Editor's note: An updated version of this blog can be found here.
With the added capabilities of our new website, and as a recent student who found VoxDev useful for my own studies, I thought I would outline some of the ways in which VoxDev can be used at universities. This blog intends to highlight useful resources to students looking for interesting examples of the econometric methods they might have been taught without much context or application, or academics preparing courses in need of some applied development economics research. It should be noted, as someone who has never taught economics (but sent this around to those who have, to check my suggestions are somewhat reasonable), these are very much the perspectives of a recent student, rather than a seasoned university economics teacher.
In this blog, I will focus mostly on articles I have edited in my time as Managing Editor which feel particularly useful for students, whether that’s because they use a particular causal inference technique* commonly taught in economic undergraduate or postgraduate degrees, or approach development issues in an insightful way. Hopefully this both demonstrates VoxDev’s value as an academic resource, and introduces you to new research by some of the amazing authors we feature.
If you have a general interest in development economics and enjoy reading a wide range of research, our newsletter is the best way to keep up to date – you can sign up here. (If you are teaching a development economics course, one option might be to sign up your student’s university email addresses.)
You can click each heading below to go straight to a list of all of our articles that use the specific technique in question.
‘Natural Experiments’
Natural experiments involve shocks, policies or other events of interest which lead to some sort of “random” variation in the exposure of individuals or groups. Often these natural experiments are historical accidents that create differences between otherwise similar groups or individuals that can help economists identify a good counterfactual.
Researchers get extremely creative in their search for these accidents. For example, one of our most popular articles of 2023 used the collapse of vulture populations in India due to accidental poisoning to estimate the impact of reduced biodiversity. Another article used the failure to deliver two out of eight boats to fishing communities in the Amazon to show how fishing communities broke out of a poverty trap. In this recent article, offshore oil discoveries and revenue sharing rules in Brazil create a natural experiment which is used to explore whether oil discoveries lead to local development benefits.
Differences-in-differences (DIDs)
Lots of establishing “causality” boils down to finding good comparisons, including a group that experience some sort of change, be it policy-driven or otherwise, and a different group which did not and can be shown to be a good counterfactual. Difference in difference designs (there are many different types!) are a go-to method in this case.
On VoxDev, this method has been used to show that:
- Cash transfer programmes stimulated the local economy in Brazil
- Deportation policies can inadvertently disseminate ideas and criminal networks between countries, unintentionally increasing migration back to the US
- School internet access improved student learning in Peru
- The expansion of the Beijing subway system drove innovation
- Foreign corruption regulation helps African communities capture resource extraction benefits
- Cash transfers reduce adult and child mortality rates in low- and middle-income countries
Regression Discontinuity Designs (RDDs)
Every economist loves a cutoff as it lets us compare people just above and just below, or areas just to the left and just to the right, to establish causal effects.
This recent article uses cutoffs in the eligibility criteria for financial aid to low-income, high-achieving students to explore how this policy impacted students studying at university in Colombia.
Another common RD strategy uses election outcomes – specifically, close elections in which a particular party barely wins some elections and only just loses others - to estimate causal effects. For example, this article uses this type of set-up to show how falling groundwater levels in India have enabled politicians to manipulate aid for electoral gain.
An example of a spatial RDD is this article, which compares places with elevations just above irrigation canals in India to places with elevations just below to explore the long-run development impacts of agricultural productivity gains.
Instrumental Variables (IVs)
A good instrument requires a strong first stage (i.e. it is strongly correlated with the explanatory variable of interest) and must satisfy the exclusion restriction (it must only affect the outcome of interest through its effect on the explanatory variable).
Once an instrument has been found, this can be used in an IV strategy to estimate causal effects. Research featured at VoxDev has used a wide variety of instruments to explore a range of interesting policy-relevant questions:
- Pollution which blows over from China as an instrument for pollution in South Korea to explore the international spillover effects of air pollution
- Least-cost paths and straight line paths to estimate the effects of transport infrastructure – Section 3 of our VoxDevLit on Land Transport Infrastructure
- Distance to major logistical hubs to show how not to disengage from a conflict - evidence from NATO’s war in Afghanistan
- The China shock and baseline variation in the industrial composition of districts – to explore the gendered effects of import competition in Peru
Randomised Control Trials (RCTs)
Well, of course, as a website covering development economics we have featured our fair share of RCT’s – this exercise in classifying our articles by research design has shown it is the most common method used by the research we feature. For a summary of the early stages of this method, and how its pioneers wound up winning a Nobel Prize, see this previous article - What does the 2019 Nobel mean for development economics? Although, RCT’s are not a silver bullet - as explained here by Maitreesh Ghatak, here by Ted Miguel, and here by Paul Niehaus.
This great article outlines results from an RCT in Ghana which expanded professional networks for female entrepreneurs, and discusses the policy implications of this work and why this type of intervention could be scalable. Another article experimentally tests the impacts of a community-led intervention in Northern Nigeria and shows how it significantly reduced rates of child marriage in adolescent girls by changing entrenched, normative behaviour.
The scope and scale of the RCT’s that researchers are able to conduct are increasing. For example, this article outlines results from five years of work with regulators in Kenya to design and implement a new regulation that sought to improve patient safety by imposing minimum quality standards at hospitals across Kenya. This article presents global evidence from five randomised trials on educating children in emergencies.
Calibrated Models
While we don’t feature purely theoretical research, many of our articles present results from calibrating models, i.e. plugging numbers from the real world into models. (As a student disillusioned with unrealistic “Econ101” models and theory, I would have certainly appreciated some examples of the important real-world issues that more advanced modelling can be used to think through.)
One area in which work using these techniques is particularly important is considering how the world will adapt to climate change, as seen in Section IVA of our VoxDevLit on Climate Adaptation. For example, this article shows that coastal favouritism in Vietnam’s infrastructure investment has significant potential welfare costs in the context of projected rising sea levels. Another recent article walks through a model that considers how trade agreements can be designed to limit deforestation, before calibrating this model to consider how much deforestation these contingent trade agreements could prevent.
Other articles have used similar approaches to explore a wide range of issues including:
- Designing public transit networks
- Phasing out coal power
- How current international trade policies subsidise pollution
For a broader discussion on the role of macro modelling, this recent podcast shows the importance of bridging the “micro-macro divide”. Combining results from statistical micro work with macro modelling techniques can improve the policy relevance of economic research.
Heterogeneity
This jargony word beloved by economists (I have become well accustomed with the synonyms) simply conveys that aggregate effects can mask differing impacts. A fantastic example is this recent article on special economic zones in China which, on average, increased local high school enrolment rates. However, this masked very different effects depending on which type of industry SEZ’s sought to promote.
Another great example is this article, which examines how changing economic conditions effects marital outcomes and family formation for women and uncovers key differences in polygynous vs non-polygynous areas.
Other VoxDev resources
While I have mostly stuck to articles when highlighting relevant readings for different methodologies, it’s worth remembering we have a ton of brilliant podcasts in which researchers outline findings from their papers, and discuss the key takeaways from their work. Why not mix up your reading list with the inclusion of some of these episodes of VoxDevTalks?
Also, our growing library of VoxDevLits summarise research on specific development topics. These are kept up to date, don’t need a university admin password to download, and focus on implications for policy. These living literature reviews also throw up lots of ideas for where future research is needed - for those looking for ambitious and important dissertation projects these might be a good place to start.
VoxDev and educating economists
In many ways, VoxDev’s usefulness as an educational resource for students and educators is a spillover from our original mission, to make economic research accessible to policy makers who may not have had the technical training of an economist. This means that, as opposed to an academic publication, the audience for our economic content does not need to understand the methods or jargon of economists (in theory at least, there is always more we can do to make research more accessible).
This makes our articles, podcasts, videos and VoxDevLits great gateways into development economics for students, many of whom will go to work in policy, who can read about the latest, high-quality applied research in an accessible manner with no restrictions or fees.
For those looking for a deeper understanding of the methods I have mentioned above, as I quickly discovered as a student there are few better places than the Development Impact World Bank Blog. In particular, I would recommend Florence Kondylis and David McKenzie’s recent “One-Stop Shop for Methodology” which is a curated list of their methodology explainers. For those at the more advanced end of university development economics, the BREAD-IGC PhD style online courses are another brilliant resource. We have covered this year’s course on environmental economics in a series of recent podcasts here. Another useful set of podcasts outside of our own is the “Trade Talks Episode Catalog for Educators” which divides their podcasts by topic and is a super useful resource. Finally, for video, econimate is a fantastic resource summarising research papers in an animated, accessible style.
Editor’s note: If you think there is anything I should add to this post, or have any comments/suggestions, please email me at [email protected]. I intend for this to become an updated resource, like our VoxDevLits, that will evolve over time as we continue to feature new research that will reflect future methods by economists that have to be added to the syllabus. Also, if you are involved in teaching and have used VoxDev as an educational resource, it is always great to hear about this – tracking this type of “impact” is always tricky and your stories help us understand our reach.
Footnote
* Many of our articles do not rely on the standard causal inference toolkit, but instead establish new facts or measure previously unseen activities (most of these articles can be found here - methods and measurement). I have not included these here but might use them as the basis of a future blogpost.