Evidence gaps: Key unanswered questions in development economics

Editor's note: This blog has been updated twice to incorporate the latest VoxDevLits.

We now have VoxDevLits on 20 topics within development economics. In each of these living reviews, a community of scholars identifies the key takeaways for policy from economic research in low- and middle-income countries. They also shed light on evidence gaps across different topics in development economics.

In this blog, I have simply collated the key unanswered questions in development economics, identified by the teams behind our VoxDevLits, in one place. These signpost areas where more research is needed and could even be interesting dissertation topics for ambitious economics students.

Since we aim to update our VoxDevLits approximately once per year, if there has been any recent research that speaks to one of the questions below, please send this through ([email protected]) so that I can pass this on to the Senior Editors ahead of the update.

Promising areas for EdTech – Abhijeet Singh

The EdTech team identify a number of key research questions related to the use of AI in education. Importantly, it is not a question of whether or not to use AI. Instead, the question is what additional governance, assessment, and usage norms are required for AI to engage students productively and enhance learning rather than reduce it?

There are another series of questions around scaling EdTech programmes, which is a key challenge for policy. For example, more research is needed on how to encourage effective computer-assisted learning usage at scale using existing resources (i.e. without hiring additional staff).

Other promising areas include short tightly scripted videos including worked examples, and low cost communication that improves coordination between teachers and parents.

The team emphasise that testing questions in EdTech can be done quickly and methodically. They call for researchers to test whether implementation bundles raise exposure and progression in the short run; scale those that do into longer trials focused on learning; and continuously simplify supports so they can be sustained with existing resources.

Read the full VoxDevLit here: EdTech.

Five important knowledge gaps from research on entrepreneurship training – David McKenzie & Christopher Woodruff

• What is the best way to screen and select entrepreneurs for training and consulting, in order to match them to the appropriate type and maximise impacts?
• How do we improve the cost-effectiveness and scalability of training and consulting, for example, by using online and generative AI approaches to training, or innovative forms of consulting?
• What are the factors that limit the adoption of proven beneficial business practices by entrepreneurs and managers?
• How do we make markets for training and consulting work better?
• How do we design and evaluate incubator and accelerator programmes?

Five things we have learned from research on Training Entrepreneurs.

Adapting polarisation research to the political realities of LMICs – Cesi Cruz & Horacio Larreguy

The majority of causal evidence on political polarisation still comes from the US or other high-income democracies, limiting relevance for settings where the risks of polarisation may be most severe.

Cruz and Larreguy highlight that this is not just a gap in geographic coverage, but in conceptual relevance. Interventions designed to counter polarisation in the Global North may not translate directly into settings with different political, social, and institutional constraints.

The authors also identify a need for evidence on the durability of interventions, i.e. long-run follow-ups to distinguish fading effects from habit formation.

Read the full VoxDevLit here: Political Polarisation.

More evidence is needed on the following areas related to Female Labour Force Participation – Rachel Heath

• On the key contextual factors that lead similar policies/interventions to have very different impacts across countries.
• On the impacts of having women in supervisory, managerial and leadership positions within firms in low-income countries.
• On improving women’s safety, and reducing the harassment they face, in public places and the workplace.
• On the gendered effects of global trends including AI, climate change and the rise of remote work.
• To establish which aspects of the intra-household bargaining process are most important for women’s labour supply.
• On which interventions boosting women’s economic empowerment would represent a profitable investment for private firms, the contexts in which they succeed, and more generally, represent high benefit relative to cost.

More evidence and key questions on Female Labour Force Participation.

What remains to be learnt about reducing deforestation – Francisco Costa & Allan Hsiao

There are a number of opportunities to build on the existing evidence base and establish how we can lessen the potential trade-off between economic development and environmental protection:

• By using new data to study firm choices and outcomes at scale.
• By developing new policy tools to ensure green trade and international coordination.
• Emphasising and evaluating politically feasible regulation.

Other important gaps include: the lack of work on the Congo rainforest, the second-largest tropical rainforest in the world; quantifying impacts on biodiversity; and active vs passive forest regeneration.

Read the full VoxDevLit here: Deforestation.

What don’t we know about the causes, consequences & responses to organised crime – Santiago Tobon & Maria Micaela Sviatschi

• Shifting adolescent expectations and behaviour at scale: How do peer mentoring, efforts to strengthen cooperative norms within schools, early deterrence interventions, labour-market simulations, and school-based vocational tracks impact recruitment, norms, perceived earnings, social status, and risk tolerance?
• The mental health burden of exposure to crime: Embedding trauma modules in household surveys and randomising therapy through schools or clinics would reveal both prevalence and cost-effectiveness.
• Intergenerational effects: Administrative data that follows children of gang-affected households into adulthood would clarify whether criminal shocks propagate disadvantage.
• Courts: Which legal reforms raise conviction certainty without eroding due process?
• Measure what matters: Researchers ought to measure the outcomes that residents experience – mobility, extortion payments, recruitment, perceptions of authority – rather than rely solely on homicide counts.
• Structural models to get at trade-offs: When one policy reduces violence, does extortion increase allowing criminal organisations to grow stronger? When recruitment falls, do gang wages rise, tempting other youth to join?
• Gender and organised crime dynamics: How gender relates to recruitment, roles, exposure to violence, and communities’ experience.

Finally, the authors highlight the lack of research on organised crime in West Africa, where qualitative reports suggest that criminal organisations are an increasingly pressing concern. 

Read the full VoxDevLit here: Organised Crime.

Key unanswered questions on Sanitation Infrastructure – Britta Augsburg, Andrew Foster & Molly Lipscomb

The team identifies the following areas as key priorities for future research and policy:

• Procurement and cost reduction strategies for expanding sanitation networks – particularly in urban settings with rapidly growing populations and informal settlements.
• Climate resilience of sanitation systems, as the frequency of extreme weather events increases.
  • Both isolated and networked systems need to be re-evaluated under climate change risks such as flooding and migration.
• Complementarities with other interventions, particularly nutrition, hygiene, and infrastructure services, to maximise the health and welfare returns to sanitation investments.
• New approaches to addressing market failures, including centralised market platforms, innovations in service delivery, and pricing mechanisms to encourage uptake and reduce costs.
• Better targeting of subsidies and behavioural interventions, recognising the heterogeneity in household constraints and community-level externalities.
• Understanding government failure and political economy constraints.
  • The limited rollout of large-scale sanitation infrastructure may reflect not only fiscal or technical challenges, but also political under-prioritisation.
  • Future research could explore how government priorities align or misalign with household sanitation needs, and what institutional reforms might better align public spending with public health returns.

Read the full VoxDevLit here: Sanitation Infrastructure.

Evidence gaps on Electricity Infrastructure – Rachel Heath

Given significant recent advances in data availability, remote monitoring technologies, and machine learning, there is great potential for future study of electricity infrastructure and its impacts.

Mini-grids: Where, and under what circumstances, are mini-grids a viable electrification solution that can bring economic benefits to the communities they serve?

Systems: This VoxDevLit highlights the importance of considering electricity interventions more comprehensively, rather than studying them as supply or demand side processes in isolation. Focusing on these system-wide, and sometimes indirect effects, of interventions is a major opportunity for future research.

Renewable Energy: Research considering the challenges and opportunities posed by renewable energy in LMIC settings, to help these countries avoid scenarios where reliability issues are exacerbated by renewable integration.

New technologies: As technological advances in metering (devices enabling real-time monitoring), and automated demand response systems, are now being deployed in developing countries, researchers have an opportunity to measure their impacts.

Learn about economic research on electricity.

What do we have to learn about the causes and consequences of informality in developing countries? – Gabriel Ulyssea

On the firms’ side:

• Does informality work as a stepping-stone for entrepreneurs with high-growth potential but who might be constrained, e.g. by credit constraints?

On the workers’ side:

• We need a deeper understanding of the determinants of workers’ choice/allocation between formal and informal jobs, what determines their permanence and evolution in either, as well as the main tradeoffs they face.
  • Regarding the tradeoffs, for example, we do not know how much workers value the greater job security provided by formal employment relative to informal jobs.
• As in the case of firms, we do not know how much informal jobs represent a stepping-stone for younger workers versus the extent to which there is an “informality trap” that makes future transitions into formal employment very unlikely.
  • Related to this point, there are still very few studies that investigate the life-cycle dimensions of informality.

And on the housing side, new to Issue 2:

• Key open questions remain about upward and intergenerational mobility, human capital accumulation, and labour market dynamics for slum residents.
• An understudied but crucial aspect of understanding informal housing is the role of the formal housing supply and the “missing market” for low-income formal housing.
• Integrating research on slums with research on urban labour markets, transportation, and firm location choices will be important to understand the trade-offs that slum residents face.

Read evidence on the informal sector in this VoxDevLit.

Future research on Refugees and Other Forcibly Displaced Populations – Sandra Rozo

Research on forced displacement is expanding rapidly, Sandra Rozo identifies several critical areas which would benefit from further investigation:

• The impact of displacement on origin communities, where initial evidence points to severe developmental consequences - understanding these dynamics is crucial for mitigating long-term harm and fostering recovery in affected regions.
• Displaced children are uniquely vulnerable - developing targeted strategies to protect their well-being and prevent the emergence of a "lost generation" is essential.
• There is a pressing need for more research on the effectiveness of programming for IDPs and returned migrants.
• Gender-sensitive approaches also warrant greater attention, particularly to address the unique challenges faced by displaced women and girls, such as heightened risks of violence and systemic barriers to accessing essential resources.

What we have learned about refugees and forcibly displaced populations.

Particularly fruitful avenues for progress on international trade research – David Atkin & Amit Khandelwal

1. Combining multiple sources and types of data – including leveraging advances in digitisation, tracking technologies and text analysis – to provide a more complete understanding of the effects of trade in the developing world.
2. Focus on specific industries where tailored firm surveys and niche datasets can overcome at least some important measurement concerns.

Crucial to these approaches is data that captures the institutional complexities, market failures and distortions of the particular setting.

Issue 2 identifies a key new set of questions given recent changes to the international trade landscape, which has seen supply chains moving away from China, countries diversifying their sourcing to improve resilience, and new environmental requirements added to trade policies. The critical guiding question given the state of the world: how can developing countries benefit rather than lose from these changes?

More questions and evidence on international trade and development.

Evidence gaps in our understanding of Foreign Direct Investment and Development – Stefania Garetto, Nina Pavcnik & Natalia Ramondo

Our VoxDevLit on FDI has chapters on how FDI intersects with key aspects of the economy, each of which highlights open areas for future research.

• The macroeconomy
  • Quantitative models currently do not incorporate knowledge spillovers from multinationals for local firms or workers, or allow for multinational engagements that involve both horizontal and vertical linkages.
  • More work is also needed to understand the role of MNEs in transmitting shocks across countries and the synchronisation of international business cycles.
• Linkages with local firms
  • What drives the impacts of links between MNE’s and domestic competitors? Is this driven by the firm’s foreign ownership or merely by it’s large size?
  • To what extent do MNE affiliates actually rely on local supply chains? Can policymakers intervene to increase this reliance?
• Environment
  • Substantial evidence shows that foreign-owned plants emit less – we need to learn more about the mechanisms behind this result.
  • Evidence is mixed on whether MNEs are reallocating their dirtier production to countries with laxer policies.
  • A key open question is how new companies from China and India might play a role now for other low-income countries, having benefited from FDI themselves over the years.
• Natural resources
  • Because a small number of large MNEs often operate in natural resource sectors, it is important to continue to consider the role of MNEs for host economies in these sectors.
• Labour markets:
  • Research needs to incorporate the connections between MNE activities and the informal sector.
  • It is important to consider how MNE jobs influence economy-wide employment and wages in the presence of unemployment or underemployment.
  • Finally, randomised control trials that help us further understand the incentives or collaboration practices that encourage enforcement and compliance with labour market standards are promising.

Read Foreign Direct Investment and Development.

Promising avenues of future research on microfinance – Jing Cai, Muhammad Meki & Simon Quinn

Given the evidence that the effects of microcredit vary across borrowers, and that different contractual forms can work in different contexts, an open research question is how microcredit can become more flexible/tailored while retaining the advantages of its more basic forms – such as transparency, simplicity, the ability to keep costs low through group disbursement and collection, and the harnessing of social capital to promote repayment.

Different microcredit contracts clearly have different uses for different borrowers – in particular, some microcredit provides for business expansion, while much goes to consumption. There may be valuable contractual innovations in designing microfinance products that more effectively provide for business investment – and conversely, other products that more directly serve a consumption need (for example, products that are intuitive to the borrower and that incorporate appropriate consumer protection).

• This seems like one particularly promising avenue of ongoing and future research: testing ways in which novel contractual arrangements can more effectively serve specific groups of microfinance clients, many of whom use microfinance for very different purposes.
• In this regard, digitisation and novel financial technologies hold significant promise for enhancing access to financial services (both in the sense of expanding microfinance access, and in the sense of allowing for more flexible performance-contingent microfinance contracts).

Evidence from economic research on microfinance.

Key questions on Taxation and Development – Anders Jensen

This VoxDevLit was organised around three important dimensions of taxation in low- and middle-income countries and identified key future research areas within each.

Constraints on effective taxation and enforcement:

• How would an optimal tax design combine enforcement interventions and reforms to the statutory schedule?
• New technologies and digitisation may enable governments to enhance enforcement, but this will depend on the quality and coverage of new data sources.
• What are the real efficiency costs of commonly implemented tax instruments?

Tax authorities:

• Systematising tax processes to follow transparent rules, not discretion, holds promise, but also presents challenges: while systematising the collection process, local officials are likely to gain more information and knowledge on existing and potential taxpayers.
  • What policies can reduce the abuse of this information by officials?
• Communication strategies can enhance perceptions of fairness, but understanding whether, and how, this might impact compliance by taxpayers requires more evidence.

Tax equity:

• The co-existence of formal and informal sectors means that some agents in the economy may benefit from tax increases – understanding who benefits is an understudied area of research.
• The economic incidence of any tax reform will depend on many factors, including the market structure and relative market-power of the different relevant agents; as these factors will vary across settings, there is a strong need to build evidence on economic incidence ‘from the bottom up’, across a variety of environments.

More on the growing body of research on taxation and economic development.

Boosting Agricultural Technology in Africa – Chris Udry & Tavneet Suri

• How should we provide incentives to either the public or private sector for the development of new agricultural technologies that are locally customised?
  • Then, how should we provide incentives for experimentation with these technologies?
• Are improvements in agricultural technology and productivity the most useful way to raise the standard of living and create a path out of poverty, or should the focus be on investments in the nonagricultural sector?
• Can the integration of rural and urban markets in Africa provide better incentives to farmers?
• Is there a way to scale down large-scale infrastructure investments to get more irrigation across Africa?
• What is the role of the state in agriculture? How extensive is crony capitalism in agriculture?
• How are policy priorities or large infrastructure investments decided specifically in agriculture?
• This VoxDevLit identifies many different constraints to farmers, especially in terms of labour, land markets and the environment, but more work is needed to understand which constraints impact which farmers.
• Lots and lots to learn on climate change!
  • How is climate change likely to affect agricultural output in Africa?
  • How will farmers and entire agricultural systems in Africa adapt to these changes?
  • Are there technologies that can help climate adaptation in agriculture? If so, how do we trial/test and adapt them to local contexts?    

More evidence and key questions on agricultural technology in Africa.

More evidence is needed on the following areas related to Female Labour Force Participation – Rachel Heath

• On the impacts of having women in supervisory, managerial and leadership positions within firms in low-income countries.
• On improving women’s safety, and reducing the harassment they face, in public places and the workplace.
• On the gendered effects of global trends including AI, climate change and the rise of remote work.
• To establish which aspects of the intra-household bargaining process are most important for women’s labour supply.
• On which interventions boosting womens’ economic empowerment would represent a profitable investment for private firms, the contexts in which they succeed, and more generally, represent high benefit relative to cost.

More evidence and key questions on Female Labour Force Participation.

Evidence-based policy on Barriers to Search and Hiring in Urban Labour Markets – Stefano Caria & Kate Orkin

There is very little evidence that quantifies the potential negative employment impacts of job search interventions on individuals not receiving the intervention – the so-called displacement effects. It is plausible that in the short-term these can be large, since firms in LMICs do not appear to have many unfilled vacancies. (Displacement effects are also a potential issue for training interventions.) Other questions include:

• Do job search interventions generate productivity gains for firms?
• What are the impacts of job search interventions at scale?
  • They could increase labour demand, as identifying talent is identified as a key barrier to growth. However, increasing search could generate congestion, and signals about workers skills could harm low-skill workers.
• How should job search programmes be targeted?
  • Initial evidence suggests that job search interventions are most effective for groups with weaker expected labour market outcomes, so would there be gains from more effective targeting?

Evidence and key questions on Barriers to Search and Hiring in Urban Labour Markets.

One key constraint holds back research on land transport infrastructure - Marco Gonzalez-Navarro & Román David Zárate

• More comprehensive and better cost data is needed.
  • Cost-benefit analyses, which require accurate cost estimates, help provide policymakers with better guidance regarding how much to build and the relative merits of different infrastructure types. The lack of cost data means current analyses ultimately focus on benefits rather than more relevant benefit to cost ratios.
• What are the impacts of climate change on infrastructure provision and placement decisions?
• How to develop optimal transportation networks across space?

Evidence and key questions on Land Transport Infrastructure.

Areas where additional work is needed to better understand bureaucracies in low- and middle-income countries – Guo Xu

It is difficult to measure the output of bureaucrats and bureaucracies. 

• Future work could use remote sensing data and economic censuses to proxy changes in GDP and other outcome variables of interest.
• Administrative data capturing all the bureaucrats in a region or country, often linked to monitoring of their actions, is increasingly being used. 

Gathering this type of economy-wide micro-data will open possibilities to evaluate system-wide reforms of what bureaucrats do and how they affect these outcomes, which is often what governments are interested in.

Understanding whether and how bureaucrats can innovate and adapt to future challenges is also crucial:

• The COVID-19 pandemic has uncovered large heterogeneity in the capacity of bureaucracies to respond.
• We need to identify what characteristics of bureaucracies are needed to respond to key future challenges, such as climate change, which may be different to those faced in the last century.
• How innovative are bureaucracies in using up-to-date knowledge to face current and future challenges?

What we do know about building effective bureaucracies.

Possible directions for research that can contribute to our understanding of adaptation to climate change – Namrata Kala

• What are the long-term impacts of climate change and the mechanisms facilitating adaptation?
• What actions and policies empower individuals to build sustainable resilience?
  • There are no single strategies that lead to complete preparedness or recovery. Reflecting on the success of graduation programmes in pushing people out of poverty traps, can composite interventions build climate resilience?
• In many settings, take-up of weather insurance remains low. What is the optimal design and outreach approach for climate insurance in different contexts?
• A deeper understanding of firms’ adaptation to climate change could include research on areas such as the use of weather insurance, management practices and organisational structures, location of work, and research and innovation in building firm’s capacity in weathering climate shocks.
• Informality adds to the vulnerability to climate shocks in developing countries and more work is needed to understand how these populations can be covered despite climate change.
• What conditions can align political motives with climate resilience?

Economic research on climate change adaptation.

Research on mobile money is growing, research on the new innovations building on mobile money has been lagging – Tavneet Suri

Although mobile money may seem revolutionary, aside from the dramatic adoption, it is far from revolutionising the role of financial markets or cash in low- and middle-income economies. This opens up a number of research areas:

• Should these economies actually become cashless, or close to cashless? How would this be accomplished?
• Will the banking system be the primary venue for this transformation, and, if so, what sets of products and services will be needed to accomplish this?
• What will encourage financial market transformation in these economies?
• Will mobile money be the first stepping stone toward new financial markets and transactions in these economies?
• Will it encourage broader, better-integrated, secure platforms for transactions?

Learn about the expansion of mobile money service providers and accounts over the past decade.