Examination of Witnesses (Questions 1-19)
DEPARTMENT FOR
INTERNATIONAL DEVELOPMENT
3 MARCH 2008
Q1 Chairman: Good afternoon. Welcome
to the Public Accounts Committee. Today we are considering the
Comptroller and Auditor General's Report, Providing budget
support to developing countries. I have great pleasure in
welcoming the new Permanent Secretary, Dr Minouche Shafik; I think
you were appointed today.
Dr Shafik: That is correct.
Q2 Chairman: You are very experienced
in this field so I do not think you will have any difficulty in
answering our questions. Obviously this is a very important and
interesting subject. I think this country is by far the biggest
bilateral donor in Africa, are we not?
Dr Shafik: We are just after the
United States.
Q3 Chairman: Certainly much bigger
than other European countries. Obviously this is very important
to us. A third of all our aid to Africa goes through budget support,
which is what we are talking about today. Does it worry you, with
such large sums of money—£341 million of tax payers'
money—there is no conclusive evidence, as we read in this
Report, of its cost effectiveness?
Dr Shafik: Would you like me to
introduce my colleagues first and then I will answer the question?
Q4 Chairman: Yes please.
Dr Shafik: This is Sam Sharpe,
he is our Finance Director at DFID; this is Rachel Turner who
is head of our office in Mozambique and can give you a sense of
what it is like at the country level to manage budget support.
Q5 Chairman: So, there is no conclusive
evidence of the cost effectiveness of £341 million of tax
payers' money
Dr Shafik: I would like to give
a little bit of evidence to the contrary. Just to give a sense
of proportionality, DFID provides money in about 120 countries
around the world; we have significant programmes in 64 countries;
we provide budget support in just 15 so we are highly selective
in where we use this instrument and we only use it where we think
the benefits exceed the risks. In terms of evidence around benefits
of the use of those funds, I think the NAO itself says that in
seven of the eight countries in which we provide budget support
there is clear evidence of an increase in the quantity of services
that are provided by the government in terms of education and
health services and I can cite several examples of that.
Q6 Chairman: I do not deny it. If
you are providing this amount of money obviously you are achieving
something—nobody denies that—but traceability is weak,
is it not? I will ask you further questions in terms of support
and the State Audit Institutions, but the actual traceability
of the money is much weaker if you are giving it direct to the
government than if, for instance, you are building a health centre
directly when you would actually know the health centre is being
built. As we read in this Report, when giving to the governments
there are clear questions about how well we can trace the effectiveness
of this money. I do not deny it has some effect, but whether it
is cost effective is what this Committee is designed to look at.
Dr Shafik: The traceability is
different and the nature of the benefits is different. Rather
than having ten schools funded by the UK, the UK actually provides
funding to transform the entire school system; it could be providing
5% or 10% on transforming the system. We think the additional
benefits are worth it. We can show clear outcomes and benefits
that result.
Q7 Chairman: You mentioned schools,
let us look at one example. Please look at figure 9 which you
can find on page 15, the pupil/teacher ratio in Rwanda. What worries
me is that service quality is being compromised by service delivery.
We can see that pupil/teacher ratio in Rwanda in 2000 was 54:1,
it is now 70:1. I agree the net enrolment rate has gone up from
72 to 95 and that is very good, but they have an example there
of service quality suffering at the expense of service delivery.
Dr Shafik: That is a very common
phenomenon. In any country when you rapidly increase the quantity
of public services you often see a deterioration in quality. In
Ethiopia school enrolments went up from seven million children
to 14 million children in just six years. That is doubling the
number of children in the county's educational system in just
six years. Having said that, we are very focussed on the issue
of quality. In many countries when we see a deterioration in quality
we are engaged in it. In Zambia, for example, we work on increasing
the quantity and the number of children enrolled, but we also
have specific targeted interventions to address quality. We are
working with the Zambians specifically and narrowly in the education
sector on increasing recruitment of teachers in order to address
this issue.
Q8 Chairman: You mentioned Zambia,
the Report at paragraph 2.4 says that the data in Zambia is so
poor that you simply cannot monitor expenditure trends effectively.
It says here, "They could not produce budget figures which
were comparable from year to year as the Zambian Government was
introducing new budget classifications". You simply did not
know what was going on. That was a particular answer that you
have just alluded to.
Dr Shafik: Yes, but in that case
the Zambian Government was switching to a more appropriate and
a better quality classification system so we could monitor what
was happening under the old system but not under the new system.
I can assure you that now the Zambian system is able to generate
data in the appropriate IMF classifications.
Q9 Chairman: At paragraph 4.14 on
data availability and quality the Report tells us that often national
statistics are so weak that it is very difficult to monitor outputs.
Dr Shafik: Yes, although the NAO
also does say that DFID provides more support to building high
quality statistics in poor countries than any other donor. We
do more than all other donors combined in Africa.[1]
We think the solution here is not in DFID itself building individual
statistical capacity in countries, but in finding a multi-lateral
solution. That is why we are working with the World Bank and are
giving them £40 million to try to sort this because with
data the key issue is comparability and consistency across countries.
We are working on that in many countries.
Q10 Chairman: I am sure that is right,
but if you look at appendix five at the back of the Report you
see these interesting traffic lights which tell us what is going
on. It says that in some cases budget support has actually worsened
the general situation. Look at Malawi for instance, that has two
red traffic lights; that is not very encouraging, is it?
Dr Shafik: There are also a very
large number of green and yellow traffic lights.
Q11 Chairman: Yes but it worries
us when it says on two occasions that it is actually making it
worse.
Dr Shafik: I think in the case
of Malawi the issues there are particularly around macro-economic
stability and the fact that government did not have a grip of
the macro-economic situation and had very high levels of domestic
public debt. I think what budget support has done in Malawi is
to provide stability on the macro-economic front and I think the
Report itself acknowledges that Malawi was one case where budget
support played a key role in macro-economic stability.
Q12 Chairman: You mentioned Malawi,
but look at Uganda, figure 13, "Variable progress in public
financial management benchmarks". The number of benchmarks
not met in 2001 was six out of 15; the number of benchmarks not
met in 2003-04 was seven out of 15.
Dr Shafik: That is correct. We
actually held back some budget support in Uganda when we were
concerned about slow progress on public financial management and
the government had lost a grip on expenditures. When we are not
satisfied with progress we take action. I can give several other
examples such as Ghana and Sierra Leone where we have taken similar
action when progress has been inadequate.
Q13 Chairman: I think that one should
start with support of civic governance. At paragraph 3.15 it says
that: "historically DFID has placed less emphasis on domestic
accountability to parliaments and State Audit Institutions and
so had supported parliaments in just 20% and State Audit Institutions
in 13% of such cases". We are often joined in this Committee
by PACs from Africa and we often find that there is a grave shortage
of resources to State Audit Institutions and to committees like
this. It worries me that you had only supported parliaments in
just 20%; is this not a good place to start?
Dr Shafik: I think it is fair
to say that in the past, when DFID started doing budget support
it did not pay enough attention to building these institutions
of accountability but I think in the recent period that has changed.
If you actually look at the numbers provided in the NAO Report
they show that we are more likely to build accountability institutions
in countries where we are doing budget support than where we are
not. In fact, we are doing that in several cases. In Ghana and
Zambia, for example, we have actually supported the televising
of PACs to the public. I think the NAO itself saw that that resulted
in much higher levels of public awareness of PACs.
Q14 Chairman: I have one last question
and that is on corruption levels. You were obviously right not
to provide assistance in Kenya given what is going on there, but
if you see that on the estimate of corruption levels you have
actually got Sierra Leone and Cambodia worse than Kenya and yet
you go on giving them budget support. Is there any consistency
in what you are doing?
Dr Shafik: The consistency is
around assessing risks relative to benefits. We do not have a
threshold of fiduciary standards at which we will not provide
budget support. What we do is assess whether the benefits here
are worth the risk. To give an example, Sierra Leone is a very
high risk environment, a post-conflict country. We gave budget
support and the question is why did we do that? We know from international
experience that the cost of avoiding one conflict in economic
costs exceeds the value of total aid in the world. To save 30
or 40 or 50 billion dollars of economic damages by avoiding a
conflict is actually quite a good investment. In Sierra Leone
we judged that giving budget support would stabilise the situation
and give the government an ability to stabilise after the conflict.
That investment was worth those benefits even when we were taking
risks.
Q15 Chairman: So your lack of budget
support to Kenya has stabilised that country, has it?
Dr Shafik: No. We do not need
budget support in Kenya to keep Kenya stable; it is not the same
sort of conflict that we faced in Sierra Leone or Rwanda.
Q16 Mr Touhig: Dr Shafik, your Department
has moved away from delivering aid to project based schemes to
channelling support through national governments. Why is that
more effective?
Dr Shafik: We think it is more
effective for several reasons. First, when you provide support
at the national level you have the opportunity to shape national
policy and the overall allocation of the government's resources
and not just the individual projects that we fund. Secondly, it
gives us an entree into engaging in much broader issues around
public financial management, accountability, audit, the role of
the PACs. When we are doing our own projects we do not have a
mechanism to improve the quality of government systems. Lastly,
we think that working through government systems is actually more
cost effective. I will give you an example from Mozambique. In
Mozambique the average government school costs 12,000 dollars
to build; a donor school on average costs 23,000 dollars to build,
so twice as much because you are having to spend time on foreign
consultants and all sorts of overheads that donors require. We
think there is a strong cost effectiveness element there.
Q17 Mr Touhig: What admin costs did
you incur under the old system that you do not incur under this
system?
Dr Shafik: The NAO Report does
actually show the trend in admin costs for DFID as a result of
moving to budget support and it shows clearly that while the initial
set up costs of budget support are high, they diminish very rapidly
over time.
Q18 Mr Touhig: Many of the countries,
as the Chairman has alluded to, have pretty weak systems of financial
management and accountability. How are you ensuring that the funding
is going to where it is intended? There is some doubt cast on
that in the Report.
Dr Shafik: We have a variety of
mechanisms and I may turn to our Finance Director to provide you
with more detail on that. We do look at a variety of mechanisms.
We have results frameworks for everything we do to make sure that
there are clear results to be delivered. We do fiduciary risk
assessments which assess on about 15 different indicators of fiduciary
risks how well that country is doing. Appendix seven lists the
things that we assess in terms of how well the government is using
the money and you will see that for 15 indicators in nine countries
only one has actually deteriorated. On 134 indicators the levels
of fiduciary risk and the government's use of money is either
stable or improving in terms of the trend. Sam, do you want to
say anything more on the methods we use?
Mr Sharpe: Maybe just to highlight
what is covered by fiduciary risk assessments. They look at how
the planning and budgeting system goes in the country, whether
the budget is comprehensive for all government expenditures, how
it is reported and recorded through the system, how procurement
goes, how accounting goes and how audit goes. It is a pretty thorough
assessment of the level of financial capacity in these countries.
Q19 Mr Touhig: It all seems rather
mechanistic and I cannot quite see that you can check that there
is actually a school there at the end of the day? How much of
this money ends up in the pockets of corrupt officials?
Dr Shafik: We have a variety of
ways of checking that there is a school there at the end of the
day.
1 Note by witness: Refers to Sub-Saharan Africa
in 2005-see NAO Report paragraph 4.15 Back
| 
