New Political Economy, Vol. 7, No.

2, 2002

DEBATE: DEMOCRACY AND DEVELOPMENT A Contradiction in the Politics of Economics

ADRIAN LEFTWICH
Insofar as they are analytically isolable from other social processes, all political systems are best understood as systems of power in that they generate, legitimise, distribute and seek to control (political) power in different ways. Likewise, economic systems are best understood as systems of wealth in that they, too, generate, legitimise, distribute and seek to control wealth in different ways. In practice, of course, systems of power and wealth overlap and, in doing so, coincide, complement, balance or con ict with each other, or a mix of these. From the point of view of achieving social stability, the macropolitical trick is to maintain some kind of equilibrium between them. For stability can be threatened where wealth and power are concentrated in the same hands (as in pre-revolutionary France or Russia and in much of Latin America during the twentieth century), thus provoking pervasive popular opposition. Equally, stability can be threatened where wealth and power are concentrated in different hands, provoking the politics of stand-off or worse (as in Malaysia in the 1960s and Nicaragua in the 1980s), or in the barely suppressed tension between wealth and power in postapartheid South Africa. It is this relation between systems of power and systems of wealth, crudely stated, and its implications for the political economy of development, which is my concern, plus the associated imperative of bringing political science to the core of development studies, and vice versa. 1 However uncomfortable a thesis it may be, I argue that democracy (especially, but not only, in its more limited but almost universally practised representative and Schumpeterian form2) is a conservative system of power. Democracy (at least in its liberal form) is also, of course, radical in that no other political systems have promoted and protected individual political rights and civil liberties to the same extent.3 In their struggles to de ne, win or protect such rights in the political domain, countless millions of people have died or suffered appallingly at the hands of authoritarian regimes. These are the human dramas which illustrate so powerfully the narrative of the struggle for democratisation from the late 18th century to the present. This should never be underestimated. Whatever the virtues of their radical properties, stable and hence enduring democracies are conservative systems of power. As Przeworski observes, social and economic conservatism may be the necessary price for democracy.4 There are at least two important respects in which this is the case. First, while democracies have established the principle and practice of at least some civil and political rights in

Adrian Leftwich, Department of Politics, University of York, Heslington, York YO1 5DD, UK.
ISSN 1356-346 7 print; ISSN 1469-9923 online/02/020269-1 3 DOI: 10.1080/1356346022013887 1 2002 Taylor & Francis Ltd

269

Adrian Leftwich
their political systems (though not all, such as gender equality, have been fully implemented), they have not, to the same extent, been able to de ne, agree on or institutionalise social and economic rights (although the record of some democracies, especially social democracies, in these matters is generally better than most authoritarian regimes). Such rights (which the Chinese, at least before the end of the iron rice bowl era, called subsistence rights) include job security or universal and equal access to health and welfare systems and adequate pension. There are, again, many reasons for this, but as a general rule it is in the nature of governments in democracies to tread warily when seeking to extend their political power in the public domain into the private domain of the system of wealth which would inevitably be required to foot the bill. This is a point to which I will need to return substantively when discussing the developmental implications of democracies or the prospects for developmental democracy. Democracy may be considered conservative in a second, and perhaps more troubling, sense, at least from a developmental point of view. Democracies have great dif culty in taking rapid and farreaching steps to reduce structural inequalities in wealth which new democratic governments (as in post-apartheid South Africa) or born-again (and again) democratic governments (as in Brazil or the Philippines) may have inherited, whether they be based on class, colour, ethnicity, religion or a combination of them. In short, and uncomfortably, the system of power which democratic politics represents seldom promotes the politics of radical change in the system of wealth, even though this may be vital for establishing developmental momentum, especially in late developing societies. Democracies, that is, have few of the characteristics of what David Apter long ago referred to as mobilization systems, as opposed to reconciliation systems5 and this is one of the key reasons why the relationship between democracy and development is so problematic and often tense. For consolidated democratic politics is characteristically the politics of accommodation, compromise and the centre; and its political logic is generally therefore necessarily consensual, conservative and incremental in the change it brings about. For many that is its virtue: for others, its vice. Why should democracy be a conservative system of power? To answer this is to remember the complex structural conditions which underpin and, indeed, de ne democratic politics. There are many, including, rst, a variety of forms of legitimacygeographic, constitutional and political. Geographical legitimacy involves citizens accepting their physical place in the territorial state in which they live (which Chechens, Basques, Sri Lankan Tamils and Southern Sudanese, for example, do not) and not wanting to secede or establish irredentist movements. Likewise, constitutional legitimacy requires acceptance of the political system, that is the constitutional provisions that provide a durable set of rules for the political game and that distribute and control power in acceptable and hence legitimate ways. These are intensely dif cult to establish, as the complex constitutional bargaining in South Africa after 1991 illustrated, andif not suf ciently acceptablecan easily produce democratic breakdown (often repeated), as in Ulster, Fiji, many parts of postcolonial sub-Saharan Africa, the Lebanon or countless other societies. Finally, political legitimacy entails agreement that the outcome of the political game, according to the operation of its rules, is and remains fair. The absence of such agreement brought about the endgame for Mr Marcos in the Philippines in 1986 and Mr Milosovic in Serbia in 2000, and deep doubts about fairness of the 1992 election brought uneasy postelectoral politics for President Moi in Kenya. Such multifaceted legitimacy has never been easy to establish or sustain. Commonly, one or more of its elements (geographic, constitutional or political) has been missing, de cient or just plain weak.

270

Debate
Moreover, such legitimacy has proved to be very dif cult (though not impossible, as in India, though even there it is highly conditional and under both religious and regional threat) to institutionalise in societies where poverty is gross and pervasive, where wealth and income inequality is not only bruising but also not declining, and where the social order is sharply striated by ethnic, religious or racial cleavages.6 Already, from these few infrastructural preconditions for democracy, it can be seen why and how democratic politics veer towards procedural conservatism. Two further intimately related and balancing binary operational conditions of democratic politics make it almost certain that democracy will normally be a conservative system of power. These operational conditions are rst, and most simply, that losers must accept the outcome of the political game, knowing that they have the right to come again, within a given and constitutionally stipulated period of time (usually in a range from two to six years time). Losers cannot return to the gun, the foco or the bomb (if that is what they did before democratisation). The distinguished American political scientist, Adam Przeworski, has theorised democratisation as a process of institutionalizing uncertainty, of subjecting all interests to uncertainty.7 By de nition, no-one can know the outcomes of electoral politics but, while this may be true at some level of theory, it is seldom true in practice, for it is highly unlikely that any group or party would come to accept the rules of the electoral game if it knew that losing meant that it or the interests it represented would lose too much. In short, no group or interest would enter the democratic political game if it believed that one possible outcome of electoral defeat would result in its effective elimination, politically or otherwise. Thus the quid pro quo for the rst operational condition (losers acceptance of defeat) constitutes the second of the two balancing binary operational conditions for democratic endurance, namely that winners must exercise restraint. Democratic electoral victory is not the same as a licence for the winners to undermine, attack or eliminate the vital interests or resources of the losers; on the contrary, there are signi cant limits to what they can do with their newly-won power. This condition means that sustained and sustainable democratic politics thus depends on victorious parties exercising restraint when in government, although the temptation (and sometimes the developmental or egalitarian need) is often to rewrite the policy book. That is to say, new or born-again democracies are more likely to consolidate and prosper if the new government does not pursue highly contentious policies too far or too fast, especially where these policies seriously threaten other major interests. Indeed, such agreed limits on policy change are often established before democratisation can be completed, in the course of negotiations about the rules of the game, and are thus part of that process itself.8 The case of Venezuela aptly illustrates this when, in 1958, two extraordinary pacts (the WorkerOwner Accord and the Pact of Punto Fijo) were concluded between the three main parties, excluding the Communist Party.9 These pacts (pacted democracy) framed the directions and limits of policy change and effectively tied the major parties into a consensus on the broad limits of developmental policy choice and provided for a sharing of power which, for almost 30 years, sustained democracy in Venezuela, something rare in Latin America during the period. They did so by guaranteeing that the main parties would all have a stake in the government and that neither they nor their supporters would ever lose too much through electoral defeat. The problem is that development is both by de nition and in practice a radical and commonly turbulent process that is concerned with often far-reaching and rapid change in the structure and use of wealth, and whichif successfulmust transform it. But the prospects of combining such transformation in the structure of wealth

271

Adrian Leftwich
with democracy are slim. For in laying the foundations for, or re-starting, development (certainly the kind of rapid and catching-up development so urgently needed in so many societies), it is inevitable that non-consensual steps will have to be taken, especially where a new developmentally committed regime comes to power facing a legacy of immense inequality in the structure of wealth and opportunity (illustrated dramatically by the cases of South Africa or the Philippines after the fall of Marcos). Land reform is a good example of the kind of non-consensual step often necessary, since it is widely recognised that this can be an important condition for both rural and industrial development. But landowners in general do not consent to land reform! As in Latin America and Asia, they have often constituted a very powerful interest with intimate connections to the dominant parties and the state. In consequence, Third World democracies have seldom been effective in overcoming such vested rural interests to achieve the restructuring of both rural wealth and power which land reform is designed to bring about. Land reform failed in the Philippines after the restoration of democracy in 1986,10 and proved impossible in Pakistan under the Bhutto regime in the 1970s.11 Indian democracy, too, has had very little success in pushing through national land reforms (though some states, such as Kerala, have been more successful). Moreover, at the level of more general redistributive policy and practice in India, there has been hardly any signi cant taxation of agricultural income and wealth.12 Indeed, the general attempt in India under the dominance of Congress governments for almost 30 years to organise development of what was in effect a continental political economy, more empire than nation, was undertaken with one arm tied behind its back by its commitment to liberal democracy, as Herring puts it,13 producing relatively limpingly slow development as re ected in what Raj Krishna once described as the Hindu rate of growth.14 In conclusion, under most circumstances, the rules and operational conditions of stable democratic politics will tend to restrict policy to incremental and accommodationist options. On the other hand, developmental requirements (whether liberal or radical) will be likely to pull politics and policy in the direction of quite sharp (and, for some, unpleasant) changes affecting the structure of wealth of the society and hence important interests within it. It is this structural contradiction between the conservative requirements of stable democratic survival and the urgent transformative imperatives of late development which makes the combination of democracy and development so dif cult, and which makes the establishment and continuity of democratic developmental states so rare. For these reasons, no-one should hold their breath for an imminent great leap forward in development performance in the Third World, following contemporary western insistence on democratisation as a condition for growth (and aid). Effective development will not depend on regime type, but on the character of the state, whether democratic or not. But that is a different story. Notes
1. 2. 3. Adrian Leftwich, States of Development (Polity Press, 2000). J.A. Schumpeter, Capitalism, Socialism and Democracy (Unwin, 1965), p. 269. R.G. Gastil, Freedom in the World, 19856 (Greenwood Press, 1986); and Charles Humana, World Human Rights Guide, 3rd edn (Oxford University Press, 1992). Adam Przeworski, Democracy as a contingent outcome of con icts, in: J. Elster & R. Slagstad (eds), Constitutionalism and Democracy (Cambridge University Press, 1988), p. 80. David Apter, The Politics of Modernization (University of Chicago Press, 1965). Adam Przeworski et al., Sustainable Democracy (Cambridge University Press, 1995); and Adam Przeworski, M. Alvarez, J.A. Cheibub & F. Limongi, What Makes Democracies Endure?, Journal of Democracy, Vol. 7, No. 1 (1996), pp. 3955.

4.

5. 6.

272

Debate
7. Adam Przeworski, Some problems in the study of the transition to democracy , in: G. ODonnell et al., (eds), Transitions from Authoritarian Rule: Comparative Perspectives (The Johns Hopkins Press, 1986), p. 58. 8. S. P. Huntington, How Countries Democratize, Political Science Quarterly, Vol. 106, No. 4 (1991/1992), pp. 60915. 9. See Jennifer McCoy, The State and the Democratic Compromise in Venezuela, Journal of Developing Societies, Vol. 4 (1988), pp. 85133. 10. B. J. Kerkvliet, Land Reform in the Philippines since the Marcos Coup, Paci c Affairs, Vol. 47 (1974), pp. 286304; and J. Moran, Patterns of Corruption and Development in East Asia, Third World Quarterly, Vol. 20, No. 3 (1999), pp. 56987. Ronald J. Herring, Zul kar Ali Bhutto and the Eradication of Feudalism in Pakistan, Comparative Studies in Society and History, Vol. 21, No. 4 (1979), pp. 51957. Pranab Bardhan, The Political Economy of Development in India (Blackwell, 1984), p. 46 and ch. 6. Ronald J. Herring, Embedded particularism: Indias failed developmenta l state, in: Meredith Woo-Cumings (ed.), The Developmenta l State (Cornell University Press, 1999), pp. 30634. Stuart Corbridge & John Harriss, Reinventing India (Polity Press, 2000), p. 173.

11.

12.

13.

14.

Two Elaborations on the Argument

EVELYNE HUBER
The general point of Leftwichs piece is very well taken. Formal democracy does not necessarily bring substantive democracy in the sense of progress towards political and social equality, but its survival may on the contrary demand the absence of efforts to advance popular participation and economic redistribution. I would like to elaborate on two points related to the general argument. First, one needs to distinguish clearly between democracies in advanced industrial countries and democracies in the developing world. Second, the same social forces that push for formal democracy generally also push for advances towards participatory and social democracy, but other variables may favour formal democracy while at the same time undermining progress towards meaningful participation and redistribution.1 In advanced industrial countries, it has been possible to institutionalise social and economic rights, though democracies in these countries vary tremendously in the extent to which they protect their citizens from poverty and reduce inequality through the tax and transfer system. The key causal variables that made the establishment of redistributive safety nets possible are long-term incumbency of social democratic parties or Christian democratic parties (particularly if in competition with social democratic parties). A further contributing factor has been a constitutional structure that concentrates power and minimises the availability of veto points.2 Democracies in advanced industrial countries also vary greatly in the extent to which people participate in the political process, and those differences are particularly stark among the lower classes. One of the main factors that reduce class differ-

Evelyne Huber, Department of Political Science, CB#3265, University of North Carolina, Chapel Hill, NC 275993265, USA.

273

Evelyne Huber
ences in rates of political participation is organisational involvement, particularly in unions. 3 Unions have also been the main bases of support of social democratic parties. Thus a strong labour movement contributes both to higher levels of political participation and to the establishment and maintenance of redistributive social policy. In developing countries, in contrast, the trade-off between redistribution and survival of formal democracy has been a real one. One of the primary reasons is that the enemies of democracy and redistribution have been stronger relative to the supporters of democracy and redistribution than in advanced industrial countries. For the most part, the same forces that have demanded inclusive formal democracy (not to be confused with the types of restricted democracy that were prevalent, for instance, in Latin America before the military takeovers of the 1960s) have also demanded social democracyand not infrequently socialism outright, and the enemies of redistribution have at best tolerated democracy as long as it did not entail a threat to their economic interests. In many developing countries, particularly in Latin America and Asia, large landowners have retained much economic and political power and have used it to obstruct democratisation and land reform to the best of their abilities. On the opposite side, organised labour and social democratic or radical left parties have mostly been weak and thus incapable of effectively pushing for democratic procedures and of competing successfully for political power through elections. Where organised labour was numerically moderately large, it was frequently co-opted or controlled by the state and thus had little autonomous mobilisation capacity to support leftist parties. If we examine the forces behind installation and survival of formal democratic regimes, we have to take into account three clusters of power: the balance of power in civil society, the balance of power between civil society and the state, and the power constellations in the international economic and political system.4 Having just discussed the rst of them, and given the limited space available, I shall concentrate on the last in the remainder of these remarks. Since the end of the Cold War, it is fair to say that the international system has been very favourable for the establishment and maintenance of formal democracy. However, it has done little to raise the quality of democracy in formal legal dimensions, and one could argue that it has been outright counter-productive in terms of fostering democratic participation and economic redistribution.5 American policy has shifted from support for any type of strong anti-communist regime, no matter how authoritarian, to a clear preference for formally democratic regimes, and newly (re)-established democratic regimes have used regional associations to put pressure on governments threatening to deviate from that norm. At the same time, the nancial crises suffered by most developing countries at some point during the past two decades have given the International Monetary Fund and the World Bank tremendous in uence on policy making in countries requesting assistance. Essentially, imposition by these institutions of their preferred policies has drastically narrowed the room for political choice and thus for democratic political participation. It has not eliminated this room, as there has been resistance to such policy imposition, but it certainly has reduced it. Moreover, the neoliberal structural reforms imposed have generally had a regressive impact on income distribution. Economically and politically, they have strengthened large capitalists and weakened both middle and working classes and their organisations. Thus they have contributed to the survival of formal democracy by reducing the threat to the essential interests of propertied groups, while at the same time severely undermining the development of participatory and social democracy. Notes
1. This argument is developed in Evelyne Huber, Dietrich Rueschemeye r & John D. Stephens, The

274

Debate
Paradoxes of Contemporary Democracy: Formal, Participatory, and Social Dimensions, Comparative Politics, Vol. 29, No. 3 (1997), pp. 32342. 2. There used to be an argument about the redistributive nature of the welfare state, with some claiming that it only distributes across generations, not social classes. Research done over the past decade has shown conclusively that some welfare states indeed are highly redistributive. See, for example, Deborah Mitchell, Income Transfers in Ten Welfare States (Brook eld, 1991); Anthony B. Atkinson, Lee Rainwater & Timothy M. Smeeding, Income Distribution in OECD Countries (Organization for Economic Co-operation and Development, 1995); and Evelyne Huber & John D. Stephens, Development and Crisis of the Welfare State (University of Chicago Press, 2001). 3. Sidney Verba, Norman Nie & Jae-on Kim, Participation and Political Equality (Cambridge University Press, 1978). 4. Dietrich Rueschemeyer , Evelyne Huber Stephens & John D. Stephens, Capitalist Development and Democracy (University of Chicago Press, 1992). 5. For the problems with the quality of formallegal democracy in Latin America, see Guillermo ODonnell, Polyarchies and the (un)rule of law in Latin America, in: Juan E. Mendez, Guillermo ODonnell & Paulo Sergio Pinheiro (eds), The (Un)Rule of Law and the Underprivilege d in Latin America (University of Notre Dame Press, 1999).

Fundamental Tensions in the Democratic Compromise

MUSHTAQ HUSAIN KHAN
In his statement Leftwich points out that for a democracy to be stable, there has to be an understanding that the losers in elections will not be wiped out by the winners. This inevitably ensures that stable democracies are conservative, based on consensus and therefore unlikely to carry out the structural changes necessary for development to take off. While this is undoubtedly a problem for developing countries trying to institutionalise democracy, a comparison of developing countries with more advanced ones suggests a more fundamental problem with trying to construct democracy at an early stage of development. This more fundamental problem has to do with who the winners of democratic contests are likely to be in a developing country, not so much that they are likely to have to compromise with their immediate contestants. Societies, of course, contain different interests, and the main economic interests constitute different social classes. Leftwich is right to argue that much of development involves structural change, and a key component of this structural change is the emergence of new classes as the economy develops, usually in the direction of capitalism. In advanced countries, a well-established capitalist class exists and dominates the economy. This fact has tremendous implications for understanding the viability of democracy in an advanced capitalist economy. The economic domination of the capitalist class and the domination more generally of the capitalist economic system ensures that democratic politics oscillates around a narrow range of options which are essentially pro-capitalist, but which may differ at the margin in terms of distribution. There are several reasons for this,

Mushtaq Husain Khan, Department of Economics, School of Oriental and African Studies, University of London, Russell Square, London WCIH 0XG, UK.

275

Mushtaq Husain Khan

of which the two main ones will have to suf ce here. First, the dominance of the capitalist sector ensures that most of the resources for running politics (rent-seeking expenditures in the economic jargon) comes directly or indirectly from capitalists. If, as a class, they are the biggest political spenders (not just in contributing to parties, but also in more subtle ways such as funding think-tanks, lobbyists and so on), it is unlikely that the outcome of the political process will systematically damage them. Second, and more importantly, the economic dominance of capitalism means that the welfare of most people, even if they are not capitalists, depends on the health of the capitalist sector. If they are workers, their employment and wage growth depends on the health of the capitalist sector. If they are professionals, the purchasers of their services are most often capitalists. If they are public sector workers or professionals, their wages and salaries come from taxation, which again depends on the health of the capitalist sector. Thus, even when the substance of politics is about how to redistribute the fruits of growth, most responsible participants understand that redistribution has to happen within a regime that is conducive to capitalist growth. In that sense, New Labour in the UK and the Third Way more broadly has simply formalised an insight that is as old as social democracy itself. These considerations, as much as, or more than, the need to nd the compromises that Leftwich identi es, are the reasons why democratic politics appears to be based on conservative compromises in most democratic countries, which are after all mostly advanced capitalist countries. The key thing is that these compromises are not necessarily destructive for growth, even though they may not be radical. The nature of the democratic compromise in developing country democracies is fundamentally different. Here, by de nition, capitalism is not the dominant sector and capitalists are not numerous enough to dominate the economy politically, even though individually they may be obscenely rich. The paucity of capitalists means that it is unlikely that they can collectively spend enough to ensure that their interests dominate in de ning the objectives of politics and rent-seeking. More importantly, and again because of the underdevelopment of capitalism, the welfare of most people is not directly tied to the welfare of the capitalist sector. True, if the capitalist sector grew rapidly, then in a generation or two most people would bene t from that, but there is a qualitatively weaker correspondence between the living conditions of most people and the health of capitalism. In such a context, political populism in extreme cases can easily kill off the capitalist sector without anyone in power caring too much, even though a very severe capitalist collapse may eventually touch almost everyone. For instance, Leftwich points out as an example of democratic conservatism the fact that the Bhutto regime in Pakistan failed to carry out land reform. This is true, but the Bhutto regime is also a very good example of Third World populism where large-scale nationalisations, clientelist job creation and a redistributive rhetoric had seriously damaging effects for local capitalism. This kind of politics is virtually unknown in advanced capitalist countries, short of a revolutionary situation. Thus the critical feature of developing countries is that, while the democratic contestants do have to make compromises with each other, there are no systematic reasons why these compromises will protect the interests of the capitalist sector or promote growth. The interests which dominate politics (and rent-seeking) in this context are both different from and much more complex than what we nd in advanced countries.1 In many cases, although not all, the groups which dominate politics (and not just electoral politics) are a motley collection coming from what may collectively and loosely be described as the intermediate classes. These include the educated classes with college or university education, the petty bourgeoisie and middling

276

Debate
to rich peasants. This middle strata may be numerically small (possibly as small as 10 per cent of the population) and economically weak, but in terms of legitimacy and organisational power they are collectively the most powerful group in most developing countries. Without their leadership and participation, electoral politics, in particular, is impossible. In practice, what we see most frequently in the democratic domain are contests between multi-class factions led and dominated by members of the intermediate classes. The objective of these contests is to replace an existing group with another one, and Leftwich is right to point out that democracy will only stabilise if the losers are assured of a chance of returning in a few years. This does not in itself identify the more fundamental problem, which is that the objectives of these competing groups may simply be to capture public resources and engage in very destructive types of corruption and primitive accumulation designed as immediately as possible to enrich their faction. This, rather than innate conservatism, is what is destroying most developing countries. Democracy is not the cause of many of these problems since value-reducing rent-seeking can operate without democracy, but the evidence certainly belies the argument that democracy in any simple way makes these problems less serious, for instance through greater accountability. Such an expectation is based on a naive view of democracy as a system whereby the people elect leaders to translate their preferences into policy. As Leftwich points out, the Schumpeterian view of democracy as competition between organised parties is more accurate, and the critical question is which groups have the organisational, nancial or other powers to dominate party politics? While democracy is a goal which is valuable in itself, the idea that democracy will help to solve the fundamental problems of value-reducing rent-seeking in developing countries is not supported by the evidence. Indeed by legitimising and intensifying contestation between essentially unproductive groups, democracy can in many contexts make the damaging rent-seeking worse. These observations are not arguments against democracy, but simply point out that the construction of political settlements which may allow more rapid economic development in developing countries has nothing necessarily to do with either promoting democracy or overthrowing it. Note
1. A comparative examination of these class and group interests in Asian countries and their consequences for rent-seeking is presented in Mushtaq Khan & K.S. Jomo (eds), Rents, Rent-seeking and Economic Development: Theory and Evidence in Asia (Cambridge University Press, 2000).

Conservative Elites and State Incapacities

JEAN GRUGEL
Adrian Leftwich offers a succinct and incisive account of the dif culties of combining growth with democracy in developing countries. In so doing, he goes

Jean Grugel, Department of Politics, University of Shef eld, Northumberland Road, Shef eld S10 2TU, UK.

277

Jean Grugel
to the heart of a strategic dilemma facing not only governments but also social actorslabour, NGOs, rural movements, etc.in the developing world. How far will mobilisation to advance the grievances of particular sectors upset the fragile institutions upholding the new systems of elected rule and destroy the chances of building a democratic state? Nevertheless, I would argue that Leftwich is wrong to attribute the barriers to equitable development to the constraints imposed by democracy. State capacity may be at least as much an obstacle to redistribution as the formal political system. Moreover, Leftwich suggests that democracy can only be understood in the limited sense of electoral politics and elite consensus. In this, he follows Adam Przeworski, the most spirited contemporary defender of minimalist democracy.1 But democratic systems have the potential to be much more than simply a conservative mode of governance. Indeed, it is precisely its radical edge that accounts for democracys enduring ideological appeal. Development, understood as a project of empowerment, citizenship and balanced growth, requires the implementation of the emancipatory potential inherent within democracy. Rather than existing in contradiction, then, democracy and development constitute mutually enhancing utopian projects. Democracy is a mode of decisionmaking about collectively binding rules and policies over which the people exercise control.2 The most democratic arrangements, therefore, are those that make possible to the greatest conceivable degree the principles of popular control and equality in policy making.3 Measured against this yardstick, the new systems of rule that have emerged since the 1980s are by and large very limited in terms of their democratic content. The reasons for this are complex and multiple. They include the fact that globalised systems of production make for a weak version of democracy because they strengthen the hand of capital vis-a ` -vis the working class, the traditional bearers of democracy.4 At the same time, the increasingly in uential institutions of global governance, most notably the IMF and the World Bank, are generating, at best, formalistic models of democracy in which the introduction of liberal democracy is tied functionally to economic growth. This is the case notwithstanding the rhetoric of consultation and ownership espoused by the governance agencies. In sum, the capacity of the proponents of limited democracy to push their vision globally and the dif culties of building systems of genuine equality cannot be doubted. Nevertheless, this should not be taken to mean that elite democracy is the only possible version. In fact, there are growing signs that the governance capacities of conservative democracy are exhaustedas in Argentina, for exampleas well as very visible popular disenchantment with it. A number of radical democratic projects are now emerging, the best known of which is perhaps the Zapatista movement in Mexico. Moreover, there are a range of movements that belong to a new and exciting public sphere, that of global civil society, engaged actively in a struggle to decouple democracy from capitalism, corporations and exclusionary systems of elite consensus with the aim of relocating it within discourses of autonomy and citizenship.5 Organisations within global civil society, then, offer both a critique of the inappropriate imposition of the institutions of liberal democracy and a source of inspiration for the generation of alternative models. Leftwich is certainly right to suggest that limited democracies, especially in developing countries, have a poor record in implementing the rapid and far-reaching programmes that effectively address structural inequalitythough, of course, it is worth remembering that dictatorships have generally performed even worse in this respect. He attributes these dif culties to the nature of the pacts, formal or informal, that make possible the introduction of electoral politics. As democratisation theorists have consistently noted, the deals struck during transitions provide the mould

278

Debate
for politics later. But attention must also be paid to the question of the state. The obstacles to development after transition are frequently embedded state practices of clientelism and predation and very real limitations on the states capacity to deliver goods and services. Perhaps more than anything else, a genuine process of democratisation relies on the transformation of the state, since it is only through the state that the promises of equality, accountability and citizenship inherent within democracy can be made real and delivered upon.6 Yet transforming the state is a herculean task, especially in the developing world. Changing the institutions of government is relatively easyespecially given the range of inducements for developing countries to introduce elections and the paraphernalia of liberal democracy by comparison with the comprehensive set of reforms that are required to democratise the state more broadly. Reforms are blocked by elites who rely on money and status to deliver excessive political in uence. Moreover, and perhaps even more pertinently, states need infrastructural power, that iscollective power, power through society, the capacity to work with and through society7in order to deliver development and democracy. This is a kind of state capacity that developing states notoriously lack. This absence of capacity and legitimacy is just as likely to be the reason why developing states fail to deliver development and democracy. To sum up, most contemporary experiments in democratisation are partial and limited. This does not mean that democracy is essentially conservative but rather suggests that processes of political and economic change are currently controlled by local or global conservative elites. Despite this, a growing number of local, national and transnational movements that contest the conservative attempt to lay exclusive claim to democracy are emerging. It is not democracy, then, that can be held responsible for the lack of development in the South, but rather conservative policies allied to state incapacity. Notes
1. Adam Przeworski, Minimalist conceptions of democracy : a defense, in: Ian Schapiro & Cassiano Hacker-Cordo n (eds), Democracys Value (Cambridge University Press, 1999), pp. 2355. 2. David Beetham, Liberal Democracy and the Limits of Democratization, Political Studies, special issue, Vol. 40 (1992), pp. 4053. 3. Ibid. 4. Dietrich Rueschemeyer , Evelyne Huber Stephens & John D. Stephens, Capitalist Development and Democracy (University of Chicago Press, 1992). 5. See Mary Kaldor, Transnational civil society, in: Timothy Dunne & Nicholas Wheeler (eds), Human Rights in Global Politics (Cambridge University Press 2000), pp. 195213. 6. For a fuller development of this argument , see Jean Grugel, Democratization: A Critical Introduction (Palgrave 2001). 7. Michael Mann, The Sources of Social Power: The Rise of Classes and Nation States, 17601914, Vol. 2 (Cambridge University Press 1993), p. 59.

Perspectives from Village Democracy in China

BAOGANG HE
Adrian Leftwich examines the structural contradiction between the conservative requirements of stable democratic survival and the urgent transformative imperatives

Baogang He, School of Government, University of Tasmania, Box 25222, Hobart 7001, Tasmania, Australia.

279

Baogang He
of late development and concludes that the contradiction makes the combination of democracy and development so dif cult. Indeed, the experience of China as a late developmental country fully supports this conclusion. To reduce structural inequality, the totalitarian regime of Mao Zedong took rapid and far-reaching steps, such as land reforms, which did not take place in a democratic India. Chinas economic success since the 1980s has also been achieved under the authoritarian regime of Deng Xiaoping and Jiang Zemin. At the local level, economic development demanded an able, upright and authoritarian local leadership. There is a trade-off between democracy and development, and peasants demand for money appears to override the demand for democracy. Leftwichs contradiction thesis, nevertheless, can be improved by drawing a distinction between the earlier and later stages of economic development for later developmental countries. Obviously, his thesis does not apply to the later stage of development. Take the experience of village election and democracy in China. As a village reaches a certain level of development and as its wealth accumulates, farmers demand public accountability, democratic control of wealth and fair distribution of collective wealth. Because of the de ciency of local democratic institutions, corruption often occurs amongst authoritarian village leaders and village funds are often misused, leading such rich villages to ruin. This con rms Mancur Olsons argument elsewhere that autocracy will rarely have good economic performance for more than a generation. 1 The demand for village election and democracy also stems from a desire for relatively equal redistribution of village wealth. When a village has accumulated enormous wealth, say from compensation payments for land, it is necessary to put the village committees under the supervision of the people so as to institute a degree of accountability.2 Here, following Huntington, 3 I would like to revise the old modernisation theory and emphasise that the mode of distribution of wealth is as important as the level of wealth in understanding the democratisation push.4 It is observed that, if village leaders distribute village wealth in a paternalist way so that each family in the village will have a relatively equal share of village wealth, the farmers demand for elections will decrease. By contrast, when inequality in wealth distribution is signi cant, particularly when corruption as a mode of distribution in favour of village cadres is rampant, elections sponsored by the central government offer disadvantaged farmers an opportunity to demand a relatively equal distribution of village wealth. Veteran soldiers, the young, the educated and experienced political and business activists demand elections because they want a relatively equal distribution of collective wealth and some control over village affairs in order to protect their interests. Poor farmers see elections as a way of gaining monetary bene t (up to 60 yuan for a vote) and advancing their interests. This structural contradiction, according to Leftwichs analysis, derives from the conservative nature of democracy. He discusses two forms of its conservatism. First, they [democracies] have not, to the same extent, been able to de ne, agree on or institutionalise social and economic rights. Second, democracies have great dif culty in taking rapid and far-reaching steps to reduce structural inequalities in wealth. However, there is a third form of conservatism. In the case of village democracy, the village assembly usually does not endorse proposals of riskinvolved investment or other developmental projects. Village and township leaders hence complain that the democratic institution of the village assembly has slowed down economic development. It should be stressed that this conservative attitude towards development is to do with the contingent opinion of villagers, electors and village representatives, not with the structure of democracy itself. If the villagers attitude toward

280

Debate
risk changes, they are likely to support risk-involved development. Moreover, Leftwichs argument about the conservative nature of democracy is only true in some areas but false in others. Here it is extremely important to distinguish the four dimensions of wealth and power (generating, legitimising, distributing and seeking to control) in discussing the conservative nature of democracy. Although Leftwich mentions these four dimensions, it is not clear whether he makes a distinction when he examines the conservatism of democracy. The democratic control of village wealth, for example, is not conservative, for village democracy ensures that farmers have the right to access village accounts and the right to question the village leaders in the assembly about the use of village wealth. Indeed, Leftwich recognises the radical element of democracy in protecting individual political rights and civil liberties. The democratic distribution of wealth is not conservative either in that farmers are able to discuss, debate and vote on the policies relating to their interests, and to demand a relatively fair distribution of collective wealth through their representatives in the assembly. These two dimensions should be distinguished from the other two forms of power and wealth in terms of production and legitimisation. It is in these two areas that the conservative tendency of village democracy is most apparent. Village democracy is conservative is so far as it does not necessarily support risk-involved development, especially when the majority village assembly representatives reject proposed development programmes. It is also conservative in the sense that it enables private capitalists and the new rich to win power through elections. In sum, their acquisition of wealth can be legitimised in this democratic process and further reinforce structural distribution inequalities. In conclusion, Leftwichs main argument can be improved by making three distinctions. The rst distinction is the earlier versus later stages of development. His contradiction thesis holds true in the earlier stage of development, but not in the later development of late developmental countries. To sustain economic development, the democratic control and distribution of wealth is required. Without a democratic guarantee, any benign authoritarian gure is likely to abuse power and the rapid development will come to ruin. Second, the democratic control and distribution of wealth should be distinguished from the production of development and the democratic legitimisation of wealth. The former tends to be radical, while the latter is conservative. Third, three forms of the conservatism can be distinguished. The two forms identi ed by Leftwich are structural ones, while the third form identi ed by this author is contingent. By making these three distinctions, I would argue that democracy contains both radical and conservative elements in different areas: radical in both the democratic control of wealth and the demand for the fair distribution of wealth, but conservative in both the maintenance of structural inequalities and the production of development. Democracy and development are thus potentially compatible or even inescapably complementary, even whilst there is permanent tension between them. Notes
1. Mancur Olson. Dictatorship. Democracy, and Development, American Political Science Review, Vol. 87, No. 3 (1993), pp. 56776. 2. This is particularly true in Guangdong. See South China Morning Post, 19 November 1999. 3. Samuel P. Huntington, The Third Wave: Democratization in the Last Twentieth Century (University of Oklahoma Press, 1991). 4. Jean Oi and Scott Rozelle have made their contribution to the literature when they emphasised the importance of the source of income. See Jean C. Oi & Scott Rozelle, Elections and Power: The Locus of Decision-makin g in Chinese Villages, The China Quarterly, No. 162 (2000), pp. 51339. Here I would highlight the importance of the mode of distribution of wealth.

281