Design of the Strategy

A realistic overall strategy should be designed up front and in consultation with donors. It should include measures to get the government immediately up and running to pay wages and provide basic services, as was recommended in Timor-Leste. The strategy should not underestimate capacity constraints and should avoid setting up an unrealistic timetable for implementing certain reforms. In practice, some reforms have taken longer to implement than expected. Experience in Bosnia and Herzegovina, for example, suggests that a basic expenditure control system can be set up relatively quickly, but that strengthening analysis and policy development capacity takes much longer. The strategy should lay down a road map of actions with appropriate sequencing and priorities, and identify technical assistance needs and providers. This can also facilitate coordination of donor activities.

In cases where there is little or no administrative capacity, long-term advisors can play a critical role in building capacity and transforming skills in the early postconflict stages. As capacity improves, these experts should play a more advisory role. In some cases, such as Timor-Leste, the IMF assisted in identifying expatriate personnel who could take up senior positions in key fiscal institutions and work closely with national counterparts to train local staff. The objective was to transfer management responsibilities to local personnel in the shortest possible amount of time. In other countries, such as Cambodia and the Democratic Republic of the Congo, these advisors assisted country authorities in implementing technical assistance recommendations and also helped train local staff. The tenure of long-term advisors varies from country to country depending on the pace of implementation of reforms and development of local capacity.

The initial steps in the process of rebuilding fiscal institutions in postconflict situations should focus on designing and implementing simple revenue policies and administration, as well as public expenditure management procedures. As local capacity develops, these can be upgraded. With respect to revenues, for example, experience in Kosovo and Timor-Leste showed that it is useful to focus on basic procedures for filing returns and paying taxes that are easy for taxpayers to comply with and for the administrative system to implement. In a similar vein, in light of scarce capacity, it was recommended that large taxpayer units be established in Albania, Liberia, Serbia and Montenegro, Timor-Leste, and West Bank and Gaza. For public expenditure management, simplified expenditure classification systems were introduced in Albania, Croatia, Democratic Republic of the Congo, and Timor-Leste.

In framing policies and procedures, it may sometimes be necessary to deviate from first principles. Postconflict countries often face a tradeoff between short-term revenue mobilization and economic efficiency objectives. In this light, many countries have had to move gradually to address inefficiencies in some taxes. Such was the case in Croatia, Liberia, Serbia and Montenegro, and Tajikistan. Thus, policies that are suboptimal from an efficiency point of view may nevertheless need to be implemented for a period of time. However, these should be presented as temporary and phased out as quickly as possible.

Where possible, short-term measures should be consistent with the longer-term objective of moving to a modern fiscal system. For example, taxes recommended for the early stages of postconflict institution building have helped lay the groundwork for a more efficient tax system that could only be implemented at a later stage. Some countries introduced a withholding tax on wages (e.g., Afghanistan, Albania, Bosnia and Herzegovina, Kosovo, and Timor-Leste) and a presumptive tax on small business (e.g., Kosovo and Timor-Leste). As capacity developed, it was expected that these taxes would be transformed into a more refined system of personal and corporate taxation. For indirect taxation, a similar strategy was employed. In Cambodia, Kosovo, and Timor-Leste, a tax on selected services in the initial stages was recommended. The hope was that the development of this tax base could set the stage for implementing a full-fledged value-added tax over the longer term. In some situations, however, such as Afghanistan and Timor-Leste, tax policy options were constrained by the limited mandate of transitional administrations.

Decentralization poses a challenge to fiscal management in postconflict countries. It is essential to ensure that the decentralization often needed to keep the peace following ethnically based conflicts does not endanger economic reforms and fiscal management. A flexible approach to decentralization can address these concerns. Revenue and expenditure assignments should be clearly identified, with an appropriate transfer mechanism to mitigate regional income disparities. In the postconflict situations analyzed here, the emphasis was on the need to develop capacity for effective fiscal management at the subnational level. Given the IMF’s limited resources for technical assistance, however, other development partners had to be relied upon to provide such assistance to subnational governments.

Implementation

The postconflict reform strategy should be flexible enough to respond to changing circumstances on the ground. The strategy may need to be adjusted in the context of emerging priorities or as new issues and weaknesses come to light. The pace at which reforms can reasonably be implemented may also necessitate a change in the strategy. For example, the slow pace of implementation and the need to track poverty-reduction spending prompted a revision of the overall technical assistance strategy in the Democratic Republic of the Congo. In Afghanistan, the weakness of public expenditure management was initially underestimated, as the security situation did not permit mission travel outside Kabul. Following a more detailed assessment by subsequent missions, the strategy was later adjusted. In Mozambique, the initial implementation plan for public expenditure management reforms was found to be unrealistically ambitious and required subsequent revisions.

A strong commitment by country authorities is an important determinant of the pace of reforms. Progress is rapid where the ownership and commitment of the country authorities is strong. For instance, progress was slow in the Democratic Republic of the Congo in the early postconflict stages until the transitional government reiterated its commitment to structural reforms. Progress on tax reform was also slow initially in Bosnia and Herzegovina, reflecting weak ownership at different levels of government, but progress accelerated when the distribution of powers regarding tax policy and administration was clarified. To secure ownership in the initial stages, the reform plan should be simple enough to allow for input from the authorities as time evolves. Where capacity exists, the sequencing of reforms should reflect the priorities of the authorities. For example, the technical assistance strategy in Lebanon assigned priority to tax policy reforms, consistent with what national authorities wanted. This garnered stronger ownership for the overall reform strategy. Furthermore, it is necessary to reach beyond the ministry of finance to ensure implementation of key measures. In Afghanistan, there would have been more rapid progress in amending the income tax and customs codes had there been earlier and more effective consultation with the ministries of commerce and justice. Public expenditure management reform would also have been more successful in that country had there been better collaboration between the ministry of finance and the line ministries.

There is scope to improve donor collaboration. The multitude of donors, coupled with weak administrative capacity, has in many instances made cooperation and coordination difficult in postconflict situations. In Timor-Leste, assistance in implementing public expenditure management reforms was provided by several donor countries, each of which had its own different expenditure management systems and traditions. Thus, personnel provided by one donor were not always familiar with another donor’s system or approach, resulting in coordination problems. In Cambodia, the donor coordination mechanism was weak because it excluded key stakeholders such as tax, customs, and treasury directors. In Afghanistan, donors provided uncoordinated advice on energy taxation.

There are several different ways to strengthen such collaboration, although no single measure can be viewed as a panacea. Creating a separate unit for donor coordination can help, as happened in Kosovo and Mozambique, by providing a vehicle to incorporate the input of various donors in the design and implementation of the strategy and for the sharing of information, with a view to avoiding duplication and disputes. In post-conflict situations where such a unit already existed, such as in Cambodia, the IMF recommended strengthening the coordination function by appointing a fulltime national program manager. Designation of a “lead donor” to ensure greater accountability can also improve collaborative efforts. In Cambodia, Mozambique, and Yemen, the IMF took the lead in coordinating donor assistance. Finally, collaboration can be improved by establishing a steering committee to focus on specific areas in need of immediate attention,¹ as was done in Afghanistan and Cambodia.

Appropriate conditionality in IMF-supported programs can facilitate implementation of reforms by raising the profile of these measures and focusing the attention of the authorities on the steps necessary for their implementation. Program conditionality complemented the reform strategy in Afghanistan. The IMF’s Staff Monitored Program included a prior action for tax policy reforms and a structural benchmark for implementing a treasury single account. Likewise, some technical assistance recommendations were incorporated as program conditionality in the Democratic Republic of the Congo and Mozambique, facilitating their implementation.

More effort by the international community needs to be directed toward developing local capacity. Weak institutional capacity was a major constraint in rebuilding fiscal institutions in all of the postconflict situations where technical assistance was provided. In the initial postconflict stages in such countries as the Democratic Republic of the Congo, this problem was addressed through long-term resident advisors, although the reforms slowed once those advisors left. Donors have provided generous contributions for training, but developing local capacity has often proven more difficult than anticipated, as in Mozambique. In many countries, slow progress reflected deep-seated problems in the civil service.

Progress in implementing postconflict fiscal reforms also has been slow in countries where the security situation is fragile. In cases such as Afghanistan, central government authority over the country as a whole was not effectively established until well after the end of the conflict. This affected the implementation of technical assistance recommendations. It took over two years in Afghanistan, for example, to record improvements in revenue administration. This reflected weak central control over provincial fiscal operations, resulting in the nontransfer of revenues to the central government. The security situation in the Democratic Republic of the Congo also slowed progress by delaying capacity-building efforts for local fiscal institutions in former rebel-held areas.