Cost Analysis for Education Projects: Resources and Reflections

You’ve got an education program, and you’re confident that it’s having an impact. But is it worth the cost? How can you know, and how can you compare it to other education programs? Cost-effectiveness analysis tells you how much you pay for a given increase in student learning or student school participation, but most evaluations don’t include it (for various reasons).

If you want to do cost analysis, here are a few resources (depending on what you want) and a few of my own reflections. (There are many more resources out there; if you have favorites, share them in the comments!)

Resources

If you want a quick introduction (10 pages or less):

• The Abdul Latif Jameel Poverty Action Lab (J-PAL) has a helpful brief, including suggestions on what to do if you can’t completely break down the costs into different components.
• A brief from the IRC and the World Bank provides a simple 5-step process for gathering cost data.
• USAID has a 2-pager to motivate how you can use cost data and the kinds of questions you can use it to answer, and how to get started with cost analysis.

If you want templates to help you capture the costs:

• J-PAL has a detailed Excel template that will spit out an estimate of cost-effectiveness for you in the final tab.
• Hey, that seems too complicated! Is there a simpler template? More detailed cost data will get you better estimates, but J-PAL also provides a much simpler template to get you a general sense of the cost of the program.
• USAID has templates as well; they’re a little less automated.

If you want more detail on how to get cost analysis right:

• As a basic introduction, you can’t beat Patrick McEwan’s 20-page primer on “Cost-effectiveness analysis of education and health interventions in developing countries.”
• USAID has a cost analysis guide (100+ pages) for education projects that explains all the methods and then a guide to doing this in practice.
• Dhaliwal and others walk readers through many of the decisions you have to make in cost analysis and their implications.

Reflections

First, remember that just like estimates of the impact of a program, cost-effectiveness estimates also come with errors. Popova and I showed that taking those errors into account can dramatically re-order lists of which programs are most cost effective. As a result, I wouldn’t put much stock in small differences in cost-effectiveness.

Second, if you’re using cost-effectiveness analysis from one place to decide whether to implement a program from one place in another place, remember that costs can vary dramatically from place to place. Working with data from a large NGO, Tulloch showed that “costs for the same intervention can vary as much as twenty times when scale or context is changed”! Imagine you’re implementing a program that involves driving to visit schools. Popova and I showed that the transportation cost per school was 27 times higher in rural Kenya than in urban India. So if you want to transport cost-effectiveness from one place to another, make sure you update the costs based on local prices. (And don’t forget that the size of the program’s impact—the “effectiveness” in cost-effectiveness—can change a lot from context to context as well!)

Third, remember that cost-effectiveness won’t be (and shouldn’t be) the only factor in deciding which programs to focus investments on. A program with small impacts that is very cheap may be highly cost effective. It’s great to do that program, but maybe not if it distracts teams with limited operational capacity from programs that will deliver big impacts.

Even with these caveats, just as I’d never make a major purchase without looking at the price tag, I’d never recommend that an education system implement a new policy or program without trying to estimate the cost and thinking through the benefits that come with those costs.

-----------------------------------------

A few other miscellaneous resources for the curious

• The Systematic Cost Analysis Consortium distributes a tool called Dioptra, which actually plugs into programs’ accounting data to facilitate cost analysis and make it maximally comparable across programs. Learn more here!
• With others, I’ve done some analysis on How to Improve Education Outcomes Most Efficiently? using cost-effectiveness. 

If you want someone else’s take on the best resources for cost-effectiveness analysis (not limited to education), Glandon and others provide their “ten best resources” for cost-effectiveness analysis in impact evaluations.