). 20 Humphreys (2005) finds evidence to support this mechanism; Ross (2006; 2012) does not. 5 Figure 7. Oil Prices and Conflicts in MENA, 1960-2009 Source: oil prices are from BP (2010); conflict data are from Gleditsch et al. (2002). 34. To some extent, causality runs from violence to price shock, since the Iranian Revolution had a sharp impact on world oil prices. But the overthrow of the Shah led to a collapse in Iran‘s oil production, which fell from 212 million metric tons in 1978 to 72 million metric tons in 1980, and stayed low for the rest of the decade (figures are from the World Bank‘s Adjusted Savings dataset). This implies that the Iranian Revolution did not cause the 1979-85 negative oil shock – and hence that the shock itself (probably caused by a combination of reduced demand in the West and increased Saudi production) may have helped trigger the spike in MENA conflicts. 21 35. The final mechanism is that oil wealth may help explain the region‘s high fertility rates, and high fertility rates could be producing a large ―youth bulge‖ in the population that raises the risk of conflict. Although there is no consensus among scholars, some studies find that populations with unusually large youth cohorts are more likely to experience violent conflict.22 Ross (2008, 2012) and Jamal et al. (2010) argue that in the Middle East, oil wealth 21 If this spike was caused by the negative oil shock, it may have reflected disruptions in the region’s economy rather than just the economies of the oil producers: several of the new conflicts were in non-oil producing states – like North Yemen and Morocco – whose economies may have been hurt by the reduced demand for migrant labor in neighboring oil producers. 22 See, for example, Urdal (2006), Yousef (2003). 6 tends to crowd women out of the labor market, since the ―Dutch Disease‖ effect may make unprofitable industries that typically employ women in low and middle income economies (e.g., export-oriented low-wage manufacturing), and because large government transfers tend to discourage women from seeking work to provide their households with a second income. 36. Figure 8 shows the correlation between oil wealth and female labor force participation in the Middle East region. At a global level, Figure 9 compares the fertility rates of oil producers (solid dots) to non-oil producers (hollow dots): while more wealth seems to lower fertility rates in both sets of countries, oil wealth seems to have a smaller impact. By reducing female labor force participation, oil wealth seems to indirectly boost fertility rates – which may in turn lead to a heightened conflict risk. Figure 8. Oil and Female Labor Force Participation in the Middle East Source: calculated from data collected by the International Labor Organization and published in World Bank (2004, 2005); Ross oil dataset. 7 Figure 9. Income and Fertility Rates in Oil and Non-Oil Producers, 2003 Source: World Bank (various). The solid dots represent oil producers (oil income/capita>$100), while the hollow dots are non-oil producers. 37. We can summarize the arguments above as follows: oil wealth tends to heighten a country‘s conflict risk, especially when incomes are relatively low. The danger of a secessionist rebellion is higher when petroleum is extracted from regions dominated by marginalized ethnic or religious minorities. Petroleum might also lead to conflict by providing rebels with targets for extortion or looting; by making countries more susceptible to macroeconomic shocks; and possibly by increasing fertility rates and hence generating large youth cohorts. 38. Each of the four causal mechanisms points to a different set of conflict-mitigating policy interventions. If we are worried about petroleum-induced secession – a problem in both Iraq and Iran, and a problem in Yemen in 1994 – suggested policies might include increased distribution of rents to the producing region; encouraging local economic spillovers to generate more jobs and higher wages in the region; and reserving a significant fraction of these positions for local residents. Similar policies, plus additional community-building and security measures, could reduce the risk of extortion and looting. 39. Finally, it should be noted that other studies center on alternative factors that may be associated with the outbreak of conflict in the MENA region (e.g., Lust-Okar and Sambanis 2007). Often they underscore the significance of factors like the persistence of, and spillover effects from, the Israel-Palestine conflict; the existence of state boundaries that may not reflect important ethnic or religious divisions (for example, in the Kurdish region); and interventions by foreign powers. Although all of these factors are important, they also are difficult to address with the World Bank‘s policy toolkit. 8 4. Violent Conflict Trends in the MENA Region, 1960-2009 40. The recent political upheavals in the MENA region have naturally increased interest in better understanding the extent, and apparent drivers, or violent conflict in the region. We present an analysis of the particular observed conflict patterns observed in the MENA region, elaborating on the foregoing international empirical literature on conflict drivers, as well as the particularities of the region‘s resource endowments/dependence and political economy regimes. 41. To identify the patterns that are distinctive to the MENA region, it is useful to compare trends in the region to global trends. According to the ACD, the total number of armed conflicts – including both intrastate and interstate violence – rose steadily from 1946 to about 1992, then declined sharply until about 2003. Since 2003, there has been a steady rise in the number of conflicts. Figure 10 displays this pattern. In recent decades both ―extrasystemic‖ (anti-colonial) conflicts and ―interstate‖ conflicts have almost disappeared. Almost all of today‘s wars are either purely ―intrastate‖ wars between a government and a rebel organization, or ―internationalized‖ civil wars, which are intrastate wars in which foreign powers intervene on one or both sides. Figure 10. Conflict by Type, 1946-2009 Source: PRIO web page, http://www.prio.no/CSCW/Datasets/Armed-Conflict/UCDP-PRIO/Armed- Conflicts-Version-X-2009/ 42. Another important trend is the relatively larger decline in major conflicts, displayed in Figure 11. The number of ongoing large conflicts fell from 16 in 1990 to 6 in 2009, while the number of ongoing small conflicts fell from 34 to 30 (Harbom and Wallensteen 2010). Smaller conflicts increasingly dominate the landscape. 9 Figure 11. Conflict by Intensity, 1946-2009 Source: PRIO web page, http://www.prio.no/CSCW/Datasets/Armed-Conflict/UCDP-PRIO/Armed- Conflicts-Version-X-2009/ 43. Figure 12 displays the number of armed conflicts by region; Figure 13 shows only the Middle East conflicts, counting wars of all types. Note that the trend in the MENA region is broadly similar to the global trend, but with one interesting difference: while the number of global conflicts rose steadily through the early 1990s, in the Middle East there was a sharp increase in the number of conflicts between 1978 and about 1985. Without this spike, the regional trend would almost perfectly mirror the global trend. We return to the possible significance of this 1978-1985 period in Section 5 below, when looking more closely at relationship between oil shocks and MENA conflicts. 10 Figure 12. Conflicts by Region, 1946-2009 Source: PRIO web page, http://www.prio.no/CSCW/Datasets/Armed-Conflict/UCDP-PRIO/Armed- Conflicts-Version-X-2009/ Figure 13. Conflicts in MENA Region, 1960-2009 12 10 8 6 4 2 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Source: Calculated from data in Gleditsch et al. (2002) and Harbom and Wallensteen (2010) 11 44. Within the Middle East, most of the conflicts have occurred in a handful of countries. One simple way to make comparisons is by adding up the number of years that each country had an ongoing conflict; if a country had more than one conflict in a given year, the additional conflict is added to the total. Figure 14 shows the distribution of conflict-years among MENA countries from 1960 to 2009. About two-thirds (65 percent) occurred in Israel (including the Palestinian territories), Iran, or Iraq. Algeria, Yemen, and Morocco accounted for another 21 percent of conflict-years.23 Most of the region‘s twenty-odd countries had little or no conflicts at all. Figure 14. MENA Conflicts by Country, 1960-2009 Tunisia Saudi Arabia Oman Syria Djibouti Egypt Lebanon Morocco N Yemen/Yemen Algeria Iran Iraq Israel 0 10 20 30 40 50 60 Source: Calculated from data in Gleditsch et al. (2002) and Harbom and Wallensteen (2010) 45. This underscores the surprising and distinctive quality of recent conflicts in Libya, Syria, Tunisia, Egypt, and Bahrain – and to a lesser extent, protests in Jordan, Saudi Arabia, Oman, and Kuwait. Over the last fifty years, these societies have been both undemocratic and had low levels of violent conflict. In all cases except Libya, the recent upheaval might not be classified by political scientists as ‗civil wars,‘ but government violence against peaceful protest movements. Only in Libya, where the military has fractured, and anti-regime forces have acquired a rudimentary political organization and military capacity, are we witnessing an unambiguous civil war. We return to a discussion of recent events, and their economic implications, in Section 7 below. 5. Oil, Political Stability, and Conflict: MENA Case Study Evidence 46. Can the prevalence of violent conflict in MENA countries be explained by the fact that many of them are significantly dependent on natural resource revenues? At first blush, when we consider the individual histories of natural resource producers in the MENA region we do 23 For the purposes of this analysis, we include the conflicts in North Yemen in the total for the Republic of Yemen. 12 not immediately observe a clear positive or negative effect of resource dependence on incidence of conflict or political stability. On the one hand, we can point to a highly resilient petrostate such as Saudi Arabia which has experienced enduring political stability and relatively low levels of domestic conflict for nearly 80 years. Others, such as Oman and Kuwait, appear to more or less fall into this category as well. On the other hand, we can point to the collapse of the Pahlavi monarchy in Iran in 1979. The governments of Libya and Iraq similarly fell at a time when their oil revenues were increasing (Herb 1999). 47. Writings in the rentier state literature theorize that significant oil wealth should result in less state-society conflict and enhanced political stability. This well-treaded argument emerged in the work of Mahdavy (1970) through his observations of pre-1979 Iran. The logic here is that access to an alternative source of revenue means that the government does not need to tax the population and instead redistributes the oil wealth through rents and subsidies. Thus, the state ―buys off‖ the population through redistribution and eliminates the potential for a ―no taxation without representation‖ conflict, yielding an autonomous state alongside a contented population that (assumedly) does not aspire to challenge its rule. 48. Throughout history, a number of MENA countries have fit this description well. One obvious case is Libya where Qaddhafi utilized oil revenues to strengthen an autonomous state and, until quite recently, successfully demobilized the population (Vandewalle 1998a). We observe the same ―buying off‖ of the population in a number of Gulf states, such as Oman where oil wealth was directly used for building hospitals, schools, and an infrastructure that created ―a cohesive state out of a fragmented tribal society‖ (Range cited in O‘Reilly 1998). Saudi Arabia (see Al-Rasheed 2002) and Iran (see Karshenas 1990) fit the bill as well. In Iraq under Saddam, however, gradual improvements in living standards were undercut by decades of conflict and sanctions. 49. An abundant source of nontax revenue may also facilitate the build-up of a strong military and internal security apparatus that represses the population, which is certainly what we observe in many of the countries listed above. Hence, it may not be the act of ―buying off‖ the population but the application of oil revenues toward repressive activities that enhances political stability and minimizes conflict in these countries. This argument that oil pays for the installation of a security state can be found in the work of Bellin (2004). The governments of Iraq and Iran throughout the 1960s and 1970s also tried to ―force‖ stability and whereas the former was quite successful in this endeavor the latter was not. Yet, Henry (2004) points out that although excessive military spending was indeed the norm in these two countries, a closer look at data reveals that other oil producers such as Algeria invested far less, in spite of the military‘s role in the politics of Algeria since independence. 50. It has been theorized that conflict may be reduced in oil dependent countries due to the rent-seeking and patronage that occurs between the government and elites. In pre-1979 Iran, Karshenas (1990) observes an increasingly autonomous state as the onset of oil wealth and the consequent rents effectively concentrated power among a small group of elites. Shambayati (1994, p. 309) writes that since business classes are dependent on government rents in oil states, this ―minimizes the potential for class conflict‖. The experiences of Saudi Arabia may provide a counter-argument. As Chaudhry (1997) illustrates, an increasingly autonomous group of private sector elites clashed with the state over matters of policymaking following an 13 economic crisis – and, perhaps surprisingly, prevailed over the state in the conflict. Shambayati also theorizes that even though significant oil wealth may eliminate pressure from one group – business elites – it can ―[lead] to the emergence of culturally and ideologically based groups such as Islamist movements, for whom economic issues are of secondary importance‖ (Shambayati 1994, p. 307). 51. Another theory posits that political stability is the outcome of the protection that some oil-producing countries get from other countries who value the role of oil in the global economy (see Basedau and Lay 2009). Bellin agrees with this, writing that ―Authoritarian regimes in Saudi Arabia, Egypt, Jordan, Tunisia, and Algeria have received western support, at times in very generous proportions, because of the belief (perhaps mistaken) among western policymakers that these regimes would be most likely to deliver on western security concerns by assuring regular oil and gas supplies to the West and containing the Islamist threat. …The West‘s generous provision of this support has bolstered the capacity and will of these regimes to hold on‖ (2004, pp. 148-9). However, this theory hardly squares with recent events in Libya, Egypt, and Bahrain. 52. Other studies of oil producers in the MENA region provide a more nuanced alternative to the arguments laid out above. For instance, Smith (2006) points to the timing of economic development strategies and the kinds of coalitions that were forged during development as a key intervening variable in the chain of causation from oil wealth to political stability. Although the empirical evidence he provides is not entirely convincing, Smith argues that two distinct trajectories emerge in oil dependent states: durable authoritarianism and vulnerable authoritarianism. Crystal‘s (1989) study of Kuwait and Qatar places similar emphasis on understanding the details of coalition-building in examining the causes of political instability and vulnerability in the Gulf monarchies. 53. Lowi breaks further from these arguments and writes that the causes of governmental breakdown in authoritarian petrostates precedes the onset of oil and derives ―from problems related to the construction of the nation; and second…the decisions of political leaders‖ (Lowi 2004, p. 83). Henry equally questions the causal effect of oil wealth on political stability in Algeria, writing that ―it was not expectations of oil revenues that undermined Algeria‘s institutional capacity. Algeria‘s institutional weaknesses were born in 1962, before the oil from Hassi Messaoud started to flow in significant quantities‖ (Lowi 2004, p. 75). 54. Will an economic crisis in a country dependent on oil revenues lead to conflict? Indeed, it is tempting to argue that tying the government‘s budget to the fate of a single commodity whose international price is highly volatile can raise the risk of political instability. Luciani (1994), for one, argues that a sharp decline in oil revenues can produce a fiscal crisis that precipitates the fall of authoritarian regimes. By contrast, Smith and Bueno de Mesquita (2010) offer evidence that oil wealth has helped soften economic crises in authoritarian states, and explains why they so rarely have made democratic transitions. 55. Lowi (2004) cites the case of Algeria as an example of an economic crisis leading to a government fiscal crisis that fuels conflict. Also writing on Algeria, Henry argues that ―The menu of state services diminished with the decline in oil revenues while the population expanded and became more educated and demanding, leading to civil instability‖ (2004, p. 14 74). The fall of Suharto in Indonesia occurred following the financial crisis of the late 1990s. Finally, there is reason to suppose that not merely fluctuations in the international oil market but poor policymaking can worsen or even provoke an economic crisis. To the extent that this is the case, we may cite oil dependent countries as being more prone to policymaking that Karl describes as ―considerably arbitrary, irrational, and volatile … the antithesis of the environment necessary to confront a boom successfully‖ (see also Dillman 2000; Karl 1997, p. 190). 56. A link between oil dependence and conflict might operate through other variables such as income inequality, rural-urban migration, or unemployment. In the case of pre-1979 Iran, Shambayati (1994, p. 309) finds that state-society conflict increased given the public‘s concerns with how the oil revenues were being divided in the country. Gilbar similarly writes that ―The increased income from oil exports in the 1970s [in Iran] gave rise to, or precipitated, a whole range of social and economic changes, some of which created serious problems: accelerated migration from the countryside to the big cities…deepening economic disparities in Iranian society; and, most important of all, the disappointment among the lower strata of Iranian society in their expectation of an improvement in their living standards‖(Gilbar 1988, p. 207). Reflecting on the experiences of Oman, Alhaj observes that ―The oil economy had created visible collective inequalities which, when combined with Arab nationalistic or Islamic religious trends, produced several acts challenging the legitimacy of the traditional political elite and the policies of the government‖ (2000, p. 100). Furthermore, because oil wealth can fuel modernization it can create ―a new educated social group and a more broadly educated and politically aware public‖ . This may lead to conflict as the public demands that the government create new institutions that allow greater political participation. 57. The dynamics of succession may also make regimes in the region more or less prone to collapse. Some research exists on how the capacity of monarchies to survive depends on their ability to successfully manage succession (Herb 1999). The monarchy in Saudi Arabia, for instance, has remained in power using a system of hereditary succession through agnatic seniority since 1932; monarchs in other states – including Libya, Egypt, Iraq, and Iran – have been overthrown. On the other hand, non-monarchical, single-person dictatorships may be more vulnerable to collapse given the greater uncertainty surrounding succession. Recent events in Tunisia, for example, were triggered by a mounting succession crisis. 6. Political Transitions and PSM and PSD Priorities in MENA 58. The particular combination of largely authoritarian rentier regimes, for the majority of countries directly or indirectly sustained by oil and gas endowments, has for the most part generated a remarkable degree of stability in MENA, especially if one looks beyond the previous headline impressions of a number of lingering high profile conflicts that are often associated with the region. This combination has also tended to frustrate significant public and private sector reform and modernization. Public policies and programs, and their implications for private sector development, have been marked by a significant degree of patronage and lack of reform appetite, but also continuity. A previous MENA governance flagship publication already highlights some of the key reforms needed in the MENA region to enhance inclusiveness and sustainability (World Bank 2003). Key messages in the report 15 were that weak governance in the MENA region leads to weak growth, poor public services, limited citizen participation, and a restrictive business and investment environment. 59. Recent political developments, coupled with a view to the PSM and PSD reform deficit, has generated significant interest in the potential entry points for supporting MENA‘s anticipated transitions while enhancing violent conflict resilience. Our foregoing analysis has suggested that it is exceptionally difficult to predict the on-set of violent conflict both globally, and in MENA in particular. Establishing a chain of causality between PSM-PSD benchmarking indicators and reform trajectories and conflict risk is therefore exceptionally difficult. At the same time, some manifestations of conflict potentially also present positive triggers of change and reform. But for domestic reformers and international development partners, the priorities and potential sequencing in the MENA context is likely to be diverse and complex. Whereas a number of stakeholders will be aiming to foster more fundamental transformations in state-society relations, others may be focused on supporting better policies and emerging PSM and PSD reform opportunities. At the same time, ―crisis management‖, both in responding to mounting political pressures and economic shocks will invariably be an important driver of priority setting and decision making in MENA over the coming years. 60. The leverage of international development partners, and multilaterals such as the World Bank in particular, is likely to be quite limited in fundamentally affecting political dynamics in the regions in the prevailing context. The current international economic climate is unlikely to provide an environment to significantly increase international aid to the region, short of shifting resources from one setting to another. However, the region does enjoy the advantage of being endowed with significant rents. Already, Saudi Arabia appears to have increased its support to Yemen. As MENA‘s governments are pressed for enhanced inclusiveness, a major challenge will be to enhance targeting of its subsidy programs. However, this targeting will need to be less viewed through purely a pro-poor lens, but an understanding of which groups are likely to be most aggrieved and positioned to trigger instability. Development partners will need to think carefully about how to move with the political grain, as well as against it, when programming interventions over the coming years. At one level, donors should ask whether proposed reforms move either in a way that enhances inter-temporal credibility and inclusiveness. They will also need to be based on an appreciation that enhanced inclusiveness in the absence of safeguarding inter-temporal credibility may actually increase conflict risk beyond the short term. 61. International agencies engaged with resource-dependent settings have frequently focused on a number of policy management areas highlighted in Table 6. While most reforms are likely to strive for policies associated with programmatic pluralism (see Table 2), the actual choice of policies is likely to be closely associated with prevailing political economy exigencies. This will be conditioned by short term policy choices, and longer term institutional reforms. Arguably those factors that move a political economy to greater inclusiveness and inter-temporal credibility associated with the programmatic regime quadrant will also reduce the potential of violent conflict risk. The challenges of course is that countries will face a number of trade-offs and time-inconsistencies in effectively navigating these trajectories to more ―virtuous‖ development paths. We briefly review a number of indicators to stimulate the discussion concerning prevailing and prospective priorities across the region. These are intended to be illustrative, with a view of encouraging 16 the analysis of more nuanced quantitative and qualitative indicators and framing potential entry points. Table 6. Policy Management Domains, Prospective Instruments/Interventions, PE Drivers Policy Symptomatic Interventions PE Drivers Management Macro-Fiscal Lack of short and MTEFs, Stabilization & Significant distributional pressures, long term Savings Mechanisms, lack of ability of incentives to sustainability Balance Sheet/Green implemented longer term (boom & bust pro- Accounting consumption and savings policies cyclicality, vulnerability to shocks, poor genuine savings) Distribution Large and Subsidy Reform, Significant distributional pressures, "inefficient" Transfers nature of MENA state-society transfers (Technology/Indexing), compact Tax Reform Diversification Limited Regulatory Reform Lack of dynamic business sector competitive private constituencies, rentier private sector, low labor sector force participation, including woman A state dependent private sector that actively mobilizes against reforms that can allow new firms to enter the market. 62. While significant analytical efforts continue to be devoted to whether or not there is evidence for a resource curse, a broader policy consensus has emerged that institutional quality fundamentally conditions outcomes (including potential conflict) and policy choices. Operational implications can therefore be arguably found at two levels. First, those that seek to promote more optimal policies with respect to the mobilization and use of resource rents. Second, those that seek to strengthen institutional arrangements to better manage resource rents as well as encourage diversification and reduced dependence on some of the adverse impacts of resource dependence (notably shocks and volatility). The literature on natural resource-led development increasingly underscores that the policy choices observed in resource dependent countries are integrally related to prevailing political economy ―equilibria.‖ Going forward, most of MENA countries will not only continue to pose this challenge of improving policy management given the prevalence of high but frequently volatile rents, but also the special challenges of navigating the ―democratic deficit‖ illustrated in Figure 2. Fiscal Management for Sustainability and Conflict Mitigation 63. The political pressures associated with the Arab spring for greater inclusiveness may have significant implications for both short and longer prioritization of PFM reforms. Fiscal institutions can represent a critical element for managing revenue cycles, notably stemming 17 from the price and production volatility or ―rentier‖ revenues. The revenue volatility associated with resource revenue dependence has seen significant emphasis placed on smoothing revenues through stabilization mechanisms, as well as savings mechanisms.24 Fiscal institutions can help discipline both aggregate public expenditure management choices, but more generally set the framework for medium term prioritization and enhancing the quality of rent dissipation in resource rich settings. A ―voracity effect‖ has been associated with the significant revenue volatility and a periodic ratcheting up of various distributional demands in resource dependent settings (Tornell and Lane 1999). 64. But the particular forms of rent dissipation in MENA to-date could perhaps be better described as low level equilibria, where rent distribution has been a central element of the state-society compact coupled with a notably degree of stability. Admittedly the level of resource rent flows, and existing proven endowments and associated financial savings, varies significantly across the region. Hence it is difficult to generalize the relationship between fiscal challenges, prevailing political economies, and conflict risk. Given prevailing resource endowments, a key strategic question for MENA countries entering potential phases of transition is whether fiscal institutional arrangements that align to both increased inclusiveness and inter-temporal credibility dimensions are in place, and whether emerging reform initiatives are actually aligned with more fundamental shifts in the nature of the state- society compact. 65. The World Bank's Wealth of Nations "balance sheet" approach provides a useful framework for assessing the sustainability of development policies for countries relying on natural resources (Hamilton and Ley 2010; World Bank 2006, 2011a). The approach emphasizes debiting the depletion of non-renewable resources from traditional savings measures, while "crediting" government expenditures on education. Table 7 and Table 8 present these indicators from a cross-regional and MENA specific perspective. The data are subject to a number of missing observations, but highlight the prominence of resource depletion (at 20% of GPD, compared to 4.4 percent of global averages). The region exceeds only Sub-Saharan Africa, although levels are significantly lower there. Where available, the data show significant diversity across countries, but suggest high levels of financial savings and gross public and private fixed capital formation across the region, notably in the petrostates. Education spending is high in the region, although a number of non-resource settings (Tunisia, Israel, Jordan) do better by these measures. A key question of course pertains to the quality and inclusiveness of spending, and its ultimate impact on economic diversification impacts. Various forms of subsidies – from energy, public employment, housing – appear to be important elements of distribution in MENA, for example as denoted by relatively low fuel prices (Table 7 and Table 8, Colum (v)).25 In Algeria and Iran, for example, housing subsidies amount to the equivalent of 4–6 percent of the GDP, making the 24 There is much to be said about mitigating the macroeconomic shocks that periodically afflict oil producers. Also see the discussion in Frankel (2010); the Natural Resources Charter at www.naturalresourcecharter.org; and the concluding chapter of Ross (2012). For decades, well-meaning economic advisors have urged low-income commodity exporters to adopt some package of ―optimal‖ policies – for example, a combination of fiscal rules and stabilization funds – that seem to have worked in Norway. Unfortunately, these first-best policies have often failed in countries with weaker governance (Davis et al. 2003; Ossowski et al. 2008). 25 Other datasets present this data on the basis of subsidies relative to GDP 18 magnitude of expenditure comparable to expenditure on education or health (Buckley and Kalarickal 2006:49). Table 7. Medium Term Fiscal Management, by Regions Wealth of Nations Subsidy Medium Terms Expenditure Frameworks Savings (% Extractives "Genuine 2008 GNI) Depletion Education Savings" Fuel ($) obs MTFF(%) MTBF(%) MTBF(%) (i) (ii) (iii) (iv) (v) (vi) (vii) (viii) (ix) Advanced Economies 18 1.6 4.6 6.8 1.44 28 96% 57% 46% EAP 47.3 8.7 2.0 28.6 0.92 19 68% 16% 0% ECA 24.8 12.7 4.1 3.2 1.06 30 87% 50% 7% LAC 22.4 8.1 4.4 6.3 0.8 31 45% 13% 0% MENA … 20.1 4.4 … 0.5 19 58% 11% 0% SA 35.3 5.7 3.0 21.8 1.07 8 88% 25% 0% SSA 16.5 15.5 3.3 -6.2 1.19 46 74% 41% 9% Average/Total 20.9 4.4 4.2 7.2 1.03 181 73% 34% 10% Non-oil & gas 8.7 2.1 4.2 10.0 1.14 160 75% 36% 11% Non-oil & gas 23.5 33.1 3.4 -7.8 0.61 21 57% 14% 5% Source: Wealth of Nations (2011:195), PRMPS MTEF Database, WDI, Coady et. al. (2010) (ix) Notes: Oil dependent refers to more than 25 % income, export, or fiscal revenue dependence (source Barma et. al. 2011) 66. Prevailing political pressures for greater inclusiveness have seen a number of MENA governments emphasize boosting wages and benefits, and increasing transfers and subsidies – especially towards aggrieved groups and politically-influential constituencies. Most of these measures seem to be short term, ad hoc and ultimately unsustainable. This implies westward movement in Table 1 (and Figure 4), as intertemporal commitments come under strain and short-term considerations overwhelm long-term planning. A central question therefore will pertain to the institutional capability of MENA‘s governments to both maintain aggregate fiscal discipline, even in the absence of significant changes in resource prices, while enhancing the selectivity and sustainability of distribution programs. 67. Various forms of Medium Term Expenditure Frameworks (MTEFs) can serve as instruments to support both aggregate fiscal discipline, expenditure prioritization, and performance based orientation. A recent cross-country dataset aims to capture the prevalence of degrees of MTEFs globally, focusing on three indicators to denote institutionalization of macro-fiscal management, expenditure prioritization, and performance orientation (see Table 7, Colum vi-ix, Table 8).26 On adoption of expenditure prioritization mechanisms (MTBFs), MENA ranks lowest globally, followed by LAC. This raises significant questions as to the medium term mechanisms MENAs governments can adopt to sustainably reprioritize distribution programs in the context of political transitions, especially as the traditional 26 These are based on an extensive global dataset recently constructed by the World Bank‘s PREM Public Sector & Governance Group (forthcoming). The three different MTEF—MTFF, MTBF, and MTPF—were coded as one if a country met the following criteria: (i) MTFF: the government has rolling aggregate, expenditure, revenue, and other fiscal forecasts. Features include the availability of a macro-fiscal strategy, macroeconomic and fiscal forecasts, and debt sustainability analysis, (ii) MTBF: the budget, spending agency or other reports explain aggregate and sectoral expenditure objectives and strategies, budget circulars detail medium-term expenditure ceilings and revenue forecasts, and budget documents contain some detail for medium-term estimates. MTPF: the budget, spending agency or other reports explain program objectives and strategies, listing specific agency and/or program output or outcome targets, as well as results. 19 (narrower) credibility of traditional executive and patrimonial regimes begins to wane. A key challenge will be averting increasingly polarizing and short term claims over rents, and ensure that public expenditure management acts as a glue rather than solvent for peaceful political transitions. Resource rents on the one hand afford governments with the ability to underwrite distribution programs to buttress political stability, but also may reduce incentives to tax the non-resource economy (Brautigam et al. 2008; Knack 2008). Low levels of taxation of a productive society will risk dampening accountability, while resource dependence may furthermore be associated with an anemic private sector. The autocratic and patrimonial state-society relationships found in much of the region have also manifested themselves in centralized rent control and distribution, in addition to recourse to repression of groups or regions. Table 8. Medium Term Fiscal Management in MENA Wealth of Nations (2008) Subsidy Medium Term Expenditure Frameworks Sub- Savings (% Extractives "Genuine Region Country GNI) Depletion Education Savings" Fuel (USD) MTFF (%) MTBF (%) MTPF (%) (i) (ii) (iii) (iv) (v) (vi) (vii) (viii) N. Africa Algeria* 58.8 30.1 4.5 21.4 0.34 1 1 0 Dijbuti … 0 3.6 … 1.62 0 0 0 Egypt 23.5 15 4.4 2.1 0.33 0 0 0 Libya* 66.8 38.8 … … 0.14 … … … Morocco 31.4 6.1 5.2 19.8 1.29 1 0 0 Sudan* 15.9 19.2 0.9 -13.1 0.56 … … … Tunisia 22.6 10.5 6.7 7 0.96 1 0 0 Other Bahrain* 45.4 26.4 4.4 15.6 0.21 1 0 0 Iran* … … 4.2 … 0.10 0 0 0 Iraq* … … …. … 0.33 1 0 0 Israel 19.8 0.5 5.9 11.3 1.50 1 0 0 Jordan 13.7 4.7 5.6 3.6 0.61 1 1 0 Kuwait* 58.7 38 3 9.7 0.24 1 0 0 Lebanon 10.1 0 1.8 14.8 0.76 1 0 0 Oman* … … 3.9 … 0.31 0 0 0 Palestinian Territory … … … … 1.34 1 0 0 Qatar* … … … … 0.22 0 0 0 Saudi Arabia* 48.3 43.5 7.2 -1.8 0.16 0 0 0 Syria* 12.6 43.5 2.6 -15.2 0.85 0 0 0 UAE* … … … … 0.45 0 0 0 Yemen* … 43.5 … … 0.30 1 0 0 Average/Total 32.9 21.3 4.3 6.3 0.60 11 2 0 Source: Wealth of Nations (2011:195), PRMPS MTEF Database, IMF (viii), WDI Notes: * Countries denotes net oil & gas exporters (2008) (see Table 1) 20 Box 2: A Revolution in Public Distribution Systems? Any transfer program must include two basic elements – unique personal identification to indicate who is eligible for the transfer, and a system for making payments. New biometric and electronic transfer technologies have opened up an option of broadly based, transparency and potentially criteria based transfers that have only been a theoretical possibility up till now, except for high-income regions with small populations like Alaska. Applied well, these may provide important instruments for buttressing state-society relationships in MENA‘s states in a way that meets popular and legitimate objectives while limiting the proneness of human based identify systems to local capture and discrimination. Greater recourse to reformed ―blue-chip‖ transfer programs has the potential to substitute for inefficient fuel, food transfer systems and welfare public employment. Over 450 million people in developing countries have had their biometric data recorded for a variety of purposes, and over the next five years this number is expected to triple, or even quadruple. In at least eleven countries it has been used to provide the identifier that underpins a transfer program. Some of these programs have been on a relatively small scale, experiments financed by donors with no link to any nation-wide registration system. Others, such as the Watan card issued by the Government of Pakistan, have drawn on national biometric population registries to enable the rapid creation of emergency support. The Watan card program is targeted to 1.5 million flood-affected households and draws on Pakistan‘s National Database & Registration Authority (NADRA) fingerprint database that covers some 96 million citizens. In Nigeria, biometric audits reduced the number of Federal Pensioners from 97,000 to 60,000, a cut of almost 40%. India is now poised to create the largest biometric database in the world. In early 2009, the government announced that they would create a national ID program for its 1.1 billion people, with the goal to enroll 600,000 people in the first five years. In December 2010 Iran launched a massive program of direct transfers using electronic transfers into bank accounts, as a mechanism to compensate citizens for the removal of huge subsidies on food and fuel. In the first stage of the program, bread prices were tripled and fuel prices increased by a multiple from very low levels of around 7 cents/liter. A total of $15 billion was transferred to 20 million families through payments of $80 per family into bank accounts; this first installment was intended to cushion the increase for two months. Subsidies reached some 60 million people, about 80% of the population, with richer citizens tending to be excluded. In South Africa, pensions and other social payments have been delivered using biometric identification, smartcards and electronic transfers for many years. A powerful argument for direct, transparent, transfers is that in their absence many countries are forced into something worse. A particular strength of electronic systems of financial management is that they automatically generate an auditable trail which, in the case of biometric registration and payment identification, runs all the way to the final recipient. A second argument that a ―transfer-out-tax-back‖ system will improve governance and accountability. The need to tax citizens has been an essential component of the process of state-building; it has both forced states to develop capacity and encouraged demands for accountability to the public for money spent. There are a number of technical issues around the design of such large scale transfers, including incentive effects, whether citizens consume or invest these proceeds, averting volatility in transfer dependent on resource revenues and their operational feasibility and accountability. The deeper political question is whether MENA‘s leadership now has a greater appetite for availing themselves of the new technologies to meet pressures for greater inclusiveness with sustainability. In the current environment, citizens may be skeptical of nationwide identification systems, suspicious of any program that helps government monitor them in any way. Alternatives have focused on programmatic roll- outs around pensions, demobilization, and disaster relief. The current situation may offer the opportunity to proactively seek to launch credible programs, rather than revert to short term reactions and ad hoc rent dissipation. Source: Gelb and Decker (2011) 21 68. The political economy perspective set out in the introduction suggests that MENA‘s prevailing regimes face some risk of veering towards patrimonial ―corner solutions,‖ or failure to effectively address pressures for enhanced inclusiveness. Proactive responses to growing pressures for greater inclusiveness could seek to leverage expansions in programmatic and credible distribution programs (see Box 2). More decentralized distribution mechanisms with a significant emphasis on legitimacy and transparency may be one option to dissipate resource rents more broadly (Gelb and Grassman 2010). These may be targeted to particular performance objectives, including those associated with addressing the region‘s prevailing gender gaps. Promises of greater decentralization may be another avenue, but also likely to have more protracted and complex time horizons owing to their inherent political nature (Eaton et al. 2010), and the lags by which MENA‘s regimes could demonstrate credible actions to face the pressures of prevailing political contagion, as well as prepare better for managing more traditional economic shocks. 69. The level of resource decentralization/devolution remains quite limited in the region (Kaiser and Vinuela 2010). A much echoed concern of course in the presence of high rent dependence are concerns whether ―centers will hold‖ in the face of significant sub-national or even sectarian resource claims. A major question therefore becomes whether measures to promote actual devolution of deconcentration act as glue or solvents for evolving state-society relations. The case for transfers to citizens has been made in resource-rich settings from Iraq to Nigeria (Birdsall and Subramanian 2004), and also through the argument that it can serve as a building block to strengthening the state-society linkages and government resilience to resource revenue dependence by building mechanisms to tax part of these transfers back (Devarajan et al. 2010). The latter will of course depend on the institutional capability of states to tax more broadly, a capacity often enfeebled in rentier regimes. Promoting Economic Diversification and Private Sector Development 70. Many point to the Dutch Disease as the core causal story that explains the resource curse phenomenon, and how macro economics policymaking might exacerbate or alleviate the Dutch Disease effect.27 Yet, scholars have paid less attention to how the role of government policy may similarly produce uncompetitive firms in resource dependent countries in the context of rentier political economies. Consequently increased attention is being paid to how how reliance on oil or other natural resources may affect government regulation of firms and the country‘s business environment that entrepreneurs and investors face. There is some evidence that increasing dependence on oil and natural gas rents leads to a lower quality business environment in a country, as characterized by higher regulatory and financial barriers to domestic firms and entrepreneurs (Mazaheri, (2010). The mirror side of this situation is the prevalence of ―limited access orders‖ for a small set of well connected elites (North et al. 2009). Amin and Djankov (2009) similarly confirm an effect for dependence on oil and natural gas exports on the adoption of micro-economic reforms overall. The effect of 27 The Dutch Disease effect argues that significant natural resource exports lead to overvaluation of a country‘s currency, a wage premium in the resource sector, and increased demand for labor in non-tradeable sectors such as construction and retail services. As a result, a country‘s non-resource exports become more expensive, its industries become less competitive (globally and at home), and imports become cheaper (Murshed 1999; Sachs 1999.). 22 resource rents on the business environment can have a major impact on the ability of domestic entrepreneurs to open a business and for small and medium-sized firms, in particular, to effectively compete in markets. This will likely have a broader effect on entrepreneurship rates in a country. Although cross-national data with wide coverage are hard to come by, a World Bank study using data for up to 90 countries finds that a lower quality business environment will result in lower rates of entrepreneurship, as measured by ―entry density‖ rates (see Klapper and Love 2010). While rents may over significant periods of time help sustain a stable dual system of restricted access for domestic private sector actors, coupled with significant and largely unproductive public distribution mechanisms, this development trajectory poses significant challenges for the pursuit of a more dynamic and inclusive growth strategy. Figure 15. World Bank “Doing Business� Rankings, Regional Averages in 2010 (By Region, 2005-2011) 0 20 40 60 80 100 120 140 160 OECD high income Eastern-Europe/Central Asia East Asia/Pacific Latin America/Caribbean Middle East/North Africa South Asia Sub-Saharan Africa Avg Country Ranking, Overall Ease of Doing Business Avg Country Ranking, Overall Ease of Starting a Business Source: www.doingbusiness.org. Lower rankings indicate greater ease of doing business and starting a business. Ranking for Libya is missing and therefore not included. 71. Policy measures that might reverse this trend are greater reform of the regulatory environment that individuals and local businesses face. A few oil-dependent countries in the MENA region, such as Saudi Arabia and Bahrain, have made notable progress in recent years in reforming the regulatory barriers to domestic business activities. Others, such as Iran and Iraq, have made less progress (see World Bank 2011b). Overall, countries in the MENA region rank low on the measure of ―overall ease of doing business.‖ With respect to the ―ease of starting a business,‖ only countries in the Sub-Saharan Africa region average a lower ranking in 2010 (Figure 15). Note that Libya‘s ranking is missing from the dataset. A study by the World Economic Forum in 2007 on national business environment ranks Libya lower than Iraq, which the World Bank ranks near the bottom of the world in both categories (see Figure 15). Another way that these countries can foster the growth of the domestic private sector is by directly providing credit to firms, in particular those that are small and medium- sized. Strengthening the financial institutions that produce information on credit history and 23 transactions for banks and lending agencies can be another reform that will make credit allocation more meritocratic. 72. The prevalence of a ―rentier‖ private sector may not just stifle more dynamic employment creation, but also fuel a wider sense of entrepreneurial frustration, if access to a level playing field in terms of finance and domestic or international market opportunities is perceived as restricted. Many of the resource-dependent states that have managed to diversify have done so owing to the presence of a strong enough lobby from the more dynamic sectors, such as Chile and Malaysia, or strong other imperatives, including longer term food security in Indonesia (Gelb 2010). While Dubai‘s developmental trajectory is in a number of ways unique, it does underscore the important demand side dynamics or other political imperatives that provide the enabling environment for a more vibrant intermediate SME sector to thrive. Creating the space in order to enhance collective action beyond prevailing patterns may therefore be an important element of altering a country‘s development trajectory rather than reliance on a state ―supply‖ led initiative. 73. Regulatory reforms may also help stabilize countries when they transition to more accountable governance. We noted above that countries with a higher density of civil society organizations have historically made more peaceful and successful transitions from authoritarian to democratic rule. Economies that are dominated by state spending, and whose private sectors are dependent on government patronage and political connections – and who therefore dwell in the left column in Table 2 also have weaker civil societies. Societies that lack networks of civil society organizations that bridge economic, social, and political cleavages are also more likely to experience conflict when a government collapses or undergoes a significant transition (Varshney 2002). Economic reforms that expand private sector activity, and allow new firms outside of traditional clientelist networks to thrive, can strengthen civil society in ways that may ultimately avert conflict. Promoting Female Labor Force Participation 74. Finally, there is much that governments can do to offset the detrimental effects of oil wealth on female labor force participation in general, and fertility in particular. Many MENA countries – including major oil producers – have made substantial investments in female health and education; these investments have in turn brought about admirably rapid gains in female literacy and sharp reductions in fertility rates. These investments have boosted the supply of well-educated female workers; yet the demand for these workers is anomalously low, which help explains why investments in female health and education have not yet produced commensurate gains in female labor force participation. Further gains must come from the creation of new employment opportunities for women, particularly in the private sector. 75. In the Middle East, non-agricultural jobs in the private sector are especially scarce. Within the non-agricultural private sector, employment opportunities for women tend to be greater in firms that are export-oriented and foreign-owned (Baslevent and Onaran 2004; Ozler 2000). As we note above, many oil-exporting MENA countries have strikingly poor business climates. Regulatory reforms that open countries to more foreign investment, and 24 encourage the diversification of exports out of the petroleum sector, could also have a disproportionate impact on opportunities for women. 76. Measures to boost public sector employment for women would also help. One sensible measure might be for states to invest their resource revenues more heavily in economic sectors that have traditionally hired more women, like education and health, and hence can raise female labor force participation without violating the cultural norms that sustain gender-segmented labor markets in the MENA region. Equally important are policies that make it easier for women to join the labor force by curtailing gender discrimination, and encouraging generous family leave and child care measures. 7. Conclusions 77. William Ascher‘s Bringing in the Future (2009) has many important ideas for encouraging policymakers to adopt policies that have short-term political costs but long-term economic and social benefits. Will recent developments in the MENA region yield long-term benefits while minimizing political costs? The previous section presented some of the major structural issues inherent to MENA‘s rentier societies, yet the recent political upheavals associated with the Arab Spring may offer the promise for significant spates of governance and political economy reform. But the fluidity of the prevailing context, and looming economic shocks, is likely to challenge MENA‘s policymakers and political elites in significant ways. Moving beyond the rentier state associated particularly with oil- endowments is likely to involve precarious transitions, including ones which may alter the historically observed conflict risks in the region. 78. PSM and PSD reforms can present valuable building blocks to enhancing the prospective political transitions in MENA, and potentially mitigating the risks of violent conflict. But development partners will need to be fully cognizant of the limited actual leverage they are likely to have in shaping these transitions, in particular in settings where resource rents are still integral to the overarching political economy equilibria. A recent World Bank (2010:x) stocktaking of Public Financial Management (PFM) reforms in the MENA region alerts one to the operational limitations of systematic political economy analysis (PEA) for informing engagement choices: [At] its core PEA analysis embodies the important observation that political and bureaucratic factors matter. Would be reformers should not mechanistically seek to transfer approaches and practices that work well in one setting to other very different institutional contexts. Yet beyond this basic and sensible warning, the PEA literature has relatively little to contribute when it seeks to address the most important questions in any reform initiative: what should be done, when does it need to happen, who should do it and how. Each reform experience is unique and endogenous to a particular time and country, as well as to an individual set of political, bureaucratic, institutional and personal dynamics. PEA may be valuable in highlighting constraints or identifying potential supporters, and it can help to inform decisions about what is to be done. But the trajectories of the most successful reforms in MENA bear little resemblance to the type of recommendations that a typical donor supported PEA would generate. PEA is not a substitute for solid strategic decision-making, and even less for astute tactical maneuvering during implementation. More is likely to be gained by investing in 25 greater tactical flexibility, in terms of field-based governance advisors and public sector specialists who can provide real-time advice and assistance in high risk, high engagement settings, than in trying to perform more rigorous political and bureaucratic analysis up-front. 79. While we have some sympathy for the real life exigencies of public sector management reform practitioners at the front line, recent political upheavals in the region do suggest that this important work can be complemented by gaining a broader strategic bearing on the context. One objective of this paper has been to position the deeper rationale for PSM and PSD reform in the larger context of enhancing resource-dependent development trajectories, and above all the implications for conflict risk and resilience within MENA‘s prevailing political context. The paper also suggests that recent political upheavals, promising typically to enable transitions away from problematic authoritarian paternalism, must be carefully assessed from their prospects and risks for enhancing or eroding the bureaucratic functionality of the states and the enabling environment they provide for dynamic private sector development. 80. The prevailing transitions suggest that PSM and PSD reform engagement must carefully assess how it relates to a country‘s overall development trajectory in terms of both inclusiveness and inter-temporal institutional credibility. Some of the examples presented in the paper have been intended to be illustrative, but do suggest a rich research agenda in terms of both benchmarking indicators to capture both the specificities of the MENA context, including learning from past democratic transitions in other regions. We would however argue that the presence of resource rents, and the particular political regimes this has shaped in the region, needs to be considered as an important intervening variable in situating reform priorities and prospects. 81. Looking forward, it will be important to monitor the potential incendiary environment prevailing in the region and how it intersects with more traditional economic shocks in the area of energy and food prices, as well as finance. Accelerated by developments in the region, oil prices have looked to return to the historic highs of 2008. Food prices have also surged in the past year. Growing concerns surrounding political risk in the region may also press on financial risk premiums in the region, both as capital flights to safety and potential sovereign risk premiums rise. 82. Temporarily at least, the uncertainty around Libya has enhanced the terms of trade of net oil exporters by driving up international oil and gas prices. The low retail prices for fuel in most MENA countries (Table 8) mean that many less richly endowed settings (e.g., Egypt, Syria, and even Iran) are experiencing scissor effects from increasing international prices on the one hand, but the low absolute prices at the domestic level. While this may have a favorable terms of trade effect for oil producers, this will also increase liabilities in terms of subsidies.28 This may yet amplify the challenges around more standard economic shocks, and delicate subsidy reform pressures in particular. 28 A similar pattern has been experienced by Indonesia in recent years, which now straddles being a net oil importer and exporter. 26 83. In the short run, it will be challenging for many of the regimes in the region to transition from the ‗hegemonic‘ (southeast) to the ‗programmatic pluralism‘ (northeast) quadrant. More likely, regimes will wind up in one of the two western quadrants – either patrimonialsm or clientelistic pluralism. This brings both risks and opportunities. On one hand, it may open opportunities for desirable reforms as governments face greater pressures to transform rents into public goods, rather than private ones. On the other, it may begin to exert significant pressure and even erode institutions that to-date have been able to function in a relatively resilient manner, typically symbolized by strong executive powers. 27 References Al-Rasheed, Madawi (2002), A History of Saudi Arabia (New York:: Cambridge University Press). Alhaj, Abdullah Juma (2000), 'The Political Elite and the Introduction of Political Participation in Oman', Middle East Policy, 7 (3), 97-110. Amin, M. and Djankov, Simeon (2009), 'Natural Resources and Reform', World Bank Policy Research Working Paper, (Washington, DC: World Bank). Ascher, William (2009), Bringing in the Future: Strategies for Farsightedness and Sustainability in Developing Countries (Chicago & London: University of Chicago Press). Aslaksen, Silje (2010), 'Oil as Sand in the Democratic Machine?', Journal of Peace Research, 47 (421-31). Barma, Naazneen, Kaiser, Kai, and Le, Tuan Minh (2010), 'Rents to Riches? The Political Economy of Natural Resource Led Development', (Washington, DC: Draft Volume, June 16th, pp. 225.). Barma, Naazneen, et al. (2011), 'Rents to Riches? The Political Economy of Natural Resource Led Development', (Washington, DC: World Bank Press, forthcoming). Basedau, Matthias and Lay, Jann (2009), 'Resource Curse or Rentier Peace? The Ambiguous Effects of Oil Wealth and Oil Dependence on Violent Conflict', Journal of Peace Research, 46 (6), 757-76. Baslevent, Cem and Onaran, Ozlem (2004), 'The Effect of Export-Oriented Growth on Female Labor Market Outcomes in Turkey', World Development, 32 (8 (August)), 1375-93. Bellin, Eva (2004), 'The Robustness of Authoritarianism in the Middle East: Exceptionalism in Comparative Perspective', Comparative Politics, 36 (2), 139- 57. Birdsall, Nancy and Subramanian, Arvind (2004), 'Saving Iraq From Its Oil', Foreign Affairs. Blattman, Christopher and Miguel, Edward (2008), 'Civil Wars', Journal of Economic Literature. Brautigam, Deborah, Fjeldstad, Odd-Helge, and Moore, Mick (eds.) (2008), Taxation and State Building in Developing Countries (Cambridge: Cambridge University Press). British Petroleum (2010), 'Statistical Review of World Energy 2010', (London: http://www.bp.com/statisticalreview, June, pp. 50). Brunnschweiler, C. N. and Bulte, E. H. (2008), 'Linking Natural Resources to Slow Growth and More Conflict', Science, 320. Brunnschweiler, C. N. and Bulte, E. H. (2009), 'Natural Resources and Violent Conflict: Resource Abundance, Dependence, and the Onset of Civil Wars', Oxford Economic Papers, 61 (4), 651-74. Buckley, Robert M. and Kalarickal, Jerry (2006), 'Thirty Years of World Bank Shelter Lending: What Have We Learned?', (Washington, DC: World Bank, Directions in Development (Infrastructure)). Bueno De Mesquita, Bruce and Smith, Alastair (2010), 'Leader Survival, Revolutions and the Nature of Government Finance', American Journal of Political Science. Buhaug, Halvard, Gates, Scott , and Lujala, Paivi (2010), 'Lootable Natural Resources and the Duration of Armed Civil Conflict, 1946-2001', (Paper presented at the Paper presented at the 36th annualPeace Science Society meeting (November)). Chaudhry, Kiren A. (1997), The Price of Wealth: Economies and Institutions in the Middle East (Ithaca and London Cornell University Press). Collier, Paul (2009), Wars, Guns, and Votes: Democracy in Dangerous Place (New York: HarperCollins). Collier, Paul and Hoeffler, Anke (2004), 'Greed and Grievance in Civil War', Oxford Economic Papers, 56 (4), 293-308. Crombrugghe, Denis de , et al. (2009), 'Institutional Profiles Database III: Presentation of the Institutional Profiles Database 2009 (IPD 2009)', (Paris: Direction Generale Du Tresor et La Politique Economique (DGTPE) (http://www.cepii.fr/ProfilsInstitutionnelsDatabase.htm), pp. 25). Crystal, Jill (1989), 'Coalitions in Oil Monarchies: Kuwait and Qatar', Comparative Politics, 21 (4), 427-43. Davis, Jeffrey M., Oates, Wallace E., and Fedelino, Annalisa (2003), Fiscal Policy Formulation and Implementation in Oil-Producing Countries (Washington, DC: International Monetary Fund). Devarajan, Shantayanan, Le, Tuan Minh, and Raballand, Gaël (2010), 'Increasing Public Expenditure Efficiency in Oil-rich Economies: A Proposal', (Paper presented at OxCarre 2010 Conference (http://www.csae.ox.ac.uk/conferences/2010- EDiA/papers/099-Devarajan.pdf)). Diamond, Larry (2010), 'Why Are There No Arab Democracies?', Journal of Democracy, 21 (1). Dillman, Bradford (2000), State and Private Sector in Algeria: The Politics of Rent- Seeking and Failed Development (Boulder: Westview Press). Dube, Oeindrila and Vargas, Juan (2009), 'Commodity Price Shocks and Civil Conflict: Evidence from Colombia', (Cambridge: Harvard University). Dunning, Thad (2008), Crude Democracy: Natural resource wealth and political regimes (Cambridge: Cambridge University Press). Eaton, Kent, Kaiser, Kai, and Smoke, Paul (2010), 'The Political-Economy of Decentralization Reforms: A Development Partner Perspective', (Washington, DC: World Bank). Eifert, B. and Gelb, A. (2002), 'The Political Economy of Fiscal Policy and Economic Management in Oil-Exporting Countries', World Bank Policy Research Working Paper No. 2899, 1-33. Eifert, Benn, Gelb, Alan, and Tallroth, Nils Borje (2002), 'The Political Economy of Fiscal Policy and Economic Management in Oil-Exporting Countries', Policy Research Working Paper 2899. Evans, Peter B. (1989), 'Predatory, Developmental and Other Apparatuses: A Comparative Political Economy Perspective on the Third World State', Sociological Forum, (Fall ), 561-87. --- (1995), Embedded autonomy : states and industrial transformation (Princeton paperbacks; Princeton, N.J.: Princeton University Press) xx, 323 p. 29 First, Ruth (1980), 'Libya: Class and State in an Oil Economy', in Petter Nore and Terisa Turner (eds.), Oil and Class Struggle (London: Zed Press, 119-42). Fjelde, Hanne (2009), 'Buying Peace? Oil Wealth, Corruption and Civil War, 1985–99', Journal of Peace Research, 46, 199-218. Frankel, Jeffrey A. (2010), 'The Natural Resource Curse: A Survey', (Cambridge, MA: NBER Working Paper 15386). Gelb, Alan (2010), 'Economic Diversification in Resource Rich Countries', (Paper prepared for IMF High-Level Seminar: Natural Resources, Finance and Development, Algiers, November 4-5, 2010, pp. 23). Gelb, Alan and Grassman, Sina (2010), 'How Should Oil Exporters Spend Their Rents?', (Washington, DC: Center for Global Development, Worlking Paper 221 (August)). Gelb, Alan and Decker, Caroline (2011), 'Cash At Your Fingertipcs: Technology for Transfers in Resource-Rich Countries', (Washington, DC: Center for Global Development (CDG), draft, March 9). Gelb, Alan, Kaiser, Kai , and Viñuela, Lorena (2011), 'How Much Does Natural Resource Extraction Really Diminish National Wealth? The Implications of Discovery', (Washington, DC: draft paper). Gilbar, Gad (1988), 'One Arab State, Many Arab States: The Impact of Population Growth and Oil Revenues', in Elie Kedourie and Sylvia Haim (ed.), Essays on the Economic History of the Middle East (London: Frank Cass). Gleditsch, Nils Petter, et al. (2002), 'Armed Conflict 1946-2001: A New Dataset', Journal of Peace Research, 39 (5), 615-37. Haber, Steve and Menaldo, Victor A. (2011), 'Do Natural Resources Fuel Authoritarianism? A Reappraisal of the Resource Curse', American Political Science Review,, 105 (1), 1-26. Hamilton, Kirk and Ley, Eduardo (2010), 'Measuring National Income and Growth in Resource-Rich, Income-Poor Countries', (Washington, DC: World Bank, PREM, Economic Premise Note, No 28, August, pp. 7). Harbom, Lotta and Wallensteen, Peter (2010), 'Armed Conflicts, 1946-2009', Journal of Peace Research, 47 (4), 501-09. Henry, Clement (2004), 'Algeria‘s Agonies: Oil Rent Effects in a Bunker State', The Journal of North African Studies, 9 (2), 68-81. Herb, Michael (1999), All in the Family: Absolutism, Revolution, and Democracy in the Middle Eastern Monarchies (Albany: State University of New York Press). Humphreys, Macartan (2005), 'Natural Resources, Conflicts, and Conflict Resolution: Uncovering the Mechanisms', Journal of Conflict Resolution, 49, 508-37. Jamal, Amaney, et al. (2010), 'Fertility and Economic Development in the Muslim World', (Princeton, NJ: Princeton University). Kaiser, Kai and Vinuela, Lorena (2010), 'Intergovernmental Fiscal Management in Natural-Resource Rich Settings: A Guide for Policy Practitioners', (Washington, DC: PREM Public Sector Group (mimeo)). Karl, Terry (1997), The Paradox of Plenty: Oil Booms and Petro-States (Berkeley: University of California Press). Karshenas, Massoud (1990), Oil, State, and Industrialization in Iran (New York: Cambridge University Press). 30 Klapper, Leora and Love, Inessa (2010), 'The Impact of the Financial Crisis on New Firm Registration', (Washington: World Bank Policy Research Working Paper 5444). Knack, S. (2008), 'Sovereign rents and the quality of tax policy and administration', (Washington, DC: World Bank Policy Research Paper 4773). Kohli, Atul (2004), State-directed development : political power and industrialization in the global periphery (Cambridge, UK ; New York: Cambridge University Press) xii, 466 p. Kuran, Timur (1989), 'Sparks and Prarie Fires: A Theory of Unanticipated Political Revolution', Public Choice, (61), 41-74. Lal, D. and Myint, H. (1996), The Political Economy of Poverty, Equity and Growth (Oxford: Clarendon Press). Lowi, Miriam (2004), 'Oil Rents and Political Breakdown in Patrimonial States: Algeria in Comparative Perspective', Journal of North African Studies, 9 (3), 83-102. Luciani, Giacomo (1994), 'The Oil Rent, the Fiscal Crisis of the State and Democratization', in Ghassan Salamé (ed.), Democracy without Democrats? The Renewal of Politics in the Muslim World (London: I.B Tauris). Lujala, Päivi (2009), 'Deadly Combat over Natural Resources: Gems, Petroleum, Drugs, and the Severity of Armed Civil Conflict', Journal of Conflict Resolution, 53 (1), 50. --- (2010), 'The Spoils of Nature: Armed Civil Conflict and Rebel Access to Natural Resources', Journal of Peace Research, 47 (1), 15-28. Lust-Okar, Ellen and Sambanis, Nicholas (2007), The Political Economy of Internal Armed Conflict in the Middle East and North Africa (New Haven: Yale University). Maddison, Angus (2009), 'Historical Statistics of the World Economy', (Groeningen, Netherlands: 1-2008). Mahdavy, Hussein (1970), 'The Patterns and Problems of Economic Development in Rentier States: The Case of Iran', in M. A. Cook (ed.), Studies in the Economic History of the Middle East (London: Oxford University Press). Marshall, Monty and Jaggers, Keith (2007), 'Polity IV Project: Political Regime Characteristics and Transitions', (http://www.systemicpeace.org/polity/polity4.htm). Mazaheri, Nimah (2010), 'Petrostates and Private Sectors: Regulating Business and Markets in Resource-Rich Countries', (SSRN Working Paper). McNeish, John-Andrew (2010), 'Rethinking Resource Conflict', (Washington, DC: World Development Report 2011 Background Paper). Messiant, Christine (2001), 'The Eduardo Dos Santos Foundation: Or, How Angola's Regime Is Taking over Civil Society', African Affairs 100, 287-309. Miguel, Edward, Satyanath, Shanker, and Sergenti, Ernest (2004), 'Economic Shocks and Civil Conflict: An Instrumental Variables Approach', Journal of Political Economy, 112 (4), 725-54. Moore, Clement Henry (1976), 'Petroleum and Political Development in the Maghreb', in Naiem A. Sherbiny and Mark A. Tessler (eds.), rab Oil: Impact on the Arab Countries and Global Implications (New York: Praeger Publishers), 112-29. 31 Murshed, S. Mansoob (1999), 'A Macroeconomic Model of a Developing Country Endowed with a Natural Resource.', (Helsinki: United Nations World Institute for Development Economics Research, Working Paper, No. 165). North, Douglass C., Wallis, John Joseph, and Weingast, Barry (2009), 'Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History', (Cambridge: Cambridge University Press), 326. O‘Reilly, Marc (1998), 'Omanibalancing: Oman Confronts an Uncertain Future', Middle East Journal (52), 70-84. Olson, Mancur (1993), 'Dictatorship, Democracy, and Development', American Political Science Review, 87 (3), 567-76. Ossowski, Rolando, et al. (2008), Managing the Oil Revenue Boom: The Role of Fiscal Institutions (Washington, DC: International Monetary Fund). Østby, G., Nordås, R., and Rød, JK (2009), 'Regional Inequalities and Civil Conflict in Sub-Saharan Africa', International Studies Quarterly, 53 (2), 301-24. Ozler, S. (2000), 'Export Orientation and Female Share of Employment: Evidence from Turkey', World Development, 28 (7), 1239-48. Ross, Michael (2001), 'Does Oil Hinder Democracy?', World Politics, 53 (April), 325-61. --- (2004a), 'What do we know about natural resources and civil war?', Journal of Peace Research, 41 (3), 337-56. --- (2004b), 'How Do Natural Resources Influence Civil War? Evidence from 13 Cases', International Organization, (58), 35-67. --- (2008), 'Oil, Islam, and Women', American Political Science Review, 102 (1), 107-23. --- (2010), 'Latin America's Missing Oil Wars', (Los Angeles: University of California). --- (2012), 'The Oil Curse: How petroleum wealth shapes the development of nations', (Princeton, NJ: Princeton University Press). Ross, Michael L. (2006), 'A Closer Look at Oil, Diamonds, and Civil War', Annual Review of Political Science, 9, 265-300. Sachs, Jeffrey D. (1999.), 'Resource Endowments and the Real Exchange Rate: A Comparison of Latin America and East Asia', in T. Ito and A. Krueger (eds.), Changes in Exchange Rates in Rapidly Developing Countries: Theory, Practice, and Policy Issues (Chicago: University of Chicago Press). Shambayati, Hootan (1994), 'The Rentier State, Interest Groups, and the Paradox of Autonomy: State and Business in Turkey and Iran', Comparative Politics, 26 (3), 307-31. Smith, Benjamin (2006), 'The Wrong Kind of Crisis: Why Oil Booms and Busts Rarely Lead to Authoritarian Breakdown', Studies in Comparative International Development, 4 (40), 55-76. Stein, Ernesto, Scartascini, Carlos, and Tommas, Mariano (2008), Policymaking in Latin America: How Politics Shapes Policies. Cambridge, United States: Harvard University, David Rockefeller Center for Latin American Studies. (Cambridge: Harvard University, David Rockefeller Center for Latin American Studies). Tessler, Mark A. and Jamal, Amaney (2008), 'Attitudes in the Arab World', Journal of Democracy, 19 (1), 97-110. Tornell, Aaron and Lane, Philip R. (1999), 'The Voracity Effect', American Economic Review, 89 (1), 22-46. 32 Urdal, Henrik (2006), 'A Clash of Generations? Youth Bulges and Political Violence', International Studies Quarterly, 50 (3), 607-29. Vandewalle, Dirk (1998a), Libya since Independence: Oil and State-Building (Ithaca: Cornell University Press). --- (1998b), Libya since Independence: Oil and State-Building (thaca; Cornell University Press). Varshney, Ashutosh (2002), Ethnic Conflict and Civic Life: Hindus and Muslims in India (New Haven: Yale University Press). Viñuela, Lorena (2008), 'Limits of Regional Mechanisms in Latin America', Maxwell Review, (Fall). World Bank (2003), 'Better Governance for Development in the Middle East and North Africa: Enhancing Inclusiveness and Accountability', (Washington, DC: MENA, http://go.worldbank.org/TPI9B58YT0, pp. 306). --- (2004), 'Gender and Development in the Middle East and North Africa', (Washington D.C: World Bank). --- (2005), 'The Status and Progress of Women in the Middle East and North Africa', (Washington D.C: Middle East and North Africa Social and Economic Development Group). --- (2006), 'Where is the Wealth of Nations? Measuring Capital for the 21st Century', (Washington, DC: World Bank). --- (2010), 'Public Financial Management Reform in the Middle East and North Africa: An Overview of Regional Experience', (Washington, DC: Middle East and North Africa Vice Presidency, Part I: Overview and Summary, Report No. 55061- MNA). --- (2011a), The Changing Wealth of Nations: Measuring Sustainable Development in the Milenium (Washington, DC: World Bank, Environment and Development Series). --- (2011b), 'Doing Business 2011: Making a Difference for Entrepreneurs (Middle East and North Africa)', (Washington, DC). Yousef, Tarik (2003), 'Youth in the Arab World: Demography, Employment, and Conflict', in Blair Ruble (ed.), Youth Explosion in Developing World Cities: Approach to Reducing Poverty and Conflict in an Urban Age (Washington D.C.: Woodrow Wilson Center, 9-24. .v). 33