International Journal of Educational Development 66 (2019) 139–146

Contents lists available at ScienceDirect

International Journal of Educational Development

journal homepage: www.elsevier.com/locate/ijedudev

Ensuring quality education? Low-fee private schools and government T

regulation in three sub-Saharan African capitals
Joanna Härmä
Honorary Fellow, Centre of African Studies, University of Edinburgh, and Visiting Research Fellow, Centre for International Education, University of Sussex, United
Kingdom

A R T I C LE I N FO A B S T R A C T

Keywords: Low-fee private schools are currently relied upon by many low-income families, particularly in urban areas
Low-fee private schools under-served by government. A common response includes calls for effective regulation if they are to be allowed
Regulation to operate. This paper goes beyond the common theoretically-based perspectives and documents key stake-
Poverty holders’ experiences, and finds that Ministries of Education in three African capital cities are failing in their roles
Urban schools
as regulators of both government and private education. Unrealistic regulations and under-resourced in-
Sub-Saharan Africa
spectorates have resulted in patchy oversight systems of which rent-seeking has become an integral part, failing
most school-going children.

1. Introduction part of efforts to achieve equitable access to education for all, of an

acceptable quality? Are governments at the present time managing to
The debates surrounding low-fee private schools in less-developed regulate them meaningfully? What are the practical outcomes of the
countries have moved on from questioning whether they represent a existing regulatory systems?
significant development in international education, to whether these This paper makes a contribution to the literature by bringing to light
should be actively promoted and supported as part of a route to the perspectives and experiences of individuals involved in the opera-
achieving Sustainable Development Goal 4 on education. There are tion and regulation of private schools in three different sub-Saharan
some entrenched views on either side, but what most appear to take as a African capital cities. It draws on qualitative evidence gathered by the
given is that the thousands of small, individually-owned schools need to author, from a range of respondents regarding the regulation of private
be regulated in order to ensure that they meet minimum quality stan- schools in Abuja, Accra1 and Kampala. The sample is small, however
dards and treat their teachers and client families fairly. Governments these key players were able to provide clear insights that cannot be
are the duty-bearers of the human right to education, and so naturally gleaned from large-scale surveys. The paper responds to and builds on
government is the body that is called on to regulate schools. Baum, Cooper and Lusk-Stover’s recent paper (2018) using official
Yet the rise in low-fee private schooling has been driven by gov- documents and administrative data to look at the issue of regulation in
ernment failures in many aspects of their role in delivering/ensuring 19 African countries. It does this by providing insights that help to flesh
the right to education, including their role as regulators, perhaps most out their findings, and it also provides part of the picture that their
obviously with regard to their own publicly-funded and provided paper was lacking, providing proportions of unregistered private
schools. The existing research demonstrates that where government has schools found during school mapping exercises in sample areas of each
failed to keep pace with growing need for additional school places location. The next section sets out the background and the methodology
(access) and (often concurrently) failed to deliver education of an ac- for this paper, before moving on to city-specific sections. The final
ceptable quality, parents who can afford to pay have been giving up on sections discuss the issues raised, with brief conclusions to close.
government schools and sending their children to private schools in-
stead (Akaguri, 2014; Baum et al., 2018; Härmä, 2013; Srivastava, 2. Background
2006; Tooley, 2013; Tooley and Dixon, 2006). This situation begs the
questions then: can government effectively regulate private schools as Baum et al. (2018) reports on their World Bank SABER work

E-mail address: [email protected].

1
The low-income areas of Accra studied are peripheral to the main city and fall within the Greater Accra Region, however ‘Accra’ is used for short throughout the
paper.

https://doi.org/10.1016/j.ijedudev.2018.10.007
Received 18 February 2018; Received in revised form 7 September 2018; Accepted 29 October 2018
Available online 07 November 2018
0738-0593/ © 2018 Elsevier Ltd. All rights reserved.
J. Härmä International Journal of Educational Development 66 (2019) 139–146

assessing the policy environment regarding private schools in 19 dif- was the context, out of 21 studied in 19 countries, with the most
ferent Sub-Saharan African countries. The paper looks at the policies stringent set of input-focussed regulations.
and regulations governing the sector, with an especial focus on market These tough regulations have proven ineffective in every sense, and
entry. The authors find that regulation is ineffective in addressing the proprietors have repeatedly ranked dealing with government as one of
negative externalities of the market; that there is a lack of capacity to their key challenges, referring specifically to the difficulties attached to
enforce regulations and that widespread rent-seeking behaviour has becoming registered (Härmä and Siddhu, 2016 and Härmä and
resulted. Furthermore, it is found that while onerous regulations appear Adefisayo, 2013), and with navigating the multiple taxes and fees that
to have some impact on whether schools seek to become registered schools are asked to pay by a range of actors (Härmä and Siddhu,
schools, they do not seem to inhibit growth in unregistered, informal 2017).
sector schools. Parents’ ostensible ability to pay does not seem to play a This paper complements Baum et al.’s paper (2018) that used mostly
role, as they find GDP per capita to be unrelated to the size of the national-level data on the registered school sector, and national po-
private school sector. What is far more significant is whether there is licies, to examine these issues. The authors observe that the paper
sufficient space in the government school system for all children. Their captures the policies, but not how well they are applied, and ‘un-
paper sought to approach the subject systematically, rooting it in a fortunately, available data across countries only allows for tracking of
theoretical framework, noting that the understanding of regulations the size and growth of ‘recognised’ private schools’ (Baum et al., 2018,
and their impacts has been driven less by empirical observations than p.105). They did however also gather further evidence in Nigeria and
theoretical reasoning. Ghana, two of the three focus countries in this paper, finding under-
DFID’s ‘rigorous literature review’ of evidence on low-fee private implementation of the regulations in both cases. This paper therefore
schooling delved into the issue of private school regulation, summar- contributes to the literature and builds on this previous conceptual
ising evidence regarding certain assumptions, the first of which is that work, and charting of official policies and administrative data. The
‘states have the knowledge, capacity and legitimacy to implement ef- paper does this by documenting the accounts of a small sample of key
fective policy frameworks for collaboration and regulation of the pri- stakeholders regarding their efforts to navigate the registration system;
vate school sector (Day Ashley et al., 2014, p.35).’ Regarding this as- also the accounts of government education officials regarding how the
sumption, the report summarises: ‘there is consistent evidence across a system is administered and the challenges faced.
range of contexts that attempts by governments to intervene in the
private education sector are constrained by a lack of government ca- 3. Methodology
pacity, understanding and basic information on the size and nature of
the private sector. Attempts… to apply regulatory frameworks suffer The interviews reported in this paper came out of larger studies in
from poor implementation’ and furthermore, ‘…the legitimacy of in- each city. These studies were similar, however they were conducted
tervention has been damaged by past attempts to… extract rents discretely, and not with the prior intention of writing in a comparative
through regulation’ (Day Ashley et al., 2014, p.35). This conclusion is manner across study sites, or on the topic of regulations and their im-
echoed by Tooley and Dixon’s (2005) work in India, which found that plementation. In each case, the focus for the author during fieldwork
regulations on paper were very different to what was observed in the was on quality-assuring the survey data. The supplemental key in-
field, even at government schools, and that bribes were a common formant interviews were conducted purely to better understand the
feature of the landscape in a context of an under-resourced in- relationship between the government and the schools. In each case,
spectorate. Tooley (2009) catalogues in considerable detail the short- advice was sought from local school proprietors on whom to interview
comings of efforts to regulate private schools in a range of contexts. from the government and from a private school association. The pri-
A second assumption tested by the DFID review was that ‘state mary objective was to discuss the issues with a respondent with the
regulation is effective and improves the quality, equity and sustain- time and the willingness to discuss, not to interview a person in the
ability of private school provision’. Unfortunately, ‘the evidence in- same position in each country. The author was aware of potentially
dicates that where state regulation of private schools exists, it is not differing perspectives offered by different ‘levels’ of staff, but the right
necessarily effective or may be selectively enforced. This may relate to a ‘type’ of information was gleaned from district-level interviews in one
lack of sector knowledge and capacity by governments in some cases. country, and central, Ministry-level interviews in another.
However, there is also evidence that it relates to unrealistically strin- The key informant interviews were entirely qualitative in nature
gent regulation which may offer opportunities for rent-seeking and and developed organically, following only a rough list of questions;
bribery to bypass recognition requirements (Day Ashley et al., 2014, some of the conversations lasted several hours. The topics covered in-
p.36)’. These findings are entirely reinforced by Baum et al.’s findings. cluded the highly sensitive issue of buying one’s way out of being bound
DFID’s review suggests that ‘the key finding across a range of countries by the regulations, which is not something that can be included in a
is that there are severe difficulties with enforcing existing regulations’; structured, standardised interview schedule. This topic could only be
but ‘…there are also concerns about private sector provision being broached once rapport had been established, and several attempts to
promoted through collaboration without adequate regulation and discuss this issue were evaded by respondents in a way that made it
quality controls’ (Day Ashley et al., 2014, p.37). clear that they did not feel comfortable discussing it. For this reason,
Härmä (2011) notes the weaknesses of the regulatory environment there is no quantitative data on this issue, which might have expanded
in Lagos, finding that the overly stringent regulations have driven the the picture from the qualitative data.
vast majority of low-fee private schools to ‘fly under the radar’: in 2011, In Abuja, the author was taken straight to the Secretariat of the
76% of Lagos’s 12,098 private schools were unregistered. At the same Federal Capital Territory (FCT)’s government, to meet with the Director
time, 40% of schools surveyed in Makoko slum stated that the im- of the Department of Quality Assurance. The local official whom we
possibility of ever meeting regulations was the key reason for them asked to help us arrange the appropriate meeting brought us to the
intending to remain unregistered (Härmä and Adefisayo, 2013, p.146). Director; it was clear that in that case, only the Director was authorised
Baum et al. (2018) report similar responses in their survey of Ajeromi- to discuss the issue of regulation with us. In Accra, the author met with
Ifelodun in Lagos. In further research in Lagos in Autumn 2016, it was the Private Schools Coordinator at one of the Municipal Education
found that school numbers were growing, with 42% of surveyed schools Offices of one of our study areas, with the approval of his supervisor,
having been established in the period between the private school census the head of the Municipal Education Office. This officer was in charge
of 2011 and the survey of 2016; and the proportion of registered of handling the vast majority of a school’s registration process. In
schools had shrunk, from 24% in 2011 to 23% in 2016 (Härmä and Kampala, the author was directed first to the Assistant Commissioner
Siddhu, 2016). This despite the finding of Baum et al. (2018) that Lagos for Basic Education, and then to personnel within the Department for

140
J. Härmä International Journal of Educational Development 66 (2019) 139–146

Table 1
Key numbers – sampling frame, surveys and interviews completed.
Abuja Accra Kampala

Number of private schools found 491 534 438

*
Number of government schools found 92 26
Number of private schools surveyed 258 242 223
Registered 87 (34%) 201 (83%) 142 (64%)
Schools
Unregistered 171 (66%) 41 (17%**) 81 (36%)
Schools
Numbers of key informants interviewed 1 government & 1 association 1 government & 1 association 2 government & 1 association
Date of fieldwork January 2016 November 2017 June 2016

* Government schools were not listed during the Abuja study.

** Only 4% of the 242 schools in Accra were unregistered and not currently in the process of applying for registration.

Private Schools and Institutions, to obtain more detailed information on divided official administrative areas that would be reflected in official
the application process. In the latter case, useful information was government data on population and school enrolment. Because this
gleaned from two ‘levels’ of staff, while in Accra and Abuja all of the work was exploratory, it was not possible to know in advance how
necessary information was gleaned from just one individual. quickly it would progress, and how much physical ground could be
In each country a representative of the main private schools asso- covered. The design is to get a feel for more than one area of each city,
ciation was also interviewed. The relevant interview for Abuja is not so work did not simply continue until an entire named area of the city
reported below, because the respondent (the head of the organisation was covered – sometimes these areas are extremely large. In each in-
for the FCT region) was the proprietor of an elite boarding school, and stance, it was unknown how many schools would be found, and how
while the association caters to all shapes and sizes of private schools, many interviews could be conducted during the surveying portion of
the respondent did not speak to the issue of application of regulations to the work, as various contextual factors influence how many interviews
small, poor schools. Rather, she gave high-level, more political answers, can be carried out per day.
no doubt connected to her own school’s relationship with the state. The total number of listed private schools in each location served as
With regard to the survey, a professional research company was the sampling frame. What was the same across study sites was the
selected competitively and the teams of enumerators were trained by number of days listing schools, and the size of the team, which lends
the author. The instrument was piloted and necessary revisions were some rough comparability. A random sample of schools were surveyed
made. The fieldwork commenced with four days of school listing, in each city, with the samples being drawn from each of the research
during which selected study areas were combed to find and document neighbourhoods proportionally, depending on what share of the sam-
every private school found serving any level between pre-primary to pling frame each area made up. The survey instrument was a structured
upper secondary. Table 1 provides details of the numbers of schools questionnaire, while qualitative data from schools resulted from side
found, and shows that two-thirds of schools in Abuja are unregistered, discussions that developed spontaneously. All participants in the re-
while the proportions are lower for Kampala (36%) and Accra (17%). search gave informed consent.
The studies were motivated by a drive to understand the local low-
fee private schooling ecosystems, which are never homogenous and 4. Findings from three sub-Saharan African capitals
include schools charging a range of fees from the very lowest possible to
barely support the school, up to fee levels that only the less-poor living This main section of the paper presents the main findings, starting
in these poor or slum areas can pay for. There was no fee limit imposed with a summary of some of the key provisions across the three case
for this study, because the aim was to gain an understanding of what study areas. Table 2 provides information on just a few of the key
schools exist within a selected low-income area. The study areas were provisions, all of which could not be discussed in the paper due to space
the lowest-income areas of the study cities based on existing poverty constraints.
maps or other available poverty data, and were confirmed through the
local knowledge of the research teams in each location. In each case, 4.1. Abuja
neighbourhoods were selected from within larger areas of the cities,
meaning that establishing the population size and other statistics of Abuja provides an interesting case of a new city, built from scratch
precisely the areas listed and surveyed is not possible using available in the last thirty years or more (to become the national capital in 1991),
data. Survey areas were often bounded by natural features such as which did not implement its own ‘master plan’ to make the necessary
rivers, which effectively divided neighbourhoods, but may not have provision for important services such as education and transportation.

Table 2
Some key requirements as set out in guidelines for private schools.
Abuja FCT* Ghana Uganda

Owned land required No No No

Minimum land area, hectares 2 (not flexible) 0.81 (not flexible) 2.02 (flexible for urban schools)
Minimum classroom size 108 square metres 61.88 square metres 45.24 square metres
Minimum number of toilets 3 10 1 for every 40 students by sex
Sports field required Yes Yes Yes
Qualified / licensed teachers required Yes No Yes
Minimum additional requirements for head 5 years’ experience 5 years’ experience Grade V Teaching Certificate in primary education
teacher
Minimum required funds in place at inception 1 million Naira (US$5000 at the time of N/A 3 months’ teacher salaries & 10% costs of school
fieldwork) development plan

* The regulations in Abuja are specific to Abuja FCT, while in Ghana and Uganda, the guidelines apply nationally.

141
J. Härmä International Journal of Educational Development 66 (2019) 139–146

This has resulted in its very large private school sector that many would be closed. Both of these points were frankly and openly stated by
communities depend on. the respondent. He further explained that the insufficient supply of
government schools was due to plots designated in the FCT Master Plan
4.1.1. What the regulations say for schools, both government and private, having been 'taken over' and
In Abuja FCT, private school regulations are set out in the that now there was little prospect of these plots being used for their
Guidelines on the Establishment and Operation of Educational intended purpose. ‘It is difficult for the government to get land to build
Institutions in the Federal Capital Territory, Abuja (Government of schools on, or to grant to approved private providers’.
Abuja FCT, 2013). These guidelines come under the Educational (Na-
tional Minimum Standards and Establishment of Institutions) Act of 4.1.3. The view from schools
1985, and were approved by the Federal Ministry of Education in 2005. From this meeting, and from visiting registered schools in the field,
The guidelines call for ‘semi-urban’ schools, which category this study’s it became clear that regulations are more flexible in practice than of-
schools fall into, to have 2–3 hectares of land, to support: at least 3 ficials would be able to let on in such an interview. No registered school
classrooms of 108 square metres each; a ‘functional library’; a first aid visited in the sampled (low-income) areas had the required amount of
room; an administrative block of two large rooms and a store; a games land, an assembly hall or a sick bay with a trained nurse, and many had
field of 30 × 30 m; farmland and/or a fish farm; 3 toilets; provision for unqualified teachers. Only 23% of sample schools are registered and
water, electricity and fire-fighting equipment; two computers; and two have legally-enforceable land ownership, and formal leases are rare.
copies of the curriculum/syllabus. Schools are encouraged to have an The risk of closure is less than the risk of demolition, but both risks are
assembly hall with capacity to seat twice the school’s enrolment, greatest amongst the poorest schools, and it was the minimum land
however this is marked as ‘optional’. requirements that were most often cited by proprietors as the key
With regard to staffing, there must be at least three teachers, all of barrier stopping them from registering their schools. One proprietor
whom must be qualified and registered; one head teacher who must be stated ‘they said you have to have 5000 square metres of land. No
qualified and registered and have not less than five years’ teaching school is able to have that, maybe only a university2. But it is good to be
experience; one trained nurse; and two support staff. Schools must keep registered… I would like to be’. But in Abuja, it seems that the chal-
basic administrative records and should have available a site plan and lenging conditions mean avoiding calling attention to one’s school – the
either a Certificate of Occupancy (for owned land) or valid lease survey evidence showed that a quarter of respondents had never had
agreement (implying rented land is acceptable). Of note, whenever the any kind of government visit or inspection.
land requirement was raised in school proprietor interviews, it was with
the belief that the land must be owned – this was true too of Accra – and 4.2. Accra
yet in the guidelines it is stated that a valid lease agreement is accep-
table. Schools are required to keep financial records, and must have a The situation in Accra is unusual in that 83% of schools surveyed
minimum bank balance at inception of 1 million Naira (equivalent to were registered, while another 13% were in the process of applying,
US$5000 at the time of the fieldwork). The school must have no more despite many schools having just make-shift, temporary structures and
than three primary grades at inception, each with a minimum of 15 no qualified teachers.
students.
4.2.1. What the regulations say
4.1.2. The view from the government
In Accra, private schools are governed under the Education Act
The private school sector is regulated by the Federal Capital
2008 (Act 778), and there are two separate documents providing gui-
Territory (FCT) Education Secretariat through the Department of
dance and regulations, falling under this Act. Ghana Education Service
Quality Assurance (DQA). The author was able to meet with the
has formulated its ‘Guidelines for the Establishment of Private Schools’
Director of the DQA to discuss the regulation of the private school
(Ghana Education Service(undated), 2018) which sets out the most
sector. For a school to obtain the complete Guidelines on the
detailed guidance, which includes that proprietors must seek approval
Establishment and Operation of Educational Institutions in the Federal
by government of the intended site two years before establishment; the
Capital Territory, Abuja, a fee of 30,000 Naira (US$150) must be paid.
site should be at least 2 acres (0.81 ha); it should not be on unsuitable
The situation of land and its development is complicated by the gov-
land (for example, water-logged or close to refuse dumps); but also
ernment’s ‘master plan’ for the continuing development of the FCT,
schools must be 3 km distant from each other. Further (and confus-
which entails not giving secure title to land, so that it can later be taken
ingly), ‘a school should not be in rented premises. A school could be in
by the government for planned development. Most people live in pre-
rented premises meant for the purpose of a school with legal document’
carity, particularly in informally developed settlements, and many
(Ghana Education Service(undated), 2018, p.1) which seems to indicate
school proprietors in this study reported that they have not invested in
a preference for owned land, but that a lease with a legal document
their buildings due to this uncertainty and the constant threat of de-
would suffice. There must also be a playing field (of unspecified size).
molition.
With regard to classrooms, these must be ‘enough’ in number, and
The respondent was asked about the amount of land required, and
must be 61.88 square metres each. There must also be a head office,
whether in practice the department is more flexible in its approach, as
staff room, library and store; 10 toilets (2 of which for staff) and 6
there appear to be many registered schools that do not have anywhere
urinals. There should also be suitable classroom furniture and teaching-
near the stated minimum. The reply was that in central Abuja they are
learning materials; a first aid box; arrangements for safety and security;
flexible due to the high cost of land, but that in outlying areas schools
drinking water and water for hand washing.
are expected to have more land, despite the fact that schools in outlying
The Guidelines state that ‘there should be enough teachers for the
areas often cater to those living in poor, informal settlements (or
start of a school’, and these teachers must have a minimum qualifica-
slums).
tion of 5 passes at the lower-secondary or ‘O-level’, but unqualified
However, schools in these peripheral areas were acknowledged by
teachers should receive in-service training after recruitment. District
the Director, and their de facto situation discussed. He stated that they
Education Directorates are to organise in-service training for these
had the legal authority at that time to close down some 550 private
teachers, with the costs borne by the school. The head teacher must be a
schools. Several considerations were stopping them from carrying out
these closures: firstly, the department simply lacks the resources to
carry out the closures and to ensure that schools stay closed. Secondly, 2
This is equivalent to only 0.5 hectare, which is in fact only one quarter of the
there is no provision for the education of the children whose schools land required.

142
J. Härmä International Journal of Educational Development 66 (2019) 139–146

professional (presumed to mean qualified) teacher who has taught for usually just starting out and will come forward for registration. When
not less than 5 years (Ghana Education Service(undated), 2018, p.2). asked why the government had not kept pace with the growing need for
The Guidelines also state that with initial registration, schools can school places, he said that ‘individuals have taken over all the land in
operate for three years, but then must apply for a permanent certificate the area. It [the area] has expanded in the last 10 years, boomed ac-
of registration. It is mandatory for private schools to apply the same tually. Government now has no access to land, so they can’t build any
code of ethics as is used in government schools. Lastly, the document more schools’. As in other countries, it was reported by respondents that
states that the GES has the authority to close down unapproved schools; government officials do not spend much time inspecting government
schools operating in an unhealthy environment; and schools failing to schools, and instead focus on private schools. In the case of Accra, it
abide by regulations as well as the Code of Conduct of the GES. District was said that ‘government school teachers are trained, so they don’t
Education Offices are responsible for monitoring private schools, and need the same support as private school teachers do’.
are meant to write quarterly reports to GES regarding their monitoring
activities. 4.2.3. The view from an association
The second document, ‘Education (Private Schools) Regulations One of the zonal representatives of the main association began by
2000 Arrangement of Regulations’ (Government of Ghana, 2000) sets explaining that the association was formed to protect private schools
out further guidelines, including that pupil-teacher ratios must not from certain policies of the government. They also provide information
exceed 45:1 at the primary level, and 40:1 at secondary levels. It further and peer learning amongst schools. The association is officially sanc-
states that ‘overall fees chargeable in a private school shall be regulated tioned by the government, making the relationship unique amongst the
by the Ghana Education Service Council in consultation with GNAPS study cities: all schools coming forward for registration are meant to
[Ghana National Association of Private Schools]…’ and that ‘no pro- have become members of the association, although many have not. The
prietor of a private school shall determine, charge or revise the level of intention is that the association is the conduit between government and
school fees without applying to the Ghana Education Service Council the private schools in terms of two-way information flows.
and receiving a written appraisal by the Council’ (Government of When asked about the relationship between government and private
Ghana, 2000, p.7). Every school must have a board of governors; and schools, the respondent stated that ‘the government has a negative at-
proprietors ‘shall be responsible for the procurement and supply of titude towards private schools. They see them as a business, and that
adequate library books, equipment, stationery and other materials re- they are earning a lot of money. They charge money for any trainings
quired for use by pupils in attendance at the school’ (Government of they invite private schools to. The government personnel take personal
Ghana, 2000, p. 10). Lastly, schools must keep proper financial records payments to give trainings… and they don’t provide information on the
and submit these annually to the school’s board of governors. various policies and changes made to them, that private school pro-
prietors are meant to know… They just want to exploit private schools’
– which they do partly by taxing them like other businesses. When
4.2.2. The view from the government
asked if the government has ever closed any schools down, she reported
The approach in Accra appears to be much more flexible in practice
that they do not engage in this, but that poor quality private schools are
than in other countries. The Private Schools Coordinator at one of the
able to continue to exist solely because government officials are willing
Municipal Education Offices described the usual pattern for schools:
to accept bribes. ‘Every time a government official comes to the school,
they tend to start operating, try to attract some students and then they
you are meant to “tip” him’.
apply to become registered. This initial start (based on the survey evi-
The respondent also reported that corruption extends to other areas
dence) can be in even the most makeshift structure with no qualified
of education governance – she said that ‘not even the high-stakes ex-
teachers.
aminations like the Basic Education Certificate Exam truly mean any-
When asked what are the main requirements that the government
thing, because there are government officials who will take bribes just
wants to see in any school, the official interviewed stated that ‘venti-
to place a student into the senior high school of their choice – this costs
lation is key, and there must be toilets and drinking water. There must
2000 Cedis (US$441.50).’ She reported also that government schools
be seats and a place to write. There must be enough teachers and
use actual examination questions in class time to prepare students to do
teaching-learning materials. The teachers must at least have a senior
well, and that ‘private schools can buy the examination questions too.
high school qualification, but there are expected to be two or three
Some schools get good exam results this way, but they totally ignore
certified teachers to help the uncertified ones3. Human resources have
curriculum areas that are not part of the exam. Then a student gets a
not really been a focus in this country, so we can’t expect all the tea-
good score, gets to senior high school, and struggles because they don’t
chers to be qualified. We are here to serve the private schools… there is
actually have any foundation’. Regarding textbooks and materials, it
no antagonism… they are our clients and we are here to help them.’
was reported that the government provides these to ‘just a few schools,
A striking feature of private school regulation in Ghana was the
and nowhere near enough books, even then… they give to their “fa-
frequency of government inspection visits to registered and un-
vourite schools”’. These ‘favourites’ are those that pay larger bribes (or
registered schools alike. Fully 67% of schools reported having been
‘gifts’), donating to the zonal education office furniture, air condi-
inspected in 2017, and a further 15% in 2016. Only thirteen schools
tioners, notice boards and other items. The respondent reported that it
(5%) reported their last inspection as being in 2015 or earlier. The
is not difficult to become registered; if schools don’t have the facilities
back-and-forth of inspections and requests for improvements continues
required, ‘people pay their way out’, even those schools with makeshift
even after schools are registered, with officials asking schools to make
wooden structures. ‘When schools do not meet any of the regulations,
incremental improvements.
the government officials just “take an envelope” and leave them alone’.
The respondent was asked about the relationship between the
government and the private schools, and stated: ‘we view it as a part-
4.2.4. The view from schools
nership. However private schools are businesses set up to make money,
Many respondents reported that the government’s frequent inspec-
so the government does not give them full support. The taxes private
tion visits are not without ulterior motives – it was repeatedly stated
schools must pay are few… They pay for annual renewal of their GES
that inspectors (both at the time of applying for registration and during
registration, and they pay all normal taxes that apply to businesses.’ The
routine inspections) take bribes to overlook serious deficiencies and
respondent also stated that unregistered schools are few, and they are
that no inspection visits take place without money changing hands.
Indeed, more than one respondent at a very poor school reported that
3
The regulations state that teachers must only have completed lower-sec- they did not come so frequently as to other schools, because the pro-
ondary school; there is no requirement for any teachers to be qualified. prietor could only afford to pay a small amount, so officials dedicate

143
J. Härmä International Journal of Educational Development 66 (2019) 139–146

their time elsewhere. Schools that in no way met the requirements for conforming entirely with the requirements’. These schools receive ad-
registration reported that they were able to get through the registration vice on how to improve, and are instructed to make the necessary
process via bribery. One respondent reported having had to move a changes.
fully functional toilet block in order to satisfy inspectors, but that this It was asked why there are so many unregistered schools in
expense was not enough, the officials were to be paid in addition. Kampala, and the response was that many simply cannot afford to put
in place all of the requirements of a registered school, for example
4.3. Kampala providing sports facilities (‘enough land for… physical education’ is
mentioned in the Guidelines, p.12). He also stated candidly that while
Kampala stands out for having up to 84% of all school enrolments department staff are meant to go out and inspect schools, they lack the
(pre-primary to upper-secondary) in the private sector (Härmä et al., person-power, vehicles and resources to be able to do so. The author
2017), and a relatively high proportion of registered schools (64%). also met with officers who described the process of becoming a regis-
tered school. She noted that the department assesses a school’s land
4.3.1. What the regulations say requirement in a flexible manner, and that schools in rural and urban
In Uganda, the Education (Pre-primary, Primary and Post-primary) areas are also assessed differently, meaning that urban schools could
Act 2008, Section 3(2b) empowers the Minister responsible for educa- register with smaller land areas.
tion to create regulations for all schools, including private schools, and It was asked if any private schools get closed down due to poor
the ‘Guidelines for Establishment, Licensing, Registration and standards, and the response was that it happens, but rarely, and that
Classification of Private Schools/Institutions in Uganda’ (Government they first work with the school, asking them to come up with an im-
of Uganda, 2014) is the result. These regulations set out responsibilities provement plan to help the school reach the minimum standard. They
of various organisations, including the Directorate of Education Stan- are given a time limit, but inevitably the inspectors tend to come back
dards (DES) and the District/Municipal Inspectorates who are jointly much later than was initially intended, due to the resource constraints
responsible for ensuring that minimum standards are kept, and for already mentioned, meaning that schools always have extra time to
providing support and supervision to private schools. comply. They give many chances and encourage proprietors to at least
The requirements for running a primary school include: a qualified arrive at a well thought-out plan, even if they do not have the capital at
head teacher; all teachers and head teacher to be registered and/or li- the time to institute it. It is only after repeated visits with no evidence of
censed4 with the Ministry of Education and Sports; administrative and intention to change, that the department will eventually close a school.
financial records to be kept; a school bank account; functional School The respondent indicated that while they do their best, they, as reg-
Management Committee; a minimum of one staff house on, or near the ulators, are not always able to act in an entirely streamlined and effi-
school compound; at least 5 acres of land (2.02 has); classrooms of cient way, pointing once again to the issue of the resource constraints
45.24 square metres minimum; security arrangements; first aid kit; that plague such regulators.
maximum of 55 students per teacher; ‘appropriate sitting facilities’; safe
drinking water; one toilet for every 40 students, by sex. At the time of 4.3.3. The view from an association
applying, schools must be able to show that they have funds for three The impressions formed were reinforced through an interview with
months’ worth of teacher salaries, as well as 10% of the funds needed to a zonal representative of a major association. The respondent was asked
implement a school development plan (that must also be in existence). if the association needs to take any action to protect private schools
Land ownership is not required and the 2.02 ha minimum is flexible from government action against them, and he reported that it is ex-
for urban schools such as those covered in this research, but proof of a tremely rare for the government to take strong action against schools
lease lasting not less than five years for a secondary school and eight and that he could recall only 2 school closures in the last ten years. He
years for a primary school are required. Schools are meant to apply for reported however that ‘the government is very unfriendly to private
permission before taking any other action. Once permission is granted, schools’ due to various policies such as making all schools pay similar
they are allowed to build the school and apply for a two-year license to fees and taxes irrespective of size and income. These official fees in-
operate. After 18 months of operation, the school must then apply for clude: registration renewal, local service tax, sign board fees, van tax
registration. Registration is granted for five years, after which time the (even if there is no school van), and property rates.
school must be re-inspected by the DES for renewal of the registration. He pointed out that in his area the government itself runs just 5
schools charging 80,000–100,000 Shillings (US$23.70-US$29.63) per
4.3.2. The view from the government term, which is approaching the costs of low-fee private schools. ‘There
In Kampala the main Ministry of Education and Sports respondent are around 20 [private] schools for every government school,5 because
was the Assistant Commissioner for Basic Education, and our initial parents do not want to use government schools, even though they have
question was ‘why are there so few government schools in Kampala?’ better teachers, they do not achieve anything, and parents are now very
The reason for the lack of ongoing school building was that ‘there are results-oriented and want to see learning going on. Parents are now
historical issues to do with land ownership which have caused gov- very desperate to access private schools and come to bargain on the
ernment schools to be few and far between. By the time the government fees, asking for reductions and suggesting what they can pay. It seems
realised they needed more schools, all the land was taken over. They to be a government policy not to bother with building any new gov-
didn’t wake up to the need until much too late, and purchasing land is ernment schools; they don’t dedicate the necessary level of financing to
difficult in the private sector.’ Government respondents in all three education’.
locations gave nearly identical answers to this question. With regard to the registration process, the association respondent
The respondent was asked what the relationship is like with the reported that schools must have enough land but that there is no
private schools? The response was that they collaborate. ‘We view the minimum set amount, and it can be rented as well as owned, which
private sector positively, as they bridge the gap in provision that we agreed with the Ministry’s responses. But he reported that there must be
know exists – but there are basic requirements and minimum standards a ‘permanent’ building (and many lack this); there must be toilets (most
that must be observed, although we know that some schools are not schools had some). There must also be enough government-registered
and certified teachers, and the head teacher must have a diploma in
4
It is assumed that this entails being a qualified teacher, but this is not made
5
clear in the document, while it is specifically mentioned for head teachers that The Ministry respondent had estimated a ratio of 3:1 of private to govern-
they must be qualified teachers and registered and/or licensed. ment schools; our fieldwork found a ratio of 27:1.

144
J. Härmä International Journal of Educational Development 66 (2019) 139–146

education. He stated that ‘only poorer schools opt for untrained tea- low-fee private schools serve low-income areas, they remain in-
chers’. He reported that the registration process is difficult, and that accessible to the poorest quintile of households, if not most of the
older schools are likely to be registered while newer ones are much less poorest two quintiles (this issue is explored in other literature, in-
likely to be, due in part to the long time the process takes – ‘but the cluding DFID’s rigorous literature review). Perhaps most importantly,
process becomes easier if you are willing to pay bribes’. He also stated better government schools would serve as the minimum bar against
that ‘once a school has been operating for at least five years, and can which private schools have to measure themselves, forcing them to up
show that all of the necessary elements are in place; the infrastructure is their game along with the public sector, or lose clients, as we are cur-
good and there are no problems with the land, et cetera, then you be- rently witnessing in Rwanda (The East African, 2017). As inspectorate
come a registered schools and get a registration number’. or quality assurance departments build in effectiveness and size, the
private sector should progressively be brought into such oversight
4.3.4. The view from schools systems.
The ambivalence of staff in the Ministry was further reinforced Currently, both ‘sides’ of the debate fall into unrealistically outlining
during the main school survey, particularly with one respondent at a functions to be performed by the state that governments demonstrably
mission-run slum school, who reported regular requests for bribes: lack the political will, transparency and resources needed to implement.
‘Inspectors will come and bother us about the smallest issues, things Roles for government are envisaged that call for transparent and by-the-
that are completely insignificant compared with issues with other book state action, be it in enforcing minimum standards for private
schools. They visit us every month, when there are schools just down schools; devising and effectively implementing contracting-out ar-
the road with much smaller compounds and much worse facilities, but rangements through public-private partnerships; or bypassing the pri-
they are registered schools and I know they manage this through paying vate sector all together through providing good government education.
bribes… things progress much more quickly if bribes are paid.’ He also Even while recognising and documenting the failures of government
indicated that inspectors neglect government schools where bribes regulatory systems, Tooley (2009) calls for targeted vouchers to allow
cannot be asked for. the poor to access private schools, despite the well-known risks of elite
This challenge is faced by individually-owned schools as well as capture and corruption. As Levin (2000) points out, the administering
mission-run schools. One school operating in poor conditions surprised of a voucher scheme requires transparent, advanced and effective ad-
the author by reporting its registered school status. The building was ministrative systems – similar to what a good government school
makeshift, constructed of a timber frame with horizontal wooden system requires. Corruption is arguably the central concern, and the
planks nailed to it, providing partial ‘walls’. The proprietor stated that central challenge to mitigate before either government schooling sys-
yes, the regulators were ‘flexible’ in terms of allowing them to become tems on one hand, or government regulation systems on the other, will
registered with such a building, but that this was achieved via the work properly.
payment of an informal fee, and that such payments are made every In the three case study locations, the sheer numbers of private
time a government representative comes to inspect. They also ask for schools present a major challenge for regulators, so what can be done to
improvements to be made, and expect eventually that the makeshift improve quality at these schools? With so few resources to go around,
building should give way to something more substantial. The experi- and with many school owners lacking specific training in school man-
ences of these two schools were similar, while the second did not face agement, anecdotal evidence suggests that tools to guide proprietors in
any accounting constraint on its ability to pay the officials the ‘fees’ school self-improvement are able to stimulate positive changes in
demanded. schools. The UK’s Department for International Development has been
running a programme using the ‘making markets work for the poor’
5. Discussion approach (Developing Effective Private Education in Nigeria or
DEEPEN, slated to wrap up in late 2018) which found that schools
The evidence from these city case studies indicates that regulation filling out a self-assessment form stimulated reflection on internal
of all schools is not working. The paying of bribes to government offi- weaknesses, and when re-visited, proprietors were found to have made
cials is an integral part of the private school-government relationship, positive changes in these areas, unsupported from outside6.
leading inspectorate staff to neglect government schools. As highlighted This is also the approach taken by the microfinance institution,
in DFID’s rigorous literature review and in each of the three study cities, Opportunity International, which encourages the formation of ‘Self-
government inspectorates are under-staffed, under-resourced and Improving School Systems’, which are small clusters of schools that
under-funded. Staff numbers are simply too small to meaningfully in- exchange insights and experience to help each other achieve the goal of
teract with the large numbers of schools under their jurisdiction. The better quality education (Opportunity International, 2018). Individual
vehicles available to them for school visits are woefully few, and there schools assess their own quality against a set of criteria based on best
are often no funds for fuel. Unpaved roads and heavy rain hamper ef- practices as derived from the work of UNICEF and the Brookings In-
forts to reach remote, rural schools during wet seasons. These genuine stitution. As part of Opportunity International’s lending programme,
challenges in carrying out their duties, mentioned repeatedly during this self-assessment is compared with a third party assessment of the
interviews, leads to demotivation and focusing on low-hanging fruit; it school to come up with an agreed school improvement plan, similarly
is all too easy to fall into step with this status quo. to what was hoped would take place in Lagos with government quality
Taking the current reality as the starting point, this paper calls into assurance staff acting as the third party assessor. In the latter case, the
question the utility of calls for private schools to be more strictly government counterparts back-tracked on their agreement to take this
regulated or even curtailed by national governments. It also highlights system forward, and in the vacuum left, as already noted, schools got on
the unworkability of calls for private schools to be actively encouraged with making improvements on their own.
to play an even larger role under close government supervision, as is so This paper, building on Baum et al. (2018) and all of the research
commonly suggested by pro-marketeers. Governments in the countries drawn on by DFID in its rigorous literature review (Day Ashley et al.,
most affected by the rise in low-fee private schooling simply do not 2014) and more, suggests that functional government regulation and
dedicate the resources needed to either curtail the sector, or to support oversight of thousands of private schools is a long way off. In the
its expansion in a controlled way. meantime, those working with governments in areas with large
As regulatory regimes become more effective, a more pro-poor use
of these resources would be to support and quality-assure government
schools, which are often deeply in need of improvement, and which 6
This information was gained through ongoing personal communication with
serve the greatest numbers of poor and marginalised children. While DEEPEN staff during 2016-2017.

145
J. Härmä International Journal of Educational Development 66 (2019) 139–146

numbers of low-fee private schools could encourage a softening of at- reviewers who helped to drastically improve this paper.
titudes towards low-fee private schools, as part of a fostering of school
improvement culture for all schools. The old binary classification of References
registered and unregistered (legal and illegal) schools has not led to any
positive outcome for children in schools, as parents continue to use Akaguri, L., 2014. Fee-free public or low-fee private basic education in rural Ghana: how
unregistered schools for reasons set out at length in other literature on does the cost influence the choice of the poor? Comp. A J. Comp. Educ. 44 (2),
140–161.
the topic. Effectively ‘outlawing’ schools, along with rent-seeking, leads Baum, D., Cooper, R., Lusk-Stover, O., 2018. Regulating market entry of low-cost private
to schools evading any interaction at all with government. schools in Sub-Saharan Africa: towards a theory of private education regulation. Int.
Those working directly with schools, such as foundations and fi- J. Educ. Dev. 60, 100–112.
Day Ashley, L., Mcloughlin, C., Aslam, M., Engel, J., Wales, J., Rawal, S., Batley, R.,
nancial institutions, could encourage the use of school self-improve- Kingdon, G., Nicolai, S., Rose, P., 2014. The Role and Impact of Private Schools in
ment tools, which could be disseminated by private school associations. Developing Countries. University of Birmingham, Institute of Education, Overseas
The author’s fieldwork in many countries has already found that some Development Institute, UK Aid, London.
The East African, 2017. Improved Public Schools in Rwanda See Private Academies Close
schools do come together in quality-assurance activities; for example in Down. Available online at:. (Accessed 14th June 2018). http://www.
many cases schools run community examinations, to keep up pressure theeastafrican.co.ke/rwanda/Business/Improved-public-schools-see-private-
on standards, and to better-inform parents. academies-close-down/1433224-4064492-dlujesz/index.html.
Ghana Education Service(undated), 2018. Guidelines for the Establishment of Private
Schools. Accra, Ghana Education Service.
6. Conclusion Government of Abuja FCT, 2013. Guidelines on the Establishment and Operation of
Educational Institutions in the Federal Capital Territory. Abuja Federal Capital
A realistic assessment of the current situation suggests that the state Territory, Department of Policy and Implementation, Education Secretariat.
Government of Ghana, 2000. Education (Private Schools) Regulations 2000 Arrangement
will not, in the short- to medium-term provide a solution to issues of Regulations. Accra, Government of Uganda.
surrounding private schooling, when the very schools in question arose Government of Uganda, 2014. Guidelines for Establishment, Licensing, Registration and
in response to state failure. It should be recognised that in responding to Classification of Private Schools/Institutions in Uganda. Ministry of Education and
Sports, Kampala.
insufficient supply and/or quality of government schooling, low-fee Härmä, J., 2013. Access or quality? Why do families living in slums choose low-cost
private schools are filling a very real need, serving those that can afford private schools in Lagos, Nigeria? Oxf. Rev. Educ. 39 (4), 548–566.
them, so condemning them en masse is futile and counter-productive. Härmä, J., 2011. Lagos Private School Census 2010/2011 Report. DfID-ESSPIN Report
LG501.
Where macro plans for addressing school quality (in this case, through Härmä, J., Adefisayo, F., 2013. Scaling up: challenges facing low-fee private schools in
regulation) are not working, spontaneous local efforts that are already the slums of Lagos, Nigeria. In: Srivastava, P. (Ed.), Low-Fee Private Schooling:
happening may provide the start of a way forward, drawing on local Aggravating Equity or Mitigating Disadvantage? Symposium Books, Oxford.
Härmä, J., With Pikholz, P., Hinton, P., 2017. Low Fee Private Schools in Low-Income
people’s natural drive for the best possible education for their com- Districts of Kampala. CapitalPlus Exchange, FSD Uganda, UK Aid, Uganda. Chicago.
munities. Härmä, J., Siddhu, G., 2017. Sensitising Proprietors to Deal With Multiple Taxation.
DFID-DEEPEN Report, Lagos.
Härmä, J., Siddhu, G., 2016. Parental Fee Default: Extent, Determinants and Implications.
Competing interest statement
DFID-DEEPEN Report, Lagos.
Levin, H., 2000. The Public-Private Nexus in Education, Occasional Paper no. 1 National
I confirm that I have no competing interests to report. Center for the Study of Privatization in Education. Teachers College, Columbia
University March 2000.
Opportunity International, 2018. What We Do: Education Quality. Available online at:.
Funding (Accessed 14th June 2018). https://edufinance.org/what-we-do/education-quality.
Srivastava, P., 2006. Private schooling and mental models about girls’ schooling in India.
This work was conducted for and funded by Capital Plus Exchange, Comp. A J. Comp. Educ. 36 (4), 497–514.
Tooley, J., 2013. Challenging educational injustice: ‘Grassroots’ privatisation in South
a non-profit organisation based in Chicago. Asia and sub-Saharan Africa. Oxf. Rev. Educ. 39 (4), 446–463.
Tooley, J., 2009. The Beautiful Tree. The Cato Institute, Washington, DC.
Acknowledgements Tooley, J., Dixon, P., 2006. ‘De facto’ privatisation of eduation and the poor: implications
of a study from sub-Saharan Africa and India. Comp. A J. Comp. Educ. 36 (4),
443–462.
Thanks are due to my long-term fellow researcher, Abiola M. Lawal, Tooley, J., Dixon, P., 2005. An inspector calls: the regulation of ‘budget’ private schools in
our valiant and dedicated research teams, and to the anonymous peer Hyderabad, Andhra Pradesh, India. Int. J. Educ. Dev. 25, 269–285 2005.

146