Editor’s Note: This blog is one of a series on early childhood development, featuring experts from Brookings and elsewhere that have been discussing the topic as part of work being conducted by the Center for Universal Education.
When it comes to early childhood development (ECD), we know quite a bit about how different programs impact young children. However, we know much less about their cost and hence cost-effectiveness. Yet having cost information is crucial: the evidence on impacts is compelling, but policymakers need to know what it will cost to reach more children with ECD programs that make a difference in their lives.
On December 2-3, the Brookings Center for Universal Education joined the Inter-American Development Bank (ADB), the International Food Policy Research Institute (IFPRI), the United Nations Children’s Education Fund (UNICEF) and the World Bank to hold a workshop examining existing efforts to cost ECD services, identify gaps in research, and discuss a way forward in building knowledge in this area. Workshop participants considered how to build ECD costing models and what can be learned from different approaches, including in the health sector. We discussed the factors that influence ECD service quality, and what we know about the trade-off between cost and ECD outcomes. We examined what happens to costs as ECD programs are expanded from pilot programs to large-scale ones (“going to scale”) by looking at examples in Colombia and Turkey. We also debated the cost-effectiveness of bundling various services for young children and their families, and explored the cost implications of strategies for reaching vulnerable populations with ECD programs.
Over the next few weeks, we aim to publish specific posts that detail the key issues and findings that emerged from the various discussions. However, for now, here are my main takeaways from the discussions and the next steps for this work:
- There is scope for improving the availability of data on the costs of ECD services. The different efforts presented at the workshop—detailed budgeting models for South Africa, plus two regional models, one by UNICEF for Western and Central African countries and the other by IDB, based on data from several Latin American countries—show that it is indeed possible to develop robust excel-based costing models. But how one goes about this, including the level of detail needed in specifying the inputs, is a function of not only what information can be readily collected but also on the intended purpose. What is needed for budgeting and planning purposes will, for example, be quite different than what is sufficient for modelling cost effectiveness of programs. There was agreement at the workshop that not only was cost information scant but that the data that did exist was not comparable across programs and countries. This was a result of the absence of protocols on cost data collection, and dispersed and uncoordinated efforts by the research community. One implication of the workshop then, was that it would be highly valuable for guidelines to be developed on how to collect cost information for ECD services. These guidelines could then inform researchers evaluating ECD programs in low and middle-income countries. Programs like the World Bank’s Strategic Impact Evaluation Trust Fund (SIEF) have placed a higher priority on collecting ECD cost data to enable analysis of both the economic impacts and each program’s cost-effectiveness, but progress so far has been limited. The availability of a template or guidelines developed with input from the researchers themselves should facilitate data collection and increase the payoff from such efforts by creating a database of cost data that are comparable across programs.
- What we count is not what counts when it comes to ECD. Most of the costing data available is on structural elements of programs—that is, inputs such as infrastructure, caregiver per child ratios, and the formal qualifications of the caregivers/teachers. There is limited cost data on “process quality,” which refers to how children spend their time and the nature of the interaction between the caregivers and the children. Policymakers are typically more concerned with structural quality even if it is more costly to implement because its elements are measurable, tangible, and can be captured in standards and subsequently monitored. Yet process quality is less expensive and far more important for children’s development and learning outcomes, but it is harder to measure, communicate, and influence. A key need moving forward is to better understand what it takes to improve process quality, so as to make the case for nurture and stimulation as low-cost, high-return investments.
- Systems matter for costs, quality, equity and sustainability. Our knowledge is expanding about what ECD interventions yield positive results for children’s development and later well-being. However, rarely do the interventions studied extend beyond small pilots, and so we have limited information to guide countries in the costs of large-scale implementation of ECD services. We know that it is unwise to extrapolate costs and benefits from a small project to a scaled-up program, and that systemic effects may amplify or undermine specific project benefits. For example, we expect that quality is likely to suffer as an ECD program expands in coverage; unit costs may decline (holding constant population characteristics) after an initial phase of cost increases when systems are being built to manage the scaling up of the provision (e.g. recruitment, training), financing, monitoring, and supervision of ECD services; and marginal costs may or may not increase depending on whether coverage is being extended to hard-to-reach populations. Those at the workshop agreed that too little attention is paid to elements of systems building—training and supervision, monitoring and evaluation, payment mechanism—that are crucial for the quality, equity and sustainability of large-scale ECD programs. Building up the knowledge base for what resources are required to develop and operate these systems is another high-payoff research area.
There is much to be gained from coordinating and pooling the efforts of ECD researchers and practitioners. At the workshop, experts from around the world acknowledged that we need to do better in collecting data on ECD cost and cost-effectiveness. Once we have this data, we can support policymakers implement valuable, evidence-based programs and get the most bang for the buck when resources are constrained and competition for budget is intense. We can also better arm advocates and practitioners to make the case for more funding and enable donors to prioritize ECD investments that are proven to work at scale in low-income contexts.
Stay tuned for other postings from workshop participants on the topic.