Is Cost-Effectiveness Analysis the Answer to the Crisis in Education Cuts?

Key Highlights

• Cost-effective doesn’t mean context-proof; some low-cost interventions like SMS nudges can backfire when context is ignored, especially in communities with low adult literacy or fragile trust in education.
• Affordability alone doesn’t guarantee impact; a well-executed, low-cost programme in Sierra Leone failed to improve learning outcomes, proving that even the best-priced solutions need to work on the ground.
• EdTech isn’t always as low-cost as it looks; some EdTech interventions have higher costs per learner than traditional programmes, once you factor in infrastructure, scalability, and actual usage.

---

As the funding landscape for global advocacy and the public sector continues to shift, many are left grappling with its implications. The education sector continues to weather major shocks and threatens the delivery of Sustainable Development Goal 4 — a paradigm shift that puts a future where every child has access to inclusive and equitable quality education is at risk. While the full impact of these changes remains uncertain, there is an increasing emphasis on demonstrating cost-effectiveness as a key response. In this context, EdTech Hub recently hosted a thought-provoking webinar titled Is Cost-Effectiveness Analysis the Answer to the Crisis in Education Cuts? The session brought together experts, Lee Crawfurd from the Centre of Global Development, Elisabetta Aurino from the University of Barcelona School of Economics and Joel Mitchell from EdTech Hub to discuss the challenges and opportunities of investing in EdTech for learning in low- and middle-income countries (LMICs), bringing out some highlights from ongoing research in some of the LMICs.

Featured Speakers 🔉

• Haani Mazari (moderator), EdTech Hub’s Asia Lead 
• Lee Crawfurd from the Centre of Global Development 
• Elisabetta Aurino from the University of Barcelona School of Economics
• Joel Mitchell from EdTech Hub

Key Insights

The roundtable conversation delved into the offerings of cost-effectiveness in EdTech as a solution to the ongoing crisis of education funding cuts. 

Relevance of understanding drivers, the real impact of costs and defining cost-effectiveness amid uncertain funding cuts

Lee Crawfurd stressed the importance of first understanding the true scope of funding cuts and how we define cost-effectiveness in education at this critical juncture, especially as the full extent of these cuts across sectors and countries remains unclear.

It’s not just the major cuts from USAID or the UK’s recent 40% reduction; several other donors, including France, Belgium, the Netherlands, Finland, and Canada, have also announced significant reductions of 20–30%. It is still unclear exactly how these cuts will be distributed across sectors and countries.

• There’s a strong case to be made that education may face deeper cuts, particularly if life-saving aid for food and medicine is protected. That inevitably shifts the burden to other areas.
• Some global responses suggest we may be approaching the end of the traditional aid paradigm. But that may be overstated: aid levels have historically risen and fallen. This could be part of a cyclical trend, not a permanent shift.
• Still, the scale of these reductions could push progress back by a decade or more. It’s not the end of aid—funding isn’t disappearing entirely—but it does mark a significant setback.

What’s driving these cuts? Crawfurd further mentions a mix of domestic pressures in donor countries: economic uncertainty, high costs of living, and growing public scepticism about the effectiveness of aid. He highlights the importance of cost-effectiveness but cautions against oversimplifying its application.

“Naturally, we spend a lot of time on research and analysis to ensure that resources are used efficiently and go as far as possible. Said Lee. “So yes, cost-effectiveness matters. But I do worry about how it’s sometimes implemented, particularly when analyses are based on pilot programs that perform well under ideal conditions but don’t deliver the same results when scaled, a crucial lens we need to apply more consistently.”

EdTech may not always be as low-cost as it’s perceived to be

The roundtable brought out additional EdTech implementation challenges linked to infrastructure requirements, highlighting some strong examples of personalised adaptive learning, like Mindspark in India or the onebillion one tablet programme in Malawi. However, both depend heavily on existing infrastructure or require new investments in hardware and electricity, which can be expensive or difficult to roll out at scale.

While it’s clear that cost matters, Lee highlighted that organisations like the NGO IIC (International Initiative for Impact Evaluation) are setting the gold standard in transparent and detailed cost reporting. IIC, for instance, has published cost estimates for 79 education programmes, including four that fall under the EdTech category. Interestingly, three of those four EdTech programmes had a higher cost per child than the average across all IIC education programmes. This points to an important caveat: ‘EdTech may not always be as low-cost as it’s often perceived to be. ’

The other thing is impact—and that’s where things get more uncertain. A study conducted in Sierra Leone during the COVID-19 pandemic evaluated a well-designed remote, one-to-one coaching programme delivered by a strong implementing partner. Despite the low cost and high-quality execution, the programme failed to improve learning outcomes. This is a stark reminder that affordability alone doesn’t guarantee effectiveness.

Navigating technological change and finding stability in digital learning

Another concern on the horizon is the pace of technological change, particularly with the rise of artificial intelligence. With EdTech tools evolving so rapidly, there’s a real risk that today’s tailored software may quickly become obsolete, overtaken by newer, more adaptive technologies. This makes long-term planning and investment a challenge for governments and implementers alike.

However, there are bright spots that cover the rapidly growing technology space, particularly in the realm of educational broadcasting. Systematic reviews of EdTech interventions consistently point to large-scale remote instruction models, such as Mexico’s Telesecundaria, as successful and scalable. These programmes, also seen in countries like India and China, fill critical teacher gaps by providing recorded lessons in subjects where local expertise may be lacking. While not a replacement for in-person instruction, these broadcasts serve as powerful complements, particularly in resource-constrained settings.

Scalable programmes like Sesame Street and Akili and Me in Tanzania also illustrate the power of EdTech when the marginal cost of reaching additional learners is virtually zero. According to Lee Crawfurd’s insights, their reach and impact demonstrate how smart design, paired with broadcast technology, can support children at scale.

The power and pitfalls of SMS nudges in rural education engagement

One other delivery modality that continues to attract interest is messaging, particularly SMS, due to its low cost and wide reach, even in low-connectivity, low-infrastructure settings. Elisabetta Aurino, drawing on joint research conducted in Ghana, shared insights from a large-scale randomised controlled trial (RCT) that tested the effectiveness of SMS nudges aimed at parents during school closures.

The intervention was straightforward: send simple, supportive messages to parents in rural Northern Ghana, areas marked by high poverty, food insecurity, and low levels of formal education. The aim was to encourage home-based learning engagement and ease the transition back to school. Messaging was seen as a promising tool because most households had access to basic mobile phones.

The findings were striking and mixed, showing that:

• Among parents who had received some formal education (primarily those who had completed primary school), the SMS nudges were effective. These parents engaged more with their children, school attendance improved, and children showed gains in social and emotional learning.
  
• For the majority of parents, particularly those with no formal schooling, the intervention backfired. These parents engaged less with their children, reported lower aspirations for their children’s education, and became more sceptical of the value of schooling.

It is crucial to go beyond the price tag and rethink low-cost interventions through context

This experience reveals that, as cost-effective as it is known to be, messaging interventions are not one-size-fits-all. While highly cost-effective on paper, they must be carefully tailored to context, especially where adult literacy is low and parental confidence in engaging with education is fragile.

EdTech Hub’s engagement in cost-effectiveness work has witnessed the Best Buys lists and global reports praising low-cost interventions. That’s why we raised a big question: How do we interpret minimal impacts from low-cost interventions? 

As highlighted by Joel Mitchell, this speaks to the philosophy of nudge messaging and whether other contextual factors contribute to an environment ripe for change, creating a behavioural ‘tipping point’ which amplifies the impact of nudges. So much depends on context, and findings from our research suggest that it cannot be attributed to tailoring messages. That was a key part of EdTech Hub’s cost-effectiveness analysis in Ghana.

Cost-effectiveness in messaging and digital personalised learning

One thing that has stood out within our deep dives into thematic areas like messaging and digital personalised learning (DPL) to unpack not just costs and outcomes, but also underlying assumptions, has been how hard it is to define a coherent category of EdTech intervention. What counts as DPL? How do we compare one tool to another? And how do issues like scale and sustainability shape cost-effectiveness over time? There’s no point in having an excellent personalised learning tool if no one uses it. At the same time, platforms like TikTok or YouTube Shorts excel at engagement, but they’re not built for education. It raises the question: How much do we prioritise marketing and adoption over the actual value of what we’re delivering? It’s a tricky balance with no easy answer.

In all this, one of the biggest challenges researchers would face is understanding what policymakers are prioritising. There’s often a tension between building capacity and ensuring efficacy, when really, the two should go hand in hand. A major gap in the evidence is that too few evaluations consider cost alongside impact. This lack of cost data makes it harder to make informed decisions and often shifts the focus toward what’s most feasible, not necessarily what’s most effective.

Other underlying barriers include connectivity, where many people who would benefit most from engaging EdTech interventions, especially in rural areas like Ghana, lack access to smartphones and face high mobile data costs. From the conversation, we also saw infrastructural challenges like limited access to electricity, which further restricts the reach of digital learning solutions. Some communities would prefer in-person interaction with face-to-face engagement, with EdTech serving as a supplementary, low-cost support rather than a standalone solution. 

So, where does that leave us? Indiscriminate use of EdTech, especially focused on device distribution, is rarely cost-effective in education, and the more EdTech interventions are planned without evidence, the worse the reputation of EdTech will be. However, the effectiveness of specific, targeted approaches has a growing evidence base for effectiveness and more integrated mechanisms for data collection within EdTech platforms. This means that in a context of increased cuts and scrutiny of education budgets, EdTech interventions have relative advantages in cost-effectiveness evidence.

Resources 📚

• Why the US Should Keep Funding Global Education by Lee Crawfurd, Centre of Global Development
• Evaluating Cost Effectiveness in EdTech by EdTech Hub
• A ‘Smart Buy’ for All? Unequal and Unintended Consequences of a Messaging Program for Child Education by Dr. Elisabetta Aurino’s January publication on the cost-effectiveness of messaging, University of Barcelona School of Economics
• Assessing the overall benefits of programs enhancing human capital and equity: a new method with an application to school meals by Dr. Elisabetta Aurino, University of Barcelona School of Economics