Can civic education workshops promote electoral participation? New paper from Liberia

From a new paper by Eric Mvukiyehe (World Bank) and Cyrus Samii (NYU) on the impact of interventions promoting democracy and participation in elections


A field experiment in rural Liberia is used to study democratic participation in fragile states. Fragile states are marked by political fragmentation, local patronage systems, and voter vulnerability. To understand the effects of such conditions on democratic expression through elections, the experiment introduced new forms of interaction between rural citizens and third-party actors: (i) civic education and town hall workshops directed by non-governmental organizations in communities over nine months and (ii) security committees that brought rural community representatives into monthly exchange with United Nations peacekeepers. Civic education workshops increased enthusiasm for electoral participation, produced a coordinated shift from parochial to national candidates, and increased willingness to report on manipulation. A program combining the two interactions had similar effects. The security committees had negligible effects. Barriers to political information and voter coordination appear to be important but resolvable problems for elections in fragile states.

Extending financial services in Kenya – three personal stories

Innovations for Poverty Action and the Bill & Melinda Gates Foundation have a call for proposals out on innovative digital financial services.

In some ways, the financial lives of the poor are more complicated and mentally taxing than those of more well-off households. While the wealthy can often absorb a small miscalculation, for the poor it can be catastrophic. Add to that irregular or unpredictable income, and financial decisions quickly become difficult to manage. Well-designed products that help individuals successfully navigate the difficulties of daily financial life may be able generate new demand for financial products. 

I have three personal stories:

  1. The cost of ATM usage in Kenya: I am charged KES 33 on every withdrawal when I use my bank’s ATM and KES 77 when I use the ATM of another bank. That hurts. What’s more expensive is the distance to my home from the nearest ATM. Given Nairobi’s specific security situation, I wouldn’t walk the distance to my house with KEs 20,000 in my pocket. So I take a taxi that costs me KES 200 – 300. So the effective cost of withdrawing cash is KES 230, often higher = 1 – 1.5%. Sure, I could take a matatu or a boda-boda – both of which offer various other kinds of risks I will need to factor in. What can make this cheaper?
  2. Is Mpesa better? Every time I use online banking to transfer funds to my Mpesa account, I get charged KES 66. So it might actually be cheaper to withdraw cash from an ATM and walk to an Mpesa station and deposit the cash, except for the convenience of online banking. On the payments side, Mpesa is near-perfect when paying bills and other business establishments. But when transfering small sums of money (say, KES 500) to someone else’s phone, the charges can amount to over 10%. Hardly a good deal. When using Mpesa, what I am really paying for is convenience and security. But is it really a poor-friendly product?
  3. Consumption smoothing in Kenya as evidenced by the night-life buzz: The pay-day weekend and the weekend after are hectic – streets are crowded, bars are full. Relatively, the third weekend is pretty silent and the fourth weekend is dead. Nothing novel here, except a market gap for a smart savings product for my friends who can be nudged to plan better.

I look forward to research findings that make my life easier!

Can WASH deliver more than just sanitation?

This column first appeared on the Guardian’s WSSCC partner zone


The abysmal state of access to safe water and sanitation facilities in the developing world is currently a major cause for alarm; 580,000 children die every year from preventable diarrheal diseases. This is due largely to the 2.5 billion people around the globe who do not have access to safe sanitation. Not only can an effective WASH intervention save lives, it can also engineer changes in the social fabric of communities that adopt these behavioural changes. This points to a key attribute of a successful WASH intervention – that through these programmes, communities not only access a new service that improves their quality of life, but they also learn from being part of a concrete intervention that emphasises equity and inclusion.

Let me explain how. Safe sanitation is essentially ‘total’. In a community, even one family practising open defecation puts the health of other families at risk. Also, unsafe sanitation practices pollute local potable and drinking water sources in the habitations. Together, this can undo any gains from partial coverage of WASH interventions. This much is now widely accepted by sanitation practitioners around the world. However, there remains a serious challenge when it comes to the implementation of this concept.

When a community is introduced to a WASH-focused behaviour change campaign, there are often variations in the levels of take-up in different families. This could be because of several barriers – financial ability, cultural beliefs, education levels, etc. In response, external agencies have many options. They can focus more on families in their behaviour change campaigns, offer them material and financial support or incentives, or exert peer pressure (which may in some cases become coercive, etc).

However, the best approach – whether facilitated by an external agent or not – is for a community to devise a collective response. The issue should be framed as a collective action problem that requires solving for the creation of a public good. In many instances, communities have come together to support the poorest families – social engineering at its finest. At its best, recognising the needs of every member of a community will lead to a recognition of the challenges that the typically marginalised groups face. It is this recognition that could prompt a rethink of social norms and relationships.

On the other hand, the power of peer pressure can be effective. Where families that are able, but unwilling, to construct a toilet and switch behaviour, the initial take-up from other families has a strong demonstration potential. In societies with caste and class differences, this can be deployed effectively to highlight choices that threaten the public good.

Encouraging the development of shared norms and collective action is also a key aspect of determining the role of subsidies in WASH programmes. As research evidence from Bangladesh shows, subsidies could be effective when targeted at communities, instead of at individuals. Where it is possible to measure progress at the community-level, subsidies can be designed and delivered accordingly. This will encourage communities to take up WASH as they would approach say, the building of a road or a school.

This is no longer just a theory. Increasingly now, various organisations have documented such successes. For example, in multiple NGO-led programmes in eastern India and Bangladesh, local community-based organisations formed initially to tackle sanitation went on to engage in collective livelihoods activities. However, as with any other, this theory too should be put to test – evaluated at different sites and for different approaches. Currently, we are not sufficiently focused on the positive social externalities a WASH intervention could generate, and as a result, are running the risk of restricting ourselves to narrow technocratic approaches. This needs to change.

This brings me to a key message I have for WASH interventions: do not hurry into scaling up. Given the urgency of the problem – about 2.5 billion people do not practice safe sanitation – this might seem completely counter-intuitive. However, there is the real risk that aiming for scale will lead to the perpetration of target-driven hardware interventions which will neither change behaviour, not create social cohesion. It is not unusual for organisations that rush to scale end up compromising on exactly those key design elements that made their pilots a success.

In conclusion, it is important to acknowledge that WASH interventions have the potential to go far beyond basic service delivery. In order to realise these gains, one must follow a very careful sequence of steps designed to promote community ownership and systematically change behaviour. The goal should be to nudge communities towards a public spirit and collective problem solving, so that WASH works as an entry point into communities, creating fertile ground for future interventions.

What Digital India will not fix

This is my latest livemint column


There is palpable excitement among many over Prime Minister Narendra Modi’s Digital India scheme. At its glitzy launch, business leaders competed with each other in eulogizing Modi’s vision and outlined their own investment plans. The event and the glamour plays right into this government’s playbook—a launch a day to keep doubters away.
It is then not ironic in the least that the Digital India week was effectively kicked off by the Prime Minister’s #SelfieWithDaughter campaign. Superficial in its intent and (likely) impotent in its impact, this is the kind of digital campaign that effectively means nothing. It isn’t the only one though. Here are some issues that Digital India will not fix.
While there is a case for targeting in certain subsidies, the case in favour of universal inclusion is strong when it comes to the public distribution system (PDS). In fact, states like Chhattisgarh have demonstrated how a near universal PDS is not just only pro-poor, but also helps improve administrative efficiency. Mislabelling of above poverty line (APL) households as below poverty line (BPL) and vice-versa is the result of an insidious system where the pay-offs are high. Neither Aadhaar unique identity number nor digital records can help once a family is declared ineligible to access a particular scheme.
Digital India might ensure the old and vulnerable get their pension on time. However, it will do nothing to fix an apathetic system that thinks Rs.200 per month is a dignified pension sum. Similarly, if Modi is intent on choking schemes such as the rural jobs guarantee scheme, the much-awaited wage transfers will have little meaning. And, the skilling missions, which probably will not target the typical poor rural labourer, will not be delivered online or otherwise, unless jobs are created in industries that can absorb youth.
The Digital India movement can help gather and analyze masses of administrative data. This could then reveal the critical insufficiencies in the delivery of public services such as education. India has over 250 million children in primary school, and learning levels in our schools have stagnated. Yet, successive government programmes focus more on civil works and teacher salaries, instead of teacher mentoring and monitoring. Further, in an administrative context, where the bureaucratic hierarchy along with the political leadership are complicit in turning a blind eye to these gaps, going digital will not improve service delivery on the ground.
Staying on the topic of sanction and enforcement, it is now clear that no amount of digital evidence can lead to the prosecution of any individual in the good books of the ruling dispensation. Independence of our investigative agencies remains a pipe dream. On the other hand, constitutional safeguards such as the right to information (RTI) are being systematically dismantled. Consider this alongside the blatant disregard for probity and competence in the appointment of heads of various public bodies and institutions, and one can see the incessant march of political partisanship and ideology at the cost of merit and integrity.
A Digital India cannot fix this. And as we have now learnt, no amount of noise on digital mediums seem likely to succeed in repatriating Lalit Modi (or Dawood Ibrahim), and from the looks of it, silence across the board on the Vyapam scam in Madhya Pradesh is unable to stem the series of mysterious deaths around it, or initiate an impartial investigation.
The lesson here is simple—the underlying architecture of the governance system has a long way to go before it can claim itself to be just, progressive and pro-poor. And reforming the administration requires concrete steps that emphasize accountability.
Post script: Digital India will continue to block websites and suppress dissent both online and offline. It is striking that this contradiction does not bother the Prime Minister. And this makes one wonder if the digital revolution is more about controlling the medium and the message than a tangible improvement in governance.
As a Mint column reminds us, by 2019, there will be 600 million smartphone users who need to be tapped early—a figure far higher than the 100 million missed-call members in the Bharatiya Janata Party’s kitty. This is not to say that there will be no marginal improvements in delivery worth talking about in the years to come. But it also suggests that the Prime Minister and his party will be satisfied if all that Digital India achieves is that it furthers their propagandist needs.

A case for cutting out implementation staff?

Implementation staff and their roles in ‘facilitating’ development for communities are under scrutiny.

Chris Blattman bases his argument on a review of the ultra-poor evaluations as well as his own research on ‘cash plus training’ intervention in Uganda:

The biggest expense across all the programs was staff time. Especially for supervision. Delivering training and cows takes skilled labor, and it’s hard to cut this back. But supervision? … should it cost 50 or 60 percent of the program? Is it more valuable than the cow or the grant itself? It’s hard to believe.

We tried to test this with cash-plus program in Uganda. Supervising the women cost about $377, about half the cost of the program and 2.5 times as much as the grant itself. After the 18-month impact evaluation, the charity brought the control group into the program. This time, however, the charity randomized the supervision: Some women got the training and grants with full supervision, some with only one or two visits, and some with none at all.

We surveyed the women a month and then a year later, and found that the supervision helped the women maintain the new businesses they started, but there was virtually no effect on consumption. We have no idea whether the supervision helps another year down the road. Maybe, eventually, it pays for itself. But the simple fact is this: taking away the most expensive part of the program had little effect on benefits after a whole year

And here is Howard White, reflecting on the evidence from Community Driven Development (CDD) projects

In many CDD projects, the decision-making and application process is facilitated by outsiders. A chunk of project resources are used not for funding things communities want, but paying NGOs to train communities in how to hold meetings and help communities decide on what they want.

Now, facilitation may be useful. It can help ensure that the voices of the marginalised are heard, that poorer communities without the skills and connections get to apply and develop skills in project management. But I do wonder if communities that already have community-level decision-making bodies need outsiders to help them hold meetings and to decide their priority needs.

On one hand, Chris is saying supervision and monitoring isn’t worth the money, and on the other, Howard is saying the same might not even be good development strategy.

So should field staff in implementing organisations be made redundant? Perhaps communities do not need technical guidance and hand-holding. They also perhaps do not need support from external resource persons in solving collective action problems.

Here are some thoughts on the subject, and potential lessons for donors and implementers:

  • Not all supervision is paternalism: Yes, the Uganda study shows that stripping out supervision and monitoring doesn’t have an effect on individual benefits, but terming all of it paternalism or even compassion doesn’t quite convince me. Implementing agencies on the ground are delivering a mix of social and technical interventions. The power relations between communities implementing agencies is complex, with the balance of power shifting from one side to the other depending on specific circumstances. There are also some other drivers of these costs. Implementers are often stuck in a donor-mandated system of financial controls and audits – which mean greater on-ground supervision. Not all of this may be productive from an outcomes perspective, but are nevertheless costs that implementers will find hard to cut out.
  • No excuses for poor facilitation: On CDD projects, Howard’s views on facilitation are really a pushback against poor facilitation of the kind that obstruct rather than allowing communities to organically evolve decision-making structures and determine collective action. In his blog, Howard describes how a group of development workers and researchers were seemingly convinced that ‘facilitation’ is an essential component of the work they do. But surely, in their heads, they were assuming the ‘right’ kind of facilitation, implementing which is a whole different matter. That kind of poor facilitation is paternalism – an idea that outsiders know better and have to drag communities towards enlightened decision-making. There is no excuse to perpetuate this practice.
  • Agencies are grappling with efficiency in operations: It would be unfair to say that field organisations do not grapple with questions of value for money, with respect to their operations. In some cases, it is donors that provide the impetus for this, but questioning the number and composition (mix of local/international) staff proposed. Working more with local organisations, as opposed to INGOs – apart from the benefits one anticipates from locally embedded partners – is also an attempt to realise more efficient field operations costs. From the implementers’ cost of delivering projects, it would be interesting to strip out the costs of desk-based supervision of the field monitors and then compare across locations.
  • Some projects need facilitation/supervision: Delivering livestock or capital drops is a very small slice in the universe of aid interventions – and of a very specific nature. It might be possible to minimise monitoring for these projects, but not so for others. For instance, a behaviour-change communication campaign on water and sanitation, or a pedagogical improvement intervention involving public school teachers may need more regular monitoring. There is no denying though that implementers should constantly try to figure out the most efficient monitoring mix, with a judicious use of technology where possible.
  • Cost-effectiveness of supervision: There is a difference between programmes that target individuals as opposed to those that target communities. Of particular relevance to the kinds of project that Chris describes is the question regarding the size of the cash/capital/asset provided. Would supervision costs grow in proportion to the size of the transfer? I think, not. Would benefits have gone up in proportion to the size of the transfer? I can’t tell. An important follow-up to the ultra-poor pilots may be seeking a better understanding of thresholds at which supervision is cost-effective.
  • Difficulty in cross-country comparisons: Issues around cost of living and wages in developing countries is often baffling. Field staff in Indian NGOs are paid very little in comparison to many African countries. The differences cannot be explained just by US$ PPP adjustments – it has also to do with the size of the labour market, skills available, etc. So as almost anyone who has worked in India has experienced, every $ goes much further in India than anywhere else. That partly explains the more promising cost-benefit ratios coming out of India. Finally, this contains useful lessons for value for money comparisons across projects and contexts. In a perfectly comparable world, we can test out the impact of providing exactly the same input, in exactly the same way, for exactly the same time-period across locations. But that is neither possible nor desirable, since…duh, “context matters!”. A temporal comparison within locations may be more appropriate – are implementers able to bring down delivery costs over time? Is it able to ensure that these savings do not come at the cost of outcomes on the ground?
  • Is cash the answer? Then there is the ‘give them cash’ brigade which is stacking up evidence in favour of literally air-dropping cash to people. As I have written elsewhere, I think there are all sorts of collective action problems and systems improvements that will not come out of cash transfers. Neither are most poor country governments capable of administering these transfers, nor should they be withdrawing from public services and focusing on individual-targeted cash transfers.

Regulator focuses on top executive remuneration in India

SEBI now requires listed companies to disclose the ratio of remuneration paid to their directors and their median staff salary. Industry experts probably know this already, but the ratios are sure to reveal interesting insights about the nature of these companies, as well as the sectors in which they operate.

Hopefully, SEBI will come out with a consolidated report on the entire industry.

The point of this post is not to generate gossip, but for juicy information on specific individuals, see full story here.

H/T – Ajit Ranade

A silicon classroom revolution?

This column first appeared on the Adam Smith International partner zone on The Guardian


Every morning, teacher Gladys Achieng takes the class register at Mathare Community Education Beijing School in Nairobi. Beijing was added to the school’s name in honour of new classrooms donated by a Chinese construction company: they built a new six-lane “super-highway” which rumbles right alongside Mathare, a sprawling slum. The highway has become a symbol of Kenya’s aspirations, much as the slum is symbolic of what the country intends to leave behind. The children answering their names at roll-call hope that education will be their ticket to a better life.

A different kind of “super-highway” – the information super-highway – was supposed to transform education in the developing world. However, in Kenya, a campaign promise by the president of “one laptop for every child” has so far failed to materialise. Whilst computers have begun to appear in schools, there is often a lack of teacher skill – or even electricity – to use them.

Buoyed by wildly optimistic talk of a silicon classroom revolution, innovators not only underestimated the human and institutional capacity required to implement their solutions, they overestimated the impact of their ideas and products on the lives of the children.

This is no longer a radical or a Luddite view. Experts are beginning to come around to the view that efforts at technology adoption and scale-up are doomed to failure unless they tackle the human element. Technological innovations cannot be seamlessly implemented across community, socio-economic or national borders. What works for the rich may not work for the poor, unless solutions are tailor-made to suit each context.

Back to Gladys Achieng’s roll-call. Here, in Mathare, a simple application of everyday technology – the ubiquitous mobile phone – has had a practical and immediate effect on school attendance. “After completing the register,” Gladys explains. “I send an SMS to an automated number. If any pupil is absent, their absence is recorded, and their parents can be notified.”

After three days of absence, the system creates an alert that may prompt a visit to the child’s home. Meanwhile, the data is used by the school’s donors – including the UK’s Department for International Development – to monitor the success of their funds in keeping children in school. The system was designed in collaboration with teachers, who insisted they didn’t want burdensome paperwork which took them away from the classroom.

“It really works,” says Gladys. “As you can see, we have 100% attendance today, as most days.”

Other uses of technology, piggybacking on existing consumer equipment and applications, have also proved successful. In remote areas, Global Positioning System (GPS) enabled mobile phones and webcams have led to improvements in school management and staff training.

But more costly, high-profile interventions have struggled. An animated education series aired on national television, while proving immensely popular with children, has shown limited impact on results. Experiments in collecting survey data using tablets (instead of paper forms) neither improved staff efficiency, nor prevented errors in data reported.

Technology cannot on its own, therefore, improve educational outcomes. Laptops and e-books will never replace qualified and accountable teachers. However, when sensitively designed, technological innovations do have an important role in terms of helping focus the efforts of implementers on the ground.

The rapid rewards of the application of technology in education are proving to be in ‘enhanced accountability’ and not ‘educational content’. Better use of conventional technology – such as digital data collection systems and GPS mapping – can provide decision-makers at national and sub-national levels with information that is relevant, comprehensible and actionable.

For the pupils in Gladys Achieng’s classroom, the technological wonderland of internet-enabled touch-screens is a long way off. But they’re all in school, and happy to be learning. That is technology at work.