WPS5196 Policy Research Working Paper 5196 Development Strategies Integrating Governance and Growth Brian Levy Francis Fukuyama The World Bank Poverty Reduction and Economic Management Vice Presidency Public Sector Governance Unit January 2010 Policy Research Working Paper 5196 Abstract A frontier challenge for development strategy is to the potential for arbitrary discretionary action --and move beyond prescribing optimal economic policies, thereby shift expectations in a positive direction; 3) For and instead--taking a broad view of the interactions `just enough governance', the initial focus is on growth between economic, political and social constraints itself, with the aim of addressing specific capacity and and dynamics--to identify entry points capable of institutional constraints as and when they become breaking a low-growth logjam, and initiating a virtuous binding--not seeking to anticipate and address in spiral of cumulative change. The paper lays out four advance all possible institutional constraints; 4) Bottom- distinctive sequences via which the different dimensions up development engages civil society as an entry point for might interact and evolve over time, and provides seeking stronger state capacity, lower corruption, better country-specific illustrations of each. Each sequence is public services, improvements in political institutions defined by the principal focus of its initial step: 1) State more broadly--and a subsequent unlocking of capacity building provides a platform for accelerated constraints on growth. growth via improved public sector performance and The sequences should not be viewed as a technocratic enhanced credibility for investors; strengthened political toolkit from which a putative reformer is free to choose. institutions and civil society come onto the agenda only Recognizing that choice is constrained by history, the over the longer term; 2) Transformational governance has paper concludes by suggesting an approach for exploring as its entry point the reshaping of a country's political what might the scope for identifying practical ways institutions. Accelerated growth could follow, insofar as forward in specific country settings. institutional changes enhance accountability, and reduce This paper--a product of the Public Sector Governance Unit, Poverty Reduction and Economic Management Vice Presidency--is part of a larger effort in the department to better address governance challenges in the design and implementation of development strategies. Policy Research Working Papers are also posted on the Web at http://econ. worldbank.org. The author may be contacted at blevy@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 - 1 - DEVELOPMENT STRATEGIES: INTEGRATING GOVERNANCE AND GROWTH Brian Levy and Francis Fukuyama 1 I: OVERVIEW There is widespread agreement among social scientists that development is a multidimensional phenomenon, with economic, political, and social aspects, and that the different dimensions of development are interconnected with one another in complex ways. The fact that these different dimensions exert causal influences on one another would suggest that policymakers dealing with issues of national development need to integrate economic, political, and social approaches, and to think strategically about how to achieve their goals. Strategic thinking implies prioritization and sequencing, seeking to do first that which is necessary or helpful to achieve later goals. A strategy begins by defining a clear objective (e.g., are we aiming in the first instance at economic growth, democratization, establishment of a rule of law, a more vigorous civil society?), and then assesses constraints that prevent the achievement of this objective. Strategic thinking looks at alternative pathways that get around those constraints; if this is not possible, it requires the redirection of energies to other goals that are more realistically achievable. Yet development policymakers and practitioners seldom think in this manner. That is, they do not integrate political, economic, and social strategies; they do not set clear priorities or sequence reforms over time; and they do not pay sufficient attention to constraints. Development strategies in official documents often read like wish lists 1 Brian Levy is Advisor, Governance at the World Bank Group; Francis Fukuyama is the Bernard L. Schwartz Professor of International Political Economy at the School of Advanced International Studies, Johns Hopkins University - 2 - of all of the good things that the sponsors would like to see happen, with little appreciation for what needs to be done first in order to facilitate the achievement of other objectives down the road. There are several reasons for this lack of strategic thinking about development. Development specialists tend to be compartmentalized by academic discipline. While economists, for example, have recently come to take politics and institutions more seriously, their first inclination is usually not to think about the political preconditions for achieving economic goals. Conversely, people in the democracy promotion field often do not pay attention to economic issues, or else relegate growth to a second-order priority. In the donor community, there is a great deal of political sensitivity on the question of sequencing or prioritization. If one admits that there is little chance for a democratic breakthrough or for cleaning up a particular corrupt bureaucracy, one appears to be endorsing a bad status quo. The one strategic issue that has generated a substantial amount of debate concerns the sequencing of economic and political reforms. Proponents of the so-called "authoritarian transition" like Huntington (1968) and Zakaria (2003) have argued in favor of putting economic development and establishment of a liberal rule of law ahead of democratization, given that a stronger causal connection appears to exist from economic growth to democracy than the other way around. Others like Plattner (1999) and Carothers (2007) have argued that such sequencing is not a practical option in most cases, and that advocates of reform in developing and transitional countries usually want both rule of law and democracy simultaneously. This debate is a worthwhile one, but too limited in certain respects. There are more dimensions of development than economic - 3 - growth and democracy, and therefore more entry points to consider when designing a national development strategy. Strategic thinking about development requires consideration of the full range of these entry points, and how progress along one dimension affects progress in the others. This paper suggests a more comprehensive framework for thinking about development strategies and for integrating political, economic, and social policies. The next section introduces each of five dimensions, and summarizes briefly what is known about their causal links with one another. Section III lays out four distinctive sequences via which the different dimensions might interact and evolve over time. The final section examines how the `sequences framework' can aid development decision-making. II: THE DIMENSIONS OF DEVELOPMENT STRATEGY There are five broad dimensions of development, one economic, one social, and three political: · economic growth · development of civil society · state-building · liberal democratic political institutions, including both rule of law and electoral democracy We will discuss each of these dimensions as entry points for development, and as objectives of development in themselves. Obviously, each of these dimensions contains within itself a vast number of goals and approaches. What we seek to do here is not to choose one over another, but to begin to understand how they are related to one another. - 4 - Economic growth. We define economic growth in a narrow, traditional sense, as increasing per capita GDP. Much of the field of development economics has centered around strategies for promoting growth, and there have been numerous approaches, fads and orthodoxies in this area over the years. Some growth strategies result in a more egalitarian income distribution than others ­ a variation that, as will become evident, can be relevant for the longer-term sustainability of a country's development path. Many East Asian fast developers like Japan, South Korea, and Taiwan pursued industrial policies involving targeted credits and managed sectoral transitions. By contrast, the 1980s and 90s saw the rise of a very different approach, the so-called "Washington Consensus" that sought to reduce state intervention in favor of market pricing. We do not intend to rehash familiar arguments over the relative merits of these and other economic strategies, except to note that, in addition to differing in their implications for income distribution, they have rather different political and institutional requirements. As Haggard (1990) suggests, industrial policies can be made to work, but require a substantially greater degree of technocratic capacity to manage properly than ones that rely on market pricing. Equally important, they are liable to capture by various political actors seeking rents, and have to be carefully insulated from them. Development of Civil Society. There is a large body of social thought detailing the social changes that take place as a society shifts from one that is primarily agricultural to one that is industrial. In the former, social relationships are often ascriptive rather than voluntary, based on kinship, ethnicity, social class, and gender. Social hierarchies are often inherited, with little opportunities for individuals to change the status into which they were - 5 - born. In addition, the division of labor is limited by the small size of the market economy and non-agricultural sector. A modern civil society emerges when social groups between the family and the state are able to freely organize on the basis of shared passions and interests. Developed societies remain hierarchical, but hierarchies ideally are more fluid and accessible; social mobility and representation increases. State-building. State-building is itself a multi-dimensional phenomenon that is the precursor of and necessary condition for either liberal rule of law or democracy. We will make use here of Max Weber's famous definition of the state: a legitimate monopoly of force over a defined terrority. State-building begins with a concentration of coercive power in the hands of the state, through the disarming or destruction of private militias and the creation of a national army and police. It also involves defining the state's territorial extent (by either incorporating or sloughing off particular geographical regions), and extending the reach of the state's enforcement power over that territory. Finally, state-building involves creating administrative capacity in the form of public bureaucracies. Core state functions (beyond internal and external security) include the ability to extract taxes, the ability to budget and spend money, and the ability to enforce the state's rules. In more established states, state-building can also refer to the expansion of state functions, improvement of state efficiency in provision of services, and control of official corruption. Liberal Democratic Political Institutions. While state-building involves the concentration of power in the state's hands, establishment of modern political institutions limits that power by reducing the - 6 - state's discretionary use of force. The "liberal" part of liberal democracy is rule of law. Rule of law is the basis for property rights and the adjudication of commerial claims, and thus is key to sustained economic development. Rule of law is also the basis for the protection of a private sphere and individual human rights. Legal rights do not have to be universal; in some societies, they are enjoyed only by elites who benefit from the full privileges of citizenship. Universalization of the rule of law permits larger markets, greater competition, and in the long run, more economic growth. A second aspect of political institutions is democracy, that is, popular sovereignty through regular multiparty elections. Democracy can involve a number of different mechanisms for holding governments accountable to the people. Besides elections, there are other mechanisms of accountability as well such as separated powers which monitor each other's behavior, and a free press and civil society outside of the formal political system that can monitor and check the government's performance. As in the case of rule of law, democracy puts limits on the state's discretion and forces it to reflect the will of at least some important proportion of the people. All five of these dimensions are goals of development in themselves, and they can exist, for the most part, independently of one another. That is, one can have growth without social development, and social development without increases in either state capacity or democracy. It is possible to have an illiberal democracy, and a liberal autocracy, and both democracies and autocracies can experience either low or high growth. While a rudimentary state is a necessary precondition for economic growth, rule of law, and democracy, it is also possible to have some or all of the latter three conditions in a weak state. - 7 - There is a sixth, intangible factor that is critical to development, which is the credibility and legitimacy of the state. Credibility has to do with expectations that the government will do what it promises, whether that is upholding individual human rights or protecting the interests of property owners. Legitimacy has to do with the degree to which the society's citizens believe that the system as a whole is just, and deserving of their support (even if they disagree with certain of the government's policies). Credibility and legitimacy are related to the five main channels of development described above ­ whether, for example, the benefits of growth are perceived to be shared -- but are not simply coterminous with any of them. They arise as byproducts of the other channels, but are not in themselves entry points for development. We know that there are certain presumed causal relationships between certain of these dimensions that can form the basis for development strategies. The presumption of causality is based on a combination of theoretical reasoning and empirical correlations between the different phenomena. Though the precise causal pathways are often not precisely understood (for example, between development and democracy) some plausible relationships include: Between state building and growth. Basic state formation is a precondition for sustained growth. Paul Collier (2007) has demonstrated the converse of this proposition, namely, that civil war and interstate conflict have very negative consequences for economic growth. Having a Weberian state at peace is a precondition not just for sustained growth, but for virtually all of the other development objectives (fair distribution, rule of law, democracy). Beyond establishment of a state that can provide for basic order, greater administrative capacity is also strongly correlated with - 8 - economic growth. This is particularly true at low absolute levels of per capita GDP (i.e., less that $1000); while it remains important at higher levels of income, the impact may not be proportionate. There is also a large literature linking good goverance to economic growth. There is a debate over the direction of causality here, with Sachs maintaining that governance is endogenous to growth; Easterly (2006) has argued that the causality goes the other way. It would seem likely that causality here is bidirectional, and that economic growth facilitates greater state administrative capacity. Between liberal rule of law and growth. There is a large literature demonstrating the correlation between property rights (and a supporting rule of law) and growth (see Acemoglu and Robinson 2005). While most economists take this for granted, there is an interesting new literature showing that growth can also occur under situations where property rights and rule of law are not universal, that is, where elites grant rents to themselves (see Haber, Razo, and Maurer 2003; Khan and Jomo 2000). Between economic growth and stable democracy. Beginning with Lipset (1958) there has been a large literature linking development and democracy (Diamond 1992). The relationship between growth and democracy may not be linear; Barro (1997) shows that it is stronger at lower and weaker at middle levels of income. Przeworski and Alvarez (1997) show that while democratic transitions can occur at any level of development, they are much less likely to be reversed past a level of about $6000 per capita. The evidence linking democracy to economic growth is much less clear; what the data show is a much wider variance in the economic performance of autocracies than democracies. Between liberal rule of law and democracy. While we do not know of empirical studies explicitly correlating these factors, we presume - 9 - it must be true because there are many examples of liberal democracy, and relatively few cases of liberal autocracy (though perhaps more in the past). Whether a causal connection exists as Plattner suggests is not clear, but may well be so. Between growth and the development of civil society. A lot of classic social theory predicates the emergence of modern civil society on economic development (see for example Gellner 1992). Adam Smith in the Wealth of Nations notes that the extent of the market determines the division of labor; as growth occurs and firms take advantage of economies of scale, social specialization increases and new social groups (e.g., an industrial working class) emerge. The fluidity and open access demanded by modern market economies undermines certain traditional forms of authority and forces the replacement of ascriptive with voluntary social groups. Between development of civil society and liberal democracy. From Tocqueville onwards there has been a large body of democratic theory arguing that modern liberal democracy cannot exist without a vigorous civil society. The latter allows weak individuals to pool their interests and enter the political system; even when social groups do not seek political objectives, voluntary associations have spillover effects in creating social capital. A highly developed civil society also poses dangers for democracy as well: groups based on ethnic or racial chauvinism spread intolerance; interest groups can invest effort in zero-sum rent seeking; excessive politicization of economic and social conflicts can paralyze societies and undermine the legitimacy of democratic institutions. The correlation that exists between economic growth and stable liberal democracy (Lipset 1958; Przeworski 1993) is presumably mediated by this channel: growth entails the emergence of - 10 - new social actors who then demand representation in a more open political system. Between the development of civil society and state-building. There are two opposing propositions as to the `sign' of the coefficient that links these two dimensions. Huntington (1968) argued that there was often a negative relationship between the mobilization of new social actors and political order when existing political institutions could not accommodate their demands for participation. On the other hand, contemporary thinking about governance reform asserts that civil society, in the form of free media and organized citizens' groups, are necessary to promote transparency with regard to the quality of government and accountability when abuses are uncovered. Plausibly, which of these opposing propositions is dominant depends on a country's institutional starting point ­ with the former proposition more relevant in settings where political institutions and state capacities are especially weak (Bates 2008). - 11 - III: ENTRY POINTS AND SEQUENCES Consider as a starting point a low-income country enjoying little or no economic growth. As the political-economic framework outlined in the previous section suggests, the challenge for reformers is not to prescribe optimal economic policies, but rather -- taking a broader view of the interactions between economic and political constraints and dynamics -- to identify entry points capable of breaking the low-growth logjam, and initiating a virtuous spiral of cumulative change. The relevant constraints, plausible entry points, and subsequent dynamics all are country-specific. There is no `one size fits all' ­ but there appear to be a few overarching patterns. To help guide the search for country-specific actions, this section will lay out four distinctive development sequences ­ each initiated by distinctive entry points, and with distinctive subsequent paths. Sequence I: State Capacity Building When a low-income country is not growing, some of the key reasons are plainly visible to lay citizens, as well as experts. Government doesn't work. Political leaders are inept or corrupt (or both). The risks to private investors and entrepreneurs -- of failed infrastructure, of a dysfunctional, burdensome bureaucracy, of political pressures to `share' profits, of violence and instability ­ generally overwhelm any entrepreneurial instinct to seek out opportunities for productive investment. In environments such as these, a natural response of a country's citizens (and its development partners) is to look for `political will' ­ for leadership ready to respond to the public interest rather than private ends, to improve policy-making, to fix the broken bureaucracy, to make wise public investments, and to put in place an environment - 12 - capable of attracting private investment. More broadly, the entry point implicit in this first of four development sequences is to build the capacities needed to transform state dysfunction into state effectiveness. Key transmission channels. Figure 1 depicts how state capacity building potentially could set in motion a virtuous development spiral. The first step is an investment in state capacity building. The second step applies this newly built capacity to promoting economic growth. This impact potentially can occur via three complementary channels: · Channel 2a highlights the direct links between improvements in state performance and economic performance. Better policies, more efficient infrastructure, a transactionally more efficient bureaucracy, transparent and participatory approaches to service provision ­ all of these can contribute to a better investment climate, and thus potentially to growth. Improving infrastructure and the bureaucracy takes time, with little opportunity for `quick wins -- policy, though, can be turned around quite rapidly. · Channel 2b highlights the expectations-driven impact of the emergence of leadership perceived to have the `political will' to build state capacity. Even before public sector performance actually improves, credibility can rise among private investors as to the productive potential of the economy ­ with the credibility gain itself sometimes sufficient to achieve renewed economic activity, and an acceleration of growth. (One common way - 13 - to secure such credibility has been to move rapidly to adopt far- reaching `stroke of the pen' economic reforms. 2) · As channel 2c suggests, political leaders can make the capacity and credibility channels mutually reinforcing by asserting a broader commitment to `fairness' ­ to inclusive growth with broadly shared benefits. Insofar as citizens believe this commitment, state legitimacy and stability will both be enhanced ­ contributing more broadly to improvements in the investment climate, and hence to growth. Figure 1: State Capacity Building as an Entry Point for Development CIVIL SOCIETY GROWTH DEVELOPMENT 2b & c 2c 2a Credibility and 3 legitimacy 2b STEP 1: BUILD STATE POLITICAL CAPACITY INSTITUTIONS Note that none of the channels described above involve political institutions. Insofar as a sequence led by state capacity building can 2 Thus, paradoxically, for reasons that will be elaborated later, insofar as `structural adjustment' can be said to have had an implicit institutional model, it fits best with the state capacity building sequence ­ though the economic reform package often included `state- capacity-destroying' elements. - 14 - lead to transformation of political institutions, the route is a long- term and indirect one ­ with social change (via, say, participatory approaches to service provision, wealth creation, and rise of a middle class) generating new pressures for the rule of law to manage inter- elite conflicts (both between firms, and between businesses and the state), and for political reform more broadly. Channel 3, the dotted line in Figure 1 illustrates this longer-term dynamic. Experiences and challenges. Readers familiar with the development literature will recognize the sequence described above as that of the `developmental state'. The past fifty years offers no shortage of countries -- and political leaders ­ that have put themselves forward as aspirant `developmental states'. It was the dominant model underlying East Asian development. It was an aspiration of many Latin American countries (including Brazil and Mexico) prior to the 1980s. The former centrally planned economies offered a (failed) radical variant for much of the twentieth century. Still today ­ though communism has collapsed, and the international economic and political environment is very different than that which prevailed during the rise of the East Asian developmental states --there is no shortage of as aspirant developmental states among low-income countries. Examples range from Ethiopia, Rwanda and Uganda in Africa, to Azerbaijan in Central Asia, and Vietnam in South East Asia. In each of these cases, governments combine efforts to win legitimacy by highlighting the competence of their rule, on the one hand, with a continuing wariness of greater political participation and accountability, on the other. Each has achieved quite strong economic performance (relative to their peers, and/or their preceding historical experience). A few development states have been spectacularly successful, with Korea and Taiwan two classic examples. In both countries, development - 15 - was led by interventionist, competent states that used pro-active industrial policies to fuel growth. In both, the initial decades of accelerated development were under authoritarian rule ­ but in both `fairness' provided a basis of legitimacy. Land reforms assured rural equity; growth was oriented towards labor-intensive exports and so employment creating; high public investment in education provided opportunity for all. And in both, inclusive growth set in motion profound social transformation ­ facilitating the emergence of a middle class, and an eventual transition to democracy. Sustained success remains, however, more the exception than the rule. Two sets of cautions seem especially relevant for a development strategy based on building state capacity. A first caution is that in the early years of a new political leadership, it can be difficult to tell whether a seeming commitment to `developmentalism' and associated state capacity building is real, or simply a useful cover for the perpetuation of longstanding patterns of corrupt, patrimonial rule. From Moi in Kenya, to Suharto in Indonesia and Marcos in the Philippines, recent history is replete with leaders who have touted themselves as `developmental', concentrated authority on the basis of its necessity for their achieving `developmental' goals, and been showered with largesse from a supportive donor community ­ only to have their reigns end in recrimination, corruption and disgrace. Such regimes often use the rhetoric of`state capacity building' as a key part of their program: it has the virtue of being long on ambition and fine-sounding objectives, but sufficiently `soft' and supposedly `long-term' in its impact that busy work can proceed for long periods of time before it becomes evident that nothing much is being achieved. - 16 - The second set of cautions concerns the move towards more pluralistic political institutions (Step 3 in Figure 1). As Figure 1 implies, developmental states typically focus first on economic and bureaucratic reforms, while neglecting reform of political institutions. Korea and Taiwan offer successful examples of a seemingly successful subsequent institutional `catch-up', but other trajectories also are feasible. Singapore, for example, generally has been included with Korea and Taiwan as an example of East Asian success ­ but for all of its contemporary economic prowess, unlike the latter two countries, Singapore shows no sign of making a transition to democracy. Alternatively, as Indonesia's difficult political evolution over the past decade illustrates, the transition to democracy can be an unusually fraught affair, insofar as it has to proceed against the backdrop of an earlier lack of investment in political institutions and associated underdevelopment of the social expectations that support such institutions. How this process will play out for China ­ whether the latter will be stably authoritarian, as in Singapore, make a relatively smooth transition to democracy, as in Korea and Taiwan, or experience significant social dislocation, as in Indonesia ­ is one of the key global imponderables for the coming decades. Sequence II: Transformational Governance The second development sequence has as its entry point an effort to transform the country's political institutions by strengthening the rule of law and establishing democratic mechanisms for selecting a country's leaders. As noted earlier, the correlation between the quality of state institutions and per capita income is clear: When per capita income is high, the rule of law, democratic institutions and the public bureaucracy all work well. Further, in high income countries - 17 - these institutions are mutually reinforcing, with the checks and balances provided by democracy and the rule of law providing a corrective mechanism that helps keep the polity from veering too far astray, even in the face of dysfunctional political leadership. Thus both democracy and the rule of law serve as checks on a leadership that might, for example, seek to pervert the functioning of the bureaucracy for narrow political ends. Against that backdrop, consider a country that has been saddled endemically with factionalized elites, and self-seeking and unaccountable political leadership. Because rule-boundedness is weak, a poorly-functioning bureaucracy is likely to be part of this dysfunctional governance syndrome. Given these endemic weaknesses, the option of the country constructing a developmental state is not likely to be credible. A natural response of the country's citizens (and its developmental partners) is to focus on political institutions themselves. Figure 2 illustrates three distinct channels through which a transformation of political institutions hypothetically could catalyze a virtuous development spiral. The first channel [steps 2ai & 2aii in Figure 2] comprises the direct impact of a transformation of political institutions on credibility and legitimacy. This transformation could comprise one or both of a move to political democracy or a strengthening of the rule of law: · A move to democracy holds the prospect that elections can be a quick route to a radically transformed political leadership ­ with legitimacy, with a mandate to pursue developmentally- oriented policies, and accountable to citizens. (As discussed below, other outcomes are, of course, also plausible.) - 18 - · A stronger rule of law holds the prospect of introducing into the governance equation both clearer rules of the game, and impartial mechanisms for monitoring and enforcing compliance with the rules. Note, though, that under the best of circumstances, the task of building the capacity of a country's justice system is a long-term one. And note also that the incentives of incumbent elites to support the strengthening of these discretion- restraining institutions are likely to be mixed. For both, the hope is that institutional changes will enhance the accountability of a country's leadership, and reduce the potential for arbitrary, discretionary action ­ thereby shifting expectations and kick-starting growth. The second channel linking political institutions and growth is via state capacity (steps 2bi & 3). 3 Political transformation has the potential to radically improve both the incentives and the means for state capacity building. The improved incentives could follow from a shift from self-seeking to more accountable and hence public-good oriented leadership, leadership that would be more likely to invest in state capacity as a way of improving development performance. And improved means for state capacity building could come about because greater accountability and rule-boundedness are not only key intended outcomes of political transformation, they also are important requisites for a well-functioning public bureaucracy. As in the case of the developmental state, better public sector capacity can directly enhance growth. 3 The numbering of the channels signals the possibility that steps 2ai, 2bi and 2ci can all be directly initiated by political transformation ­ but experience suggests that actual gains in bureaucratic capacity and perceptions of fairness take time. - 19 - Figure 2: Institutional Transformation as an Entry Point for Development CIVIL SOCIETY GROWTH DEVELOPMENT 2aii 4 3 Credibility and 2ci legitimacy STEP 1: 2ai TRANSFORM BUILD STATE POLITICAL CAPACITY INSTITUTIONS 2bi The third channel linking political transformation and growth works via the development of civil society (steps 2ci and 4). As will be discussed further in the context of the fourth sequence, the transformation of political institutions and the development of civil society are strongly complementary. Insofar as many citizens had historically distrusted state authority even to the point of, at the limit, direct resistance, a transformation of citizen-state relations and the emergence of a rich array of civil society institutions could enhance the legitimacy of state institutions, thereby bolstering investor confidence ­ and hence growth. Note that both channels 2bi and 2ci (and also strengthened rule of law) affect growth only over the longer-term, and so are shown as dotted lines in the figure. Reformers looking to `transformational governance' as a quick-fix to kickstart growth must thus rely on early elections. If reformers hold and win these elections ­ and if the - 20 - elections indeed have a credibility-enhancing impact and kick-start growth -- the temptation could be strong to enjoy the fruits of these early success, with weakened momentum to address the longer-run institutional challenges. Later sections of the paper will consider this last variant of transformational governance in more depth. Experiences and challenges. To illustrate the relationship between political transformation and economic performance consider three divergent sets of experiences. The first comprises post-communist countries that have recently acceded to the European Union (or hope to do so in the relatively near future). As has been well documented, in at least three ways the prospect of joining the EU proved to be a powerful engine of reform. First, the perceived benefit of EU membership provided a powerful incentive for change. Second, the EU acquis and related political expectations provided a clear, explicit institutional model for accession countries to follow. Third, the EU connection provided a strong mechanism for signaling credibly both the direction and commitment of countries to continuing reform. The 2004 accession of 10 countries to the EU ­ and the anchor it provided for continuing, steady, economic performance ­ ranks as a major institutional success story globally over the past half century. A second group of countries that have used improved political institutions to transform governance are in Latin America. Democracy was of course more deeply rooted in Latin American than in other parts of the developing world, but suffered massive setbacks with the onset of military regimes in the mid-1960s. Virtually the whole hemisphere returned to democracy by the early 1990s, and the quality of democracy has, by various measures, been steadily improving. This is measurable by increasing rates of voting over the past three decades across the continent, but also through qualitative transformations by which power - 21 - has been pushed down to lower levels of the political hierarchy, giving citizens greater voice and participation. Brazil, Colombia, Peru, Venezueala, and Bolivia are among the countries that have undergone decentralization programs in recent years. The result of greater citizen participation at lower levels of government has in many cases increased the quality of governance. Bolivia's 1994 Law of Popular Participation, for example, has led to the reallocation of public resources in a pro-poor direction. In Colombia, directly elected mayors with increased municipal autonomy have engaged in a variety of innovative programs to control gangs and drugs. And in Brazil, traditional patronage politics has been curbed through innovative programs like Porto Alegre's participatory budgeting; according to one observer, Brazilian "voters have developed a habit of using the ballot box to punish mayors or governors who devoted a disproportionate share of their revenues to public employment rather than services or investment." (Reid, 2007, p.292) The third set of experiences comprises the democratizing reforms undertaken by 40 sub-Saharan African countries in the early 1990s. (Bratton and van de Walle, 1997) Africa's transition to democracy happened as comprehensively and as rapidly as that of Eastern Europe and the former Soviet Union. The experience of South Africa illustrates that even without the anchor of the European Union democratization can nonetheless provide a springboard for broader improvements in development performance, via each of the channels identified in Figure 2: · The transformation of expectations (steps 2ai-> 2aii) turned out to have been a slower process than many had hoped. To be sure, legitimacy was hugely enhanced via the shift from apartheid to democratic majority rule ­ in part because South Africa's - 22 - political transformation translated rapidly into a radical shift of the pattern of public spending in favor of the poor majority. But only after over a half-dozen post-apartheid years of among the best managed macro-economic policies in the developing world did markets overcome their fears of instability and re-rate South Africa's creditworthiness upwards. Growth accelerated slowly in the latter 1990s tgo just over 5% by 2007. · The impact of political transformation on state capacity (steps 2ai => 2bi =>3) has been less clear. South Africa's move to democracy created the opportunity for a new vision to motivate its public sector (which in the years prior to the transition to majority rule had gone from being the relentless implementers of the architecture of apartheid to become a cynical, demoralized and self-seeking bureaucracy) ­ but it also required the bureaucracy to transform itself from a bastion of white privilege into a public sector whose complexion resembled the country as a whole. Yet Thomas Carothers (2002) analysis suggests that the South African experience of democratization supporting development is something of an African outlier. He argues that `transition' to democracy need not be a seamless path to well-functioning, competitive political and market institutions but could settle into either of two less benign syndromes: "`Dominant-power politics', where countries `have limited but still real political space, some political contestation by opposition groups, and most of the basic institutional forms of democracy. Yet one political grouping dominates the system in such a way that there appears to be little prospect of alternation of power..... The state's main assets -- as a source of money, jobs, public information (via state media) ­ are gradually put in direct service of the ruling party. ...the judiciary is typically cowed....citizens tend to be disaffected from politics and cut off from significant political participation beyond voting." - 23 - Carothers notes that Africa's wave of democratization in the early 1990s ended up producing many dominant-power systems ­ some cases (e.g. Burkina Faso, Cameroon, Equatorial Guinea, Gabon, Mauritania and Tanzania) where a hitherto one-party state liberalized but with only limited subsequent political opening, and others (e.g. Zambia) where new regimes ended up in dominant-party structures. Note that, for all of its limitations insofar as the renewal (via partial democratization) of a dominant power system enhances state capacity, growth could nonetheless accelerate via the mechanisms summarized in Figure 1. "In `feckless pluralism', (the second syndrome) countries enjoy alternation of power between genuinely different groupings, but "democracy remains shallow and troubled. Political participation... extends little beyond voting. Political elites from all the major parties are perceived as corrupt, self-interested and ineffective. The alternation of power seems only to trade the country's problems back and forth from one hapless side to the other." Carothers notes that `in sub-Saharan Africa, alternation of power remains rare generally'. He reports feckless pluralism to be an especially common pattern in Latin America, and widespread also in the postcommunist world. The prospects of feckless pluralism generating growth via state capacity building appear bleaker than for the dominant power politics syndrome. As will become evident, though, it could nonetheless be growth-enhancing via the `just enough governance' sequence ­ to which we now turn. Sequence III: `Just Enough' Governance; Growth as Entry Point Both of the dynamic sequences considered to this point have had as their first step one or another kind of institutional reform. Though this focus on institutions is very different from the earlier pre-occupation of development practitioners with economic policy, both the institutional and economic approaches share a common presumption ­ namely that far-reaching reform was necessary to unlock development in - 24 - hitherto poorly-performing countries. Thus, both institutional and economic reformers generally were advocates of what were perceived as `best practice', optimal policies. In sharp contrast, recent empirical work on `growth accelerations' has demonstrated that far-reaching reforms ­ either institutional or economic ­ need not be necessary to kick-start growth. Careful empirical analysis by Hausmann, Pritchett and Rodrik (2004) of data worldwide identified over 80 growth accelerations over the 1950- 1990 period that lasted for eight or more years . They found that "The onset of economic growth does not require deep and extensive institutional reform ...Moderate changes in country-specific circumstances (policies and institutional arrangements, often interacting with the external environment), can produce discontinuous changes in economic performance...." (Rodrik, Analytic Growth intro, pp. 8-9, 17)..... Once growth is set into motion, it becomes easier to maintain a virtuous cycle with high growth and institutional transformation feeding on each other." (Rodrik, Getting Institutions Right, 2004, p.10) Consistent with this pattern, Figure 3 thus delineates a third development sequence ­ one where the first step is the initiation of growth itself. - 25 - Figure 3: Sustaining growth through incremental institutional reforms STEP 1: CIVIL SOCIETY GROWTH DEVELOPMENT 2iii 3 2i 2ii POLITICAL BUILD STATE INSTITUTIONS CAPACITY Steps 2i-2iii in Figure 3 illustrate how a `just enough governance' development strategy `virtuous cycle' might evolve incrementally. These steps are depicted by dotted lines, here intended to signify that in a strategy of `just enough governance' sustaining growth remains the primary objective. As growth proceeds, though, one or another institutional constraint might threaten to short-circuit expansion ­ perhaps weaknesses in the delivery of infrastructure or key public services, perhaps a rise in corruption as public officials seek their share of the growing economic pie, perhaps rising social alienation with a growing sense on the part of citizens that government doesn't care about their everyday problems, perhaps the need for more sophisticated laws and institutions to underpin an increasingly sophisticated economy. With a `just enough governance' strategy, the goal is not to anticipate and address in advance all possible institutional constraints. Rather, the focus is on addressing specific capacity and institutional constraints as and when they become binding. - 26 - Sustaining growth thus becomes something of a `high-wire' act ­ continual crisis management, endlessly putting out fires in an environment which to the casual observer seems quite dysfunctional, but nonetheless defies the odds by sustaining continuing dynamism. Note that the `just enough governance' sequence should not be construed as a relatively painless (if nerve-wracking) development strategy. For one thing, there are multiple countries where the economic environment is too dysfunctional, and institutions are too weak, for incremental reforms to be enough to kickstart growth. For another, not all binding constraints can necessarily be eased through incremental measures. Moreover, as growth proceeds, society changes profoundly ­ with new pressures, and new opportunities building up. At some point channel 3 (identical in Figures 1 and 3) is likely to become dominant ­ laying the stage for a non-incremental transformation of political institutions. Experiences and challenges. The `Bangladesh paradox' offers a vivid example of the relevance of a `just enough governance' development strategy. Over the past decade, Bangladesh has made major gains in development performance. In the three decades since the country's independence in the early 1970s, per capita income has more than doubled; the poverty rate has fallen from 70 to 40 percent; child immunization rates have risen from negligible levels to close to 80 peercent; life expectancy at birth has gone from 45 to 63 years. (World Bank, 2006) Yet over the same period, the country also rated consistently as among the most corrupt in the world: Other governance indicators also rated poorly. Key to Bangladesh's paradoxical achievement were a series of `just enough' development reforms: good macro-economic management, targeted trade policy reforms which provided just enough openness to enable a take-off of the garment export sector - 27 - (and similarly-focused policies which facilitated take-offs in other specific sectors) 4 ­ plus strong civic engagement via an unusually strong role for the country's well-developed non-governmental institutions in the provision of public services. This strong overall performance has occurred against the backdrop of continuing crisis ­ which included, in the twelve months prior to completion (in early 2008) of an initial draft of this paper, a chaotic lead-in to a presidential election, subsequently aborted by a de facto military coup; the arrest on corruption charges of the country's two leading politicians; and repeated ambiguity as to when and how the military authorities will restore democracy, resulting in a new round of street protests. Throughout, growth has continued. A similar combination of seemingly chaotic governance and improving economic performance is evident in other countries. Consider Thailand, which since the 1980s has been one of the exemplars of the East Asia Miracle ­ but which never (its strong Ministry of Finance aside) exhibited the same levels of state capacity as, say, Korea and Taiwan, and which over the past decade has traversed a winding path from military rule to far-reaching constitutional reform, rising corruption and populism, a reversion to military rule, followed by another round of constitutional debate. The 1997 East Asian financial crisis temporary halted Thailand's almost two-decade-long growth surge, but (despite political and economic turbulence) by 2002 it had picked up again to an annual rate above 5% where it has remained. 4 Two examples which resulted in major gains in rural incomes comprise the introduction of a winter rice crop, and expansion in aquaculture. Remittances comprised another key source of economic growth. - 28 - In the short-term, too, Kenya seemed to illustrate the potency of `just enough governance'. In 2001, the country witnessed the electoral defeat of a corrupt long-term incumbent, Daniel Arap Moi, and the promise of far-reaching constitutional reform. Instead, constitutional reform stalled ­ and, within three years of the exit of Moi, the country was subjected to a new round of high-profile allegations of corruption against the successor government. Nonetheless, the ouster of Moi proved sufficient to trigger improved economic performance, with growth -- which had stubbornly remained below 3% prior to 2003 -- accelerating to 6% by 2006. Yet the turnaround seemingly was short- lived, with renewed political turbulence overtaking the country in the aftermath of bitter disputes over who won a 2008 presidential election. The Bangladesh, Kenya and Thailand examples point to a puzzle vis-à- vis a `just enough governance' development strategy: what is the long- run trend ­ and what is the short-run `noise'? Is the trend sustainable development -- with governance weakness the `noise' in the system? Or is the seemingly strong growth simply a short-term bubble ­ with governance turbulence signaling the hazards ahead? Certainly, HPRs empirical analysis confirms that many `growth accelerations' are followed by subsequent reversals. Indeed, natural resource driven boomlets in particular are notorious for the capacity destruction they can inflict on fragile public sectors ­ as the prospect of huge rents induces a country's elites to undermine already fragile institutions of accountability. Yet for all the caveats, growth in each of the three examples signals that, at least in some circumstances ­ those where the underlying growth drivers are strong and/or capacity and institutional constraints are not overwhelmingly constraining -- `just enough governance' would appear to be a viable short-term development - 29 - strategy. A complex economy requires complex institutions ­ so at some point, the longer term constraints will need to be addressed. Even so, as Figure 3 suggests, the longer a `just enough governance' strategy can be sustained, the broader is likely to be the constituency with a stake in stronger institutions, and hence the better may be the prospects for more far-reaching institutional reform. Sequence IV: Bottom-up Development; Civil Society as Entry Point There are many cases in which virtually all channels except for civil society are blocked: there is little or no economic growth; state capacity is weak and government corrupt; democracy and rule of law are either non-existent or not readily fixable because political power is in the hands of actors with no desire to change the status quo. In this case, the primary driver of development will be the mobilization of civil society, which will increase demands for greater democracy and rule of law, as well as a state that can deliver basic public services. The effect of this sequence on economic growth is often indirect. Civil society often demands political representation rather than growth in the first instance, while authoritarian or nominally democratic rulers oftentimes hope they can buy off potential opponents through their ability to deliver growth. It might be tempting to think about bottom-up development as a kind of residual strategy to be pursued only when all other channels are blocked. (This is the way that it appears in the decision tree at the end of the article.) In fact, promoting the development of civil society is an end in itself, and can be a critical complementary component to several of the other strategies. This is clearest in the case of the transformational governance sequence. While some forms of democracy promotion involve restructuring democratic and rule of law - 30 - institutions through constitutional change, or strengthening formal institutions like court systems or political parties, most practitioners in the democracy promotion field devote considerable energy to the development of a vigorous civil society ­ labor unions, watchdog groups, business roundtables, media organizations, and the like ­ because they are assumed to be part of a broader liberal democratic political order. Experiences and Challenges. There are a number of cases where bottom-up development of civil society was the primary route to development. For example, Poland during the 1980s was a communist country that had a nascent civil society, in the form of an independent trade union (Solidarity), private agriculture, and the Catholic Church. Neither rapid economic growth nor top-down political change appeared possible, but in 1989, the thawing of Cold War relationships suddenly made possible a dramatic breakthrough to liberal democracy and a market economy. Poland had a considerable advantage over other post-communist states in making this transition because it had social actors outside the party nomenklatura that could fill the ensuing political vacuum. Ukraine was arguably another country in a situation where all channels but the civil society route were blocked in the first years of the 21st century. Ukraine had nominally democratic political institutions, but rule of law was very weak, and the political process heavily manipulated behind the scenes by various oligarchs and shadowy economic actors. All branches of the state lacked capacity and corruption was rampant. Ukraine's formal institutions had received considerable external support from outside donors during the 1990s, very little of which had any evident impact on their performance. While economic growth had resumed by 2002, it was due less to good - 31 - economic policy than to favorable external conditions which affected virtually all countries in the global economy in that period. Under these circumstances, political change occurred not through reform of existing institutions, but by mobilization of civil society that put pressure on the government to open up the political process and make itself more genuinely accountable. The Orange Revolution of December 2004 occurred when the incumbent president sought to manipulate the vote for a new president, a fraud that was documented by election observers and then publicized by Ukraine's nascent independent media. Civil society groups were sufficiently mobilized to protest these results and put enough pressure on the government to force a second election, which brought the Orange coalition to power. Figure 4: Bottom-Up Reform STEP 1: DEVELOP GROWTH CIVIL SOCIETY Credibility and 2aii legitimacy 2i 2ii BUILD STATE CAPACITY POLITICAL INSTITUTIONS Though the dotted lines in Figure 4 suggest how civil society development leading to bottom-up democratization might support growth, in practice the economic consequences are not clear, at least in the - 32 - short run, any more than they are for transformational governance. Ukraine has done well economically for most of the period since the Orange Revolution, but that improvement in performance was driven by external factors and likely would have occurred even in the absence of democratic change. Nor is it clear that state capacity has increased, as levels of corruption remain high. On the other hand, Ukraine today has a freer and more open media than neighboring Russia, something one presumes would benefit government accountability in the long run. How the Sequences Inter-relate The four sequences described above are conceptual constructs ­ `ideal types' used to bring analytical order to messy, multi-faceted reality. The aim is to highlight some key features of country-level processes, not to suggest that these features capture the whole of a country's development evolution. On the contrary, there are likely be variations over time as to which of the four sequences best characterizes a country's development dynamics. And, at any point in time the reality may be a hybrid of more than one sequence. Consider how conditions change over time. As already noted, the `state capacity building' sequence seems highly likely to evolve over time into something different, insofar as its success creates new economic actors that seek more complex and open economic and political institutions than a state-centric model of development can provide. Similarly, a `just enough governance' trajectory is likely to be temporary, either because, as with state capacity building, its success generates demand for more robust institutions, or because it comes up against an institutional constraint that cannot be eased incrementally. Only the `transformational governance' sequence offers (in theory) the prospect of long-term institutional stability ­ although this - 33 - presupposes unusually far-reaching and effective institutional reforms up-front. Now consider the extent to which the four sequences indeed are distinct from one another. The differences between state capacity building and just enough governance are stark: the former is top-down, tightly-controlled and ­ insofar as it is implemented effectively ­ highly orderly; the latter is haphazard, seemingly chaotic. The strengths of one are the weaknesses of the other. Only in settings in which a broad range of institutions are already mature might they be perceived to converge ­ though, at that point, neither `state capacity building, nor `just enough governance' accurately describe the realities on the ground. Overlap between the `transformational governance' and `bottom-up reform' sequences on the one hand, and the other two sequences on the other is best understood through the lens of Carothers' two partial variants of institutional transformation introduced earlier. Insofar as transformation of political institutions and of civil society results in the de facto dominance of political actors who perceive themselves to enjoy stable incumbency and take a long-term view ­ i.e. Carothers' dominant power politics variant -- the result could approximate quite closely the `state capacity building' sequence. But insofar as the democratic and civil society transformations are more turbulent ­ i.e. look more like feckless pluralism -- they might nonetheless unlock a dysfunctional equilibrium. This could help unleash quite rapid economic growth, with the resulting chaotic dynamism resembling quite closely the `just enough governance' sequence. This last set of overlaps suggests the following speculation as to the relevance of `transformational governance' in at least some low- - 34 - income developing countries. As Carothers highlights, often the outcome of efforts to foster `transformational governance' falls short of initial ambitious intentions. But might it be too negative to describe the results as `dysfunctional'? Perhaps in many low-income countries transformational governance might better be viewed as catalytic ­ as a (risky) path to the `state capacity building' or `just enough governance' dynamic sequences. Perhaps, in such countries, economic growth, combined with a seeming excess of order or a seeming excess of chaos may be in the (medium-term) nature of things, rather than an aberration that requires `fixing'. IV: THE `SEQUENCES FRAMEWORK' AND DEVELOPMENT DECISION-MAKING This final section explores how the sequences framework, laid out in Section III can aid development decision-making. A key underlying issue is the balance between choice and constraint: how much scope do policymakers have to choose which route a country pursues ­ and how much is historically conditioned, is path dependent? To motivate this question consider the illustrative examples of Ethiopia, Korea and China. 5 Over the past four decades, Ethiopia has lurched from semi-feudal monarchy, to hard-line communist, to market-oriented modernizer. Yet throughout, the seemingly transformed polity has been highly centralized, building on the inherited foundations of centuries of top- 5 The use of these three examples should not be taken to suggest that low-income countries generally thrive by beginning first with the state capacity building sequence. That hypothesis does not account for Bangladesh's success with `just enough governance', or for the kickstart that transformational governance provided to Kenya. Indeed, systematic data analysis points to a wide diversity of institutional starting points among low-income countries. See the appendix "Applying the Indicators ­ a Typology of Countries" in Levy (2007). - 35 - down monarchical control, and correspondingly profoundly hierarchical social relations. Arguably, path dependence accounts for the repeated turn of Ethiopia's political leaders to state capacity building as the country's preferred development strategy. Through to the mid-1980s Korea also was an exemplar of a top-down developmental state, built (as in Ethiopia) on a centuries-long history of centralized control. (Henderson, 1968) Yet over the past two decades, Korea increasingly has broken from its top-down, state-centric legacy and transited into an open, democratic society, governed by increasingly robust checks and balances institutions. Protagonists of deterministic path dependence might interpret Korea's transformation less as a result of determined political leadership than as endogenous adaptation, a consequence of economic success. But there is no getting away from the role of leadership in accounting for China's astonishing evolution over the past seventy years ­ from the collapse of warlordism in the face of a communist-led peasant revolution, to failed Maoist experimentation, to Deng Xiaoping's 1978 exhortation that `to get rich is glorious', which catalyzed a globally unprecedented growth surge ­ all, of course, within a broader frame of robust, deeply rooted central state capacity. Certainly, as these three examples suggest, the sequences framework laid out in Section III should not be viewed as a technocratic toolkit from which a putative reformer is free to choose and implement his preferred option. On the contrary, to a significant ­ but not wholly determined ­ extent, choice is constrained, with country-specific history and politics shaping the trajectory of change. Recognizing that history constrains options, what is the scope for maneuver, for initiating or sustaining cumulative processes that can move development forward? The final subsection of this paper explores - 36 - how the broad perspective provided by the sequences framework can help address this question in a way that supports more skillful development decision-making. Making Choices: Governance Reforms as Development Strategy Figure 4 suggests a decision tree to help clarify how priorities for development reform might vary depending on specific country circumstances. The first fork in the decision tree distinguishes among countries according to their current development performance. Is a country's current, short-term development trajectory a postive one ­ in which case the challenge is to sustain an ongoing process? Or is the challenge to kickstart development from a more `stuck' place? For countries already on a dynamic path, the key point of departure should be to sustain existing momentum, on the principle that "if it ain't broke, don't fix it". Following this admonition is not as easy as it sounds. For reasons that should now be clear, the governance realities even in the most dynamic of low-income countries often will be profoundly unsettling: perhaps an underside of weak accountability alongside a dominant bureaucracy; perhaps a panoply of dysfunctions, each threatening imminently to short-circuit dysfunction. Certainly, over the medium-term the weaknesses will need to be addressed ­ but an excessive, too-rapid response could itself risk short-circuiting the momentum already underway. Country-specific political economy analysis will be key to assessing where the balance of risk lies as between doing too much or too little to address continuing weaknesses. In deciding how to proceed, a first step might be to explore the extent to which endogenous processes already are inducing reforms of lagging institutions ­ whether, for example, countries moving along a state capacity building trajectory also are - 37 - Figure 5: Country Circumstances and Development Sequences ­ A Decision Tree - 38 - beginning to invest in strengthening checks and balances institutions. If this is not happening, development reformers might usefully initiate work on the lagging governance dimensions ­ proceeding incrementally so as not to kill the goose currently laying golden eggs. The second major branch of Figure 5 lays out options for low- income countries stuck with stagnant economies. A key judgment underlies reform choices vis-à-vis this group of countries: Should the agenda for reform push institutional change to the maximum extent feasible ­ or seek, rather, the minimum changes necessary to kickstart dynamism in a specific country setting? Comparing the four development sequences, transformational governance generally calls for the most far-reaching up-front changes, and just-enough governance the least far-reaching, with state capacity building and civil society development (depending how it is approached) somewhere in between. The decision tree in Figure 5 is constructed on the assumption that the preferred option generally is the least-disruptive one that is capable of unlocking dynamism, given a country's specific circumstances. This assumption aligns well with the sequences framework, which underscores that causation is cumulative, and that a well-chosen first step can bring a cascade of dynamism in its wake. (The assumption is contestable, of course; others might prefer rather to `maximize' reforms while a `window of opportunity' is open. The judgments always are country-specific; there can be no fixed formula.) Thus, for low-income countries mired in stagnation, a key initial question is whether or not incremental reforms can unlock dynamism. 6 As the decision tree suggests, `just enough governance' is the obvious way 6 The `binding constraints' approach to reform, proposed by Hausmann, Rodrik and Velasco (2006) offers a promising analytical starting point for addressing this question. - 39 - forward in countries whose political economy offers only scope for reform on the margin ­ and where incrementalism can work. If both incremental and non-incremental options are politically feasible, the choice is more contestable. For stagnant countries where incremental reforms are unlikely to unlock dynamism, the relevant questions concern the incentives and constraints of political leaders. The decision tree suggests two options for reformist leaders. Where they enjoy the legitimacy and authority ­ as well as desire ­ to undertake far-reaching pro- development reforms, the least institutionally disruptive option would be to focus on state capacity building and, perhaps, also associated efforts to foster far-reaching economic reform. But as the `no' branch of this last fork in the decision-tree suggests, even willing leaders may not enjoy the requisite authority and legitimacy. In such settings, `transformational governance', including actively fostering the development of civil society, comes onto the agenda as an entry point capable of kickstarting development dynamism. In practice, of course, political leaders seeking to kickstart development via an institutional reform sequence (of either the state capacity building or transformational variety) confront a more complex calculus ­ and more of a continuum of options -- than the heuristic framework of Figure 5 suggests. How boldly transformational should their institutional reform efforts be? And for how long? Even if leaders enjoy the legitimacy needed for state capacity building and associated bold economic reform, might there nonetheless be a case for a bolder, more transformative approach to governance to `lock-in' change while a window of opportunity is open? Conversely, even if legitimacy seems to be lacking, leaders might choose to push ahead - 40 - along the state capacity building/economic reform branch in the hope that success will yield political dividends down the road. Careful political economy assessments of the interests and likely responses of influential social actors can help support decision- making. Also key will be a readiness to adapt flexibly as the reform strategy unfolds. Indeed, as suggested earlier, even where a country starts out with a bold strategy of institutional change, it may turn out to be sustained only to the point that the economy accelerates ­ a de facto reversion to `just enough governance', and hence a less far- reaching departure from the country's historical (path dependent) trajectory than might initially have been intended. The final fork in the decision tree comprises countries that are stuck deep in a pit of dysfunction ­ with a stagnant economy, and decaying (or even, at the limit, collapsing) institutions, fragmented authority, and non-developmental leadership. In such settings, the decision-tree suggests that civil society development emerges as the preferred entry point. For external actors, with less directly at stake, the relevant approaches generally should be appropriately modest and developmentally-oriented. Interventions (typically donor- supported)to foster civil society by channeling resources directly to communities, and engaging service users and communities in overseeing how the resources are used, can be quite effective in helping to alleviate poverty. They might also plant valuable seeds of social learning as to different ways of handling relationships between a country's elites and its broader populace. But history suggests that such modest approaches are unlikely to unlock developmental dynamism in profoundly dysfunctional settings. In these latter settings, far- reaching domestically-driven social and political mobilization has proven decisive. Examples include: Ataturk's modernization of Turkey; - 41 - China's communist-led peasant revolution; Korea's military government; colonial independence movements; the fall of President Marcos; the collapse of communism in Eastern Europe and the former Soviet Union; South Africa's transformation from apartheid to democracy; the `yellow' `orange' and `rose' revolutions in Georgia, Ukraine and the Kyrgyz Republic etc etc. Some of these aimed to foster development via a state capacity building sequence; others via transformational governance. Consideration of the drivers of far-reaching social and political mobilization ­ and its divergent trajectories, prospects and risks -- , falls outside the scope of the present paper. Concluding Comments The past fifteen years have seen an explosion of interest among both scholars and practitioners in governance, and its link to long-run development performance. Our goal in this paper has been to contribute to this discourse by laying out a framework that highlights the interactions between governance and growth, offers a broad view of the diverse strategic choices available to development decision-makers, and helps clarify how the preferred choice might be conditioned by a country's unique historical circumstances. We identify four distinctive dynamic sequences, which differ from one another both in their points of entry and, more broadly, in `what comes before what' as development unfolds. We explore how this `sequences framework' can help guide choice, given the assumption that the preferred option generally is the least disruptive one that is capable of unlocking dynamism, given a country's specific circumstances. Other assumptions also are plausible. And even with our assumption, there is much room for disagreement as to a country's preferred development strategy ­ in part because the - 42 - dynamic sequences are heuristic constructs, with the reality likely to be a hybrid of more than one sequence. Our hope is that by making explicit multiple dynamic sequences -- and how decision-makers choices among them might vary according to a combination of country-specific empirical realities and preferences as between incremental and non-incremental change ­ we have helped lay the groundwork for a new generation of empirical work on the relationship between governance and development. We are aware that some readers will find this paper excessively sweeping, insufficiently anchored empirically, simplistic. But we worry less about these risks than the alternative that seems all too common in the current discourse: partial approaches; `cookie-cutter' recommendations that fail to account for countries' radically disparate circumstances; inattention to dynamic processes, to cumulative causation. - 43 - Bibliography Acemoglu, Daron and James A.Robinson, 2006. Economic Origins of Dictatorship and Democracy: Economic and Political Origins (Cambridge: Cambridge University Press). Acemoglu, Daron, Robinson, James A. et. al., 2001. "The Colonial Origins of Comparative Development: An Empirical Investigation," American Economic Review 91(5): 1369-401. Barro, Robert J., 1997. Determinants of Economic Growth: A Cross- Country Survey (Cambridge: MIT Press). Bates, Robert H, 2008. When Things Fell Apart: State Failure in Late- Century Africa, (New York: Cambridge University Press). Bratton, Michael and van de Walle, Nicolas, 1997. Democratic Experiments in Africa: Regime Transitions in Comparative Perspective (Cambridge: Cambridge University Press) Carothers, Thomas, 2007. "The "Sequencing" Fallacy," Journal of Democracy 18(1): 12-27. Carothers, Thomas, 2002. "The End of the Transition Paradigm," Journal of Democracy 13(1): 5-21. Collier, Paul, 2007. The Bottom Billion: Why the Poorest Countries are Failing and What Can Be Done About It (Oxford: Oxford University Press). Diamond, Larry, 1992. "Economic Development and Democracy Reconsidered," American Behavioral Scientist 15(4-5): 450-499. Diamond, Larry, 2008. The Spirit of Democracy: The Struggle to Build Free Societies Throughout the World (New York: Times Books). Easterly, William R., 2001. The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics (Cambridge, MA: MIT Press). Easterly, William R., 2006. The White Man's Burden: Why the West's Efforts to Aid the Rest Have Done So Much Ill and So Little Good (New York: Penguin Press). Fukuyama, Francis, 2007. "Development and the Limits of Institutional Design," in Natalia Dinello and Vladimir Popov, eds., Political Institutions and Development: Failed Expectations and Renewed Hopes (Cheltenham, UK: Edward Elgar). Gellner, Ernest, 1994. Conditions of Liberty: Civil Society and Its Rivals (London: Hamish Hamilton). Haber, Stephen, Armando Razo, et. al., 2003. The Politics of Property Rights (Cambridge: Cambridge University Press). - 44 - Haggard, Stephan, 1990. Pathways from the Periphery: The Politics of Growth in Newly Industrializing Countries. (Ithaca: Cornell University Press) Hausmann, Ricardo, Lant Pritchett and Dani Rodrik, 2004. "Growth Accelertions" National Bureau of Economic Research Working Paper Number 10566. Henderson, Gregory, 1968. Korea: The Politics of the Vortex (Cambridge: Harvard University Press) Huntington, Samuel P., 1968. Political Order in Changing Societies (New Haven: Yale University Press). Khan, Mushtaq H. and Kwame Sundaram Jomo, eds., 2000. Rents, Rent- Seeking and Economic Development: Theory and Evidence in Asia (Cambridge: Cambridge University Press). Levy, Brian, 2007. Governance Reform: Bridging Monitoring and Action, (Washington, DC: World Bank) Lipset, Seymour Martin, 1959. "Some Social Requisites of Democracy: Economic Development and Political Legitimacy," American Political Science Review 53: 69-105. Plattner, Marc F., 1999. "From Liberalism to Illiberal Democray," Journal of Democracy 10: 121-134. Pzeworski, Adam, 1993. Przeworski, Adam, Alvarez, Michael et. al., 1996. "What Makes Democracies Endure?" Journal of Democracy 7(1): 39-55. Reid, Michael, 2007. Forgotten Continent: The Battle for Latin America's Soul (New Haven, CT: Yale University Press). Sachs, Jeffrey and McArthur, John W., 2001. Institutions and Geography: Comment on Acemoglu, Johnson, and Robinson (2000) (Cambridge, MA: NBER Working Paper 8114). -- and Warner, Andrew, 1995. Natural Resource Abundance and Economic Growth (: NBER Working Paper 5398). Schmitter, Philippe C., Guillermo O'Donnell, and Lawrence Whitehead, eds., 1986. Transitions from Authoritarian Rule: Tentative Conclusions About Uncertain Democracies (Baltimore, MD: Johns Hopkins University Press). World Bank 2005. Equity and Development, World Development Report, 2006 (Washington, DC: World Bank and Oxford University Press) World Bank, 2006. BANGLADESH Zakaria, Fareed, 2003. The Future of Freedom: Illiberal Democracy at Home and Abroad (New York: W. W. Norton).