Discussion Paper 57G: Capacity building for decentralised education service delivery in Pakistan

This paper should be cited as: Watson, D. 2005. Capacity building for decentralised education service delivery in Pakistan. (ECDPM Discussion Paper 57G). Maastricht: ECDPM. http://www.ecdpm.org/dp57G

David Watson
Accredited Governance Consultant
Department for International Development (DfID)

In association with

Adnan Qadir Khan
Public Policy Consultant,
Punjab Province, Pakistan

The UK Department for International Development (DFID) financed the field work for this paper

A case study prepared for the project
‘Capacity, Change and Performance’

Study of Capacity, Change and Performance:
Notes on the methodology

The lack of capacity in low-income countries is one of the main constraints to achieving the Millennium Development Goals. Even practitioners confess to having only a limited understanding of how capacity actually develops. In 2002, the chair of Govnet, the Network on Governance and Capacity Development of the OECD, asked the European Centre for Development Policy Management (ECDPM) in Maastricht, the Netherlands to undertake a study of how organisations and systems, mainly in developing countries, have succeeded in building their capacity and improving performance. The resulting study focuses on the endogenous process of capacity development - the process of change from the perspective of those undergoing the change. The study examines the factors that encourage it, how it differs from one context to another, and why efforts to develop capacity have been more successful in some contexts than in others.

The study consists of about 20 field cases carried out according to a methodological framework with seven components, as follows:

Capabilities: How do the capabilities of a group, organisation or network feed into organisational capacity?
Endogenous change and adaptation: How do processes of change take place within an organisation or system?
Performance: What has the organisation or system accomplished or is it now able to deliver? The focus here is on assessing the effectiveness of the process of capacity development rather than on impact, which will be apparent only in the long term.
External context: How has the external context - the historical, cultural, political and institutional environment, and the constraints and opportunities they create - influenced the capacity and performance of the organisation or system?
Stakeholders: What has been the influence of stakeholders such as beneficiaries, suppliers and supporters, and their different interests, expectations, modes of behaviour, resources, interrelationships and intensity of involvement?
External interventions: How have outsiders influenced the process of change?
Internal features and key resources: What are the patterns of internal features such as formal and informal roles, structures, resources, culture, strategies and values, and what influence have they had at both the organisational and multi-organisational levels?

The outputs of the study will include about 20 case study reports, an annotated review of the literature, a set of assessment tools, and various thematic papers to stimulate new thinking and practices about capacity development. The synthesis report summarising the results of the case studies will be published in 2005.

The results of the study, interim reports and an elaborated methodology can be consulted at www.capacity.org or www.ecdpm.org. For further information, please contact
Ms Heather Baser (hb@ecdpm.org).

Contents

Acronyms and glossary
Summary
1 Introduction
2 Devolution in Pakistan
3 Previous experience of capacity building in Punjab
4 Key capacities in the context of devolution
5 The influence of contextual factor
6 The influence of external interventions
7 Stakeholders and education service delivery in Punjab
8 Internal management factors
9 Capacity building experiences to date
10 Conclusions on capacity for capacity building
Appendix: List of individuals interviewed
References
Notes

Acronyms and glossary

ADB Asian Development Bank
AEO Assistant education officer (junior district education department field officer)
CCB citizen community board (new voluntary institution charged with monitoring service delivery and launching development projects)
CIET Community Information Empowerment and Training (opinion survey organisation)
CSO civil society organisation
CSP Civil Service of Pakistan (elite cadre abolished in 1973)
DAC Development Assistance Committee (of OECD)
DC Deputy Commissioner (until 1999 principal district executive with executive and judicial powers)
DCO District Coordination Officer
DfID Department for International Development (UK)
DMG District Management Group (elite cadre effectively replacing CSP)
DSP Decentralisation Support Programme (ADB capacity building programme: including TA oninter-governmental transfers; nationwide + province-based)
DTCE Devolution Trust for Community Empowerment (spearhead organisation for CCB development)
EDO Executive District Officer (head of department)
EMIS education management information system
FDTA Fiscal Decentralisation Technical Assistance project
GoP Government of Pakistan
HRD human resource development
M&E monitoring and evaluation
MNA Member of the National Assembly
MoE Ministry of Education
MPA Member of the Provincial Assembly
MPDD Management and Personnel Development Department (of Punjab government)
NAEP Northern Areas Education Programme (DfID-supported, 1998-2003)
NCHD National Commission for Human Development
NGO non-governmental organisation
NIPA National Institute of Public Administration
NRB National Reconstruction Bureau (principal architect of devolution policy and regulations)
NRSP National Rural Support Programme
NWFP North-West Frontier Province
OECD Organisation for Economic Cooperation and Development
PASC Pakistan Administrative Staff College
PEF Punjab Education Foundation (body to be supported in PESRP to promote dialogue on private education)
PESP Punjab Education Support Programme (possible DfIDP TA project related to governance and management of the education sector in Punjab)
PESRP Punjab Education Sector Reform Programme ($100 million sector adjustment credit - subject to possible future enlargement: World Bank credit wef 2003).
PIHS Pakistan Integrated Household Survey
PMIU Project Monitoring and Implementation Unit (of PESRP)
PPP public-private partnership
PRMP Punjab Resource Management Programme (public policy reform support programme: World Bank, phase I $200 million + $4 million TA)
PRHS Punjab Rural Household Survey
PRSP Poverty Reduction Strategy Paper
PSRP Public Sector Reform Programme
TA technical assistance
UNDP United Nations Development Programme
WB World Bank

Summary

This report examines capacity building for decentralised education service delivery in Pakistan. It is one of over 20 case studies connected with the project 'Capacity, Change and Performance' being compiled by ECDPM for GovNet, the Network on Governance and Capacity Development of the OECD's Development Assistance Committee. It attempts to apply the methodological framework adopted for the studies to consider the capacity building challenges posed in improving the delivery of devolved state education services in Pakistan. Two other related papers have been produced: an equivalent study in Amhara Regional State of Ethiopia, and a comparative analysis of the two cases, in which conclusions are drawn about the relative significance of the issues raised. DfID has sponsored this set of cases within the GovNet brief.

Devolution and education policy in Pakistan

Pakistan is classified by the World Bank as a low-income country with a GNP per capita of $410, and a Human Development Index ranking of 144 out of 175 countries. It spends only 1.8% of GDP on education (1998-2000), which amounts to 7.8% of the government budget. The average literacy rate is 44%, but this disguises the fact that only half as many women are literate as men. The primary school completion rate is 59%, but the female enrolment rate is only 74% of that of boys, and much lower in rural areas. Overall, it is one of the world's worst performing countries in terms of education coverage and outcomes. As a result of state neglect of education, the private sector has filled the gap. In Punjab, there are 63,000 state schools, and between 18,400 and 55,000 private schools.

Devolution in Pakistan has a long history, dating back to the mid-19th century in what was then colonial British India. Up to partition in 1947, the British granted local governments only very circumscribed functions. They were headed by an all-powerful deputy commissioner - 'an integral part of the central bureaucracy' - even after Independence. There is some evidence that their elected members were used to legitimise the 'essentially unitary' 1962 Constitutional arrangements or to build constituencies for the military. Non-party local electoral arrangements resulted in the localisation and personalisation of politics at local level. When provincial assemblies were revived in 1985 they were dominated by local politicians. Indirectly elected provincial ministers - faced with non-party constituencies - therefore started to use development funds to increase their chances of re-election. The continued lack of political ownership of local government has led, among other things, to a tendency for discretionary development programmes to be controlled by upper levels of government. Tensions between provincial and local tiers of government led to the suspension of local bodies from 1993 to 1998.

The eventual emergence of education as a national priority (in the National Education Policy of 1998 and the corresponding Punjab Education Sector Reform Programme, PESRP) has to be seen against a backdrop of neglect. Historical budget allocations have been low, resulting in chronic infrastructure shortages, and the low status of teaching as a profession. In Pakistan's political environment (where patronage is significant), politicians have preferred development schemes such as roads, electricity or water supplies, which have shorter payback times than education, and building new schools rather than improving existing ones. Political influence has been disruptive: new schools are sited in politically advantageous but educationally inefficient or irrelevant locations, and teachers are transferred as political favours.

Recent changes

The military government that took over in late 1999 introduced constitutional amendments, including limitations on the powers of provincial governments to amend Local Government Ordinances without Presidential approval. In August 2001 the President introduced a devolution plan, to be implemented through the National Reconstruction Bureau (NRB). The aim of the plan was - through community empowerment - to enhance bottom-up accountability, and thereby improve service delivery.

The role of the provinces was imprecisely defined in this plan. Emphasis was instead placed on new bodies outside government, including citizen community boards (CCBs) that were to apply pressure on (local) government, and identify and develop projects and initiatives. The latter qualify for funding from the 25% of local development budgets reserved for that purpose. The plan ushered in many changes, including Local Government Ordinances establishing over 6000 union councils. Their elected heads (Nazimeen) were to be members of district councils that would elect a district Nazim to supervise development and direct a district coordination officer (DCO) with no judicial powers. Seats on both union and district councils are reserved for women and vulnerable groups.

A major study, Devolution in Pakistan (ADB/WB/DfID, 2004), described this process as 'incomplete devolution', in part because authority over middle to senior grades of staff has not been devolved. Provincial (education) offices regularly give directives to district education offices, whose staff still 'look upwards' to the level of government that determines their careers. Given that much of district budgets (over 90%) are salaries and allowances, this also means that districts have little fiscal discretion. District roles in relation to provinces and federal authorities have been inadequately specified (at least in education), and conflicts between Members of National and Provincial Assemblies (MNAs/MPAs) and district councillors persist. As a contribution to the current debate, we present a list of functions related to education that we suggest should be performed at district level, obstacles to their achievement, and the measures included in the PESRP (see Table 1 on page 11).

Previous experience in capacity building

Despite this inauspicious context for building the capacities of provincial and local governments, numerous attempts have been made over the past ten years, through the public training institutes, as well as through development projects and technical assistance. We have not encountered any formal or 'official' definitions or discussion of capacity or 'capacity building'. The NRB Chair sees it as a process of popular empowerment to provide the confidence to hold government to account. The NRB sees no potential for capacity building in the bureaucracy without consistent pressure from communities for improved service delivery and for public accountability. Otherwise, particularly in the public sector, it is seen as (often off-the-job) 'training' to provide knowledge and skills to perform jobs better. Due to the over-supply of and reduced appetite for training, there are some indications that cynicism about capacity building is beginning to set in.

The disappointing results of capacity building initiatives tend to lend credibility to the NRB perspective. Needs analyses appeared to have ignored fundamental obstacles to the application of knowledge and skills derived from training. Several evaluations identified obstacles to the translation of training programmes into improved individual and organisational performance, or indicated that there was no evidence of behaviour change. There is no evidence that these evaluations and their conclusions have been shared or discussed, and in only one case was a decisive management action taken as a result. While the ADB's Decentralisation Support Programme (DSP) is attempting to develop a 'demand-led' market for training, there are no indications that applicants (including district governments) will be informed of the negative results of earlier efforts, and thus forewarned of the obstacles to improving performance.

The role of donors

Faced with a series of unsatisfactory outcomes from expensive capacity building initiatives, donors are becoming more aware of the shortcomings of their support. As a consequence they are adopting innovative approaches to:

stimulating the demand for capacity enhancement;
policy dialogue on operational aspects of devolution;
supporting 'third-party' (objective) evaluations of training;
experimenting with organisational development approaches to enhance local government capacities; and
analysing how existing devolution measures should be complemented in order to complete the policy reform process.

Factors influencing the development of capacity for service delivery

Based on the experience of capacity building policies and initiatives in Pakistan, the factors that appear to have influenced these largely discouraging outcomes include:

The bureaucracy still exhibits characteristics forged during colonial times. It is an elite, with a bias towards generalists, beset with a cadre system that is counter-productive, 'bottom-heavy' (with excessive numbers of unskilled staff), and prone to 'political' pressure and interference.

Morale among senior staff at district level has suffered from the significant deterioration in (real) remuneration, particularly compared to the private sector, their reduced authority in relation to elected Nazimeen, and conflict of loyalty to their provincial 'masters'.

There is little scope for staff management at district level. Few variables are within the competence of district authorities. Many of the 'stock' of district education officers are in any case unsuited for their managerial responsibilities: they are approaching retirement, have no management background, and are pawns in political manipulation from council members and the influence of MPAs.

The chemistry of the relationship between the (new) political player at district level (the Nazim) and the (newly constrained) district coordination officer is important as a determinant of how services are delivered. Provincial 'ownership' of federally championed reforms appears weak, yet it is integral to the success of devolution, and has considerable implications for the CCBs, and the work of the National Commission for Human Development (NCHD).

Incentives to perform either do not exist, or work perversely. The Devolution study devoted much attention to the issue of incentives to performance at all levels of the delivery system, particularly at the district level. It concluded that devolution has not made much impression on prevailing patterns of incentives. Mechanisms such as school monitoring committees to give communities a 'voice' at district and union levels are weak, and the new school councils in Punjab are largely untested.

Public expectations of the state education system are modest. Socio-economic, socio-political, literacy and gender disparities are pronounced. The recent National Rural Support Programme indicates that this has affected the capacity of local communities to forge social partnerships for a common purpose.

The budgeting system is dysfunctional, split into the recurrent budget (controlled by the province) and the development budget (very limited, often allocated without regard to the routine budget, or development priorities, but divided equally among council members, who then allocate it to 'pet' constituency schemes).

Teachers are important actors at provincial and district levels. In the past they tended to dominate the (now defunct) school management committees, but their performance, morale and professionalism have been affected by many factors, including the perceived (low) status of teaching as a career; poor supervision; weak local accountability; poor school infrastructure; and the absence of in-service training and support. As a result, teacher absenteeism is a now major problem.

Given the woeful performance of the state education system, the private sector is a major player in Punjab, but it is not adequately and constructively regulated and its capacities have yet to be enhanced by provincial governments. There are prospects for a rejuvenation of the Punjab Education Foundation under the PESRP.

Capacities for capacity building are limited. Predominantly 'static' traditional facilities provide training off-site, using methods that are unlikely to lead to behaviour change. There appear to be major problems in integrating public and private sector capacities, due to the absence of financial incentives within the public sector (they receive guaranteed budgets). This financing 'regime' inhibits attempts to forge partnerships between public and private/NGO training providers. The lack of consultants with public sector experience is another inhibiting factor. Socio-political and cultural rigidities affect the speed of response of trainees and organisations, especially at local government and community level, indicating the need for intensive work at that level, supported by professional facilitators.

Conclusions

No definitive causal connections can be made between improved performance, (past) capacity building inputs, or the changes brought about. Incentives for principal stakeholders and players matter in capacity development, yet they were ignored before the Devolution in Pakistan study brought them to centre stage.
The factors that limited the effectiveness of capacity building efforts in the past appear to be as real now as they were then.
Given the magnitude of the political imperative for community empowerment, speed is of the essence, but social re-engineering takes time. The recent experience of the National Rural Support Programme network with community organisations is salutary, but appears not to have been internalised in the strategy for establishing and supporting citizen community boards. The CCBs are crucial to the federal government's agenda, as well as designated conduits for one-quarter of future local development budgets.
Cynicism about 'training' is (justifiably) setting in; 'capacity building' is becoming a devalued currency. The recent $60 million programme of overseas scholarships for federal and provincial officials sends the wrong signals, having been approved by the World Bank in the absence of firm government commitment to far-reaching reform of the public sector.
The government faces a real capacity building dilemma. There is an urgent need for effective and rapid development of capacities in civil society (CCBs and elected councillors) and the public sector (to say nothing of the private education sector). However, there are severe constraints on enhancing capacities to design and deliver programmes that will induce sustained behaviour change and more effective organisational performance.
One fundamental underlying factor is that there appears to be little experience among key players of systematic training and organisational development. Another is that little learning about capacity building is taking place (about what works, what doesn't and why).
It appears timely for development partners to present to high-level audiences in Pakistan some relevant international experiences of effective capacity building, and how countries facing similar constraints have made some progress.

This report does not offer recommendations. At the request of DfIDP a separate report was prepared and delivered during the fieldwork. Subsequent dialogue has covered the need for, and the potential role of donors in sponsoring an international conference in Pakistan on the subject of capacity building. The final ECDPM synthesis study will contain a comprehensive set of recommendations on optimal approaches to capacity building, for government and donor consideration.

Epilogue, April 2005

At the request of DfID Pakistan, between December 2004 and January 2005 the authors compiled a supplementary policy paper based on the Pakistan/Punjab report for consideration by Punjab provincial government.¹ The paper focused on two positive case studies of capacity building in Punjab: the PMIU of the PESRP, and the Policy and Strategy Unit of the Strengthening Decentralised Local Government in Faisalabad Project.

1 Introduction

Pakistan is classified by the World Bank as a low income country with a GNP per capita of $410, and a Human Development Index ranking of 144 out of 175 countries. Only 1.8% of GDP is spent on education (1998-2000), which amounts to 7.8% of the budget. The average literacy rate is 44%, but this disguises the fact that only half as many women are literate as men. Its primary completion rate is 59%, but the female enrolment rate is only 74% of that of boys, and much lower in rural areas. Overall, it is one of the world's worst performing countries in terms of education coverage and outcomes. As a result of state neglect of education, the private sector has filled the gap. In Punjab, while there are 63,000 state schools, there are now estimated to be between 18,400 and 55,000 private schools.

This report examines capacity building for decentralised education service delivery in Pakistan. It is one of over 20 case studies connected with the project 'Capacity, Change and Performance' being compiled by ECDPM for GovNet, the Network on Governance and Capacity Development of the OECD's Development Assistance Committee. It attempts to apply the methodological framework adopted for the studies (see inside front cover) to consider the capacity building challenges posed in improving the delivery of devolved state education services in Pakistan. Two other related papers have been produced: an equivalent study in Amhara Regional State of Ethiopia, and a comparative analysis of the two cases, in which conclusions are drawn about the relative significance of the issues raised. DfID has sponsored this set of cases within the GovNet brief.

At a series of meetings in late 2003 ECDPM and DfID's Policy Division agreed to look at education service delivery in two decentralising, but very different country contexts - Pakistan and Ethiopia. This focus corresponded to DfID's concerns about pro-poor service delivery. These cases are the only ones in the DAC sample that analyse capacity issues in the context of public service delivery and decentralisation. The approaches adopted in the Ethiopia and Pakistan studies were similar,² including the preparation and agreement with DfID Pakistan and its principal counterparts in education services of ToR; the recruitment of a national consultant (Adnan Khan) with experience in the sector; and the preparation of fieldwork meetings and visits by Mr Khan.

The fieldwork was conducted by the two consultants from 1-14 July 2004. An issues paper was discussed with DfID Pakistan at the start of this phase. Meetings were held with many individuals at federal, provincial (Punjab) and district levels (see Appendix). The two districts in Punjab we visited, Sheikhupura and Okara, fortunately offered a complete contrast in terms of political and bureaucratic contexts and the challenges they face.

Figure 1. The provinces of Pakistan.

2 Devolution in Pakistan

In order to understand the context of and the constraints affecting attempts to improve the capacity and performance of the education service delivery system under the current (devolving) policy context in Pakistan, it is necessary to outline the origins and evolution of political players and governance structures in Pakistan.

2.1 Historical background

Devolution in Pakistan has a long history. 'In the areas which are now Pakistan, local governments … were introduced from scratch, following the annexation of Sindh in 1843 and of Punjab in 1849 … These local governments were formed in a 'top-down' manner in urban and rural areas with extremely circumscribed functions, and members who were not locally elected but appointed by the British bureaucracy' (Cheema et al., 2004). The post of Deputy Commissioner, which combined executive and judicial powers, survived virtually intact until 1999; the DC was 'an integral part of the central bureaucracy who, as chief executive of the district, emerged as the most important agent at the local level'.

After Independence, there was initially no enthusiasm for decentralisation - local government had played little role in the Independence movement. The only levels at which there was party political mobilisation was at federal and provincial levels. Martial law in 1958 saw a purge of politicians and dissolution of higher-tier elected governments. Locally elected union councils in rural areas and town/union committees in urban areas were established. However, the retention of executive powers under the DC, and severe fiscal constraints meant that these bodies were firmly under central authority. Their 80,000 members or 'Basic Democrats' were, under the 1962 Constitution, the electoral college of the President, as well as national and provincial assemblies, thus legitimising his 'essentially unitary' Constitution, or as he called it, 'controlled democracy' (ibid, p.7). The 'part-bureaucratic, part political system was used to distribute resources as patronage to secure a mandate for President Ayub Khan' and to build a constituency for the military, especially in rural areas, who had a majority of the electoral college. However, urban political discontent, especially in Punjab, led to the ascendancy of Bhutto's People's Party in the late 1960s, which held power until 1977.

Local governments were revived under General Zia, again as part of a legitimisation strategy. Once again, however, there was no substantive empowerment of local governments (i.e. they did not have access to, or authority over, significant resources). Military officers headed provincial governments. The tradition of non-party local elections was maintained. In the eyes of observers, this resulted in the localisation and personalisation of politics at the local level (ibid, p.10), and, after the revival of provincial assemblies in 1985, the use of development funds to boost the chances of ministers' re-election. Weakening party machines - one result of the Bhutto and Zia regimes - still keep this habit alive. Political parties in Pakistan are not known for their collective decision making. According to Cheema et al., 'individual bargaining between power-brokers' tends to be the norm.

The continuing lack of political ownership of local government has also meant that discretionary special development programmes controlled by upper layers of government have become one means of controlling allocations to local government. Tensions between provincial and local tiers led to the suspension of local bodies between 1993 and 1998, propelled paradoxically by democratic forces at provincial and federal levels. Politically controlled transfer of bureaucrats was a gambit adopted not only by the army, but also by subsequent civilian governments (ibid., p.14). Ministries and departments became 'employment agencies' with a free hand in recruitment, and the number of civil servants doubled between 1977 and 1987.

A new plan to devolve power was introduced in January 2000 and implemented after local government elections in August 2001. The driving motive of the reform was to improve service delivery responsiveness and accountability, and to empower communities. This new structure is illustrated in Figure 2.

Figure 2. Structure of government in Pakistan.

2.2 Recent changes

The main changes introduced in the 2001 devolution plan, as summarised by Cheema et al. (2004), are presented in Box 1.

The process of devolution has been characterised by a number of factors:
1. The extent of fiscal decentralisation (i.e. local government authority over resources) remains limited because no new taxes have been devolved to local governments; most district expenditures are 'establishment charges' that cannot be altered by the districts, because most middle- and senior-ranking employees at district level are, and remain, provincial employees, who cannot be fired by, or have salaries adjusted by the districts.
2. Inadequate efforts were made to integrate the newly elected local governments with the soon-to-be-elected provincial and federal governments, or to specify clearly their roles and interrelationships.³ Consequently, conflicts have arisen among local politicians and Members of the Provincial and National Assemblies (MPAs/MNAs), and competition for electoral credit for development schemes.
3. The extent of devolution varies within sectors. In the education sector, for example, universities and higher education remain provincial responsibilities.
4. Emphasis has been placed on the creation and strengthening of new bodies outside government, including citizen community boards (CCBs) that are to apply pressure on (local) government, and identify and develop projects and initiatives. The latter qualify for funding from the 25% of local development budgets reserved exclusively for CCB projects.

A number of other important factors have not changed since the devolution policy was introduced in 2001:

There has been no devolution of federal powers to provincial or local governments.
Local taxation instruments and bases remain basically the same.
Terms and conditions of employment for provincial staff stationed at district level. Over 90% of education department budgets are for teachers' salaries, yet these they are effectively controlled by the provincial Education Department and not the districts.
Although DCOs are accountable to the Nazims, their promotion and transfer are still matters for the provincial secretariat.
The tendency of provincial sector offices to give directives to senior staff in various sectors at the district level, without consulting the Nazim or DCO. Senior district education staff still feel accountable to the provincial offices, which decide on matters such as appointments, transfers and postings, and promotion.
Heads and members of union councils are still popularly elected.
Local governments still have no constitutional status or protection, except for that provided by the new 17th amendment, which allows provinces a period of six years to make changes to local government legislation (ordinances), with the concurrence of the President.

Box 1: Changes introduced in the devolution plan of 2001

A new elected government was created at the district level and politically linked to local governments at the sub-district (tehsil and union council) level. Union Nazimeen (mayors) and Naib Nazimeen (deputy mayors) are members of district and tehsil councils, respectively.

Under the new electoral arrangements, the heads of union councils are members of district councils (two-thirds of the membership); they are also the electoral college for the district Nazims and Naib Nazims. The other one-third of seats is reserved for 'peasants and women'. There is now a minimum educational qualification for politicians.

The district government is headed by the Nazim, who is elected (indirectly) by the union Nazimeen on the district council. The post of deputy commissioner is abolished and replaced by the District Coordinating Officer (DCO), who reports to the Nazim. This is a major change, since DCs reported to the non-elected provincial secretariat in the past. The DCO has no judicial powers. Thus local level administration is notionally accountable to elected officials at local level. For the first time, local governments are responsible for the delivery of most local services. Districts are responsible for establishing new schools, upgrading existing schools, and for maintaining and inspecting schools.

Planning, budgeting and development functions related to services have been devolved to the districts. Provincial financial transfers have become non-lapsable (i.e. they may be carried over to the following financial year) and 'rules-based'. The provincial finance commissions typically reserved 40% (in 2002/3) of their consolidated fund for districts, and established distribution rules that take into account the relative 'poverty' of districts (although its interpretation varies between provinces).⁴ No final formulae have yet been announced, even though this is a requirement of the legislation.

Source: Cheema et al. (2004).

2.3 Education service performance and standards

Pakistan has one of the worst records of any country in terms of investment in education - estimated by the World Bank to be only 1.8% of GDP in 1998-2000, compared with 4.8% in Ethiopia. Indicators of the poor performance of the education sector include very low enrolment rates (especially of girls, and at secondary level)⁵, the high primary dropout rate, the rural/urban divide, the extent of teacher training; teacher absenteeism,⁶ poor school infrastructure (two-thirds of schools in Punjab do not have boundary walls or adequate furniture, and one-third lack toilets), and the reliability of supplies.

Partly as a result of the deplorable state of the state-financed education system, and public dissatisfaction with its quality, the demand for private education at all levels, including primary, is booming. Even poor people are attempting in ever-larger numbers to send their children to private schools (especially at primary level). No less than 3.7 million children in Punjab are enrolled in private schools (3 million in primary schools), two-thirds of which are in urban areas. One advantage of private schools (cited by parents) is that 94% are co-educational.

The only indication of public satisfaction with service delivery⁷ is contained in the results of a nationwide CIET/NRB survey conducted in 2002 (GoP, 2003c).⁸ The survey also cast light on public contacts with politicians, the extent of teacher absenteeism, and popular involvement in voluntary groups at local level. Just over half the respondents were satisfied with government education services in their area. Children from vulnerable households were less likely to be enrolled in school. The gender gap for enrolment in Punjab (7%: 73:66) was the lowest in the country. Nearly one-third of children in households sampled attended private schools: nearly 90% of their parents were satisfied with the education provided (compared with only 73% of parents of children attending state schools).

Corruption is reportedly a problem in education service delivery. The CIET/NRB survey found that 7% of head teachers knew of teachers who had had to pay for their posting. Transparency International's (TI) 2003 survey of perceptions of corruption ranked Pakistan 92nd in terms of the prevalence of corruption, with a score of 2.5/10.00 (marginally better than the 2001 score of 2.3). The TI survey noted that '70% of respondents who interacted with educational institutions cited the existence of irregular methods of gaining admission. Teachers and the management committee were cited as the most involved. The main causes of corruption identified were a lack of accountability and low salaries' (World Bank, 2004b, p.90).

3 Previous experience of capacity building in Punjab

3.1 Concepts of 'capacity building'

Based on our literature survey and discussions with GoP officials and consultants/ practitioners at provincial and district levels, it appears that at least in Punjab, 'capacity building' is usually equated with off-the-job training, perhaps with the introduction of new systems, with or without corresponding procedures manuals and equipment. It appears that the emphasis is on conveying information on the assumption that participants will be more knowledgeable and skilful, and that they will be able to do their jobs better as a result.

Several experienced practitioners and observers of development in Punjab interviewed during this case study confirmed the above conclusion, and commented that there are fundamental misconceptions about what 'capacity building' entails. Only one interviewee, the head of the National Reconstruction Bureau (NRB), had a clear perception of capacity building as 'popular empowerment' via citizen community boards (CCBs), and the provision of information to ordinary people, to enable the users of services to have a greater 'voice'.

On the basis of our observations, needs analyses - if undertaken at all - are often based on brief observations of the broad types of skills the participants will need. Needs analyses appear to be rarely linked to the job context in which officers work, and do not explore factors such as:

the clarity of their assigned roles;
the degree of authority possessed by trainees/incumbents of posts;
their skills, qualifications, experience or confidence;
inter-cadre or inter-collegial and inter-institutional relationships;
the availability of support staff and facilities; or
motivational 'signals' given to incumbents, such as incentives to perform (or disincentives for non-performance).

In the Pakistan context there may be numerous perverse incentives. There may be (perceived) risks associated with the innovation implied in applying new skills, imaginative action or exceptional effort, or there may be few opportunities to apply training received (sufficiently soon after receiving it). Factors such as political interference, (poor) remuneration, the existence of (top level) commitment to change or systemic reform (if needed to permit application of training) appear to be rarely taken into account in the needs analyses we have seen.

Training methodology is not given much treatment in the literature on capacity building in Pakistan. Yet methodology matters - the effectiveness of learning depends on the extent to which objectives (in terms of changed awareness or behaviour) are clearly defined, shared and 'owned' by trainers and trainees. It also depends on whether the information imparted is relevant, accurate, up to date, intelligible and digestible, and on whether - if behavioural change is sought - new skills or abilities are demonstrated, build on existing practices or knowledge, are practised by trainees under supervision, and if feedback is provided to trainees on whether their performance meets - or falls short of - required standards. None of the material we have seen appears to tackle these issues explicitly. Only one programme provides insights into an effective training methodology in action: that of the CIET in training its survey enumerators (see Box 2).

Box 2: The CIET survey: an example of systematic training

CIET, the Community Information Empowerment and Training organisation, adopted a systematic approach to prepare its enumerators and supervisors for the nationwide survey of social service delivery in 2002. The training was based on pre-tested survey instruments and focus group guides; trainees included teachers and statistics department staff, NGO staff, university students and graduates. Only those who did well in the initial training were selected for field teams. The trainees practised administering questionnaires, including role playing interviews and handling difficult situations. Half a day was spent on field-based practice, under close supervision and monitoring, and feedback was provided. The final day featured a full mock data collection exercise in the field. In cases of unsatisfactory performance, the training was extended until the trainers were satisfied.

In all, 200 enumerators were trained for a ten-district pilot programme, and over 500 for the 2002 nationwide survey. Guidance, monitored practice and field simulations were also used to train staff to lead focus group discussions. Careful work was a key performance criterion for data collection - staff who performed badly were dismissed and replaced - and on-site supervision was always available. In one district (Balochistan), the process of 'socialisation' of results is being demonstrated, lessons are being learned and a 'toolkit' is being produced to help extend the process elsewhere, at CCB and district (planning) staff levels.
Source: GoP (2003c, pp.12-22).

We found no evidence of discussions amongst GoP stakeholders or donors on what capacity building is, what the lessons of experience are, and what they imply for a future strategy to support devolution. There have been no seminars on the notion of capacity building itself. Thus, all the new capacity building activities appear to have taken place without reference to earlier experiences in Punjab, Pakistan or elsewhere. The results of evaluations have apparently not been disseminated or used in subsequent exercises of a similar nature.

3.2 Major capacity building initiatives in Pakistan

There have been many national (federal government), provincial, and province-specific capacity building programmes in Punjab. Federal programmes impinge significantly on the emergence of devolution policy and practice in Punjab.

Before effective and systematic capacity building can take place, needs assessments are required. These should assess the performance deficiencies indicating that capacities are inadequate, identify capacities that exist and those that are needed, and identify the nature and magnitude of the 'gaps' between needs and current capacities. They should also encompass and draw conclusions from the contextual and motivational factors mentioned above. We describe the apparent approach to capacity building adopted because this is a determinant of the effectiveness and magnitude or extent of the building of capacities. To the extent that data are available, we mention the (ex-post) impact of capacity building in terms of changed performance of trainees or the system in which they were working. This aspect is sometimes confused with the opinions of trainees on the relevance or appropriateness of the training.

Box 3: Capacity building providers

Capacity building institutions in the public, NGO and private sectors in or accessible to Punjab include the following:

Federal: Civil Service Academy, National Institute of Public Administration, Pakistan Administrative Staff College.
Provincial: management and personnel development department.
NGO: Institute of Rural Management (under the NRSP).
Private sector: various management consultancies.
Academic (Punjab only): University of Education, University of Punjab, more than 20 private universities.

3.3 Capacity for capacity building

In view of the acknowledged huge need for capacity building in Pakistan's devolution, the lack of capacity for capacity building is a major issue that does not appear to have attracted the degree of attention it merits. Box 3 lists the main providers; we then describe some of the problems that appear to be emerging, and initiatives to address them.

Evidence regarding the performance of capacity building organisations is limited. There have been no objective assessments of the effectiveness of any of the training programmes offered by federal or provincial training organisations. In Punjab, none provides on-site training outside Lahore. There are, however, indications that in supporting district government capacity building a major challenge will be to find consultants or trainers with experience of local government systems and manner of operation. For example, the Fiscal Decentralisation Technical Assistance (FDTA) project completion report noted that very few national consultants had any local government experience, and so had major difficulties not only in understanding the 'as is' situation and the feasibility of applying new or modified financial systems, but they were also unable to offer (worked) examples of the application of new approaches in the off-site training they provided. The (ex-private sector) consultants engaged in studies for the Punjab Resource Management Project (PRMP) reported a similar 'culture shock' in encounters with public sector officials at provincial and especially district levels.

Recent interventions to strengthen capacities for capacity building are summarised in Box 4.

Box 4: Initiatives to build the capacities of capacity builders

Public Service Capacity Building project
Besides offering overseas scholarships, the project also includes an institutional strengthening component for training institutes such as the Pakistan Administrative Staff College (PASC), National Institutes of Public Administration (NIPAs), and the new National School of Public Policy (NSPP). The NSPP will serve as 'the premier training institution for civil servants', and will support the Civil Service Reform Unit of the Ministry of Establishments. The project acknowledges that one of the lessons from earlier attempts at reform is that 'sustainable capacity building requires that broader civil service reform issues (incentives, compensation, etc.) be addressed and that consensus building amongst stakeholders should be undertaken' (World Bank, 2004b, p.29).

Management and Personnel Development Department (MPDD), government of Punjab
The MPDD was formed from the merger of several provincial training institutes, and has received (limited) external assistance. Based in Lahore, and financed by the provincial budget, its core staff are on deputation (rotation from departments) and it uses officers as visiting lecturers. It has no routine mechanisms for evaluating the effectiveness of its courses, for analysing training needs, or for assessing the demand for training.

Decentralisation Support Programme (DSP), Punjab province
The DSP encourages coalitions or partnerships between public training institutes (PASC, NIPA and MPDD) and private sector or NGO providers to bid for packages of courses that will form part of the DSP's capacity building component.

NGO capacity building
Several organisations are helping to build the capacities of NGOs and community organisations in terms of management, and their work with communities to improve their self-reliance and local development. However, larger NGOs (including Action Aid and members of the National Rural Support Programme network) stress that these approaches require on-site liaison between them, trainees and communities over long periods of time. Few short-cuts are available. Capacities at this level will evolve gradually and uncertainly. There is little traditional membership of voluntary organisations in Pakistan. According to the 2002 CIET survey, very few respondents had heard of the CCBs, but after hearing about them from enumerators, 50% of men and 30% of women professed themselves willing to join them.

4 Key capacities in the context of devolution

In order to define key capacities in education service delivery, one has to be clear about the education-related functions performed at various levels of government. We consulted various actors at federal, provincial and district levels about their functions, particularly in the context of devolution, who commented that their distribution was at best ambiguous, at worst unclear. The literature on the issue is not very helpful.

4.1 Federal roles (in education)

One interviewee, an authority on constitutional matters and rules of business, indicated that no functions are assigned exclusively to provincial governments: there is no formal distinction between federal and provincial governments. Education is on the 'concurrent' list of responsibilities (i.e. federal/provincial).

In a recent review of public expenditure, the World Bank (2004b) cited what it interpreted as a 1973 Constitutional provision that 'education is a sub-national responsibility in Pakistan', but commented that 'the federal government has assumed responsibility for formulating national policy, planning, curriculum development, managing centres of excellence, and financing university education'. The report goes on to say that 'federal government should remain concerned with providing strategic direction, setting minimum national standards and providing finance to provinces conditional on their compliance with these standards' (p.87).

Academic observers have commented that 'there has been no devolution of federal powers to provincial or local governments' (Cheema et al., 2004). Our discussions indicated that no serious steps had been taken to reflect on the optimal distribution of tasks between federal and provincial governments, although in early 2001 (before the promulgation of the Local Government Ordinance) the MoE did undertake a series of consultations on the distribution of functions among the three levels of government. The record of one of these events indicates that while there was debate about provincial and district roles and the interactions of those levels, there was no discussion of what should happen to the roles and relationships between the federal ministry and provincial departments of education (SAP, 2001).

A recent paper from the Decentralisation Focal Point officer in the MoE (Qureshi, 2004) indicates that a decentralisation audit is planned, which will identify issues and 'the roles and responsibilities of the MoE and provincial government ...'. It also states that the Ministry and its Academy for Education, Planning and Management have embarked on a series of district (education) capacity building needs analyses,⁹ and the delivery of programmes to meet them. There appears to be a gradual transfer of a few functions to the provinces - e.g. for liaison with donors regarding new education projects - while existing projects will continue to be managed centrally.

There appears to be no immediate prospect of major structural change in federal ministries to respond to their changed (ostensibly reduced) roles under devolution. There is no mention of such a change in the recently approved $61 million Public Sector Capacity Building programme of the World Bank, even though functional devolution and tax re-assignment from federal to provincial governments is a commitment in the PRSP, and stated as a GoP intention at the Pakistan Development Forum in 2003 (cited in ADB/WB/DfID, 2004).

4.2 Provincial government roles (in education)

It has been difficult to find a definitive statement of what exactly are provincial functions pertaining to education services. They differ between provinces. The Local Government Ordinance (2001) for Punjab is silent on this matter. There is a statement of provincial and district functions in a province-by-province study of decentralisation in education (SAP, 2002), which notes the actual functions accorded to provincial and district governments in the early days of devolution.¹⁰

On the basis of our consultations, the key roles that provinces should play, and thus should be the focus of efforts to develop capacities, include the following:

providing sectoral policy guidance, measuring and monitoring the districts' performance in education service delivery;
establishing and maintaining an education management information system;
designing and assisting with the implementation of capacity building schemes and approaches for district and sub-district staff and politicians, including encouraging comparisons of 'best practices' in terms of service delivery, performance, and ideas for improving the community/provider interface;
ensuring the equitable distribution of (federal and provincial) resources among districts taking into account national minimum service standards, applicable (variable) unit costs of adhering to them, and past performance in service delivery (these could be in the form of conditional grants);
(re)assigning some provincial taxes to the districts to provide them with more revenue-buoyant tax bases - including the retention of user charges;
curriculum development (including textbooks and learning materials);
identifying and designing (with district inputs) sectoral development programmes suitable for federal or international funding;
providing (pre-service) teacher training and support for district-based in-service training; and
supporting districts with policy related to regulation and monitoring of private sector schools.

In the view of several commentators, provincial governments are de facto still functioning as employers of senior executive staff deployed at district level, managing sectoral development programmes funded from federal/ provincial sources, and issuing directives to districts on policy matters, which they should not do in the present devolved environment.

The Devolution in Pakistan report suggests that the GoP demonstrate its commitment to devolution from federal to provincial, and from provincial to district levels (ADB/WB/DfID, 2004, vol.1, p.63).

4.3 District level functions

On the basis of our discussions, there appears to be lack of clarity (amongst provincial and district staff) about priority district education functions. In particular, there is a general lack of appreciation of the wide range of functions implied in the current devolved arrangements. As a contribution to the current debate, Table 1 lists the priority functions that we suggest need to be performed at district level, the current impediments to these functions, and the measures included in the recently commenced Punjab Education Sector Reform Programme (PESRP) or in other programmes.

Table 1. (Suggested) priority district functions, impediments to performance, and measures in the
Punjab Education Sector Reform Programme (PESRP) designed to address them.

High-priority function	Current impediments	Measures in the PESRP
Preparing an education strategy for the district related to the district development plan	Very new demand implied by devolution: no experience or skills at this level of doing so.	No direct capacity building inputs in this area the current (WB-supported) package. DfIDP is pursuing dialogue in preparation for a possible complementary Punjab Education Support Programme (PESP).
Planning, construction and distribution of (new) schools, by type	Low volume of development budget transferred from provincial level. Low priority attributed to education in annual development plans by politicians.Little experience of procurement and tendering; even under GoP regulations school councils are not empowered to engage in construction.	Not an immediate priority in the PESRP (see below).
Rehabilitate and supplement infrastructure for existing schools	As above. Low priority even in education development plans for this function: political preference is for new facilities.	Major priority in PESRP: approx two-thirds of first years' programme budget. This is to be provided in the form of conditional grants, subject Terms of Partnership agreements (ToPs).¹¹ As an incentive, future allocations to districts will vary according to performance. to
Preparing education sector (development) budget	As above. Annual plans within first three years of PESRP finally determined by steering group at provincial level, influenced by MPAs - may not reflect immediate district priorities.	Capacity building planned under the PESP, or later in the PESRP.
Preparing education sector (recurrent) budget	Over 90% of available resources required for salaries.Most staff and teachers involved are provincial (and a few federal) employees.	Textbooks to be supplied free to state schools (6-8% of budget), plus a transport allowance to facilitate their distribution from tehsil stores.
Recruitment and employment/ deployment of staff and teachers	As above (only very junior staff within authority of districts).No general recruitment feasible for several years because of recruitment ban. Teachers' basic qualifications only diploma level; teaching seen as 'last resort'. Political interference in teacher recruitment, transfers and postings.Political power of teachers' unions at provincial level.	Minimum entry qualification = degree graduate. New policy introduced on (time and school-specific contract basis), within district discretion.
Performance monitoring and discipline of staff and teachers	Poor information available from school level on this.Provincial employees virtually impossible to dismiss.Political protection for teachers who complain of disciplinary action by the school councils.School councils not oriented or empowered to monitor or act. See 'inspection' below.	Possible assistance under PESP.Problem remains with majority of existing (provincial) teachers.Some school councils may be empowered on pilot basis to act in relation to recalcitrant teachers (National Rural Support Programme/NGOs will provide support in six districts).
Technical (pedagogical) inspection and monitoring of schools	Weak, irregular or corrupt supervision of schools and teachers by assistant education officers and 'learning coordinators'	Possible assistance under PESP.
Procurement and distribution of supplies to schools	Very limited financial authority (under legislation dating from 1962). Under devolution, EDO(E) is a category 2 officer, and authorise expenditures only up to Rs.2500 (approx. $35).Inadequate resources for non-teaching expenditures, so previous procurement experience is limited.Laborious expenditure authorisation/ procurement system.School-based budgeting only in pilot stages.	Under new amendment as a 'chief purchaser', the authority of EDO(E) is raised to Rs.600,000 (approx. $7000).¹² Possible support from PESP?Being tackled eventually under PRMP? Problem identified in World Bank (2004b).

High-priority function	Current impediments	Measures in the PESRP
Awareness raising schemes and campaigns (esp. re girls' education)	Inadequate resources hitherto.	Rs.150 million per year earmarked in PESRP for campaigns. Stipends for girls Rs.350 million in year 1, and Rs.450 million in year 2 in 15 worst districts for female enrolment.
Promotion of the role of school councils in oversight and school management	Unsuccessful experience with school councils set up under SAP.	Major emphasis: Rs.1.2 million in first 2 years. Funds to be transferred to SC bank accounts; concessions made avoiding procurement regulations. Only a few prohibitions on use of funds.
In-service training (IST) and career development for head teachers and teachers	Dependence on provincial initiatives hitherto. Large scale, expensive but ultimately abortive IST effort 2002-3.	Rs.500 million in first two years.
Registration, supervision, inspection and support for private schools	Dysfunctional registration system; no effective supervision or inspection capacity; rent-seeking behaviour reported.'Old' Punjab Education Foundation dysfunctional.	'New' Punjab Education Foundation, with an independent board, will be resourced up to Rs.400 million in first two years to support private sector.
Compilation of data for an education management information system (EMIS)	No district 'ownership' of system up to now. Generic weaknesses in data collection (e.g. enrolment counted on just one day each year)	Developments in PMIU in database of province-wide state school infrastructure and development plans. Limited support for EMIS, including equipment and staff capacity building.
Conduct exams (5th and 8th classes)	No functioning national assessment system.	Work ongoing at federal level.
Internal audit and supervision of external audit queries	Late submission of external audit reports.Weak internal audit function/ capacities.	Ditto.Will be tackled under the Punjab Resource Management Programme (PRMP)?
Handling legal claims from teachers	Time-consuming duty at present: districts have maximum one legal adviser: overwhelmed	Unlikely to be addressed in short term; 'stock' of teachers will continue litigation tendency.
Preparing reports on strategy, policy implementation and problems to education monitoring committees	Monitoring committees are present and functioning in only a few districts.Despite the provisions of the Punjab Local Government Ordinance 2001 forbidding contacts between elected members and council staff, or pressure on them regarding staff or administrative matters. In practice this is the main means of communication.	Some attempts have been made to establish monitoring committees, but it is likely that the main effort will be made in the Devolution Trust for Community Empowerment/CCB capacity building programme.

5 The influence of contextual factors

In this section we attempt to trace the influence of contextual factors - historical, political, social and bureaucratic - on the emergence of capacities for performance in education services. Section 2 sketched the main elements of the emergence of decentralisation in Pakistan, but it did not capture in detail the influence of the British colonial period on its bureaucracy, which has been studied exhaustively (see Kennedy, 1987).

5.1 The bureaucracy

A number of (largely historically derived) characteristics of Pakistan's bureaucracy have most import for performance and capacity:

It is an elite. In the foreword to the study by Kennedy (1987), Hashmi comments that 'Since its birth, Pakistan has been governed by its elites, mainly bureaucratic, military and political. The bureaucratic elite who had formed an important element in the power structure since the very creation of Pakistan gradually became more assertive, frequently gaining their steadily increasing power at the expense of the political elite … higher bureaucracy has ruled the country; in fact, Pakistan can be described as a bureaucratic polity'.

There is (still) an entrenched preference for generalists. This has affected both recruitment and the pre-eminence enjoyed by the generalist cadres (see below) - the Civil Service of Pakistan (CSP) and its successor, the District Management Group (DMG). Kennedy (ibid., ch.7) concludes that the relationship between generalists and technocrats is 'zero-sum' - e.g. a gain (of a job category) by the technocrats is offset by a loss to the generalists.

It is beset by a cadre system, introduced in the late 18th century, in which officers are assigned to an occupational group, where they remain for the remainder of their service. In Kennedy's view, this has impeded the process of administrative reform, and has contributed to administrative inefficiency since it is only marginally related to job performance, and encourages a type of 'politics' in the bureaucracy. Although the CSP and its cadres were abolished by the administrative reforms of 1973 (and replaced by 'occupational groups'), as were reserved posts for members of certain cadres, and 'service organisations', Kennedy notes that there has been no change in 'the contextual importance of the cadre system to the bureaucracy [...] Reforms did not challenge the nature of the cadre system per se, but rather the prevailing relationships between members of different cadres'.¹³ He notes that while the DMG has declined, the significance of financial cadres has increased. However, 'the limited successes of administrative reform have been paid for dearly by the bureaucracy of Pakistan'. The negative effects have included the politicisation of the bureaucracy and declining morale; 'shunning of responsibility for making decisions'; and the 'depletion of the gene pool' in that the 'best and the brightest' no longer regard the civil service as the optimal career choice.¹⁴

The World Bank analysis highlights other characteristics of the bureaucracy, including:

The service is bottom-heavy, in terms of numbers of junior compared to senior staff,¹⁵ and in terms of the relatively favourable pay of the lower grades. This in turn, gives rise to …
Pressure for patronage-based employment of unskilled staff. According to the World Bank study, this has had a damaging impact on performance of the civil service. The study noted that a retrograde step was taken earlier this year when, according to a decision of the federal cabinet, the federal Public Service Commission is no longer to be involved in supervising the process of recruiting staff to grades 11-16. Thus provincial service commissions would henceforth be free to recruit staff without supervision from federal level.
Political pressure is a major factor in the continuing high level of transfers of senior staff. The Bank reports that 'the problem is clearly chronic at the provincial level'.¹⁶ This extends to districts, because the provinces retain the power to appoint, post and transfer all staff above grade 16 (which includes education officers and head teachers). 'This is a highly politicised arrangement to influence transfers throughout the districts. By the threat of an unpleasant transfer, or the promise of an attractive one, a senior staff member can be put under pressure to accede to the transfer of junior staff', including schoolteachers (World Bank, 2004b, p.101).

5.2 Political influences on education

Section 2 outlines a political history marked by the absence - until the last three years - of any significant focus on, or interest in, real resource-backed devolution of power to local government. In this context, another reality has impeded the development of basic education services provided by local governments. The low level of resources allocated to the sector has been a major factor in poor performance, and reflects the low political priority attached to state education in the past.

There is no evidence of improved allocations of (untied and unconditional) resources to basic education at the local level.¹⁷ The district staff we met were grateful that the Punjab Education Sector Reform Programme (PESRP) had made a conditional grant of Rs.150 million to each district (as the first annual instalment of development funds) - it was the first significant injection of resources for basic education other than for new schools they could remember. This relates to the underlying problem of 'visibility' and the apparent attribution of development schemes to individual politicians.¹⁸ Given their limited time in office, they prefer highly visible, fast 'payback' schemes. Improving the quality of education takes time, but apart from building new schools, education is not an attractive 'political' investment. Urban infrastructure, rural water supply and electricity provision, tend to be more in demand by voters, and therefore politically more attractive.¹⁹

5.3 The socio-economic context

The socio-economic context in Pakistan has an influence on the development of key capacities needed to improve education service performance and delivery. Factors include the relatively low literacy rates, particularly among women and in rural areas, widespread poverty and gender disparities - women are among the most vulnerable groups in Pakistan.

There appears to be a dearth of information about, and low public expectations of, public service standards in Punjab. One (rural district) Nazim reported that he had worked with the CIET survey team in interpreting the results, which indicated that rural people were more content with education services than urban dwellers. He attributed this to the fact that relatively less choice is available in rural areas (most private schools are in urban areas), so that expectations are more modest.²⁰

All of these factors affect the feasibility of enhancing the accountability of service providers, which international experience indicates is one of the primary determinants of performance. We heard how the operation of school management committees (SMCs) in Punjab had been negatively affected by the parents' unwillingness to criticise teachers, let alone head teachers (who chair the SMCs), who were regarded as educated professionals to whom deference should be accorded. Teachers' unions have been historically averse to parents having a say in 'professional' matters such as how schools are run, and are opposed to contractual employment arrangements that would result in tighter accountability (see section 7.2).

Politically active families still dominate provincial and, to some extent, local political life. There is no tradition of mass action or popular pressure on either politicians or service providers, especially in rural areas. Political accountability is weak. There is a tradition of vote-buying, either directly or indirectly - by constituents who request the assistance of politicians in securing (public sector) jobs, or politicians who seek to buy allegiances based on their success in 'delivering' new projects.

6 The influence of external interventions

As is clear from earlier sections, external agencies have played a major role in attempting to build capacities and improve education system performance in Pakistan and Punjab. This section identifies the main external interventions, and assesses the evidence about their effectiveness.²¹

6.1 Social Action Programme (SAP)

The largest and most complex attempt to influence education service delivery was the World Bank's Social Action Programme (SAP), launched by the government in 1992/3 to redress historically low allocations of funding for, and the poor quality of social services. The Bank provided financial (and some TA) support to the GoP on an expenditure reimbursement basis, to provide an incentive and leverage for increased spending on social services that would benefit the poor, including basic education, preventive health care, water and sanitation. Total investments in SAP I (up to 1996/7) and SAP II (to 2002/3) are estimated at $10 billion, about 80% of which was from GoP (ADB/WB/DfID, 2004). The objective of SAP II (1998-2002) was to build government capacities for planning, monitoring and implementing social service programmes.

Evaluations by the World Bank (2002) and Asian Development Bank (ADB, 2001) agreed that the impact of SAP in terms of service delivery performance was disappointing, especially in the education sector. Education was the largest sector, and Punjab benefited disproportionately compared to other provinces.²² The ADB (2001) reported that SAP I had not had its intended impacts in terms of gross enrolment rates or numbers of new schools. Between 1990 and 1996 the primary enrolment rate declined marginally rather than increasing by 27% as planned, but gender differentials were improved, and there were nearly 15,000 more primary schools in 1996/7 than in 1992/3 (an increase of 12%, but only half that planned). The number of teachers increased by 40,000, but the proportion of female teachers remained constant.

During SAP II most performance indicators remained stubbornly flat, or even declined. Between 1995 and 2002 primary enrolment in Punjab fell from 78% to 76%, as did female enrolment from 64% to 61%.²³ The national net primary enrolment rate remained constant between 1998 and 2001 (42%), while in Punjab it increased by 1% to 45% (for girls it increased from 40% to 43%, mostly in rural areas).

Third party validation reports²⁴ were reportedly useful in tracking reforms in SAP II, but showed negative trends: between 1998 and 2001 compliance with objective recruitment criteria fell from 72% to 64%, and compliance with procurement procedures fell from 59% to 36%. According to a survey in 2001 the rate of male teacher absenteeism in Punjab was 22.4% - the worst in the country (PRHS, cited in World Bank 2002, p.35).

On the positive side, SAP did have some impact in terms of changing mindsets and raising awareness of the extent of past neglect. Henceforth, education and health sector expenditures were to be given some protection, and efforts were to be made to improve existing facilities rather than to build new ones. Checks on absenteeism, and audits of the existence of schools and teachers on payrolls, led to some innovations, including location-specific contracts for teachers, and scholarships for girls (funded by provincial governments).

A full treatment of the reasons for the failure of the SAP is beyond the scope of this report, so we focus on the role of external agencies and capacity issues. The main factors appear to be as follows:

the unbalanced 'ownership' referred to above was undoubtedly a factor;
the excessive complexity of SAP II in view of the limited capacities of provincial governments (which the Bank was warned about in SAP I);
unrealistic institutional goals and the short timeframe for achieving them;
procurement implications, especially at district level, never took into account the lack of experience and basic capacities;
manuals were not issued to guide those responsible for complying with them;
the focus on a few components (i.e. primary, not secondary education) led to imbalances in sector investment;
programme performance in improving the quality of education was to be judged by comparing standard student test results, but the arrangements for conducting such tests were not put in place or thought through;
performance indicators were monitored only once, although this should have been an annual process (but no annual targets were specified), and there were no regular reviews, and thus no opportunities for discussing problems;
the Bank proved unwilling to use its leverage, and to impose sanctions. Even though several donor missions²⁵ pointed out the GoP's failure to meet its commitments (related to non-salary budgets, procedures, recruitment policies, absenteeism, continuity of tenure for key staff, and decentralisation of authority), the Bank continued to fund the SAP.²⁶

6.2 Other external interventions

We cannot comment definitively on the apparent relationships between other externally supported programmes and interventions and education system performance, because there is no evidence. The Northern Areas Education Programme (NAEP) impact assessment commented that there was no evidence to verify improvements in management behaviour. The Fiscal Decentralisation TA project had not been subject to an additional ex-post assessment that could have offered evidence of how it had contributed to the operation of newly decentralised financial systems. DfID's (relatively modest) support for performance-based budgeting in Punjab has apparently not led to major changes. However, the work in 2003 and 2004 will stand the province in good stead as the Punjab Resource Management Programme (PRMP) again begins to take up the challenge. All other externally supported programmes are still at an early stage of implementation.

Generally positive external view of decentralisation
Despite the World Bank's initial nervousness towards decentralisation in the later stages of SAP, it appears that the new government's decentralisation policy has attracted significant external support - from the ADB for the Decentralisation Support Programme (DSP) and from the World Bank for the Punjab Education Sector Reform Programme (PESRP) and the Punjab Resource Management Programme (PRMP). In response to an invitation from the GoP to assess progress with decentralisation, and to suggest measures to facilitate future progress, the principal donors undertook a major study, Devolution in Pakistan (ADB/WB/DfIDP, 2004), which proved extremely costly in terms of staff time. A decision was made early in the process to pursue the study without the use of consultants. This was a remarkable break with past practice, and a measure of the strategic significance of the study. Whether it will have an impact on future policy evolution, and in particular on the role of incentives in local government performance (a major focus of the study), remains to be seen. Anecdotal comments from some of those involved in the study indicate that their work may have begun to have a positive impact, within the external organisations, on staff attitudes to decentralisation support (in Pakistan and elsewhere).

Encouraging (evidence-based) policy dialogue
External agencies are playing a useful role in providing data and forums for policy dialogue. The 'third party' evaluation of the Punjab In-service Teacher Training (ISTT) programme would not have happened if DfIDP had not offered financial support. In response to the disappointing conclusions of the study, the GoP's (then) education secretary cancelled plans for further ISTT programmes based on the cascade model used.

The DSP has begun to arrange policy dialogues on the implementation of decentralisation. Just before the fieldwork for this case study, the Punjab DSP organised a forum on the operational difficulties of establishing and financially supporting CCBs (with plans for three other provincial dialogues, leading to the submission of a consolidated report to the National Reconstruction Bureau, NRB). It is difficult to estimate the eventual impact of these dialogues on future policy, especially in circumstances where the NRB has been very wary of initiatives that will open up important elements of the decentralisation strategy for scrutiny by the bureaucracy.

In Faisalabad the district council has approved the establishment of a 'strategic policy unit', based on proposals of the DfID-supported consultant team. Its role will be to develop a coherent strategic vision and to pull together sectoral plans for the development of the district. This is the first such unit (and attempt to develop a district strategic plan), and was initially criticised by the NRB as being a 'parallel' institution to the council's Mushavirat committee, which ostensibly does the same things. These comments ignore that fact that these committees, which are part of the Punjab Local Government Ordinance (Section 140), are dysfunctional due to political resistance to their operation.

Programme management units
The issue of programme management and implementation units (PMIUs) is beginning to emerge in Punjab, but has not yet attracted systematic study. The PMIU is the most appropriate vehicle for interaction between external TA teams and counterpart departments or organisations in the government structure. The organisational development model adopted in Faisalabad is much rarer. Here, consultants work alongside counterparts within targeted departments, and facilitate their efforts and interactions with colleagues within and outside the district government, rather than undertaking tasks on their behalf.

We understand that there was discussion of how the PMIU of the PESRP could best address the major task of setting up a database. The use of international consultants was considered, but ultimately rejected in favour of the establishment of and provision of (donor) support to a team drawn of provincial government staff to undertake the task. The latter option appears to have worked well so far.²⁷

The role of donors
Faced with a series of unsatisfactory outcomes from expensive capacity building initiatives, donors are becoming more aware of the shortcomings of their support, and are adopting innovative approaches to:

stimulating demand for capacity enhancement;
policy dialogue on operational aspects of devolution;
supporting 'third-party' (objective) evaluation of training;
experimenting with organisational development approaches to local government capacity enhancement;
analysing how existing devolution measures should be complemented to complete the policy reform process.

7 Stakeholders and education service delivery in Punjab

We turn now to the roles, attitudes, abilities and interrelationships of some of the stakeholders as determinants of the interplay between capacity, change and performance in the delivery of education services in Punjab.

7.1 Central/federal level

National Reconstruction Bureau
The NRB is the spearhead organisation for the devolution reforms. It has produced an impressive set of guidelines, which include model local government ordinances, a 'step-by-step' guide to fiscal devolution and trainers' guides for use in capacity building activities. It has championed the establishment of CCBs, seeing them as vital if - for the first time in Pakistan - real demand-led pressures for accountability and improved service delivery are to stimulate capacity building in local government. To this end, the NRB supported a survey of social services delivery (GoP, 2003c) and has forged alliances with civil society. Recognising the importance of the media, the NRB has suggested that they report positive success stories in devolution. Professional networks are also important, and the Bar Association has supported the NRB in launching class action lawsuits to give legal backing to demands from key groups.²⁸

It is clear that the NRB is aware that it still has to overcome strong vested interests, mainly among the bureaucratic elite, who wish to derail devolution and preferably revert to the status quo prior to 2001. This has affected the NRB's willingness to engage in dialogue with provincial governments over proposals for institutional reform, and to support innovations introduced by district governments (the NRB initially objected to the establishment of a strategic policy unit in Faisalabad)

The NRB commissioned a study (unavailable at the time of fieldwork) of the capacity building/orientation efforts it pioneered in 2001/2 via provincial governments. The study alleged that large volumes of training funds provided by the NRB were diverted to politically preferred construction projects, that there was mass absenteeism and forgery in relation to participation in training activities, and that the results were well below expectations. This has affected the NRB's willingness to trust provincial governments, and its perceptions of the feasibility of undertaking effective capacity building via 'supply-side' oriented activities in or via government institutions.

Federal Ministry of Education
See section 4.1 on the role and (lack of) restructuring of the Ministry.

The Devolution Trust for Community Empowerment (DTCE) and the National Commission for Human Development (NCHD)
These are new (central) institutions spawned by the philosophy of the NRB. They believe that active efforts from the bottom up are crucial in order to galvanise communities and alliances among citizens, the private sector and local governments to work together, and to put pressure on bureaucrats and politicians to deliver.

Ministry of Local Government and Rural Development
Despite its name, this Ministry is not the champion, regulator or capacity builder of local governments, nor is it a conduit for interactions with the federal government.²⁹ It manages development schemes financed by the funds allocated to MNAs from the federal development budget to be implemented in their constituencies.

7.2 Provincial level

Chief Minister
Several interviewees referred to the key role of the Chief Minister in keeping education services on the political agenda in Punjab over the past few years. He has been a driving force in ensuring that the PESRP was designed and financed, and has championed the use of EMIS data in problem identification and planning. He has also introduced regular forums where he meets Nazims to discuss common issues.

Members of the Provincial Assembly (MPAs)
The Chief Minister's forums have gone some way to repair the damage done to provincial/district government relations due to the MPAs' alleged interference in district affairs, including attempts to derive benefit and kudos for development schemes or employment of staff at district level. There is considerable evidence that MPAs still influence the allocation of resources for district development, project locations, as well as staffing matters, including appointments, transfers and discipline. Provincial bureaucrats have also tended to exploit their own direct and indirect influence on staff in district departments. One complicating factor is that the Chief Minister's 'electorate' comprises the MPAs, which limits the extent to which he can pressure or influence them to modify their behaviour.

Department of Education
The Department establishes policy for the districts in terms of education. This includes registering and regulating private schools. Government authorities acknowledge that the private sector provides valuable services that the public sector does not have the capacity to replace in the medium term (see section 4.2). However, there is no effective forum for dialogue between public and private sectors. This will be addressed by the Punjab Education Foundation under the PERSP.

The private sector
There is ample evidence that the private sector represents a significant component of overall education service delivery capacity in Punjab. Even the (admittedly under-reported) data from the Punjab EMIS indicates that in 2003 there were nearly 18,400 private schools, nearly half of which were co-educational. The Private Schools Management Association (see Box 5) claims that the actual figure is three times higher. Ismael et al. (2000) cite a 1996 survey that found that the public perceives private schools to be qualitatively better than the state sector. Success factors included: co-educational schools; female teachers; much lower absenteeism; lower pupil-teacher ratios; and better infrastructure than in state schools.

Box 5: The Private Schools Management Association

The Association claims to represent 90% of the 95,000 private schools in Pakistan, which employ about 1.5 million teachers. In Punjab there are 55,000 private schools, compared with about 63,000 state schools. Representatives of the Association's Punjab branch claimed that the provincial government wished to control the private sector, and was not interested in dialogue or consultation. They complained about the plethora of provincial (and some federal) taxes they were obliged to pay for what they felt was inadequate support.³⁰ They can gain access to the (provincial) Minister of Education, but their requests are never acted upon.³¹ Compared to other provinces, the Punjab provincial government was more reluctant to change its policies and support to the private sector, while it has raised the fees it imposes on private schools.

Some private schools have started their own in-service training programmes for teachers. The Association claims that private schools are generally in better condition than state schools (many of whose premises are rented), and had offered the state sector free use of their facilities for youth education and training, but it was never taken up. They welcomed the promised public-private sector interface that the reformed Punjab Education Foundation will provide under the PESRP.

Teachers' unions
Teachers' representatives echoed some of the private sector's complaints about the Department of Education. Communication tended to be at the behest of department officials, unless they see an item in the press about an issue in which they have an interest, when they request a meeting. They also complain that the unions were not consulted in the design of the PESRP, and that teachers are routinely expected to be involved in preparing electoral rolls, and in running elections and censuses.

The representatives acknowledged that teacher absenteeism is a serious issue, especially in rural areas, and alleged that this always involves some form of connivance with local education officials. In the past, politically promoted school siting decisions³² and the employment of relatives of MPAs as teachers or support staff have left a heritage of under-utilised schools and a fear of politically inspired retribution for any action to rationalise staffing or school distribution. Among the unions, opinions differ on the merits of contracts - the main teachers' union is against them, but others support them as a means of enhancing the professionalism and reputation of teachers.

According to our informants, capacity building is 'all things which are required for improving education outcomes'. Above all, recruitment and promotion (including to managerial or administrative positions) on merit should be a decisive factor in boosting future capacities, but in their experience it was the exception rather than the rule.

7.3 District level

The Nazim and the District Coordination Officer (DCO)
These two functions are discussed together because although their status and tasks are discrete and clearly defined (in the Punjab Local Government Ordinance), their relationship can have far-reaching implications for the operation of the district's bureaucracy, and the political accountability of the executive (see Box 6).

During two district visits we observed two completely different relationships between these key protagonists. In one case, the relationship was adversarial, mutually suspicious and distrustful. Our visit had been difficult to coordinate and to conduct - perhaps symbolic of the day-to-day difficulties within a district administration. In the other case, we were received by a plenary meeting, chaired by the Nazim, of planning and education-related principal officers, including the DCO. A full programme of interviews and discussions was efficiently arranged.

Executive district officer (education), EDO(E)
The EDO(E) coordinates and supervises education service delivery and is responsible to the education monitoring committees at district and union levels. He is accounting officer of the department and reports to the district accounts committee.

Box 6: The Nazim and the DCO

The Nazim is the indirectly elected head of the district council. He/she (there are two female Nazimeen) is responsible for providing a vision for district-wide development, drawing up strategies and annual plans (with the district administration), and ensuring that all functions ascribed to the district are performed (including administration and financial monitoring) by issuing orders to the DCO. The Nazim reports to the council at least twice a year, presides over the Mushavirat committee, and is responsible for internal audits. He/she is prohibited from employing a political adviser, and is financially accountable for decisions made on his authority. He/she is subject to recall for misconduct by the Chief Minister after deliberation of the Provincial Assembly, and by a majority of union council members. He/she may request the provincial government to transfer the DCO in cases of unsatisfactory conduct.

The District Coordination Officer is a federal or provincial employee (grade 20 minimum, although the post is so unattractive that in Punjab the grade is more often 19), usually of the District Management Group cadre. The DCO is responsible for coordinating the work of district offices (except the police or magistrate's office), and for providing information to council members. He is principal accounting officer, responsible to the PAC of the Provincial Assembly (since most funds are received from and accountable to the province). The DCO reviews the performance of executive district officers (heads of department). He may appeal to the (provincial) local government commission in cases where he feels the orders of a Nazim are unlawful.

8 Internal management factors

8.1 Incentives

According to the recent study Devolution in Pakistan (ADB/WB/DfID, 2004), the pattern of incentives facing stakeholders in devolution is the principal factor that affects both performance, and absorptive capacities for capacity building efforts. These incentives are rooted in accountability relationships, financial and budgeting systems, and the ways tasks are assigned. The authors of the study express concern that unless this issue is addressed, the current devolution policy will fail to deliver the service improvements it is intended to deliver, and will succumb to the same fate as the SAP, for similar reasons.³³ A recent ODI study assessed the state of service delivery in Punjab, and compared it to the stated intentions of the devolution reform programme. It noted that 'translating ... good intentions (of policy) ... involves changing behaviour that has been entrenched in the bureaucratic and political establishment' (Williamson, 2004, p.10).

In 'unpacking' the incentives facing Nazimeen, the Devolution study relates the prospects in Pakistan to the (very mixed) past record of international experience in devolution, in terms of the allocation of resources for, and improved delivery of, education (and other) services. It concludes that Nazimeen need to be provided with both fiscal³⁴ and popular (electorate and user) 'voice' incentives. Service providers, e.g. the EDO(E) and his/her staff under the DCO, and school staff , need both managerial incentives (from the Nazim and DCO) and citizen/user voice incentives. The study's analysis of incentive-related issues (which fills 50 pages of volume II; pp.25-75) depicts a veritable maze of possible permutations of political interests, conflicting bureaucratic loyalties, interference between layers of government, and perverse, dysfunctional accountability mechanisms.³⁵ It concludes that, overall, devolution has not yet made much impact on prevailing patterns of incentives.
In the following, we distinguish between general 'systemic' issues related to incentives, and specific issues related to education services. The most glaring examples of dysfunctional, perverse or otherwise problematic 'incentives' to performance include:

Regarding the effectiveness of voice mechanisms for users and citizens:

Service delivery monitoring committees are largely ineffective. If they meet at all, their capacities and motivation appear minimal. Reports from EDOs are not forthcoming or are too general to make an impact; district budgets are unintelligible; committee members are preoccupied with their 'own' (rather than union or district) development schemes. In education, the primary concern is interference with the essentially administrative issue of posting teachers. One district we visited was an exception to this pattern, up to a point.³⁶
Similar problems have tended to emerge in the workings of the school councils in Punjab. It remains to be seen how PESRP's component on strengthening the school councils (in terms of membership, funding and the provision of information to members) will address these problems.³⁷

Regarding the incentives facing senior managerial staff:

DCOs are accountable to the Nazim and to the provincial executive, whose staff they remain. They are subject to transfer at the behest of both the province and, under certain circumstances, the Nazim, whose interests rarely coincide. Morale has suffered as a result.
There is no incentive for or feasibility of restructuring staff complements or skill sets at the district level, since all discretion for hiring, firing (very difficult), promotion and postings/transfers rests at provincial level.³⁸
Non-salary budgets are so small that the autonomy to allocate financial resources according to local priorities, which devolution was intended to provide, remains illusory.
Experience of preparing budgets at district level is only entering its second year. Recurrent and development budgets are compiled separately. The latter is predominantly a politically driven process, where funds are often allocated to Nazimeen to spend on pet schemes that may - or may not - accord with provincial or even district priorities.³⁹ The PESRP earmarked allocations will redress this trend. The development fund allocations to MNAs and MPAs compound the problems of coordination (and to an extent motivation) faced by district managers.
'Vertical' programmes (such as literacy, and a special programme for girls' education) are controlled by the province, with minimal flow of information to the districts where programmes are being implemented.

The situation facing service providers (head teachers and teachers) is no more favourable in terms of incentives:

teachers have not received effective in-service training and support;
head teachers have yet to receive adequate guidance in supervision of teachers. There is a lack of operational funding for schools, and protracted procedures to obtain funding for maintenance or repairs. Remote postings cannot be rewarded materially because districts do not have the discretion (or the budgets) to pay allowances;
'visiting' supervision in the form of learning coordinators or assistant education officers is infrequent and ineffective. Our informants confirmed that teacher absenteeism must involve a degree of collusion with those inspecting schools; and
teachers are prey to, and/or seek out political influence.⁴⁰

8.2 The existing 'stock' of EDOs

Almost all executive district officers (EDOs) have a teaching background (thanks to the working of the cadre system). However, due to the workings of another aspect of the bureaucracy - seniority-based promotion - they tend to be approaching retirement age and to have been principals in higher education. It is very rare for anyone managing primary and secondary education to be mid-career, or to have been directly engaged as service providers at that level. Past efforts to introduce 'new blood' into this crucial cadre have met with no success,⁴¹ and the potential for capacity building in the future is limited.

9 Capacity building experiences to date

On the basis of the foregoing analysis, this section assesses the considerable experience with capacity building with regard to devolved education service delivery in Punjab.

9.1 Capacity, change and performance in the Pakistan context

It is not feasible to come to any empirically supported conclusions regarding the relationship between these three concepts in the context of the case study. This is because, as the Devolution study observed, 'devolution is recent [and] there is no consensus on which data sources to use to measure intermediate and outcome indicators' (ADB/WB/DfID, 2004, vol.I, p.2). The CIET study also noted that 'uneven-ness in the reliability of information makes it all but useless for … evaluating performance' (GoP, 2003c, p.1), and evaluations are rarely conducted in Pakistan.

Despite this dearth of evidence on which to base any conclusions, our impression is of a delivery system undergoing major changes in its 'polity' that are still being worked through. There is some hope that over time the system will become more responsive to increased popular pressure from below, yet 'supply-side' constraints are proving difficult to overcome. The bureaucracy is largely self-interested, and representative democracy is dysfunctional in part because of collusion with the bureaucracy. These factors pose formidable obstacles to any effort to improve education service performance.

9.2 Prospects for enhancing education service performance

In this section we highlight some of the factors that appear to influence the feasibility of enhancing performance via capacity building activities in Punjab.

Incentives for change and performance improvements.
Incentives should lie at the heart of any future strategy for capacity building to support the changes and performance improvements that devolution was intended to bring about. However, the analysis in volume II of the Devolution study leaves the reader pessimistic as to the feasibility of creating an environment conducive to building the capacities (awareness, skills, abilities and relationships) of politicians, staff and the users of education services. It remains to be seen whether the newly constituted Presidential committee to review devolution will address the issues outlined in the study.⁴²

The factors that impeded the earlier attempts to introduce change and improve performance appear to be as real and as relevant as ever.
The World Bank review of SAP II concluded that its impact had been very disappointing, due to factors such as: the unstable political context in which it was implemented; the economic crisis and consequent resource constraints and financial accountability drives; poor ownership of the programme at provincial level; policy dialogue was confined largely to the (federal) bureaucracy, without adequate involvement of the political leadership; the introduction of recruitment bans; and the lack of continuity of key senior staff. These should be compared to the factors impeding change, performance and capacity building listed in Box 7.

Box 7: Factors impeding organisational change in Punjab

Many factors appear to explain the generally modest success - or indeed failure - of the capacity building programmes in Punjab. Participants in a training programme organised by the Institute of Public Finance in 2003 suggested many negative factors, including the lack of:

Stability: There have been eight governments in about 17 years, each with a different vision, goals and priorities. Civil servants are distracted by managing the impacts of these changes. No system is allowed to settle down to provide a focus for implementing government priorities.
Political maturity: The decentralised government at both provincial and district level many (political) representatives operating without a strict party system. Local spending on members' pet schemes has limited the availability of funds for priority programmes.
Long-term planning (but poor links to budgets): The government has elaborated many long-term and strategic plans, yet activities and annual budgets are rarely linked to these plans.
Continuity: Civil servants are regularly reassigned - three years appears to be the norm. To implement new systems, even to pilot them, it is important to have continuity until the worst impact of the changes is worked through.
Capacity: There is considerable concern that there is inadequate capacity to implement new systems without detailed instructions.
Trust: This is apparent in a number of ways, typical of governments with newly decentralised functions. To maintain their roles or jobs, staff commonly exercise control over district functions through the need for approval in the planning and budgeting process. Agencies and districts are said to have no flexibility or authority to change staffing levels upwards or downwards.
Integrity and objectivity: The incremental approach to budget preparation encourages departments to bid high every year knowing that Finance will reduce the allocation. The annual confidential staff report is rarely used for staff development purposes; honest appraisals of performance rarely recorded.
Line management. Departmental or sector staff have no control over hiring and firing or the deployment of staff.

Source: Cited in a review of the ADB's Fiscal Decentralisation TA programme (Gray, 2003b).

Pressures of time, scale and geographical coverage are becoming associated with the policy imperative of devolution.
The devolution policy has stark implications for community and local government capacity building, which have not attracted the attention they deserve. The President has a vision of political and resource empowerment of local communities. He and other champions of devolution are aware of the urgent need for mass awareness raising and for equipping local governments with the means to respond to new popular pressures for improved service delivery performance. This is all the more urgent given the perceived hostility to this major reform at provincial and district levels amongst cadres of the bureaucracy (e.g. the DMG) whose roles and interests have been adversely affected by it. Speed and coverage are therefore of the essence. The CCB/DTCE initiative is politically crucial, yet it must heed the lessons of international - and domestic - experience of community and local-level mobilisation, including the need to move slowly; to respond to local initiatives and interests; and to build trust and confidence in the (new) CCBs at union level. It is urgent that the CCBs play their role. Yet there are over 6000 unions, and 25% of local development budgets are earmarked but 'log-jammed' waiting for CCB schemes to absorb them. But evidence of past progress is not auspicious. Since well before the advent of devolution the National Rural Support Programme network of NGOs has supported the establishment of community organisations. The evidence indicates that intensive, sustained community-level effort is needed over a considerable period of time before new organisations are sustainable. This begs the question of whether the current plans for mass orientation will be adequate to 'kick-start' so many CCBs as a new type of community organisation.⁴³

The extent of provincial 'ownership' of some federally championed programmes is questionable.
Ownership was clearly a factor that diminished capacity building performance in SAP, and we heard critical comments about the proposed NCHD and DTCE approaches at provincial and district levels. The review of Fiscal Decentralisation TA training noted that provincial finance departments were excluded from dialogue related to detailed design of the systems and training content.

Difficulties in assessing needs.
Many parameters have changed, and new roles are ill-defined or not yet assigned.
This is particularly true at district level, where there is no real strategic context (yet) for the assessment to what ends they are working), and unclear and ill-defined operational relations to other officers and - especially - politicians.

The limited extent of learning from past experiences.
The only evaluations we encountered were donor-financed. Objective, ex-post evaluations of capacity building initiatives are not routinely conducted, and those that have been undertaken do not appear to have been widely shared, discussed or considered in the design of future initiatives.

The lack of experience and familiarity with systematic HRD programmes.
The discourse on 'capacity' seems to be based on very little previous experience with effective skills-oriented human resource development (HRD) programmes. This may be a significant factor, and was mentioned by international resource persons who had formerly been involved in systematic HRD and the broader process of building organisational capacities. We saw how traditional civil service training is oriented towards knowledge, administration and procedures. Very few people we met have experienced skills-based training (except in the armed forces). There is thus an (unconscious?) expectation that 'capacity building' (equated solely with training) will be remote from the workplace (geographically and in terms of content), often one-directional, and not closely related to the work context. There may also be a tendency to expect that senior staff are too busy to attend 'training', and so to send subordinates. In the case of orientation of or consultation with senior staff in new policies or procedures, this will undermine their future 'ownership' or the operability of innovative measures.

Attitudes to capacity building if equated with 'training'.
Cynicism about 'capacity building' is beginning to set in. Training is beginning to mean 'allowances', and conveniently meets the need to spend available budgets. There appears to be pressure to spend budgets allocated to training: these activities can be lucrative for trainers, especially the senior protagonists. There are incentives for 'mass-production' or 'cascade' training approaches, even in the absence of positive evidence of their effectiveness, and indeed in the face of evidence of the contrary.

The extent to which 'capacity building' is used in 'horse-trading' by proponents of public sector reform.
The World Bank has approved a major public service capacity building project featuring overseas training, in the absence of a clear government commitment to reform. Several purported reforms already agreed to or started are mentioned in the project memorandum (besides professional development, and reform of training institutions to which the project is to contribute), but there is no evidence of substantial changes or action being taken as a result. The approach adopted appears to be one of 'consensus building', influencing the climate for later negotiation, or simply 'keeping the door open' to the prospect of more significant, decisive action later; creating 'champions of change' at senior levels; creating a 'critical mass' of internationally trained policy analysts and decision makers, creating a 'paradigm shift' in the way the country is governed and managed. The programme is seen as a 'long overdue investment' (World Bank, 2004a, p.13). However, the document acknowledges that there are 'significant' risks in terms of the potentially slow pace of further major reforms. It is difficult to escape the conclusion that this form of 'capacity building' (via overseas scholarships) may be in danger of becoming a devalued currency, especially if it does not produce the more favourable environment for future reform which is intended.

10 Conclusions on capacity for capacity building

The government of Pakistan faces a real capacity building dilemma. Although there is a clear need for orientation and training of large numbers of public officials, elected councillors and CCB members (including skills development leading to significant behaviour change), the capacities for such capacity building are limited by:

predominantly 'static' traditional facilities that provide training off-site, using methods that are unlikely to lead to behaviour change (unless followed up on-site);
major problems in integrating public and private sector capacities due to the absence of financial incentives within the public sector (they receive guaranteed budgets from the Department of Finance). This financing 'regime' is inhibiting attempts to forge partnerships between public and private/NGO training providers.⁴⁴ The lack of consultants with public sector experience is another inhibiting factor; and
socio-political and cultural rigidities that affect the speed of response of local government trainees and community organisations, indicating the need for intensive work at that level, supported by professional facilitators. It is unclear from where such facilitators are to come, however.

The manner in which public institutions are funded will mean that they have no incentive to appraise critically the value and relevance of what they do, since they do not depend on their clients for revenue. This slows learning about 'what works'. Thus, the planned or ongoing efforts to build capacity for capacity building in public sector institutions will take years to yield results, unless the financial 'regime' under which they operate is changed. We are aware of no such prospects; indeed, no concerted efforts have been made to re-orient private sector consultants to local government systems and operational realities. Consulting firms claim to provide (and clients expect) consultants who are able to respond in a variety of environments. The reality appears to be at variance with this, however.
It remains to be seen whether the planned support programmes for (public) training institutions stress the importance of practical application of concepts after exposure to knowledge- or awareness-raising, or skills-oriented training. This could be facilitated by exposing staff to HRD practices in the private or military sectors, both of which are renowned for their practically oriented systematic training approaches.

It appears timely for development partners to collaborate with Punjab authorities (and the National Reconstruction Bureau) in mounting an event or conference in which international experience in 'capacity building' could be shared. There are major successes on the international stage, exposure to which - with appropriate follow-up - could be influential in the development of future policy on approaches to this issue in future. The other cases in the ECDPM study, and the experiences and conclusions that will be distilled from them, could provide useful 'raw material' for such a gathering.

Appendix: List of individuals interviewed

Federal Government of Pakistan
Kh Zaheer Ahmed, Secretary, Ministry of Local Government and Rural Development
Daniyal Aziz, Chairman, National Reconstruction Bureau
T.M. Qureshi, Focal Point, Decentralisation and Gender, Ministry of Education

National Commission for Human Development
Zulfiqar Ahmad, Chief Operating Officer

Devolution Trust for Community Empowerment
Paul Oquist, Chief Technical Adviser (UNDP)

Provincial government of Punjab
Khushnood Lashari, Additional Chief Secretary, S&GAD
Shahid Rasheed, Secretary, Department of Education
Md Imtiaz Tajwar, Special Secretary,
Elementary Education
K.I. Gillani, Programme Manager, PMIU, Punjab Education Sector Reform Programme (PESRP)
M. Khurram Khan, Deputy Director, PESRP
Amjad Saleemi, Manager, Finance, PESRP

Sheikhupura district government
Tawakkul Virk, Nazim
Tariq Najmi, District Coordination Officer
Prof. Javed Iqbal, Executive District Officer, Education
Amjad Javed Assistant Director, Administration, Education Dept
Shafiq Durrani, District Education Officer, Colleges
Mukhtar Gill, DEO (Secondary)
Ch. Fazal Ahmed, DDEO (Tehsil Sheikhupura)

Okara district government
Syed Sajjad Haider, Nazim
Syed Abbas Raza Rizvi, Naib Nazim
Iftikhar Hussain Babar, District Coordination Officer
Abdul Jabbar, EDO, Education
Md Alam Jaffri, EDO, Literacy
Awais Ahmad, District Officer, Planning

International cooperation agencies
Mark Poston, Education Adviser, DfIDP
Jackie Charlton, Senior Governance Adviser, DfIDP
Nargis Sultana, Senior Programme Officer (Education), DfIDP
Douglas Porter, Senior Governance Specialist, ADB
Jacob Christian Gulmann, Economist, ADB
Zahid Hasnain, Economist, World Bank
Asya Akhlaque, Public Sector Reform Specialist,
World Bank
Farhan Sabih, Governance Adviser, UNDP
Fauzia Saeed, Country Director, ActionAid
Ann Cockcroft, Country Representative, CIET

Punjab Resource Management Programme
Dr Shujat Ali, Programme Director
Asadullah Sumbal, Programme Coordinator
Azhar Rauf, Deputy Director

Decentralisation Support Programme (DSP)
Babar Yaqoob Fateh Muhammed, Provincial Programme Director (Punjab)

Strengthening Decentralised Faisalabad District Government Project
S.M. Khatib Alam, Project Manager (GHK Country Director)
Bridget Smith, Educationist

All Pakistan Private Schools Management Association (Punjab)
Kashif Adeeb, President
Adeeb Jawadani, President
Imran Akhtar Malik, Secretary General

Society for the Advancement of Education
Fareeha Zafar, Director

Centre for Education and Consciousness
Baela Raza Jamil, Chairperson and Technical Adviser to Ministry of Education

Punjab Association of Subject Specialist Teachers
Chaudhury Muhammad Amin, President,
Abdul Aziz, subject specialist teacher

References

Academy of Educational Planning and Management. 2003. EMIS Development in Pakistan Workshop Report, December 2003.

ADB. 2001. Project Performance Audit Report The Social Action Programme (Sector) Project Loan 1301-PAK November 2001

ADB/WB/DfID. 2004. Devolution in Pakistan (3 volumes: I. Assessment and Recommendations; II Recent History; III Technical Considerations). Asian Development Bank, World Bank, DfID.

Aga Khan University Institute for Educational Development. 2003. Policy Dialogues on Key Issues in Education.

AKU-IED and SAHE. 2004. Effectiveness of In-Service Teacher Training Programmes offered by University of Education, Lahore Evaluation Report, February.

Alam, Khatib and Naveed Aalia. 2001. Community Led Approach to the Delivery of Quality Basic Education. Faisalabad Area Upgrading Project. November

Ashraf, A. 2003. Northern Areas Education Project, Management Development: Final Evaluation, November.

Ashraf, N. et al. 2002. Report of the President's Task Force on Human Development Islamabad, January.

Asia Foundation (undated) Lessons from Public Private Partnerships: The Education Sector in Pakistan.

Cheema, A., Khwaja, A.I. and Qadir, A. 2004. Decentralisation in Pakistan: Context, content and causes (draft), prepared for Decentralisation in Developing Countries: A Comparative Perspective, by D. Mookherjee and P Bardhan (forthcoming).

Dean, D. 2003. Final Evaluation Report on EMIS. Northern Areas Education Project November.

District Government, Multan (with MSU). 2002. Proceedings of District Social Sector Development Forum, March.

DTCE. 2004. Quarterly Newsletter, vol. 1(1). Devolution Trust for Community Empowerment

ECDPM. 2003. Capacity Development: The why's and how's. Capacity.org, Issue 19. Maastricht: ECDPM.

Government of Pakistan. 2003a. Education for All (2001-15). Pakistan Ministry of Education.

Government of Pakistan. 2003b. Education Sector Reforms Action Plan 2001-5 January 2003.

Government of Pakistan. 2003c. Social Audit of Governance and Delivery of Social Services: Baseline Survey 2002. A. Cockcroft et al. for the CIET/NRB.

Government of Pakistan. 2004. Entity Improvement Plan of the Establishment Division (Public Sector Capacity Building Programme) January.

Government of Punjab (with MSU) 2002. Strengthening School Councils in Punjab. Multan Workshop Report, January.

Government of Punjab. 2001. Punjab Local Government Ordinance (XIII of 2001). Revised edition.

Government of Punjab. 2003. Punjab: Education Sector Reforms (Mimeo) June.

Gray, D. 2003a. Performance-based Budgeting Training for Government of Punjab, Report for DfIDP, April.

Gray, D. 2003b. Fiscal Devolution Technical Assistance Project Completion Report, for ADB, October.

Ismail, Z. et al., 2000. Alternative Delivery Mechanisms for Social Services: Case Studies from Pakistan. Social Policy and Research Centre, June.

Jamil, B.R. 2001. Public-Private Partnerships in Education: Perspectives from the Political Economy of Education, Paper presented at National Symposium on International Experiences with Decentralisation and Education, Quetta, June.

Kennedy, C.H. 1987. Bureaucracy in Pakistan. Karachi: OUP.

Morgan, P. 2003. DAC Study on Capacity, Change and Performance: Background Paper on Methodology. Maastricht: ECDPM.

Morgan, P. 2005. Organising for Large-Scale System Change: The Environmental Action (ENACT) programme, Jamaica. ECDPM Discussion Paper 57J. Maastricht: ECDPM.

NCHD. 2004. Support for Achieving the Millennium Development Goals. National Commission on Human Development, Progress Report, August 2002-February 2004.

Qureshi, T.M. (undated) Decentralisation in Education in Pakistan. Ministry of Education (mimeo).

Rashid, A. (ed) 2000. Engaging with Basic Education in Pakistan. SAHE Education Watch Report.
Social Action Programme. 2001. Devolution and Decentralisation: Implications for the Education Sector. Technical Group Meeting Report, Islamabad.

Social Action Programme. 2002. Devolution and Decentralisation: Identification of Progress in Implementation in Education. SAP Multi-Donor Support Unit.
UNDP. 2003. Essential Institutional Reforms Operationalisation Programme: Evaluation of First Phase, H. Fredborg and Z. Ismael, UNDP Pakistan, January.

Unsworth, S. 2003. Better Government for Poverty Reduction: More effective partnerships for change. DfID Consultation Paper.

Watson, D. and Khan, A.Q. 2005. Capacity and Capacity Building in the Punjab for Devolved Education Service Delivery: Comparing current approaches with international experience, deriving possible pointers for policy and future action in the Punjab. Discussion Paper for DfID Pakistan (available on request from Nargis Sultana at DfIDP).

Watson, D. and Yohannes, L. 2005. Capacity Building for Decentralised Education Service Delivery in Ethiopia. Discussion Paper 57H. Maastricht: ECDPM.

Watson, D., Yohannes, L. and Khan, A.Q. 2005. Capacity Building for Decentralised Education Service Delivery in Ethiopia and Pakistan: A comparative analysis. Discussion Paper 57I. Maastricht: ECDPM.

Williamson, T. et al. 2004. Devolution and Service Delivery in Punjab: Strengthening Health, Water and Sanitation Services. ODI Centre for Aid and Public Expenditure.

World Bank. 2001. Pakistan: Reforming Punjab's Public Finances and Institutions. August.

World Bank. 2002. Completion Report Social Action Programme (II), April.

World Bank. 2003a. Education in Punjab (presentation), March.

World Bank. 2003b. Private Schooling in Punjab, World Bank Technical Paper by T. Andrabi.

World Bank. 2004a. Making services work for poor people. World Development Report 2004.

World Bank. 2004b. Pakistan: Public Expenditure Management, Vol.I, PREM Sector Unit, January.

Zafar, F. 2003. Fiscal Devolution in Education: Case study reflecting initial responses. Society for the Advancement of Education (SAHE) for MoE.

Notes

1 David Watson and Adnan Khan (2005) Capacity and Capacity Building in the Punjab for Devolved Education Service Delivery: Comparing current approaches with international experience. Discussion Paper for DfID Pakistan (available on request from Nargis Sultana, DfIDP).

2 Nargis Sultana, senior education programme officer, DfID Pakistan, joined the consultants during the fieldwork. The Ethiopia case study and a comparative analysis of the two cases have been published as ECDPM Discussion Papers 57H and 57I, respectively.

3 See section 4 for a discussion of these ambiguities and suggested priority roles at various levels.were 60% and 76%, respectively.

4 Punjab weighted its 39.8% distribution 75% according to population and 10% according to poverty. NWFP accorded a weight of 25% to poverty (Cheema et al., 2004, p.25)

5 In 2001/2 the net enrolment rate for girls aged 5-9 was just 38%, according to the PIHS. The study Devolution in Pakistan (ADB/WB/DfID, 2004) compared gross primary enrolment rates in Pakistan and other countries with similar incomes in 2002. Pakistan is 26% below the average value for its income level (at 72%), and 33% below the average for females (at 61%). The World Bank (2003a) estimated that in Punjab nearly 50% of 5-9-year olds (5.6 million) do not attend school, two-thirds of 10-14-year olds (6.5 million) are not enrolled, and primary and elementary dropout rates were 60% and 76%, respectively.

6 In one district, over 60% of teachers were regularly absent, although it has recently 'succeeded' in bringing this figure down to 20%.

7 Commenting on the dearth of reliable information on the performance of service providers, the CIET/NRB (GoP, 2003c, p.1) states that 'uneven-ness in the reliability of information makes it all but useless for ... evaluating performance or allocating resources from a central level'.

8 Nearly 60,000 people were interviewed, half of them women. Over 700 focus groups, with 1300 head teachers and health facility heads, added qualitative information.

9 These needs included sensitisation of officers and communities on their roles; administration skills (including community mobilisation and negotiation with political functionaries); academic skills (auditing schools' academic performance, in-service teacher training), including literacy training; planning and financial issues (including M&E, school-based budgeting, empowerment of head teachers, capacity building in PPPs; and MIS training for district level operation and feedback.

10 At the provincial level in Punjab, these functions included policy formulation; performance-based M&E; inter-district transfers (whether of teachers or financial resources is not specified); quality assurance; curriculum development; teacher training; and budget coordination/consolidation. At the district level, tasks included recruitment of staff; non-development budget and re-appropriation by districts; registration of private schools; audit, inspection and upgrading of institutions; and preparation and execution of annual development plans.

11 ToPs oblige districts not to reduce non-salary routine expenditures below 2002/03 levels; to distribute textbooks; to use budgeting subheads appropriately; to hire teachers on a contract basis; to use their own resources for infrastructure improvements; to inform school councils on forthcoming funds and uses; and to encourage payment of bonuses to teachers who merit them.

12 See Zafar (2003, p.24).

13 Inter-cadre rivalries 'determine the shape of federal divisions and departments, as structures are created to meet the needs of the various groups and cadres for promotion opportunities rather than on the basis of policy priorities or operational needs' (World Bank, 2004b, p.94).

14 The unattractiveness of senior posts is also related to the 'compression' of pay scales over the years, from about 62:1 in 1960 to 10:1 today (World Bank, 2004b, p.99). Top public sector remuneration has declined in real terms by one-third in the last 50 years, widening the gap between public and private sectors, where real pay doubled over the same period.

15 In Punjab, 60,000 civil servants are in grades 17-22, 124,000 in grades 12-16 (which includes some junior teachers), and over 700,000 in grades 1-11 (World Bank, 2004b, p.97).

16 In North-West Frontier province the average tenure of a departmental secretary is 7.5 months; in Sindh it is 15 months.

17 50% of development resources are allocated to roads, 12% to electricity, 0.3% to education and 0.2% to health services. In one district there were only three functioning CCBs, yet 25% of the annual development plan capital budget have to be reserved for schemes initiated by CCBs.

18 This may also relate to the fact that politicians and senior bureaucrats tend to send their children to private schools.

19 This problem is discussed at length in the World Development Report (World Bank, 2004a; see also Williamson et al., 2004). The public funds allocated to each MNA are reserved for their preferred projects in federal sectors (electricity, irrigation, etc.), although those allocated to MPAs may include education projects.

20 The CIET survey, the Devolution in Pakistan study (ADB/WB/DfID, 2004), Williamson et al. (2004), the Presidential Task Force (see Appendix 2), and the NRB all noted the lack of awareness of what citizens could expect from public service providers.

21 In the course of this case study the authors compiled extensive information on 12 major capacity building programmes at both national and provincial levels. A synthesis of this material is available on request from ECDPM.

22 The ADB notes that education received over 60% of SAP I funding, 55% of which went to Punjab.

23 Data from the Pakistan Integrated Household Survey (cited in World Bank, 2002, p.33). The PIHS net enrolment rates were disputed by GoP.

24 Since SAP, some provincial governments continued to compile validation reports, but not Punjab.

25 The SAP II report notes that the missions, some with over 50 members, on one occasion prepared an 150-page aide memoire listing unprioritised actions for the GoP to tackle.

26 This may be symptomatic of the Bank's difficulties in maintaining cohesion among donors.

27 The PESRP's PMIU is examined in a DfID Pakistan-commissioned analysis of capacity building in Punjab. See the 'Epilogue' at the end of the summary for details.

28 For example, a group of district Nazims in NWFP took the provincial government to court to redress rights that had been infringed.

29 We understand that DfIDP has supported discussions on the establishment of local government associations as a means of improving local/federal government interactions.

30 The Association alleges that in Punjab over 5000 private schools have had to close over the past two years because of the tax burden.

31 These included requests for land on which to build schools, or for public buildings to rent; access to teacher upgrading programmes; and easier visa procedures to facilitate international exchanges.

32 According to our interviewees, since the PESRP did not analyse the appropriateness of school locations before allocating infrastructure and furniture, some resources will be wasted.

33 They attribute the failure of SAP in part to its neglect of the incentives for key policy makers and service providers (ADB/WB/DfID, 2004, p.2).

34 Budget certainty, autonomy in preparing budget, and incentives for local revenue-raising.

35 Nazimeen, often from politically well connected families, may dominate council and union members because they can influence budget allocations, so that there is little effective 'downward' accountability to their constituents - district and union council members.

36 The district has established a 'checking committee' to deal with the problem of teacher absenteeism. In 2002 over 60% of teachers were regularly absent, but this was reduced to 20% with tighter supervision and surprise visits.

37 In Faisalabad, due to the forceful and imaginative personalities of the Nazim, DCO and finance and planning EDOs, the school councils and monitoring committees are working better than average.

38 Even the new-found flexibility of hiring teachers on contract was made conditional (by provincial fiat) - only graduates should be hired. One district we visited had no realistic prospect of hiring graduates, and thus of filling contract postings. It had sought dispensation but had been refused.

39 In one (relatively liberal) district we visited, the recurrent education budget was nearly 70% of the total budget; the development budget for education was just 13%.

40 nterviewees suggested that teachers who serve as local government election officers are attractive as allies to local politicians. In turn, teachers find it easy to gain politicians' support in evading disciplinary action for absenteeism, or obtaining favourable postings. Several informants described the Education Department as 'the most politicised service in the Punjab'.

41 We heard of two attempts, including the establishment of an 'education management cadre'. However, a combination of opposition from teachers' unions and ostracism by provincial (generalist) bureaucrats put paid to these experiments.

42 We heard about this from a Nazim who is one of four requested by the President to be a member. It will otherwise be composed of chief ministers and chief secretaries.

43 According to the NRSP (March 2004), of the 11,414 community organisations their members had worked with in 6 districts of Punjab, 2413 were inactive, and another 1579 met irregularly, and none were 'ready for any activity'. Thus 35% of these organisations had not (yet) taken off.

44 No public institutions have made bids to participate in DSP-funded programmes. One senior official commented that he felt 'demeaned' by being expected to bid for work from the DSP.