Information about Asia and the Pacific Asia y el Pacífico
Chapter

Chapter 6. Fiscal Developments and Challenges

Author(s):
Sumio Ishikawa, Sibel Beadle, Damien Eastman, Srobona Mitra, Alejandro Lopez Mejia, Wafa Abdelati, Koji Nakamura, Il Lee, Sònia Muñoz, Robert Hagemann, David Coe, and Nadia Rendak
Published Date:
February 2006
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details
Author(s)
Alejandro López-Mejía and Robert Hagemann 

Fiscal reform has been the cornerstone of Cambodia’s macroeconomic program since the restoration of political and economic stability in the late 1990s. Fiscal revenue improved, social spending was increased considerably, and domestic financing of the budget was largely avoided. Despite this progress, fiscal revenue and social spending still lag behind the average of countries at a similar development stage. More generally, substantial reforms are still needed to improve both the efficiency and fairness of Cambodia’s public finances, so that they can play a more helpful role in poverty reduction.

Section A provides a fresh look at fiscal developments in Cambodia in the 1990s and Section B reviews fiscal reforms since 1999. Section C discusses the challenges ahead for reforming Cambodia’s public finances, which are to be addressed through the Royal Government of Cambodia’s (RGC) recently adopted “Public Financial Management Reform Program” formally launched on December 5, 2004 (see Royal Government of Cambodia, 2004).

A. Developments Since the Early 1990s25

The Early Challenges: 1990–94

Cambodia’s public finances have evolved substantially during the past 10 years. As the country emerged from international isolation, public revenues were less than 5 percent of GDP, comprised overwhelmingly of duties and taxes on traded goods (Figure 6.1). Public spending was more than double the state’s receipts, and both the current and overall budget balances were in deficit. During the early years, and despite substantial foreign financing of the deficit, resorting to central bank financing led to high rates of inflation (see IMF, 2004b). Priority was thus given to mobilizing public revenues to meet the growing expenditure requirements for reconstruction and basic public services.

Figure 6.1.Public Finances, 1993–2003

(In percent of GDP)

Sources: Cambodian authorities; and IMF staff estimates.

The revenue efforts undertaken in the early 1990s were initiated from extremely low levels of revenue collection. After all, a modern taxation regime did not exist under the socialist regime in the 1980s. Under these circumstances, during 1990–93, total revenues represented only 4–6 percent of GDP, and were 4–5 percent of GDP lower than expenditures. Moreover, with limited external financing, central bank financing averaged 3–4 percent of GDP annually, contributing to hyperinflation of over 100 percent.

In 1992, a major fiscal reform was initiated with the assistance of foreign experts, followed by a significant revenue effort by the new government formed after the UN-sponsored free election in 1993. Revenues increased to about 9.5 percent of GDP in 1994 and, together with substantial concessional lending, eliminated reliance on central bank financing. The reform program covered all aspects of Cambodia’s revenue system in tax policy and tax and customs administration. The increase in revenue was attributed to growth in the tax base and policy measures (Figure 6.2). Despite efforts, the contribution of administrative improvements to revenue growth was negative.26 The most effective policy measures related to revenue from international taxes, which increased from almost zero in 1991 to about 5 percent in 1995. The main measures were (1) introduction of a consumption tax on imports in 1993, generating about 1 percent of GDP in 1994; and (2) a higher ad valorem duty rate on petroleum products, which was increased gradually from 3–5 percent in 1992 to about 50 percent in 1994, and generated duty collections of about 1 percent of GDP in 1994—compared to virtually nothing in 1991.

Figure 6.2.Total Tax Revenue

(Central government)

Sources: Cambodian authorities; and IMF staff estimates.

Attempts to Modernize Amid Political Instability: 1995–98

During 1995–98 further attempts were made to deepen fiscal reform. New taxes were introduced in 1995–96, such as a tax on income from wages, a 20 percent excise tax on gasoline, a higher duty on petroleum products, and an increase in the turnover tax. To improve tax policy and tax administration, a new Law on Taxation (LOT) was adopted in 1997.27 Government efforts also focused on broadening the tax base, especially in the area of domestic taxes, through significant improvements in tax administration.28 Efforts included the establishment of a Large Enterprise Bureau in the Tax Department and a computerized database of large tax payers.

Despite these efforts, during 1995–98 total tax revenue stagnated at about 6 percent of GDP, mainly due to a deterioration in customs administration, while nontax revenue decreased to 2 percent of GDP. Against a backdrop of political and military instability, the authority of the Ministry of Economy and Finance (MEF) to collect revenue was undermined. Revenue performance in turn weakened due to the granting of widespread ad hoc exemptions of customs duties, smuggling, an accumulation of tax and nontax arrears, and significant losses in forestry revenue. Indeed, reflecting the overall deterioration of governance, forestry revenue during 1995–98 amounted to only 0.4 percent of GDP per year, compared to potential revenues of about 3–4 percent of GDP per year.29 Under these circumstances, coupled with difficulties in restraining military and security spending, in 1998 the government used central bank financing of the budget for the first time since 1994.

B. Fiscal Reforms Since 1999

Revenue Developments Since 1999

Fiscal performance improved dramatically in 1999. Total revenue rose from 8.1 percent of GDP in 1998 to 10.2 percent in 1999. The reform of the tax system since January 1999 aimed at moving away from direct taxation of trade and incomes to indirect taxation, relying mostly on a 10 percent VAT that replaced the turnover tax and consumption tax on imports. The VAT boosted revenue and improved the efficiency of the tax system by simplifying the tax structure, widening tax coverage, and reducing cascading. The fiscal reform momentum that started in early 1999 was strengthened during 2000–02 under the PRGF arrangement aimed at strengthening the revenue structure and overall administration. Accordingly, revenue had not been increased through changes in tax rates since 1999, except for increases in petroleum taxes and excises on cigarettes and beer in 2002 to offset a reduction in customs duties relating to tariff restructuring as part of a comprehensive trade liberalization program. Nevertheless, revenue rose only marginally in 2002, to 11.2 percent of GDP. Revenue collection was undermined in 2003 by political constraints, and fell to 10.4 percent of GDP, due largely to weaker-than-budgeted customs revenue collections.

Although most of the growth in revenue since 1999 is attributed to the expansion of the tax base, recent efforts to enhance tax and customs administration should not be underestimated (see IMF, 2003a, Annex 2). Technical assistance, notably that provided through the Technical Cooperation Action Program (TCAP),30 was instrumental in improving administration, including the expansion of VAT coverage by increasing the number of taxpayers.31 Indeed, during 2001–02, improvements in customs administration contributed for the first time in a decade to an increase of revenues (Figure 6.3). Efforts focused on ensuring a more effective use of preshipment inspection services, and increased transparency to reduce hidden costs in customs procedures. Moreover, anti-smuggling operations were strengthened through enhanced interagency cooperation and the establishment of anti-smuggling units in key border provinces. Unfortunately, these efforts recoiled in 2003 due to the political impasse.32

Figure 6.3.Customs Revenue

Sources: Cambodian authorities; and IMF staff estimates.

Improvements in tax administration during 2001–02 included an initial exchange of information between government departments and strengthening tax auditing strategies and capabilities (Figure 6.4). As a result of these initial steps, tax arrears have started to be collected. Efforts to collect tax arrears were intensified using a variety of enforcement measures, including the freezing of bank accounts, the temporary confiscation of imports, and delicensing. The collection of arrears (taxes plus penalties and interest) has increased continuously during the past several years: $1.5 million in 2001, $5.8 million in 2002, and $18.7 million in 2003 (0.4 percent of GDP).

Figure 6.4.Domestic Taxes

(Central government)

Sources: Cambodian authorities; and IMF staff estimates.

Box 6.1.The Law on Investment

The 1994 Law on Investment (LOI), recently amended (see below), has seriously eroded the revenue base. Revenue collected by the Customs and Excise Department—currently accounting for about 75 percent of total tax revenue—would have been even greater were it not for the substantial erosion of the dutiable base. Indeed, forgone import duties are approximately equal to collected duties. The law provides very generous tax incentives to investors compared to other countries. In particular (1) tax holidays are permitted up to eight years, (2) profits are taxed at a reduced rate of 9 percent (instead of the normal rate of 20 percent) after the end of the holiday period, (3) reinvestment of profits is tax free, and (4) repatriation of earnings and other incomes is tax free.

Exempted Import Duties by Exemption Regime(In percent of GDP)
1998199920002001
Total exempted import duties14.75.26.37.3
Diplomatic missions and international organizations0.10.10.10.1
Investment law provisions3.63.94.95.8
International aid0.60.30.30.3
Nongovernmental organizations0.10.20.10.1
Other0.40.80.80.9
Memorandum item
Total import duties14.16.26.06.2
Sources: Cambodian authorities; and IMF staff estimates.

Includes all tax exemptions on imports (i.e., customs duties, excises, and value-added tax).

Sources: Cambodian authorities; and IMF staff estimates.

Includes all tax exemptions on imports (i.e., customs duties, excises, and value-added tax).

Amendments to the LOI, enacted (along with requisite amendments to the Law on Taxation) by the National Assembly in early 2003, will potentially broaden further the scope of exemptions, depending on the implementing regulations currently being drafted. Under the amended LOI, investors can now choose to be subject to a special depreciation schedule rather than require tax holidays. The amendments to the LOI eliminate the cumbersome matrix determining the length of the tax holiday, the reduced tax rate on profits of investment companies (grandfathered for those already entitled the rate of 9 percent), and the dividends tax exemption, and they increase transparency by defining clear procedures for granting exemptions. However, the new LOI continues to grant generous exemptions—extensive tax and customs duty exemptions, including projects located in Export Promotion Zones—and differential tax holidays by economic sector that could complicate tax administration and go beyond the maximum three-year tax holiday advised by IMF staff. The extension of duty exemptions on a wide range of imported inputs to almost all qualified investors is a significant relaxation compared with the previous requirement that qualified investors satisfy certain export performance criteria.

Much remains to be done to broaden the revenue base, however. In particular, revenue from international taxes would have been much higher in the absence of large tax and customs duties exemptions. Indeed, during 2002–2003, about 60 percent of all imports were exempted from customs duty, mostly under the 1994 Law on Investment (LOI). In addition, although a clearing of the backlog of applications for tax exemptions during late 2002 reflected improved transparency, it seriously undermined the domestic tax base, more than halving the projected profit tax collections for 2003. As noted in Box 6.1 (see above), recent amendments to the LOI appear to have broadened further the generosity of investment incentives.

Regarding nontax revenue, several measures were implemented over the last four years, but actual collections have been consistently below expectations. Indeed, nontax revenue only increased from 2.7 percent of GDP in 1999 to about 3.2 percent in 2002, and declined slightly in 2003 (Figure 6.5). The main policy measures introduced in 1999 include (1) transparent collection of garment quotas through regular auctions, (2) introduction of a quota management fee and a garment export license fee, and (3) increasing the timber royalty to an average of $54 per cubic meter, helping to maintain forestry revenue at the 1998 level despite a large reduction in the volume of logging. In 2002, however, forestry revenue declined sharply as all logging activity was suspended pending the establishment of a forestry concession system based on sustainable practices. The main nontax revenue policy measures since 1999 comprised the revision of the contract terms of the entrance tickets at the Angkor temple complex in 2000 and 2002, and the introduction of visa stickers in December 2001. In addition, in 2002, royalty fees from casinos were increased and the share of garment quotas to be auctioned was raised. Despite these measures, there remain significant estimated arrears from telecommunications, civil aviation, and state-owned enterprises and immobile assets leased to the private sector.

Figure 6.5.Nontax Revenue

(In percent of GDP)

Sources: Cambodian authorities; and IMF staff estimates.

Expenditure Developments Since 1999

During the past several years, public expenditure policy has been supportive of macroeconomic stability and was restructured to focus spending toward priority areas. The level of current expenditure remained consistent with a current budget surplus of 1–1.5 percent of GDP during 1999–2002, ensuring sufficient funding for local development projects and helping to avoid any domestic financing of the budget. A military demobilization program initiated in late 2001 was helpful in reorienting expenditure to priority areas (agriculture, rural development, health, and education). Spending on military and security was reduced from 4 percent of GDP in 1998 to 2.5 percent in 2003, while spending on priority sectors rose from 1.5 percent of GDP to 3.6 percent. Still, despite these improvements, education and health expenditures still lag behind the average of countries at a similar stage of development.

The overall public sector wage bill has averaged below 40 percent of current expenditure. But given a fairly sizable public sector workforce (civil administration and military and security combined), funding of basic operating expenses, and low revenue mobilization, remuneration remains far too low to recruit and retain qualified persons. Monthly remuneration in 2002 ranged from $21 to $40 depending on the educational attainment of a civil servant. Even with the government’s decision to raise salaries by 15 percent beginning in 2005, the typical civil servant earns far less than even an unskilled garment worker.

Despite the few spending improvements noted earlier, public expenditure management remains weak, and reforms have been difficult to achieve to date. The use of the banking system for government transactions is still very limited, several offsetting arrangements exist with suppliers whereby overdue taxes are offset against government obligations, and the practice of separating U.S. dollar and national currency revenues and expenditures persists. As a consequence, the cash management system remains fragmented, leading to poor budget execution, marked differences between the budget plan and actual spending, and a significant bunching of commitments toward the end of the year.

Because the bunching of expenditures is partly due to extensive pre-auditing by the MEF, an attempt to resolve this problem was made through the introduction of the Priority Action Program (PAP) in 2000. The PAP was intended to ensure that the priority sectors of health, education, rural development, and agriculture could gain access to their full budget allocation by obtaining 25 percent of the budget allocation automatically on a quarterly basis, replacing pre-audits by post-audits. However, despite some initial success in increasing the disbursement ratios, the old pre-audit arrangements have not been successfully replaced, the PAP budget plan has typically not been fully committed until late in the fiscal year, with full disbursement requiring close to two years.

Mindful of the inconsistency between the objectives of PAP and its poor implementation, the government took steps during 2003 to accelerate disbursement of PAP. In September 2003, the RGC established a task force to accelerate PAP disbursements to the health and education sectors. The task force has been somewhat effective in ensuring that a portion of available cash was earmarked for disbursement to these sectors, and by October 2003, 100 percent and 80 percent of committed 2002 PAP expenditures had been disbursed to the ministries of health and education, respectively. An acceleration of PAP disbursements to the ministry of education was achieved in 2003, reducing the lag between commitment and full disbursement at both the ministries of education and health to 23–24 months. Notwithstanding the efforts of the task force, disbursing the backlog of previous years’ commitments caused unavoidable delays in liquidating PAP spending budgeted for 2004; as of November 2004, only 24 percent and 60 percent of committed 2004 PAP resources for education and health, respectively, had been disbursed.

Other efforts to improve budget planning and execution were initiated during the past two years. Beginning in 2002, efforts focused on improving cash management, regarded as the root of the shortcomings in the budget execution. As a result, revenue accounts held by line ministries began to be integrated with the National Treasury’s single account, and the operations of the Cash Management Committee were improved by implementing a specific format for cash management procedures and enhancing the coordination between the National Treasury (NT) and the Foreign Currency Unit (FCU) of the MEF.33 More recently, the FCU was transferred to the NT, helping to ensure better information on and improved management of the unit’s financial resources. In the treasury area, working groups at the NT had been making efforts to implement standardized accounting procedures and methodologies for the public sector. In addition, with assistance from the Asian Development Bank, a medium-term expenditure framework began to be developed, aimed at integrating more tightly medium-term spending plans with the priorities spelled out in the government’s National Poverty Reduction Strategy (NPRS).34

C. Remaining Agenda

Despite significant progress to date, Cambodia faces a daunting challenge to transform its public financial management system into one that is capable of adequate service delivery. The reforms implemented under the 1999–2003 PRGF, with technical assistance provided under the TCAP, were aimed at bolstering relatively rapidly the operational capacity of key institutions of macroeconomic management to help boost revenue mobilization, and to improve expenditure execution and cash management. Durably transforming the public financial management system into one that delivers cost-effective higher quality public services was always known to be a long-term project.

Recognizing this challenge, the RGC has recently launched its Public Financial Management Reform Program (PFMRP). The PFMRP, conceived and being implemented using a sector-wide approach, aims to complete the full modernization of Cambodia’s public financial management (PFM) system by 2015 in four successive stages, or platforms: (1) ensuring that the budget is a credible instrument for implementing the RGC’s development strategy, (2) introducing effective financial accountability, (3) improving the linkages of priorities and service targets to budget planning and implementation, and (4) holding budget managers accountable. The long-term objective is to transform the public financial management so that it is consistent with best international standards (Royal Government of Cambodia, 2004). This long-term program, while focusing on the institutional and policy reforms needed to modernize the public financial management system, will also address the medium-term revenue challenge posed by the NPRS.

It is within the context of this recently launched initiative that the Cambodian authorities and their development partners will have to tackle, in a systematic and coordinated manner, a number of challenges in order to modernize the public financial management system and make it an effective instrument of macroeconomic and social policy. The remainder of this section focuses on these key challenges.

The Revenue Challenge

Despite improvements in revenue performance during the past few years, Cambodia is a long way from a self-sufficient and sustainable fiscal position. Cambodia’s fiscal revenue ratios—especially tax revenue—remain very low compared with other countries at similar stages of development (Table 6.1). Increasing fiscal revenue to above 14 percent of GDP by 2009 will be necessary to meet the expenditure needs underlying the government’s development strategy while also allowing for increased debt service payments arising from the expected completion of external debt rescheduling agreements with the United States and the Russian Federation.35 In this context, fiscal policy will have to continue aiming at a sustained current budget surplus of about 1 to 2 percent of GDP, with a declining overall deficit (excluding grants) that is fully financed by external concessional resources.

Table 6.1.Comparison of Tax Revenue Structure with Other Selected Countries(In percent of GDP; unless otherwise noted)
Fiscal Revenue1
Tax revenue
of which:Per Capita GDP (US$)
Total revenueTotalDirect taxesIndirect taxesTrade taxesOther revenue2
PRGF Asian countries16.212.83.17.32.33.8427
Bangladesh8.67.11.25.82.01.5370
Lao P.D.R.11.49.22.03.81.04.6310
Nepal11.29.22.14.13.02.0240
Sri Lanka17.014.72.410.22.12.3830
Vietnam20.615.65.56.33.45.0410
Mongolia328.520.85.313.42.17.7400
PRGF sub-Saharan African countries17.613.73.95.54.62.0323
Of which: selected agriculture countries14.411.41.83.83.32.1277
Cameroon17.812.23.36.52.50.7570
Guinea-Bissau19.510.80.00.00.08.7160
Mali14.714.02.65.05.40.7210
Niger8.98.31.51.94.40.6170
Tanzania11.810.62.76.71.21.2280
Togo13.812.30.42.86.10.8270
Cambodia (in 2001)11.78.41.04.62.83.3259
Source: IMF staff country reports.

Data refer to average of 1999–2001.

Includes nontax and capital revenue.

Other revenue includes only nontax.

Source: IMF staff country reports.

Data refer to average of 1999–2001.

Includes nontax and capital revenue.

Other revenue includes only nontax.

The medium-term revenue objectives can be achieved by containing further erosion of the revenue base and strengthening customs and tax administration. Broadening of the revenue base needs to be focused on reducing the scope of existing tax and duty exemptions. A simple and transparent investment regime with lower tax rates would be more attractive to potential investors than a system that provides for large exemptions, especially when lack of capacity makes enforcement problematic. On customs administration, the automation of customs procedures, as part of a broader trade facilitation initiative (World Bank, 2004), would improve the capacity of the Customs and Excise Department and reduce hidden costs. Against a backdrop of evidently rising smuggling, however, substantial efforts are needed to strengthen anti-smuggling capacity. Ongoing reforms at the tax department include strengthening audit capacity, establishing taxpayer services, and computerization. To eventually achieve maximum benefit from these initiatives, however, the tax department will have to introduce significant organizational reforms, to restructure the department along functional lines. All of these should lead to a steady reduction in tax avoidance, additional expansion of VAT coverage through increases in the number of taxpayers, and enhanced compliance through the acceleration of VAT refund procedures.

Cambodia continues to rely significantly on nontax revenues. However, greater transparency over the terms governing the use of state assets will also be key to transfer the appropriate amount of revenue to the budget. In particular, efforts should be made to (1) ensure that procurement procedures are based on competitive bidding, and applied to all public acquisitions and contracts, (2) respect the government’s commitment to have all contracts reviewed by MEF and approved by the Minister of Economy and Finance, and (3) publicly disclose the terms of all (past and future) contracts, subject them to audit by the National Audit Authority, and carry out periodic external audits of some contracts, disclosing the results.

An evident challenge facing the authorities is maintaining fiscal sustainability, which in turn depends importantly on improved revenue mobilization. Even if the revenue target is achieved, Cambodia’s fiscal position appears to be only marginally sustainable over the medium term due to the implications of an eventual rescheduling of external debt agreements. Indeed, to retain public debt sustainability, the primary deficit would need to be reduced from about 4 percent of GDP in 2003 to below 2.25 of GDP over the medium term (Table 6.2). Since debt service would increase by about 1.5 percent of GDP by 2009, about 70 percent of the increase in revenue will be needed just to meet additional debt service obligations and to facilitate the reduction of the primary deficit. Accordingly, given that the ratio of debt service to government revenue would average 7.75 percent of GDP over 2003–07, a fundamental adjustment in expenditure priorities would be required.

Table 6.2.Indicators of Debt Sustainability(In percent of GDP, unless otherwise indicated)
200120022003200420052006200720082009
Public debt–stabilizing primary deficit12.32.32.32.32.32.32.32.32.3
Primary deficit
Including grants (projected)2.73.64.33.93.83.43.02.62.2
Excluding grants5.46.56.86.15.85.44.84.43.8
Public external debt outstanding67.268.470.868.547.548.948.948.447.3
Public external debt service (cash basis)
In percent of revenues1.72.04.03.87.58.58.38.79.0
In percent of exports of goods and services1.00.91.20.91.92.02.22.42.5
Memorandum item
Public external debt service (accrual basis)
In percent of exports of goods and services3.12.82.92.31.92.02.22.42.5
Sources: Cambodian authorities; and IMF staff estimates.

Assumes 2.5 percent interest rate and 6.2 percent annual GDP growth.

Sources: Cambodian authorities; and IMF staff estimates.

Assumes 2.5 percent interest rate and 6.2 percent annual GDP growth.

Improving Expenditure Management and Service Delivery

Enhancing public expenditure management is at the heart of the government’s reform program in this area. Several institutional reforms are particularly important for success: (1) improving the realism of the annual budget; (2) streamlining the approval and control processes of expenditure execution to enable line ministries to better fulfill their mandates while remaining accountable; and (3) shortening the delays in disbursing budget appropriations to line ministries through better cash management, including through full centralization of government accounts in the treasury single account, as well as by taking the necessary steps to reduce the use of cash in government transactions.36

The first priority is to ensure that the national budget is realistic so that it becomes a strategic instrument for public policy. In this respect, an early priority of the PFMRP agenda is to develop capacity within the MEF to forecast revenues and expenditures. Thus, a key early action in the public financial management reform agenda is to develop an overarching resource mobilization policy and framework (including tax, nontax, and debt).37 In order to construct this critical building block, agency-specific responsibilities and cross-agency interactions have been identified, and a timeline has been laid out. An integral part of this challenge is, of course, to improve the macrofiscal framework and budget forecasting capabilities (both within-year and medium-term) of the fiscal authorities. Avoiding the accumulation of additional spending arrears has to begin with realistic fiscal projections.

The second priority is devolution of spending decisions to the line ministries while simultaneously holding them accountable for results. Key to granting more spending autonomy to line ministries without increasing the fiduciary risk to public funds is to improve the budget execution and cash management systems, as well as the effectiveness of public financial control mechanisms. These in turn hinge on substantially improved budgetary information. To this effect, an early ongoing priority is the redesign of the public accounting system—one that meets international standards, notably those set out in the IMF’s Government Finance Statistics manual—and introduction of an improved budget tracking system (a financial management information system).

Key Features of the Public Financial Management Reform Strategy

The public financial management reform strategy in Cambodia embodies a number of aspects critical for success.

  • Government ownership: The PFMRP in Cambodia combines the detailed reform strategies of each of the key agencies. Moreover, management of the reform process itself rests with the PFM Reform Committee within the MEF. The committee, chaired by a senior official of the ministry, comprises all the relevant agency heads, who in turn have full ownership of reforms at their agencies.
  • Donor coordination: Coordination is critical to enhancing capacity building and avoiding duplication and/or conflicting advice to country officials; either of these slippages results in wasted resources and delays in achieving the needed reforms. To ensure better coordination, Cambodia’s development partners involved in providing technical assistance in the public financial management area have formed a Development Partners Committee to ensure that providers of technical assistance work in a coordinated and harmonized manner.
  • Incentives to civil servants: Against a backdrop of totally inadequate formal wages paid to public employees in Cambodia, most donor-financed projects are accompanied by salary supplements paid to select staff. In an environment of low public wages, insufficient motivation, and moonlighting, this is often the only way to get the job done. However, a number of features of the salary supplementation systems used in Cambodia by development partners are widely seen as counterproductive and need to be addressed: they are not merit- or performance-based, they are subject to abuse and misuse, their levels and modes of disbursement vary widely across government agencies, and they retard progress toward a proper civil service pay reform.

Against this background, the PFMRP envisages a merit-based pay initiative (MBPI) for civil servants that is expected to play a key role in the reform. The MBPI approach strikes an appropriate balance between the need to attract and motivate able counterparts on the one hand and the risk of undermining civil service capacity on the other. Several features of this initiative are novel. First, under the proposed initiative, the government contribution to the wage supplement will increase over time so that by the end of the program the merit-based payments are fully financed by the government. This ensures that the salary supplements are sustainable and eventually integrated into the government’s salary structure. Second, the eligibility criteria for supplementary pay are transparent and strictly based on merit. This ensures that people with equal qualification and performance receive equal pay, which should reduce frictions within the civil service. Other features of the MBPI include a decompression of the pay scale across those civil servants participating in the public financial management reform program, taking into account cash and noncash benefits, and incorporates the overall medium-term impact on the wage bill.

D. Conclusion

Since Cambodia’s emergence from international isolation and war, fiscal policy has been at the center of the reconstruction process. Most importantly, it has been and must remain the cornerstone of the policy framework for macroeconomic stability. After more than a decade of struggle to reform the public finances, notable progress has been achieved. But serious shortcomings remain. These shortcomings are addressed head-on and comprehensively, and in a coordinated manner, under the PFMRP. If successfully implemented, the PFMRP holds the promise of enhancing considerably the contribution that the public sector needs to make to reduce poverty in Cambodia.

Appendix. Quantifying the Revenue Impact of Administrative Improvements

A number of different approaches can be used to estimate the impact of changes in tax and customs administration. The impact of some administrative reforms can be directly measured, such as reinforced efforts to collect tax arrears. This method can be characterized as a “bottom-up” approach. In most instances, however, estimation requires a more indirect approach. In this paper, the revenue yield from tax administration (TA) measures is derived by subtracting from revenue collected two other components that contribute to the revenue yield in any period.

The change in the total yield from a revenue item (e.g., tax or import duty) between two period comprises three components: (1) the change in the size of the imposed base in the absence of policy measures, (2) rate changes or policy-determined changes in the base, and (3) administrative improvements. Given observed revenue, disentangling the relative size of each component requires estimation of at least two. For present purposes, the contribution attributable to the change of the tax base is estimated by assuming a degree of automatic responsiveness of a revenue item to changes in the applicable or a proxy base (i.e., the elasticity of the revenue item). In this paper, a unitary elasticity is assumed, implying that on unchanged policy, the revenue yield should hypothetically increase (or decrease) at the same rate as the change in the proxy tax base. The contribution from policy measures is the impact of policy measures. Finally, the revenue impact of administrative changes is derived by subtracting the latter two estimates from actual revenue. Symbolically,

where TR is total revenue collected (actual); e is the elasticity of taxes and is assumed to be equal to 1; ΔB is the change in the tax base proxy (i.e., nominal GDP for domestic taxes and retained imports, excluding those related to the garment sector, for customs); and PM represents policy measures taken in time t.

Given the counterfactual nature of the estimation methodology, the results presented in this appendix should be taken as illustrative. For example, the impact of improvements in tax administration could have been larger if buoyancy was assumed to be less than 1, which could be the case in growing economic sectors with widespread exemptions (e.g., the garment sector).

FN25For a broader discussion of reforms in the 1990s, see IMF (2000).
FN26See the Appendix for a brief description of the methodology used to quantify the impact of administrative improvements.
FN27The main measures in the LOT, including the VAT, only began to be implemented in 1999, however. The LOT divided the Cambodian tax system into real and estimated regimes. The real regime, which defines the base of VAT taxpayers, covers incorporated businesses regardless of size of turnover, but excludes unincorporated businesses, many of which may in fact be large enterprises.
FN28However, improvements in tax administration over this period should be viewed relative to the very low level of state revenues in the base year. Even with these notable improvements, by 1999 there remained serious noncompliance and tax arrears problems under the “real regime” and a high degree of discretion allowed in tax assessments under the “estimated regime” (see IMF, 2000), which facilitated corruption.
FN29Proper resource management would have yielded lower revenue since logging took place at highly unsustainable rates.
FN30The TCAP, launched in May 2001, was a multidonor program of technical assistance designed to build capacity in the key institutions responsible for the formulation and implementation of macroeconomic policies. The TCAP also comprised technical assistance in the statistics and legal areas as well. See www.imf.org/Cambodia.
FN31The coverage of the VAT in Cambodia is much wider than the former turnover and consumption taxes, exempting only (1) public postal services, (2) hospital and medical services, (3) public transportation, (4) insurance and financial services, (5) imports for personal use exempted from customs duties, (6) nonprofit activities in the public interest, and (7) imports of goods related to diplomatic and international organizations.
FN32In particular, smuggling of gasoline and diesel is estimated to have been quite substantial in recent years. A recent analysis of the smuggling of petroleum products estimates lost revenue of about 260 billion riel ($65 million), about 1.6 percent of GDP. See IMF (2004d, Chapter 10).
FN33The FCU carries out treasury functions for those parts of the budget denominated in foreign currency.
FN34See IMF–World Bank Joint Staff Assessments, March 6, 2003, and October 22, 2004.
FN35Figures in this chapter assume that concessional rescheduling had been reached in mid-2004 on comparable terms to the 1995 Paris Club agreement (i.e., flow rescheduling on Naples terms), although such rescheduling has not yet been agreed. They assume a 40-year maturity with a 16-year grace period, and interest rate of 3 percent for the U.S. debt. Similar terms are assumed on the debt owed to the Russian Federation, after an initial up-front discount of 70 percent. For a discussion of other possible rescheduling scenarios, see IMF (2004a, Annex II).
FN36The PFMRP is of course much more comprehensive.
FN37See Activity 3 of the “Consolidated Action Plan” of the PFMRP (Royal Government of Cambodia, 2004).

    Other Resources Citing This Publication