Oral evidence - DFID's allocation of resources

Oral evidence: DFID’s Allocation of Resources, HC 533
Tuesday 9 February 2016

Ordered by the House of Commons to be published on 9 February 2016.

Watch the meeting – Parliament TV: Tuesday 9 February 2016

Members present: Stephen Twigg (Chair); Fiona Bruce; Dr Lisa Cameron; Stephen Doughty; Pauline Latham; Jeremy Lefroy; Wendy Morton; Albert Owen; Mr Virendra Sharma

Questions 108-158

Witnesses: Brenda Killen, Deputy Director, Development Co-operation Directorate, Organisation for Economic Co-operation and Development, Dr Paul Clist, Lecturer in Development Economics, University of East Anglia, Simon Gill, Director, Budget Strengthening Initiative, Overseas Development Institute, and Ben Jackson, Chief Executive, Bond, gave evidence.

Chair: Good morning, everyone, and welcome. This is our third evidence session in our inquiry into DFID’s allocation of resources. Can I welcome our four distinguished witnesses this morning? We have about an hour to take evidence from you, and we have 11 questions, so can all of us, both at our end and yours, bear that in mind in terms of length of questions and answers? Sometimes, there will be an obvious member of the panel to go to; some of the questions will lend themselves to a short answer from each of you. I am going to ask Fiona to kick off.

Q108 Fiona Bruce: Thank you, Chair. Good morning. Thank you for coming to see us. The evidence our Committee has received has been broadly positive about the new aid strategy’s focus on fragile states. Can you tell us what the main implications of this focus are for the way DFID works?

Ben Jackson: We welcome the focus and the 50% target. Our perspective is that there is a lot of devil in the detail to be clarified. That is one of the important things that we hope the Committee can probe on, because a number of implications flow from that. For example, the 50% target in terms of fragile and—as I understand it—conflict‑affected areas means a lot more emphasis on middle‑income countries, which, in fact, we have been calling for for a long time. However, it is quite a radical shift from the position that DFID had before of focusing on the poorest and graduation, and so on.

Q109 Fiona Bruce: That is a very interesting point, and I wonder how that risk of not focusing so much on poverty reduction can be mitigated.

Ben Jackson: There is a piece around geography: which countries is DFID now going to focus on? There are implications in the aid strategy, but we definitely need clarity about that. What mode are they going to be in? In the humanitarian area, it is not always cut and dried, but we understand: you get in, you sort the thing, and then you withdraw, so there is a geography piece. There is a purpose piece, which is also crucial. We get the point that, whether it is national security or prosperity, it is not that you eradicate poverty over here, and it cannot be coherent with that, but what is the primacy of purpose? That is the key thing that concerns us overall about the aid strategy, including the issues on security and fragility. We have four objectives in there, one of which is poverty. Which takes primacy? That is a second point that we need a lot of clarification on: what is the leading purpose of aid going forward? Otherwise we, you and others cannot judge its performance.

Q110 Fiona Bruce: I rather have the sense that this is a work in progress. I am wondering whether, Chair, it might be appropriate for us to ask if we could have a letter in a short time as to how that progress is developing.

Ben Jackson: For sure, yes. It slightly depends on when you are going to report, and we are waiting for some big reviews to come back from DFID, so the sequencing here is very important, Chair.

Q111 Fiona Bruce: Just a final point: the last time DFID scaled up on fragile states, ICAI criticised it for failing to recognise the ramifications, particularly in terms of timescales. Are you better equipped this time to scale up? Do you have a timeframe within your work in progress?

Ben Jackson: On scaling‑up, for example in the Middle East and Syria and so on, there have been huge changes in working in countries around the region, which in general has been super‑impressive. My members in civil society have also stepped up to the plate on that. DFID has shown its ability to do that, and the co-ordination of other Departments has been strong in many areas, but you are absolutely right: there is a question about how that is going to work on an ongoing basis. This is a new strategy. It is not just a Middle East response strategy; it is a general strategy. It is absolutely right to ask how that works, and how that is going to work across Government, which is the other crucial point here. By the end of the five years, 25% of British aid is going to be spent by other Government Departments. Potentially, that is really good news; it is exciting, but how will it be delivered? How is it going to be managed? How is it going to be co-ordinated, and how is it going to be held accountable?

Q112 Chair: Simon, do you want to add anything?

Simon Gill: I have three proposals from our experience of working in fragile states. There is a need for a longer‑term focus. Our experience, and one of the things we put in the submission, is that DFID staff’s tours of duty in fragile states are still short. There is lots of turnover; they do not necessarily get the best people in those places. We have teams in South Sudan; one has 30 years’ experience engaging with the region. Most have over five years’ experience, and you need people with that depth of knowledge and political understanding to make a real difference, and also people who have the trust and understanding of that context. That is one proposal. Despite the rhetoric, there is still a need to look at posting length in fragile states.

The second point that we make, which I think others on the panel have also made, is about an adaptable and flexible approach. Bringing in a blueprint approach, and saying you need best practice in a place like South Sudan, Liberia or the DRC is not the answer. You need programmes that are flexible and adaptive; the jury is still out on how you make those things happen. Although we are talking about tangible and demonstrable results, institutional support is still needed. I was in Sierra Leone twice last year. The ambition to deliver resilient and better healthcare in Sierra Leone, and helping the Government manage that, is dragged down by the fact that the Government still struggle to manage their own cash on a daily and weekly basis, which affects their ability to get predictable financing to the Ministry of Health. Yes, you need health interventions, but you also need complementary interventions in an institutional sense. I have other things to say, but I will leave it at that for now.

Q113 Chair: Briefly, Paul or Brenda, do you want to come in at this point, or do you want to wait?

Brenda Killen: Yes, just to agree with the comments that have been made. I particularly want to underline that this is not just about spending money in fragile and conflict‑affected states; it is about DFID’s ability to bring a conflict lens to all programming that is across Government. It is also about working with the multilateral system, and we have learned about the skills that DFID has to do this, and the importance of being flexible and able to work in regions. We know that conflict does not stop at the borders; they need to be able to work on the drivers of conflict, as well as looking at states. As some of you have said, the specificity of this really matters. We need more detail.

Chair: Brenda, that leads nicely into Albert’s questions.

Q114 Albert Owen: I want to concentrate on the definition of fragility. Brenda, you said in your written evidence that it is complicated and has multiple dimensions, and will change over time. My question is: how can the definition of fragility accommodate these complexities?

Brenda Killen: As you have said, it is complex. Breaking it down by the way in which fragility presents itself gets us away from this more old‑fashioned approach of, “You are either fragile, or you are not”. It helps to direct aid, other resources and other support to the drivers of the fragility and the conflict, and to design programmes around that. It can also be a much better way of getting towards building resilience and getting a poverty target, but understanding that not all fragile states are the same, and that the responses are not the same, is important.

Q115 Albert Owen: How do you think the Department can appreciate the various types and causes of fragility? Does it have to have a mindset change, as compared with the work it is already doing?

Brenda Killen: The Department for International Development has a great deal of knowledge of the different drivers of conflict and fragility. It needs to use the different drivers, and the fact that the SDGs recognise things like violence to build expertise in some of these thematic areas, but also to bring that to other parts of Government that are engaged and working with the multilateral system to help better tailor packages of assistance to different countries and themes. As I said before, this highlights the fact that they need to work across regions as well.

Q116 Albert Owen: Simon, you said in your written evidence that you would consider adopting a narrower definition of fragility. Many of us are concerned that DFID would have a renewed focus, which is too narrow, if it were to do that. Can you explain?

Simon Gill: Depending on the definition you use, many countries are fragile or can be described as fragile. Part of the remark is coming from the fact that it depends what DFID really wants to do. It could say that all the countries it is currently focused on are fragile, and a definition would justify that. We are concerned that, from our evidence, “fragile states” in a very narrow definition—taking, say, the G7+ definition of fragility, or their self‑defined collection of countries that are in in that group—are the countries that get less aid money per person. If you want to concentrate your financing on those who are most needy, a very narrow definition of fragility is better, but we recognise on the other hand that the causes of fragility are multidimensional and multi‑faceted, as Brenda has said. It depends on what you are trying to do, and your definition can justify a broad approach across a wide spectrum of countries. You can include lots of countries in that, depending on how you define things.

Q117 Chair: Does anybody else want to comment?

Dr Clist: From my perspective, DFID have come to approximately the right answer, but probably for the wrong reasons. It is very sensible to be thinking about fragile states; there is lots of good work by Tim Besley and Torsten Persson, built on by Ines Ferreira, trying to operationalise what we really mean by fragile states, and it is pretty clear that there are two dimensions. One is political violence; the other is lack of government effectiveness. I know the OECD is looking at multidimensionality. I think these two axes make a lot of sense and are quite a workable way of thinking about fragility.

If you wanted DFID to spend 50% of their aid budget in politically violent countries, that would probably be quite a bad idea. That would be an excessive focus on places that are much riskier to spend aid, and where spending at that level is potentially more likely to do more harm than good. For that reason, we should have a definition of “fragile states” that is not just a definition of fragile states but also includes a lack of government effectiveness—and DFID’s definition is actually heavily weighted towards lack of government effectiveness, rather than political violence.

Q118 Chair: Did you want to respond, Ben?

Ben Jackson: I have just a very small point. This is slightly more political, but our understanding is that the definition that has been put against the 50% is “fragile and conflict‑affected”. It is very material if you are Egypt, or particularly if you are Jordan. Everyone has their problems, but it would not be counted as that; it is affected by what is happening in the region. That is the political logic here, which is fine, but again, there is the point about what is under the radar and what is clear. A couple of years ago, we would have been saying that middle‑income countries like Jordan would be way off the list. If India has been crossed off the list, Jordan is, for sure. That is fine, but it is not clear. There needs to be some probing of the actual definition against the 50%. These are all more sophisticated points, but there is a basic point there about what countries are in.

Q119 Albert Owen: We will be asking questions, but we really need to know your answers.

Ben Jackson: My answer is that it is not clear.

Q120 Wendy Morton: Following on from that, the new aid strategy has made this clear geographical shift towards the Middle East. We have been talking about that; we have been talking about fragile states. We have received several suggestions about factors that should determine where DFID delivers aid. In light of the new aid strategy, I would like to know—I will ask you first, Ben—what DFID’s priorities should be when choosing which countries to work in, and what their strengths are over other aid deliverers.

Ben Jackson: This comes back to the point about primacy of purpose. There needs to be absolute clarity. We will definitely look for this in the review results that are coming. The overriding one is still the focus on how our aid programme can reduce poverty and support the fulfilment of the new SDGs. That needs to be the driving focus. In an area like the Middle East—in Syria or Iraq—the emphasis, given the context, needs to be on emergency relief, tackling conflict, and so on. That needs to be the key thing. It is partly about where the need is, and also our ability to add value and change, which was the point that was being made here. It needs to be a balance, because we need to show that we can have results. If you go too far down that scale, we would accept that you then go only to those countries where it is “easiest” to make that impact, and that is the balance that needs to be struck. Given the aid strategy, we accept a number of the points that have been made there, but it is really critical now for DFID to be clear about which countries they are operating in, and how that relates to the primary objective of poverty reduction and having impact.

Q121 Wendy Morton: Does anybody else want to add anything to that?

Simon Gill: Three things, actually. There is a myth that fragile states are too risky; that you cannot deliver results or make an impact. The World Bank research says, “Yes, you can”, and some of the projects in fragile states perform better than elsewhere. Secondly, some of DFID’s comparative advantage is not always geographic: it is in its various modalities, such as the way it supports civil society; the way it works alongside Government; and the way it works in a complementary way with multilaterals. One of the things that worries me is that sometimes DFID abrogates responsibility. It gives money to the World Bank to deliver, and there are instances where the UK can play a unique brokering, facilitative role alongside the Bank and others. I would not want to see DFID give that up by saying, “Well, in particular fragile states, we are giving money to the World Bank to deliver on our behalf.” The World Bank can do good things, but it cannot do everything, and it needs the balance that an independent broker can bring. The other thing is that the networking abilities that DFID has of being engaged in the multilateral system but also working bilaterally is a real strength.

Q122 Wendy Morton: I would like to follow up with a second question, which is that the new UK aid strategy also includes a commitment to end the traditional general budget support. I would welcome the panel’s views on that. What do you feel the implications are, and do you feel this is the right move for DFID? Maybe we could start with Brenda this time.

Brenda Killen: Yes, I noted that in the aid strategy. It is quite a switch from the position that DFID was taking in the past; they were very much an advocate of general budget support. The key thing here is to make sure that DFID finds a way of delivering aid effectively. General budget support is one way of working with country ownership, working through country systems, and being transparent and predictable. There are other ways of doing that without using general budget support. Certainly, when working more with fragile states with more concerns about systems, I can see the desire to move away from general budget support, but that does not take away the need to look at how the support maps onto the aid effectiveness principles and the principles that the UK has strongly been behind in the past.

Q123 Wendy Morton: Does anyone else want to add anything?

Dr Clist: It is a real shame to get rid of general budget support as an instrument. There are two really useful bits of context for that. The first is that most aid is not policy‑sensitive. If you are thinking about the volume of aid that countries get, it is mostly determined by population, their level of income, and some links with different donors. That is from a 2011 World Development publication that I have. Better governed countries do not, in general, in relation to their poverty level, get more aid.

The second bit of background is that most aid is very fragmented and very volatile. In that context, general budget support has those two advantages that we are looking for. First of all, I showed in a 2012 paper in the Review of International Organizations with Oliver Morrissey and Alessia Isopi that general budget support is given to better‑governed countries, so, if you are better governed, within the volume of aid that you will get, you will get a higher percentage of general budget support. A World Development paper written last year by Nadia Molenaers and colleagues showed that when there are things like human rights abuses, failed elections and so on, general budget support tends to be suspended. It is a policy lever that donors can use to influence outcomes in various different ways. The second thing is that, where most other aid is volatile, general budget support is predictable, it has low transaction costs, and it does not take too much time from the Government, so I think it is a real shame that that instrument is not an option, at the very least.

Q124 Stephen Doughty: Just to follow up on that, is there not a bit of a paradox in ending general budget support at the same time as increasing focus on fragility and fragile states? One of the reasons for general budget support originally was increasing long‑term state capacity to respond to the wishes of its citizens, and if you move completely away from that and go into that fragmented situation that you are talking about, you can actually sow the seeds of further fragility and insecurity going on. Does anybody have any comments on that?

Simon Gill: I had a previous incarnation with DFID. My first job in 1999 was going to Sierra Leone. DFID wanted to provide emergency support to Sierra Leone; we gave it in the form of budget support. We had safeguards: it was set against teachers’ salaries, and we checked the teachers’ salaries. It was very well‑documented and very carefully done. Last week, I was in Ethiopia. I know Ethiopia is on a slightly different scale from Sierra Leone, but we were looking at financial systems in Ethiopia. We went into the Ministry of Health, and the Ministry of Health’s finance department are managing 157 donor projects. They have financial reporting to do on each one of those projects.

We talk a lot from a UK perspective about how we manage the fiduciary risk of our funds going into countries like Ethiopia and others. In fact, our concern about fiduciary risk and what we need in terms of our own reporting potentially undermines the ability of the Ethiopian Government to manage its own money. There has to be a balance there. It would be a shame if budget support is not a tool that we can use, because in certain situations, if it is well‑managed, targeted, and safeguards are built in, it can have a really useful effect in the way that you cited, Stephen.

Brenda Killen: I would just like to come in and support my colleagues here in citing the benefits of general budget support. I hope it will continue to be an option, that that will be reconsidered, and I have been assuming in this that things like sector budget support are still on the table, so there is a possibility to work with institutions, strengthen those institutions, and give them an incentive to further strengthen their transparency and accountability.

Q125 Jeremy Lefroy: What is the approach of other major donor countries to general budget support at the moment? Are there any that are still doing it? If so, which ones?

Brenda Killen: There is generally a move away amongst the bilaterals from general budget support. Some donors are still using sector budget support—so, more tightly earmarking support through the budget to specific sectors like health—but, generally, there is a move away across the DAC members.

Simon Gill: But, Jeremy, in the EU, the “state‑building contract”—that is the term they use—is used extensively across about 90 countries, and that has all the characteristics of general budget support, so they do, and by implication we support them in that.

Q126 Chair: Why is there such a dichotomy between the approach countries are taking bilaterally and the EU’s approach?

Simon Gill: I am not sure I can answer that.

Dr Clist: Is it not the risk of political problems? Essentially, if you are giving general budget support, you are always worried that there will be a news report saying that some money has been misspent, and by implication you then funded that, even if it is at the margin—you did not fund anything. That is the split, and that is one benefit that multilaterals have: they can think more long‑term, rather than short‑term, on those costs.

Brenda Killen: They are removed from the immediate backlash in the newspapers, but they also can look at the evidence of how they can spend money quickly and effectively and achieve other aims, particularly strengthening institutional and state capacity, which is key in fragile states.

Q127 Jeremy Lefroy: Just very briefly, would one of the problems with general budget support not be—as we saw with Rwanda—that when it is removed temporarily, it causes very substantial friction in political relations between countries, more than would be the case with the removal of sector budget support or a specific programme, because it is seen as a direct attack on a country’s Government, as opposed to a part of the Government or an NGO?

Chair: I am just going to take Paul to answer that, and then we will move to the next question, because you mentioned earlier that that was an advantage of general budget support.

Dr Clist: It is a policy lever that you can use, and the vast majority of aid is not a policy lever.

Q128 Jeremy Lefroy: That would be seen as a colonial imposition. You call it a “policy lever”; some other people might call it interference.

Dr Clist: Specifically in the case of Rwanda, I believe all that money was repurposed in a different form. That is exactly what happened in Kenya around the elections; that is typically what happens. The money is given, just in a very different way, and so it is essentially a way of punishing the political elite by saying, “It is not going under the control of the Government”, but it is still spent in‑country. For me, it seems like something that can actually be used.

Q129 Pauline Latham: 10 out of 11 aid orphans identified by the OECD are also considered fragile. How can DFID co‑ordinate its efforts with other donors to be able to tackle the existence and persistence of aid orphans, and how should DFID support aid orphans?

Brenda Killen: I can take this, to start with. This really needs to be taken into account in looking at where DFID spends its money. Aid orphans are not important geopolitically. Afghanistan will never be an aid orphan, but countries like the Central African Republic or Chad become aid orphans, and as we have seen in the Central African Republic, something that was problematic but not of much concern 10 years ago is now driving challenges that are affecting the whole region and beyond. DFID needs to take this into account, look at those countries where it will have a bilateral relationship, and be much more open about where it should have a bilateral relationship, where it can bring added value, but also make sure that the multilateral system is efficiently resourced to be able to support aid orphans. As I said before, taking a regional approach as well within DFID’s programmes can help to ensure that those countries do not fall through the cracks.

Q130 Chair: We cannot have everyone on this one, because we are way behind time. Simon?

Simon Gill: There are things that you can do. I would, as I said before, worry if you just gave money to multilaterals to support aid orphans. There are aid orphans; I know that Stephen has a special interest in CAR, for example, as one. For the bit of ODI that I represent, we have been given an accountable grant by DFID; we provide flexible institutional support to countries. That could include providing financial management support to aid orphans and countries like that. DFID has a very successful Girls’ Education Challenge fund, which is a centrally‑run fund that provides flexible funding across a range of countries. There are modalities like that that DFID could look at. I do not think it is going to stretch all of its funding to every single bilateral country, but there are different mechanisms it could use to touch on aid orphans and support them in that way.

Q131 Pauline Latham: Do any of you think that the new definition of fragility might mean that we are less likely to see as many aid orphans in the future? No?

Brenda Killen: It is political.

Q132 Pauline Latham: Political from our point of view, or political from their point of view?

Brenda Killen: Political from the bilateral donor’s point of view.

Q133 Chair: Simon, you mentioned my interest in the Central African Republic; I think that is shared by a lot of other members of the Committee. Brenda, in terms of the bilateral relationship, do you think the French are rising to the challenge of support for CAR?

Brenda Killen: They are, but I hope that the UK engages with its partners to make sure that the full package is present in these countries, because the UK brings certain capacities and knowledge that not all the other bilaterals have. There can be a bilateral focus on certain countries, but a general concern that the fragile states—especially on a regional basis—are receiving the support that they need.

Q134 Dr Cameron: This question is initially for Paul, and then others can contribute. We have been told by Save the Children that payment by results incentivises achieving the easiest results. Is the idea of payment by results therefore compatible with the complexities of DFID’s renewed focus on fragile and conflict‑affected states, and how would it impact upon the “leave no one behind” agenda?

Dr Clist: That is definitely a concern that I have heard several times from various different NGOs; there have been conversations with procurement officers saying, “If you need to achieve results, maybe we can take out these easiest communities”. I have never heard it actually go through, but that is a very common concern. The most difficult part of doing payment by results in fragile states will be working out where it actually, genuinely has had a real effect. To see where payment by results has done well, we really need more than one data source. If we only get the one that is incentivised, there is always that slight question mark over whether results have been achieved. We can come onto that later.

Q135 Dr Cameron: How would that impact upon the “leave no one behind” agenda? Presumably, those are some of the more difficult projects or programmes.

Dr Clist: It very much depends on how you agree the tariff, and how well that can be done, but within any tariff, it will always incentivise efforts for the easiest people within whatever is being incentivised.

Q136 Dr Cameron: Does anyone else wish to comment?

Ben Jackson: Our concern from the NGO community is the impact, both in terms of effectiveness and, also, value for money. Payment by results is not the same thing. It might be one vehicle to do that in certain instances. It is very important, from the NGO point of view, that we underline that we are all for effectiveness. We are all for impact. We do a lot of work, and we admit there is a lot of work to be done on how we improve that, but the simple thing about evaluation is that you need to be clear about the objective and then the right measures to get you towards that.

Payment by results, in our view, is only a very narrow measure. There are simple things; it is partly about the complexity of impact, but also just timeframes. Through payment by results, you can incentivise building lots of wells. If a year or two years later, none of those wells are working and the contract has been paid, how is that good value for money? The other concern that we have is that it has a danger of biasing towards certain institutions—particularly those with a lot of capital—that can deliver and have the money to be able to do that. On average, none of my members, even large members such as Save the Children, do not have that. There is also charity law—all these things have to come into play so that you have that money sitting there, and risking that. That does not mean it is good value for money.

If you are paying for an expensive company or operator to do that because they happen to have the capital, it does not mean that we, the taxpayers, get a good deal. There is the effectiveness piece: let us look in a “horses for courses” way at measuring effectiveness in relation to particular outcomes, rather than saying, “Here is one methodology; let us ram that through.” That is not a good way to do things, and it raises some very profound questions about value for money.

Brenda Killen: The decisions about which method to use need to be driven by what is right for the situation, rather than by the vehicle. Payment by results is challenging in fragile situations where, as the aid strategy recognises, there needs to be room for flexibility. When we start the journey, we do not know exactly where we are going to end up, and we need to be able to react quickly. There is also a challenge on the aid effectiveness side. Bilateral donors have a terrible reputation for not providing the money they have committed, and often in results‑based programmes, the Governments do not fully understand that more resources are on offer; they just chalk this up to, “We must only plan to spend, say, 40% of donors’ funds”.

Simon Gill: I have three quick points. Payment by results depends on a very strong statistical base and a good monitoring and evaluation framework. That is often not the case in many of the countries you are most concerned about, the fragile states. Secondly, our experience tells us that blending work—working at the core, the centre, as well as at the sectoral ministry and at the local level with different actors—is important. Sometimes, payment by results forces you to work in a very binary way, and it is much more complicated than that. Thirdly, we are working in South Sudan and Liberia. Stuff happens—Ebola, conflict, all sorts of things—so development is not linear, and you need flexibility and adaptability to be built into that.

Q137 Dr Cameron: You have pre‑empted my further questions about flexibility and ability to adapt within the payment by results system. The other issue I was considering was, in terms of the level of engagement that DFID should have in monitoring programmes, should it just be at the results stage, or should it be straight through the programmes?

Brenda Killen: In fragile states, having the ability to keep an eye on what is happening right the way through is important. Payment by results works best if the donor stands back and lets the results be delivered, and that is the only way to learn whether PBR is working or not.

Dr Clist: The worst of all possible worlds is to have payment by results with heavy donor oversight throughout the project. You then pay all the costs of the bureaucracy, and all the costs of delayed payment, volatility, risk, etc.

Simon Gill: One of the things I worry about is that DFID has done good work on smarter delivery and better programming, but their M&E framework is still very rigid, and I worry that the rigidity of their M&E framework might squeeze the life out of some of the better delivery initiatives. That is a real concern of mine.

Q138 Jeremy Lefroy: One of the reasons why payment by results has become very popular—we understand that nearly 80% by value of all new centrally‑procured contracts between April and October last year were PBR—is because there has been this feeling that money is spent and there is no long‑term accountability. If you were to press DFID regarding what was being done with an awful lot of money 10 or 15 years ago, you would find it very difficult to find what impact from the money spent 10 or 15 years ago, except in a few cases of long‑term projects, is seen today. If PBR is not the answer, or only partially the answer, what is the answer to ensure that we know that money spent now is not just going to be viewed as a two or three‑year programme that, in 10 years’ time, will be completely forgotten and will have left no impact whatsoever, despite the expenditure of a large amount of British taxpayers’ money?

Dr Clist: One of the problems with payment by results is that it will often look like it is a success when it is not, because it is essentially a one‑way bet. If the results have not been met, you do not disburse anything. If the results have been met, you disburse everything, and therefore it looks like a success, and naïve evaluations will often claim success when there genuinely is not. I am happy to give examples of that.

In answer to your question, one of the answers is to be a little bit more pragmatic and realistic. If you take a longer view of how difficult development is, the vast majority of things are not going to have transformative effects instantly. Doing development is difficult; money will be spent very well in some cases, save lives and be transformative in other cases, and in other cases, despite best efforts, there will not be an effect. Sometimes, that is the best and most sensible thing to do. Especially in fragile states, you need that risk appetite.

Q139 Jeremy Lefroy: I understand that, and I accept that. That is a very good point. Would you agree with me that one of the problems is that if you take, say, a company with a balance sheet, its balance sheet tends to reflect what it has done over 10, 20, or 30 years? If you look at CDC, for instance, it has a balance sheet; you can go back and find out the investments it made 10 or 15 years ago, and how they are doing. In the case of DFID, we do not have that. Because we operate in a grant environment, the money is spent, and that is it. I have sometimes advocated that we should be saying, “Look, we have spent in Tanzania, let us say”—and I declare an interest as Chairman of the All‑Party Parliamentary Group on Tanzania—“£1 billion over 15 years. Let us have a look at what impact that has had,” rather than saying, “We have a £50 million project here, and a £20 million project here; let us get results out of those.” Then, we can take a much longer‑term approach to this, but we find none of this long‑term development expenditure accounting in DFID, in the way that we would automatically expect in a commercial organisation.

Ben Jackson: I hear what you are saying. That is a very interesting idea. To be fair to DFID, I do not think it is perfect, but there is strong evaluation and accountability at different levels. I do not think that the way they work is so different from a company. You can have something going through a balance sheet; they deliver the results of an aid project, and there is a bit more profitability. You can look at those accounts, but it does not necessarily build up the big picture. They do also invest a lot in the kind of work that is funded here, the research and evaluation and so on, which does look at trends and longer‑term things. None of that is to say that we can be complacent and sit back.

Sometimes, we over‑exaggerate the fact that we do not look at the effectiveness of aid. Yes, aid and the effectiveness of aid are complex; they are more complex than other things, so they are more difficult to evaluate, but there is a lot that is done already at those different levels. Every year, you have your reports and your log frame, and all that kind of stuff within that framework, and then you have higher‑level evaluation and so on. You make a valid point that perhaps relates to your point about what DFID itself is doing to aggregate some of that experience; that could be at country level, or it could be thematic.

Q140 Jeremy Lefroy: I understand, but if I take, for instance, the equivalent in France—the Agence Française de Développement—it has a balance sheet of €17 billion. It tends to do things through soft loans over a period of time. That means it can examine impact over a long period of time. I am not necessarily saying it could not be done with grants instead of soft loans, but it is able to look at the impact over a considerable period of time, whereas there are programmes that we were looking at in the last Parliament that have now finished, and that is gone, and we do not really know what has happened.

Ben Jackson: Yes, I agree.

Simon Gill: Jeremy, just to encourage you, the EC have done a suite of things around budget support. I do not want to drag us back into that discussion, but there are longitudinal studies of the impact of budget support in Tanzania, Uganda, and Mozambique over the last decade or so. I have been involved in a project that was completed last year, looking at PFM reform in Uganda over the last decade, which was commissioned by DFID. There are broader pieces of research and evaluations out there, and I would commend those to the Committee, because you might find them really useful.

Brenda Killen: There is an incredible opportunity now to build a long‑term framework around the SDGs and the indicators with the universal framework, to make sure that we have something that tracks over the long term and over the goals involved.

Jeremy Lefroy: I think trying to get DFID to look much more long‑term at impact is something for our report.

Chair: We will come back to that in a moment, actually.

Q141 Stephen Doughty: My question follows on quite well. The ICAI report on fragile states said that transformative impact in fragile states will take a generation to achieve, and is dependent on in‑country state capacity, which draws a little bit on what I asked earlier. From what we have just been saying about length of programming and so on, do you think, given that the average DFID programme is about three years, I think—and touching on what Simon was saying about the M&E framework and so on—that DFID is actually geared to deal with the type of long‑term programme that is probably needed in fragile states?

Simon Gill: No.

Q142 Stephen Doughty: What would be the optimum model, do you think?

Simon Gill: There are two things. We have programmes in fragile states, and having a slightly longer‑term horizon enables you to get good people to commit for a long period of time, and to build up those relationships and levels of engagement. Where we are operating, we have relationships that have almost preceded the programme, bringing in people who have a good knowledge. It is also about building up understanding with your counterparts in Governments, or in the Government. A five-year‑plus programme length would be much better. I had a previous incarnation in the Lao Swedish Forestry Programme. I joined in year 21 of that; it had gone through four phases. The Swedes had taught the Lao how to cut the trees down, and in phase 4, they were helping them replant them. Those sorts of projects make a transformational impact, but you are talking about two decades, and I do not see the evidence of those sorts of programmes in DFID any more.

Q143 Stephen Doughty: Paul, do you have any comments on this?

Dr Clist: I completely agree. A five‑year minimum term would be much nicer. Having people who are in‑country for much longer periods of time would be good. I am essentially echoing what other people are saying.

Q144 Stephen Doughty: How do you think DFID could better cope with the shift in political priorities that often comes with a change of Government? How can it ensure some degree of continuity, in terms of some of those longer‑term fragile impacts? I am thinking particularly of somewhere like Yemen. There was quite a bit of focus on Yemen back when I was there in 2008-09. A lot of that drifted away; it has come back now, right to the top of the agenda. How do we ensure that we do not lose focus on critical environments like that that have the possibility to go very wrong in the longer term?

Ben Jackson: This comes back to the point around the aid strategy, and the way that has shaken the glass in terms of these other objectives and the geographical spread. We need a much more thorough, clear and long‑term interpretation of what that means; you cannot do that in 18 pages. It is about a shift in geographical priorities.

Of course, there is emergency response, although emergency response is not quite as black‑and‑white as that; as we were discussing at the Syria conference last week, we are talking about employment and so on. We get that, but there is now a critical moment, and DFID needs to say, “So, these are the countries that we are going to look at. In terms of the new themes about prosperity and so on, these are the key objectives”—because they are hugely broad there—“which enable us to make commitments for the long term, and for other stakeholders, NGOs and others, to support and work with that.” If it is vague, it is really going to undermine the subjects of the two previous questions, around long‑term commitment, engagement, and accountability. That is a concern.

Q145 Chair: Brenda, in the OECD’s evidence, you highlight the potential for a conflict between the focus on value for money and the programming risk that is inherent in working in fragile states. How do you think DFID can strike the right balance between risk, on the one hand, and the value for money agenda that we have just been talking about, on the other?

Brenda Killen: With value for money, as with any number of other things we have been talking about, it is important to define what we are talking about and what we are trying to achieve. Value for money is not just doing things as cheaply as possible; it is about the most effective way of achieving a given outcome. It forces everyone to look, to innovate, to draw on the evidence and make properly‑informed decisions about how to deliver an objective, but it should not just be about cost‑cutting.

Q146 Chair: Does anyone else want to come in on the balance between risk and value for money? Brenda, do you think that on the ground DFID staff get this right? Do they have the proper support and training to get this right, or is there a “safety first” approach?

Brenda Killen: Certainly, what we hear on the downside is about the pressure on staff to fill out forms, get more reports written, and doing more analysis, rather than just taking some risks. Risks are things that can be measured, managed, and need to be balanced against the “do nothing” case, or the certainty of failure. There maybe needs to be a more open acknowledgement that there are risks, and that the success of a programme can be judged by how well failures or problems are dealt with, rather than avoiding them at all costs, which is not value for money.

Simon Gill: The way that DFID divides administrative and programme costs can cause problems. They are very tightly‑defined on administrative costs, so therefore they contract stuff out to consultants, because they cannot do it, and often that sets against value for money. DFID needs to build up its own institutional knowledge, but sometimes it would be cheaper if it did stuff in‑house. I started off working as a technical co-operation officer in the 1990s. I was a DFID‑funded TCO. That sort of job does not appear any more; it is done by consultants, so you are not gaining institutional knowledge for the organisation, which is what you need, particularly in the context of fragile states. I am not saying everything in the 1990s was brilliant, but if you go backwards, we were doing some things right then. We are not doing the same things now, for different reasons: Treasury rules, regulations, etc.

Q147 Chair: I would like to move us on now to look at some of the issues around policy coherence across Government, and policy co-ordination. With the new aid strategy, a greater proportion of ODA is planned to be spent by other Government Departments. Do you have thoughts on how DFID can ensure cross‑departmental policy co-ordination, but also coherence in this environment? Who would like to go first on this one?

Ben Jackson: As I touched on earlier, it is a big concern. To be clear, we are not saying that everything has to be spent by DFID, and that it is bad per se for things to be spent by other Government Departments. There are two or three different aspects to this.

First, starting with the last point, there is the technical expertise and the management capacity that needs to be there to make sure that money is well spent. That needs to be in place. Again, I am not saying that all knowledge lies with DFID, but DFID has done this for decades, and there needs to be a way in which—through secondments and those kinds of things—that expertise can be drawn on. The problem is that the central pool of capacity at DFID is very stretched, because in the strategy, there is further bearing down on those “overhead costs”. Personnel is one of the pieces.

It is not just about the pieces across Government; again, how are the big funds that have been announced going to be managed? They can operate, but focusing on just one, the Prosperity Fund, as an example, we really have some big questions there. All it says in the strategy is that it is to be managed by the National Security Council. The National Security Council does a great job, but it is not an operational vehicle of any kind. Also, why is the Prosperity Fund being run by the National Security Council? It has to be asked. There is also half a line about a new value for money committee, co‑chaired by DFID and the Treasury. How is that going to happen?

There are questions about the management and operationalisation—the real business of delivering things across Government. There is the co-ordination, so that it is the sum of the parts. We have the SDGs as the perfect long‑term vehicle to provide the framework, and I have to say to the Committee that we are very concerned about the way that that is expressed in the strategy. We are hoping that is just because it was written quickly. We have just agreed this new global framework, and it seems to get a third‑rate billing in there, to be frank. We are very much looking forward to that being reasserted and made clear.

Who is actually driving the different levers? There is definitely a propensity to have slightly separated funds run by the National Security Council—whatever that actually means in practice. There are some big questions there, and the problem with this is that not only is money being spent through those, but also that there is scope for money not to be well‑spent. Every time that a pound is not spent in the way it should be, the whole of the aid budget within DFID is to be held accountable, and that is the last point about accountability. We are maybe going to come onto that.

Chair: Thank you. You have anticipated both of my supplementary questions.

Ben Jackson: Apologies, Chair.

Q148 Chair: They were about the National Security Council and about the SDGs; I was going to ask them before I brought in other colleagues. In particular, I am interested in your views on the role the National Security Council will play, if you have a view on that, and secondly, how we can ensure that this is coherent with the sustainable development goals.

Simon Gill: I am not sure I can directly answer your question. Just from my experience in Ethiopia last week, they would value support from the NAO, say—they value hard‑edged skills—but the key point that I would make is that it has to be driven by demand from the developing country, not a pool saying, “The NAO have to go out”. DFID officers are going to have to act as gatekeepers—not in a bad way, but to channel requests for help, and then channel them back to the UK. The other point I would make on this whole agenda is that there is a lot in the aid strategy about transparency and the IATI guarantee. DFID comes out well in terms of aid transparency; other Government Departments in the UK come out badly, and I would want either your Committee or some Committee to hold other Government Departments to account, so that they are as transparent as DFID is. I am not saying DFID is perfect—it has further to go—but that is really important, in my book.

Chair: That relates directly to Virendra’s next question, which we will take in a moment. I just want to give colleagues a chance to respond.

Brenda Killen: Responding generally on these funds and processes to bring different parts of Government together, and having the whole of Government involved in delivering the aid programme and contributing to the SDGs, this has pros and cons. It is certainly in the spirit of the SDGs: universal, whole‑of‑Government, whole‑of‑society solutions are needed to very complex problems around the world. It has the potential to bring a development lens into the work of other Government Departments. We can look at the excellent example set by countries like Germany, who have adopted the SDGs as a universal agenda and looked in every single sector, from federal level right down to city level, at what they can do to deliver the SDGs. It is about bringing the outside world into this decision‑making. DFID has a role to play there, with the development lens. Like others, I was surprised to see so little reference to Agenda 2030 in the aid strategy. The aid strategy appeared to be a Treasury document. I would have expected this to be something either led by DFID, or taken from higher up. The Prime Minister was a leading light on developing Agenda 2030, so something could have been led from there. There needs to be a clear sign that all these parts of Government have something to add to delivering the SDGs, bringing the development lens in, and that what they are doing is to serve the delivery of the SDGs. As others have said, we need more detail, and we certainly need more transparency about how the decision‑making will take place.

Q149 Chair: It is fair to say that we share a lot of those concerns, but it is also fair to say that when we put them to Justine Greening, the Secretary of State, she has provided very strong reassurance around the centrality of the SDGs to her approach. However, there is little doubt that you are right: the Treasury led on that document, and DFID had an input into it.

Brenda Killen: There is just one more concern I wanted to raise.

Chair: Very briefly, because we are severely behind time.

Brenda Killen: Spending being shared across Government should not go hand‑in‑hand with a reduction in the staff in DFID, because the expertise that exists in DFID to guide what other Government Departments do is crucial to make sure that it is all contributing to the SDGs.

Q150 Albert Owen: Simon, you have touched on what I was going to ask about transparency. The MoD and the National Security Council are involved; by nature, they are secretive and methodical, and take a long time and are less flexible. One question that has not come up at all—and we have not had it in our brief, really—is DFID’s relationship with the devolved Administrations. We have projects led by Scotland, by Wales, and by Northern Ireland. My worry is that DFID’s approach might be, “They are dealing with that. Let us leave them to it.” I would be interested in your comments on that.

Simon Gill: They need to be really joined‑up and complementary. There are things that the devolved Administrations can do that DFID cannot do. It is about complementary roles; let us not all do the same thing. There may be unique things that Scotland can do in Malawi.

Q151 Albert Owen: Has thought gone into that?

Simon Gill: It is not apparent in the aid strategy. It does not come out.

Ben Jackson: To be fair, a lot of conversation goes on about how those fit in and so on. They are all so different in nature in the different devolved nations; in Wales, for example, there is much more emphasis on NGOs and so on, and there has just been a whole review of that and an attempt to pull that together. We have been working as a London‑based organisation to make sure there is co-ordination there. It is a very valid point; to be fair to DFID, quite a lot of energy is going into trying to make sure that they do add up.

Q152 Stephen Doughty: I was doing a bit of quick research the other day, and there are only seven countries that DFID supports where there is not also a UK defence‑led team—either the UK armed forces doing training or security sector work, or whatever. They are in every other country where DFID operates, bar seven. What is your experience? Particularly given this new focus on fragile states, do you think there is enough co-ordination between the MoD’s work in some of these countries and DFID’s?

Chair: I am just going to have one answer on this, I am afraid. Do you want to do it, Simon?

Simon Gill: I am not sure. I was in Ethiopia; they are all on the same compound. I am sure they are all fairly joined‑up, but I cannot really comment, I am afraid.

Chair: Does anyone feel they can? Do not worry; if you cannot, you cannot.

Q153 Mr Sharma: Partly my question was answered, but there is still room to develop that question further. It is wonderful that we have a universal agenda for everybody involved to deliver. Do the other Departments that are not used to delivery of that service have the skills, the resources, and adequate training? If they have, then that is wonderful. If not, how can DFID play a part in bringing others up to those standards?

Ben Jackson: Just in terms of the practicalities of this, the length of experience does vary. Covering other Government Departments, in certain areas—the Foreign Office, for example—it is very close, and in some areas of the MoD, there are quite well‑worked out patterns in terms of delivering humanitarian aid and so on. It is complex; it is not all the same. At the other end, you have the Home Office, in terms of newly‑arriving refugees and that money now being counted. As far as I know, there is very little experience there. We need a more nuanced approach that looks at that and the experience of management of aid, and how that is going to be accounted for.

That also relates to this point about transparency. In the aid strategy, it says that other Departments that are not at the highest level of IATI and transparency are going to get there in five years. I am sorry; my members, for several years now, have been obligated via their DFID grants to meet all that. We are very happy to do that, and we do a lot of work meeting that. There is a lot of work involved in doing that. To say that it will take five years for the MoD to get to those standards, and that that is acceptable, is not acceptable to us. It is out of kilter with what is expected of many other partners, like ourselves. The short answer is that a lot has to be done in terms of the skills; gearing up and saying, “You are now an aid‑delivering Department with international obligations. How are you going to get the right people, management, transparency, and so on?” Particularly in two or three of the Departments, the gap between where they are now and where they need to get to is larger than in other Departments that have done more of this.

Q154 Chair: Who else would like to come in on this one?

Simon Gill: What I worry about DFID is that hard skills from the MoD, FCO and HMT are useful, but the soft set of skills that those people will need to make them effective operators in a fragile state will need contextualisation and briefing from DFID staff. Brenda’s point earlier was that DFID are also constrained heavily; their ability to do that might be constrained, so if they are going to be facilitative in brokering with other Government Departments, that takes time and effort and energy, and needs to be recognised.

Mr Sharma: My own fear relates to what you initially said in answering another question: whether this is an opportunity for the consultants to get in. The people who are already employed need to be given the training and the skills so that we can have a long‑term attitude change in the Department. That was my fear: that the easy option might be to get the consultant, get the job done, then forget it. My way of thinking is that the training should be done for the existing staff members, so they can be a resource and an asset for the long run.

Chair: I will leave that as an assertion on all our behalves, rather than a question.

Q155 Jeremy Lefroy: Finally, on the OECD DAC rules—and this is specifically for Brenda—what do you think is going to come out of the high‑level meeting later this month, in terms of the definition of ODA?

Brenda Killen: At the moment, we are updating the areas in the ODA definition that are unclear, because it was defined 45 years ago. The two areas where there is still uncertainty are private sector instruments and aid spending on peace and security. We are reaching an agreement on a list of areas where aid being used to work with the military on peace and security is in line with delivering the SDGs—in particular, SDG 16. There has been a healthy debate around this and making sure that the line is clear, so that the primary purpose of aid is to deliver poverty reduction and economic wellbeing, and to deliver the SDGs. This is what we expect at the high‑level meeting next week.

Q156 Jeremy Lefroy: Does anybody else have any quick comments about the definition of ODA?

Ben Jackson: We do have concerns, along with our civil society colleagues in other OECD countries. You have to look historically. We have been talking about the long view; I have been doing this stuff long enough to remember the successful battles we have had over 20 years; Britain, under Governments of different colours, played an important role, for example, in terms of pressing down on the commercialisation of aid. That is the use of aid to win contracts, such as in the bad old days of the ATP, and all those kinds of things, where aid was used to win contracts because the French did it, and all that. It led to bad‑value aid and bad‑focused aid.

The problem is, particularly on the security front and so on, that you need to be very careful about how you widen that definition. I do not want to sound cynical, but unfortunately you have to look at those countries or those donors that would be looking at that and saying, “How can we use not the spirit of the intention but the letter of the law to say, ‘Let us open that out’—particularly in countries where aid is highly constrained—and say, ‘This piece of expenditure over there can be justified in that way.’” We do have concerns about that. It is not just about how British aid is used. If you change those rules, you then risk diluting the value of aid overall. It can be done, but we need to be very careful about it, and we do have concerns about what certain donors’ drivers are, in terms of getting that widening of the definitions.

Q157 Chair: Does anyone else want to speak on this one? Paul, do you want to come in on this one?

Dr Clist: No, I think I am fine.

Q158 Chair: Brenda, how significant a change do you think it is going to be?

Brenda Killen: At the moment, these are areas around the edges of the definition of aid where there are differences in behaviour, and so we need to get coherence across the donors. This is about defending aid; it is about setting a clear line. If it is fuzzy, we can have all sorts of things happening, and I understand my colleagues’ concerns. We need to be very clear in the area of peace and security, referencing this to the humanitarian principles, to make sure that this is driven by the right spirit. We then have other ways of monitoring whether that is actually happening in practice, through the peer reviews, but also through the aid effectiveness and development effectiveness work. We have a ministerial on that coming up at the end of the year, and I expect that there will be a focus on, “Within the new definition of aid in Agenda 2030, are resources being spent effectively?”

Chair: Can I say a massive thank you to all four of you? We have covered a hell of a lot of ground in just over an hour, and I am really grateful. Fiona, were you indicating you wanted to ask one more thing?

Fiona Bruce: No, I was only demurring at your language, which perhaps would be better not recorded, but I agree. Subject to that, we have covered a huge amount of ground.

Oral evidence: DFID’s Allocation of Resources, HC 533 7