International Development Committee

Oral evidence: Sub-Committee on the Work of the Independent Commission for Aid Impact’s Report: DFID’s Approach to Anti-Corruption and its Impact on the Poor, HC 864
Wednesday 10 December 2014

Ordered by the House of Commons to be published on 17 December 2014.

Written evidence

Independent Commission for Aid Impact

Watch the meeting Wednesday 10 December 2014

Members present: Sir Malcolm Bruce (Chair); Fiona Bruce; Jeremy Lefroy; Fiona O’Donnell

Questions 1-42

Witnesses: Graham Ward CBE, Chief Commissioner, Independent Commission for Aid Impact, Mark Foster, Lead Commissioner, ICAI, and Lance Croffoot-Suede, Team Leader, ICAI, gave evidence  

Q1   Chair: Good morning, gentlemen.  Thank you for coming in on what is I suppose now a routine process, but not always routine in terms of the subject matter.  I am in the Chair pro tem, because Fabian Hamilton has a debate, which meant he had to vacate the Chair for this.  Just for the record—obviously we know who you are—could you introduce yourselves? 

Graham Ward: Graham Ward.  I am Chief Commissioner of the Independent Commission for Aid Impact.

Mark Foster: I am Mark Foster, Commissioner for the Independent Commission.

Lance Croffoot-Suede: Lance Croffoot-Suede, Team Leader on this report.

Q2   Chair: Thank you very much.  You have been pretty scathing of DFID in its work in fighting corruption.  In terms of what you found, and I know you have looked at particular examples in Nigeria and Nepal, which by the way the Committee is planning to visit in February, how concerned were you with what you found and what was the predominant factor in that?  You found corruption; how pervasive was it?  How serious was it?  Perhaps you could identify what particular aspect of corruption you felt was most concerning.

Graham Ward: Thank you, Chairman.  “Scathing” is not a word that we used and, if I may just try to put that word in the right context, we would all agree that corruption is recognised to be a critical issue affecting the poor around the world and it does present a formidable challenge to international development and to the way in which people can be lifted out of poverty, and countries can be lifted out of poverty.  We found that there is a lot of really important anti‑corruption work that the international community undertakes, often with the leadership of DFID.  We found that that should continue, but that it would benefit in addition from a greater direct focus on the needs of the poor.

              We found that DFID had not developed an approach equal to the challenge of corruption, so the main thrust of our report is to urge DFID and the UK Government to put more explicit effort into tackling corruption in the countries where it provides assistance, with a particular focus on the forms of corruption that afflict the everyday lives of the poorest people in society, by which I mean the mothers who have to pay a bribe to get medical assistance for their children, the parents who have to pay a bribe to get education for their children, the pensioners who have to pay a bribe to receive their pension.  Given the relative deficiency in the evidence base in this area, we also urge DFID to do more to gather information about what works over both the shorter term and the longer term.

              In terms of the evidence that we looked at, this was a very wide‑ranging report.  We reviewed over 900 documents in detail.  We spoke to over 1,000 people during our fieldwork.  We surveyed predominantly more than 5,000 people, and really predominantly those were face-to-face surveys, so we had a lot of contact in these countries.  We used that to polyangulate our findings—I know people normally talk about triangulation, but I would say polyangulation, because there were so many different aspects that we brought together in relation to this.  Contrary to the views that have received quite a lot of currency from some of the commentators, our findings are not based solely on any one element of that evidence; they are based on a judgment arising from all of those different sources of evidence.

              My fellow Commissioner, John Githongo, who of course you know, was very sorry not to be able to join us and very sorry that he could not lead this report, because he has been suffering from illness and he is tied up with a long‑running court case in Kenya.  John, as you know, is one of the foremost anti‑corruption practitioners in the world.  I did ask John to cold‑review our report post‑publication, because I thought that would be of real value to us and hopefully also to you.  He said that he supports the report; he supports the conclusions; he supports the recommendations, all of them, in every case.  He believes that this report would be of real use to practitioners in government, in the police, in the judiciary and in civil society, but that some expert commentators may find it to be challenging.

              Our team comprises, again contrary to some of the stuff that we have seen written up, experienced anti‑corruption practitioners in many countries.  We can go into that in a moment.  These are people who have actually done anti‑corruption programmes on the ground, in different countries.  We believe that, given the Government’s push for greater transparency at a global level, which we support, we want full specific accountable programming with a focus on the poor to be complementary to the UK’s push for zero tolerance to corruption.

              What we have done is search out a large amount of evidence.  We have reviewed in detail DFID’s own internal evidence and the evidence of a lot of other people and, on the back of that, we have made constructive recommendations for how the day-to-day lives of the poor, we believe, could be improved as far as corruption is concerned.  Mark was in Nigeria.  You asked specifically about Nigeria and Nepal.  Lance was in Nepal specifically, so perhaps I could ask them to say a few words to answer that aspect of your question, Chairman. 

Mark Foster: With regard to the situation in Nigeria, clearly Nigeria is a country that has a pretty robust reputation with regard to the whole area of corruption at multiple levels in its society.  What we were trying to understand was, as we looked at the programmes that DFID is undertaking in that very difficult environment, which ones appeared to be really focusing in on the issue of corruption and which ones were acting more indirectly on the activities. 

There clearly are a number of things that DFID is doing with regard to, for example, trying to increase transparency in the oil industry and make people more aware of what is happening to oil revenues in the country.  We saw some positive aspects of work in the individual states, around trying to provide more transparency around budgets for state governments and trying to encourage civil society to hold those governments more to account.  We saw a number of aspects.  We also looked and saw some of the work that is being done to improve trust in the police force, and to improve the investigatory and prosecutorial powers of the judiciary. 

All those things we saw as being good.  Individual parts of it were having individual point success, but the reality was that the whole thing was not really tackling the issue of corruption holistically.  We also felt that, because many of the approaches were indirect, there was no real strong position being taken by the UK in its work in Nigeria around our perspective on how important it is that corruption is rooted out at the multiple levels in society. 

Also, as we look particularly at where one of DFID’s real strengths is, which is around the delivery of service programmes around health, education and critical areas, many of which are being done at scale in Nigeria, the opportunity to use those programmes to tackle the kinds of lower‑level corruption experienced by the poor that Graham has mentioned, was not being taken.  Our key takeaway that I took away from our time on the ground there was that the office has a focus on it, but it could really do with stepping up that focus to meet the scale of the challenge in a place like Nigeria.  Nigeria receives an awful lot of support from the UK, and our view is that, unless this aspect is really optimised in the mix, there is a risk that our ability to move forward the whole country with regard to development will be compromised.

Lance Croffoot-Suede: In terms of Nepal, many of the things we witnessed in Nepal were similar to what we witnessed in Nigeria, so I think we should focus on some of the aspects of Nepal that differentiate it.  One of the principal aspects is the presence of many other donors that are very active in terms of international development programmes within Nepal.  We had much interaction with many of the different donors in very different contexts. 

In addition, in Nepal you have a country that has somewhat successfully emerged from conflict, and DFID’s work during the period of conflict has evolved in a very laudable way to now consist of programming for a country that is in a much more stable political environment, relatively speaking.  In the context of the period of conflict, DFID and other donors made great strides in terms of working with communities.  As you will see in our report, we talk about CSP, for example.  It came under the Enabling State Programme.  As a result of the great strides in terms of working with communities, one of the examples, in terms of leveraging that knowledge and those relationships to promote anti‑corruption programming that would benefit the everyday person and the poor, and try to mitigate the effects of everyday corruption, those opportunities were not necessarily being captured or focused upon. 

In addition, in Nepal as opposed to Nigeria, DFID and other donors worked closely, relatively speaking, with the host Government.  In connection with the relationship that DFID and other donors have with the host Government, there were a lot of situations in which the attitude towards discussing corruption with the Government, challenging the Government on corruption issues, is very difficult indeed.  As a result of those three different factors within Nepal, we saw a situation where opportunities to move forward, in terms of anti‑corruption programming that would benefit the poor, the opportunity to move forward in a forceful manner, in terms of making progress in getting the host Government to take responsibility for the corruption within their own government agencies and the possibility to work with other donors to harness all of this activity, was not being taken up with the sort of focus or alacrity as one might hope it would be if one had a real focus on the poor.

Q3   Chair: I do not think any of us would be in any doubt that petty corruption is endemic right across many of the developing countries.  Indeed, Hugh Bayley and I saw an artist in Dar es Salaam doing wonderful cartoons of people reaching into their pockets as soon as they saw a police officer, of people having to pay a bribe in order to get on a bus in order to pay a fare and stuff like that.  We know that is the case.  The question is to what extent you identify DFID’s failure to contribute to solving that problem.  Indeed, what is DFID’s capacity against such an onslaught? 

You know that the report has been criticised.  You have anticipated that in your response.  To pick up some of those, Professor Matthew Stephenson of Harvard Law School described the report as “just dreadful … neither the report’s condemnatory tone nor its primary recommendations are backed up with adequate evidence or cogent reasoning.”  It goes on to say, “It is, in most respects, a cautionary example of how incompetent execution can undermine a worthwhile project.”  Then it goes on to say that it is too long.  Dr Heather Marquette described it as “shabby”.  It should be “thrown out”, calling it “a waste of time and, at worst, a dangerous diversion”.  It has “no understanding of current thinking on measuring corruption, no understanding of the limited value of even high‑quality micro-level survey data taken on its own, and no understanding of the quality thresholds required to make such bold recommendations”.  You have clearly scratched some raw nerves amongst the academic community, but those are not amateurs who are engaged in the field and feel, by the sound of it, pretty angry.

Mark Foster: First of all, we do actually both appreciate and respect the fact that our report does raise some debate.  This is a very important issue and, frankly, the more focus and debate there is about it is a good thing. I would also say that we respect the expertise that many of the people bring to the table in terms of their commentary on it.  For some of the comments they make, we can see where they are coming from.  For others, we disagree with the point that they are making.  We saw some different things on the ground, in our work, and we reflected that in the evidence we saw and in the report that we produced.

              There are a couple of other areas where, to be honest with you, we feel they missed the point we were making.  For example, both cases make quite a big issue around our focus on the point of petty corruption and seem to imply that we are saying there should be a redirection of funds towards focusing on petty corruption, away from other either indirect or top‑down vehicles.  We are not saying that at all; we are saying it is an “and”.  We actually say there is an opportunity for more to be done, particularly given the fact that DFID is working so closely on so many of those service delivery programmes. 

Our sense is certainly that there are some aspects of critique in there.  We note, for example, the critique around whether we have focused on this demand/supply approach, for example.  That is the language that is used in this work across the world.  It is the language being used by DFID themselves in Nigeria.  In terms of the structures that are being used to try to tackle corruption, we do believe it needs to be a combination between things that are working on the government delivery side of things and things that are working with regard to citizens, about how they receive and respond to those initiatives.  In fact, if anything, our main takeaway, which came through very strongly from our survey work, was the fact that there should be more focus on the demand‑side work, alongside structured, focused, coherent and comprehensive supply‑side work.

Q4   Chair: Just before I bring in Fiona Bruce, obviously this kind of report is an absolute gift to the detractors of aid and development.  That is not any reason why you should not do it.  We know that the Mail is no friend of development.  Their headline was, “Our foreign aid fuels corruption”.  All I am going to ask you is whether you think that is at all a fair summary of what your report is saying.  How would you respond to that?

Mark Foster: Clearly it is not a fair assessment of the report.  Our report was fundamentally recommending that there should be more spent around this area to really tackle the issues of corruption.  We did highlight in the report one particular example that we found, where we were concerned that the way that the streams of money were flowing was encouraging some potentially corrupt behaviour.  It was this example in Nepal that we may come on to later on. 

The fact that that became the headline and the centre of gravity for that review is what it is.  We knew, when we were putting it in there, that it was likely that that would be a lightning rod for some of the commentary, but the fact of the matter is that is what we saw.  We saw that there.  We felt we had to report it as we saw it.  As I say, when I look at any response to any of our reports by aspects of the media, they pick things that are interesting to them and do not pick other things that we think are important.  It is true that our green and amber/green reports get much less coverage than our red and amber/red reports.  That is the way it is, but we feel that we had to present things as we did. 

Chair: I do not think the Mail needs more response than that.

Graham Ward: Can I just say that, clearly in a free democratic society with a free press, the press will comment as they see fit?  Clearly that is right.  From ICAI’s point of view, we very much see our duty to report the truth as we see it, from the evidence that we have gathered, to you and through you to Parliament and to the taxpayer.  That is our overriding duty.  Would you like Lance just to talk about this Nepal issue?

Chair: Just briefly. 

Lance Croffoot-Suede: There are a few points that I will make very briefly.  It is important to address the questions about the commentary.  There are a few misconceptions that underlie much of the commentary, we would say.  This is a sensitive subject.  We welcome the debate, but it must be said that our recommendations, specifically recommendations 2 and 3, explicitly state that we want DFID to add to its programming.  We are not recommending that DFID changes programming such that it eliminates or diminishes any sort of programming.  The additional programming that we are recommending and the additional work to fill the evidence gap with regards to everyday corruption is exactly that: it is a recommendation to add to the existing programming; it is a recommendation to include more explicit anti‑corruption programming. 

Those two facts, plus the rather lengthy introduction to annex 2 in our report, which introduces the surveys, go to great lengths to explain that these surveys are meant for performance management and accountability purposes.  These surveys were not at all designed to, in any way, touch upon the sorts of surveys that DFID already does and does fairly well, as we say in our report.  These are not evaluation surveys for evaluation purposes.  These are for performance and accountability purposes. 

Professor Stephenson’s comments, for example, do not seem to reflect the surveys in terms of what they were designed to do and how they were actually qualified, in terms of what portion of the evidence base they consisted of in the report.  There is various language in the report, which I do not need to point the Committee to, which explicitly talks about the fact that the surveys were just one part of the jigsaw of evidence, with regards to how we came to our conclusions and recommendations.  It is quite important for the Committee to understand that, because this is a sensitive subject, for some reason or another, there has been interpretation of our recommendations, of the intent of the surveys and of the content of the surveys that is not necessarily in line with what is actually in the report and in the surveys themselves.

Q5   Fiona Bruce: Thank you very much for that clarification that you are suggesting that elements are added to the work already undertaken.  Could you comment further on your note that you acknowledge that interventions must operate on a 10‑to‑15‑year timescale?  Can I roll into that two other questions, just so that we can make progress, because I know that there are a large number of questions this morning to be asked?  Just comment on the relative emphasis that you see should be given to the macro‑level and micro‑level work of DFID on this issue.  Finally, comment on the criticism that has been raised that perhaps your comments that there should be targeting by DFID on petty corruption is making too strong a policy case, bearing in mind your remit not to stray into that area.

Graham Ward: Perhaps I can deal with the last point first, and then I will ask Mark to go back and deal with the earlier ones.  Actually, we did not see it as being a policy point.  We saw it as a point to do with the implementation of policy.  The policy, as I understand it, is that DFID has taken the decision to put a lot of effort into fighting corruption.  We heard the Secretary of State enunciate this at a speech that she gave yesterday evening to Transparency International UK at New Scotland Yard, just down the road.  What we are talking about is the implementation of the policy.  We are saying that the focus on the poor is a really valuable additional element of the implementation of the policy.

Mark Foster: To try to join the dots between the various levels of intervention, first of all, we very much respect and we acknowledge in the report DFID’s leadership and focus on the global initiatives that relate to the whole anti‑corruption agenda.  Certainly from looking at that, we can say that they have been engaged in it.  Everyone you spoke to talks about DFID’s engagement in that.  Where we critiqued it was around trying to understand how engagement at that level percolates down to having a real impact on the ground, and that is a fair critique. 

              What we were then trying to say is that, if you look beneath that main level, the next level is the interaction at the political level with the leaders of these countries.  That is where our recommendation around the FCO, DFID and joined‑up explicit positioning with those leaders, around an anti‑corruption push, is something that we should be doing, as we are with the anti‑bribery act in the private sector that I come from, for example.

Fiona Bruce: Indeed, as we are talking about the issue of persecution.

Mark Foster: Exactly.  There is an opportunity to strengthen that level, appreciating what has already been done, and being clearer about that.  At the next level, which is the level around reforming agencies and bodies, such as the police, the judiciary, the people who actually execute the Government’s policy and things like anti‑corruption commissions, etc, we believe that that is also an important part of the structure.  Again, our view would be to see what is really working in that area.  Reinforce the pieces that are working and just check, as we found in some cases that, for example, the model police station example did not appear to be very effective around this particular aspect, despite being pursued in multiple places across the patch. 

The final piece was to say that we recognised this point about the service delivery capacity of DFID, the real strength around health, education and key programmes on the ground with communities—it is a particular community linkage point—and said that that is a missed opportunity.  If you join that all together, the opportunity to keep the leadership at a global level, be more explicit at the leadership level, tighten up doing the right things and then add more to the work of services, we feel that would be a really powerful story played out over multiple years.  That is again our point about the 15 or 20‑year process. 

This is not going to be a quick fix.  If anything, some of the concerns we had were that unrealistic expectations were being set for some of these programmes, with regard to this.  We believe it should be recognised that it is going to be a long haul.  Small steps will be made, but what could we do in five years in a place like Nigeria, or 10 years, 15 years or 20 years, with the right kind of benign and appropriate counterpart?

Graham Ward: The existing and flourishing of petty corruption is a real barrier to effective service delivery in these core areas that we have debated together so many times, in the past.  Anything to add, Lance? 

Lance Croffoot-Suede: A couple of things.  In the report, we talk at great length about how certain programmes that DFID in Nigeria has funded for some time now have achieved some of these interim goals.  In fact, if DFID were to see these programmes as 10, 15 or 20‑year programmes, it would better fit the reality on the ground that already exists and already is proving to be successful, if one looks at the achievements in an interim fashion and makes them more modest. 

In addition, in terms of the policy points, the decision of ICAI not to dictate what the emphasis should be, micro versus macro, goes exactly to the third question, which is that ICAI is not in a position of dictating policy to DFID.  As a result, our report points out a gap in programming that we believe exists, in terms of targeting everyday corruption.  We talk about the fact that there should be expanded programming and additional direct anti‑corruption programming.  It is not our place to dictate the policy in terms of what the emphasis should be, micro versus macro.

Q6   Fiona O'Donnell: The report says that you should have evidence to justify anti‑corruption measures but, at the same time, you say that we need many more programmes to tackle everyday corruption.  You acknowledge that the evidence was not there to support the recommendation.  Also, you do not appear to have investigated any standalone anti‑corruption programmes.  I am just wondering how you then justify the recommendations that they have to be evidence‑based, if you are saying that there is no evidence there of what works.

Graham Ward: The fundamental recommendation is that there is clearly a gap in terms of direct programming on day-to-day corruptions that affect the poor.  We believe that it is worth investing in that area, because it is so debilitating, as I mentioned before, as far as the individual poor people are concerned.  That is something that we think is worth investigating. 

              It is relevant that the thrust of the DFID response to these recommendations, which they partially accept, is that the direction of travel of the actions that they are taking in their response to our recommendations is very much the same as the direction of travel and thrust of the recommendations themselves.  They are not really questioning whether it is appropriate to expand the programming to see whether things can improve, as a result of doing that.

Mark Foster: Also, we did identify, admittedly not in the countries that we visited, more direct work in Jamaica and Bangladesh, where we saw, both with the police and with the health service work in those two countries, more explicit focus on tackling anti‑corruption as it relates to the health service in Bangladesh or tackling anti‑corruption as it relates to the police in Jamaica—that kind of explicit and direct thing is possible, but one would argue that there are some very difficult environments to do that as well.  Our point was that there is evidence, albeit we did not see so much of it in the two countries we visited.  We see that. 

              Also, as Lance has said, we saw evidence of interim success.  We saw evidence that things were on paths that were creating pieces of the jigsaw puzzle.  We felt that that was positive, but was insufficient against the challenges that we are talking about here.  That is why our final recommendation around gathering evidence, gathering knowledge and making a stronger organisational thrust against this is so important, because we think that, actually, there are points of light here that could be drawn together, from which better targeted workable programming could arise.

Fiona O'Donnell: I am just conscious of timing, Chair.  We have quite a few questions to get through.

Lance Croffoot-Suede: Two seconds.  It is important to note that the report very explicitly talks about the fact that there is an evidence gap and that DFID, as a leader in overseas development programmes, should be looking to be innovative.  We tried to and did consult with the poor in innovative ways.  We spoke, as you have heard—directly the team—with 1,000 stakeholders.  The evidence that ICAI gathers, and ICAI does gather its own evidence in connection with these reports, is evidence that ICAI relies on in terms of formulating its recommendations.  We are recommending to DFID that DFID takes the lead, fills the evidence gap, targets more programmes towards everyday corruption, because the stakeholders themselves, whether the poor or officials in the anti‑corruption agencies themselves, who spoke on the basis of anonymity, are asking for more direct anti‑corruption programming, because the time is right, they believe, in their countries now—and it may not have been five years ago—to attack this problem in a more direct and aggressive manner.  When we recommend things, we are recommending them in the context of the need to step up the attack on corruption, which will require innovation.

Q7   Fiona O'Donnell: Can I quickly check whether it was 1,000 face‑to-face interviews or 5,000?  I have written down 5,000.

Lance Croffoot-Suede: There were 1,000 face-to-face interviews by the team itself, and then there were 5,300‑odd interviews done by the survey firms that we worked with.  Those people met face-to-face with beneficiaries.

Q8   Fiona O'Donnell: Just to be clear, there were stand-alone examples you looked at.

Graham Ward: Yes.

Q9   Fiona O'Donnell: That is good.  This problem with there being a lack of evidence and then justifying that you say DFID is failing to do enough, is it actually just a lack of evidence and enough evidence for failure, in particular in relation to the paper U4, which seems to have caused a bit of debate?  Does that paper actually give evidence that interventions have failed, or is it simply that there is no evidence of what the outcomes are?

Graham Ward: We were frankly rather perplexed by the evidence you received from U4, because they criticise us for holding them up as a justification for criticising the DFID programme.  We do not, in our report, hold them up as a justification for any criticisms of DFID’s programming at all.  What we do is refer to published work by U4 in relation to the transparency of information around the corruption area as good examples of improving transparency and improving awareness of the problem.  Frankly, we were perplexed by this submission and we believe that it is misdirecting to those who read it.

Q10   Fiona O'Donnell: Your 2011 anti‑corruption review focused more on fraud than anti‑corruption, but you have criticised DFID for over‑emphasis since then on counter‑fraud, though we have said that you do not set policy.  I suppose DFID really has to take responsibility for that decision, rather than saying you are recommending policy when it suits their purposes.  You acknowledge that DFID has increased its focus on anti‑corruption in recent months but, given ICAI’s focus has similarly changed since 2011, are you surprised that DFID is following suit in this?

Graham Ward: We are not surprised that DFID has gone about implementing the recommendations in our report in 2011, because indeed their management response to that report was that they would implement recommendations, so that is entirely in line with what we would have expected.  Also, they have made these policy statements about fighting corruption and having zero tolerance of corruption, which indicates that they will be working to achieve those ends.  The other comments that we have made here are in recognition of that policy thrust and what has been done in respect of it.  What we do in the report here is acknowledge—and do it in very positive terms—the work that DFID has been doing in terms of implementing the report that we issued in 2011.

Mark Foster: To add to that, this is another situation of an “and” rather than “either/or”.  We say in the report that they have made progress on the counter‑fraud side of things, both with the strategies and the Cx3 and other bodies that are created.  The question mark is around the “and”, around how much is being done on the anti‑corruption issue in country.  Our view is that more could be done, particularly our recommendation 2 on having specific strategies.  When you look at the strategies that have been created, they are very good strategies with regard to protecting our funds, increasingly good.  The question mark was how much they were anchored in context around what was changing in the corruption world at large in these countries, and how much that then gave direction for programming that sat underneath it.

Q11   Fiona Bruce: Again, I am going to roll a number of questions into one, if I may.  First on the question of beneficiaries’ surveys, you have used the word “unfortunate” when you said it is unfortunate if the way that the report is drafted implies that the beneficiaries’ surveys were the principal source of evidence.  At the outset, you did tell us that you have a number of other aspects of evidence and information, but it is perhaps more than unfortunate if the way that the report has been presented undermines its credibility. 

Can I give you a couple of examples to specifically respond to?  For example, you say that prosecutors in Nigeria who received training from Justice For All “overwhelmingly felt that their capacity to prosecute corruption cases had improved”, but your online survey of 377 prosecutors had a response rate of only 18%.  You also say that you “challenge the belief of successful impact” of model police stations in Nigeria, but is your survey the only evidence of a perceived increase in the experience of requests for bribes by police?  If so, do you accept that perhaps it is risky that coming to any firm conclusion is questionable, on the basis of self‑reported perceptions?

Graham Ward: I will just say, before Mark comments, that I did say at the beginning that our conclusions are drawn from a polyangulation of evidence.  We have perhaps suffered here by being open in terms of the methodologies and results of all of the different bits of evidence that we explicitly gathered.  What we have not been able to do, and I do not believe anybody is very much able to do it, is to present an algorithm that takes the evidence, weights it and then drives through logic circuits to the professional judgments that are made in coming to conclusions and recommendations.  There is a danger in commentators salami‑slicing the evidence, because that is not what we did.  We brought it all together and formed a conclusion on the basis of what we brought together. 

Mark Foster: That is good overall context.  First of all, taking the examples explicitly, with regard to the prosecutor example, the first thing to say is that we spoke to a number of prosecutors when we were out there.  That was part of our process.  It was from speaking to the prosecutors that we drew out our hypothesis as to whether or not this programme was making a difference or not around the critical issue of more cases being brought to court, etc, more things being pursued and therefore whether the overall thrust of the programme was functioning.  We used the survey to see if we asked more people—and we actually asked all the people who had gone through that prosecutor training, so we asked the whole lot—who would come back to us with a view as to whether or not they thought it.  Rather than just taking a view from the people we had spoken to, we could get a wider view. 

Yes, 18% of them came back to us with a response, and that response, to some extent, validated the concerns that we had heard about from the individual conversations that we had had.  That was the basis upon which that piece of evidence fitted into the jigsaw puzzle of our perspective. 

              On the second example, which was the example of the police stations, interestingly in this case, the trigger for us doing the survey was actually DFID’s own survey work and the work being done by the programme, both of which had identified a challenge with the issue of trust in the areas around these model police stations.  As was said to me by one of the programme leaders, “We can’t get it”.  We keep on doing more and more work around these model police stations, and yet the indicators on our own surveys keep going in the wrong direction around trust.  We thought we had better had a look at that.  As a result of that, we chose to do a survey on the model police stations, as best we could, to find out whether or not our view would be validated by that.  That is where we did the face-to-face surveys with the majority of the people in Nigeria in particular to validate that. 

We did, in that process, also try to ask some people around a place that had not got a model police station, just to see whether there was any particular difference.  The reality was that there was not.  Whether you were around a model police station or not, the fundamental feedback we received from our survey totally matched what was coming through from DFID in the programme’s own reviews.  That is what we reflected in the report.

Lance Croffoot-Suede: I would just add a couple of things, if that is okay.  In addition to what Mark has said, to be sure, we spoke with heads of anti‑corruption agencies; we spoke with various NGOs and other citizens regarding the various initiatives undertaken by the anti‑corruption agencies, including the prosecutor training.  We spoke with various stakeholders about myriad issues and, as Graham has said, we have not set forth what an algorithm might or might not be, in terms of coming to our recommendations and conclusions. 

In addition, the report does talk about the fact that, for example, the police station surveys do not appear to corroborate the reports that were told to us orally by a divisional police officer himself and also by a programme manager about how well a particular police station had been functioning, in terms of rebuilding the trust in the community.  When evidence is being given to us from myriad places that indicates one sort of direction of travel, and then a survey is done at low cost that indicates the opposite, our point is that those light‑touch surveys could be useful in terms of trying to figure out which is the correct answer.

Q12   Fiona Bruce: Could I just come back?  Would you now present aspects of this report differently, in light of the criticism that you have received, so that you could clarify perhaps some of the representations that have been made and the impressions given?

Mark Foster: Like all things, with the benefit of hindsight, there are things one might go back on.  First of all, we stand incredibly strongly behind the core of what this report is saying, which is on the need to step up action at all levels in the area of anti‑corruption.  If people read the report and understand what it is saying about the core recommendations and the evidence that that is based upon, we feel very strong and robust about that. 

It is clear that, by our efforts to show a number of innovative, we thought, triangulation approaches we were using to test our thinking in a number of places, and to expose that in the report, it meant that it went to some length in the annex about that.  That gave the impression that that was more of our evidence base than it was.  If we were to have our time again, we would condense the presentational part of that, if that would have avoided some of the apparent misconception about our approach. 

Q13   Fiona O'Donnell: Can I just quickly ask, because we have DFID coming in afterwards?  Did the methodology in this report vary from previous reports?  Did the standards you set for responses to surveys hit all the quality assurances tick boxes? 

Graham Ward: The report was quality-assured, in the same way as we have quality‑assured previous reports.  I take comfort from John Githongo’s cold review of the report.  We continue to believe that the conclusions and recommendations that we have presented to you are fully justified by the evidence that we gathered in the round.

Chair: We are running out of time, but I want to bring in Jeremy Lefroy and then a final question about Nepal.

Q14   Jeremy Lefroy: Very briefly, I entirely support the comments you made about the importance of this kind of work and that it needs to be looked at and needs to be challenged.  At the same time, would you not see that, actually, there is room for more work looking at the integration or the impact of higher‑level problems, sometimes corruption, on what is seen as petty corruption?  I will just give one instance. 

In my experience, and I will not go into the details, but towards the end of every month in many countries there is a problem that salaries are not paid on time.  As a result of salaries not being paid on time, there is an implicit almost directive to, let’s say, the police, “You had better go and get your salary yourself in your own way”.  Whatever kinds of safeguards you bring in, like saying to the public, “You must get a receipt if you are caught for speeding”, what often happens is the police with speed guns are conveniently situated a long way away from police stations, and people are then told, “To get a receipt, you have to go back 15 kilometres to get the receipt”.  Therefore, it is much easier if you do not have a receipt. 

There is a clear connection between the lack of proper financial controls at a higher level and the incentive to corruption at a lower level, with everybody getting in there.  Is there not a danger that, if we talk about grand corruption and petty corruption as two separate things, that actually it is a little bit simplistic?  That is not in any way to decry the importance of the kind of work that you have done. 

Mark Foster: First of all, we totally agree with you.  In fact, what the report says is not that we say that we should only be focusing on petty corruption and that there is not a connection between petty corruption and the bigger things that you mentioned.  It is quite the opposite.  Our view is that the whole system is a stack and a linked series of dependencies around how this all works.  Our only critique is that we think that DFID and other donors should be tackling the whole stack, in an integrated fashion, that is contextually aware of the country that they are operating in. 

Our sense is that, absolutely, if you think that the critical lever to push, in that case, is more policy at the higher levels of the police force around police pay, for example, or transparency over fining or whatever it may be, we would absolutely support that that should be a top‑down‑driven initiative alongside whatever you are doing to raise awareness among the citizenry that they should be holding the police to account and not accepting this stuff as a fact of life.  Our point is—and this is a very important thing—we were saying “and petty corruption”—more, not instead of the rest of the focus on the activities.

Q15   Chair: A few years ago, there was a governor in one of the northern Nigerian provinces who basically did a deal saying he would pay civil servants on time, but would be very, very harsh on anybody who took a bribe, bearing the point out that it can be done if there is the leadership from the top. 

Finally, we as a Committee are planning to visit Nepal, partly because of climate change and partly because of the corruption issue.  We can engage with you before we go with more details, rather than do it now.  You are saying that it is highly problematic for DFID to support Government systems that are known to be corrupt.  Does that imply that you think DFID should stop budget support altogether in Nepal, in those circumstances, or cut it or in some way suspend it?

Graham Ward: We are not making that recommendation, Chairman.  That would be a policy recommendation and it would be inappropriate.  What we are recommending is that, in the light of the facts, we believe DFID should consider whether or not they wish to continue.  We would then see that in the round, in terms of all the work that they are doing in Nepal.

Lance Croffoot-Suede: When the Committee or the Sub‑Committee goes to Nepal, it will be quite evident that the various donors in Nepal are ever present, in a positive way, longstanding, and have significant connections and relationships with the host Government.  The issue is: is there more leverage available to DFID and other donors in order to get host governments to take ownership of the corruption problem themselves?  It is about the grey area, in terms of there potentially being some constraints, self‑imposed on donors, in terms of not wanting to address corruption directly or forcefully at times, for fear that that will damage existing relationships.  The thought is that there is more leverage in these relationships for donors, not just DFID, to incentivise and motivate host governments to more seriously try to root out corruption within their own ministries.

Q16   Chair: The other thing is that, perhaps not surprisingly, this report has led to questions in the Nepalese Parliament, basically accusing DFID, I suppose, of precisely what the Daily Mail has, I get the impression, of fuelling corruption.  First of all, do you think that is justified?  Perhaps more positively, is there a role for Nepalese parliamentarians in taking this forward?

Mark Foster: Absolutely, and your visit and other things could be part of a process that is around what I would call a joined‑up view from the UK about this.  Back to this multi‑level thing, the more that Nepalese parliamentarians can take this seriously, be part of the change process, lead it and demand it, because their constituents are demanding it as well, the better that will be.  At the end of the day, that is how change will occur around this and that group has a very important role to play in leading on this.

Graham Ward: The role of parliamentarians in holding Government and officialdom to account is absolutely key.

Lance Croffoot-Suede: I would add that, with a greater focus on the poor and with innovative attempts to receive feedback from the poor, and the ways in which we attempted to do so, the issue that we discuss in the report regarding the Nepalese Government ministry and support by multiple donors would have been rooted out; it would have been addressed.  We met with poor people, time and time again, who openly, in open fora with 50 people present, told us about how they were being put in untenable positions, about how they had to resort to corruption and bribery in order to get the government support.  We then surveyed 3,000 such people and received similar feedback.  With a greater focus on the poor, with a greater focus on everyday corruption, the sorts of things that we talk about in the report would not happen.

Chair: We can all accept that, first of all, this kind of corruption hits the poor hardest and it is pervasive.  We know that.  The question, I suppose, is the extent to which DFID can and should be part of the process of addressing it.  We would agree, and ICAI should not hold back on that, that clearly we should be exploring more and that is what we will do with DFID, on that basis. 

All of us, in every country we go to, the one thing you come away with, which is the most depressing thing, is that there are many good things happening—quite often the tax base is rising, commitments are improving and there are good projects—and yet somehow the money does not quite get to where it needs to go, because it leaks all the way down.  As Jeremy says, poor people having to pay bribes they do not have the money for to get services that should be provided is clearly restricting development.  We share the need to address the analysis.  We now have an opportunity to see how DFID is going to react to your report, so thank you very much.  Perhaps we can liaise with you before we go to Nepal just to follow up the question.

Graham Ward: It would be a pleasure.

Examination of Witnesses

Witnesses: Richard Calvert, Director General, Finance and Corporate Performance, DFID, Liz Ditchburn, Director, Policy and Global Issues, Department for International Development, and Phil Mason, Senior Anti‑Corruption Adviser, gave evidence.

Q17   Chair: Good morning.  Welcome.  Again, we know who you are but, for the record, we would be grateful if you could introduce yourselves. 

Richard Calvert: I am Richard Calvert, Director General for Finance and Corporate Performance.

Liz Ditchburn: Liz Ditchburn, Policy Director for DFID.

Phil Mason: Phil Mason, Senior Anti‑Corruption Adviser.

Q18   Chair: Obviously you were sitting in on the previous evidence session, so you know what was being addressed.  Perhaps I could just ask you what your general reaction is.  I think it would be fair to say that, of all the ICAI reports, this is probably one of the ones that would cause you the greatest reaction and difficulty.  Perhaps I could just ask you first of all how you have reacted to it.  What do you feel about the impact of it?

Richard Calvert: The first thing to say is corruption is a huge challenge for all of us, and a huge brake on progress in reducing poverty, so the issue is an incredibly important one.  It is also an incredibly complex one.  In some of the reaction to the report, we have seen that it is difficult to dip into this area without provoking some strong reactions.  In the kind of study that ICAI has undertaken, it is impossible to do a comprehensive analysis, either of the whole of DFID’s work in this area or indeed of the whole landscape.  I do not think that ICAI would pretend that the report has attempted to do that. 

From our perspective, what it has done is provided us with a challenge on how we are approaching this area.  Corruption has been a huge issue for us for many years, and tackling corruption at all levels has been a big issue for us.  For us, the report targets some particular areas.  From our perspective, we would probably say there is a broader picture of DFID activity on anti‑corruption that we would highlight.  The report perhaps underplays some of the work we do internationally, some of the work we do in seeking to tackle the root causes of corruption and some of the work we do in the UK to tackle corruption.  Obviously we have understood this morning from the evidence that, from ICAI’s perspective, a lot of that work is implicitly endorsed.  What their report is doing is challenging us to do more in specific areas.  That is something that we broadly accept.  There is more to do in this area.

              There are also aspects of the report that certainly confirm things we would understand ourselves.  It is clear that the strength of the evidence base around tackling corruption is not as strong as it should be.  That is not a DFID‑specific issue; it is an issue in the international community and all donors would recognise that.  It is also fair to say that we would accept that tackling corruption is not, even within HMG, just a DFID issue; it is an issue for the Government as a whole.  We have been looking to work in a more cross‑Government way, both in partner countries and here in the UK.  The Government has been strengthening its co‑ordination and joined‑up nature of its activities in tackling corruption. 

              There are some really important and useful challenges in the report.  From our perspective, there is much that we do in this area that the report does not particularly address.  The fact that we have partially accepted the recommendations shows that, across all the areas addressed by the report, we believe there are some important points and some things that we should respond to seriously.  Equally, there are some areas where we feel the analysis probably does not give us a simple straightforward menu for action in the way that perhaps it might.

Q19   Chair: I do not know whether this is a fair question, but do you accept that, where you are working with local agencies, whether it is local government agencies or public agencies, in countries where this kind of whether corruption, lack of transparency or petty bribery is endemic, maybe you do not do enough, when you are running those programmes, to build into it some requirement?  You could almost be saying to your partners, “This has to be different, if we are your partner, than would be the norm”.  You cannot be everywhere if you are operating in a country like Nigeria.  That is the implication, it seems to me, from ICAI: that you need to build into your support mechanisms both commitments from your partners and some kind of follow‑through, that says it is time we began to get through to people that you do not have to pay a bribe.  Jeremy Lefroy’s point is that the people who are actually delivering the programme need to get paid on time. 

Richard Calvert: We would accept that that is incredibly important, but we would not accept that that is new for us.  Building in strong systems around accountability and ensuring that funds get through to where they are needed is a core part of our business and should be a core part of our business across all of our service delivery.  I will perhaps let my colleagues say a little bit more about that.  What we would not accept is the implication that our focus on tackling institutional corruption and tackling corruption at an international level has meant that we do not think about corruption and the impact of corruption at a local level.  The ICAI work does, to some extent, acknowledge that, but we would say that, actually, it should be a core part of our service delivery work.

Liz Ditchburn: If I can add and maybe complement on the initial question as to our reaction, one of my reflections, having thought a lot about this over the last few months, is that we think we are working in three ways and that all of those ways are inter‑dependent and need to add up to a coherent whole.  That includes the work that we do in countries, it includes the work that we do here in the UK, and it includes the work that we do internationally.  We conceive of that as an integrated package.  There are interdependencies within that.  We need to take that as a whole.  Particularly because the report is interested in the balance and the ways in which we do more of one thing or more of another, that whole is also very important. 

It feels to us, in terms of responding to the report, that the report took a rather different starting point, which is to say that we are interested in the impact on the poor, so let us look at what is happening in countries.  I suppose we would say that the totality of our work, in countries, the UK and internationally, we have to look at all of that through the lens of how it impacts on the poor.  The legitimate challenge and the reflection that we, as an organisation, are going through and thinking about is whether we can show better how all of those activities do indeed impact on the poor.  We know some things about how they might impact, but the report challenges us to do even more to follow that through. 

For example, if we are talking about recovered stolen assets, we need to follow through and we will be following through where those resources go.  Do they go back to the countries?  Are they used in ways that impact on the poor?  We would say that all of those things can and should impact on the poor.  That is the lens that we put to everything that we do; it is not just about what we do within country.  There is a sort of difference of world view almost that is implicit in the report.  We absolutely agree that everything that we do needs to impact on the poor.  You know our poverty reduction mandate; everything that we are doing is applied in that way.  The report is challenging us to think about that again to make sure that we are doing the right things that can impact on the poor.  What we disagree on is what the best way of achieving that is.

Q20   Fiona O'Donnell: This is the question I want to ask about, Liz.  You identify the approach that ICAI took in terms of impact on the poor, but why is corruption, as experienced by the poor, not discussed in your theory of change?  If we have this integrated approach, why is that not part of it?

Liz Ditchburn: It is in our theory of change.  Our theory of change talks about three things that we need to do to reduce corruption.  One is to reduce the opportunities for corruption.  That takes us, in terms of the evidence that we see, to a lot of work around systems strengthening.  For example, all of the work that has been described in the report and from DFID’s perspective on public financial management, on the kind of sector‑specific opportunities to strengthen systems, is one set of things around opportunities.  Incentives, changing the incentives, so looking at the kind of scrutiny that we can apply through transparency, is also a set of actions that we take, and then sanctions.  That is law enforcement, prosecution and pursuit, all the way through to recovery and return of assets.

Q21   Fiona O'Donnell: I get the triangular approach.  I am just still not seeing where this will impact on the poor as part of that framework.

Liz Ditchburn: All of those things apply to it as experienced by individuals as well.  That way of thinking about the world, in terms of opportunities, incentives and sanctions, applies. 

Fiona O'Donnell: It is that trickle‑down hope that things will benefit the poor, rather than actually having that at the core of your framework.

Liz Ditchburn: No.  For example, in something like service delivery in a health programme or an education programme, it is absolutely integral to that.  Giving people information about school budgets enables them to hold people to account to identify where procurement has paid more for textbooks than it should do.  That is working on opportunities and incentives, and it is directly impacting people in terms of their experience of that school.  Maybe Phil might want to say more about how the theory of change absolutely comes down to the impact on the poor.

Q22   Fiona O'Donnell: I am just conscious that we have a lot of questions and there is another part to this, which maybe, Phil, you might want to come in on.  You touched on this as well, Liz: the challenges of actually measuring the impact on the poor.  Are there any ways in which you feel that DFID can improve the way that it measures that, so that you have the evidence, so that we do not get another ICAI report like this?

Phil Mason: Measurement is an extremely difficult aspect of corruption, given the secrecy in which corruption happens.  It is always going to be problematic.  It is not like health status, where you can measure the health status of someone.  All donors would recognise that.  One of the areas that we have been trying to think about is trying to get some statistics that are actually what you might call directionally unambiguous.  In other words, when the number goes up, you know that that actually is a positive trend. 

One of the problems we have with measurement is that, if you are looking at, for example, the number of prosecutions an anti‑corruption commission has done, does a bigger number, if it doubles the number of prosecutions, mean it is actually getting twice as good at tackling corruption or is it just that more corruption has been surfaced?  Trying to find a way in which you can find a number that really gets to the heart of whether you are improving on the baseline or not is our problem.  The U4 and ourselves did do an evidence map in the paper that attempted to try to show what the evidence base looks like, and it is patchy, because context matters.  Every single context changes the way in which some of the interventions will work or be more successful, so it is a problem, but we are on to it possibly more than many other donors are.

Q23   Fiona O'Donnell: You would have heard me asking ICAI about their 2011 report.  I just wondered from your side why DFID is focusing more on fraud rather than corruption.  Why is there not a balance?

Richard Calvert: I do not think we would say that we are focusing more on counter‑fraud than corruption.  The 2011 report was a really important challenge to the way we thought about fraud in the organisation.  We have seen a real step change in the way the organisation has thought about protecting its funds, the work we have done on local strategies, on building a much more systematic approach to due diligence and strengthening our counter‑fraud capacity.  We have seen a real step change since 2011.  Our work on anti‑corruption has actually been an important part of the Department’s work for many years.  The 2011 ICAI report was not so much a challenge around that area of our work.  We have seen the anti‑corruption work scale up over the last few years, as with many other things, but the difference is on counter‑fraud work we had a real challenge to our approach in 2011, which we responded very significantly to.  The anti‑corruption work has been more of a steady expansion.

Phil Mason: We had for example, by 2006, introduced the UK‑funded police units, which we saw as an integral part, as Liz said, of the kind of work that the UK has to do here to protect poor people in developing countries, when the money gets stolen from their country and ends up in the UK.  That was early work that was quite innovative.

Q24   Fiona O'Donnell: Are you talking to the Treasury about the sharing of tax information?  Other countries have announced pilots; we have not done that yet.  How is that work progressing?

Phil Mason: It is certainly progressing.  We are funding quite a lot of the OECD’s collective global work on all of this.  The beneficiary ownership and the automatic tax information exchange agreements are exactly the kinds of areas that DFID is promoting as part of the settlement, as it were.

Q25   Fiona O'Donnell: Is there a pilot planned?

Phil Mason: I do not know the details about whether there is a pilot or not, but we are helping, through funding developing countries, ensuring that developing countries can benefit from the openness that this new regime is going to allow.

Q26   Jeremy Lefroy: On many occasions, DFID is actually either directly or indirectly responsible for delivering often quite large‑scale programmes to the poor, whether in health or education.  Those programmes will always run the risk of petty corruption.  One is then faced with the problem that, if you have come across that corruption, do you close the programme with a direct impact on the very people who we are helping or who we are trying to support?  How do you deal with that?  I have seen recently one example that brought about a real conflict between the need clearly to deal with corruption, absolutely, and, on the other hand, the very valuable service that is being provided to extremely marginalised and poor people, which is then taken away, because of the problems with corruption?

Richard Calvert: We have an overall approach of zero tolerance to fraud in relation to UK funds.  In a number of countries, albeit reluctantly, we have closed programmes where we felt that UK funds were at risk of being used for purposes for which they were not intended.

Q27   Jeremy Lefroy: If I could just intervene there, when that happens, how do you assess the really direct impact on the poorest from a programme being closed, which is perhaps not an additional programme but an essential programme, perhaps part of an overall government strategy?

Richard Calvert: As with all these situations, it depends on the context.  If we are putting funds through government systems and we no longer judge that the government systems are a secure route, then we can look at alternative channels.  We can look to work through NGOs or through alternative routes.  If it is a non‑government delivery channel, again, the judgment is whether the weakness is with the delivery channel and if there is an alternative way to achieve the same purpose.  What you cannot avoid is potentially some impact.  If programmes stop, there can be consequences, but what we would look to do is find an alternative route of getting to the same purpose.

Liz Ditchburn: That analysis of putting everything into the round is clearly what we have to do.  The report invites us, as we accept in recommendation 3, to consider the huge opportunity that these service delivery programmes represent for also strengthening systems and for building in work that is around anti‑corruption.  Again, we really believe that that opportunity exists and we often take it up.  We will look to make sure that we are always taking it up where we can.  That direct experience of poor people in their engagement with service delivery, we believe, is a really critical opportunity to do things like strengthening budget accountability, making more information available and building exactly the kinds of very basic, on‑the‑ground, often simple, but potentially quite powerful, things. 

Again, where we perhaps differ from ICAI is in saying that, yes, we should exploit all of those opportunities but, actually, in terms of the whole package, some of the other work we do is very important for that context too.  For example, if people in a country see members of their elites being prosecuted in the UK and assets being returned, we think that that is also an incredibly powerful signal, which impacts on how people then feel able to challenge within their own more local circumstances.  We see an interaction between all of these different levels. 

Q28   Chair: Can I move on to the Nepal section of the report?  According to the ICAI report—this is the Local Government and Community Development Programme, run through the Government and funded by DFID—citizens needed to pay bribes to government officials or to forge documentation in order to receive funds.  Apparently, they are saying, DFID was aware that government officials were producing false documentation or documentation falsely promising compliance, and aware that funds released late in the financial year created difficulties for beneficiaries.  There was media coverage about these and there were negative consequences for the poor.  That does not seem to be very satisfactory.  If your objective is to do no harm, you are actually compliant, if that is true, in actually doing something that is exactly that: it is pushing people into corruption.

Liz Ditchburn: The biggest point that ICAI makes here is on the idea that we might have developed some innovative ways of supporting citizens, so citizens being able to claim their entitlements more effectively without having to pay bribes, for example.  We have developed these innovative and effective ways of doing that, but we have not followed them through into government programmes.  That is the picture that we do not recognise, so we are really glad that ICAI has recognised that innovative work, both in LGCDP—sorry, I am only remembering the acronyms.  I cannot remember all the long names underneath them, but there are these two programmes in Nepal that have really developed these innovative ways.  Indeed, the whole aim of the programme is to take that learning, build it into government systems and improve the government systems through the block grants. 

Our money does not go into that block grant system directly, but we do want to influence it.  We think that that is indeed what we are doing.  The learning from the innovation is being taken up by Government.  We are also strengthening government systems even more at the systemic level, in terms of overall systems across Government, but we just do not recognise this picture of not taking the learning and building it into government systems.

Q29   Chair: When they say that you were aware that government officials were producing false documentation, is that true or not true?

Liz Ditchburn: The office is aware that there have been accusations around corruption within the block grants system.  Explicitly on compliance relating to donor funding, I am not aware. 

Q30   Chair: If they were aware, what did they do about it, and should they not have done something about it?

Richard Calvert: If we believed there was a threat to DFID funds, then clearly we would be in a situation—

Q31   Chair: That is not the point.  The point here is not the threat to DFID funds.  It is the fact that poor people are having to pay bribes in order to get access to the funds.  In terms of the British taxpayer, you can argue that the funds are delivering the programme in all those things, but in terms of the beneficiaries, namely the poor, it is at a cost of endemic corruption.

Richard Calvert: The strategy that Liz is describing and certainly which our office would recognise is that of course we recognise this corruption in the country and corruption in aspects of government funding.  The work that we were involved in was seeking to take some of the lessons of ways to strengthen local accountability and broaden that out into wider government systems.  From our perspective, we do not regard what we are doing as breaching any sort of do‑no‑harm principle; in fact, rather the opposite, we are looking to embed local accountability in wider government programmes beyond those that are directly donor‑funded.

Q32   Chair: Is there an opportunity here, given these questions were being asked in the Nepalese parliament, for you perhaps to revisit the programme and bring in some additional practice?  Indeed, are you planning to do that?  If we get there in February, are we going to be given that opportunity?

Liz Ditchburn: We believe that the approach we are taking of having developed this innovative work and making sure that that innovative work is taken up in broader government systems is the right approach.  I understand that the second phase of this programme has started fairly recently, so we think that this is a really critical time for us to be very close to that.  A large number of people were trained.  The numbers I have are 4,000 community activists and over 400,000 local people supported through that process to better be able to claim these entitlements, without risk of corruption.  What we will want to do now is make sure how many of those people who have been trained are now working within the new government phase.  That is an area that we want to follow up.  It is a question of following that very closely at this stage.  We cannot yet say that those people who have been trained are not involved.  We cannot yet say that those people who have been trained all are involved.  It is the question that we will be very close to and will be monitoring as part of this phase, as it kicks in.

Q33   Fiona Bruce: Turning now to Nigeria, there DFID staff told ICAI that model police stations have reduced corruption.  Do you now accept that, as ICAI found, they have made no difference and that the perception of corruption is increasing?  We have also heard and understand that the Justice For All programme’s own perception survey showed very poor trust in the police.  Should your staff not have known that corruption was not being ameliorated?  Even if they did not make changes at that time, what lessons has DFID now learned from the ICAI report on this issue and what changes will be made as a result?

Richard Calvert: What this shows, and we have alluded to this a couple of times, are the challenges of having really good data on what is working and what is not working.  In a country like Nigeria, a huge country much of it very remote, the experience of different people in different parts of the country is going to vary.  ICAI has used some of its local survey methodology in a way that challenges us on what we understand about the impact of that programme.  I do not think that either ICAI would claim or we would recognise that that is a comprehensive picture of the impact of that programme across Nigeria.  On the other hand, it underlines the difficulty that Phil has referred to as well of having really strong systems for gathering data.  Local accountability is a key part of all our service delivery programmes in Nigeria.  Getting better and more comprehensive data on that is and has been a priority for us.

Liz Ditchburn: I think ICAI would recognise that they are not presenting that particular piece of survey data as a full conclusion that nothing is changing.  What ICAI is doing, which we agree with, is saying that we should all look at whatever information is available to us.  We use all that information to reach a conclusion.  We are just finalising the latest annual review of that programme, and that will give the full picture, taking into account all of the data that we can access.

Q34   Fiona Bruce: Have you actually looked again at the systems that you are using?  You talked, Mr Calvert, about really strong systems and getting better data.  Are you going to improve your systems and your data collection, and will that be in this report, in the light of ICAI’s report?

Richard Calvert: There are things we want to do at project level, but there are also important things we want to do centrally and with other donors.  As Liz says, the team is looking at this project now.  We will certainly be looking at how they can get a strong picture of impact in future, but it also comes back to the challenge of developing a stronger research base to find ways in which we can improve measurement of corruption, particularly at a local level, which is something that we and other donors are committed to doing more. 

There are aspects of our work on corruption that are much easier to measure.  It is much easier to measure some of the work, for example, that we are doing with tax or revenue authorities to stop payment of salaries to ghost workers or to identify stolen assets reclaimed.  Those aspects are relatively straightforward to get some quantitative data on.  It is much harder at a local level, particularly in remote areas.

Q35   Fiona Bruce: Can I ask on a separate point that obviously so much corruption results from a lack of the rule of law and poor legal systems?  How much of your efforts are spent on improving legal systems within countries, right from the micro to the macro level?  Can I give you a very pertinent example?  The Committee has just returned from Tanzania and we were told, at local level, that whilst ivory poachers are caught—they may even be taken to court—prosecutions are rarely followed through.  We were told by a lawyer who is working in the country that one of the problems is that the laws generally there are unclear.  Often one has to advise on custom or practice.  We heard from businessmen who said that the problem is that we have an unpredictable Government, often imposing regulations at short notice.  We heard people saying that much of this ivory poaching is not only condoned, but there is involvement at the highest Government level or those connected with Government.  There is macro to micro there. 

It is a real issue about having clear legislation and having that enforced.  It is a real issue that affects the poorest because, if the ivory disappears at the current rate from that part of the world, the tourist industry will go and, with that, will go the wellbeing of an awful lot of people who rely on it.  What kind of approach will you take to tackle that and strengthen the systems involved to stamp out that corruption?

Richard Calvert: It underlines very well the top‑to‑bottom nature of the response to corruption, but Phil.

Phil Mason: Tanzania is a very good example, in fact, of how DFID has learned some of the lessons of the evidence mapping, which argues around the inter‑dependencies of some of our interventions.  One can spend years working to support a police force but, if you do not reform the judiciary, then all of that is not terribly useful.  In fact in Tanzania, the strengthening of Tanzania’s anti‑corruption action project does precisely this: it is a set of interventions that addresses six parts of that chain.  It helps to work with the anti‑corruption commission and the prevention of corruption bureau.  It works with the director of public prosecutions.  It works with the financial intelligence unit.  It works with the police, it works with the judiciary and it works with the national audit office. 

The whole point of all of that is that precisely the weaknesses in the system, for example the intelligence about ivory poaching—if you just get that but you do not reform the police or the judiciary, then it is not going to work.  The Tanzania office took the view that, if we are going to help Tanzania sort these problems, it cannot be a set of interventions in one agency; it has to be this menu.  This six‑office menu is the model to go.  It is one of the few in our programme that attempts to draw the threads between the chain.

Q36   Fiona Bruce: Unfortunately, things are getting worse, not better, are they not?

Phil Mason: The more you address corruption, sometimes more corruption then ends up coming through the courts and the police system.  It can appear that corruption is getting worse because, actually, more is coming to light.  One has to be very careful about judging the trajectory of a country’s performance.

Q37   Fiona Bruce: I do not want to take up any more time, but I am speaking specifically regarding the ivory trade.  Things are getting far worse.  Do you have a 10‑to‑15‑year plan to address that?  As you have just said, this is something that needs to be tackled over time. 

Phil Mason: The programme in Tanzania is innovative, in a sense that it stitches together action across the whole suite of offices.  I would be fairly confident that, if that were showing success, it would continue into a multi‑year programme.  All we know about transition and transformation on anti‑corruption is that it is a long haul.  Some of the studies have shown, particularly in European history in the 19th century, that this is a 40‑year activity to stitch together.

Q38   Fiona Bruce: In terms of recommendation 2, for example, you would accept that ICAI’s recommendation applies to such work.

Phil Mason: The long term is absolutely vital, yes.

Liz Ditchburn: What we are saying in our response to recommendation 2 is that, as we refresh these strategies which, as Richard said, were a really important response by the organisation to the previous ICAI report—those strategies were very much part of our response to the previous report—as we refresh them now and, over the course of 2015, we will be continuing to reflect, then we will look at this question.  We all agree that tackling corruption is a long‑term challenge.  What we want to look at in those strategies, as we refresh them, is how we can use a longer‑term timeframe more effectively and how we can express that in the strategies in a way that makes sense.

Q39   Jeremy Lefroy: ICAI says that evidence of effectiveness of anti‑corruption interventions is weak in terms of quality and quantity.  Is it not an essential part of your work and a condition of providing assistance that you know whether what you do is effective?

Richard Calvert: It is, and we have acknowledged already that some of the methodology underpinning good tracking of the impact of anti‑corruption is not there.  It is not strong.  It is not a DFID‑specific issue.  Within the international community, it is hard to track corruption really effectively and Phil has set out some of the reasons for that.  It is also an area though that, with other donors, we want to invest more in.  It is an important part of our research priorities.  We would expect to do more in that area, but that is not to say that we do not attempt to monitor the impact of programmes already.  Clearly we do but, in terms of some of the methodological bases for that, there is room to improve there.  We think that that is a fair observation from ICAI.

Q40   Jeremy Lefroy: We understand that you have commissioned some work on this since 2011 and the last report.  Can you share any of that with us?

Phil Mason: I can share the action that is going ahead.  There were two pieces of work.  There is the evidence mapping work in 2012, which our resource centre, U4, did with our research team, which attempted to set out what the literature did say about what works and what does not work, and that came up with the very clear picture that Richard describes.  There are no silver bullets and golden bullets in all of this. 

The second piece of work, which again we have asked U4 and ODI to look at, is some of the impacts of corruption on the poor.  This will certainly help inform the redevelopment of programmes to get back down to the grassroots.  That is imminent; it is being finished off at the moment and should be out, I would hope, in the first quarter of next year.

Q41   Jeremy Lefroy: What sort of impact do you think that will have on DFID programming?

Phil Mason: It will be core, in addition to the kind of day-to-day work that they can get from the U4, in terms of practical knowledge and experience of examples of what has worked.  This will hopefully set the frame for some of the issues that we have to think about when we are reprogramming.

Q42   Jeremy Lefroy: Your management response says that you have “developed a rigorous portfolio of monitoring and evaluation, including using beneficiary feedback, which [is] used to improve current or future programming … including anti‑corruption”.  Do you accept there must be some shortcomings, as ICAI saw no evidence of feedback being used to evaluate the impact of anti‑corruption activity or to correct existing programmes?

Phil Mason: We accept that there are always new ways in which one can monitor the feedback from beneficiaries.  We are very interested in looking at clever ways in which you can maximise the anti‑corruption effects.  That may well be through citizens monitoring their budgets locally.  It may well be through just getting a decent national audit office to do its work properly.  Sometimes, the kind of work that the randomised control trial evidence starts to show is that you can get both those outcomes, depending on the context.  A good national audit office that is predictable and actually does its work achieves a higher anti‑corruption impact than citizen monitoring, but it only works if you can absolutely guarantee that the national audit office will turn up and do the audit.

Richard Calvert: What we have seen across DFID as a whole, as the Committee knows, is a big investment in evaluation and in moving from a position, some years ago, where we essentially had a semi‑independent evaluation unit to a position now where we expect country teams to embed a strong evaluation programme across their programmes and take ownership of that.  Some of that will be long‑term impact evaluation; some of it will be shorter term.  That gives us a much stronger base, albeit some of that will take some years to deliver results, but it gives us a much stronger evidence base to understand the impact of what we are doing.

Liz Ditchburn: The only other comment I would make is, as this Committee knows, beneficiary feedback is something that ICAI has been very strong on from its beginning.  The NAO has also been encouraging us.  Over the last three years, we have been challenged very much to build beneficiary feedback far more systematically into our programmes.  That is what we are doing, but it is a work in progress.  We have developed a lot of tools and we have a clear requirement for all annual reviews to show how they have or have not used beneficiary feedback.  We are in the process of taking all that forward, rather than being able to say, “We did that 10 years ago and here is the evidence.”

Jeremy Lefroy: We certainly saw, in our visit to Tanzania recently, evidence of the impact of a good public accounts committee, which has perhaps received some support from the UK in the past or the present when, as a result of their report, Parliament voted for certain Ministers and senior officials to be sacked, as a result of IPTL, the power station corruption scandal.  What we also found was that citizens were glued to the television throughout the broadcast proceedings of that parliamentary session everywhere, I think we would say.  People were talking to us not quite in the street but, when they knew we were Members of Parliament, they were saying that it was very important that this kind of work goes on.  That was encouraging.

Chair: We were all spoken to individually by the hotel staff.  The line I got was: “Our Government is stealing our money.  What’s your Government going to do about it?”  I guess that it is a good citizens’ revolt, but they want some results too, I think.

Thank you very much.  We want to ask briefly some questions on the private sector, so thank you.  When we do go to Nepal, you can be forewarned that we will obviously want to have a look at these programmes, talk to some of the beneficiaries and see whether things have changed or, indeed, what they have to say about it. 

              Oral evidence: The Work of the Independent Commission for Aid Impact, HC 864                            2