Governance, decentralisation, and poverty: the case of Pakistan.

Ismail, Zafar H. ; Rizvi, Sehar

Human development and human rights share a common vision and purpose: to secure the freedom, well being and dignity of humanity. Human development is as essential for human rights as the latter is for the former. Historical evidence suggests that the more civilised societies were those that gave a higher priority to both, for example, the Greek, the Roman and the enlightened years of early Islam. The freedom from want is perhaps the one inalienable right of humanity which stands between dignity and indignity and which must be mitigated against by both state and individual.I For the first time in history mankind adopted these and other human rights when the Universal Declaration of Human Rights was adopted by the United Nations in 1948. Today all but one of the six core covenants and conventions have been ratified by at least 140 countries and state sponsored anti-poverty programmes were initiated globally, most with the help of civil society [UN DP (2000)].

The lessons from history clearly establish that for improving the social endowment of the poor the state and the individual are key players, money is a necessary requirement, good governance is critical to success and the participation of each component of society as a whole ensures sustainability and continued benefits beyond the initial attempts to mitigate the impact of poverty. In other words decentralisation is a key element to improve the lot of the people, particularly the poor and the disenfranchised.

Governance is generally conceived of as the exercise of economic, political and administrative authority to improve the quality of life of the people and is a continuing process where divergent opinions and desires are satisfied through compromise and tolerance in a spirit of cooperative action for the mutual benefit of the larger whole. It has three dimensions: one, the political regime; two, the systems and procedures for exercising authority; and three, the capacity of governments [World Bank (1994); UNDP (1997); OECD 0995); Commission on Global Governance (1995)]. The basic objective of good governance is to improve the quality of life of people, to ensure that their needs are met equitably and that disparities in income are reduced over time and is best attained through the effective participation of people, that is through decentralising the apparatus of the state to the closest level to people, namely local governments. The incidence of poverty is one measure of whether governance in any nation or state is good or bad.

Politics is about the creation and distribution of power, be it within organisations, tribes, communities or society at large. Governance is the mechanism which controls the relationship between the two extremes--the governed and the governors. And the political process lies at the core of governance and this can be said to be efficient only if elections are free and fair, the elected are accountable; authority is divided between the legislature, the bureaucracy and the judiciary; and, power is decentralised.

By definition, decentralisation distributes power, resources, decisions and capacities from central to sub-national governments (mainly municipalities) and communities [IADB (1999)]. The processes of decentralisation focus on the "formal institutions of government", both political and administrative [Parker and Serrano (2000)]. it has come to be widely regarded as an important instrument for supporting the factors needed to create effective local governance and promote grass roots development. It is increasingly being realised that central governments cannot do it all, and that the active involvement of both communities and representative local governments is a requisite for a thriving local development.

While much of the drive for decentralisation may be political, the main economic justification has been to improve efficiency. And the important aspect of efficiency in decentralisation is information--lower tiers that have better information about what kind of services they want and are willing to pay for. Within the region, support for decentralisation has reflected general beliefs about its role in strengthening democracy, overcoming social imbalances and improving the provision of services. But what must always be kept in mind is that the form of the decentralisation process chosen must reflect specific national conditions, policies and priorities.

Decentralisation is complex and encompasses a wide range of elements [Parker (1995)]. Evidence from many countries would seem to suggest that three important conditions are necessary if decentralisation is to lead to improved governance [Manor (1999)]. Firstly, significant powers and responsibilities for local service delivery should be devolved to representative local authorities in line with their capacities i.e. political decentralisation. Secondly, sufficient resources, through a combination of local taxes and grants from higher level governments, must be provided to enable local governments to fulfil their responsibilities, That is, fiscal decentralisation. And finally, proper channels for accountability are needed to encourage strong accountability between bureaucrats and elected representatives, and between elected representatives and their electorate i.e. institutional decentralisation. At its best, decentralisation does not simply transfer centralised functions to the local level but simultaneously reorganises the roles and responsibilities of central governments, local government and communities, and opens government processes to greater involvement by the people.

Institutional factors, more often than not, impede effective decentralisation. in Pakistan, rural self-government worked well soon after independence through the panchayat system. The rural population abandoned this in favour of a more centralised system, which resulted in denial of access to basic services. While lack of institutional capacity was cited as a reason for disbandment of the participatory system, the newer system left a majority of the citizens with no voice and participation and no access to basic public services. Indonesia on the other hand, is now nurturing self-government in rural areas though its village development and poor villages grants programmes [Shah (1997)]. Contrary to common misconceptions, the success of decentralisation policies requires a strong responsive and accountable government at the national level. The success of decentralised structures on the other hand critically depends upon the higher level enabling environment and citizen participation and less so on the local institutional capacity and information network as confirmed by the Colombian experience [Fiszbein (1995)].

Poverty is an outcome of the interaction of economic, social, legal and political processes mediated through a range of institutions. By removing these barriers to poverty reduction, the state can help to empower the poor to improve their lives. The interaction between the quality of governance and democracy, the rule of law and the extent of devolution, decentralisation and autonomy determine the success of state institutions in being "pro-poor." In essence, therefore, there are four ingredients in the relationship between governance and poverty: one, democracy, two, the ride of law, three, bureaucratic performance, and four, pro-poor institutions.

DEMOCRACY

In any society the key dimensions to poverty are both being unheard and powerless. Democracy requires one, regular and peaceful elections; two, highly inclusive participation by all; and three, freedom of expression, of the press, to form and join organisations, and to information, it is argued that democracy will reduce these barriers to poverty eradication through empowerment. However, this is not necessarily achieved, principally because the poor are excluded from access due the lack of education, knowledge, opportunity, economic power, information, etc. Unless elected to local office the likelihood of the poor having access to the elected is non-existent. Thus the poor need a direct voice in the interventions that affect them on a day to day level, and to build up their assets to prevent their exclusion. On the other hand, democracies appear to be more favourable for the income, health, and education dimensions of poverty. Further, a free press can create awareness in the public and increase transparency and accountability in governance. This in turn helps in reducing corruption, reduces the arbitrary use of authority and increases the potential for meeting the needs of the poor. All of which combined contribute to higher rates of growth.

Democracies, more often than not, fail to deliver on pro-poor programmes. Bardhan (1997); Bardhan and Mookharjee (1999) and Easterley (1999) state that the poor performance of democratic institutions in this respect may be due to the capture and manipulation by the elites and the middle-classes. The former as a consequence of economic power and the latter their ability to lobby and corner resources. Systems, procedures and politicians add to the low pro-poor performance. Inherently democratic processes are slow, cumbersome and could lead to deadlock, and politicians need to show results, thus the temptation to succumb to populist programmes which are visible rather than best for long term growth and equity. But because they are inherently more transparent and accountable, on the whole, however, it cannot be argued "that less democracy is better for economic growth and poverty reduction" [ADB (2000)].

Unfortunately, Pakistan does not meet at least three criteria for good governance. The lethal combination of powerful personalities (or a coterie of a few families who are linked together by inter-marriage) and weak institutions(which are of their own making) has resulted in the subversion of law, the deprivation of people's sovereignty, the lack of accountability and a process which remains largely personal and informal and allows rulers to operate outside established institutional frameworks. Democracy, when present, is confined largely to the ballot box--and that too for the legislatures only. Political parties do not hold elections for office. Decision-making is controlled by the powerful elite. Sovereignty is equated to powerful governments, not free citizens.

Because of the highly personalised control over power and resources, institutions are weakened and decisions are arbitrary and prone to informal transactions. The elite power structure consists of landlords, who have a disproportionate share of seats in the legislatures and thereby frustrate reform, a handful of large industrialists who cumulatively control resources which are equal to or greater than the federal budget, bureaucrats who have gained power as a consequence of weak political structures and the military, which have overtly ruled for half or more of the time and covertly for considerably longer. The last of the quartet justify their rule as a consequence of a breakdown in civil rule and political governments failing to legitimise themselves.

HDC [1999] argues that thus democracy has not been translated into visible benefits for the poor and queries "Can people be truly free when they have no control over their destinies?" The moral foundations of most governments in Pakistan are weak, to say the least. The are founded on electoral fraud, money politics, criminalisation of the political system and increasing corruption. It has weakened societal structures to such that all governments (elected or not) appear to have the same face as a result of inter-marriage and cross-part representation by each of the elite families. They also constitute a fair portion of the bureaucracy, the judiciary and the military as the youngest are "encouraged" to seek a career to safeguard familial interest. Thus they control the economy, and benefit disproportionately.

Notwithstanding these shortcomings there are positive gains which have accrued to society in Pakistan as a consequence of democracy. Proponents for democracy also argue that the behaviour of the politician is the consequence of societal structures. This is largely based on fiefdom-tribal, biradari (brotherhood), and familial. Owing to rampant illiteracy and a lack of awareness of alternatives, votes are cast on the basis of these loyalties. They also argue that over time there has been a change which has led to a more open atmosphere of debate and compromise which has resulted in a freedom of speech and of expression, some improvement in human rights. They also point to the freedom enjoyed by the press. The press have gained such strength that attempts to muzzle it by recent governments did not prove successful.

In a comparison of performance between democratic and authoritarian governments since independence from India, Ismail (1998) and SPDC (2000) have shown that while governments under authoritarian rule in Pakistan were good for economic growth, they were not necessarily as successful in improving human endowment. It can be argued this is a consequence of the complex and cumbersome mechanisms of accountability and consensus formation in democratic regimes which lead to investment into social capital. These checks and balances put a limit to rent-seeking, the need to trim sails for re-election, and thereby encourages the development of social infrastructure, such as schools, clinics, water supply schemes, etc. This is borne out by international experience [Bardhan (1997); Przeworski and Limongi (1993); Sen (1999)]. While authoritarian regimes do not face these problems they, nevertheless, face serious dangers of abuse of state power. Rehman [forthcoming] says that owing to these very irritating procedures democratic governments cannot make big mistakes while experience in Pakistan shows that autocratic governments have made blunders which were exposed ex-post. He, therefore, concludes that even a bad democracy is, in the long-run better for nations than good authoritarian rules. The question that emerges from his arguments is have democratic governments in Pakistan been pro-poor?

Narrowly understood as the right to vote in elections, democracy is an indirect and often incomplete mechanism for benefiting the poor. They need a direct voice in the interventions that affect them, as well as the ability to organise themselves politically and vote selectively. For instance, while Mr Bhutto is attributed to have made the mass of the population more aware of their political rights to select their candidate, mechanisms to ensure that the poor had an effective voice were not instituted, voting on the basis of peer pressure was rampant, and democracy was not introduced within the political parties themselves. Thus the leaders and the candidates were self-selected. Further, an analysis of the card carrying membership of the political parties shows that very few, if any, were from below the poverty line. Action, therefore, is needed to bring down legal, political and social barriers that work against particular groups and to build up the assets of the poor to prevent exclusion.

An effective opposition ensures that leaders are responsive to their electorate. A free press and an active non-governmental advocacy sector draw public attention to current events. Civil liberties linked to press freedom and unimpeded access to the law increase transparency and accountability, reduce corruption and increase the potential for meeting the needs of the poor. Political rights, civil liberties, and press freedom collectively provide greater protection for the poor from the arbitrary use of discretionary powers and authority by officials, both elected and appointed.

To make democracies more pro-poor, governments and external agencies need to collaborate to ensure that anti-poverty measures and programmes are more effective. This would require that the institutional environment ensures that the poor engage in collective action on their own behalf. Therefore, the changes needed dictate that institutional reforms should consist of, one, a pro-poor political environment, two, accountability of public officials, three, appropriate programmes and institutional mechanisms to support them, and four, benefits that are legally recognised and enforceable at law, if required.

RULE OF LAW

Legislation creates the environment which can lead to both stable and higher levels of economic growth as it can be predictable, define property rights, guarantee long-term security, boost investment, and lower transaction costs. By both stimulating growth and guaranteeing property rights, it increases the prospects of poverty reduction through an increase in incomes and empowering the poor. The daily lives of poor people are deeply affected by how well the legal system works. Harassment, lawlessness, and violence are ever present threats to the poor people. To mitigate this aspect of poverty, the poor should be able to access entitlements without complication and be free of the fear of lawlessness and harassment. Further, the laws must not be discriminatory, either in design, application, access or execution.

Ignorance of the law (as it is written, and the poor have low literacy levels) and poverty (as they cannot afford the time and money costs of legal transaction) are important causes of lack of access to the law and its protection. Further, language, ethnic, caste and gender barriers, their greater vulnerability and other exclusionary practices exacerbate the knowledge problem and their distrust of the justice system as a whole. Lacking the wherewithal, the poor people are often forced to forgo the protection of the law, even while engaged in legal activities.

While the constitution enshrines the rights of people, only some basic rights are justiciable, others, (economic, social and cultural), contained in the Directive Principles of Policy, are not. However, the Supreme courts have found innovative mechanisms for ensuring that even these are reviewed and that too rapidly through a system of participatory justice. Single applications of public interest are received directly by the highest levels and cases are disposed off relatively rapidly.

Because many of the problems affecting the poor are common civil, society organisations can help the poor through class action, thus seeking redress for issues which affect a group of the poor. Alternative dispute resolution (ADR) mechanisms of mediation and arbitration could help explain the law and reduce the delays and corruption which are today an integral part of dispute settlement. They could also provide more predictable outcomes as the arbitrators would be more aware of ground realities regarding the dispute.

The government's proposed devolution plan is expected to create such mechanisms and through the involvement of the local councillors and vigilance committees ensure a higher degree of redress to the poor.

BUREAUCRATIC PERFORMANCE AND INSTITUTIONAL DECAY

Perhaps the greatest barrier to poverty reduction is the ossification of institutions which have over time drawn away from the poor through capture by the elite and interest groups. North (1991) states institutions are "the humanly devised constraints that structure human interactions." They consist of informal constraints (sanctions, taboos, customs, traditions, codes of conduct) and formal regulations (constitutions, laws, property rights). Institutions thus include a wide range of observed phenomena, ranging from social networks to the state, that interact with each other.

Both economic growth and poverty reduction are dependent on the quality of state institutions and macro-economic management. Investment is attracted to environments which are stable, provide a consistent set of policies and where contracts can be easily enforced. There is considerable evidence to link higher-quality governance to a higher pace of economic growth and to higher levels of health and education, which are crucial in helping the poor become upwardly mobile. The latter then improves the quality of governance. Countries with higher-quality governance, particularly in managing the economy, have been able to respond to external shocks more rapidly and have thus been able to dampen the impact on the poor.

The efficient allocation of resources to ensure wide spread and equitable economic and social development of a nation is the basic responsibility of any government and this can only be achieved through good governance. Another aspect of good governance is achieving appropriate regulatory balance for growth and for increasing income-earning opportunities for the poor. In Pakistan the state straddles all economic, social and cultural activities through a variety of agencies, ranging from government to Quangos. Its contribution to equitable and sustainable economic and social development is minimal compared to the resources they pre-empt and the inefficient manner in which these are consumed. The inefficiency is in the role and composition of government: rather than ensuring the delivery of basic social services, redistributing resources and ensuring economic, social and societal stability, they have stepped into trade and production and over-zealous control of the private sector. These have generated inefficiencies, rents accruing only to vested interest groups; income disparities are rising as is unemployment and corruption.

"Corruption is the misuse of public power, office, or authority for private benefit" [UNDP (1999)]. Petty corruption occurs when public servants are grossly underpaid and may demand kickbacks and speed money to feed their families. While people may tolerate such demands, this is not generally approved. This impacts more on the poor. Empirical evidence suggests that the burden of corruption, measured as a proportion of income, is much greater on the poor. Grand corruption involves the higher levels of officialdom, both elected and appointed, which make decisions regarding large contracts, make or change policy or laws through enabling powers contained in badly drafted acts. For example, these could and do operate though the issuance of quasi legislation in Pakistan, the SRO (2), which either provides relief from taxation or gives specific economic benefits to a vested interest group. This benefits the rich, but affects the poor adversely.

Poor governance leads to, nay encourages and breeds, corruption in a number of ways, for instance through bribery and extortion, speed money, influence peddling and nepotism, and fraud and embezzlement. Low levels of corruption exist where the institutional arrangements between the executive, the legislative and the judiciary have effective mechanisms to identify, prevent and punish wrongdoing, provide few opportunities for corruption and society does not accept this. In other words, systems are transparent and officials are accountable. This is seen in decentralisation of functions. High levels of corruption occurs when the mechanisms to combat corruption are either weak or are not used, there are no checks and balances, there is extensive government control and regulation, and where, because it is so pervasive, society accepts and tolerates this as the norm, equating this to gifts. Its incidence is highest where high degrees of centralisation or concentration exists. However, Obasanjo (1994) makes the distinction between accepting a gift and accepting bribes: "The distinction between gifts and bribes is easily recognisable. A gift can be accepted openly; a bribe has to be kept secret".

Corruption is uneconomical, unsafe and threatens people, especially the poor, and governments. Corruption can be brutal to the poor who have no resources to compete against those with the wherewithal to pay bribes. It denies the poor their fair share and increases poverty by redirecting relief resources to corrupt government officials and business men. It encourages gender disparity as it redistributes from the poor (who are mostly women) to officials (mostly men). It damages the economy and slows, sometimes reverse, development. It can be used to pose environmental and health hazards and render legislation ineffective. Corruption is thus an issue of governance--the failure of institutions.

Most planners and economists agree that the best mechanism for sustaining growth into the future is to ensure effective governance. This may be realised through reforms in civil services, improving the capacity of institutions to frame coherent policies and ensure their consistent implementation over time, improving tax and fiscal administration, developing suitable measures for legal recourse, increasing the role of the private sector, decentralising public effort, devolution of fiscal powers, improving resource mobilisation at all tiers of government and controlling unproductive public expenditures.

Pakistan's public institutions faced with a basket of shortages and shortcomings. There is a skill shortage for identifying and enunciating coherent and coordinated policies; the work ethos discourages initiative and efficiency; systems and procedures are cumbersome and outdated; coordination mechanisms exist but are not implemented; staff is inadequately trained both abinitio and subsequently on-job because of the very short-term assignment to posts; staff selection, posting and promotions are not merit related; and the incentive structure is designed to penalise the performers. Administrative capacity and capability are an important component of good governance. By affecting the quality of public service delivery, administrative capacity and capability directly affects the well-being of the poor. Merit based recruitment and promotion linked to performance, market-based wages and autonomy from the political process are perhaps the most crucial elements for improving bureaucratic performance. Combined they will help reduce corruption and bureaucratic delays.

Central to the creation of institutional capacity is the effective decentralisation of functions. This means that effective financial powers, administrative authority and commensurate responsibility must be given to lower echelons but within a tight regulatory framework. The current practice of requests for supplementary grants would, therefore, have to be done away with and the current powers to authorise expenditures in excess of budget allocations would need to be exercised in the same manner as the budget allocations itself, that is by the legislature in assembly through a process of dialogue "ex-ante".

Analysing the institutional decay in Pakistan SPDC (2000) concludes "Institutional reform must be a component of structural reforms within the civil services. The objective should be to make the service more professional and accountable. While these reforms could be painful in the short-run the benefits from these reforms would be substantial in the long-run. These would include, but not be limited to, a smaller size of the civil bureaucracy, a lower per unit cost of service and infrastructure, and greater efficiency in government. The current status of "permanent non-terminable" employment must be replaced by a system where the inefficient or the corrupt can be weeded out within a short time frame. Staff skills need to be developed to use modern management techniques".

The remedy prescribed by several studies have stated that much of the faults in the civil services can be remedied through a redefinition of its role. Firstly, the civil services should be taught that they are meant to implement policy and not make it. Secondly, it must be limited to maintaining law and order, providing a regulatory framework for economic activities, promoting human development and encouraging participatory governance. Thirdly, they must be open and transparent in their dealings.

Efficiency can be achieved in a number of ways. One may be the use of appropriate mechanisms which ensure the greatest value for money. Other ways may include pro-active legislation (rather than retro-active) ensuring target based management, or through the use of information to analyse shortcomings and suggest ways to improve. Yet others may be the use of regulatory mechanisms which ensure that private sector provision of services and development of infrastructure does not create rents for the vested segments of society.

DEVOLUTION AND DECENTRALISATION

State institutions are seen to be monoliths too far removed from ground realities and there to serve the interest of the elite and vested interest groups. They are not seen to be pro-poor as systems and procedures are archaic, time consuming, cumbersome and access is effectively denied to the poor through rationing of visiting hours. Most studies have established that for making them more pro-poor is either through decentralisation (ceding power from the centre to a local government or agency with the centre government keeping some measure of oversight) or through devolution (more complete and permanent form of decentralisation, in which the power of the centre is more limited). Studies around the world have shown that effective decentralisation does benefit the poor as development and safety net programmes can be more effective in meeting local needs if they can draw on the advantages of effective community involvement (without exclusion of any group of people) as they rest on local information, local accountability, and local monitoring.

Yet there are important caveats to decentralisation. In settings where the local power structure is unequal, such as in feudal or tribal societies, or where the concentration of economic power is skewed, local government institutions are liable to elite capture and to be swayed by a large and vocal middle class through their ability to lobby. Further, evidence suggests that decentralisation could also lead to a worsening of the gender balance and girls may have less access to education, health care and other basic services. By itself decentralisation cannot make state institutions pro-poor. It must be accompanied by systemic changes and changes in work ethos. It must be adequately supported and safeguarded by a strong political will from the centre. It is, therefore, necessary to ensure that the relationship between local government agencies and communities are meaningful and designed to help the poor and disenfranchised segments. To ensure that the poor benefit decentralisation must be approached cautiously. Effectively implemented it has the potential to improve immediate development outcomes, improve cost-effectiveness of both implementation and delivery, and be the catalyst for broader institutional reforms which benefit and empower the poor.

Some hidden benefits from effective decentralisation and its partnering with community involvement would be a reduction in corruption, an increase in cost-effectiveness of services, and an increase in targeting efficiency for social safety net programmes designed to ameliorate the lot of the poor. Empowered, and hopefully thereby enlightened communities themselves would be able to identify the truly needy and the credit worthiness of individuals in any such schemes, particularly those involving access to micro-credit.

Barriers to devolution are more likely to be presented by financial resources (owing to existing patterns in the allocation of fiscal powers and intergovernmental fiscal relations); human resources (owing to problems in quality, remuneration and motivation); reluctance to devolve power (owing to a lack of political will and the urge to draw economic rents); weak institutional capacity (as a consequence of archaic rules and regulations governing human resource management, the lack of continuing skill development programmes, the induction of alternative technologies suited to local conditions, and the dichotomy in employing two sets of staff--those recruited locally and those on deputation from the provincial ser5vices or unified local council services with their divided loyalties and rapid turnover); and the use of antiquated systems and procedures (most dating from the days of the British 'raj').

With no explicit constitutional existence, the lowest tier of government can be superseded at the discretion of the provincial governments. However, social sector services are grassroots services which need to be provided at the local level by effective and continuous local governments with the participation of the communities. This would, ensure the inclusion of the poor themselves in the management of the infrastructure as they, particularly the women, are likely to have more time to devote to these tasks. This effective devolution can only be guaranteed by amending the Constitution (as India has done) to ensure that local councils cannot be set aside by the provincial governments, that they have a list of distinctive responsibilities, and that they have the resources to fulfil these responsibilities. This may require creating an arrangement similar to the National Finance Commission (NFC) which sets out the resource-sharing arrangements between the federation and the provinces every five years.

In the last decade and a half two major programmes were started with the avowed intention of both devolution and decentralisation. Both the Five Point Programme as well as SAP, are failures both in the reform of the institutions for and the delivery of public services. In both the programmes, the responsibility for delivery of services rested with the provincial governments while policy and financial controls lay with the federal government. SAP at least attempted to delegate some responsibility for operation, maintenance and monitoring to local communities. The management committees have hardly functioned in any meaningful manner, largely because these are not legal bodies nor are they representative. As they are transient in nature their sustainability is questionable. Nevertheless these management committees have had some degree of success in a number of areas, for example, the School Management Committees in the Punjab.

These general failures and shortcomings in both the Five Point Programme and SAP suggest that the need is for far-reaching institutional reforms. Perhaps, the answer lies not in partially delegating responsibility, but in effective devolution of full authority and responsibility over local matters. The time has come to consider the establishment of effective integrated local governments which are constitutionally mandated with executive and legislative arms and authority to tax and spend.

This implies devolution of federal and provincial functions and powers to local governments. This, however, is a political decision, and determined through an inter-play of historical, administrative and technical factors. Conceptually the question as to 'what to decentralise' can be determined by the presence or absence of three characteristics: externality (spillovers), chargeability (extent of self-financing), and technicity (relating to economics of scale and institutional capacity).

The current government strategy to alleviate poverty hinges on the devolution of power to the local governments. The general opinion about the Plan is that it is deficient in a number of ways. Firstly, the plan is likely to further compound the confusion by adding tehsil and village councils to the roster of local bodies. It is argued that in many parts of the country a number of villages are not an integrated community entity, while others are single household entities and some, comprising of nomads, are also mobile. Secondly, planning for and production of services requires a minimum critical mass of area, population and resources and fiscal viability, while the distribution/delivery of services requires closer contact with the citizenry. An upper tier, which meets the critical mass and economies of scale considerations, and a lower tier, which meets the economies of scope (for people's participation and accountability), service efficiency and equity considerations. The district and the metropolitan/municipal corporations appear to fulfil the requirements of critical mass and economies of scale for the production of major services. The union and municipal/town committee appear to fulfil the requirements of economies of scale, service efficiency and equity. However, the larger urban centres will need to have the more elaborate governing structure of metropolitan governments.

Proponents of the Plan argue that there is merit to the creation of the tehsil and village councils and of the Citizen Community Boards (CCBs). With responsibility for delivering municipal services and for acting as the bridge between the populations in the urban centres and the rural areas, the tehsil councils would be able to plan in an integrated manner for the development of both. They could thus ensure that the urban areas would become the market centres for agricultural inputs and output from the surrounding rural hinterland, and also centres for agro-based industries. Further by locating basic infrastructure, such as the higher tiers of education and health infrastructure in these towns, they would reduce the rural-urban migration. The village councils could well become the focal point for establishing the facility level supervisory agencies that were created under the SAP with only some degree of success. As these Boards will be selected by elected local governments, they should be more representatives and the revised Local Government Ordinances will have provided them with the legal mandate needed to ensure their sustainability and continuing existence. The structure, composition, authority and responsibilities of these Boards (at the facility or sub-regional levels) would need to be clearly indicated and defined. Their role should be pro-active rather than reactive. Because these CCBs would be elected by the local people the likelihood of ensuring that the poor are represented effectively on these Boards would increase. Thus the poor will begin to have a voice.

One can only wait and see.

Authors' Note: The views expressed are their own and do not reflect those of the Centre.

REFERENCES

Asian Development Bank (ADB) (2000) Asian Development Outlook 2000. Manila: Oxford University Press.

Bardhan, Pranab (1997) The Role of Governance in Economic Development, A Political Economy Approach. OECD Development Centre Studies.

Bardhan, Pranab, and Dilip Mookherjee (1999) Capture and Governance at Local and National Levels. (Mimeographed.)

Commission on Global Governance (1995) Our Global Neighbourhood. New York: Oxford University Press.

Easterley, William (1999) Happy Societies: the Middle-class Consensus and Economic Development. Washington, D. C.: World Bank (Mimeographed.)

Fiszbein, A. (1997) The Emergence of Local Capacity: Lessons from Colombia. World Development 25:1029-43.

Human Development Centre (1999) Human Development in South Asia, 1999--The Crisis of Governance. Karachi: Oxford University Press.

Inter American Development Banik (IADB) (1999) Workshop of Decentralisation and Local Development. Summary of Major Themes; Stockholm, Sweden 25-28th May.

Ismail, Zafar H. (1998) Impediments to Improvement of Social Sectors in Pakistan. Social Policy and Development Centre, Karachi. (Conference Paper No. 27.)

Manor (1999) The Political Economy of Democratic Decentralisation. Washington, D.C.: World Bank.

North, Douglass (1991) Institutions. Journal of Economic Perspective 5, 7-112.

Obasanjo, Olusegun (1994) Corruption, Democracy and Human Rights in Africa. Keynote address to the Africa Leadership Forum on Corruption, Democracy and Human Rights in Africa. Contonou, Benin, 19-21 September.

Organisation for Economic Cooperation and Development (OECD) (1995) Participatory Development and Good Governance, Development Cooperation Guideline Series; Paris.

Parker, A and R. Serrano (2000) Promoting Local Governance through Social Fund and Decentralisation. Washington, D. C.: World Bank.

Parker, A. (1995) Decentralisation: The Way Forward for Rural Development? World Bank, Washington, D. C. (Policy Research Working Paper No 1475.)

Przeworski, Adam, and Fernando Limongi (1993) Political Regimes and Economic Growth. Journal of Economic Perspectives 7:3.

Rehman, I. A. (forthcoming) Social Policy and Development Centre, Karachi. Sen, Amartya K. (1999) Development as Freedom.

Shah, A. (1997) Balance, Accountability and Responsiveness: Lessons about Decentralisation. World Bank Conference on Evaluation and Development, April 1-2.

Social Policy and Development Centre (2000) Social Development in Pakistan--Annual Review 1999: Towards Poverty Reduction. Karachi: Social Policy and Development Centre and the Oxford University Press.

United Nations Development Programme (UNDP) (1997) Governance for Sustainable Human Development: A UNDP Policy Document; Management Development and Governance Division. New York: UNDP and Oxford University Press.

United Nations Development Programme (UNDP) (1999) Fighting Corruption to Improve Governance. New York: United Nations Development Programme, Management Development and Governance Division, Bureau for Development Policy.

United Nations Development Programme (UNDP) (2000) Human Development Report 2000. Oxford/New York: Oxford University Press.

World Bank (1994) Governance: The World Bank Experience. Washington, D. C.: The World Bank.

* The views expressed in the paper are those of the author's alone and do not necessarily represent those of the Ministry of Finance or of the Government of Pakistan.

(1) The first informal recognition of this lies in the alms-giving encouraged by all religions and beliefs, and was formally codified by Islam through zakat and ushr (a form of redistributing wealth and agricultural income respectively) as a religious duty [the Koran]. In fact, this was applied with such zeal, and the partnership between state and individual was so effective that in the days of Caliph Umar Farooq (634-644 AD) the collections from zakat and ushr could not be distributed as no one could qualify as a mustaheq (deserving poor). The state system was unfortunately allowed to lapse. Individual and private efforts to mitigate poverty and endow the poor with social capital, however, continued throughout the Muslim comity.

(2) HDC (1999) refers to this euphemistically as the Sure Route to Opulence.

Comments

"Governance, Decentralisation, and Poverty" is a large subject, it is also clear that, notwithstanding the complexity of these topics, the authors have still handled it well. In the brief span of 11 pages they have encompassed a very wide range of issues: the 1948 UN Declaration of Human Rights, on the one hand, and the Citizens Community Boards, on the other. As we know, these Boards were promised by the National Reconstruction Bureau in their May 2000 document. "Devolution of Power and Responsibility". The authors have also quite appropriately clarified in their abstract that the central concern of their paper is to show how governance and decentralisation impinge on poverty, and to discuss what kind of mechanisms need to be set-up in the country to improve the situation so that institutions become pro-poor. This is not an easy task for anyone, and I must compliment them on having raised and discussed so ably these very difficult issues.

I would also like to say that because the authors chose to range over a very wide domain, they inevitably found it difficult to give adequate time to any one of the major issues they raised or to look at them in greater depth. In fact, just because of the way they have organised their paper, they have had to use rather general statements to make their points. It also seems to me, and I might be wholly wrong, that in discussing these controversial issues, they seem to be somewhat reluctant to set out clearly the conclusions of their analysis. It is of course one of their virtues that they always present fully both sides of the questions they raise, but, then, having done that, they withdraw from the conflict and seen to find comfort in sitting on the fence. But, perhaps this a matter of individual temperament; perhaps it is good to be cautious and modest even though that does not necessarily help social analysis.

Let me now to turn to some of the basic ideas in the paper. As we can see from the text, the core of the paper consists of an extensive discussion of five issues: one--democracy; two--the rule of law; three--bureaucratic performance; four- decentralisation and five--pro-poor institutions. The paper has argued that it is these five ingredients that constitute the essence of the relationship between governance and poverty. I will therefore now consider these ingredients one by one.

Democracy First. I propose to deal with the authors' discussion of democracy by responding to a question, perhaps a rhetorical question that they have raised. "The question", the authors point out, "that emerges from these arguments is have democratic governments in Pakistan been pro-poor?" The answer to this question, at least in my view and in the short and medium-term perspective is not very difficult to give. But before I do so, let me say straight away that I do not think "democracies tend to be more pro-poor than authoritarian governments", nor do I agree with the view that "democracies, more often than not, fail to deliver on pro-poor programmes". I doubt very much if it is at all possible to establish a causal link, positive or negative, between democracy and poverty (1) except perhaps when one is discussing democracy in an ideological vein.

As far as the specific question is concerned, I have no hesitation in saying that the experience of Pakistan in the years between 1988 and 1998, when more than a dozen unprincipled coalitions captured and lost control of the state machinery, was entirely bleak. In this period, growth rates slowed, unemployment increased, expenditures on education and health as a percent of the GDP stagnated and pverty increased. (2) The country also lost both the financial support and the moral respect of the international community. The fact of the matter is that in this period the economic decision-making machinery simply collapsed under the weight of its mediocre bureaucracy and its self-serving policies.

Regretfully 1 have to say that democracy was wholly betrayed by the politicians, their intellectual advisors and their pusillanimous bureaucracy. I doubt very much if we can name a principle of democracy that was not subverted by them. I believe that the country will have to begin all over again its experiment with democracy. But to do so, democracy will have to be redefined. The new democracy will have to address the two imbalances that threaten the solidarity of the country: the ethnic imbalance by qualifying majority rule and devolving real power to provinces and the economic imbalance by increasing the revenues of the state through appropriate taxation to increase social spending. Democracy is necessary because this is the only worldly political arrangement in which human beings can live with some dignity, relatively free from the fear of the matins of a modern state, and under which a citizen has some hope of protecting his/her economic interest.

The Second Ingredient: The Rule of Law: It goes without saying that rule of law is a critical element in the organisation of an orderly and prosperous society. Rule of law however is a very complex concept and its applications to social and economic affairs can not be divorced from considerations of efficiency or equity; it cannot be used just to defend any structure of property relations or to preserve any low-level social equilibria in the name of stability. At the same time it is also difficult to discuss property rights without having clarified what property means in a constantly changing technological, economic, legal and social environment and how rights to it can be secured in a lawful manner. I do not think that in this gathering 1 need to labour the points:

* that the structure of political coalitions is linked to the structure of property rights.

* that not all structures of property rights are consistent with widespread economic prosperity, and

* that some distributions of property rights are the biggest impediments to collective action by the poor.

In this context, I must also say that even though the authors have begun their study by a reference to human rights they have not made any serious attempt to give them an operational sense nor have they discussed any where the crucial role civil liberties play in organising collective action. I am afraid no discussion of poverty can be complete or meaningful without asking ourselves "By whom am I governed?" and "How much am I governed (3)?"

The Third Ingredient: Bureaucracy and Corruption. It is now part of our folklore that bureaucracies everywhere are self-centered, and that their utility functions are not always consistent either with those of their principals, i.e., the political governments or with those of the general public. The excesses of our bureaucracy cannot however be explained only by the considerations of the collective choice theory; we need to look both at history and the structure of power relations to understand them.

I hope I am right in saying that both in the colonial and the post colonial days districts have been managed by the representatives of the state more or less in collusion with the dominant local groups. The difference between the colonial and the current arrangements is that administrators are no longer the masters. In fact, over the years and all over the country bureaucracy has lost most of its autonomy and is now playing an uneasy, subordinate role. Quite clearly, this change has come about both as a result of the political and economic changes in the last half-century as well as the loss of the constitutional protections available to the colonial bureaucracy. This loss has been a particularly traumatic experience for the national bureaucracy, and it seems to me that a large number of civil servants are still in the process of adjusting to their new, insecure status.

Given what I have said before, I hardly need add that I agree with a great deal of what the authors have said about corruption, bureaucracy and good governance. But their statement that "the civil services should be taught that they are meant to implement policy and not make it" does not seem to me to be quite right. I would say that in the day to day conduct of the business of a government, it is rarely possible to draw a sharp line between policy and implementation: in real life what matters is not so much who makes a proposal but who approves it and who monitors its implementation. Bureaucracy cannot but participate in the making of a policy.

The Forth Ingredient: Decentralisation: I can recall that decentralisation has been a favourite themes of some civil servants and scholars since the 1950s, when US social scientists introduced Village-Aid programmes into Pakistan. I can also recall that since then it has been said over and over again by their critics that these are essentially political matters and that devolution and decentralisation can not make much headway so long as gross asymmetries of power, wealth, and coercive capacity persist, i can therefore only reiterate that without a political arrangement dedicated to civil liberties, an efficient and effective judiciary, periodic elections and extensive deregulation (i.e., reduction in state controls and state interventions), technical adjustments such as changes in financial caps, amendments in establishment rules and some reshuffling of some marginal functions will not help.

The Fifth ingredient: Pro-Poor Institutions: I confess that, like the authors of the study under discussion, I also look to the future for pro-poor institutions. But I must not lose heart. A journey of a thousand miles begins with a small step: only a few days ago elections to 21 union councils were held under the new dispensation and more are already in an advanced stage of preparation. The problem with these elections however is that non-party elections once again push citizens back into forms of relations such as baradari systems, caste systems, relations of economic dependence that more often than not retard rather than promote higher and wider loyalties. The redeeming feature of these elections is that they have provider a wider opportunity to women to participate in them. I hope that this concern for the welfare of women will continue, and that soon reforms will be introduced to ensure that in the future no one can deny them what is lawfully theirs: their title to their land, their right to the produce of their land and their wages for their labour.

(1) India, the largest democracy: 34.4 percent of population lived in conditions of absolute poverty. [NSS (1977)]. Pakistan calorie-based approach: 1993-94: 20.8. Nearly 16.5 percent of Americans lived in poverty despite the country's immense material wealth and generally higher per-capita income. The Human Development Report 1998. Population below $1 a day. Brazil 23.0 percent 1995, India 52.5 percent 1992, Pakistan I 1.6 percent 1991.

(2) Pakistan Economic Survey 1999-00.

(3) Isiah Berlin: Four Essays on Liberty.

Akhtar Mahmood

Formerly Secretary,

Government of Pakistan,

Islamabad.

Zafar H. Ismail and Sehar Rizvi are respectively Head of the Governance, Training, Monitoring, and Evaluation Units and Social Policy Specialist at the Social Policy and Development Centre.