Uganda: Selected Issues and Statistical Appendix
Author: International Monetary Fund
Publication Date:
05 Sep 1998
eISBN:
9781451838626
ISBN:
9781451838626
Language:
English
Keywords:
ISCR; CR; VAT; enterprise; taxpayer; government; tax; salary reform; wage bill; University employee; reform program; wage expense; Minimum wages; Wages; Ghost workers
Download PDF (4.2 MB)

This Selected Issues paper and Statistical Appendix describes how to improve value-added tax (VAT) compliance in Uganda. The paper highlights that although the VAT in Uganda has a single positive rate and broad coverage, its initial threshold of U Sh 20 million may have been set too low, and a number of items that should have been exempted were zero rated. This paper presents a brief survey of the financial sector of Uganda. Public sector reforms and the privatization program are also discussed.

Abstract

This Selected Issues paper and Statistical Appendix describes how to improve value-added tax (VAT) compliance in Uganda. The paper highlights that although the VAT in Uganda has a single positive rate and broad coverage, its initial threshold of U Sh 20 million may have been set too low, and a number of items that should have been exempted were zero rated. This paper presents a brief survey of the financial sector of Uganda. Public sector reforms and the privatization program are also discussed.

III. Public Service Reform5

23. At the time of political independence in 1962, Uganda could claim to have “the best public service in Africa south of Sahara” (Langseth, 1995). However, by the late 1980s, the public service was damaged by the years of civil strife and economic mismanagement. When the National Resistance Movement (NRM) came to power in 1986 after seven years of civil war, it inherited an ineffective public service. The main weaknesses of the public service included excessive centralization of power, the bloated size of the public service, and the inadequate pay and benefits of its employees.6 The other problem areas were an obscure and compressed salary structure, slow promotions, and a discretionary and nontransparent system of noncash benefits and allowances.

24. With the shift in the strategy of the new government toward macroeconomic stabilization, the development of the private sector and reduction in the role of the government through public service reform became an integral part of the reform agenda. The 1990 Civil Service Reform Program aimed at reducing the role of the government from being a major provider of jobs to creating an environment that supports private sector development, especially through the more efficient delivery of public services and provision of public goods, such as education and health care.

25. In its first stage, the main objective of the reform of the public service was to reduce its size to manageable levels and to improve its performance by rationalizing salaries, including their increases and the monetization of noncash benefits. This stage was implemented during fiscal years 1993/94–1996/97 (July–June). In the second stage, which is ongoing, the emphasis is on improving the performance and quality of the public service and adapting it to the requirements of decentralization. This paper reviews the evolution of the public service reform since the early 1990s.

A. Reduction of Public Service and Salary Reform

Size of public service

26. In January 1990, public service employment (comprising ministry headquarters, the teaching service, and police and prisons) totaled some 320,000 people.7 They were part of the central government and were paid from the budget. University employees and employees of the local government were also paid from the budget. Following several years of preparatory work on regrading jobs and redefining the needs of government employment, the central government size was reduced by more than half to just over 123,000 by December 1995; it increased afterward to almost 125,000 by December 1997, mainly because of increased hiring of primary teachers (Table 3 and Figure 1). Most of the reductions occurred during fiscal years 1993/94–1994/95.

Table 3:

Central Government Employment 1990–97 1/

article image
Source: Ministry of Public Service.

Includes the ministry headquarters, the teaching service, and police and prisons.

Figure 1:
Figure 1:

Central Government Civil Service Staff, 1993–97

(Indices, December 1993 =100; end of period)

Citation: IMF Staff Country Reports 1998, 061; 10.5089/9781451838626.002.A003

27. By end-December 1997, 75 percent of total central government employment represented the teaching service, and the rest were accounted for by police and prisons and by the traditional civil service (Table 4). The steepest cuts took place in the traditional civil service and support staff. The number of ministries was reduced from 34 in the early 1990s to 21 (in July 1996, this number rose to 22 as the Ministry of Finance, Planning and Economic Development was divided into two). During the 1990s, employment in universities rose moderately (to just over 5,000). However, with the move toward decentralization, which became a part of the new constitution of October 1995, district-based staff increased to nearly 35,000 by December 1997; 23 percent were directly hired by local governments, 66 percent represented decentralized staff transferred from the central government to the districts, and 11 percent were “delegated” staff, that is, staff deployed by the central government in nondecentralized services (such as referral hospitals) or temporarily lent to the districts to assist their operations. Thus, the total size of the central government, universities, and district administration stood at just over 164,000 people by end-December 1997. These numbers do not include the armed forces, which were subject to restructuring and size reductions under a separate program (the army demobilization reform program) that was carried out at about the same time as the public service reform.

Table 4.

Uganda: Size of the Ugandan Civil Service, 1993-97

article image
Source: Ministry of Public Service.

28. Payroll management has also changed with the onset of decentralization. The local governments were provided with block cash grants for paying their wage bill; they were not accountable to the central government for the size of their staffs, as long as their wage expenses were within their cash allocations. This system also applies to the universities and the armed forces. These three categories (local government employees, universities, and the armed forces) became identified as the “cash-limited” public service, whereas the central government payroll (traditional civil service, the non-university teaching service, police and prisons, and support staff) was identified as the “number-limited” public service (Figure 2).8 Discussion and policy with regard to the size of the public service in the context of reform is focused on the “number-limited” public service. The size of the “cash-limited” public service in some cases is not public information (e.g., the armed forces), or the information is available with a lag (e.g., district staff).

Figure 2:
Figure 2:

Size of the Civil Service, June 1993-December 1997

(In numbers of employees)

Citation: IMF Staff Country Reports 1998, 061; 10.5089/9781451838626.002.A003

29. In the early 1990s, before the decentralization, it was generally considered (although it was never formal government policy) that the appropriate size of the public service (excluding the armed forces) should be about 150,000 (compared with the 320,000 employees at that time). With decentralization, this concept has become outdated and has not yet been replaced by a firm view of the optimal size of the “number-limited” public service—partly because the process of shifting functions and staff from the center to the districts has not been completed (see below), and partly because the launching of the Universal Primary Education (UPE) program in early 1997 has necessitated a significant increase in the number of primary school teachers.

30. Monitoring the size of the (“number-limited”) public service is the responsibility of the Ministry of Public Service. The principal indicator is the monthly payroll statement, which records the number of paid positions in each category of public employment. The tracking system has occasionally been affected by various problems. For example, when staff or teachers were transferred from one establishment to another, they were dropped from the payroll of the original establishment, but re-registered on the payroll of the other establishment with a significant delay, thereby giving an inaccurate picture of the size of the public service in the interim. In a number of instances, ghost workers have reappeared on the payroll, and the tracking system has not been able to ensure their early detection. In other instances, ad hoc policy decisions (for example, the automatic hiring of graduates from Makarere University) have been inconsistent with the overall policy on public service size. These factors contributed to an upward drift in the size of the public service in 1996. The Ministry of Public Service has examined these factors with a view to improving the tracking system. Monitoring has generally been strengthened, and the upward drift of the public service (excluding primary school teachers) has been reversed. However, the monitoring of primary school teachers in the context of the UPE program was initially weak. In particular, the sharp increases in primary school teachers in 1997 were not fully reflected in the payroll, resulting in wage arrears to teachers. With the completion of the teacher-student census late in 1997, and with assistance from a World Bank Educational Structural Adjustment Credit (ESAC), the authorities intend to substantially improve the payroll management of the UPE program.

31. In its approach to staff cuts and improvements in payroll management, the government opted for a variety of methods including removal of ghost workers, retrenchment (both mandatory and voluntary), and freezing of employment growth (through less hiring).

32. The most important method of reducing the size of the public service was the retrenchment of workers. From 1988/89, retirement rules started to be enforced, that is, those above the retirement age were required to leave the civil service. However, the number of staff reduced through this method was relatively small, mainly because civil service employees in Uganda are relatively young. This reduction was supplemented by the laying off of temporary workers, by the nonrenewal of contracts, and by the abolishment of the Group Employee Scheme in 1995. The Group Employee Scheme had allowed civil service managers to hire nonpensionable, short-term employees and was also a source of considerable abuse. The elimination of the scheme thus not only reduced employment (by about 30,000) but also increased the transparency of the public service. Voluntary retrenchment (including an early retirement scheme) was offered with the incentive of a retrenchment package.9 Mandatory retrenchment was also utilized; this provided a retrenchment package, but it was less lucrative than the one provided to the workers who left voluntarily. The retrenchment costs (for both voluntary and mandatory retrenchment) for established workers were largely financed by foreign donors. Table 5 illustrates average retrenchment packages provided between 1995 and 1997. For both established and support employees, the amount of the average retrenchment package decreased between 1995 and 1997; the decrease was dramatic in the case of established employees.

Table 5.

Uganda: Retrenchment Packages, 1995–97

article image
Source: Ministry of Public Service.

Financed by donors.

Financed by the government of Uganda.

33. Employment reductions also occurred through removal of ghost workers. This method was used after the survey conducted in 1989–90 revealed a large number of such workers. About 42,000 ghost workers were removed from the government payroll between 1991 and 1994 (Langseth, 1995). This was accomplished with the help of the Payroll Monitoring Unit, which was established to check all entries and exits from the payroll. However, the payroll monitoring was not fully effective, and the problem of ghost workers recurred in 1996. The Ministry of Public Service conducted an ad hoc investigation as a result of which about 2,000 ghost workers were removed from the payroll between October 1996 and February 1997. An improved monitoring system was established in January 1997 in order to prevent the reoccurrence of ghost workers.

34. The government also opted on occasion to a hiring freeze. However, this has proved a blunt instrument, and the urgent need to fill managerial positions in certain ministries has on occasion caused problems. In practice, the hiring freeze was only partially implemented. In some instances employees (e.g., new graduates) have been hired but not put on the payroll pending the lifting of freeze; the consequence has been the need to pay wage arrears when the freeze has been lifted, complicating budgetary management.

Salary reform

35. The reduction in the size of the public service was intended to make feasible a leaner but better-paid public service. In that connection, the main objective of the salary reform was to provide better incentives for public service workers to improve their performance by monetizing noncash benefits, raising basic salaries, and rationalizing the salary structure. This reform was intended to increase transparency and remuneration in order to improve incentives and performance.

36. The “number-limited” public service (excluding the support staff) is hierarchically divided into eight grades (from U8 to U1; U8 is the lowest), and a salary range is attached to each grade. In the early 1990s, workers were remunerated with a basic (monetary) salary plus noncash allowances. The monetization of transportation and housing benefits started in fiscal year 1995/96 and was completed in 1996/97. Table 6 below shows monthly average salaries for employees in the traditional civil service during 1994/95–1996/97. The percentage increase in nominal wages between 1995/96 and 1996/97 is much higher than between 1994/95 and 1995/96 because the monetized benefits are included.

Table 6.

Uganda: Average Nominal Monthly Wages and Housing Benefits, 1994/95–1996/97

(In Uganda shillings)

article image
Source: Ministry of Public Service (1995), Annex B.

37. In terms of cash payments, the salary structure was relatively flat and did not appear to sufficiently reward higher skills. The uniform 10 percent increase in wages between 1994/95 and 1995/96 did not appreciably affect the spread between the lowest-paid and highest-paid traditional civil servants. However, as benefits-in-kind (which had been unevenly distributed) were monetized, the effective spread between the lowest and highest paid employees became more obvious. These developments are illustrated by the shift in the distribution of civil servants by wages in Figure 3. The cumulative distribution of the number of workers for a given salary level is on the x-axis, and the share of wages received by those workers is on the y-axis. The shift in the curve between 1994/95 and 1996/97 indicates that the distribution of wages became more unequal, that is, the highest-paid civil servants received a larger share of the total wage bill. Table 7 provides summary statistics on the wage distribution. It shows that the top 10 percent of employees received 18 percent of total salaries in 1995/96, but 25 percent in 1996/97, owing to monetization of in-kind benefits. Figure 3 and Table 7 show cash wages (and do not include noncash benefits) and, therefore, are not a basis for a strict intertemporal comparison.

Figure 3:
Figure 3:

Distribution of Civil Servants by Wages, 1994/95–1996/971

Citation: IMF Staff Country Reports 1998, 061; 10.5089/9781451838626.002.A003

1/ Fiscal year July-June.
Table 7.

Uganda: Degree of Inequality in Distribution of Salaries, 1994/95–1996/97

article image

38. Another objective of salary reform was to pay civil servants at least a minimum acceptable salary for the most junior grade of the public service.10 In 1994/95, a national minimum wage for civil servants was determined at U Sh 70,000 a month (derived from the food basket/cost of living survey in 1989/90 and using 1990 prices) with no noncash benefits. Soon thereafter, the government recognized that the concept of the minimum wage was not motivating staff or attracting qualified outside workers. The focus on the minimum wage did not solve the problem of a compressed salary structure; if salaries of civil servants in higher grades were not increased at least proportionally to increases in minimum wages, the salary structure would become further compressed. The concept of a minimum wage was replaced by the concept of a “living wage.” In addition to providing for basic needs, the “living wage” also takes account of the social costs (such as health, education, transportation, and clothing) that were appropriate for each position (i.e., the living wage increases with grade). The determination of the “living wage” was based on the household survey conducted in 1992 (derived from the average total monthly household expenditures by different income groups and adjusted for inflation) and it included full monetization of noncash benefits. In 1996/97, the move toward application of a “living wage” was implemented, but because of resource constraints the targets were not reached for all groups. Figure 4 provides information on wages in relation to the targeted “living wage” for various grades in 1996/97. The highest priority was accorded to increasing the salaries of the lowest-paid civil servants. Actual wages were less than 80 percent of the targeted “living wage” for staff in grades U5a–U1, that is, for the middle and senior managers. Even though it is not likely that salary levels determined on the basis of the living wage concept accurately reflect productivity levels, the failure to meet the expectations of mid-level civil servants had some demoralizing impact. Raising salaries of the middle and senior managers became a priority for the future, budgetary resources permitting. In January 1998, some wage adjustments were made for mid-level civil servants.

Figure 4:
Figure 4:

The Actual Wages as Percentage of the Targeted Living Wage in 1996/97, by Occupations and Grades 1

Citation: IMF Staff Country Reports 1998, 061; 10.5089/9781451838626.002.A003

1/ Fiscal year July-June.

39. In order to enhance the efficiency of the public service, an outcome-oriented performance is being introduced by linking the allocation of resources to objectives in the provision of public services. More specifically, before the start of each fiscal year, each part of the public sector will clearly state its objectives, activities to be funded from the central budget, and, if feasible, relevant performance indicators.

B. Decentralization

40. At the end of the civil war in 1986, the authority to make decisions and manage public functions in Uganda was almost exclusively concentrated in the central government. Decentralization became an early goal of the new government. It was intended to transfer substantive political, administrative, and financial authority to make decisions and manage public functions from the central government to local governments. The objective of political decentralization was to democratize local governments and increase local participation in decision making. The objective of administrative decentralization was to build managerial capacity at the local level and to reorganize the public service to increase efficiency in the delivery of services. The objective of financial decentralization was to increase the self-sufficiency of local governments. Effective implementation of these interlinked aspects of decentralization was intended to increase transparency, accountability, and the efficiency of service delivery.

41. In 1987, the government introduced political decentralization through Resistance Councils established at the following levels: villages, parishes, subcounties, counties, and districts.11 The government has also been exploring implications of administrative and financial decentralization since 1986; a program was formalized in the 1993 Local Governments Statute, enshrined in the new constitution in 1995, and consequently in the Local Governments Act of 1997. The program aimed at placing delivery of most public services with local governments (with certain exceptions, such as national defense); the central government was to maintain mainly a policymaking role. Implementation of administrative decentralization started in 1994, when employees of the central government that worked in the districts became employees of the districts (with some exceptions) and, hence, accountable to the district government. The current phase of public service reform deals with a reassessment of the size of the central government in the light of decentralization and the shift of many of the functions from the center to the districts.

42. Prior to decentralization, the central government was handling the recruitment, training, and promotions of district employees.12 However, since district matters were not always the center’s highest priority, the recruitment and firing needs of the local governments were not met. Local governments solved this problem by hiring temporary workers; although they were often very poorly qualified, their contracts could at least be periodically extended beyond the original date. In the absence of training programs, the quality and efficiency of the provision of services in the districts suffered. With administrative decentralization, local governments gained the right to make their own employment decisions. Instead of being controlled by the center, recruitment was determined by resources provided to each district (“cash-limited” employment). Local governments were no longer limited by the previous organizational structure, and were free to establish and eliminate posts and departments. Most councils (basic administrative units of the local governments) responded by retrenching workers and eliminating departments; the average number of departments per council went down from 20 to 8 (Lubanga, 1997). At the same time, the recruitment of qualified staff has increased during the last several years (Table 8). Also, with the new organizational structures and new staffs, methods of providing service have changed. Several types of services, such as road maintenance and construction, have been contracted out to the private sector.

Table 8.

Uganda: Number of Staff Employed by Professional Category in the District Planning Units, March 1997

(By number of years)

article image
Source: Lubanga (1997).

43. A scarcity of accounting and financial management skills is one of the key constraints on effective decentralization. For example, even though with decentralization the local governments acquired the right to manage their own payrolls, in practice they are still not able to do it.13 The goal is that the majority of districts will be able to prepare their own monthly payrolls within three years. The transparency of local government operations is further weakened by the poor recording of data on taxpayers and other crucial statistics and by unsatisfactory financial reporting.

44. Financial decentralization is being implemented gradually. Local governments still finance their expenditures mainly from transfers from the central government budget. Additional resources derive from local revenues (taxes, fees, and user charges) and foreign donors’ grants.14 Local governments can keep over 60 percent of the revenues that they collect through taxes, fees, and user charges. They currently receive two types of grants from the center: unconditional (block) grants and conditional grants. Unconditional grants are spent on financing decentralized services, and their allocation is at the discretion of local governments. Conditional grants are provided to finance high-priority expenditures, such as salaries of teachers and medical personnel. As decentralization progresses, conditional grants are intended to be phased out. Moreover, while previously transfers from the Ministry of Finance to the districts were channeled through line ministries, they currently go directly to districts. Part of the central government’s current budget has been decentralized to cover all districts. The government intends to start decentralization of the development budget in 1998/99.

45. Problems related to financial decentralization include inadequate locally generated revenues, a shortage of technical and administrative skills, an undeveloped system of public accounting, and a lack of monitoring and transparency in the use of resources. In addition, the efficient provision of decentralized services is occasionally disrupted by unexpected cuts in the unconditional grants, owing to resource constraints at the center.

Bibliography

  • Chew, David, 1990, “Internal Adjustment to Falling Civil Service Salaries: Insights from Uganda,” World Development, Vol. 18 (July), No. 7 pp 1003–14.

    • Search Google Scholar
    • Export Citation

  • Lindauer, David L., 1996, “Government Pay and Employment Policies and Economic Performance,” in Rehabilitating Government: Pay and Employment Reform in Africa, ed. by D. Lindauer and B. Nunberg (Aldershot, England and Brookfield, Vermont: Avebury Publishing), pp. 17–32.

    • Search Google Scholar
    • Export Citation

  • Lubanga, Francis, 1997, Decentralization Program in Uganda: Results and Problems (Kampala, Uganda), Ministry of Local Government.

  • Langseth, Petter, 1995, “Civil Service Reform in Uganda: Lessons Learned,” Public Administration and Development, Vol. 15 (October), pp. 365–90.

    • Search Google Scholar
    • Export Citation

  • Ministry of Public Service of Uganda, 1995, Public Service Reform, Status Report 4. (Uganda, Kampala).

  • Ministry of Public Service of Uganda, 1997, Public Service Reform, Status Report 12. (Uganda, Kampala).

  • Ministry of Planning and Economic Development of Uganda, 1997, Statistical Abstract, (Uganda, Kampala).

  • Ministry of Public Service of Uganda, 1994, “Civil Service Reform: Context, Visions, Objectives, Strategy and Plan,” Uganda, Kampala.

    • Search Google Scholar
    • Export Citation

  • Musisi, Kiwanuka, 1997, “Achievements, Shortcomings, and Way Forward of Decentralization in Uganda: The Views of a Practitioner,” a paper presented at the Decentralization Review Workshop, Kampala, Uganda.

    • Search Google Scholar
    • Export Citation

  • Nsibambi, Apolo, 1997, “Implementing Public Policy Reform with Special Reference to Africa” (unpublished; Kampala, Uganda).

5

Prepared by Zuzana Brixiova.

6

The Public Sector Review and Reorganization Commission, which was established by the NRM, identified the weaknesses of the civil service based on a survey of 25,000 civil service employees conducted in 1989–90.

8

As the terms imply, the “number-limited” public service encompasses those areas where budget resources are made available based on approved establishments; in the case of the “cash-limited” public service, transfers are made from the budget and the number of employees is not constrained as long as their wage bill conforms to the resources made available.

9

In the case of early retirement, the departing person could either receive the pension earlier, or receive a lump-sum package at the time of retirement and the pension when reaching retirement age. An examination of the fiscal implications of early retirement, and of public service reform on pensions, is beyond the scope of this paper.

10

More specifically, the minimum wage was defined as equivalent to the cost of a basket of goods and services needed for sustenance.

11

Currently, there are 45 districts (including Kampala). There are no layers of government between the center and districts.

12

Article 104 of the 1967 constitution established that the President is responsible for hiring and firing all grades of the established civil servants in the local governments (Lubanga, 1997).

13

The exception is the Rakai district, which now keeps its own payroll records and issues its own checks.

14

The Local Governments Act of 1997 also allows the local governments to borrow from the banking system, but this option has been used only very little so far.

Same Series

Other IMF Content

Other Publishers

Cover IMF Staff Country Reports
Uganda: Selected Issues and Statistical Appendix
Author: International Monetary Fund
Volume/Issue:
Volume 1998: Issue 061
Publisher:
International Monetary Fund
ISBN:
9781451838626
ISSN:
1934-7685
Pages:
90
DOI:
https://doi.org/10.5089/9781451838626.002
  • View in gallery

    Central Government Civil Service Staff, 1993–97

    (Indices, December 1993 =100; end of period)

  • View in gallery

    Size of the Civil Service, June 1993-December 1997

    (In numbers of employees)

  • View in gallery

    Distribution of Civil Servants by Wages, 1994/95–1996/971

  • View in gallery

    The Actual Wages as Percentage of the Targeted Living Wage in 1996/97, by Occupations and Grades 1

Central Government Civil Service Staff, 1993–97

(Indices, December 1993 =100; end of period)

Size of the Civil Service, June 1993-December 1997

(In numbers of employees)

Distribution of Civil Servants by Wages, 1994/95–1996/971

The Actual Wages as Percentage of the Targeted Living Wage in 1996/97, by Occupations and Grades 1