An Interview with Ahmed Mushfiq Mobarak, Professor of Economics – Yale University

Ahmed Mushfiq Mobarak is a Professor of Economics at Yale University with concurrent appointments in the School of Management and in the Department of Economics.

Mobarak is the founder and faculty director of the Yale Research Initiative on Innovation and Scale (Y-RISE) . He holds other appointments at Innovations for Poverty Action, the Jameel Poverty Action Lab (J-PAL) at MIT, the International Growth Centre (IGC) at LSE.

Mobarak has several ongoing research projects in Bangladesh, Brazil, Chile, Kenya, Malawi and Sierra Leone. He conducts field experiments exploring ways to induce people in developing countries to adopt technologies or behaviors that are likely to be welfare improving. He also examines the complexities of scaling up development interventions that are proven effective in such trials. For example, he is scaling and testing strategies to address seasonal poverty using migration subsidies or consumption loans in Bangladesh, Nepal and Indonesia. His research has been published in journals across disciplines, including Econometrica, Science, The Review of Economic Studies, the American Political Science Review, Proceedings of the National Academy of Sciences, and Demography, and covered by the New York Times, The Economist, Science, NPR, BBC, Wall Street Journal, the Times of London, and other media outlets around the world. He received a Carnegie Fellowship in 2017 .

Mobarak is collaborating with the government of Bangladesh, NGOs and think-tanks such as BRAC and BIGD, the major Bangladeshi telecom providers, Innovations for Poverty Action, UNDP, other economists, epidemiologists, computer scientists, and public health researchers to devise evidence-based COVID response strategies for Bangladesh and for other developing countries. The approach and results have been covered by BBC, Foreign Policy, New York Times, Washington Post, Vox, and media in India, Pakistan, Bangladesh, Nigeria, Austria, Sweden, Denmark, among others. The work is supported by the Bill and Melinda Gates Foundation, Givewell.org, the Global Innovation Fund, and Yale Macmillan Center. Professor Mobarak’s research work can be viewed on his website, “https://faculty.som.yale.edu/mushfiqmobarak/ ”and his views can be viewed via his Twitter Account. He tweets @mushfiq_econ .

Covid19 :

Q1-) When the Covid-19 pandemic struck the developing world, estimates were that the starvation and food insecurity problem will cause a havoc. According to the World Food Program, a United Nations agency. Altogether, an estimated 265 million people could be pushed to the brink of starvation by year’s end which meant an additional 130 million people. Although we are yet to reach the end of the year and the poverty numbers did get worse but the starvation numbers didn’t get as bad as the projections. What’s the reason behind this?

Professor Ahmed Mushfiq Mobarak(Prof AMM) - When people face risks – whether from a virus or from food deprivation - they are forced to adapt and adjust to do what’s best for their families, within the boundaries of the constraints they face. The quick projections made by international organizations may have paid a lot of attention to constraints without adequately accounting for citizens’ adaptation. In low and middle income countries, where a large share of the population work in the informal sector, analysts need to collect direct survey data to understand how people’s lives are changing, and in what ways they adapt. It’s difficult to make accurate projections based on prior administrative records.

Q2-) Experts said that the West’s policy of going for a complete lockdown was fine as they have the resources to support their people. However Sweden did not go for a lockdown and not did that save livelihoods but the potential second wave in the country has been much lighter than those countries in the west who went for the a complete lockdown during the first wave. Does this mean that a complete lockdown was a flawed policy economically and health wise and should the developed countries go for a lockdown once again.

Prof AMM - The “western” countries that fared the best overall were the countries that had clear, consistent strategies, and were able to get the virus under control. We have now learned from data in the U.S. that it wasn’t necessarily lockdowns that was keeping rich people at home, it was the fear of the virus. When the rich remain at home and don’t spend money, that hurts all the workers in that community. To get the local economy going, you really needed to get the virus under control, to make everyone feel confident and comfortable. People think most carefully about the risk they are facing, and therefore these questions about lockdown vs not are somewhat misplaced. We should really be thinking more effectiveness of virus control, consistency in mask-wearing, people’s adherence to social distancing norms.

Q3-) How do you rate Pakistan’s response to Covid19 where the state did not go for a complete lockdown and also gave cash handouts to the most vulnerable fraction of the society. On the other hand India that went for a complete lockdown is not only unable to save lives but livelihoods as well. What are the lessons these two Covid19 responses offer to the world especially to the developing world ?

Prof AMM - We shouldn’t try to over-interpret the comparison between two countries. These are only two data points, and we cannot confidently make any inferences about the relative effectiveness of different policies from two data points. We know that getting money in the hands of poor and vulnerable people quickly is critical to allowing people to adhere to social distancing guidelines. We also know that implementing policies in a haphazard way without thinking about the constraints citizens face and how they would react such policies is not smart. We saw this in India where a sudden lockdown – including a shutdown of all public transport – led to immense suffering for migrant workers and their families. The United States also announced a travel ban from Europe abruptly without coordinating with European governments or giving travelers adequate opportunities to react and adjust. This led to tremendous crowding at airports, and likely some super-spreader events in terminals.

The lesson here is that disjointed, disorganized decision-making without proper implementation and planning were government and leadership failures in both India and the United States thatled to unnecessary suffering for their citizens.

Productivity :

Q4-) Educational attainment among females has increased rapidly as a result (from 22% in 1986 to 42% in 2007 according to UNICEF, and the rate now surpasses male education). What are the key factors that encouraged girls to enroll in schools and how has this increase helped Bangladesh as an economy as well as a society?

Prof AMM - When it comes to education decisions, low enrollments in schools could either be a problem either on the demand side or on the supply side.

On the supply side, there may not be enough schools, or the schools could be too far away, or they may not have enough teachers. These are supply side constraints that need to be addressed through investment in schooling or teachers.

Another possibility could be that parents decide on whether to educate their kids based on the economic returns of education. In an environment, where educating your child is costly (and cost considerations often include opportunity costs because the child could be working on family farm instead of attending school) and these costs are not matched by higher economic returns or wages, parents are likely to hold back investments in education. This is a demand side issue.

Demand for education is therefore dependent on the returns to education in the economy. Based on data, it seems that supply side problems in education have largely been addressed. Nationwide surveys in India indicate that over 95% of kids live within walking distance of a school. If gaps in education continue to be observed, the low take up of education is probably not going to be solved by increasing the supply for education.

One can use the example of Bangladesh to identify some of the underlying causes that are driving low education take up. The period of impressive growth in education in Bangladesh that you just noted, was accompanied by the growth in the country’s garment industry. The garment sector was essentially non-existent in Bangladesh up until 1979 but since then it has reported an impressive growth of about 17% per year over a period of 30 years. The garment sector has experienced exponential growth in the past four decades and now garment exports account for approximately 90% of Bangladesh’s goods exports.

Now the question is, how is growth in the garment sector linked to the growth in education? The garment sector is a large employer for Bangladeshi women – nearly 50% of the employees in the garment sector are women. As this sector grew, employment opportunities were created for women in a country where they traditionally did not work outside their homes, thereby increasing returns for education. This is because, securing a job at a garment factory required basic literacy and numeracy and the promise of a higher wage, motivated parents to educate their children.

Our research show that the increase in school enrollment was observed in areas where factory growth had occurred and the prospect of a higher wage was the largest driver of increases in school enrollment for girls. In fact, we documented that not only did parents put their children in school but that parents were more encouraged to put their daughters in school relative to their sons after this growth in the garment sector. We saw that parents are more likely to enroll their five to ten-year-old girls in schools because of their potential to earn as a garment factory worker when they turn eighteen.

As part of our study, we also tracked the girls who were enrolled in schools and observed that girls were less likely to be married by the age of 16-17 in areas that experienced a growth in factories. Additionally, factory growth also delayed childbearing and decreased fertility rates.

The trends observed in Bangladesh are in line with global patterns – one of the major reasons driving lower fertility rates in the world is increased labor force participation of women.

Q5-) According to the World Bank, Bangladesh has one of the world’s fastest agricultural productivity growth, at around about 2.7% and, according to an estimate, Bangladesh has the highest cumulative growth in agriculture per person at about 70%. How did Bangladesh achieve this?

Prof AMM - When we talk about agricultural productivity growth anywhere in the world in the last 50 years, it has mainly been driven by one factor, which is technology. Around the world, Asia has increased agricultural productivity a lot more than Africa, while South America has increased its productivity a lot more than Asia. Technology can mean a variety of things. For example, there are new and improved varieties of seeds. There are also fertilisers and irrigation techniques which help farmers to become less rain-dependent. There’s also mechanisation.

Mechanisation requires larger farms because it doesn’t make sense to invest capital in smaller farms. This is why the largest agricultural productivity growth has been seen in Brazil, where the entire sector is mechanised and resembles the agricultural sector of developed countries.

In South Asia, the biggest factor that has led to the increase in agricultural productivity has been improved varieties of seeds. These are called the green revolution technologies.

In Bangladesh when I was growing up, we were worried about food insecurity. We had about 100 million people that we as a country were struggling to feed. This, fortunately, is no longer a concern due to the boost in agricultural productivity.

The existence and take-up of green revolution technologies helped us eradicate much of Bangladesh’s food insecurity and is one of the reasons Bangladesh’s growth in agricultural productivity has outpaced Pakistan’s.

The quick spread of green revolution technologies in Bangladesh was driven by a couple of of factors. First, the high [population] density allowed for information to disseminate quickly across farmers. Farmers in our region learn by talking to other farmers in their neighbourhood and therefore high population density facilitated more frequent communication and allowed the technology to spread quickly.

The second factor that has been a part of Bangladesh’s story is that from our early history, we have had an impressive network of NGOs that the government has allowed to operate relatively freely. These NGOs have run several programs that focused on technology dissemination.

Q6-) How has the deployment of resources via microfinance (Grameen Bank) helped Bangladesh in terms of economic empowerment, and how can this be emulated for developing countries — especially Pakistan?

Prof AMM - The role of microfinance, particularly in regard to Grameen Bank, has been overplayed, perhaps due to the Nobel Prize. The story that is told is that everyone is an entrepreneur and these small loans allow people to unleash their entrepreneurial spirit. The story sounds nice but the data suggests that the vast majority of people who take these loans do not start businesses.

The reason is that most people don’t actually want to be entrepreneurs; instead, they want jobs. However, what’s useful about microfinance is that people need credit. They might not need credit to start a business or be an entrepreneur, but they might need credit for consumption-smoothing, or to invest in something that is costly, for instance, a refrigerator or a television -- poor people need entertainment too.

Over here in the United States, I can go to the local Best Buy and, depending on my credit score, purchase a television for monthly instalments. In fact, this is a big part of their business model. However, it is difficult to find electronics stores in rural Bangladesh offering provisions for monthly instalments, so people can use microcredit to make such purchases. While this explains why people are happier, it does not explain development. Moreover, the story about microfinance being the key input for entrepreneurship is probably not the right one but many NGOs in Bangladesh that were founded on the premise of providing loans have developed a dense network of operations around the country.
Given their presence in every district it has became easier for us to then run other development programmes using that infrastructure. Thus, while while those NGOs might have started with microfinance, they facilitated the implementation of other programmes aimed at improving health, sanitation, education, etc.

Q7-) It is said that one of the key factors behind the cost of the labour being one of the lowest in Bangladesh is underpaid women workers. What measures are being taken to address this issue ?

Prof AMM - We must be careful about how we use the word “underpaid”. Using underpaid implies that women in Bangladesh are being paid less relative to something else, although I am not clear on what the something is. As the garments sector grew, we increased women’s participation in labor force and allowed them to earn a wage, when previously they were at home engaging in productive labor but not earning a wage. This means there was an incredibly productive workforce that was just being underutilised.

The garment sector that differentially preferred women or gave them labour market opportunities was able to tap into this underutilised resource. It is not surprising that these wages are going to be lower because they were partially set by what the outside options for the workers are. In short, it can be said that the wages stayed low because of an abundance of women workers who didn’t have many other high paying options.

However, given the inequality in the country and in the world, it is incumbent on Bangladesh to think about the ways in which we can give these workers a larger share of returns being generated by the global trade of textiles. However, in trying to fix this we run into a demand supply mismatch. Capital is scarce while labour is not, especially female labour, and most of the returns go to factory owners. We have seen in worker history that labour rights do not come easy – it requires organization and regulation. In Bangladesh, a lot more needs to be done to improve labor rights not only when it comes wages but also when it comes to ensuring safe working conditions.

Bangladesh has had several unfortunate incidents of fires or accidents in [garment factories]. The most famous one is the Rana Plaza disaster. It was clear that factory owners were not paying sufficient attention to workers safety and health.

Q8-) Pakistan has had many boom and bust cycles in the past due to unsustainable deficits whenever the growth rate goes up. The increase in growth is mainly spurred due to consumption. The current government plans to spur the growth and revive the economy via large-scale housing projects and construction. What’s your view on this?

Prof AMM - I should be careful in first stating that I am not an expert on Pakistan’s economy and as a result I am answering this question not with any intimate knowledge of Pakistan’s economy, but with my intuition about economics. If I had to pick a sector to prioritise in order to generate growth, construction wouldn’t be it, because of laws surrounding land ownership, property taxes, and the fact that it is easier for us to park resources in property [if the property taxes are not high].

In South Asia, we probably have too much emphasis on investment in properties that may not be as productive as investing money in the capital market [like in the US]. Investing money in capital markets enables capital to flow to companies that are innovative and competitive and are creating new markets.

In contrast, constructing a building is not innovative nor does it create a new market. Thus, we need to address why is such an attraction towards investing in properties among ordinary Pakistanis.

Is it because tax laws favor people putting money in apartments or houses instead of other productive investments? Over-investment in properties can often create asset bubbles which precipitated the 2008 financial crisis in the U.S.

Q9-) Rapid growth enabled Bangladesh to reach the lower middle-income country status in 2015. In 2018, Bangladesh fulfilled all three eligibility criteria for graduation from the UN's Least Developed Countries (LDC) and is on track for graduation in 2024. This will eventually change the tariff-free concessions it receives from the European Union. How is the state planning for it?

Prof AMM - I think that in the changes that have occurred, the role of the state has been somewhat limited. The growth has not driven by the heavy handed involvement of the state, however in certain circumstances they have provided an enabling environment that has facilitated the growth of industries.

It is important for the state to provide general infrastructure that is required for firms to grow. For many years, especially prior to the current government’s administration, one of the key failures of the state was lack of investment in energy. Access to power is a basic infrastructural need for all firms and Bangladesh had some of the highest rates of power outages.

Firms in Bangladesh were reporting more than one power outage on average per working day but in the last decade, the government has made massive strides and invested aggressively in this sector, thereby addressing the issue of power outage and creating an enabling environment for firms to succeed.

Moving forward the government needs to continue identifying constraints that bind firm growth and find strategies that could relax these constraints and promote firm growth.

For instance, Bangladesh continues its struggle to devise a comprehensive urban planning policy that makes its cities more habitable. Cities have been engines of growth for the country -- over a quarter of Bangladesh’s GDP is produced by the Dhaka Metro Area. Still, lack of careful planning has made it one of most unlivable cities in the world. Dhaka struggles with traffic congestion, lacks space, etc. and these factors can be impediments to growth. Lack of easy access to the market or ports can inhibit investment and thus the state will need to invest in comprehensive city planning that allows for easy movement within and across cities.While there is some progress, there is room for improvement.

To your point about government income, country status and how it changes concession — when Bangladesh’s status changes,we are likely to lose some concessions that are designed for low income countries. But this news could also attract new investors who are drawn to Bangladesh’s sustained and buoyant growth.

Q10-) Bangladesh has a healthy savings-to-GDP ratio due to which it is able to fund investments via domestic savings. How was the trend of savings promoted so successfully in the country despite cultural and conservative religious views?

Prof AMM - I think that there are many factors but let me just focus my answer on one of them. Migration is an important component of Bangladesh’s economy. We have already talked a lot about ready-made garments but Bangladesh’s second largest export is human resources for which people send remittances back to the country..

There are two types of migration: internal migration and international migration. Internal or seasonal migration is often undertaken by poor agriculture workers who move to the cities to generate additional income during lean agricultural seasons. However, Bangladeshis, just like Indians, Pakistanis, Sri Lankans and Nepalese also undertake international migration. Thousands of Bangladeshis travel to the richer labor markets of Middle East, Singapore, Malaysia for work.

When it comes to international migration, migrant workers know that they will have to come back home at some point, because there is no path to permanent residency or citizenship, in countries like the United Arab Emirates or Malaysia.

This means that migrant workers have an opportunity to spend some of their most productive years abroad to live cheaply and generate some savings. These savings are sent back to Bangladesh in the form of remittances that is often used to pay back debts of migrant workers or invest in land. This is an important component of our national savings.

Climate Change :

Q11-) Countries like Pakistan and Bangladesh contribute very little to global warming but they’re expected to be hit the most by climate change. According to the World Bank, climate change poses a huge threat to Bangladesh’s agricultural productivity growth. What needs to be done in terms of policy making to avert this disaster?

Prof AMM In both Pakistan and Bangladesh, the weather patterns might change drastically as a result of climate change and it is unclear how this will affect agriculture.

However, Bangladesh is likely to be hit more severely as a result of climate change relative to Pakistan. Bangladesh is a low lying delta because of which some portion of its landmass is underwater for parts of the year.Given that rising sea levels is one of the first observed impacts of climate change Bangladesh is at high risk of going underwater as a result of climate change.

Climate change is a global issue that requires collective action. Therefore countries like Pakistan and Bangladesh will need to devise strategies that focus on promoting collective action across the globe.

There is widespread recognition in the world — although not universal — about climate change and its impact. Scientists have near-consensus on the reality of climate change and countries that are most likely to be affected now need to get together and speak in unified voice to drive for a policy change.

Political Economy :

Q12-) Do you feel that ever-evolving regional politics may have an impact on Bangladesh’s industrialisation? Currently, its textile industry is said to be dependent upon Indian investments and raw material. How will that change if it moves towards a more China-oriented foreign policy?

Prof AMM - Both China and India are important sources of raw materials for Bangladesh, however, it is unlikely that Bangladesh will face sudden supply chain issues moving forward. When it comes to political economy and governance, scholars in Bangladesh have spoken about the ‘Bangladesh miracle’ i.e. the high rate of growth the country has experienced in spite of poor governance.In the last 20-25 years, Bangladesh, despite experiencing high rates of corruption, immense political instability, and frequent regime changes, has recorded sustained economic growth. The country has become more stable and there is less investor uncertainty about what is going to happen. However, in recent years, the country has also begun shifting a little towards religious fundamentalism and there was a major terrorist incident a few years ago that took everybody by shock. Since that incident the government has taken decisive action to combat domestic terrorism, however, in their effort to wipe out criminal activity, the government has drifted towards authoritarian rule that penalizes any dissent. This could be bad for economic growth.

Bangladesh needs to avoid the trap other countries with authoritarian rule have fallen into. Although the stability provided by a strong government can promote growth, authoritarian tendencies often do not incentivize innovation and countries struggle to expand their production possibility frontier.

Q13-) How can democracies achieve unified objectives that often authoritarian regimes do, while not losing the essence of an open society?

Prof AMM - This can only be possible if there is freedom of expression and the press is allowed to question decisions. These are important tenets of democracy that are not only required for people's well-being but are also essential for supporting a vibrant, innovative society that is capable of reaching its potential.

Policy Making and Governance :

Q14-) How should research insights be linked with state-level policy formulation in South Asia so that the policies are efficient?

Prof AMM - I think this is an area where we have really missed the trick all over South Asia except for India. There are two gaps here.

The first gap is in the education system and India seems to be one big step ahead. India’s education system has successfully created talent who can compete globally. There is not a dearth of talent from the Indian diaspora in the global academic and tech landscape. However, other parts of South Asia are struggling to produce this kind of talent.

Countries need to produce that kind of talent. Indians, from being in some of the leading institutions in the world, gain knowledge that they then apply to promote growth in India. This is especially evident when leading economists like Raghuram Rajan and Kaushik Basu have direct influence on policy design in the country.

In contrast, we have had problems in Bangladesh and Pakistan where not nearly as much talent has been produced. This lack of talent is evident in every sector – not just academia.

If you want to have a digital economy and if you want to have proper IT policies in order to grow certain sectors, it’s useful to not just have talent in economic departments. It is also useful to have talent in Silicon Valley, and Pakistan and Bangladesh are way behind India in this area. Globally India overtook the US at some point in IT service exports.

The second problem is self-inflicted because of historical and cultural biases or beliefs, which exist in all countries, including India. For example, one of the top macroeconomists in the world is Professor Atif Mian. He should be tapped to advise the Government of Pakistan on their economic policies. People all over the world, even in the US, listen to his advice, however Pakistan has barred him from their advisory council because of his religious beliefs. That’s a completely self-inflicted wound.

Interactions between government and academia can lead to net welfare gains and systems and processes need to be put in place to allow for this exchange of ideas. The Covid-19 crisis for instance, that has forced Bangladeshi government to connect to its diaspora talent and rely on them for advice on the economy, epidemiology, and medical sciences. I hope we can build on these connections to promote future growth.

Q15-) The World Bank is working on several projects of making local governance functional in Bangladesh. How can a functional local governance system help countries like Bangladesh and Pakistan?

Prof AMM - A lot of the key services that need to be provided to increase productivity have to be done at the local level. For instance, when it comes to education, making sure that kids in a particular sub-district are learning the right thing should not be centrally managed.

While designing policy, it is important to have a localized understanding of the constraints that prevent success. The local government is a body that is often in the right position to solve these problem given their proximity to the issues. However, local governments need to have enough capacity and capability to successfully implement programs.

In Bangladesh, local government capacity is quite limited. It’s a country that in terms of actual fiscal capacity, is not decentralised but we have managed to fill up the gap to an extent. A lot of services that are traditionally thought of as local government services — like health clinics for primary healthcare or education services — are provided by NGOs.

For example, think about BRAC, which is the world's largest NGO. It has a number of offices in almost every district in Bangladesh, so they provide services that you normally only associate with the government; they have schools, they have clinics. They are very cooperative and they are doing things that normally are in the remit of the government, but ultimately, who provides them is not so important as long as they get provided. So if you empower the NGOs, you can get great results.

As a result, in Bangladesh, I can go to NGOs in rural areas and get a lot of work done, whereas in Indonesia, when I tried to implement the same project, it was impossible to make progress without getting an official letter from the state.

Q16-) How to think about scaling up all the micro-interventions that have been done in randomised controlled trials (RCTs) in Pakistan or Bangladesh, such as the one you’ve done regarding improving the sanitation services? How can we scale them up in a highly undeveloped area such as interior Sindh?

Prof AMM - To be clear — RCT is just a statistical evaluation method, so we should separate it from what you can learn via interventions. When interventions get randomised, we get clean answers, but the fact that the RCT was done and something was randomised in my sanitation project may be a small part of the overall development story.

The key finding the sanitation project generated was that sanitation investment decisions appear to be interlinked. What people around you are doing has a big effect on your decisions and it's because social norms are very important for people.

For example, if everybody around you is going out [and defecating] in the open, there is no shame in you going out in the open. However, when norms around open defecation change, and the majority of people no longer go out in the open, the minority that continues to open defecate is motivated to change their behavior since now there is shame and stigma associated with open defecation.

That’s the insight, and what we learnt further is how the social stigma works. It's not [as though] people are looking at the richest people or the leaders of the community and following their behaviour. In fact, if relatively poor people invest in toilets, then it becomes very shameful for other community members to not have one. There is social influence but, it is more downward-looking than upward-facing.

Those are the types of insights that came out of the sanitation evaluation. It just so happened that it was done on the basis of RCT, which allowed us to convince the academic world. But even without RCTs, BRAC that simultaneously was running its own sanitation programme, generated similar insights.

After that study was done, I collaborated with BRAC to incorporate elements of social influence in programs that promote hygienic sanitation practices.

That’s how change occurs.

Some of the things that we learn from this style of research is that the demand side is also very important. Behavioural change is very difficult; it is not just about building toilets.Even that kind of insight allows the world to startthinking about behaviour change and information campaigns and designing sanitation program that have built in persuasion strategies.

In India, the problem in the past has been [that] the sanitation programme being run by the bureaucracy was confined to building toilets only and not sufficient attention was paid to behaviour change. That’s why it's researchers are now trying to identify how demand for hygienic sanitation can be created.

Hopefully, the government of Sindh can use these insights when it comes to policy design. Whether these insights are generated through RCTs or not is f secondary.

You have to go to the ground level and not just build infrastructure, but spread awareness if you want to scale up the findings of your RCTs in the sanitation services. The general problem with scaling up RCTs is that RCTs by definition are run in controlled circumstances on a small scale and therefore a lot of complexities that arise at scale are often not encountered.

It is important for us to systematically assess and think through what those complexities might be. In Bangladesh for instance, we ran these randomised control trials that covered around 75,000 people. The local politicians and union chairmen in that area started reacting to the intervention and wanted to come and give speeches and take credit. If they came and gave speeches, voters would start thinking that they had something to do with the implementation of the program.

This could change the relationship between the politicians and the voters, so it’s an example of complexityAs we scale up RCTs, it is important to think about what other changes might occur.

Economic Teaching :

Q17-) Recently, Angus Deaton released a paper about the diminishing role of RCTs. Will the RCTs remain a prevalent narrative of economics teaching in the US?

Prof AMM We should [remember] that within economics, there are different fields. RCT has played an important role in developmental economics, and I think that the [extreme response to Deaton’s] argument — that randomised control trials are the only way to get credible evidence and they should be put on a pedestal — has problems

But the strong reaction against the old RCTs — which is that these are useless and people are spending a lot of money doing useless things —is also incorrect. I do think that randomised control trials have, by increasing the quality and credibility of empirical work, pushed us forward but, as we have just discussed, they have limitations.

Just because something shows up as a positive result in a pilot scale randomised control trial, doesn’t mean that when we scale it up, people’s lives would continue to be affected the same way. This because the scaled up intervention will only generate direct effects that were measured in the RCT, but it also generate indirect effects.

Markets and prices may also change at the larger scale and lots of other people who are not part of the beneficiary list may start getting indirectly affected. Designing evidence-based policy therefore requires policy makers to think about both direct and indirect effects of interventions

Q18-) What are your thoughts on tribalism in economics? Initially, it used to be based on the positions on the role of the state or market or other actors; it has also increasingly become about the methodologies one uses in economics. What’s the reconciling pattern or point amongst economists?

Prof AMM - I think that the field and the discipline have become less tribal. The example I will use is in behavioural economics. Twenty years ago, behavioral economics was not mainstream at all and would elicit strong reactions. However, over time, fields such as public finance and development economics have integrated behavioral economics to make better predictions about human behaviour.

I think that the field is becoming less tribal by becoming more and more interdependent on various disciplines, thereby drawing methods and insights from one sub-field and bringing it into another.