CDD/ODI Policy Brief No.

2
CDD/ODI Policy Brief No. 2, November 2005

The politics of Ghana’s budgetary system

Tony Killick

This paper offers a political-economy independent sources confirm the existence of large
analysis of the expenditure side of Ghana’s discrepancies.
budgetary system. It starts by setting out • There were strong systematic biases in these
evidence on the workings of the present outcomes. The budget under-estimates spending on
system and then asks why the revealed large salaries and over-estimates everything else. These
failings are tolerated. It offers a political biases not only show a malfunctioning of the budget
answer to that question and then sets the system. They are also profoundly anti-
situation in the wider context of the developmental, for it is various non-salary
condition of the public service and the expenditures which determine a ministry’s ability
reasons for its continuing parlous state. It to deliver outputs to the public. There are generally
concludes by speculating on what might huge shortfalls in actual spending on such items.
be the drivers of budgetary change in • There is evidence of large leakages in allocated
Ghana. funds between their release from the centre and
arrival at the point of service delivery. A pilot tracker
The main themes of the paper are (a) on
study of expenditures by the Ministries of Education
the expenditure side, the budgetary
and Health found that an average of only 51% of
process is so weak as to be essentially
the non-salary resources which the Ministry of
ritualistic, with limited bearing upon
Education thought had been allocated to any given
reality; (b) there is a major ‘democratic
primary school actually arrived there. About half
deficit’ in budgetary processes which
of non-salary monies leaked out of the system. The
helps explain why such a weak system
Policy Brief

position in Health was even worse. For non-salary

continues to be tolerated; (c) the
items, clinics suffered a leakage of no less than 79%.
problem is compounded by the
Because of this, patients were forced themselves to
continuing system-wide deficiencies of
pay more than half of the total cost of their treatment
the public service.
– much more than the Ministry had intended.
The expenditure budget as façade • Lastly, there was no clear evidence of an improving
A government’s ability to run its own trend in the situation described above.
budget is perhaps the most basic test
of the capabilities of the state. How One question which arises is how the inefficiencies just
well does the contemporary Ghanaian described can be reconciled with the proven ability of
state pass this test? Evidence governments to maintain overall expenditure control
specially gathered for this study finds for macroeconomic management. While the fiscal
that: record in this area is patchy, successive governments
have shown that, when they have a mind to do so, they
• There are regularly large are able to avoid inflationary deficits. In particular, the
deviations between budget estimates NPP administration has done this quite well in most
and actual spending. For the two years. The answer is that overall control is reconciled
principal Ministries for which we with inadequate expenditure control systems by cash
managed to get data the mean budgeting. Expenditures are only authorised when the
deviations were ±42% (Education) exchequer actually has the money to cover the spending.
and ±68% (Health). Other This is a device which has been used elsewhere too
Page 1
 CDD/ODI Policy Brief No. 2

and is a sensible way of coping in the face of a weak Box 1: The Budget Process
budget system. However, it is also, by common consent,
The budget process is currently made up of eight steps,
a highly inefficient way of ordering the public finances.
beginning in May and ending in November.
It means that line MDAs are unable to plan ahead and
have little or no control over the resources allocated to Step I: Macro-Economic Framework
them. The development of a macroeconomic framework which
forecasts aggregates like GDP, domestic revenue and the
The alarming situation described above suggests that availability of donor resources. It primarily involves
the expenditure budget is largely a ritualised façade, personnel in the Ministry of Finance and Economic Planning
bearing little relation to actual state spending. In the (MFEP), assisted by Ghana Statistical Services, the Bank of
most basic of its tasks – running its own budget – the Ghana, the Planning Commission and the Revenue Agencies.
state is revealed as seriously deficient, a defect which
Step II: Policy Review
has persisted over successive governments. The budget
Simultaneously with Step I, MDAs undertake a review of
had more meaning in earlier decades but has since
their policies, objectives, outputs and outcomes in order to
deteriorated. How might we understand the forces estimate their broad expenditure requirements. Reports are
underlying this deterioration and the reasons why submitted to MFEP and these constitute the information base
present weaknesses are tolerated in an otherwise rather for Step III.
sophisticated society? Why are there not stronger
pressures for improvement from within the polity and Step III: Cross-Sectoral Meetings
Groups of MDAs are brought together to discuss their policy
civil society?
review reports in order to achieve coordination among them.
Why are the failings tolerated? The Chief Director of a lead ministry is responsible for the
production of a report to MFEP.
First, there is the closed, non-transparent nature of the
budget preparation process. There are currently eight Step IV: Ceilings
stages in this process (see Box 1). MDA expenditure ceilings are determined by MFEP in the
light of revenue estimates. These ceilings are conveyed to
As can be seen, these are conceived entirely as closed- MDAs as part of the guidelines for the preparation of the
door, intra-governmental processes, from which the MDAs’ budgets.
wider public and interested non-state parties are largely
excluded. Step V: Review of MDA Strategic Plans and Budget
MDAs review their strategic plans including outcomes,
What about Ghana’s Parliament? Does it make good objectives and costs.
the democratic deficit? Parliament, of course, has to
approve the budget and it has constituted a Finance Step VI: Prioritization and Presentation of MDA Budgets to
Select Committee to undertake detailed scrutiny of MFEP
MDAs prioritize their activities and prepare their estimates
budget proposals. By general consent, however,
to fit within the guideline ceilings.
Parliament has been unable to exercise effective scrutiny
and control. Timetabling is part of the problem: the Step VII: Budget Hearings
budget comes to it too late for detailed consideration. MFEP conducts budget hearings with each MDA to ensure
Another part of the problem is shortages of the that their estimates are within the ceilings and are consistent
supporting resources that would be necessary if with their strategic plans.
Parliament were to be able to do its job properly. It is
Step VIII: Finalization and Approval of Estimates
unable to hire expert advisers and does not have enough
MFEP submits estimates to Cabinet and then to Parliament
information with which to do the job properly. In for approval. The deadline for submission to Parliament is
consequence of these deficiencies, Parliamentary 30th November.
scrutiny tends to be hurried, superficial and partisan. Any consultations with non-government parties occur outside
the above processes
Inadequate flows of information are not just a problem
for Parliament.

Page 2
 CDD/ODI Policy Brief No. 2

The tradition has been for officials to keep information in-built forces for self-improvement. Moreover, it
close to the chest and to look with suspicion on those should not be taken for granted that a government
asking for it. Budget documents have tended to be would feel moved to take action even in the face of
released only on a ‘need to know’ basis. This explains criticisms. For example, when the reports of
why those in charge of the country’s schools and Parliament’s Public Accounts Committee are
medical facilities typically do not know to what budget submitted, remedial action is slow at best.
allocations they are entitled. Civil society organisations
A political interpretation
complain that information is withheld and of non-
cooperation from MFEP officials. Political students of Africa have made heavy use of
the concept of the neopatrimonial state, i.e. one in
Without necessary information and with non-
which there is only a weak sense of the public good
transparent processes, low public awareness of the
or of public service, and where the resources of the
deficiencies of the budgetary system is not surprising.
state are at the disposal of the president and his
Nor, therefore, is the absence of public pressure for
ministers. As explained in Policy Brief 1 in this series,
improvements. However, it would be wrong to suggest
observers have described Ghana’s state in this
that nothing is changing. Outside the formal processes
language, drawing attention to the way public
sketched in Box 1, there is some consultation with
resources – jobs and the power to allocate rents,
expert individuals and organisations by the Minister
provide services, and determine policies and their
of Finance. But while these consultations may be
beneficiaries – are captured by personal or private
beginning to become established as a regular – and
networks in the hands of dominant ‘patrons’.
expected – feature, they are not yet institutionalised.
They are also almost entirely Accra-based. There is no Now consider the condition of the budget just
scope for District authorities to have direct inputs, even described. We can ask, what kind of budget process
though much of their revenue comes from the central is likely to be most agreeable to politicians working
government. There remains a ‘democratic deficit’ here. within the patrimonial tradition, in which public
The budget process has largely failed to adapt to the resources are used to reward supporters and for the
country’s democratisation over the last decade or more. pursuit of private advantage?
This contrasts with the situation in otherwise First, it will be a closed system – as it remains in
comparable countries, like Tanzania and Uganda, where Ghana today – because the less that the public knows
budget hearings are far more open and transparent. about what decisions are being made and why, the
There are other signs of improvement. More easier it will be to utilise public resources for the
information is gradually being made available. exercise of patronage, and the less the risk of being
Interested parties are being given access to additional held to account. Second, the budget will be ritualised
data. The annual Budget Statement has become more – again as it is in Ghana – meaning that the formal
informative and efforts are made to distribute this processes of budgetary expenditure planning have
nation-wide. There is a budget web-site. It is said that limited bearing upon how public monies are spent in
the standard of Parliamentary scrutiny is gradually reality. Maintaining the ritual is important, so that
improving and becoming less partisan. taxpayers will believe their monies are being used in
their interests and are accountable. In an aid-
These, however, are improvements at the margin.
dependent situation, the façade is particularly
Essentially, the process remains closed and difficult to
important in order to keep the donors happy. But the
scrutinise. Getting hard information is still difficult.
reality of ritualisation is that ministers and high
Those living outside Accra are particularly at a
officials are able to set aside what the budget says
disadvantage. Given continuing poor levels of
and dispose of public monies according to quite other
education, people’s ability to contribute meaningfully
decision processes.
to processes of accountability is also questionable,
particularly in more rural locations. This is not, in other The view expressed above may be regarded as
words, a situation which is likely to generate strong excessively negative and it should indeed be
Page 3
 CDD/ODI Policy Brief No. 2

emphasised that there are signs of improvement. On posts in the service, due to the failure to address the
the other hand, analysis in terms of a patronage-based fundamental issues of pay and conditions’ and refers
model of politics helps to explain two other features of to ‘the extremely weak human resource management
the contemporary scene. First, we have noted earlier of what capacity … is left within the service’.
the inflexible structure of expenditures, with salaries
That part of the reform agenda relating specifically to
and other personnel costs tending to take priority over
the MFEP is itself an umbrella of programmes, called
expenditures which would permit MDAs to promote
the Public Financial Management Reform Programme
development objectives, with salary expenditures
(PUFMARP). This has ten components, no less,
consistently over budget, usually by large amounts, and
including programmes for budget preparation through
the rest squeezed to accommodate this over-spending.
the introduction of a Medium Term Expenditure
The point here is that such biases are what a patrimonial
Framework (MTEF), the improvement of expenditure
model of politics would predict: the giving of
and cash management systems, procurement reform,
precedence to the award and protection of civil service
fiscal decentralisation, etc. In some of these areas there
jobs over the longer-term promotion of development.
has been genuine progress, for example in
The public service problem strengthening cash management and new legislation
on public procurement procedures. Overall, however,
Looking further for explanations of the poor condition
it appears generally agreed that progress has been
of the budget, the depleted condition of the civil service
sparse and slow.
has a strong and direct – if ultimately superficial –
bearing on the problem. The Government’s Ghana The problems of the PUFMARP programmes have
Poverty Reduction Strategy is no doubt right in referring been similar to those of the wider range of public sector
to ‘unacceptably poor conditions’ in the public service reforms. They have largely been the brain-children of
and in stating that, ‘Significant improvements to the aid donors, leading to the usual complaint of weak
latter represent a component without which the political commitment. They have been fragmented and
government reform programmes and the Ghana Poverty ill-coordinated. In aggregation, they have been too
Reduction Strategy are unlikely to succeed’. ambitious, exceeding the implementation abilities of
MFEP and other agencies involved. In order to avoid
The numerous reviews and evaluations that have addressed
the difficulties of political commitment and skill
the issue of public service reform agree on several
shortages, they have been biased in favour of
diagnostic points, as summarised in Policy Brief 1. The
technological fixes of doubtful sustainability.
general record of donor-supported institutional-reform
programmes over the past decade suggests a persistent Underlying all this, however, is the wider failure of
inability to bite the bullet of large-scale structural successive governments to act decisively to restore the
change. The reduction in civil service numbers resulting civil service to something like its former condition,
from the first World-Bank-sponsored reform was when it had the reputation of being the most efficient
reversed following the first Rawlings electoral victory, in sub-Saharan Africa. The view presented above of
resulting in a chronic problem of widespread over- the politics of the budgetary system helps explain why
manning coexisting with shortages of key skills in many successive leaderships have not been very interested
important areas. in change, because the situation has suited them rather
well. Reform would mean greater transparency, more
A 2003 examination of public sector reform by a leading
accountability, a slimming-down of the civil service
firm of financial consultants is written in similar vein. It
and less scope for discretionary decision-making.
describes the various reform efforts as ‘not as successful
Moreover, we suggest below, it would be likely to lose
as may have been expected’ and cite in evidence capacity
votes.
gaps, lack of ownership, inability to institutionalise change,
low morale in the public service, weak human resource Conclusion: what drivers of budgetary change?
management and ‘Doubts about the relevance and worth
To sum up, the existing expenditure budget processes
of the reforms’. It reports ‘capacity gaps in almost all critical
have been shown to be so weak as to be essentially
Page 4
 CDD/ODI Policy Brief No. 2

ritualistic, with limited bearing upon reality. It also providing them, and the points of service delivery
remains a closed system, creating a ‘democratic below them, with more information about their
deficit’, which helps explain why such a weak system budgetary entitlements.
continues to be tolerated. The problem is compounded
The question arises, what incentives are there for
by the continuing deficiencies of the civil service and
ministers to move in these directions? The purpose of
the political disincentives for governments to improve
these various reforms would be to improve
the situation.
accountability and reduce the scope for using public
Our final question is through what means this situation resources to further essentially private and/or sectarian
might be improved. The argument of Policy Brief 1, objectives, but why should ministers want to restrict
that the best prospects for change lie in the progressive themselves in such ways? Well, Ghana’s patronage-
democratisation of the country’s institutional and driven politics is already being modified by a
political structures, is applicable here. The budget case deepening of democratic values. It seems likely that
offers an interesting variation on this theme, however, ministers’ values reflect this shift, so they may exercise
for the present budgetary system reflects a ‘democratic leadership in moving in the directions suggested above.
deficit’, meaning that budgetary practices have not They may be encouraged to do so (a) by the
caught up with the democratisation that has occurred extraordinarily costly nature of present inefficiencies
elsewhere. and (b) by the fact that none of the suggested steps
seems politically risky and could be genuinely popular.
The first task, then, is for budgetary practices to catch
up with the wider processes of democratisation. A Perhaps the most serious obstacle is the difficulty of
number of measures suggest themselves, none of accomplishing the reforms of the civil service without
which appear politically impossible: which it would be difficult to make much progress.
Here, the political logic is not very encouraging. In
• Since there is already a degree of informal prior
this area, democratisation may not contribute to a
consultation about the budget, this could be taken
solution. For reasons summarised in Policy Brief 1,
further and be institutionalised. It could also be
civil service reform is likely to be seen by politicians
widened, to bring in more interested parties, and
as a vote-loser. Preferring to provide benefits that are
become more systematic and less Accra-centred.
visible in the short term and do not incur political costs,
One suggestion here is to build on the consultation
governments may be reluctant to pursue it seriously.
processes that were adopted for the preparation of
the GPRS. Another possibility is to adopt the Politicians’ incentives structures may at present be too
Tanzanian practice of opening up budget hearings, adverse. On the other hand, the deepening of
with all interested parties having the right to democracy will increase the electoral import of
participate. governments’ abilities to deliver promised services to
• Greater consultation would, however, need to be voters. That, in turn, will raise the incentives to reform
accompanied by improved provision and the public service, for it is unable to deliver well in its
dissemination of information. It could also be present parlous condition.
reinforced by a campaign to raise public awareness
For public interest associations and political activists
and improve media coverage of the issues.
in Ghana, the implications are reasonably clear. To the
• Revising the budget timetable could also help.
extent possible, support should be given to those
Wider participation would require more time, and
democratic pressures that strengthen politicians’
more time is also needed for Parliament to be able
incentives to improve governmental performance, and
to scrutinise the budget properly. Parliament also
not to those that encourage short-termism and
needs more resources, including funds to allow it
caution VV
to obtain expert advice and to hire assistants.
• Another obvious step would be to include the
District Assemblies in the process, bringing them
directly into the machinery of consultation and
Page 5
CDD/ODI Policy Brief No. 2

About the publication:

These Policy Briefs have resulted from

collaborative research and analysis
conducted by a multi-disciplinary team from
the Ghana Center for Democratic
Development (CDD-Ghana) and the London-
based Overseas Development Institute (ODI)
in 2003 and 2004. Historically-informed and
less technocratic, they take a fresh look at
where Ghana is coming from, where it actually
is, and where it may be headed.

The Briefs come in two versions: a general

analysis and four case studies on particular
topics. Together they provide an overview of
how social, political and economic forces have
interacted in the country, and with what
effects.
A CDD-Ghana Publication
CDD-Ghana, 2005

Correspondence:
The Publications Officer

Ghana Center for Democratic Development,

CDD-Ghana
P.O. Box LG 404, Legon - Accra, Ghana

Tel: (+233-21) 776142/763029/784293-4

Fax:(+233-21) 763028
E-mail: [email protected]
Website: www.cddghana.org

Page 6