Corruption & DBS: Conditionality and sequencing

Corruption makes conditionality ineffective, and sequencing important

Efficient use of budget support requires a good institutional environment. This applies in terms of general macro-economic governance, but specifically as regards the quality of the PFM system. Budget support commonly takes the form of policy-based aid, where dialogue and conditions are attached, with a view to improving the institutions of a partner country. Corruption introduces added considerations and challenges, to the question of how to spur institutional reform in partner countries. This section explores how corruption affects the impact of conditionality, and the degree and sequencing of PFM reform.

• Conditionality ineffective in corrupt countries
• Sequencing PFM reform and budget support to avoid increased corruption
• An example of sequencing problems: PFM reform in Cambodia

Conditionality ineffective in corrupt countries

Money can't buy policy. A number of studies conclude that conditions and dialogue attached to programme support, have failed to produce the intended reform in partner countries. Though partner countries assent to reforms, they do not implement them, and donors fail to punish non-implementation by withholding aid. Thus, there is by now an informed consensus that conditionality is ineffective in delivering reform.

Exceptions to this rule have been noted, however. Where there is domestic support for reform, but the support is hard to mobilize (by some called a collective choice dilemma), donor pressure might produce the momentum needed to bring about reform. In other words, where reform is in the domestic political balance, donors might tip the scales in favour of reform.

In countries where corruption is sufficiently entrenched and widespread, donor pressure is unlikely to have this kind of an effect. If a large part of the political establishment derives significant benefits from rent-seeking, or key segments of the population benefit from patronage, mobilizing domestic support for dismantling the current state of affairs will be unrealistic.

Conditionality is therefore likely to be ineffective in highly corrupt countries. A worst case scenario is that a country implements only conditions that serve to strengthen the position of the ruling elite. These arguments also suggest that including anti-corruption reform as a condition attached to budget support, will probably not have much of an impact in highly corrupt countries.

Where budget support conditionality is ineffective due to corruption, donors should promote reform by other means. One way to make donor sanctions more credible, is to require pre-qualification for budget support, where some level of institutional development has to be attained before budget support is given. This would imply selectivity in the use of budget support.

Sequencing PFM reform and budget support to avoid increased corruption

Reform of the public financial management system, is particularly important for the effectiveness of budget support. One argument used for budget support, is that channelling funds through the PFM system of a country, will serve to strengthen that system. It should be pointed out, however, that this is not an inevitable result of budget support. In line with the general observations on conditionality, the prime determinant of PFM reform will be domestic political factors. Reform will thus only be implemented if it has domestic political backing.

To efficiently allocate funds, the PFM system as a whole must be well-functioning, from budget planning, through execution, to audits and control. Given the poor state of the system in many developing countries, ideally all parts of the PFM system should be improved. Given resource constraints, this is hard to do, and reform is therefore often done partially and sequentially. In sequencing PFM reform, one should be careful that the sequencing is not done in a way which creates greater incentives for corruption. The following example from Cambodia illustrates that agreements between donors and partner countries on the sequencing of PFM reform, do run this risk in practice. To the extent that sequencing of this kind is necessary for other reasons, budget support should not be phased in until the necessary checks and balances are in place to prevent its misuse.

An example of sequencing problems: PFM reform in Cambodia

At present, Cambodia has a very weak PFM system, with significant weaknesses in budgeting, accounting, auditing, cash management, and a general lack of skilled staff and capacity. Cambodia is also a society characterized by extensive corruption. Though the country is not among the 146 countries ranked in the 2004 Transparency International Corruption Perceptions Index, Cambodia ranks on a par with Malawi and Uganda on the 2002 KKZ corruption index.

To improve the PFM system of Cambodia, a group of donors including the World Bank, has reached an agreement on a Public Financial Management Reform Programme with the Cambodian government. The programme was to commence in September 2004, and was divided into four stages or platforms (see below figure).

Sequence of platforms - Public financial management

As the figure shows, the first stage focuses on getting a credible and reliable budget in place. The second stage consists of steps to improve internal control and accountability, the third stage focuses on linking policy to budget planning and implementation, and the fourth on accountability, decentralization and integration of the PFM system through information technology.

The first two stages illustrate the disadvantage of having to pursue a stepwise approach to PFM reform in an environment characterized by corruption. In the first stage, resources will be channelled towards budget realism. This may divert resources and attention from other important functions of the PFM system, such as internal control. And relaxing whatever control mechanisms may be in place (though not intentionally), will make the risk of corrupt activities smaller, and therefore lead to an increase in corruption in the PFM system. One should thus not be too surprised if the incidence of corruption in fact increases during the first phase of the reform programme.

This has implications for the use of budget support. An increased flow of funds into a PFM system in which internal controls are being relaxed, increases the potential benefits while reducing the risk of corrupt activities. Providing budget support during the first stage of such a reform process, would thus add to the problem of relaxed controls, and lead to an even greater increase in corrupt activity. A prudent policy would therefore be to not give budget support until the necessary control and accountability mechanisms are in place, that is after the completion of stage two, at the earliest.