VoxDev Development Economics

Many years ago, Tim Besley (LSE) had a "crazily simplistic" idea to categorise states by their capacity, and in his early work he suggested three groups. When ten years later, he gave an algorithm some data on the measures of state capacity, and left it to categorise states, it also found three groups.
Which states are in which groups barely changes. Does that make a country's group its destiny? In the latest VoxDev Talk, based on his lecture in the IGC-BREAD series on political economy, Besley tells Tim Phillips why it does not. But he also explains why giving citizens the power to vote doesn't necessarily help solve the problems that create weak states. 
The lecture behind this episode:
Besley, Tim. 2026. "State Capacity, Conflict and Development." Lecture 4, BREAD-IGC virtual PhD-level course on political economy, 29 September. Lecture slides. Watch the recording: [LECTURE RECORDING URL].
The research behind this episode:
Besley, Timothy, Chris Dann, and Torsten Persson. 2021. "Pillars of Prosperity: A Ten-Year Update." CEPR Discussion Paper 16256. Gated.
Besley, Timothy. 2020. "State Capacity, Reciprocity, and the Social Contract." Econometrica 88 (4): 1307-35.
Besley, Timothy, and Torsten Persson. 2011. Pillars of Prosperity: The Political Economics of Development Clusters. Princeton: Princeton University Press.
Besley, Timothy, and Torsten Persson. 2010. "State Capacity, Conflict, and Development." Econometrica 78 (1): 1-34.
To cite this episode:
Phillips, Tim, and Tim Besley. 2026. "Why democracy alone does not build effective states." VoxDev Talks (podcast).

About the guest

Sir Tim Besley is School Professor of Economics and Political Science and W. Arthur Lewis Professor of Development Economics at the London School of Economics and Political Science. He is known for research spanning political economy, development economics and public economics, with a focus on how governments can design and deliver economic policy more effectively. He served on the Bank of England's Monetary Policy Committee from 2006 to 2009 and was an academic director of the LSE-Oxford Commission on State Fragility, Growth and Development. He is a past president of the Econometric Society, the Royal Economic Society, the European Economic Association and the International Economic Association, and is affiliated with the Centre for Economic Policy Research.
Research cited in this episode

State capacity. In Besley's work, state capacity has three dimensions. Fiscal capacity is the breadth of the tax system, legal capacity is the ability to enforce contracts and property rights, and collective capacity is the ability to provide services such as health and education. Besley and Persson (2011) built the first two on Adam Smith's recipe for prosperity, "peace, easy taxes, and a tolerable administration of justice".
Quasi-voluntary compliance. Margaret Levi's term for what happens when people pay tax because they accept the arrangement, and not only because they fear punishment (Levi, Margaret. 1988. Of Rule and Revenue. Berkeley: University of California Press). Besley (2020) models this as a reciprocal contract; citizens pay their taxes and the state provides public goods.
Circular causation. Gunnar Myrdal, who shared the 1974 Nobel Memorial Prize in Economic Sciences with Friedrich Hayek, described development as a system in which "if one changes, others will change in response" (Myrdal, Gunnar. 1974. "What Is Development?" Journal of Economic Issues 8 (4): 729-36). Besley uses the idea as a metaphor for state capacity and development. One does not cause the other; common processes lie behind both.
Institutions. Acemoglu, Johnson and Robinson used differences in European settler mortality to estimate the effect of institutions on income (Acemoglu, Daron, Simon Johnson, and James A. Robinson. 2001. "The Colonial Origins of Comparative Development: An Empirical Investigation." American Economic Review 91 (5): 1369-1401). Besley's addition is that state capacity moves more slowly than political institutions, and accumulates like a form of capital.
Three kinds of state. Common interest states are peaceful and have strong capacities in every dimension. Special interest states, which Besley, Dann and Persson (2021) call redistributive states, are repressive and have some capacities; Besley's example is China. Weak states are conflictual and have weak capacities.
Finding the clusters. Besley, Dann and Persson (2021) use six variables for 67 countries in 1990, 2000 and 2016. These are the income tax share of total tax, a legal quality index, a collective capacity index, the share of years since 1975 in repression, the share in civil war, and GDP per capita. They summarise the data with principal components and then apply hierarchical clustering, which lets the algorithm choose the number of groups. Eight of the 67 countries changed cluster between 1990 and 2016. 
Elections or executive constraints? Besley separates two stories of democratisation. One is the extension of the vote, which in Britain ran through the Reform Acts of 1832, 1867 and 1884 and the votes for women in 1918 and 1928. The other is the limit placed on what those in power can do. In the lecture slides, all three kinds of state score above 90% of years at maximum openness between 1950 and 2016. At maximum executive constraints, weak, special interest and common interest states score about 15%, 35% and 93%. Besley stresses that this is a correlation.
Why rulers accept limits. Besley gives two accounts. In the first, elites give up power under the threat of revolution (Acemoglu, Daron, and James A. Robinson. 2006. Economic Origins of Dictatorship and Democracy. Cambridge: Cambridge University Press). The second is about ideas. Besley describes an early-stage paper, presented in Chicago, which finds common elements between speeches in the House of Commons and John Locke's Two Treatises of Government.
War and common purpose. Besley and Persson argue that common interest public goods, such as fighting external wars, encourage investment in state capacity (Besley, Timothy, and Torsten Persson. 2009. "The Origins of State Capacity: Property Rights, Taxation, and Politics." American Economic Review 99 (4): 1218-44). Besley's post-war example is the National Health Service, founded on 5 July 1948 and kept by the Conservatives when they returned to power. He develops the wider argument in Besley, Timothy. 2021. "Is Cohesive Capitalism Under Threat?" Oxford Review of Economic Policy 37 (4): 720-33.
Civil society. Besley's point that effective states are built from the ground up draws on Robert Putnam's study of Italy's regional governments, which links their performance to a tradition of civic engagement (Putnam, Robert D., with Robert Leonardi and Raffaella Y. Nanetti. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press).
Pivotal moments. Besley was an academic director of the LSE-Oxford Commission on State Fragility, Growth and Development, whose 2018 report Escaping the Fragility Trap looked at moments of potential change, such as the end of a conflict. His example is Rwanda since the genocide, where he sees real progress while acknowledging that many questions will be raised about it.
More VoxDev Talks episodes

Armed groups as states: Taxation, security, and civilian support. Raul Sanchez de la Sierra on eastern Congo, the case Tim puts to Besley, where militias do much of what people expect from a state.
Taxation, civic culture and state capacity. Besley's earlier VoxDev Talks episode on why people pay their taxes, and the reciprocal contract behind high compliance.
No taxation without administration: What makes tax authorities work. Anders Jensen and Jonathan Weigel on how tax authorities are organised and staffed; their paper is on the reading list for Besley's lecture.
Related reading on VoxDev

State capacity and development clusters. Besley, Dann and Persson on 25 years of data, and why countries form persistent groups according to their levels of internal peace and state capacity.
Does low state capacity set a ceiling on tax rates? Evidence from the DRC. Augustin Bergeron, Gabriel Tourek and Jonathan Weigel report the kind of field experiment on tax that Besley wants the next generation to run.

This is an episode from VoxDev's new podcast series, Ideas in Development. This series has a separate podcast feed, where you can find every episode of Oliver Hanney’s conversations on evidence.
YouTube: https://www.youtube.com/watch?v=_DYOzIhbZlI 
Apple Podcasts: https://podcasts.apple.com/us/podcast/does-africa-need-development-economics/id1866874059?i=1000793373513 
Spotify: https://open.spotify.com/episode/3naR6n03DiTzzokWqaJ6cs?si=-CrMmOO1RYi_spJwpm3HNA 
Audioboom: https://audioboom.com/posts/8960400-does-africa-need-development-economics 
Substack: https://ideasindevelopment.substack.com/p/does-africa-need-development-economics 
If every development economist stopped their research tomorrow, would Africa's growth rate in 2040 look any different?
Ken Opalo, associate professor at Georgetown's School of Foreign Service and author of the Substack An Africanist Perspective, joins Oliver Hanney to argue that the constraint on African growth is not knowing what to do but knowing how to get it done. He also explains why Kenya's evidence boom has coincided with declining policymaking, and what happens when policymakers outsource their ambitions.
In this wide-ranging conversation we cover:

Why the SDGs are the wrong goal for an ambitious minister.

How narrow evidence reshaped education policy in Tanzania and Kenya.

The craft of policymaking.

When a policymaker should actually listen to a development economist.

How to learn the right lessons from growth success stories.

What the third wave of philanthropy should fund.

Ideas in Development is a VoxDev podcast on economic policy and research.

In parts of eastern Congo, the organisation that collects your taxes, settles your disputes and holds the monopoly on cigarettes is not the state.
Raúl Sánchez de la Sierra (LSE) has spent years investigating how Congo's armed militias work. He tells Tim Phillips how he won the trust of its commanders without building the bomb they requested. He was allowed to see administrative records, and what they reveal about how the group raises and spends its money. He also interviewed recruits and their neighbours to find out who chooses to fight and why. It's a fascinating insight into a system of government that many economists never see.
The research behind this episode:
Sanchez de la Sierra, Raul, Louis Dunia, and Hilary Yu. 2025. "Moral Violence: At the Heart of One of Congo's Major Armed Groups." Working paper.
Henn, Soeren J., Christian Mastaki Mugaruka, Miguel Ortiz, Raul Sanchez de la Sierra, and David Qihang Wu. 2025. "Monopoly of Taxation without a Monopoly of Violence: The Weak State's Trade-Offs from Taxation." Review of Economic Studies 92 (2): 1126-56.
Sanchez de la Sierra, Raul, Kristof Titeca, Haoyang (Stan) Xie, Albert Malukisa Nkuku, and Aimable Amani Lameke. 2024. "The Real State: Inside the Congo's Traffic Police Agency." American Economic Review 114 (12): 3976-4014.
Sanchez de la Sierra, Raul. 2020. "On the Origins of the State: Stationary Bandits and Taxation in Eastern Congo." Journal of Political Economy 128 (1): 32-74.
Humphreys, Macartan, Raul Sanchez de la Sierra, and Peter van der Windt. 2019. "Exporting Democratic Practices: Evidence from a Village Governance Intervention in Eastern Congo." Journal of Development Economics 140: 279-301.
To cite this episode:
Phillips, Tim, and Raul Sanchez de la Sierra. 2026. "Armed groups as states: Taxation, security, and civilian support." VoxDev Talks (podcast).

About the guest

Raul Sanchez de la Sierra is Associate Professor of Management at the London School of Economics and Political Science. His research spanning political economy and development economics draws on sociology, anthropology and psychology to study the origins of the state, the internal organisation of armed groups, the hidden workings of state agencies and the psychology of the choice to use violence. He was one of the founders of Marakuja Kivu Research, a Congolese research organisation based in Goma.
Research cited in this episode

Community-driven development in eastern Congo. Raul's first work in the region was as a research assistant on the impact evaluation of a large village governance programme funded by the UK's Department for International Development. It aimed to bring elected committees to villages run by chiefs. Villagers objected when chiefs were barred from standing, and objected again when a chief stood and lost. Humphreys, Sanchez de la Sierra and van der Windt (2019) report the results.
Stationary bandits. Mancur Olson's idea: a roving bandit takes everything and moves on, but a bandit who stays has an interest in next year's harvest, so he taxes rather than plunders and keeps rivals out (Olson, Mancur. 1993. "Dictatorship, Democracy, and Development." American Political Science Review 87 (3): 567-76). Sanchez de la Sierra (2020) tested it in eastern Congo; his VoxEU column On the origins of the state: Stationary bandits and taxation in eastern Congo summarises the evidence.
When the state takes back territory. Henn et al. (2025) study a Congolese army operation that stopped armed groups taxing villages they had controlled for years. Unable to tax, the same groups turned to pillage. Raul uses it to show how policy that looks only at the legal map can backfire.
Who joins an armed group. Most studies of combatants interview veterans, who are survivors and have been changed by fighting. Sanchez de la Sierra, Dunia and Yu (2025) set up a phone alert with the armed group's 55 human resources commanders, so survey teams could interview recruits as they enlisted and, at the same time, a random sample of civilians in their home villages. Raul says the team spoke in person to about 300 new joiners.
Measuring what people will not say. To limit misreporting, answers were given in private; the team also measured skin conductance, a physiological signal strongly associated with emotional arousal, collected supervisor and peer ratings seven months after enlistment, and asked second-order questions such as why respondents think others joined.
Parochial altruism. A willingness to make sacrifices for your own group combined with hostility to outsiders. Choi and Bowles modelled how the two traits can evolve together through war (Choi, Jung-Kyoo, and Samuel Bowles. 2007. "The Coevolution of Parochial Altruism and War." Science 318 (5850): 636-40). Raul argues his setting goes further, because the act of being prosocial towards the group depends on fighting the enemy.
Moral disengagement. Albert Bandura's term for the psychological steps, including dehumanisation, that let people harm others without feeling they have broken their own moral code (Bandura, Albert. 1999. "Moral Disengagement in the Perpetration of Inhumanities." Personality and Social Psychology Review 3 (3): 193-209).
The economics of corruption. Shleifer and Vishny separate the question of whether bribery is wrong from the question of when it is less efficient than taxation (Shleifer, Andrei, and Robert W. Vishny. 1993. "Corruption." Quarterly Journal of Economics 108 (3): 599-617). Raul uses that approach in his study of Kinshasa's traffic police.
Kinshasa's traffic police. Sanchez de la Sierra et al. (2024) follow bribes up the chain of command in Kinshasa's traffic police agency. Station commanders set officers quotas of cars to bring in, and most of the agency's revenue comes from payments made behind closed doors at the station, not from bribes on the street. The team randomly reduced the quotas and measured what happened to traffic jams, accidents and enforcement.
Pitfalls of policy advice. Acemoglu and Robinson argue that advice which ignores the political equilibrium can backfire, because it changes who holds power (Acemoglu, Daron, and James A. Robinson. 2013. "Economics versus Politics: Pitfalls of Policy Advice." Journal of Economic Perspectives 27 (2): 173-92).
More VoxDev Talks episodes

No taxation without administration: What makes tax authorities work. Anders Jensen and Jonathan Weigel on how tax authorities in developing countries are organised and staffed, from the formal side of the state Raul's militias imitate.
Corruption in public procurement: How politicians co-opt bureaucrats. Sarah Brierley on why merit-based hiring alone does not stop corruption, a companion to Raul's work on the hidden structure of Kinshasa's traffic police.
Related reading on VoxDev

Can low-capacity governments work with local leaders to increase tax revenues? Evidence from the Democratic Republic of Congo. Pablo Balan, Augustin Bergeron, Gabriel Tourek and Jonathan Weigel find that city chiefs in Kananga raised more property tax than state collectors.
Does low state capacity set a ceiling on tax rates? Evidence from the DRC. Augustin Bergeron, Gabriel Tourek and Jonathan Weigel find that, in a property tax experiment, lower tax rates raised compliance and revenue, and stronger enforcement raised the rate at which revenue peaks.

A tractor can plough an acre of Kenyan maize land in about an hour and a half. By hand, it takes a farmer nearly a week. Few smallholders can afford to buy a tractor, so they rent one. The tractor’s owner often lives in town, and employs an operator to drive the tractor to the fields and do business. To minimise risk, the owner keeps their tractor close to home, even when the best work is far away.
Sophie Nottmeyer studied what changed when owners could observe exactly where the tractor was. She used GPS records from about 1,200 tractors fitted with a tracking app from the start-up Hello Tractor, covering close to a million fields, and satellite images of the land they worked. As owners learned to use the app, their tractors travelled further, finding better business opportunities. A year after adoption, they worked about 55 km further from home than in their first month.
Nottmeyer presented the research at STEG's sixth Annual Conference in Nairobi in January 2026. She tells Tim Phillips what satellites reveal about the fields the tractors reached, and why a fuel subsidy that did the same job would cost hundreds of times more.
The research behind this episode

Nottmeyer, Sophie. 2025. "Trac(k)tors of Change: Monitoring, Tractor Mobility and Agricultural Mechanization in Kenya." Working paper, November 2025. A summary is available as a STEG policy brief: "Following the tracks: how digital monitoring expands access to agricultural mechanisation across space" (2026).
To cite this episode

Phillips, Tim, and Sophie Nottmeyer. 2026. "Tracking Kenya's tractors." Conversations on Transformation (podcast).
About the guest

Sophie Nottmeyer is a Research Economist in the Directorate General of Economic Research at Banco de Mexico. She was previously a Postdoctoral Researcher at Universitat Pompeu Fabra in Barcelona, and holds a PhD in Economics from CEMFI in Madrid. Her research spans market access, organisational frictions and informal institutions, and how they interact with geography and climate. She has also worked with remote-sensing scientists to estimate maize yields in Kenya from satellite data.
Research cited in this episode

Hello Tractor is an agricultural technology company based in Nairobi. In 2018 it introduced a GPS tracking device for tractors, fitted by a mechanic, with a mobile and web app that shows owners where their tractor is in real time. The app reports how many jobs the tractor did, how many acres it worked, how long it ran and how far it drove, and it can replay the tractor's exact path across a field. In 2023 owners paid US$125 to install the device and US$60 a year to use the app. Nottmeyer's data cover every job done by tractors that adopted it between 2018 and 2025, about 917,000 in total.
Moral hazard arises when one party cannot observe what another does on its behalf. Kenyan tractor owners hire operators, who are often paid from the revenue they collect. Because the owner cannot see how many acres were ploughed, an operator can under-report jobs, take unauthorised work or use fuel for other purposes. Owners respond with surprise visits, fuel checks and rules that the tractor must come home every night. All of these work only if the tractor stays nearby. The problem is sharper because many owners are not farmers at all; according to Kenya's 2019 census, 33% of tractor owners live in cities.
Misallocation describes an economy in which capital or labour does not go where it would produce the most. Chang-Tai Hsieh and Peter Klenow showed in "Misallocation and Manufacturing TFP in China and India" (Quarterly Journal of Economics, 2009) how large the cost of such gaps can be. Nottmeyer measures misallocation as the spread in the returns that tractors earn in different places. If a tractor would earn more in a distant district than near home, but never goes there, the economy loses output.
Counting Kenya's tractors is harder than it sounds. No official source records tractor ownership or use, Kenya's last nationwide agricultural census was in 1979, and the Food and Agriculture Organization stopped publishing its global series on agricultural machinery in 2009. The roughly 1,225 tractors in Nottmeyer's data represent between 1.73% of the tractors counted by the 2019 census and 8.75% of the last FAO figure, from 2008.
Comparing each tractor with itself is how Nottmeyer deals with the lack of a control group. The data begin only when a tractor adopts the app, so she uses each tractor's first month as its baseline, on the assumption that owners take time to trust the information and change how they manage operators. The average distance from home was 70 km in that first month. It rose by about 30 km after six months, 55 km after one year and 112 km after two years. The pattern is similar for early and late adopters, for older tractors whose owners had run them before fitting the device, and for tractors that return home every night. Nottmeyer does not claim these estimates are strictly causal, but the alternative explanations she tests do not account for the pattern.
Satellite measures of crop growth stand in for yields, which nobody records at field level in Kenya. Nottmeyer uses Sentinel-2 images from the European Union's Copernicus programme to calculate the Normalized Difference Vegetation Index (NDVI), which measures how green and dense vegetation is. Ploughing leaves a field bare at the start of the season, so the gap between the lowest and highest NDVI in a year captures both land preparation and crop growth. She matches 33,256 fields worked by monitored tractors with similar fields nearby. In the year of the visit, the gap rises by about 0.4 standard deviations more in the visited fields, and part of the gain persists in later years. Because some of the comparison fields may also have been ploughed by other tractors, this is a lower bound on what mechanisation does.
A spatial model of tractor location turns these results into national figures. Owners choose where to send a tractor by weighing the returns in each place against the cost of distance, and monitoring lowers that cost. When all Hello Tractor tractors behave as if they had five years of experience with the app, the spread of returns across places falls by 15.9% and total output rises by 2.18%, compared with tractors that have just adopted it. If every tractor in Kenya had the technology, output would rise by 1.3%.
A fuel subsidy is the obvious alternative policy, because it makes long trips cheaper. Nottmeyer calculates that to achieve the same gains, a subsidy would have to cover almost all the fuel a tractor uses per kilometre, at a total cost of about US$9.33 billion. Fitting all 70,930 tractors in the 2019 census with the device and a five-year subscription would cost about US$30.1 million.
Monitoring in other industries follows the same logic wherever an owner hands a vehicle to someone else to operate. Thomas Hubbard's "The Demand for Monitoring Technologies: The Case of Trucking" (Quarterly Journal of Economics, 2000) studied the onboard computers that US trucking firms installed to track drivers. Erin Kelley, Gregory Lane and David Schonholzer ran an experiment with Nairobi's minibus owners, reported in "Monitoring in Small Firms: Experimental Evidence from Kenyan Public Transit" (American Economic Review, 2024), and found that tracking devices helped owners manage drivers and raised profits.
More Conversations on Transformation episodes

Moving to profitability. Carolyn Pelnik asks why street vendors and motorcycle taxi drivers in Kampala stay in less profitable parts of the city, another case in which mobile workers and assets do not go where they would earn most.
Slack and economic development. Felix Samy Soliman explains why firms in Kenya leave labour and capital idle, and how much growth that unused capacity could deliver.
Further reading

Misallocation and Product Choice, a STEG working paper by Stepan Gordeev and Sudhir Singh, uses farm and crop data from India to estimate how much output is lost when inputs are spread unevenly between farms, and shows that the answer depends on which crops farmers can choose to grow.
Tillers of Prosperity: Land Ownership, Reallocation, and Structural Transformation, a STEG working paper by Shuhei Kitamura, finds that when the land reform enforced by the Allies after the Second World War gave tenants ownership of their farms, they adopted low-cost farm technology, relied less on family labour, and more young people left for the cities.

Nearly every child in a low- or middle-income country now goes to school, but about 70% of 10-year-olds still cannot read or understand a simple text. Researchers have tested ways to fix the problem, and governments want them to work. So why has the needle not moved?
Noam Angrist (Oxford) and nine co-authors argue that the missing piece is the study of implementation itself. In this week's VoxDev Talk, Tim Phillips asks Noam about their framework, just published in Science, that sets out how small tightly controlled experiments can scale to large real-world trials, then into government policy and finally into classrooms. 
As Noam describes, it's not easy to convert good intentions and a clear need to real change. Listen to this week's episode to find out why.
The research behind this episode:
Angrist, Noam, Tahir Andrabi, Stefan Dercon, Sharnic Djaker, James Habyarimana, Michelle Kaffenberger, Clare Leaver, Akshay Mangla, Nompumelelo Mohohlwane, and Abhijeet Singh. 2026. "An Agenda for Implementation Science in Education." Science 393 (6816): 1084-87.
To cite this episode:
Phillips, Tim, and Noam Angrist. 2026. "Implementation science in education: Scaling what works." VoxDev Talks (podcast). 

About the guest

Noam Angrist is Academic Director of the What Works Hub for Global Education and a Senior Research Fellow at the Blavatnik School of Government, University of Oxford. He co-founded Youth Impact, an NGO headquartered in Botswana that scales and tests health and education programmes. His research spans the scale-up and implementation of education programmes, global learning data, and A/B testing in the social sector. He led key parts of the education pillar of the World Bank Human Capital Index.
Research cited in this episode

Learning poverty. A World Bank measure: the share of children who cannot read and understand a simple text by age 10. A 2022 report by the World Bank, UNESCO, UNICEF, FCDO, USAID and the Gates Foundation estimated it at about 70% in low- and middle-income countries, up from 57% before the pandemic. It is the gap between schooling and learning in a single number.
Structured pedagogy and teaching at the right level. Two of the approaches with the strongest evidence for improving learning cheaply. Structured pedagogy gives teachers detailed lesson plans, materials and coaching. Teaching at the right level groups children by what they can do, not by their age or grade. The Global Education Evidence Advisory Panel reviews this evidence for cost-effectiveness; Angrist led the research behind its first report.
The economist as plumber. Esther Duflo's 2017 Richard T. Ely Lecture to the American Economic Association. She argued that economists who help governments design policy must work on the details of implementation, tinkering and adjusting as plumbers do. Duflo, Esther. 2017. "The Economist as Plumber." American Economic Review 107 (5): 1-26.
The voltage effect. John List's term for the drop in impact when a programme that worked in a pilot is expanded. List, John A. 2022. The Voltage Effect: How to Make Good Ideas Great and Great Ideas Scale. New York: Currency.
The What Works Hub for Global Education. A research centre based at the Blavatnik School of Government, funded initially by the UK Foreign, Commonwealth and Development Office and the Gates Foundation. It studies how proven education programmes can be implemented and sustained at scale. More at wwhge.org.
Efficacy and effectiveness trials. Terms borrowed from health research. An efficacy trial asks whether something can work under controlled conditions. An effectiveness trial asks whether it works in the real world: at scale, delivered by government, at a cost governments can pay, and for different groups of children. Most education evidence today comes from efficacy trials.
Adaptive trials and A/B testing. Instead of one large, static experiment, researchers run many small randomised tests, change the programme between rounds, and test again. Angrist's shorthand is three Rs: rigorous, as in a standard randomised trial, but also rapid and regular. Using data already in government systems keeps the cost of each round low.
How research affects policy. An experiment with 2,150 Brazilian municipalities found that mayors will pay to learn research results, and that telling them about evidence on tax reminder letters made adoption of the policy more likely. Angrist cites it as an example of studying evidence uptake directly. Hjort, Jonas, Diana Moreira, Gautam Rao, and Juan Francisco Santini. 2021. "How Research Affects Policy: Experimental Evidence from 2,150 Brazilian Municipalities." American Economic Review 111 (5): 1442-80. Angrist also describes a forthcoming experiment in Bangladesh that showed policymakers different kinds of evidence; early results suggest global causal evidence carries more weight than many assume.
Second-best policy. If the best policy is politically impossible, a workable alternative that still improves learning may be the better target. Angrist credits Stefan Dercon and Lant Pritchett with making this case. Part of the problem is visibility; a new school is easy to point at, and learning is rarely measured clearly.
Outcomes funding. Funders pay providers for measured results, such as learning gains, rather than for inputs. The What Works Hub is studying two schemes: the Ghana Education Outcomes Project and the Sierra Leone Education Innovation Challenge, an $18 million programme across 325 schools from 2022 to 2025.
The policy-practice gap. During COVID-19, governments adopted radio, television and phone-based teaching. Angrist and Dercon matched government policy data to household surveys to see who received it. Angrist says nearly all governments adopted new distance education policies, but just over 20% of households in Sub-Saharan Africa and about half in the Middle East and North Africa received them. Latin America did far better. Angrist, Noam, and Stefan Dercon. 2024. "Mind the Gap between Education Policy and Practice." Nature Human Behaviour 8 (12): 2261-63.
Difference-in-differences at scale. Once a programme reaches every school, there is no control group for a randomised trial. Difference-in-differences compares changes over time between groups. Comparing districts or schools that did and did not get a programme does not work after a national rollout; comparing a cohort that received it with an earlier cohort that did not still does. Angrist says a recent review he worked on found more than 65% of such studies in education use the location design.
Tusome. Kenya's national literacy programme, a structured pedagogy approach scaled to every public primary school from 2015. It improved learning at scale, but its effects did not last as many hoped. Piper, Benjamin, et al. 2018. "Scaling Up Successfully: Lessons from Kenya's Tusome National Literacy Program." Journal of Educational Change 19 (3): 293-321.
Teaching at the right level in Uttar Pradesh. The approach, developed by the Indian NGO Pratham, was refined over years of randomised trials before government scale-up. Angrist says the Uttar Pradesh rollout reached more than five million children in over 100,000 schools, and that a new quasi-experimental evaluation finds it worked. The earlier scale-up trials: Banerjee, Abhijit, Rukmini Banerji, James Berry, Esther Duflo, Harini Kannan, Shobhini Mukerji, Marc Shotland, and Michael Walton. 2017. "From Proof of Concept to Scalable Policies: Challenges and Solutions, with an Application." Journal of Economic Perspectives 31 (4): 73-102.
More VoxDev Talks episodes

The gap between education policy and practice. Angrist's previous visit, on the COVID-19 data behind the policy-practice gap.
Schools are failing to deliver learning. Lant Pritchett on why schooling is not education, the conversation Tim refers to at the start of this episode.
Increasing learning at scale in Ghana. Adrienne Lucas on testing four learning programmes at scale with the Ghanaian government.
Related reading on VoxDev

Implementation matters: Measure it and account for it. Noam Angrist and Rafe Meager show that implementation explains most of the variation in the effects of targeted instruction across studies.
Educating children in emergencies: Global evidence from five randomised trials. Evidence on low-cost phone tutoring tested across five countries.