Back to Top

Resource-based policies in education: a brief introduction to an important theme in education economics

The effects of resource-based policies in education, such as class-size reductions or increased per-pupil funding, is a greatly debated topic in the economics of education. At a very basic level, the debate may seem inane: without any funding at all, there would be no education at all. Yet the question is whether there are diminishing returns beyond certain levels of funding, whatever they may be, where further resources make little difference in the institutional framework in which schools operate.

The argument in favour of resource-based policies is pretty obvious. If we assume that schools strive to maximise results efficiently, more money should generate improvements. For example, with more money, you can hire teacher assistants, thereby increasing the attention each child obtains, or decrease class sizes, which may improve the learning environment overall.

Yet if schools have few incentives to maximise learning outcomes, increasing resources may merely lead to wasteful spending. Indeed, schools operate under conditions that may blunt incentives to improve outcomes. Sceptics’ point, then, is not that resources do not matter at all, but rather that poor incentives reduce their effects.

The problem when empirically analysing the effects of resources is that their distribution is not random: often politicians allocate more money to pupils with lower achievement, for example, such as pupils from low socioeconomic backgrounds. This often generates a zero or even negative correlation between resource-based policies and outcomes, but this correlation does not reflect the causal impact. Instead, it highlights the importance of obtaining variation in resources that are not affected by outcomes, or in other variables that affect outcomes.

And research that obtains such variation often finds that resources, holding everything else constant, improve pupil outcomes in developed-world contexts, at least among poorer pupils. For example, studies indicate that smaller classes appear to have positive effects in some contexts in America, Israel, and Sweden, while increased funding levels improve outcomes in England and America. Yet there are exceptions to this conclusion. For example, class size does not appear to matter in Norway – and the impact in Israel has disappeared in newer data. It is also not clear that funding-based policies benefit less deprived pupils more than marginally. In addition, of course, all resource-based policies do not necessarily have the same effects – and we have yet to determine which ones hold most promise.

In developing-world contexts, the case in favour of funding-based policies is even less clear-cut, as studies indicate they generate small gains overall. This may be because incentives are weaker in developing countries, perhaps due to their having less robust institutions, or because schools are less likely to know how to spend resources efficiently.

Furthermore, whether or not resources improve achievement holding everything else constant may not be the most policy-relevant issue to consider, as reforms that increase resources sometimes have unintended consequences. One example is that large class-size reductions may decrease overall teacher quality as the teacher pool expands to enable the reductions. Similarly, public investments may crowd out parental and pupil investments in education, thereby decreasing the effects of the former. Another issue to consider is whether the effects of resources warrant the increased costs, a question that must be settled in cost-benefit analyses.

Overall, therefore, the debate on resource-based policies has shifted in the past decades. Today, few would argue that resources ‘do not matter’ at a general level. This is a direct result of the improvements in methodology, which have allowed researchers to separate causation from correlation. Still, as highlighted above, there are many outstanding questions that require further research to answer.

Gabriel Heller-Sahlgren

This introduction to the issue of resource-based policies in education was orinally pubished in Volume I, Issue 2 of CfEE's Monthly Research Digest. You can view it in flipbook, download a pdf, and/or subscribe here.

Further reading

Angrist, Joshua D. and Victor Lavy. 2017. ‘Maimonides Rule Redux.’ NBER Working Paper No. 5888.

Ganimian, Alejandro J. and Richard J. Murnane. 2016. ‘Improving Education in Developing Countries: Lessons from Rigorous Impact Evaluations.’ Review of Educational Research 86(3): 719-755.

Gibbons, Stephen, Sandra McNally, and Martina Viarengo, ’Does Additional Spending Help Urban Schools? An Evaluation using Boundary Discontinuities.’ Journal of the European Economic Association 16(5): 1618–1668.

Jackson, C. Kirabo. 2018. ‘Does School Spending Matter? The New Literature on the Old Question.’ Unpublished manuscript, Northwestern University.

Jepsen, Christopher. 2015. ‘Class Size: Does it Matter for Student Achievement?’ IZA World of Labor 2015: 190

Leuven, Edwin. and Sturla A. Løkken. 2018. ‘Long Term Impacts of Class Size in Compulsory School.’ Journal of Human Resources (forthcoming).

de Ree, Joppe, Karthik Muralidharan, Menno Pradhan, and Halsey Rogers. 2017. ‘Double for Nothing? Experimental Evidence on an Unconditional Teacher Salary Increase in Indonesia.’ Quarterly Journal of Economics 133(2): 993-1039.

Blog Category: 
About the author