We’re building a new UNICEF.org.
As we swap out old for new, pages will be in transition. Thanks for your patience – please keep coming back to see the improvements.

Evaluation database

Evaluation report

2011 South Africa: Tracking Public Expenditure and Assessing Service Quality in Early Childhood Development in South Africa

Executive summary


“With the aim to continuously improve transparency and use of evaluation, UNICEF Evaluation Office manages the "Global Evaluation Reports Oversight System". Within this system, an external independent company reviews and rates all evaluation reports. Please ensure that you check the quality of this evaluation report, whether it is “Outstanding”, “Good”, “Almost Satisfactory” or “Unsatisfactory” before using it. You will find the link to the quality rating below, labelled as ‘Part 2’ of the report.”


The South African government has invested much in Early Childhood Development (ECD) in the last few years and enrolment in ECD programmes has increased rapidly. Government support for ECD has taken mainly two forms: (i) expansion of Grade R mainly in public schools, funded by the Department of Education (DOE); and (ii) subsidies by the Department of Social Development (DSD) to private community-based ECD facilities serving mainly children too young for Grade R. It has now become important to take stock of progress, to assess the efficiency and equity of public expenditures in the sector and to evaluate the quality of services offered to children.


This study was a first attempt to come to grips with the magnitude, scope, quality, and resources in ECD, and particularly at understanding whether public resources reach the intended beneficiaries (children) and are used effectively to achieve the intended objective, viz. the early development of children so as to provide an improved foundation for learning once these children enter the formal school system. It also allowed an analysis of the fiscal incidence of spending on ECD, i.e. whether public spending on ECD was reaching those most in need of it.


UNICEF, the Department of Social Development, the Department of Education and the National Treasury commissioned the survey reported on here in ECD facilities in three provinces. Province 1 is a richer province, Province 1 a moderately poor one and Province 3 a large and very poor province, thus the survey reflects the broad spectrum of experiences in ECD. Given the multiplicity of institutional contexts, funding structures and oversight mechanisms in different parts of the ECD sector, the survey comprised three separate samples, each with an own different questionnaire, focused on Grade R in public schools, registered community based ECD facilities and unregistered community based ECD centres. It can thus in essence be seen as three linked surveys. Questionnaires were designed by a team from the University of Stellenbosch, led by the principal author in cooperation with the technical steering committee and the survey team. As has now become common, the facility survey contained elements of both a Quality of Service Delivery Survey (QSDS) and a Public Expenditure Tracking Survey (PETS), so as to track and measure resources, their use and transfers at different levels of the administrative structure, and to allow an analysis of the types and quality of services offered. The survey was fielded by Citizens Surveys, a data collection company and fiscal data were obtained from the three provinces and from National Treasury by UNICEF.

Findings and Conclusions:

Broadly speaking the survey presents a relatively encouraging picture of the sector. Overall quality of services appears to be moderate, as reflected in teacher-pupil ratios, training and experience of staff members, planning of classroom activities and programme quality. However, several ECD facilities have limited space and poor infrastructure, they receive inadequate community support, there are issues around adequacy of nutrition, and few facilities put enough effort into development of children.

Unregistered facilities more often provide low quality services and often have inadequate infrastructure and unsafe classrooms. Reasons provided for not applying for registration were that they had just started recently, that they did not qualify (usually due to poor infrastructure) or even that they did not know about the grants and registration. The picture that emerges is of young, small and struggling organisations. If this was largely a refuge for money-hungry entrepreneurs one would expect high pupil-teacher ratios and high principal salaries, but the opposite is the case. Teacher-pupil ratios hover in the mid-teens, lower than for registered community facilities and public schools. The growth of such unregistered ECD facilities is inhibited by poor infrastructure and a lack of adequate demand for such services by poor people who cannot afford unsubsidised services.

The flow of funds to registered community based ECD facilities has expanded greatly in recent years. Furthermore, there are limited controls in place by the DSD to ensure appropriate behaviour by service providers, and weak bookkeeping is endemic in ECD organisations. In such circumstances, one would expect large-scale opportunistic behaviour. Variables to capture opportunistic behaviour were constructed and their association with measures of programme quality tested. Such measures included high absentee rates (which could imply manipulated enrolment rates), excessive schools fees, “excess” principal salary (40% higher than their opportunity cost), “excess” income (income exceeding expenditure),
and “missing” information on grant receipt. Measures were also created for competent management and transparency, such as the presence of a petty cash book and completed child progress reports, and an index of self help.

There were some significant correlations and overlaps between these markers, but correlations were generally low and overlaps were not much greater than would have been expected in the absence of correlations. It was not possible confidently to identify any cases of opportunistic behaviour or dishonest practices. It is probably best to attribute the patterns observed to low signal to noise ratios in the markers and thus to remain somewhat agnostic about these categorisations. There are some weak indications of possible opportunistic behaviour, but more investigation and preferably site audits would be needed to confirm this. Given the informal nature of transactions in this sphere, even an audit may miss some forms of opportunism, such as siphoning off of community funds that are paid in cash without any paper trail (e.g. receipts to parents or community donors). The survey, though, provides little direct evidence of large scale and systematic financial abuse of the subsidy system.

There are at least two possible reasons for the lack of correlation between poor financial management or possible dishonesty and service quality. It could be that providing a basic good quality service is not expensive and that the premium relating to delivering a quality service is thus relatively low. Alternatively, providing a very poor quality service could be observed by parents and they may have sufficient choices available to take their child to another facility. Further studies, perhaps including focus groups with parents, may yield more on this.

Four possible explanations are offered for the relatively muted evidence of misbehaviour in this sub-sector, given the apparent opportunities for it: (i) Due to the relatively recent development of this sector, some loopholes in the system may not yet have been detected and exploited. (ii) Accountability to fee-paying parents puts pressure on service providers to provide a service perceived by parents to be of good quality and at relatively low cost. (iii) Funds flowing to these facilities are relatively meagre, considering the cost of providing the service. Additional funds raised by school fees or other means would give more scope, but also brings involvement of parents with an interest in keeping fees low while demanding a good service. (iv) Because recipients of DSD funds need such funds to keep their concerns going and they usually know what subsidies they qualify for, DSD is under greater pressure to disburse funds. This rule based situation and the dependence of private organisations on the public funds is quite different from many other PETS where public expenditure is tracked from higher administrative levels to public schools.

The small margins, private service providers, relative transparency in the value of the transfer, and accountability to parents thus all play a role in keeping the system on track. South Africa’s rule-based direct funding of personnel in schools, including Grade R, mitigates against diversion of funds, whilst the rule-based subsidies to registered community based facilities may have successfully kept diversion of funds in check.

There are nevertheless possibilities of abuse and its extent is likely to grow over time. It is important to retain the strengths of the present system (e.g. accountability to parents) and to put mechanisms in place to improve the present system as it grows. It is vital to tighten and improve regulatory controls before the loopholes in the system are widely exploited by opportunists within ECD facilities or even within the DSD itself.

Also, there is clearly room for improvement in service delivery, given the demonstrated importance of ECD in providing a solid foundation for children’s education. ECD principals at community-based facilities cite lack of resources and specifically insufficient classrooms, play grounds and toilet facilities, issues surrounding the security of children, lack of funding, shortages of qualified teachers and training, the lack of involvement of parents and the poor and disadvantaged backgrounds of children as problems. Many of these items are most appropriately and effectively provided at the local level and ideally by the municipality.


The final chapter provides 27 concrete recommendations based on the survey and other information about policy options and issues. Most recommendations are directed at the DSD. Where the DoE is also involved, an asterisk (*) after the recommendation number will indicate this.

Full report in PDF

PDF files require Acrobat Reader.



Report information

New enhanced search