Richard Crook
In this paper, we shall be concerned mainly with forms of decentralisation other than federalism. Decentralisation is a very general term, conventionally used to describe the extent to which the political, administrative or fiscal powers of a central government have been shared or distributed, amongst territorially defined sub-national agencies or authorities. This spatial distribution usually forms a series of nested hierarchies coinciding with the territorial subdivisions. But within this broad definition, governmental decentralisation in practice exhibits an enormous variety of forms, based on different principles, and with widely differing purposes. Three main types can be identified: de-concentration, delegation and devolution. Only the latter is based on the principle of power sharing between national and sub-national units, in which the sub-units are granted legal, financial and/or political autonomy over agreed areas of activity (c.f. Turner and Hulme, 1997).
One critical variation in decentralisation schemes concerns the issue of delimitation of areas, which profoundly affects both the scale and viability of decentralised authorities and their political significance. Delimitation decisions are particularly critical in multicultural societies. Even if a decision has been made to use decentralisation to “recognise” cultural diversity, and thereby help to legitimise government, the decision itself raises questions of which kinds of cultural segment to recognise. There may be competing identities on offer – ethnic, regional, cultural, religious, language, historic political units incorporated within the state, or simply localised traditional communities (especially in Africa). Decisions taken concerning the relative weight of political and cultural factors as measured against considerations of technical and economic efficiency, and resource viability relative to the functions allocated, are therefore highly political (Smith, 1985).
But delimitation may also be used for the opposite purpose – to cut across and demobilise cultural units that are perceived as threatening. If a regime is nervous about providing an institutional base for sub-national, regional or ethnic political rivals, or even potential separatists, then it will often adopt a decentralisation scheme which deliberately fragments potential local power bases into smaller, weaker, non-politically significant units.1 This is often combined with central funding and control mechanisms that permit spatial redistribution and/or centrally focused patronage linkages. The Nigerian military regime in the 1990s, for instance, used local governments to transfer resources from the oil-producing to the non-oil producing areas, whilst in Côte d’Ivoire the government’s fear of regional political opposition was reflected in the extreme weakness and fragmentation of the large number of small communes. In Uganda, the delimitation of local government areas has divided the main ethnic power bases that were seen as the causes of 20 years of conflict and civil war. In effect, delimitation of decentralised authorities can turn a “decentralisation” scheme into a device for enhancing central power!
What kind of contribution might governmental decentralisation make to the management of culturally-based political conflict? This question cannot be answered without a brief analysis of the nature of cultural pluralism and the circumstances in which it leads to conflict.
Comparative analysis suggests that the mere presence of cultural pluralism (differences of ethnicity, language, culture etc.) does not automatically lead to conflict. It should be recognised that probably a majority of contemporary states exhibit some degree of cultural pluralism. Yet not all countries are divided by the kind of mobilised ethnic or cultural conflicts that threaten the integrity of the state through violence, undemocratic behaviour, separatism, and even civil war. Cultural pluralism only threatens the stability of a state when it is politically mobilised, and institutional structures are unable to mediate the kind of competition that it produces. We have therefore to explain the factors that underlie such variation.
Modern sociological theories stress the constructed and situational characteristics of ethnic and cultural identities, which are similar to the processes through which “nationalism” arose. Competitive political mobilisation of these highly variable identities tends to be “triggered” by a combination of factors: (i) the configuration of the cultural pluralism itself (which can vary from dominant group through bi-polar to multi-polar and cross-cutting multiplex), and (ii) the extent to which the political and economic context offers “cultural entrepreneurs” with some incentive or “pay-off” to organise identity-based political competition (Young, 1976 and 1998).
Thus if a context of uneven ethno-regional development, power transition and electoral competition is combined with dominant bi-polar or dominant core culture configurations, then powerful incentives exist both for cultural mobilisation and eventually irreconcilable conflict. This is particularly so where the political structure and culturally-based voting patterns (caused by political mobilisation) produce a prospect of permanent exclusion or subordination for the minority or dominated groups, no matter how “democratic” the system
(e.g. Fiji, Nigeria in the 1960s, Northern Ireland, Sri Lanka).
In multi-polar systems, such as former Yugoslavia, Uganda, or Nigeria after the civil war, a different logic is at work. The situation has to provide a pay-off to mobilise ethnically, even though one group can never win power on its own either through autonomy/secession or guaranteed shares of state benefits (Nigeria’s “federal character” principle, Bosnia and Herzegovina’s constitutionalised ethnic balancing).
Decentralisation is one amongst a number of strategies or “institutional designs” that have been proposed as a way of managing the conflict that arises from politicised cultural pluralism (Ghai, 1998). To succeed it must address some of the “triggering factor” issues noted above. Its strengths and weaknesses in relation to these may be summarised as follows.
The most obvious and frequently cited advantage of decentralisation is that it is the only strategy that addresses the issue of exclusion or subordination of mobilised minority or sub-national groups. It does so by offering some possibility of an autonomous sphere of political power or community, within which diversity can flourish, and the premium put on central power be reduced. Because it is inherently a spatial and political strategy, it is probably only relevant where there is some element of geographic concentration of the cultural segments, even at very local levels. Where the alternative is civil war, it can preserve the integrity of the state, provided there is some minimal agreement on the political community that forms the state. Federalism can underpin power-sharing arrangements by preventing the central state from appearing as the exclusive property of one cultural group, and thereby reduce the incentive to polarise identity-based attacks on the state, or to mobilise secessionist movements.
On the other hand, decentralisation is not only a device for assuaging demands from culturally and geographically defined groups. If decentralisation based on recognising cultural segments is seen as too dangerous insofar as it encourages the separatism of large sub-national groups and ethnic mobilisation, then the contrary “fragmenting” logic of decentralisation can be used to defuse polarisation and encourage alternative political arenas that reinforce centripetal forces and coalition building. It should also be noted that it has been used to dis-empower and exclude groups from the central state, as in apartheid South Africa or various military/authoritarian regimes.
If democratic decentralisation (primarily to devolved or mixed authorities) is used as a device to give autonomy and political space to ethno-regional groups or to minorities, then there remain important questions about its effectiveness in ensuring fully inclusive participation, providing an equal standard of public services and economic development throughout the national territory, and protecting the civil and political rights of all groups. Particularly if the decentralisation is based on culturally defined territorial areas then, unless there is complete homogeneity in each local area, the system will create new “minorities within minorities”, who may be subjected to domination by local elites who capture power by “democratic” means. Does this mean that power sharing and minority protection devices have to be put in place at the local levels, that simply reproduce and entrench the problems of the national level, as in Bosnia and Herzegovina? Or can the problem be avoided by the greater opportunities for participation offered by decentralised governance?
It has long been recognised that representative or liberal democracy, despite its commitment to formal political equality, can simply reproduce existing social and economic inequalities. This is particularly so when institutions of social and economic domination substantially overlap with or coincide with the power structures and those who control them (Rueschmeyer, Stevens and Stevens, 1992). “Democratic deficits” can arise just as easily in local as in national democratic arenas. Smith, for instance, has argued that decentralisation is an inherently conservative strategy, on the assumption that local elites are likely to be predominantly hostile to reform and to the empowerment of locally subordinate groups, precisely because of the proximity and intensity of their relationships and the dependency of the mass of the poor (Smith, 1985, 193). It is therefore harder for excluded groups to organise countervailing collective action against the social and economic power of local elites.
Smith’s view perhaps too readily dismisses the fact that the configuration of local social and economic structures will vary considerably even within one country. It is these variations that determine the likelihood of conservative elite capture. Echeverri-Gent (1992) proposes a revised model of what he calls the “paradox of participation”, a paradox which, he argues, arises because the poor and disadvantaged remain disadvantaged in their capacity to benefit from the enhanced opportunities for representation provided by devolved, democratic decentralisation. He shows that that in West Bengal elite capture by an “anti-poor” class was not inevitable in a competitive situation where a “counter-elite” (in this case the Communist Party) was able to mobilise a broad alliance of the poor, middle peasants and salariat, and to supplant the old landlord class.
Another factor which affects the likelihood of elite capture is the relationship between local elites and the central political power holders, as determined by the political purposes of the decentralisation scheme itself. What kind of alliance does the ruling elite have, or seek to build, with local or sub-national elites? Is the ruling elite facing potential challenges that it must deal with either through attack, circumvention or cooptation? If there are already powerful and autonomous local elites with whom the central government is seeking alliances (often associated with a decentralisation plan aimed at buying off ethno-regional groups) then a conservative “elite-dominated” outcome is very likely. But even the absence of strong local elites does not guarantee an inclusively democratic outcome, if the central government’s motivation is to use decentralisation to project its power more effectively into local areas, as with patronage-based party-building efforts in countries such as Bangladesh, Nigeria or Kenya in the 1980s and 1990s.
In effect, the comparative evidence suggests that elite capture of democratic decentralisation is a problem unless there is strong support from the central authorities themselves to prevent it happening. This is most likely in situations where the central ruling elite challenges or tries to circumvent locally powerful groups. The motives can range from party and ideological rivalry, class and ethnic conflict, through to the deep distrust often found in federal systems between institutional elites at different levels of government. The desire to challenge entrenched regional and provincial power holders can follow a popular revolution (as in post-Marcos Philippines) or the victory of reforming social democratic or communist parties, as in Brazil or the Indian states of West Bengal and Kerala (Crook and Sverrisson, 2001).
Can democratic participation at the local level prevent elite capture and/or more conflict between majority/minority groups within a decentralised system of government? Participation may be defined as covering a range of activities, both representative and direct or community based.
The main forms normally involve electoral activities, contacting/influencing, and associational membership. The evidence from a wide range of studies of decentralisation suggests that in general, democratic decentralisation does lead to increases in the number of people who participate, and broadens the scope (social representativeness) of that participation (Crook and Manor (1998) on Ghana, Côte d’Ivoire, Bangladesh and Karnataka, India; Blair’s review of Bolivia, Honduras, India (Karnataka), Philippines, Ukraine and Mali (Blair, 2000), and Crook and Sverrisson’s review of 12 cases in Asia, Latin America and Africa (Crook and Sverrisson, 2001)). This is particularly so in societies with an established history of political party activity, as in South Asia. Participation has even increased in countries with a long history of one-party or dominant-party rule, where the introduction of democratic local government was part of an overall liberalisation reform (e.g. Côte d’Ivoire and Mexico in the 1990s). But such reforms are less successful where local government already has a very bad record or there is no tradition of or experience of electoral politics.
As is well recognised, elections are not sufficient on their own to deepen democracy, and the forms of the participation that engage people between elections are clearly crucial if formerly excluded groups and minorities are to be included. Here the record in both Asia (particularly South Asia) and Africa is much more encouraging, particularly where there are also links with traditions of community action and self-help (as in much of Africa) or mobilisation of disadvantaged groups (as in India and some Latin American countries). In Africa one has to note a very different cultural expectation of local government, which prefers to see successful “sons and daughters” of the area participating and thus bringing patronage (Jaglin, 1993; Crook, 1994 and 1996; Olowu, 1989).
Crook and Manor, and Blair found in their range of case studies that groups such as women, the less well educated and youth were drawn into political activities such as public meetings, and contacting, and even into representative councils. This was especially so where there were affirmative action or quota systems, as in India where 30% of seats are reserved for women and additional quotas for “Scheduled Castes”, or South Africa and Namibia (c.f. also Scandinavia). Even quotas however do not always lead to increased effective contribution, and India illustrates the enormous difficulty of overturning deeply entrenched social structures that sustain elite domination of local politics (Webster, 1994; Centre on Integrated Rural Development for Asia and the Pacific (CIRDAP), 1992, 105).
Direct participation is classically contrasted with representative government, as the “Athenian” ideal in which all citizens have an equal chance of actively taking part in decision making. It is now understood to refer to a wide range of small-scale community-level institutions and projects based on this ideal, ranging from village and neighbourhood councils to projects for local selfmanagement of common property resources, and women’s micro-credit groups.
The association with decentralisation in less-developed countries grew with the participatory development movement of the late 1970s and early 1980s (Chambers, 1983), a movement which argued that “development” could only be equitable and effective if people (the beneficiaries) controlled the process themselves, rather than governments or experts. This trend was paralleled in new thinking about local government in the UK and Europe, through the introduction of reforms such as neighbourhood housing associations, citizens’ juries and consultative assemblies, and participatory social audit and health campaigns. The new British government’s policy initiative, “New Deal for Communities” launched in 1998, encapsulates the new approach very comprehensively (Goetz, Gaventa et al., 2001; Burns, Hambleton and Hogget, 1994; Batley and Stoker, 1991). In practice, many direct participation reforms involve linking community-level activities with the formal institutions of local government. One of the key problems of representative government even at the local level is the lack of social trust in the legitimacy of government institutions – a district or municipality may be seen to be as remote and untrustworthy as central government; hence the issue of how to make a transition from the community level of action to wider arenas.
In Brazil, Bolivia and the Philippines, for instance, decentralised governments are required by law to incorporate or formally associate community, neighbourhood or “peoples” organisations with the deliberative procedures of local government, as well as to give them a role in the administration of various services and projects. The Brazilian experiments with “participatory budgeting”, which began in the cities of Porto Alegre and Belo Horizonte as an attempt to combat the legacy of basismo, are perhaps the best known and most successful (see below)(Nickson, 1995).
The Philippines law stipulates that municipal councils should include “private sector” representatives, including in every case a representative of the workers’ and women’s sectors, and at least one from organisations of the “urban poor”, indigenous communities, the disabled and others. But there are real political problems with this reform in that the local governments regard the Non-Governmental Organisations (NGOs) and Peoples Organisations as former adversaries, whilst the Peoples Organisations tend to see the new local government reform as an attempt to restore the power of the pre-Marcos feudal bosses and “warlords” under the banner of local and regional rights (Brillantes, 1996). This kind of conflict and distrust between NGOs and government is not peculiar to the Philippines.
The most direct of all forms of participation – village-level community or selfhelp activities – are particularly popular in Africa. In Ghana and Uganda, for instance, they are incorporated into the formal local government structure. They are popular with donor-funded schemes for encouraging “community driven development”. Unfortunately, this kind of participation is most prone to reflect existing inequalities and power structures. In Asia, it would seem there is less willingness to contribute cash or free labour time. Where it does occur, it may have a forced aspect to it, or the costs to poor people may be recognised through the provision of “food for work” programs, as in India, Bangladesh or Nepal (CIRDAP, 1992).
In general, it may be concluded that although participation in decentralised and democratic forms of government can widen the scope or social basis of participation, it is still very hard for it to prevent existing local elites from capturing control of the new institutional power opportunities (Blair, 2000; Crook and Sverrisson, 2001). Even strong forms of affirmative action do not always translate into actual effective influence in the decision-making processes by the special representatives, as can be seen in the record of local government in the Indian state of Karnataka. In Bangladesh, Nigeria, and Kenya, conservative elite capture of decentralisation was actually facilitated by the government’s desire to create and sustain a power base in the countryside, and to prevent opposition forces coalescing. In Ghana a similar motivation took over after an initial commitment to radical reform faded. Côte d’Ivoire and Mexico exemplify a third scenario in which decentralisation was used by a ruling party to renew and extend the party’s support without any real commitment to pro-poor policies which might disturb the entrenched but dependent networks of influence and patronage.
In the developing world there are only a few examples of decentralisation reforms that have benefited disadvantaged groups – the Indian states of West Bengal and Kerala, and a number of municipalities in the Brazilian states of Rio Grande do Sul, Minas Gerais and Ceara. Their common characteristic is that in each case conservative local elites were challenged locally by groups supported externally by an ideologically committed, reforming government and/or party (Crook and Sverrisson, 2001; Heller, 2001).
A policy conclusion might be that decentralisation schemes should avoid delimiting the devolved areas in such a way as to give an enhanced power base to elite groups (particularly in a situation of ethnic stratification). If this cannot be avoided for reasons of cultural balancing or the need to grant autonomy to groups threatening national unity, then the central government must have a party organisation capable of sustaining popular control over, or even displacement of, local elites within the devolved areas. This needs to be combined with representative institutions that assure broad participation of all groups, both through electoral and affirmative action schemes. But affirmative action cannot be relied upon to produce the desired result unless the “quota representatives” of previously oppressed or excluded groups have considerable support from parties or other extra-local forces.
Enhancing participation at the local level through decentralisation is clearly not sufficient in itself to make governments more equitable, inclusive or responsive. Participatory activities can only effectively influence the outcomes or performance of local government if they are mediated through mechanisms of accountability, which ensure that the governments act in accordance with the wishes of representative bodies and implement authorised spending policies correctly. Transparency, or “open government” which gives citizens full information on what is being decided and how, is a kind of public accountability mechanism which only operates in systems which already have the basics of democratic accountability in place.
Political accountability is the requirement that governments be answerable for their actions, both in law and to the public. It is conventionally divided into “internal” or institutional accountability and “external” or public accountability (although O’Donnell (1999) uses the terms “horizontal” and “vertical” for more or less the same concepts).
Internal accountability mechanisms include such devices as political accountability to elected bodies, formal hierarchies of administrative, fiscal and legal accountability, and professional value systems. External accountability mechanisms include elections, public opinion as articulated through the media and civil society organisations, political party and constituency linkages, and new forms of “citizen auditing” (see below).
Decentralised government has certainly contributed to the development of some of the above mechanisms, most notably in the field of participatory citizen vetting and monitoring. Democratic decentralisation also introduces elections and elected office holders to arenas where they have not existed before or where they have been uncompetitive/authoritarian. Elections are still the most important form of accountability in that they introduce at the local level both a demand for internal political accountability, and new expectations of public accountability not just at election time, but also between elected representatives and their constituents, and through the new participatory devices mentioned above.
There is in fact, a very wide variety of electoral systems and methods for appointing local officials and holding them to account. Comparative studies suggest that electoral democracy at the local level is most effective where the executive authorities are elected indirectly from amongst the body of elected councillors, and are subject to continuous monitoring through committees or “administrative commissions” of the council (Olowu and Smoke, 1992; Mawhood, 1993; Crook and Manor, 1998; Gasper, 1989). Direct popular election of mayors (as in Nigeria, Bangladesh, and most of Latin America) on the other hand, at least in the developing country context, seems to be associated with some of worst cases of corruption, patronage politics and unaccountable behaviour. Another device that undermines accountability in less developed countries is political appointment of local chief executives by central government. This is usually done to ensure that central government loyalists are in charge locally and can enforce compliance with central government policies.
It is also important that there is a balance between political control and legal/administrative constraints. In Indian states such as Karnataka, West Bengal, or Maharashtra, elected politicians are definitely “in control”, in that they have the status, the social embeddedness, and the capacity to deal with civil servants, discuss policies and if necessary push them through against a reluctant bureaucracy (CIRDAP, 1992). But over-powerful political executives can cause problems too, and their actions need to be balanced by a wellestablished civil or public service that has sufficient autonomy to prevent politicians from breaking the law or financial regulations.
The ratios of representation at the local level should also be as “close” as possible, and encourage close ward-based relationships between elected representatives and their constituents. It is for this reason that list systems are not conducive to effective accountability in local governments. In Latin America, for instance, a major problem is the political under-representation in local governments, characterised by representative ratios of between 1:20, 000 and 1:80, 000 in the urban municipalities, and over 1:100, 000 in some of the big cities (Nickson, 1995).
Finally, it is worth stressing that competitive and regular elections remain a basic requirement. There is little point in executives reporting to elected bodies unless those bodies are prepared to challenge, demand information and debate alternative strategies; hence the need also for a free press and media. It is no doubt rare to find all of the benign contextual conditions combined. The development of genuine forms of accountability and participation at the local level is a challenge to power structures that can easily snuff them out if they remain isolated. That is why decentralisation is best introduced within a broader context of democratic and governance reforms at the national level, so that they are part of a “virtuous circle”.
The idea that decentralisation will provide more transparent, honest and legitimate government is linked not just to the expected benefits of local accountability. It is also a product of expectations regarding the “process” aspect of institutional performance. These concerns may be subdivided into those relevant at the institutional level, such as decision making, imposition of taxes and allocation of resources, and those which involve relations with the public, such as the behaviour of officials, collection of taxes, and elections. Transparency is more properly about whether the public knows more about how and on what basis decisions have been taken. Whether they then find those processes acceptable will affect the extent to which the decisions themselves are accepted.
There is some evidence that democratic decentralisation and its accompanying enhancement of transparency make corruption (in its narrower sense) more visible, particularly where, as in India, it is combined with the effects of political accountability, free press reporting and party competition.
However, it may well have the perverse effect of causing the public to think that corruption has increased when, as reported by Manor for Karnataka, it actually decreased (Crook and Manor, 1998). In other countries however, for example Bangladesh, Nigeria, Tanzania, Papua New Guinea, Philippines, Kenya and Thailand (United Nations Economic and Social Commission for Asia and the Pacific (UNESCAP), 1991), corruption and financial mismanagement is reported to have in fact increased at the same time as becoming more visible. And democratic decentralisation creates a new political class, which is not only responsible for public money but can also cost substantial sums insofar as allowances and other expenses have to be paid. In poor countries this can amount to quite significant proportions of local budgets (see especially Nigeria and Papua New Guinea (Ramm, 1993; Awotukun, 1995; Gboyega, 1998)).
The most visible aspect of corruption for members of the public in less developed countries is not, however, the graft associated with contracts or patronage allocations, but the extent to which they are dealt with fairly and honestly by government employees in every-day routine transactions at clinics, licensing offices, tax offices and the like (petty corruption). There is hardly any empirical research on this topic, but Crook and Manor (1998) found some evidence of improvement in the attitudes and behaviour of employees in Karnataka. Again, this was principally due to the willingness of members of the public to complain, and the accountability mechanisms of the new councils.
On the other hand, it is not certain that greater transparency will reduce the incidence of patronage practices and nepotism. Some analysts have indeed argued that the closer government is to local society, the greater the likelihood of corruption as officials are more likely to be pressured by familial and friendship networks which they find it difficult to offend (Tanzi, 1995). The history of local government in Latin America would seem to confirm this picture. According to Nickson (1995) the vast majority of the region’s 2.7 million local government employees are clientelistic appointments – whilst local government in Africa has the same image.
There is a trend in the literature to suggest that the most effective way to control corruption by decentralised authorities is to make the relationship between costs and benefits more transparent and direct. The intention is to make government decision makers “bear the financial and political consequences of their decisions” and force them to internalise costs. That is, they would not be able to shift the cost of services for local beneficiaries onto the shoulders of non-beneficiaries, or to “export” taxes (Tanzi, 1995; Bird, 1994; Werlin, 1992). A further important mechanism would therefore be extensive employment of “user charges” for services directly provided by the local authorities, and therefore the responsibility of those decision makers (Bird and Vaillancourt, 1998, 12).
These “New Public Management” approaches, already extensively used in the industrialised economies since the 1980s, have not been tested very much in the Less Developed Country context, except in the area of “user fees” (see Paul, 1994). The key problem with the approach would seem to lie in the concept of a direct relationship between providing only services that can be paid for by available revenues or user charges, and the consequences for providers. Only if local governments were forced to be genuinely self-sufficient would there be real consequences, which would probably lead to the demise of most Less Developed Country local governments outside the big towns, and a process of “de-development”. Ultimately, therefore, these are arguments against government decentralisation and in favour of the market.
Responsiveness of government is more than just accountability, although the two are inevitably linked. An “accountable” institution is not necessarily responsive – and perhaps should try not to be in the case of judicial or auditing authorities. The assertion that decentralisation will be more responsive to the needs of ordinary citizens is derived from the notion that local, more participatory forms of government and development activity will offer more than just greater effectiveness in promoting economic development. Institutional responsiveness has been defined as the achievement of “congruence between community preferences and public policies” such that the activities of the institution are valued by the public (Fried and Rabinovitz, 1980). Of course, this begs the question of “responsive to whom?” Does decentralisation benefit particular social groups or areas of the country, more than others?
Decentralised government has been the focus for a number of “citizen voice” initiatives over the past decade, initiatives that attempt to enhance the responsiveness of government by linking with reforms in the way government agencies themselves work. Enhanced citizen voices cannot be effective unless government agencies are ready to hear them, and to change their behaviour accordingly (Goetz, Gaventa et al., 2001). The most successful initiatives have therefore involved participatory partnerships between citizens and government agencies, which are categorised below.
| ƒ | Citizen-based monitoring and evaluation: these are devices for citizens |
|---|---|
| to review standards of service provision, and assess their likely impact | |
| on particular social groups. “Women’s Budget Initiatives” for instance, | |
| have been developed in a number of countries, such as South Africa, | |
| Uganda, and Canada, inspired by a state-level scheme in Australia. | |
| ƒ | Citizens’ Auditing: e.g in India, the best known and most radical |
| program is that launched by a politically organised community group, | |
| the Mazdoor Kisan Shakti Sangathan (Peoples Right to Information | |
| Campaign) in Rajasthan, which uses public “naming and shaming” | |
| meetings to demand public information on local government spending. | |
| In Bolivia, officially recognised Neighborhood Vigilance Committees | |
| monitor municipal investment decisions, and have the power to use a | |
| legal denuncia against a local council if they suspect corruption. | |
| ƒ | Joint management of sectoral programs: joint state-civil society |
| management and delivery of services such as forestry, watershed | |
| protection, or primary healthcare are an increasingly common form of | |
| responsive governance and have achieved acknowledged success, e.g | |
| West Bengal’s Forest Protection Committees, local healthcare rationing | |
| in Nova Scotia and Saskatchewan (Canada), and education user | |
| groups in Denmark. |
ƒ Participatory planning and budgeting: in a number of countries, participatory planning mechanisms have been institutionalised through legislative reforms or government-sponsored programs which provide frameworks for participatory planning by local citizens. The most successful to date has been the participatory budgeting system pioneered in the Brazilian cities of Belo Horizonte and Porto Alegre after the election of the Workers’ Party (PT). Researchers have found that the process did result in more responsiveness to the needs of poor areas (favelas) and outlying districts, and created a “culture change” in both government officials and citizens (Paixao, 1996; Avritzer, 2000).
Making government agencies more capable of hearing citizen voices, and therefore able to be more responsive has also been the focus of various reform efforts. On the one hand, there is the New Public Management approach, which emphasises incentives, sanctions, and performance measurement. If the goal is to make public servants more client-oriented and responsive to the public, then individual incentive structures which mimic the market must be introduced into the public service so as to reward clientcentric behaviour.
On the other hand, a number of programs around the world have sought to improve and sustain public service responsiveness through creation of a new public service ethos amongst its employees, based on a “human relations” management approach. One of the most successful was the Health Agent Program in Ceara State, Brazil. The case is frequently cited as an example of what can be achieved through the inculcation of a strong sense of dedication and job commitment amongst “front-line” employees. Unfortunately, decentralisation did not play a key role, in that one of the explanations for its success is that the governor ensured that the state health ministry was in charge of recruitment, thus by-passing municipal patronage networks (Tendler, 1997).
Other methods for making local governments more responsive include better and more accessible information about services (“One stop shops”), and new rights for citizens to demand the delivery of publicly guaranteed standards, either through citizens’ charters or even legally enforceable constitutional or statutory rights to a certain level of service.
What factors are most associated with success in making these kinds of initiatives produce more responsive government? Systematic comparative research suggests that citizen voice initiatives need to be built into the legal and formal structures of the system, and cannot be left up to civil society initiatives. Citizen groups need the power to demand formal investigations or seek legal redress for non-delivery of services (Goetz, Gaventa et al., 2001).
In this paper we are concerned primarily with the question of whether decentralisation can make a contribution to the management of systemthreatening conflict in multicultural societies. Therefore, if it is used as a device to give autonomy and political space to ethno-regional entities or to otherwise excluded minorities, then there is a potential problem over its ability to protect the rights of oppressed or disadvantaged minorities created within new, culturally defined political authorities. And if there is ethnic stratification associated with uneven development, there is also an issue of spatial equity. Will decentralisation ensure equal standards of human and economic development throughout the national territory, or will it exacerbate regional inequalities?
In international law, the rights of minorities within states are protected as individual rights under the United Nations Declaration on the Rights of Persons Belonging to National, Ethnic, Religious and Linguistic Minorities. In other words, they are only protected as individual citizens in accordance with the standard human rights conventions. “Group” rights are seen as equivalent to the concept of self-determination and therefore fall under the restriction limiting the “right of self determination” to “peoples under colonial rule”. Many indigenous or aboriginal peoples around the world are now therefore demanding various forms of decentralised autonomy within the states where they live, not as minorities but as “peoples under colonial rule” (Stavenhagen, 1998). But in general, ethnic or cultural groups living in the kinds of cultural pluralist situations described above are, if they are mobilised, looking to varieties of consociationalism, or decentralisation to deal with their problems.
The system adopted in Bosnia and Herzegovina as part of the Dayton Peace Agreement is a classic example of the difficulties of trying to build minority protection into an ethnically defined system. The Dayton Peace Agreement created for Bosnia “probably the most decentralised state in the world” (Carl Bildt quoted in Chandler, 1999, 67). At every single level down to the municipality, formal rules enforce multiethnic power sharing and guaranteed places for “minorities”. This is an acknowledgement of the fact that, in spite of “ethnic cleansing”, communities are still mixed, with minorities in each area. The system therefore succeeds in simultaneously exacerbating both sets of the key problems caused by cultural pluralism. On the one hand, it gives maximum autonomy to ethnically defined or ethnic majority areas that are capable of threatening the very integrity of the state. On the other hand, it gives maximum incentive to cultural entrepreneurs at the local level to mobilise ethnic categories in order to secure access to local government resources and patronage with veto powers to minorities. It is therefore a very good warning of the dangers of basing decentralisation on a logic of ethnic power sharing (Crook, 2001).
There is very little evidence that decentralised governance leads to more redistributive or equitable economic policies. As noted above, comparative studies of decentralisation in developing countries found only a few examples, such as the Indian states of West Bengal and Kerala, and some progressive municipalities in Brazil. The key issue is that of elite capture. The difference between the positive cases and countries such as Bangladesh, Nigeria, or Mexico is not that the latter fail to allocate sufficient funds to the decentralised authorities, or that they lack centrally funded development and anti-poverty programs. The real difference is that the latter governments fail to ensure that central funds are used in a responsible and accountable manner, and fail to ensure implementation of pro-poor policies where these formally exist. The explanation for this is to be found in the politics of central-local relations: in none of these three countries was it likely that decentralisation would empower any kind of challenge to local elites resistant to or uninterested in the development of pro-poor policies.
Insofar as decentralisation enhances the political and fiscal autonomy of territorial sub-units, it is by definition likely to exacerbate spatial inequalities. The more dependent on local revenues and assigned taxes the system is, the more this is likely to be so. The only mitigating factor is that new decentralisation programs may result in net transfers to more remote rural areas, purely as a function of the setting up of government authorities in those areas and the allocation of minimum funds (Crook and Manor, 1998). Spatial equity can therefore only be assured if there are centrally directed transfers of funds or reallocations of tax revenues using equalisation formulae. These are particularly crucial in the cases of extremely distorted regional revenue bases found in states dependent on oil or mineral producing enclaves, which is why decentralisation is particularly dangerous in such situations.
Equalisation formulae are in fact quite common, and are certainly easier to manage if decentralised authorities have minimal fiscal autonomy. In developing countries, central government transfers normally account for the largest single element – usually the majority – of local government revenues, other than the big city authorities. A few industrialised countries such as Ireland and the UK also have very high transfer levels – 80-90% of local government revenues (Litvack and Seddon, 1999; Nickson, 1995; World Bank, 2000; International Monetary Fund, 2000). Decentralisation in such settings therefore creates new tensions between local and central areas over allocation of revenues, which in a state riven by mobilised cultural conflict can again cause extra problems, and provides another reason to be cautious about the supposed benefits of decentralisation schemes based on culturally defined areas.
Avritzer, L., 2000. Civil Society, Public Space and Local Power: A Study of the Participatory Budget in Belo Horizonte and Porto Alegre Brazil. Paper delivered to the Ford Foundation Civil Society and Governance Conference, Amsterdam, 24 September 2000.
Aziz, A. and Arnold, D.D., eds, 1996. Decentralized Governance in Asian Countries. London and New Delhi: Sage.
Awotukun, A.M., ed., 1995. New Trends in Nigerian Local Government. Ile-Ife: Department of Local Government Studies, Ofafemi Awolowo University.
Barkan, J.D. and Chege, M., 1989. Decentralizing the State: District Focus and the Politics of Reallocation in Kenya. Journal of Modern African Studies, 27(3), 431-453.
Batley, R. and Stoker, G., eds, 1991. Local Government in Europe: Trends and Developments. Basingstoke: Macmillan Education.
Bird, R., 1994. Decentralizing Infrastructure: For Good or for Ill (Working Paper; Background Paper for World Development Report, 1994), Washington, D.C.: World Bank Institute.
Bird, R. and Vaillancourt, F., eds, 1999. Fiscal Decentralization in Developing Countries. Cambridge: Cambridge University Press.
Blair, H., 2000. Participation and Accountability at the Periphery: Democratic Local Governance in Six Countries. World Development, 28(1), 21-39.
Brillantes, A.B., 1996. Local Governance and Decentralization: The Philippine Experience. In A. Aziz and D.D. Arnold, eds. Decentralized Governance in Asian Countries. London and New Delhi: Sage, 197-215.
Burns, D., Hambleton, R. and Hoggett, P., 1994. The Politics of Decentralization: Revitalising Local Democracy. Houndmills: Macmillan Press.
CIRDAP (Centre on Integrated Rural Development for Asia and the Pacific), 1992. Impact of Decentralization on Rural Poverty: An Asian Perspective. Dhaka: Centre on Integrated Rural Development for Asia and the Pacific.
Chambers, R., 1983. Rural Development: Putting the Last First. London: Longman.
Chandler, D., 1999. Bosnia. Faking Democracy After Dayton. London and Sterling VA: Pluto Press.
Crook, R.C., 1994. Four Years of the Ghana District Assemblies in Operation: Decentralization, Democratization and Administrative Performance. Public Administration and Development, 14, 339-364.
Crook, R.C., 1996. Democracy, Participation and Responsiveness: A Case Study of Relations between the Ivorian Communes and their Citizens. Public Administration, 74 (3), 695-720.
Crook, R.C., 2001. Strengthening Democratic Governance in Conflict Torn Societies: Civic organizations, Democratic Effectiveness and Political conflict. IDS Working Paper no. 129, May 2001.
Crook, R.C. and Jerve, A.M., eds, 1991. Government and Participation: Institutional Development, Decentralization and Democracy in the Third World. Bergen: Chr. Michelsen Institute.
Crook, R.C. and Manor, J., 1998. Democracy and Decentralization in South Asia and West Africa: Participation, Accountability and Performance.
Cambridge: Cambridge University Press.
Crook, R. and Sverrisson, A., 2001, Decentralization and Poverty Alleviation in Developing Countries: A Comparative Analysis or is West Bengal Unique?
IDS Working Paper 130, June 2001.
Echeverri-Gent, J., 1992. Public Participation and Poverty Alleviation: The Experience of Reform Communists in West Bengal. World Development, 20(10), 1401-1422.
Fried, R.C., and Rabinovitz, F.F., 1980. Comparative Urban Politics – a Performance Approach. Englewood Cliffs, NJ: Prentice Hall.
Gasper, D., 1989. Multilevel Planning and Development Administration in Botswana. Mount Pleasant, Harare, Zimbabwe: Department of Rural and Urban Planning, University of Zimbabwe.
Gboyega, A., 1998. Decentralization and Local Autonomy in Nigeria’s Federal System: Crossing the Stream while Searching for the Pebbles. In: J. Barkan, ed. Five Monographs on Decentralization and Democratization in Sub-Saharan Africa: Occasional Papers 45-49. Ames: University of Iowa, International Programs.
Ghai, Y., 1998. Decentralization and the Accommodation of Ethnic Diversity. In: C. Young, ed. Ethnic Diversity and Public Policy. Chapter 2.
Goetz, A-M. and Gaventa, J. et al., 2001. Bringing Citizen Voice and Client Focus into Service Delivery. IDS Working Paper 138, July 2001.
Heller, P., 2001. Moving the State: The Politics of Democratic Decentralization in Kerala, South Africa and Porto Alegre. Politics and Society, 29(1), 131-163.
IMF (International Monetary Fund), 2000. International Financial Statistics Yearbook, 2000. Washington, D.C.: International Monetary Fund.
Jaglin, S. and Dubresson, A., eds, 1993. Pouvoirs et cités d’Afrique noire: decentralisations en question. Paris: Karthala.
Korten, D.C., 1984. People Centred Development. West Hartford: Kumarian Press.
Leonard, D.K. and Marshall, D.R., 1982. Institutions of Rural Development for the Poor: Decentralization and Organizational Linkages. Berkeley: Institute of International Studies, University of California.
Litvack, J. and Seddon, J., eds,1999. Decentralization Briefing Notes. Washington, D.C.: World Bank Institute.
Mawhood, P., ed., 1993. Local Government in the Third World. 2nd ed. Pretoria: Africa Institute of South Africa.
Moris, J.R., 1991. Institutional Choice and Local Development: What Kind of Social Science Do We Need? In: R. Crook and A. M. Jerve, eds. Government and Participation: Institutional Development, Decentralization and Democracy in the Third World. Bergen: Chr. Michelsen Institute, 170-206.
Nickson, R.A., 1995. Local Government in Latin America. Colorado and London: Lynne Rienner.
Olowu, D., 1989. Achievements and Problems of Federal and State Transfers to Local Governments in Nigeria Since Independence. Washington, D.C.: Coordination and Development Administration Division, Economic Development Institute, World Bank.
Olowu, D. and Smoke, P., 1992. Determinants of Success in African Local Governments: An Overview. Public Administration and Development, 12(1), 1
17.
O’Donnell, G., 1999. Polyarchies and the (Un)rule of Law in Latin America. In:
J. Mendez, G. O’Donnell and P. Pinheiro. The (Un)rule of Law and the Underprivileged in Latin America. Notre Dame: Notre Dame University Press, 303-335.
Paul, S., 1994. Does Voice Matter? For Public Accountability Yes. World Bank Policy Research Working Paper 1388, December 1994. Washington, D.C.: World Bank.
Paixao, P.R., 1996. Participative Budgeting in Belo Horizonte: Democratization and Citizenship. Environment and Urbanization, 8(1), 213
222.
Ramm, J-F., 1993. Rethinking Decentralization in Papua New Guinea. Nagoya: United Nations Centre for Regional Development.
Rueschemeyer, D., Stephens, E.H. and Stephens, J.D., 1992. Capitalist Development and Democracy. Cambridge: Polity Press.
Smith, B.C., 1985. Decentralization: The Territorial Dimension of the State. London: George Allen & Unwin.
Stavenhagen, S., 1998. Indigenous Peoples. In: C. Young, ed. Ethnic Diversity and Public Policy: A Comparative Enquiry. Basingstoke: Macmillan, United Nations Research Institute for Social Development, Chapter 6.
Tanzi, V., 1995. Fiscal Federalism and Decentralization: A Review of Some Efficiency and Macroeconomic Aspects. Paper presented at the 1995 Annual Bank Conference on Development Economics, World Bank, Washington, D.C., 1-2 May 1995.
Tendler, J., 1997. Good Government in the Tropics. Baltimore: Johns Hopkins University Press.
Turner, M. and Hulme, D., 1997. Governance, Administration and Development: Making the State Work. Basingstoke: Macmillan.
UNESCAP (United Nations Economic and Social Commission for Asia and the Pacific), 1991. Fiscal Decentralization and the Mobilization and Use of National Resources for Development: Issues, Experience and Policies in the ESCAP Region. Bangkok: United Nations Economic and Social Commission for Asia and the Pacific.
Webster, N., 1994. Panchayati Raj and Democratic Decentralization in West Bengal, India. Paper presented at the Conference on Democratic Decentralization in Africa and Asia, Institute of Commonwealth Studies, University of London, July 1994.
Werlin, H., 1992. Linking Decentralization and Centralization: A Critique of the New Development Administration. Public Administration and Development, 12(3), 223-235.
World Bank, 2000. The World Development Report 2000. Washington, D.C.: World Bank Institute.
Young, C., 1976. The Politics of Cultural Pluralism. Madison: University of Wisconsin Press
Young, C., ed., 1998. Ethnic Diversity and Public Policy: A Comparative Inquiry. Basingstoke: Macmillan, United Nations Research Institute for Social Development.
1 Compare with Barkan and Chege’s analysis of the Kenyan situation, in which they propose a hypothesis that “the probability that decentralization will serve the political interests of the regime varies inversely with the power and resources of the ethno-regional base on which the regime rests” (Barkan and Chege, 1989, 21).