Listen 3 minute audio clip on evidence and the political agenda: WMA (260kb)

Are some issues more conducive to evidence-based policy-making than others in your country? What kinds of policy processes are particularly important?
I think the most important policy context in Sri Lanka is the conflict. My organisation is not really involved in this because the whole area of the peace negotiations is extremely sensitive, so there is a highly specialised way of inputting into that process. There are many organisations in Sri Lanka that do that considerably better than we could. We have a Poverty and Conflict (PAC) programme, which looks at the nexus between poverty and conflict, but it is more concerned with looking at how communities perceive conflict, what their recovery processes are and that sort of thing, and how these years of conflict have influenced people going in and out of poverty. That conversation does not really come into the policy discussions. I have been thinking about whether it should come into the policy discussions, but I do not feel confident enough about my understanding of the whole policy arena around the peace process to enter into that kind of conversation. Still, that is the most important policy process I think.

The conversation that we are involved in - which is basically a kind of poverty and development conversation - is quite interesting in Sri Lanka because Sri Lanka is now almost a middle income country. Poverty does not exist here in the way it exists even in India, I imagine, and it certainly does not exist in the way it exists in Africa. This means that it is actually very specialised and, because it is not that widespread, understanding poverty is about understanding the incidence of poverty. There are certainly regions that are highly marginalised, and even within areas which appear to be fine using an aggregate headcount, if you then disaggregate according to regions or into lower level administrative units etc., you will find that there are some pockets of poverty. So for us, it is really about trying to understand that, and that is the kind of debate that we are engaging in.

How do you use evidence to put that kind of very nuanced argument on the political agenda?
I am not sure that we have actually done that, but we have made some impact. The Centre for Poverty Analysis (CEPA) is a very new organisation. It has only been around for five years and I have only been with CEPA for one and a half years. I think that CEPA spent its first five years building up its credibility and it is only now that it is beginning to think about how to use the body of knowledge or evidence that it has gathered to influence policy. Its main impact has actually been through its Poverty Impact Monitoring programme. By the very definition of the programme, it was about working with programmes and projects involved in development to see what their poverty impact has been. The very nature of that work enabled CEPA to have an influence because we knew our subject. When you are monitoring or evaluating the impact of a programme and seeing whether it actually has an impact, you have conversations with the decision-makers within that programme. It is within that very small sphere that CEPA has really had any kind of policy influence.

More generally, the thing to note is that policy is not made in a rational way in Sri Lanka. It really depends on a huge number of things. Of course, a certain amount of evidence influences policy, so if you were making energy policy, for instance, or transport policy, some of the evidence about people's ability to access markets, restrictions in transportation, or the demand for energy etc., would be used. That kind of data and statistical evidence plays a part in policy-making, but I do not know whether people really evaluate the alternatives in any kind of rational way.

We are doing an action research project and one of the conversations that we had recently was to find out why they chose one particular option. Access from the deep south to markets in Colombo is really difficult, so that access needs to be improved, but there are many ways in which that access could be improved, such as improving the railways and some roads, or putting in an airport. We decided on an expressway, which seems to us to come at a certain cost, not only in terms of money to the government, but also in terms of social and environmental costs. I do not know whether these were counted in the decision-making when looking at alternatives. There is also the question of who made that choice? We were discussing this with the senior government official, who is actually doing an investigation for us about the institutional aspects of this programme. He said that one option had been mooted by the government of Sri Lanka and then the Asian Development Bank (ADB) suggested an alternative option, which was the expressway. I am not sure how much of this is true yet, but apparently, somebody then did a feasibility study, which has not yet been seen, and it was then agreed that the expressway was feasible. It seems to be a matter of the players applying different kinds of pressures at different points in time and, if you want to influence things, you need to be there at the right time, right at the beginning.

Do you think there is a space for your evidence to influence the big players' pre-determined decisions?
I think that we may need to collect different kinds of evidence. That is how I am beginning to think anyway, but I really do not know. At the moment, our evidence is at a purely programmatic or decentralised level. With the exception of this project, we have not been collecting any kind of big evidence or big information and we are not engaged with the big policy issues.

Can you give some examples of the types of evidence you were thinking of?
A lot of the work we do is about looking at it policy from the bottom up, so it is based on people's perceptions and impacts on people. We are not doing, for example, any huge macro-level analysis or impact analysis. We do not work at that level at the moment and I am not sure how much our credibility is affected by that, and by the fact that we do not do quantitative analysis in any major way. At the moment, we are grappling with the question of the level at which we should try to influence policy, as well as the question of what type of evidence we need to collect to influence policy at that level. Finally, I think one of the things that we need to think about is whether we want to go down that route.

From your experience of working in different areas, are there different types of approaches that have worked better than others?
I will speak from my own experience, not about CEPA, for a moment. I was working for Practical Action (formerly IDTG) to influence energy policy and, at that time, this meant tackling the issue at different levels. At one level, we demonstrated the feasibility of alternatives, so we did micro-hydro projects which showed that it was feasible to generate decentralised energy. We made alliances with other decentralised energy providers, like people who provided solar power or people who provided power gas, and we formed a network called the Energy Forum. That was one level.

At the same time, we got people who were on the electricity board, which is the major electricity provider, into what we called TAC (Technical Advisory Committee). This meant that they were always there, supporting and backing our energy programming and aware of what was happening. I think the major issue came when the government got a World Bank funded programme called the Energy Services Delivery Project, looking at energy efficiency and alternative energy sources. The World Bank came to Sri Lanka with a very strong solar bias, mostly because solar power is a very American business I think, so the World Bank is very solar oriented. We were trying to put other alternatives forward, like micro-hydro projects, but they did not want to hear about it in our conversations. So we did this very unconventional thing, which was to take them on a tour of the solar stuff, not the other alternatives. We provided the transport and we went along with them and, by the end of the day, we were kind of good buddies with this group, which meant that we were able to invite them to come and see some micro-hydro projects the following day, if they had space in their diaries. They did, and we managed to take them and provide them with that practical evidence, rather than research evidence.

However, we also positioned ourselves as a group that could actually find information about the potential for decentralised hydropower in Sri Lanka. We had a consultant do a big survey of hydropower potential, and things like that, so that we effectively became positioned as the organisation working on decentralised energy and that enabled us to look at the issues and suggest alternatives. The other thing we did was run a training programme where we trained engineers to do micro-hydro technology, to build turbines and to carry out turbine maintenance.

There is always a time, a key moment, even though it is sometimes hard to say why it happens, but when that time came, we already had this whole system in place. There were already people able to supply the turbines, there were people trained to do the installations and there was an understanding of decentralised energy options. When we started, it was illegal to generate electricity and only the electricity board was allowed to do this. The policy was to make decentralised energy generation legal. In a way, that is an example of something that was quite easy, because the changes that we wanted were in some way tangible. In CEPA, we find that many other changes are much softer and we are still struggling with the best way to have an impact on those.

Is it because the issues are not just about pure resources but are also about rights and deprivation as well?
No, I think that the issues are different. They are not so much about something that it is possible to demonstrate. The issues are much more analytical, so you need to have people buy into the analysis rather than being able to show them something. I think that may be the reason, or it may be simply that we have not thought it through or engaged enough.

In the Sri Lankan context, are international examples of best practice persuasive or is it better to rely on small pilot projects locally?
I do not know. I think that there is a great scepticism, whether overt or covert, about donors, about NGOs and about international NGOs, and there always has been. Sri Lanka has always had a mentality that questions why these organisations think they can tell us what to do, and this attitude got worse with the tsunami. I was living here in the UK when the tsunami struck and I rang a friend of mine who is working on all this disaster stuff in Geneva - he works for the UN. He said at the time that the tsunami was the first disaster, the second disaster was going to be the aid. I gave him a big lecture about being cynical and not recognising people's good will and so on, but I think that it proved to be so true. There is a lot of money spent on doing a lot of nonsense. They were not thinking about the long-term so the aid undermined local capacity. It is very sad.

I feel a bit vindicated in terms of my own instinct about some of these issues by what happened in Sri Lanka, though I have not done any real research in this area. I am starting to read the tsunami evaluation reports that you have all done and I have been thinking that we should say some of those things as well. I am not saying that good things did not happen, but good things could have happened with much less money.

I think that there is generally a very strong anti-NGO feeling amongst the government as well - much more so than there used to be. I do not know whether this antipathy has just grown in the past ten years while I have been away, or whether most of it has happened post-tsunami, but at the moment there is a very strong anti-NGO feeling and a very strong tongue-in-cheek scepticism about donors. This means that you have to be very careful when you are using international best practice. Something may be ok in Bangladesh but how do you think it is going to work in Sri Lanka?

So does the government listen to pure academic or university-based research?
There is the commonwealth finance ministers' meeting and the commonwealth foundation, which has a sort of NGO presence. I was at a meeting where the Deputy Minister of Finance (the President is the Minister of Finance) hosted a lunch for all the NGO people. He said straight away, "I am ready to listen to what you say, but you have to remember that governments are accountable to their electorates. We are elected, who elected you?" Of course, he did not quite say it like that or in those words, but that was his message: to whom are the NGOs accountable? Government ministers get elected and if people do not like what they do, they get voted out. If people do not like what NGOs do, what happens?

Do you think that that was a genuine sentiment or an excuse to get on with their own thing and not really listen to the NGOs?
I think that there may be a certain amount of that, and the fact that we have elections so frequently tends to foster a sense of short-termism. It sounds very democratic but it also means quite short-term planning and thinking, because politicians want to get elected in two or five years' time, so they will do only those things that are likely to win votes. This it lends itself to very populist decision-making, I think.

I often wonder how people can actually influence policy in Sri Lanka. The peace process is a different matter because in that case, it is a very visible agenda, but the other policy processes are really quite difficult to influence. The Department of Census and Statistics has been very clever in using a lot of the census data to point out where poverty is and so on. That is now having a fairly strong impact and there is evidence-based policy based on that. The Department has identified the 119 poorest administrative units with the result that now all the programmes are targeted to those. They have identified the fact that the main poverty alleviation cash transfer programmes were not being targeted properly, so that has led to a change in the way people are trying to redesign the cash transfer programme. Things like that are happening, but I think that the government is more keen to look at its own evidence, generated by its own departments, than it is to look at other research.

Does that mean that they are more likely to be persuaded by quantitative findings than by qualitative findings?
They are definitely more persuaded by quantitative findings, but they also tend to trust each other more than they trust outside organisations. Nonetheless, the departments that produce the statistics do it with a robust methodology.

Back to interview list