Uganda's decentralised primary education: Musical chairs and inverted elite capture in School Management Committees

Abstract

Incl. bibl., abstract Decentralisation policies in Africa increasingly place responsibilities and resources for the provision of public services in the hands of public bodies at the lowest level, for example in School Management Committees (SMCs). This paper questions whether elite capture, which is considered a major reason for the ineffectiveness of the management of public services at a national level, also characterises SMCs. On the basis of field research in Uganda, it is argued that elite capture does not trickle down to the lowest levels in the management of public services. [Cop. John Wiley & Sons, Ltd]

Full text

UGANDA’S DECENTRALISED PRIMARY EDUCATION: MUSICAL
CHAIRS AND INVERTED ELITE CAPTURE IN SCHOOL
MANAGEMENT COMMITTEES
GERARD PRINSEN
1
*AND KRISTOF TITECA
2
**
1
School of People, Environment and Planning, Massey University, Palmerston North, New Zealand
2
Conflict Research Group/Centre for Third World Studies, Ghent University, Ghent, Belgium
SUMMARY
Decentralisation policies in Africa increasingly place responsibilities and resources for the provision of public services in the
hands of public bodies at the lowest level, for example in School Management Committees (SMCs). This paper questions
whether elite capture, which is considered a major reason for the ineffectiveness of the management of public services at a
national level, also characterises SMCs. On the basis of field research in Uganda, it is argued that elite capture does not trickle
down to the lowest levels in the management of public services. Copyright #2008 John Wiley & Sons, Ltd.
key words — decentralisation; education; Uganda; governance; elite capture; public services
INTRODUCTION
Decentralisation policies in Africa shift responsibilities and resources for public services to lower tiers of the public
administration to improve the effectiveness and accountability of service providers (Ju
¨tting et al., 2004). The
ultimate step in decentralisation is taken when users of public services are directly involved in the management of
public services—as in School Management Committees (SMC), Health Unit Management Committees and Local
Water Committees.
In spite of the widespread implementation of and international support for decentralisation policies, there has
been scepticism about its effectiveness or its relevance as leverage to change governance practices. One of the
major problems that has been identified is the phenomenon of elite capture, where elites take control of
decentralisation policies and resources (World Bank, 2003: p. 73; OECD, 2004; Ahmad et al., 2005). This has also
been the case for essential services like education and health, wherein the poor are neglected in favour of the
wealthy (Darrow and Thomas, 2005: p. 475).
It has been documented how several local governmental bodies (Francis and James, 2003), non-governmental
organisations and community-based organisations (Mansuri and Rao, 2003) are prone to elite capture, but there is
very little research on this issue for user committees. Although they are described by Manor as a damaging second
wave of decentralisation, he concedes that his concerns are‘... limited by the shortage of solid empirical evidence
on user committees’ (Manor, 2004: p. 193). This paper aims to reduce this shortage of empirical research by
studying elite capture in SMCs, which are the ‘statutory organs that govern the schools on behalf of Government’
(GoU Ministry of Education and Sports, 1999: p. vi) and the lowest level of Uganda’s decentralised public
management.
public administration and development
Public Admin. Dev. 28, 149–164 (2008)
Published online 19 March 2008 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/pad.487
*Correspondence to: G. Prinsen, School of People, Environment and Planning, Massey University, Private Bag 11222, Palmerston North,
New Zealand. E-mail: g.prinsen@massey.ac.nz
**Correspondence to: K. Titeca, Conflict Research Group/Centre for Third World Studies, Ghent University, Universiteitsstraat 8, 9000 Ghent,
Belgium. E-mail: kristof.titeca@ugent.be
Copyright #2008 John Wiley & Sons, Ltd.

Elite capture in SMCs can take various forms and this research analyses from different angles. This is done firstly
through a theoretical introduction on elite capture and decentralisation and then through a series of empirical
questions. These should not only give us empirical insights into the specific degree of elite capture in SMCs, they
should also allow us to analyse the role of informal and formal norms in causing or preventing elite capture. Firstly,
the issue of membership is discussed, in order to ascertain whether SMC members belong to the local elite.
Secondly, effective elite capture would suggest that, once occupying public offices, elites hold on to these positions
for as long as they can— the ‘president-for-life’ being the ultimate example. This research verifies if SMC members
turn SMCs into lifelong fiefdoms. Thirdly, there is the possibility that individuals within the SMCs capture all
power for their own benefit and resist the delegation of tasks, which leads to the question of the distribution of
power. Lastly, there is also the possibility that groups at a higher level capture local SMCs. This is especially the
case for two groups with particular power over SMCs, namely politicians (local, regional and national) and
education officials. Politicians can take direct control over SMCs (through becoming a member); while both
politicians and education officials can capture resources destined for SMCs (cf. Reinikka and Svensson, 2004). The
potential capture of politicians and education officials will therefore be explored in the last two sections.
These questions are answered via a complementary analysis by two researchers who investigate SMCs from
different perspectives. To understand the ‘inside’ of SMCs in three rural districts, one researcher conducted group
interviews with 17 SMCs and took individual questionnaires from 127 committee members.
1
The schools were
randomly selected from schools within 1 h drive from the district capital (ANNEX 1). The group interviews
included open questions, Venn diagrams, and individual brainstorming on cards coupled with plenary pairwise
ranking, and peer-reviews. The questionnaire had 13 items; 11 closed questions on members’ profile and opinions
and 2 open questions.
In a second perspective, in-depth qualitative research was conducted in one of the three above districts, now
looking from the ‘outside’ at SMCs.
2
The data-collection mainly used unstructured and semi-structured
open-ended interviews, including a household survey, with a range of actors: individual district officials (politicians
and bureaucrats), SMC members, as well as parents of pupils. A ‘snowball sampling approach’ selected these
actors, taking care as to include informants from different social and political networks. By combining these two
perspectives, the paper unites the ‘breadth’ of a quantitative approach with the ‘depth’ of a qualitative approach
(Carvalho and White, 1997; Kanbur, 2005).
BACKGROUND
Decentralisation
Decentralisation is generally defined as a transfer of authority, responsibilities and resources by a central
government to lower levels of government or other organisations. Rondinelli’s typology defines forms of
decentralisation on a continuum in which central government control over the transferred authority, responsibilities
and resources is gradually reduced. The typology thus ranges from deconcentration to delegation to devolution to
privatisation (Rondinelli et al., 1983: p. 14; Rondinelli, 1999: p. 4). Although many other analysts take Rondinelli’s
typology as point of departure, most exclude privatisation (and deregulation) as forms of decentralisation, if only
because this transfer of public resources to private entities implies that the objectives of these entities are
determined by their members and not by the public as a whole. The process therefore positions itself outside the
realm of public administration and public policy (Ribot, 2001: p. 1).
Most research reveals that operational practices of decentralisation in Africa countries are rather diverse and the
various types of decentralisation occur in mixed forms, parallel or in isolation (Smoke, 2003: p. 8). Nevertheless,
1
This research was carried out in three rural districts in Uganda between September and November 2004 by Gerard Prinsen, as part of his PhD
research at Massey University in New Zealand. Like most districts in Uganda, these three districts are multi-ethnic and primarily agricultural
districts, relying on farming for employment and income.
2
This research was carried out in one of the above three districts between October and December 2005 by Kristof Titeca, as part of his PhD
research at Ghent University in Belgium. This paper obscures names of districts and persons for reasons of confidentiality.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
150 G. PRINSEN AND K. TITECA

most analyses of decentralised governance are categorised by the degree to which central government empowers
local government bodies to make decisions on policies and the allocation of resources (financial as well as human),
as well as the direction in which they are principally accountable, varying from central government to the local
communities (Manor, 1999; Oxhorn, 2004). Obviously, there are sector-specific considerations that influence the
institutional arrangements of decentralisation. In the education sector, central governments usually retain a
dominant role for efficiency or equity reasons in fields like setting standards, textbook production, teacher
recruitment and overall financing. Local governments and school committees acquire authority and resources in
areas like construction and maintenance of buildings and paying teachers’ salaries. In other areas, a mixture of local
and central decision-making is required to optimise efficiencies (McLean and King, 1999).
However, aside from such sector-specific considerations, the character and institutional forms of
decentralisation are more fundamentally determined by central government’s motives to initiate decentralisation.
Rondinelli upholds in 1999 that decentralisation is primarily driven by political concerns of national politicians
about a loss of political support or legitimacy. Several of the economic benefits of decentralisation, as better
allocative efficiency or increasing citizen’s willingness to pay for public services, are secondary motives with
‘potential’ to solve these problems of centralised governance but for which meeting the necessary preconditions ‘is
a tall order, but achievable’ (Rondinelli, 1999: p. 6–8). Others point out that most decentralisations seem mostly
driven by mixed motives, but conclude likewise that national politicians are mainly driven by the political
predicaments at national level, like ‘a loss of popular confidence’, rather than the unproven economic benefits or a
desire for democratisation at local levels (Manor, 1999: p. 34; Smoke, 2003, Oxhorn, 2004; Jutting et al., 2005).
Decentralisation and elite capture
In this light, it is not surprising that the degree to which central government genuinely cedes authority and control
over resources continuously appears as a key determinant of outcomes of decentralisation. ‘Waves of
decentralisation’ (Selee and Tulchin, 2004: p. 297) in the past century have usually been deconcentrations initiated
by national elites to establish a larger and stronger presence of government officials at local levels that helped the
state ‘to get a handle on its subjects and their environment’ (Scott, 1998: p. 2). Decentralisation policies are then a
next phase in ever-evolving networks of patrons and clients, rather than a genuine reform of the public
administration. Yet there may be reasons for ‘guarded optimism’ (Rondinelli et al., 1983: p. 8) that contemporary
decentralisation practices offer scope for more efficient service delivery or create opportunities for genuine local
control over resources (Ribot, 2001: p. 4; Devas and Grant, 2003). Nonetheless, there are also many cases where
contemporary decentralisations merely disguise older patterns in which national elites use decentralisation to build
alliances with entrenched local elites ‘to create and sustain power bases in the countryside’ (Crook, 2003: p. 86;
Boone, 2003; Ribot et al., 2006).
Alternatively, when the position of national elites is weakened, local elites may capture decentralised public
services and escape the control of the national state. Rao and Ibanez (2001) demonstrate how decentralised funds
appears to be driven by a local elite’s interests and dominated by a small group of better educated and networked
individuals. This mirrors other research which shows how decentralised resources are captured by local elite groups
(Baland and Platteau, 1999; Crook and Sverrisson, 2001; Johnson and Start, 2001). Local elites are more likely to
mobilise themselves successfully for these positions and can more effectively present themselves to the donors,
central government and the local population as the effective conduit for handling these funds (World Bank, 2003: p.
73). Obviously, this increases the risk of capture of public resources by these local elites, and the exclusion of
disadvantaged groups (Bardhan and Mookherjee, 2000).
It comes therefore as no surprise that virtually all analyses consider it essential that a supervisory role of central
state and effective local accountability mechanisms are required to limit the risks of local elites capturing the
decentralised authority and resources. Failure to meet both or either of these conditions is indicated as the reason
why decentralisation does not fulfil its potential for more efficient and effective delivery of public services or more
accountable and democratic local governance: ‘the picture is rather gloomy in most African countries’ (Manor,
1999; Wunsch, 2000, 2001; Jutting et al., 2005: p. 644). Rather than expanding democratic spaces (Heller, 2001:
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
UGANDA’S DECENTRALISED PRIMARY EDUCATION 151

p. 139), decentralisation therefore tends to favour nepotism and favouritism as the guiding principles of service
delivery (Platteau and Gaspart, 2003).
Decentralisation and institutional change
A useful perspective on the outcome of governance reforms and elite capture is the classic work on institutional
change of Douglas North (North, 1990, 1991). Whereas his perspective is most often used in explaining economic
performance of organisations, it also offers important political insights into the outcome and effectiveness of
governance reforms: it points out how the dynamics of organisations such as SMCs are being strongly influenced by
a general institutional framework, consisting of formal and informal rules
3
(Powell and Di Maggio, 1991). Formal
rules are, for example, political and judicial rules, economic rules and contracts. Laws and constitutions fall under
this category (North, 1990: p. 46–53).
Informal norms can be categorised into three groups: (1) extension, elaborations and modifications of formal
rules, (2) socially sanctioned norms of behaviour and (3) internally enforced standards of conduct. Sanctions,
customs and taboos fall under this category (North, 1990: p. 36–45). Informal rules are crucial in institutional
change as they change slowly, even if new formal rules are introduced (Kingston and Caballero, 2006: p. 11). These
rules are not codified and somehow flexible (Patterson, 2003). As a result, the overall outcome of a governance
reform ‘tends to be a restructuring of the overall constraints — in both directions — to produce a new equilibrium
that is far less revolutionary’ (North, 1990: p. 91). Consequently, decentralised public services will be far less
effective if certain informal constraints work against the introduction of new formal rules. Governance reforms
therefore have to be supported by the institutions in the environment surrounding the organisation. These
institutions are not always functional with regard to achieving the organisation’s goals, which Campbell (1995,
quoted in Hall and Taylor, 1996: p. 949) describes as a ‘logic of social appropriateness’ instead of a ‘logic of
instrumentality’.
SMCs and elite capture in Uganda
Uganda’s far-reaching decentralisation process is seen as an exemplary in the region (Langseth, 1996: p. 1, 19;
Ndegwa, 2002). Particularly in the education sector, there has been a rapid decentralisation of the management of
primary schools (CAI and Smith, 2000). Universal primary education (UPE) was introduced in 1997, fulfilling an
earlier election commitment of President Museveni (Stasavage, 2005). UPE abolished the school fees for four
children per family and allocated government grants directly to primary schools (GoU Ministry of Education and
Sports, 1998: p. 4; Suzuki, 2002). Overall, UPE led to a large increased student enrolment in primary education
from 3.4 million in 1996 to 6.9 million in 2001 (World Bank, 2002: p. 6). In 1998, the SMCs were reconfirmed as
the ‘Statutory Organs at the school level and they represent government’ and are thus formally in control of
decentralised education (GoU Ministry of Education and Sports, 1998: p. 17). However, as a legal body, SMCs had
a rather patchy history.
Uganda’s government introduced legislation that established the SMCs in December 1969 (GoU Ministry of
Education, 1969). However, these SMCs hardly came into operation due to Amin’s coup in January 1971 and the
ensuing political conflict in the 1970s. In fact, a detailed account of the administration of education in Uganda of
1971 makes no mention of SMCs and elaborates that public schools are supervised by District Councils and
managed by Boards of Governors on the basis of 1963 legislation. Nine of the 13 members could be nominated by
the religious group that had founded the school, but all required approval of the Minister. It required ‘one of the
longest debates on education in Parliament’ to establish that central government was taking control over schools
from voluntary organisations—against the recommendations of a Commission of ‘outstanding Ugandan men of
learning’ (Ssekamwa and Lugumba, 1971: p. 16, 78). However, the same source concludes that most of these
Boards were not very effective. This left schools effectively in the hands of the Ministry of Education.
3
North makes a clear distinction between institutions and organisations. Whereas institutions are considered a set of ‘rules of the game’, the
organisations are considered as a single entity which are the ‘players’.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
152 G. PRINSEN AND K. TITECA

In the ensuing years of civil war and economic collapse in the late 1970s and first half of the 1980s, all manner of
support by government for schools disappeared. Parent–Teacher Associations (PTAs) effectively kept the primary
schools afloat and many PTAs assumed responsibility for the funding and management of primary schools
(Reinikka and Svensson, 2004), a practice confirmed in our districts of research: parents were participating strongly
in the management of the schools; funding was collected from them and managed on a central account by the
district education office, from which instructional materials and other goods were bought.
This funding mechanism was in place until the introduction of UPE, after which the government resumed the
responsibility for the funding and support of primary schools and indicated that SMCs were to have the formal
responsibility for primary schools. This power shift from PTA to SMC was underscored drastically by the banning
of PTAs at the onset of the UPE, a measure that was lifted later in the face of protests (Dauda, 2004: p. 30).
Nevertheless, it made clear that, with the resumption of public funding of schools, the government ensured that the
school management was also accountable to local government and not, or not exclusively, to parents or religious
groups. Unsurprisingly, several PTAs are not pleased with the loss of influence and the limits placed on their powers
to raise money from the parents (Saito, 2003: p. 158). On the other hand, practices in our districts of research
suggest that with the introduction of UPE, parents are no longer keen on participating voluntarily in the
management of the schools, as they expect everything to be done by the government.
Decentralised education budgets
Decentralised education budgets in Uganda are significant in volume and have weak accountability procedures,
making them prone to elite capture. All SMCs receive capitation grants between 2.70 and 4.40 USD per child per
year, depending on the child’s class, to cover expenditure in instructional material, extra-curricular activities,
maintenance and utilities and administration costs (GoU Ministry of Education and Sports, 1998: p. 4). About a
quarter of central government’s budget is disbursed downwards via grants, salaries and funds to the education
sector, 70% of that to primary education (Saito, 2003: p. 152). The fact that donor agencies contribute about half of
these funds (Dauda, 2004: p. 31) does not matter at local level.
Added to this are budgets that district councils contribute to primary education. Saito’s study of three districts
reveals increasing contributions to education projects from the mid 1990s to 2001, peaking between 43 and 70% of
total district expenditure (Saito, 2003: p. 166). As for accountability processes, it is claimed that UPE funds are
swindled ‘with impunity’ (UPPAP, 2002) and that education is the sector most affected by political corruption
(Centre for Basic Research, 2005: p. 58). Reinikka and Svensson (2004: p. 688) suggest that well-connected
citizens, politicians and district officials sit together in order to decide how these funds should be used. A survey
from 1991–1995 indicates that only between 13 and 20% of the central government grants reached the local
primary schools. The government also increased its funding for the primary education sector in the early 90s
(between 1991 and 1995, funding increased by 40% in real terms). The central government provided the salaries of
the primary school teachers and non-wage spending through capitation grants and funding for rehabilitation
(Reinikka and Svensson, 2004).
The levels of embezzlement have dropped considerably in recent years and the average school in 2001 received
80% of its entitlements (Reinikka and Svensson, 2005: p. 261). Our districts of research also confirm to this
tendency, whereas key-actors (both within the government, education sector in general and civil society) strongly
complain about the levels of corruption during the 80s and 90s, they also claim that this has strongly improved in the
recent years. This tendency is confirmed in our districts of research. An analysis of 15 schools which were part of
our research shows that how the central government grants allocated to the district— where they are publicly
displayed in the office of the DEO— reached these schools for the full 100%, without any money being embezzled
on the way to the school.
4
Once in the school, the SMCs are responsible for spending these funds within
well-specified bandwidths on specific issues: 50% on instructional materials, 30% on co-curricular activities, 15%
on utilities and maintenance and 5% on administration (GoU Ministry of Education and Sports, 1998: p. 6).
4
Formally, 5% of the grant is being kept by the district for inspection purposes, and 10% is kept by the Ministry of Education for buying
textbooks.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
UGANDA’S DECENTRALISED PRIMARY EDUCATION 153

Decentralisation and corruption in Uganda
According to the USAID 2005 Democracy and Governance assessment of Uganda (2005: p. 37) ‘corruption in
Uganda is pervasive, institutionalised and on the increase’. General reports on Uganda’s politics emphasise the ‘the
neo-patrimonial character of Ugandan political culture’ (Robinson, 2006: p. 33, see also Barkan et al., 2004). This
corruption problem is also reflected in public perceptions on the extent and the intensity of the phenomenon. The
2003 second National Integrity Survey found corruption taking place on both a national and local level; the highest
levels of corruption taking place among the traffic police (83%), tender boards (79%), electoral commissions (78%)
and Uganda Revenue Authorities (77%) (Deininger and Mpuga, 2005).
This political culture has a strong impact on processes of decentralisation. Francis and James (2003: p. 336)
argue that in practice, ‘(...) those with vested interests are capable of turning the institutions and opportunities
created through decentralisation to their own advantage’ (see also Ellis and Bahiigwa, 2003: p. 1010). Juma (2002)
argues that the decentralised structures of local government have come together with the National Resistance
Movement (NRM) structure and ‘have become symbols of decentralised corruption as well as instruments of
patronage for political loyalists of NRM regime’ (Juma, 2002: p. 19).
The 2004 Joint Annual Review of Decentralisation of the Ministry of Local Government points out the fact that
the decentralisation principles are not fully grasped with, increasing the risk ‘for the process of decentralisation to
degenerate [...] into a scramble for local influence and local power’ (GoU Ministry of Local Government, 2004:
p. 4, quoted in Steiner, 2006: p. 62). In such a political context, decentralised public services run a high risk of being
captured by local elites. This view is also reflected in national government policies, in which references to local
corruption and incapacity of local institutions have been used to roll back progress on decentralisation through
recentralising key local government appointments and eliminating sources of local government revenue such as the
graduated tax (cf. USAID, 2005).
RESEARCH FINDINGS
This section aims at answering the research questions which were brought up in the introduction and which
investigate whether SMCs are prone to elite capture and which formal and informal constraints play a role in this.
Firstly, do SMC members belong to the local elite? Secondly, have SMC members captured the committees?
Thirdly, what is the distribution of power? Fourthly, do politicians capture the committees? Fifthly, do education
officials capture the committees? Through answering these questions, we will be able to analyse whether SMCs are
prone to elite capture and which formal and informal constraints play a role in this.
Do SMC members belong to the local elite?
Parents only elect two of the nine SMC members directly among themselves (the headteacher is the tenth,
non-voting, member). Three of the SMC members are to be formally appointed by the Ministry of Education and
the remaining four members, including the chairperson, are to be appointed by the District Council’s Education
Committee (GoU Ministry of Education, 1969, art. 1.2). This arrangement of 1969 has not been changed in
subsequent education policies (it was reprinted in the official SMC manual of 1999) and formally reiterates central
government’s desire to keep control or provide a framework for central government to make alliance with local
elites.
However, in practice the headteacher and a sitting SMC propose to the district authorities a list of the other seven
candidates for appointment to represent the Ministry of Education (three) and the District Council’s Education
Committee (four). The Ministry’s manual for headteachers advises to follow this procedure (GoU Ministry of
Education et al., 1996: p. 85, 87). The interviews confirmed this practice, which suggest that SMCs can hardly be
considered a framework in which local and national elites establish alliances. Rather, it seems that central
government deliberately leaves the control over decentralised education in the hands of local political dynamics.
Table 1 sheds light on the profiles of SMC members. About half the committee members are in the age bracket
30–44 years and about a third of the members are women. Especially the latter is interesting because the regulations
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
154 G. PRINSEN AND K. TITECA

that guide the election process for SMCs do not stipulate a one-third quota for women, unlike, the regulation
regarding the composition of District Councils (GoU Republic of Uganda, 1997, Article 21 (e): p. 15) or
management committees for Health Units (GoU Ministry of Health, 1998: p. 7). This contradicts other research of,
for example, Agarwal (2001), who found that women in Community Forest Groups end up being underrepresented
if there is no quota set aside.
Table 2 reveals that, by their own accord, 90% of the (s)elected members of the SMC have at least completed
primary education and half of them completed secondary education. This education level is confirmed by two other
sources of information. First, it was noted that only about a quarter of the SMC members required assistance to fill
in the English questionnaire. Second, Table 3 reveals that about one-third of the SMC members are active or retired
civil servants, who must be literate. Furthermore, in these rural districts, the overwhelming majority of parents must
be subsistence farmers, but only 40% of the SMC members identify themselves as such. The majority of SMC
members can be considered the ‘better-off’ and identify themselves as farmers with employees, traders, salaried
employees or civil servants.
Have SMC members captured the committees?
Legally, SMC members’ office tenure is limited to 2 years and members are re-eligible once for another term (GoU
Ministry of Education, 1969: p. 2.1–2.3). In practice, almost all SMCs claimed that their effective term in office was
4 years (Table 4). The graph reveals that three quarters of the members has an office term within the legal limits and
the even distribution suggests that members are replaced at a regular interval. (The fact that a quarter of the
respondents has an office term of more than 5 years suggests more of an irregularity than it actually is because this
quarter consists in its majority of headteachers, who are ex-officio members of the SMC).
The impression that SMC members do not turn their seats into personal life-time fiefdoms is further confirmed
by interviews with current and former SMC members. Most members saw it as an honour to be member of the
SMC, as it gave them some social esteem and standing. Nevertheless, this social value was not deemed very
significant when compared to other local bodies (such as local councils). Moreover, a lot of hard work is required,
Table 1. Sex and age representation of SMC members
Male (%) Female (%) % n
18–29 years 2 2 6 7
30–44 years 38 13 50 64
45–60 years 21 13 35 44
61 years or older 8 2 9 12
Total 69 30 100 127
Note:n¼127 elected and ex-officio members of 17 SMCs in three districts in Uganda.
Source: Field research in 2004.
Table 2. Educational background of SMC members
%n
Uncompleted primary education 10 10
Completed primary education 21 21
Uncompleted secondary education 19 19
Completed secondary education 17 17
Secondary education plus other training 30 30
Other 2 2
Total 100 99
Note:n¼99 elected members of 17 SMCs in three districts in Uganda.
Source: Field research in 2004.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
UGANDA’S DECENTRALISED PRIMARY EDUCATION 155

taking up many volunteer hours. Former and current SMC chairpersons gave a variety of reasons for rescinding
their positions: the lack of financial reward, the hard work and the lack of support from the parents. As a result, very
few members aspire a re-election.
While SMC positions are not captured as personal fiefdoms, an analysis of SMC members’ other public
positions (e.g. in churches, soccer clubs, water committees, village councils etc.) does reveal an interesting pattern.
The inside questionnaire asked SMC members to indicate if they held other positions in committees or boards,
14 specified local bodies were listed for ticking. Table 5 reveals that only a few (9%) of the SMC members claimed
Table 3. Professional background of SMC members
%n
Farmer 40 40
Farmer with employees 9 9
Trader or business (wo)man 9 9
Employee in a business 4 4
Active civil servant 12 12
Retired civil servant 18 18
Student or homemaker 3 3
Other 4 4
Total 100 99
Note:n¼99 elected members of 17 SMCs in three districts in Uganda.
Source: Field research in 2004.
Table 4. Duration of office tenure of SMC members
%n
1 year or less 26 33
1–3 years 27 34
3–5 years 22 28
More than 5 years 22 28
Other 3 4
Total 100 127
Note:n¼127 elected and ex-officio members of 17 SMCs in three districts in Uganda.
Source: Field research in 2004.
Table 5. Number of positions held in other committees or boards by SMC members
%n
No other positions 9 12
One other position 1 1
Two other positions 35 45
Three other positions 27 35
Four other positions 17 21
Five other positions 4 5
Six other positions 5 6
Seven other positions 1 1
Eight other positions 1 1
Total 100 127
Note:n¼127 elected and ex-officio members of 17 SMCs in three districts in Uganda.
Source: Field research in 2004.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
156 G. PRINSEN AND K. TITECA

not to hold other positions, and a stunning 90% claim to hold in effect three or more positions in public bodies.
5
Only 1 in 127 respondents held a mere two positions.
6
This pattern is further confirmed by a survey (n¼64) on the
life histories of SMC members, which found that on average, SMC members first were an executive member of two
other institutions before becoming a member of an SMC. Only 5% directly became an SMC member, without
having been a member of any other executive committee. This suggests that SMC members need to have a certain
experience and maturity before they can be elected in the SMC. Moreover, of the 20 SMC chairpersons which were
interviewed, 8 had no children in that particular school. This confirms that having children in the school is not a
strict requirement for the school, but that SMCs are looking for local elites as ‘all brains must come together for the
school’.
7
If SMC membership is a typical example of membership of public bodies at village level, then these findings
suggest that public bodies at village level in Uganda are controlled by three or four dozen individuals. However,
rather than seeing one or two persons dominate a particular local body on a permanent basis, the control over local
bodies seems to be rotating among the members of a select group of villagers who are likely to be the locally
better-off, and about a third of them have a background in the civil service. They move over the available positions
of public bodies in the village as if they were musical chairs. With most positions having some sort of electoral
process, it can be argued that the music is played—and stopped — by each organisation’s constituency.
These findings are strongly confirmed by in-depth research in two villages (which both have a primary school
with a SMC) that looked from the ‘outside’ at the SMCs. In every village, the whole adult population was
interviewed (village I, n¼222; village II, n¼204). In village I, 25.6% of the adult population are holding executive
positions in local institutions, while the remaining 74.4% does not hold a single executive position. These executive
members hold on average three positions per person. Out of the adults with no executive position, only 6.6% had an
executive position in the past. More or less the same image arises in village II: 23.5% dominate executive positions,
and hold on average three positions per person. Among the adults without executive position, 14% had an executive
position in the past. This accumulation of positions suggests that the image of ‘recycled elites’ who have to do ‘hard
work’ is apt (Chabal and Daloz, 1999: p. 38).
Distribution of power
Findings from individual outside interviews mostly argue that power in SMCs is concentrated in a tandem of two
persons: chairpersons and headteachers. This is probably fed by two factors. First, according to most of the outside
interviewees there are many SMC members with low education levels unaware of their rights and duties. Three
quarters of the parents believe these uneducated members are brought in because they are ‘easier to control’ by
elites within the SMCs. They can be easily manipulated by the better informed members like the chairperson and
the headteacher. This is reinforced in many cases where ordinary members lose interest in the activities of the SMC.
This leaves the more ‘powerful’ chairpersons and the headteachers to work on their own. It could therefore be
argued that if chairpersons or headteachers embezzle funds, this is possibly as much due to their pursuit, as by a
pattern in which other SMC members abdicate from responsibilities. This was strongly the case before the
introduction of UPE, as there was little training and sensitisation of the SMCs. After the introduction of UPE, the
government had all advantage in ensuring a proper check on the government money being used. As a result, it
strongly increased funding for trainings for the SMC members. All key-actors confirm how members consequently
became more aware of their rights and responsibilities, leaving less room for mismanagement by the headteachers
and chairpersons.
5
A caveat may be in place. When puzzled by the overwhelming number of 90%, we considered that the high percentage may arise partly from the
possibility that some respondents mistook the question ‘Are you also a member of a Committee or Board in other organisations?’ with simple
membership of other organisations.
6
The 127 respondents in Uganda form part of a wider sample of 420 respondents in local committees in Uganda and Tanzania. Of those
respondents, 344 (82%) claimed to hold three or more other positions and only one—indeed this very same SMC member in Uganda —claimed
to hold just one other position.
7
Interview local teacher and PTA chairman, 9 November 2005.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
UGANDA’S DECENTRALISED PRIMARY EDUCATION 157

Second, the potential power of the tandem chairperson-headteacher is reinforced by the way in which financial
management in schools is executed. The joint signatures of headteacher and SMC chairperson are required to
authorise expenditure and payments (GoU Ministry of Education and Sports, 1999: p. 82–83). Outside interviews
confirm that this coalition is the main vehicle of corruption in SMCs— many examples were given of cases wherein
the headteacher and chairperson embezzled money. For example, the chairperson signs an empty cheque for the
headteacher, and in return gets financial and material rewards. Again, much of this had to do with the lack of
information for the other SMC members, in which they were not aware of the limits on the power of the headteacher
and chairperson; some members were also easily bribed with ‘lunch money’.
8
Nevertheless, this administrative
requirement of a double signature is one of the checks-and-balances in the schools’ financial management that
actually limits the potential for one single person to rule — and SMCs seem to adhere to the rules. It seems unlikely
that increasing the number of signatories will stop embezzlement, it will rather make the system more cumbersome
than transparent.
Observations during the inside group interviews with the SMCs recorded the number of persons who made at
least a twice above average (that is total effective interview time, divided by the number of participants) verbal
contribution to the interview (Table 6). Findings indicate that only rarely the meetings were dominated by one
person, generally two or three persons shared the floor — often indeed the headteacher and the chairperson
(a parent) as well as a parent representative. Likewise, notes were kept on who assumed effective leadership during
the participatory exercises (Table 7). In three quarters of the interviews, the headteachers or the chairpersons led or
facilitated the group dynamics. Overall, findings do not show a pattern in SMCs of a concentration of power and
control in one single person or function.
Table 6. Number of persons with twice above average contribution to SMC group
interviews
n
One person 1
Two persons 7
Three persons 7
Four persons 2
Total 17
Note:n¼17 SMCs in three districts in Uganda.
Source: Field research in 2004.
Table 7. Informal leadership during group interviews with SMCs
n
Headteacher 9
SMC chairperson 4
Other member 1
No individual leadership 3
Total 17
Note:n¼17 SMCs in three districts in Uganda.
Source: Field research in 2004.
8
Interview coordinating centre tutor Teacher Development Management System 07 June 2007.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
158 G. PRINSEN AND K. TITECA

Do politicians capture the SMCs?
Uganda elects councillors at three levels: in a village council, a sub-county council and a district council. Table 8
indicates that about a quarter of the SMC members also hold elected positions in the village council; every SMC has
at least one. However, the village councils do not hold much power or resources (SMCs’ budgets far exceed village
councils’ budgets) and do not prove very functional: in the villages of research, none of the village councils were
working.
9
It is more relevant that, out of the 17 SMCs, 7 have a sub-county councillor among their members—
sub-county councils are the lowest level of the hierarchy that do hold some effective sway for the education sector
(Saito, 2003: p. 149). There are no district councillors among the SMCs of research.
All in all, there are relatively few politicians present in SMC’s. Rather than having politicians trying to ‘capture’
the SMC, the research suggests the inverse: SMC’s continuously try to involve higher-level politicians beyond the
village horizon. The interviews reveal that—contrary to the regulations — most actors (parents, SMC members,
politicians) assume that it is formally required to include local, higher-level politicians in the SMC. When SMC
elections are due, politicians will be approached. In spite of their often substantial efforts in net-working and
travelling, most schools fail to attract a (sub-)district councillor to their SMC. And even if these higher-level
councillors allow themselves to be ‘enlisted’, their attendance at meetings or the provision of extra resources to the
school is often a source of sotto voce disappointment. An analysis of SMCs’ efforts to involve powerful persons in
their school brings to mind the image of ‘inverted elite capture’: it is the committees that try to capture the elite, not
the other way around. One SMC chairman puts it bluntly: ‘Politicians don’t approach us to get support —it is we the
SMC, who try to approach politicians!’.
10
Although SMCs mostly fail to involve councillors formally, they continuously try to establish informal
relationships with them by inviting for special occasions such as fund-raisings, school days or meetings. Repeatedly
villagers indicated that councillors need to be involved for two reasons. First, councillors are expected to assist in
addressing problems with an SMC. Not only can councillors command attention from parents, they are also
expected to be effective in solving corruption cases because they are better educated or experienced. Additionally,
councillors are expected to have better access to district education authorities and they are less vulnerable to
reprisals. Second, councillors are expected to lobby for additional funding for the school. All schools lack resources
and schools compete for a small pool of ‘additional’ resources.
The first additional resources can be found in a School Facilities Grant that the Ministry of Education allocates to
districts for investment in schools (Saito, 2003: p. 157). A second source is a council’s own budget from which it
can allocate discretionary grants to schools (GoU Ministry of Education and Sports, 1998: p. 18). This money
comes from LGDP resources or from locally generated revenue.
11
A third source is material support from NGOs.
Table 8. SMC members who also hold a political office
%n
Holds no political office 69 87
Member of village council 25 32
Member of sub-county council 6 7
Member of district council 1 1
Total 100 127
Note:n¼127 elected and ex-officio members of 17 SMCs in three districts in Uganda.
Source: Field research in 2004.
9
A similar analysis was found by, for example, Karlstrom, who found that out of seven LC1’s in southern Uganda, only one was described as
‘very active’ (Karlstro
¨m, 1999: p. 112–114).
10
Interview SMC chairman, 24 November 2005.
11
The abolition of the Graduated Tax in June 2005, which was the principal source of local revenues, however affected local governments and
their service delivery activities. Compensation is provided by the central government, but this often falls short of replacing revenue shortfalls
(USAID, 2005: p. 39–42).
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
UGANDA’S DECENTRALISED PRIMARY EDUCATION 159

This support should, however, not be overestimated. Firstly, within the ‘outside’ district of research, NGO support
was only given to two out of fivecounties. Secondly, this material support usually consisted of a few bags of cement,
some sheets of corrugated iron or extra books; never something that would normally be beyond the SMCs’ own
budgets. In fact, the ‘inside’ interviews suggest that SMCs appreciated the access to advice and information just as
much. Although this NGO support mainly passes through the DEO, district and sub-county councillors also play a
role in the identification of beneficiaries. An informal relationship with a councillor is therefore considered to be a
strong advantage in the allocation of these funds, something which is strongly confirmed by councillors themselves.
As one sub-county councillor argued ‘I am lobbying my colleagues to support schools in the parish in case I want to
move the money to my parish. (...) Of course we try to get funding to our schools of interest. Moreover, whatever
problems schools have, they first come to me, before going to the sub-county’.
12
These are the perceived benefits for the SMC, but what are councillors’ considerations when looking at SMCs’
requests for involvement? Rent-seeking from SMCs is unlikely to bring much, whatever is to be obtained needs to
be shared with headteachers and chairpersons. This might be one of the reasons where relatively few councillors are
a member of the SMC: they face high expectations on their time and there is little economic ‘added value’. On the
other hand, schools can be useful for a politician’s profile and legitimacy. Supporting schools improves a politician’
visibility and legitimacy by tying their name to ‘delivered’ goods through openings, school days or fund raisings.
Also for national politicians, schools are a good ‘investment’ for local political legitimacy. All MPs and ministers of
the district had donated goods to schools. However, contrary to local politicians, parents do not expected them to get
involved in solving management problems in the schools.
These practices can be described as ‘pork barrel’ politics, a common form of clientelism in which politicians
take care of their own constituencies by providing them with whatever they can get out of the governmental ‘pork
barrel’.
13
In other words, school funds are allocated not on the basis of need or merit, but to build political support
for a politician in an area. Anyone outside of this favoured circle will have difficulty accessing these public
resources. As one councillor in charge of education comments: ‘There is no rationality in the council. Everyone
fights for his own area! (...) Some of the politicians are development minded, but others are sectarian. They only
give School Facility Grants to areas which give them votes. (...) They also look if the headteacher and SMC are of
the same political colour as them’.
14
In this respect, some schools claim they have not been able to access appropriate funding because they lack
political connections: they have no councillor representing their interests, or they do not support those in charge of
the education budget. Although pork barrel politics is legal, those excluded feel they have been treated unfairly. In
this sense, funding for SMCs is often subject to a political bias and ‘captured’ by politicians (cf. Pasuk and Baker,
1996).
Do education officials capture the SMCs?
SMCs receive their grants from the Ministry of Education via the district education department in their bank
account in nine payments per year (GoU Ministry of Education and Sports, 1998: p. 4), and are accountable to the
district education office. Although the regulations accompanying the introduction of UPE do not state that the
SMCs report to the DEO or the District Education Committee, the 1996 manual for headteachers does (GoU
Ministry of Education et al., 1996: p. 83), as well as the older, still standing, 1969 regulations (GoU Ministry of
Education, 1969: p. 14.1). In practice, each of the three DEOs stated that the reporting on budgets and expenditure
must be done by the headteachers with the SMC chairpersons’ signature to the District Education Committee (and
DEO) before next tranches of grants can be released.
The respondents particularly stress the importance and power of the district-based DEO, even though recent
policies also establish that the sub-county authorities should supervise and support SMCs (GoU Ministry of
12
Interview sub-county councillor 23 November 2005.
13
Pork barrel politics can be defined as ‘(...) publicly funded projects promoted by legislators to bring money and jobs to their own districts, as a
political favour to local politicians or citizens’. (Brinkerhoff and Goldsmith, 2002: p. 6).
14
Interview district councillor, 6 December 2005.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
160 G. PRINSEN AND K. TITECA

Education and Sports, 1998: p. 13). But even sub-county authorities themselves indicate unanimously in the
interviews that they lack time, resources and knowledge to deal with SMCs. Likewise, the interviews show that only
4 of the 17 SMCs mention the sub-county as an actor of medium importance with whom they deal, but all 17 place
the DEO on or near the top of their list of contacts as a major actor. SMCs are acutely aware that their core
government funding (capitation grants) continues to come via the DEO’s office, not the sub-county.
Besides the DEOs’ capacity to influence decisions about additional resources from the School Facilities Grant,
the inside interviews also show that DEOs are far more influential than councillors as an intermediary between
schools and (international) NGOs with resources. Fifteen of the 17 SMCs mention that they receive materials or
advice from NGOs. However, only two SMCs received assistance directly from NGOs, whereas 11 SMCs accessed
this assistance via the DEO. (The profile of the first two SMCs did not differ from the other SMCs, so coincidence
rather deliberate action seems to explain the direct contact between these SMC and the NGO). In only two cases
councillors were indicated as the intermediary. It should also be noted here that NGOs are largely dependent on
DEOs in their targeting of schools. And because this targeting largely takes place outside public scrutiny, DEOs
have ample room to introduce personal preferences for certain schools. These tendencies are strongly confirmed by
outside interviews: most actors think the DEO has an overall responsibility to support the school, as he or she has
more power and resources.
As the DEO is the main actor in a district’s education policies and resources, accusations of embezzlement are
frequent and in one of the three researched districts, the DEO had recently been dismissed for corruption. In another
district, it was rumoured by several interviewees that the DEO had set up a scheme with the headteachers of the two
biggest primary schools in which part of the UPE money was being divided between the DEO, the headteachers and
the SMC chairpersons.
15
CONCLUSIONS
This paper started with the observation that current decentralisation policies in Uganda increasingly place the
responsibilities and the resources for the delivery of public services in the hands of public bodies at the lowest
possible levels, in this case SMCs. These decentralisation policies take place in a wider governance context in
which public bodies of the central state have proven to be rather ineffective in the delivery of these services, an
ineffectiveness that several analysts attribute to elite capture. Through a series of empirical questions, this paper
investigated whether SMCs are prone to elite capture and which formal and informal constraints play a role in this.
Our research finds that SMC members’ personal profiles do not prima facie suggest they are the local elite: most
fall in the younger age bracket 30–44 years and a third of them are women. Furthermore, formal constraints are an
important and reasonably effective check against elite capture. First, election procedures are respected with regard
to the election of SMC members. Second, although the SMC members differ from the average parent in that most of
them have a regular cash income, almost half finished secondary school and a third have a professional background
in the civil service (giving them a potential ‘elitist’ character), the formal rules with regard to term limits prevent
them from personally capturing control over the committee because virtually all of them are replaced in elections
after one or two terms in office. Third, the formal constraint in which both headteacher and SMC chairperson are
overall responsible of the SMC also prevents capture of power in the hand of one single person. In structure as well
as in practice, the management of schools reflects a pattern in which power and responsibilities are generally
divided between these two persons. Whereas there is a danger of embezzlement by these two actors, increased
training and sensitisation of SMC members after the introduction of UPE proved to be an important check against
the capture of the SMC by headteacher and SMC chairperson.
Apart from these formal rules, there are also several informal constraints at play in the functioning of SMCs. The
most striking informal rule— or better an ‘internally enforced standard of conduct’ and an ‘extension of formal
rules’ (North, 1990: p. 40)— is the practice of ‘inverted elite capture’: SMCs always try to capture the higher level
15
These statements were frequently and independently made by key actors, for example, interview PTA chairman (21 October 2005), former
SMC chairman (11 November 2005) and (4 December 2005), local district councillor (6 December 2005).
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
UGANDA’S DECENTRALISED PRIMARY EDUCATION 161

elites into their organisation. Notwithstanding this informal practice, most SMC fail to succeed. Rather than
becoming a member of SMCs, political elites use ‘pork barrel’ politics to ensure the political support of certain
schools, through which resources get ‘captured’ for a particular school. In this sense, informal (political) standards
act as an elaboration of the formal technical rules for the distribution of funds, which may affect coherent and
effective policy, in that resources for particular schools can be intercepted by powerful politicians contrary to the
interests of underprivileged schools
16
(Johnson and Start, 2001: p. 12–15). Whereas this informal practice may
have favourable outcomes for individual SMCs, it undermines policy intentions with regard to SMCs. A similar
dynamic can be seen for district education officials. Through their formal and informal power, DEOs seem to have
more opportunity than politicians to steer resources to personally favoured schools and thus lay a basis for
clientelism. This opportunity emerges particularly when DEOs are the exclusive intermediary between NGOs and
schools.
Whereas it is clear that individuals have not personally taken control of the SMC, there is strong evidence that at
village level a group of about three or four dozen individuals controls all positions in local public bodies (ranging
from church to soccer club) because almost all SMC members accumulate three or four positions in these various
bodies. In the course of the years, they seem to rotate over these positions as if they were musical chairs, mostly
sanctioned by elections. Therefore, there seems to be an informal constraint, not only within the SMC, but within
the whole village that only a ‘rotating elite’ group is eligible for SMCs.
On balance, the combined findings from these two research projects find that there are two tendencies at work
which influence the functioning of SMCs and the outcome of elite capture. On the one hand, informal norms lead to
the problem of ‘indirect capture’, in which resources of SMCs are subject to political (in the case of politicians) or
personal bias (in the case of the DEO’s). Moreover, our findings suggest a rotating elite controlling local
institutions. On the other hand, formal rules effectively prevent this local elite taking direct, permanent or personal
control over SMCs.
REFERENCES
Agarwal B. 2001. Participatory exclusions, community forestry, and gender: an analysis for South Asia and a conceptual framework. World
Development 29(10): 1623–1648.
Ahmad J, Devarajan S, Khemani S, Shah S. 2005. Decentralization and Service Delivery. World Bank Policy Research Working Paper 3603.
Baland JM, Platteau JP. 1999. The ambiguous impact of inequality on natural resource management. World Development 27(5): 773–788.
Bardhan P, Mookherjee D. 2000. Capture and governance at local and national levels. American Economic Review 90(2): 135–139.
Barkan J, Simba S, Ng’ethe N, Titsworth J. 2004. The political economy of Uganda. Background paper commissioned by the World Bank in
fulfilment of Purchase Order 7614742, Final Draft of July 6, 2004.
Boone C. 2003. Decentralization as political strategy in West Africa. Comparative Political Studies 36(4): 355–380.
Brinkerhoff D, Goldsmith A. 2002. Clientelism, Patronage and Democratic Governance: An Overview and Framework of Assessment and
Programming. ABT Associates Inc: Cambridge.
Creative Association International Inc. Smith I, 2000. USAID/Uganda: Education Sector Investment Program Review, for: The Global Bureau,
Human Capacity Development Center; US Agency for International Development.
Carvalho S, White H. 1997. Combining the quantitative and qualitative approaches to poverty measurement and analysis. World Bank Technical
Paper No. 366: Washington, DC.
Centre for Basic Research. 2005. The Impact of Political Corruption on Resource Allocation and Service Delivery in Local Governments in
Uganda. Centre for Basic Research: Kampala.
Chabal P, Daloz JP. 1999. Africa Works. The International Africa Institute in Association With James Currey: Oxford.
Crook R. 2003. Decentralization and poverty reduction in Africa: the politics of local-central relations. Public Administration and Development
23(1): 77–88.
Crook R, Sverrisson A. 2001. Decentralisation and poverty alleviation in developing countries: a comparative analysis or is West Bengal unique?
IDS Working Paper 130.
Darrow M, Thomas A. 2005. Power, capture and conflict: a call for human rights accountability in development cooperation. Human Rights
Quarterly 27: 471–538.
Dauda C. 2004. The importance of de facto decentralization in primary education in Sub-Saharan Africa. Journal of Planning Education and
Research 24(1): 28–40.
16
During the field research, key-actors gave many independent examples on schools which were lacking basic provisions and should be awarded
so on an objective basis, but lack the necessary political connections and are therefore neglected.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
162 G. PRINSEN AND K. TITECA

Deininger K, Mpuga P. 2005. Does greater accountability improve the quality of public service delivery? Evidence from Uganda. World
Development 33(1): 171–191.
Devas N, Grant U. 2003. Local government decision-making: citizen participation and local accountability, some evidence from Kenya and
Uganda. Public Administration and Development 23(4): 307–316.
Ellis F, Bahiigwa G. 2003. Livelihoods and rural poverty reduction in Uganda. World Development 31(6): 997–1013.
Francis P, James R. 2003. Balancing rural poverty reduction and citizen participation: the contradictions of Uganda’s decentralization
programme. World Development 31(2): 325–337.
GoU Ministry of Education. 1969. The Education Management Committees Amendment Rules 1969 No. 224. Government of Uganda: Entebbe.
GoU Ministry of Education and Sports. 1998. Guidelines on Policy, Roles and Responsibilities of Stakeholders in the Implementation of
Universal Primary Education (UPE). Government of Uganda: Entebbe.
GoU Ministry of Education and Sports. 1999. School Management Committee Training Manual Facilitator’s Session Plans. Government of
Uganda: Entebbe.
GoU Ministry of Education, Ekatan V, et al. 1996. Headteacher Training Series Module 4. Government of Uganda: Entebbe.
GoU Ministry of Health. 1998. Training Module for Sub-County Health Committes and Health Unit Management Committees. UNICEF/
Ministry of Health: Entebbe.
GoU Ministry of Local Government. 2004. Report on the Joint Annual Review of Decentralisation 2004, Parts I–III. Ministry of Local
Government: Kampala.
GoU Republic of Uganda. 1997. The Local Governments Act 1997. Government of Uganda: Entebbe.
Hall P, Taylor R. 1996. Political science and the three new institutionalisms. Political Studies XLI: 936–957.
Heller P. 2001. Moving the state: the politics of democratic decentralization in Kerala, South Africa, and Porto Alegre. Politics and Society 29(1):
131–163.
Johnson C, Start D. 2001. ‘Rights, claims and capture: understanding the politics of pro-poor policy’, Overseas Development Institute, Working
Paper 145.
Juma Okuku A. 2002. ‘Ethnicity, State Power and the Democratization Process in Uganda’, Discussion Paper 17. Uppsala: Nordiska
Afrikainstitutet.
Jutting J, Kaufmann C, McDonnell I, Osterrieder H, Pinaud N, Wegner L. 2004. Decentralisation and Poverty in Developing Countries:
Exploring the Impact. Paris: OECD DAC.
Jutting J, Corsi E, Kauffmann C, McDonnell I, Osterrieder H, Pinaud N. 2005. What makes decentralisation in developing countries pro-poor?
European Journal of Development Research 17(4): 2005.
Kanbur R. 2005. Qualitative and quantitative poverty appraisal: the state of play and some questions. In Q-squared: Qualitative and Quantitative
Poverty Appraisal: Complementarities, Tensions and the Way Forward, Kanbur R (ed.). Q-squared Working Paper No 1.
Karlstro
¨m M. 1999. Civil society and its presuppositions: lessons from Uganda. In Civil Society and the Political Imagination in Africa —Critical
Perspectives, Comaroff JL, Comaroff J (eds). The University of Chicago Press: Chicago; 104–123.
Kingston C, Caballero G. 2006. Comparing theories of institutional change. Paper presented at the Annual Conference of the International
Society for New Institutional Economics. Boulder, Colorado.
Langseth P. 1996. Civil service reform: a general view. In Democratic Decentralization in Uganda: A New Approach to Local Governance,
Villadsen S, Lubanga F (eds). Fountain Publishers: Kampala.
Manor J. 1999. The Political Economy of Democratic Decentralisation (Series Directions in Development No. Series Directions in
Development). Washington DC: World Bank.
Manor J. 2004. User committees: a potentially damaging second wave of decentralization. European Journal of Development Research 16(1):
192–213.
Mansuri G, Rao V. 2003. Community based (and driven) development: a critical review. World Bank Research Observer 19(1): 1–33.
McLean K, King E. 1999. Decentralization of the education sector. In Decentralization Briefing Notes. World Bank Institute Working Papers,
Litvack J, Seddon J (eds). World Bank: Washington DC, USA.
Ndegwa S. 2002. Decentralization in Africa: A Stocktaking Survey. Africa Region Working Paper Series No. 40. World Bank: Washington DC.
North D. 1990. Institutions, Institutional Change and Economic Performance. Cambridge University Press: Cambridge.
North D. 1991. Institutions. The Journal of Economic Perspectives 5(1): 97–112.
OECD. 2004. Lesson Learned on Donor Support to Decentralisation and Local Governance. AC Evaluation Series: Paris.
Oxhorn P. 2004. Unraveling the puzzle of decentralization. In Decentralization, Democratic Governance, and Civil Society in Comparative
Perspective—Africa, Asia, Latin America, Oxhorn P, Tulchin JS, Selee AD (eds). Johns Hopkins University Press: Baltimore.
Pasuk P, Baker C. 1996. Thailand: Economy and Politics. Oxford University Press: Oxford.
Patterson AS. 2003. Power inequalities and the institutions of Senegalese development organizations. African Studies Review 46(3): 35–41.
Platteau JP, Gaspart F. 2003. The risk of resource misappropriation in community-driven development. World Development 31(110): 1687–1703.
Powell W, Di Maggio P (eds). 1991. The New Institutionalism in Organizational Analysis. The University of Chicago Press: Chicago and
London.
Rao V, Ibanez AM. 2001. The Social Impact of Social Funds in Jamaica: A Mixed-Methods Analysis of Participation, Targeting and Collective
Action in Community-Driven Development. Working Paper, The World Bank: Washington, DC.
Reinikka R, Svensson J. 2004. Local capture: evidence from a central government transfer program in Uganda. The Quarterly Journal of
Economics 119: 679–705.
Reinikka R, Svensson J. 2005. Fighting corruption to improve schooling: evidence from a newspaper campaign in Uganda. Journal of European
Economic Research 3(2–3): 259–267.
Ribot J. 2001. Local actors, powers and accountability in African decentralizations: a review of issues (Paper commissioned by UNRISD and
International Development Research Centre of Canada). New York: UNRISD.
Ribot J, Agrawal A, Larson A. 2006. Recentralizing while decentralizing: how national governments reappropriate forest resources. World
Development 34(11): 1864–1886.
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
UGANDA’S DECENTRALISED PRIMARY EDUCATION 163

Robinson M. 2006. The political economy of governance reforms in Uganda. IDS Discussion Paper No 386. Sussex: IDS.
Rondinelli D. 1999. What is decentralization? In Litvack J, Seddon J (eds), Decentralization Briefing Notes. World Bank Institute Working
Papers. World Bank: Washington DC, USA.
Rondinelli D, Nellis J, Cheema S. 1983. Decentralization in Developing Countries. A Review of Recent Experiences. World Bank: Washington
DC, USA.
Saito F. 2003. Decentralization and Development Partnerships — Lessons from Uganda. Springer: Tokyo.
Scott J. 1998. Seeing Like a State—How Certain Schemes to Improve the Human Condition Have Failed. Yale University Press: New Haven.
Selee A, Tulchin J. 2004. Decentralization and democratic governance: lessons and challenges. In Decentralization, Democratic Governance,
and Civil Society in Comparative Perspective— Africa, Asia, Latin America. Johns Hopkins University Press: Baltimore.
Smoke P. 2003. Decentralisation in Africa: goals, dimensions, myths and challenges. Public Administration and Development 23(1): 7–16.
Ssekamwa J, Lugumba S. 1971. Development and Administration of Education Uganda. Fountain Publishers: Kampala, Uganda.
Stasavage D. 2005. The role of democracy in Uganda’s move to universal primary education. Journal of Modern African Studies 43(1): 53–73.
Steiner S. 2006. An evaluation of the impact of decentralisation on poverty— the case of Uganda. Dissertation zur Erlangung des akademischen
Grades Doctor Philosophiae. Leipzig, unpublished.
Suzuki I. 2002. Parental participation and accountability in primary schools in Uganda. Compare 32(2): 243–259.
Uganda Participatory Poverty Assessment Project. 2002. Second Participatory Poverty Assessment Report.
USAID. 2005. Democracy and Governance Assessment: Republic of Uganda, 2005. ARD Inc.: Vermont, USA.
World Bank. 2002. Achieving EFA in Uganda: The Big Bang Approach. World Bank: Washington, DC.
World Bank. 2003. World Development Report 2004: Making Services Work for the Poor People. World Bank/Oxford University Press:
Washington, DC.
Wunsch J. 2000. Refounding the African state and local self-governance: the neglected foundation. Journal of Modern African Studies 38(3):
487–509.
Wunsch J. 2001. Decentralization, local governance and ‘recentralization’ in Africa. Public Administration and Development 21(4): 277–288.
ANNEX 1.
Key data on the sample of School Management Committees (SMC)
Total number of group interviews 17
Total number of individual interviews (questionnaires) 127
Statutory total of SMC members (incl. 1 headteacher as ex-officio member) 10
Average number of SMC members in sample 7.5
Mode of SMC members in sample 6
Ratio of nin sample: nin district (%) District 1 ¼9:73 (12%)
District 2 ¼4:416 (1%)
District 3 ¼4:112 (4%)
Copyright #2008 John Wiley & Sons, Ltd. Public Admin. Dev. 28, 149–164 (2008)
DOI: 10.1002/pad
164 G. PRINSEN AND K. TITECA