VIEWS: 316 PAGES: 42



              Final Report

                  Prepared by

          Theodore R. Valentine, Ph.D.
      Crown Management Consultants Limited
             3rd Floor, Twiga House
                 Samora Avenue
                  P.O. Box 3092
            Dar es Salaam, Tanzania
Budget Analysis to Facilitate Pay Reform                       Final Report – 08 May 2004

                                           Table of Contents

LIST OF ABBREVIATIONS                                                                       IV

I.     INTRODUCTION                                                                         1

PERFORMANCE                                                                                 2

Government adherence to Annual Salary Adjustment Plans                                       3
 Employment trends                                                                           4

JERG implementation                                                                          5

SASE implementation                                                                          5

Reinforcing pay and incentives regime problems?                                              6

III.     BUDGET ANALYSIS: CROSSING THE LINE BETWEEN PE/OC                                   8

Sitting Allowances and Honoraria                                                            12

MACROECONOMIC VARIABLES                                                                     14

TECHNICAL AND MANAGERIAL CADRES                               19

Major issues in SASE implementation                                                         21

Need for an objective assessment of the Scheme                                              22

CEILING                                                                                     23

MoF, the MTEF and the adequacy of Wage Bill allocations                                     24

PSM, Establishment Controls and the Optimization Challenge                                  24

The Way Forward                                                                             25
  Strengthening the Policy Process                                                          25
  Establishment of an independent Compensation Commission                                   27

VII.     CONCLUSIONS AND IMPLICATIONS OF THE REPORT                                         28

Recommendations                                                                             30

Budget Analysis to Facilitate Pay Reform                      Final Report – 08 May 2004


APPENDIX 2. LIST OF PERSONS INTERVIEWED                                                    35

BIBLIOGRAPHY                                                                               36

                                           Table of Tables

   TRENDS IN PAY PARAMETERS                                                4
   FY01/02 - FY03/04                                                      10

                                           Table of Figures

   ALLOWANCES, FY01/02 EQUALS 100                                          11
   COUNTRIES, FY2002/03                                                    16
   COUNTRIES, FY2002/03                                                    17
   REVENUE FOR SELECTED COUNTRIES, FY2002/03                               18

Budget Analysis to Facilitate Pay Reform                 Final Report – 08 May 2004

                                 LIST OF ABBREVIATIONS

CG        -       Central Government

GoT       -      Government of Tanzania

IMWG -           Inter-Ministerial Working Group

JERG      -      Job Evaluation and Re-grading

LCC       -      Local Compensation Cost

LG        -      Local Government

MDAs      -      Ministries, Departments, and Agencies

MoF       -      Ministry of Finance

MTPRS -          Medium Term Pay Reform Strategy

OC        -      Other Charges

PE        -      Personnel Emoluments

PEDP      -      Primary Education Development Program

PIF       -      Performance Improvement Fund

PIM       -      Performance Improvement Model

PSM       -      Public Service Management

PSR       -      Public Service Reform

PSRP      -      Public Service Reform Programme

SASE      -      Selective Accelerated Salary Enhancement

TGS       -      Tanzania Government Service


I.     Introduction

1.     Inasmuch as low pay is a major contributing factor in the low capacity
of the public service to adequately attract, retain and motivate quality skilled
and professional personnel, the capacity of the government to adequately
enhance pay and to strengthen the incentives regime pose challenges to:

       a. Sustain the public service reform effort;

       b. Improve governance; and

       c. Improve public service delivery and accessibility.

2.     The Government of Tanzania (GoT) has long recognized the potential
role of pay reform as a facilitator of broader public service reforms and
improved service delivery, sustained public service reforms, and addressing
issues of wage bill controls. Hence, the decision to adopt a Medium Term Pay
Policy (MTPP) for the Public Service and to develop and implement a Medium
Term Pay Reform Strategy (MTPRS) to operationalize the MTPP.

3.     While there has undoubtedly been much success in implementing the
MTPRS, in recent years concerns have grown among major stakeholders that
the pace of implementation has slowed. If the pace slows further it may give
way to inertia which could induce a return to a demotivated public service,
rekindle efforts among public servants to supplement base salaries with
alternative sources of income, and induce deviant work behavior.

4.     Addressing these concerns, the Public Service Reform Program
(PSRP) commissioned this assignment to analyze the budget for the purpose
of facilitating pay reform. The major objectives of the assignment, as
specified in the Terms of Reference (TOR), are to enable the GoT to:

       a. Assess the ability of the Public Service to increase public servants‟
          salaries to levels that will enhance motivation and productivity;

       b. Ensure that the recommended salaries are competitive for attracting
          and retaining competent staff; and

       c. Review the structure of the Government budget to identify some
          scope for allocating more resources to salary payments.

5.     Reflecting the major tasks set out in the ToR,1 the structure of this
report is as follows:

       a. Section II presents an overview of issues related to the

           For the complete ToR for the assignment, see Appendix 1.
Budget Analysis to Facilitate Pay Reform                  Final Report – 08 May 2004

            implementation of the MTPRS.

        b. Section III examines the pattern of expenditure of budget resources
           between Personnel Emoluments (PE) and Other Charges (OC) and
           considered what expenditures, and suggests which OC
           expenditures if any have an element of PE, and thus may be shifted
           to PE.

        c. Section IV examines trends in the PE and OC expenditure ratios in
           relation to GDP and makes proposals on the appropriate wage bill-
           to-GDP ratios in the medium term.

        d. Section V focuses on how the Selected Accelerated Salary
           Enhancement (SASE) can best play its role as a temporary
           mechanism to attract and retain key personnel in MDAs while
           government has yet to reach the Medium Term Pay Targets,
           without creating too much distortion in the pay structure.

        e. Section VI examines existing arrangements and mechanisms for
           decision-making on annual salary adjustments and recommends
           appropriate adjustments with a view to ensuring the timeframes are
           aligned to the budget decision making process.

        f. And finally, section VII presents conclusions and implications of the

II.   Medium Term Pay Reform Strategy: Its implementation, potential
and limitations in contributing to improved performance

6.      Enhanced public service pay and an improved incentives regime,
especially for qualified technical and professional cadres, are crucial if
significant improvements are to be achieved in the accessibility, delivery and
quality of essential public services. However, given the prevailing budgetary
constraints, there were no simple options for enhancing salaries and
smoothing the salary structure in the reform of the compensation system for
the public service. As a matter of fact, in the short term, the options for
effective all-encompassing pay policy measures are very limited. Hence, the
necessity for the GoT to formulate and implement a comprehensive and
systematic MTPRS.

7.     Over time, the Government has adopted several measures that
evolved into a more systematic approach to pay reform. These measures

        a. The simplification of the pay structure and payroll systems and
           improved the transparency of the remuneration structure (1996) by:

            i.   reducing the number of job grades; and

Budget Analysis to Facilitate Pay Reform                                  Final Report – 08 May 2004

              ii. consolidating remunerative allowances into the salary structure;

        b. The adoption of a MTPP for the Public Service (1998);

        c. The development and adoption of a MTPRS to operationalize the
           MTPP (1999);

        d. Completion of the Job Evaluation and Re-Grading (JERG) exercise
           (2000); and

        e. The adoption of the Selective Accelerated Salary Enhancement
           (SASE) scheme (2000).

8.      Much has been achieved in the area of pay reform, including:

        a. Improved transparency of the remuneration system, which
           facilitated improved wage bill management and greater
           predictability of the budget outcomes;

        b. Improved rationality of the incentives regime;

        c. Improved equity (both horizontal and vertical) in the remuneration
           structure; and

        d. Enhanced real compensation levels across the public service.

9.      However, it is in the last area where much remains to be done in line
with the MTPRS. Though salary adjustments are now made on a much more
regular basis than in the past and reviews of pay are more systematic, there is
still a major divergence between the commitment made by the GoT in
adopting of the MTPP and MTPRS implementation practices.

Government adherence to Annual Salary Adjustment Plans

10.   Medium term pay targets were initially adopted for the period FY99/00
–FY04/05, with its annual salary adjustment trajectory (Valentine, 1999). The
extent to which the strategy was implemented and impediments to the
implementation process are now areas of growing concern.

11.    Analysis of the extent to which annual salary adjustment plans have
been adhered to show that for all salary grades, the actual salary scales
were below the targets and the differences were much more significant than
for the previous year. The differentials range from 13% in the lower salary
grades to as much as 21% to 32% for the middle and upper salary grades,
with the divergence between the target salary levels and actual salary levels
growing over time.2

            For more detailed discussions, see: Valentine (2002) and Valentine (2003a).

Budget Analysis to Facilitate Pay Reform                                 Final Report – 08 May 2004

12.    Why have actual salary adjustments fallen short of annual target salary
levels? The explanation for the deviation between the actual annual salary
adjustments and the target salary levels lies in part in actual versus the

        a. Employment trends;

        b. Wage bill trends; and

        c. Salary differentials/relativities.

13.     An overview of these trends is presented in Table 1.

Table 1: Contrasting Proposed Policy Directions of the MTPRS with Actual
Trends in Pay Parameters

   Contrast         Employment Trends             Wage Bill/GDP Ratio                Salary Structure
Implementation     Personnel on the           The Wage Bill /GDP ratio       The immediate
Proposals           Central Government          projected to rise gradually     implementation of the
Under the           (CG) payroll                from its (then) expected        recommendations of the
MTPRS               employment were             4.4% in FY99/00 to 4.9%         JERG was required to
                    projected to decline by     by FY04/05.                     upgrade professional and
                    14.7%, from about                                           senior technical positions,
                    262,080 in FY99/00 to                                       changing job relativities.
                    about 223,512 by the                                       Salary differentials
                    end of FY04/05.                                             (increment and grade
                                                                                differentials) would be
                                                                                increased through the salary
                                                                                structure to improve the
                                                                                awards for skills, experience,
                                                                                and responsibilities.
Actual             Personnel on the CG        The Wage Bill/GDP ratio        JERG was recently
Implementation      payroll rose by 5% to       actually declined from          implemented.
Under the           about 273,811 in            4.4% in FY99/00 to 4.2%        The salary structure remains
MTPRS               FY03/04.                    in FY01/02. Increased           very compressed with low
                                                back to 4.4% in FY02/03         salary differences,
                                                and remained constant in        particularly in the lower- and
                                                FY03/04.                        middle-salary grades.
                                                                               Differentials vary from year-
                                                                                to-year to accommodate the
                                                                                wage bill fit.

Employment trends

14.      In contrast to the projection that personnel on the CG payroll
would decline by 14.7%, employment has actually increased 5.0%. The
difference between the projected and actual employment trends is so stark
that it requires further analysis.

15.    The difference in the employment situation does not reflect backsliding,
but a re-thinking as to what constitutes a right-sized public service in
Tanzania. Given the commitment to improve the quality and accessibility of
social services and the need to build capacity among the professional and
technical personnel:

Budget Analysis to Facilitate Pay Reform                                   Final Report – 08 May 2004

        a. Under the Primary Education Development Program (PEDP), there
           is a planned net growth in the primary teachers by 9,100 in 2002;
           11,651 in 2003; 10,563 in 2004; 7,286 in 2005; and 7,249 in 2006;3

        b. After years of maintaining an employment freeze, there is now an
           active attempt to recruit young professional and qualified technical

16.   The slight increase in the wage bill that did occur over the past few
years was required to facilitate the expansion of employment in primary
school teachers.

JERG implementation

17.    Getting the job-grade structure right is important for improving the links
between pay and performance. By appropriately rank-ordering jobs in line with
the perceived value of each job‟s content, through JERG, the fairness of the
compensation system can be improved in the eyes of most employees
(though probably not for those whose jobs are downgraded). That is,
changing relative compensation through rationalizing job-grades (rank-order)
can alter workers‟ perceptions of the system‟s fairness.

18.   The JERG has finally been implemented at the CG level, after three
years of inaction on grounds of lack of availability of finances to cover its
implementation costs.

SASE implementation

19.    The pace of roll-out of the SASE scheme has been much slower than
anticipated. Whereas it was originally envisaged that the scheme would be
operationalized over the period FY2000/01 – FY2004/05, to date only four
MDAs have benefited from the scheme. Detailed discussions of the issues
related to SASE-scheme implementation are presented in Section V.

20.    It is sufficed to say at this point that to address these concerns and to
chart a path to get the MTPRS back on track, the Government undertook a
review and revision of the MTPRS and the SASE.4 Among other things, the
MTPRS implementation review studies proposed that MTPRS should be re-
launched with a new timeframe (FY02/03 – FY07/08), a new target pay
structure, and implemented with new vigour.

21.  The GoT has agreed on the revised time horizon for the MTPRS and
SASE scheme implementation. However, it was directed that SASE scheme

            For a detailed discussion of the PEDP, see: United Republic of Tanzania (2001).
         For discussion, see:Valentine (2002) and Valentine (2003a).

Budget Analysis to Facilitate Pay Reform                               Final Report – 08 May 2004

participation would be restricted to the current beneficiaries. For the other
MDAs, PO – PSM should make proposals on alternative means to achieve
the objectives which SASE is intended to do.

Reinforcing pay and incentives regime problems?

22.    There is a legitimate and growing concern among stakeholders that the
slower than expected pace of MTPRS and SASE scheme implementation
may inhibit efforts to improve work motivation and performance throughout the
public service.

23.    In the past public service capacity suffered increasingly, not only
because of the reduced financial capacity of the GoT to fund its social and
economic programs resulting from budgetary stress, but also because many
public servants pursued an exit strategy, leaving government employment
altogether or by exiting on-the-job, using work hours and public resources to
engage in their own income-maintenance strategies. With the increasing loss
of human and motivational capital, the failure to develop professional and
management cadres had significant negative implications for the quality of
public sector performance.

24.    The Government and public servants can take comfort in the fact that
the present situation is a far cry from the past, when public service pay was
allowed to be eroded significantly. Still the fact that there has been significant
slippage in the implementation of the MTPRS and the perceived lack of
Government commitment to adhering to salary targets may reinforce the need
for some public servants to seek alternative sources of incomes to
supplement their public service base salaries. The literature on deviant work
behavior, which analyzes problems in the workplace in private enterprise in
advanced economies (including employee theft and strategies to combat such
behavior), is instructive here.5 Deviance within the economics context, is seen
to take two forms:

        a. Production deviance, which includes work slow downs, misuse of
           paid work time to engage in non-work activities, or “time theft”;

        b. Property deviance, where it refers to the (mis)appropriation, misuse
           of public institutions‟ tangible and/or financial assets for personal

25.   In deviant work behavior analysis, demotivation, reduced work
performance, counterproductive behavior and misuse/misappropriate of work
resources for personal gain can be induced by the perception that the
employer has breached his contract with employees as individuals or as a
group. The contract may be deemed as a social contract or a psychological

         For discussion of deviant work behavior, see: Greenberg, 1990; Niehoff and Paul, 2000;
Robinson, 1995; and Robinson, 1996.

Budget Analysis to Facilitate Pay Reform                       Final Report – 08 May 2004

contract; it may be implicit or explicit. When income-maintenance strategies
and poor performance is viewed within this context, the analysis is robust
regarding how to address the work behavior and poor performance problem.
It is important to recognize that such behavior is not just a developing country,
public sector phenomena. It is more widespread, affecting private and public
sectors in more advanced countries as well. When taking into consideration
the increased economic deprivation that many public servants in developing
countries have experienced over a significant period of time, and the
increased perception among many public servants that the reform process
was merely a cost-cutting and downsizing exercise, the challenges to
improving performance become clearer.

26.   These income-maintenance strategies took on various forms. While
some had no direct bearing on work performance and service delivery, though
they might affect commitment to employment and work motivation, others did.
Those strategies that had the greatest impact on service delivery fall into four

        a. Work sharing, where public service personnel agreed implicitly/
           informally between themselves to share work loads to allow workers
           to pursue other income maintenance activities during official work
           time, lowering actual work-time and generating personnel rotations
           at staffing levels far below those formally expected;

        b. Cost sharing, where public servants solicited payments directly from
           customers/clients before providing (otherwise) free public services,
           and/or request of payments in excess of the formal service fees or
           user charges set by public institutions;

        c. Revenue sharing, where revenue from formal user charges were
           misappropriated or mismanaged for individual gain; and

        d. Resource sharing, where public resources, supplies and equipment
           were misappropriated, pilfered or mismanaged for individual gain.

27.   Taken collectively the above factors contributed to the worsening of the
GoT‟s already precarious fiscal situation, further constraining the
government‟s capacity to compensate public service personnel adequately.

28.    Taken together these individual income-maintenance strategies should
be recognized as important in shaping public service delivery systems as are
planned structures and actions.      The consequences were the further
exacerbation of the erosion of the public services, particularly in the social
sector, above and beyond that observed by the direct reduction of
government budgetary allocations. These consequences could be felt in
terms of their effects on:

        a. Declining efficiency in the delivery of services;

Budget Analysis to Facilitate Pay Reform                   Final Report – 08 May 2004

        b. Declining quality of services;

        c. Reducing accessibility of services;

        d. Increasing the cost of provision of services; and

        e. Increasing the incidence of user charges, both formal and informal.

29.    In addition to reduced work effort, low and declining remuneration
engendered the perception among many public servants that the GoT had
reneged on its social contract may have contributed to the lessening of
goodwill, increased ill-will and increased the motivation to engage in
counterproductive behavior. This perception engendered behavior that was
antithetical to improved performance, improved service delivery, good fiscal
management and sound government resource allocation. Viewed in this
context implementing the MTPRS is not simply an income issue or budget
issue, it is also a governance issue.

30.     Some analysts postulate that within the fiscal constraint there is limit to
the amount of good governance that a low-income country can afford (Haque
and Aziz, 1998). Good governance, the level of effectiveness and efficiency
of the public sector, and by implication the degree of deviant work behavior,
then becomes an optimization problem. This begs the question as to how
could a government operating under severe budgetary stress such as
Tanzania achieve a higher, yet affordable level of governance. The GoT in its
implicit decision to improve salary levels in line with the MTPRS has indicated
a revealed preference for a low level of governance.

III.    Budget Analysis: Crossing the line between PE/OC
31.   Developing a suitable medium-term pay reform strategy will depend
upon how well the government is able to strike a balance between various
reform and expenditure objectives. Take PE and OC expenditures as an
example. On the one hand, paying workers better without providing them with
the necessary complementary inputs required to perform their task does little
to improve service delivery. On the other hand, providing low paid, low
motivated workers with more complementary inputs is likely to do little to
improve service delivery, as the inputs may be neglected, misused or pilfered.
Improvements in the incentive regime must be made early in the reform
process otherwise training, investments in working conditions and information
technology, and re-organization will only lead to brain drain or frustration.

32.    Evidence in other countries shows that when salary growth stagnates
there is a tendency for public services (and public servants) to use non-salary
benefits (allowances and fringe-benefit) drift as a means (compensating
measure) to increase total compensation. The current remuneration structure
for the Tanzania Public Service is one where remunerative allowances do not
feature within the remuneration structure most were consolidated/eliminated
during the 1996/97 pay reform exercise.

Budget Analysis to Facilitate Pay Reform                                    Final Report – 08 May 2004

33.      Many cadres within the Public Service are now advocating the
(re)introduction of various remunerative allowances. Currently the scope for
doing so within the MTPP is limited. However, it does not mean that avenues
for allowance drift and the de-rationalization of the compensation structure
which accompanies this drift is closed. There is scope for allowance drift
through the OC budget, where what would otherwise be considered duty-
facilitating allowances within OC can be exploited for remunerative purposes.
Where this does occur, an element of PE creeps into the OC budget and the
transparency and the rationality of the OC budget may be compromised. For
purposes of transparency in the budget process, it may be better for such
allowances (or the funds that facilitate their payment) be shifted from OC to
PE where they can be appropriately accounted and the compensation
structure rationalized.

34.    Table 2 presents data on total employment allowances for civil
servants for the period FY01/02 – FY03/04.6 These data indicate that there
has been a significant growth in employment allowances paid to civil servants.
Employment allowances are paid from the OC budget. This growth is far in
excess of the increase in OC budget, which would indicate that over the years
an increasing element of OC has become a de facto substitute for PE.

35.    Employment allowances have increased significantly and consistently
over the period, far out pacing the growth of the wage bill (Figure 1). While
the wage bill has grown in nominal terms by about 15% per annum, total
employment allowances have grown by an average of 53% per annum.7

            The ToR proposes that the analysis of budget trends be undertaken for a five-year period, to
assess the extent to which there has been a drift in the payment of allowances to compensate for slower
than expected growth in average public service salaries. However, the data does not allow for analysis
over the total period. A new computerized budgeting and accounting system was installed about three
years ago. Various budget items have been reclassified and recoded. A consistent and comparable data
set is available only for the period FY01/02 through FY03/04.
           As the difficulties in undertaking the present analysis indicates, data and information systems
in the area of personnel compensation remains weak. Acquire data on employment allowances proved
to be a major challenge, delaying to completion of the study. Such data/information is critical to
compensation and overall budget management. Basic data on employment allowances by MDA should
be provided to and assess by PSM on a regular basis. This need not be an expensive or time consuming

Budget Analysis to Facilitate Pay Reform                                       Final Report – 08 May 2004

Table 2: Total Employment Allowances for Civil Servants (MDAs & Regions),
FY01/02 - FY03/04
                                           Budget            Budget             Budget        Change      Change     Change
  Item                                    Allocated         Allocated         Allocated      FY01/02 -   FY02/03 -    over
  code           Description               FY01/02           FY02/03           FY03/04       FY02/03     FY03/04     period
 250301    Leave Travel                4,886,633,509      6,409,441,014     6,354,213,581      31%          -1%       30%
 250302    Moving Expenses             2,011,967,461      3,180,684,477     3,658,135,090      58%          15%       82%
 250303    Uniforms                      880,444,597      1,595,163,712     1,323,427,580      81%         -17%       50%
 250304    Laundry and Domestic           20,300,100         34,210,000      29,480,000        69%         -14%       45%
 250305    Servants
           Medical & Dental Refunds      801,189,310        733,431,233     1,491,222,787      -8%         103%       86%
           Hotel Accommodation in
 250306    Lieu of Quarters              279,698,601        290,850,114     274,188,000        4%          -6%        -2%
 250307    Expatriate Personnel          330,860,501      1,495,885,800     986,788,250       352%        -34%       198%
           Medical Practitioners
 250308    (Chinese Doctors)               21,450,100        26,898,000     1,079,542,800      25%        3913%      4933%
 250309    Housing                       448,379,060        574,246,131     2,843,936,813      28%         395%       534%
 250310    Ration Allowance            11,067,526,400    11,233,692,000    12,875,236,000      2%           15%        16%
 250311    Per Diems – Domestic        15,006,255,330    26,726,084,444    35,685,135,708      78%          34%       138%
 250312    Internship                     257,403,001       386,338,440      229,640,004       50%         -41%       -11%
 250313    Extra – Duty                 5,024,910,192     8,617,911,710    11,518,603,546      72%          34%       129%
 250314    Invigilators                    33,400,000       134,980,000       53,000,000      304%         -61%        59%
 250315    Foreign Service              6,103,987,900     6,729,599,264     7,148,865,200      10%           6%        17%
 250316    Outfit (Clothes)               43,400,500        147,457,230      330,905,848      240%         124%       662%
 250317    Training – Domestic          3,580,463,196     6,999,905,662    25,215,881,836      96%         260%       604%
 250318    Training – Foreign             614,622,569     3,816,169,486     3,790,714,068     521%          -1%       517%
 250319    Part-time Teaching              29,420,000        33,252,454      126,301,700       13%         280%       329%
 250320    Utilities                      648,808,796       952,930,078     1,746,218,135      47%          83%       169%
 250321    Constituency                   922,860,100     1,794,400,200     4,340,480,000      94%         142%       370%
           MP Personal Assistant's
 250322    Allowance                     132,859,400        156,103,020      177,308,380       17%         14%        33%
 250323    Sitting                       977,671,549      1,649,618,686     4,264,437,045      69%        159%       336%
 250324    Acting                         34,014,100         89,149,721      114,140,535      162%         28%       236%
           Compassionate Leave –            1,400,000      201,000,000        546,663
 250325    Foreign                                                                           14257%       -100%       -61%
 250326    Release From Institutions    1,400,000,000     1,500,000,000     1,989,800,000       7%          33%        42%
 250327    Night Duty                        5,000,000       45,154,000       63,992,100      803%          42%      1180%
 250328    Instructor                    263,566,100        245,729,700      241,079,700       -7%          -2%        -9%
 250329    Professional                  165,306,000        309,643,800      160,539,700       87%         -48%        -3%
 250330    Research                        31,000,000        68,000,000      144,600,002      119%         113%       366%
 250331    Enrolment                     853,833,100        512,976,000      561,000,000      -40%           9%       -34%
 250332    Risk Allowance                     -                2,980,000      53,162,900         -        1684%          -
 250333    Entertainment Allowance          -                          -          -             -            -         -
 250334    Per Diems – Foreign          912,569,915       2,839,557,368     6,261,900,608     211%        121%       586%
 250335    Passages                      60,000,000         328,213,930      321,600,000      447%         -2%       436%
 250336    Responsibility Allowance                         228,916,000      363,334,000        -          59%         -
           Total Amount               57,851,201,387     90,090,573,674    135,819,358,579     56%         51%       135%
Source: Accountant General‟s Department

Budget Analysis to Facilitate Pay Reform                                       Final Report – 08 May 2004



                                                                           Total Employment Allowance
                                                                           Total Wage Bill
  Fiscal Year



                          0             50           100             150                           200      250

36.     The total increase in employment allowances over the three-year
period is 135%. Expenditures on total employment allowances increased
relative to that for the wage bill, equalling:

                     a. 16.3% of the total wage bill size in FY2001/02;

                     b. 21.9% of the total wage bill size in FY2002/03; and

                     c. 29.3% of the total wage bill size in FY2003/04.

37.    If these items had maintained their size relative to the wage bill for
FY01/02 (16.3%) over the period, TSh60.35 billion could have been added
to the wage bill without affecting the size of the overall budget. That is,
by shifting these funds from OC to PE, the wage bill could have been
13.0% higher in FY03/04 within the same overall budget ceiling.

38.    The items that can most easily be identified as likely serving as
substitutes for salary enhancement are:

                     a.       Housing allowance, grew 534%;

                     b.       Per diems – Domestic, grew by 138%;

                     c.       Extra Duty allowance, grew by 129%;

Budget Analysis to Facilitate Pay Reform                               Final Report – 08 May 2004

        d.    Utilities allowance, grew by 169%;

        e.    Constituency allowance, grew by 370%;

        f.    Sitting allowance, grew by 336%;

        g.    Acting allowance, grew by 236%;

        h.    Night duty allowance, grew by 1180%;

        i.    Research allowance (military), grew by 366%; and

        j.    Per diems – Foreign, grew by 586%.

39.      These items, combined with Risk allowance and Responsibility
allowance which were (re)introduced in FY02/03, account for TSh67.255
billion in the current FY03/04. Their share of total employment allowances has
increased consistently and sufficiently over the period from:

        a. 41% in FY01/02;

        b. 48% in FY02/03; and

        c. 50% in FY03/04.

40.    The Tanzania situation is not unique. Similar tendencies have been
observed in the case of the Kenya public service. Although there had not
been salary awards over a period of years, in Kenya the CG wage bill
continued to grow, primarily as a result of growth of: Extra Duty Allowances,
Fees, Commissions and Honoraria. The 13 MDAs with the fastest rate of
growth of such expenditures had an average growth of roughly 67.6% per
annum.     Hence, these expenditures served as a major source of
compensation creep.8

Sitting Allowances and Honoraria

41.    While the excessive growth in employment allowances is significant, it
is only one avenue by which OC may be used to provide compensation
enhancement.        Among the other work-related alternative sources of
compensation accessible to professional, technical, and managerial personnel
in the public service are:

        a. Payments (fees) for undertaking various short-term assignments/
           consultancies on the job;

        b. Salary supplements for undertaking various assignments on donor-
           funded projections and programs; and

         For discussion of allowance creep and wage bill management in Kenya, see: Valentine and
Wheeler (2003).

Budget Analysis to Facilitate Pay Reform                               Final Report – 08 May 2004

        c. Sitting allowances (attendance fees) or per diems paid to public
           servants to participate in forums (donor-funded workshops,
           seminars, conferences and round-table discussions); and

        d. Allowances for travel outside of the duty-stations.

42.   Various cross-country studies indicate that where salary growth
stagnate sitting allowances, attendance fees and honoraria begin to feature
prominently in public servants income-maintenance strategies, influencing
work patterns and time spent regular office duties.9

43.    The numerous workshops, seminars and conferences, may have as
their overall objective to enhance capacity building, and undoubtedly much
capacity building does take place. However, such activities are viewed by
many personnel as merely providing a means for indirect salary supplements,
taking the best personnel and senior management away from their regular
duties, with little positive benefits to public service performance.

44.     Concerns were been expressed by various government officials and
representatives of the donor community that workshops and conferences
have reached the saturation period. Frequent workshop participants have
little opportunity to apply the lessons learned at the last workshop before
attending the next one. The potential benefits to capacity building are being
loss because of the frequency of the workshops.

45.     Reliable data on sitting allowances and the amount that they contribute
to raising public servants compensation are difficult to obtain. However, when
sitting allowances are combined with employment allowances, the share of
allowances in total monetary compensation is likely to rise significantly than
employment allowances would imply alone.

46.   It should, however, be recognized that not all potential work-related
sources of compensation have a positive affect on work performance. There
may be unintended consequences of some resulting from some elements of
the compensation system.

47.    While compensating for the slow pace of salary enhancement, work-
related to alternative sources of compensation may induce and reward
counterproductive behavior by given rise to:

        a. Decision-making by committees rather than responsible individuals;

        b. A proliferation of committee (and workshop) meetings;
         Evidence across a long number of countries indicates that payments of sitting (attendance)
allowances and honorariums feature prominently in income-maintenance strategies in areas of the
developing world as diverse as Indonesia (Cherikovsky and Bayulken, 1995), Bolivia (World Bank,
1999), Cameroon (Israr et al, 2000), Uganda (McPake et al, 1999), Tanzania (Schiavo-Campo, 1998;
Valentine, 2000b), and the francophone countries of the Central Africa Region (Roenen et al, 1997).

Budget Analysis to Facilitate Pay Reform                  Final Report – 08 May 2004

        c. Rewarding personnel for spending more time away from their
           regular public service duties;

        d. Greater incentives for personnel to:

            i.   be away from their desks;

            ii. favor assignments on donor-funded projects and programs; and

            iii. delayed decision-making.

48.   The result may have been a lessening of attention to service delivery
and other matters.

49.    The major challenge of the pay reform is to ensure that improved pay
results in improved performance and changes in employee motivation and
behavior. If this does not occur, enhanced salaries will simply mean higher
cost of delivering inadequate and poor quality services. In addition to this,
GoT would find itself saddled with the costs of higher levels of allowance
creep/drift with no additional benefits. It is more desirable from the standpoint
of transparency and accountability and more effective and efficient from the
standpoint of efforts to improve work performance and service delivery for the
Government to enhance public servants‟ salaries directly rather than to
provide inducements from them to find means within the existing expenditure
framework to devise their own compensation enhancement strategies.

50.   The discussion in this section points towards two distortions in the
budgeting system:

        a. OC is not transparent, as there are lots of elements of PE hidden in
           its structure. Incorporating PE aspects of OC into the wage bill will
           improve the transparency of the budget and budget spending; and

        b. As there are elements within the OC budget that are being exploited
           for purposes of indirectly providing compensation enhance, the
           wage bill is currently under estimated.

51.    Addressing these distortions will improve the transparency and
rationality of public expenditures and improve the attention to improved work
performance and service delivery.

IV.     Central Government Wage Bill Relative to other Macroeconomic

52.     There are three measures of the relative size of the wage bill:

        a. The wage bill-to-GDP ratio;

        b. The wage bill-to-recurrent expenditure ratio; and

Budget Analysis to Facilitate Pay Reform                 Final Report – 08 May 2004

        c. The wage bill-to-domestic revenue ratio.

53.    The most often used measure is the wage bill-to-GDP ratio, particularly
for purposes of international comparisons of wage bill size and in reference to
the determination of wage bill ceilings under structural adjustment programs.

54.    A priori, there is no precise optimal relationship between the wage bill
and GDP for a particular country, as required for fiscal balance and
sustainable growth. These ratios may vary significantly between countries
and between geographical regions in SSA. However, from the standpoint of
appropriate public expenditure management and fiscal balance, it is generally
seen as prudent to have a wage bill-to-GDP ratio less than 5.5 percent and a
wage bill-to-recurrent expenditure and/or wage bill-to-domestic revenue ratio
of less than 33 percent.

55.    Figures 2 – 4 below present data on public service wage bills relative to
other macroeconomic variables for eight sub-Saharan African countries that
are currently undertaking public service reforms, including Tanzania.

56.    With a wage bill-to-GDP ratio of 4.3% in FY2002/03, Tanzania is the
country with the lowest wage bill-to-GDP ratio in the survey and also is the
only country whose ratio was (and is) less than 5.2% (Figure 2). Before 1999,
Mozambique had a similar ratio to that of Tanzania.            However, the
Government of Mozambique, with the strong support of donors have made a
concerted effort to raise its wage bill-to-GDP ratio, in order to finance its
expansion of public service employment in basic social services to implement
its poverty reduction strategy program (PRSP) and to pay reform programs
(See Text Box 1).

Budget Analysis to Facilitate Pay Reform                                     Final Report – 08 May 2004


                                                       WB-to-GDP ratio





                       Ghana   Kenya   Malawi   Mozambique         Rwanda   Senegal   Tanzania    Zambia


                     Text Box 1. UNDP CALLS FOR HIGHER PUBLIC SECTOR WAGES
    Maputo, 29 Aug (AIM) - The Administrator of the United Nations Development Programme
    (UNDP), Mark Malloch Brown, on Wednesday warned that higher wages for state employees
    should be a necessary part of a public sector reform.
    But Malloch Brown pointed out "Our experience internationally is that such reforms involve
    public sector wages, if you are to combat corruption and increase professionalism".
       "Wages are a critical component of public sector administration", he stressed. "If you don't
    pay people enough for a decent life, you are inviting corruption".
          (AIM) pf/ (869)

57.     Figure 3 indicate that Tanzania‟s wage bill-to-domestic revenue ratio is
more or less average for the group of countries surveyed. The major issue
here with respect to Tanzania is its low domestic revenue effort in comparison
to the other countries in the survey. With a domestic revenue-to-GDP ratio of
slightly over 12%, Tanzania‟s own revenue mobilization effort is far lower than
those of many of the other countries, with the exception of Mozambique and
Rwanda, two post-conflict countries.

Budget Analysis to Facilitate Pay Reform                                                                 Final Report – 08 May 2004


                                                                                       WB-to-Domestic Revenue ratio






                           Ghana       Kenya      Malawi     Mozambique             Rwanda            Senegal         Tanzania   Zambia

58.    With Tanzania‟s low domestic revenue effort, its wage bill-to-domestic
revenue ratio (which averages about 34%) is comparable to those for Malawi
and Zambia which have both high wage bill-to-GDP and domestic revenue-to-
GDP ratios. Still, its wage bill-to-recurrent expenditure ratio is among the
lowest of the countries in the comparative study (see Figure 4). This
information points towards scope for a slight increase in the wage bill in the
short term, even with the low domestic revenue effort. In the medium to long
term, maintaining fiscal prudence will require that the domestic revenue effort
be enhanced to finance both higher levels of OC as well as the increases in
the wage bill emanating from expansion of employment in the social sector
and pay reform.

59.    In the case of Tanzania, public service employment in the social
sectors is being expanded to implement the PRSP. However, as a result of
“the government‟s (overly) cautious wage policy”, 10 wage bill-to-GDP ratio has
changed very little. It is doubtful whether modest increases in the wage bill
planned will adequately finance both the planned increases in personnel in
the social sector and the agreed MTPRS.

                          For discussion, see: International Monetary Fund, 2001.

Budget Analysis to Facilitate Pay Reform                                                       Final Report – 08 May 2004


                                                      WB-to-Recurrent Expenditure ratio









                     Ghana   Kenya   Malawi   Mozambique             Rwanda          Senegal         Tanzania   Zambia

60.     A more appropriate wage bill-to-GDP ratio for Tanzania at this stage is
likely to be about 5.2% or there about. This ratio is developed in detail in a
recent study of MTPRS implementation issues (Valentine, 2003a). The most
recent MTEF publication also adopts is ratio as a target for the FY2006/07
(United Republic of Tanzania, 2003). While a gradual approach to raising the
ratio is prudent to limit the (potential) fiscal shock of a significant wage bill
increase, given past experience and the significant difference between actual
and target salary levels, a more rapid increase in the ratio would like restore
stakeholders‟ confidence in government‟s commitment to the MTPRS.

61.     Other things being equal, an increase in the wage bill will lead to an
increase in all three measures of the size of the wage bill. However, if this
were to occur today, it would not imply that the wage bill has become less
affordable or sustainable. If the TSh60.35 billion excessive increase in total
employment allowances that occurred between FY01/02 and FY03/04 were
reallocated from OC to the wage bill, the wage bill for the current fiscal year
would rise from TSh464.00 billion to TSh524.35 billion. The wage bill-to-GDP
ratio would 5.0%, which is line with proposal made in revising the MTPRS
salary targets (Valentine, 2003a) and is not much out of line with proposals
made in the MTEF (United Republic of Tanzania, 2002). Such a reallocation
of budget expenditures would improve the transparency and rationality of the
budget and (likely) have a positive impact on service delivery, if it were
appropriately targeted enhancing salary levels of qualified skilled professional,
technical and managerial personnel that the public service has a difficult time
attracting, retaining, and adequately motivating.

Budget Analysis to Facilitate Pay Reform                Final Report – 08 May 2004

V.      The SASE and capacity building of qualified professional,
        technical and managerial cadres

62.    The medium term pay targets were adopted as part of the MTPRS,
with a focus on enhancing the compensation of middle and upper-middle level
personnel. This was clearly a major step in the right direction, as compressed
salary differentials were seen to aggravate the public service‟s attraction,
retention and capacity building problems.        However, phasing-in salary
enhancement over time, which was required given the limited resource
envelop of the government had a major drawback, by necessity would yield a
gradual approach to pay reforms, perhaps too gradual to bring about the
changes in work behavior required to lead to significant improvements in work
performance, service delivery and capacity building desired. Extra budgetary
support would lend impetus to the GoT‟s pay reform effort by accelerating the
pace of salary enhancement and jump-starting the PSR effort.

63.     The SASE scheme was designed, developed and operationalize to
overcome the dilemma of trying to significantly improve the public service
attraction, retention, and motivation capacity in an era of steady but gradual
salary enhancement. The scheme, which was expected to serve as a
cornerstone of the MTPRS, targets salary enhancement towards key
professional, technical and managerial personnel whose efforts are critical to
the improvement of service delivery, management of the reform efforts, and
the production of strategic government outputs.

64.   A SASE scheme is selective in two ways. First, accelerated salary
enhancement is targeted at posts that are likely to have the greatest impact
on service delivery, the reform effort and the strategic outputs of government.
And second, the scheme is phased-in, starting with public service ministries,
departments, and agencies that have:

        a. Leading roles in change management and potential impact on the
           socio-economic well-being of the average citizen; and

        b. Well advanced formulated strategic plans and are poised for
           implementation of their respective reform programs.

65.     Why selectivity?    There were too major reasons for proposing
accelerated salary enhancement on a selective basis. First, with the
Government being under severe budgetary stress, there is little scope for
significant across-the-board salary enhancement in the short- to medium-
term. Therefore, in the short- to medium-term pay reform would contribute to
raising public service performance by undertaking salary enhancement on a
selective basis. The consensus view among stakeholders was that salary
enhancement must be targeted towards those posts that will afford a greater
impact on performance improvement and capacity building.

66.    Second, by linking qualification of to the restructuring of MDAs and the
continuing provision of salary supplements for the incumbent in a SASE-

Budget Analysis to Facilitate Pay Reform                   Final Report – 08 May 2004

scheme funded position to the outcome of objective staff assessments, the
SASE scheme beneficiary selection process would be consistent with and
help to facilitate the strategy for performance improvement management
adopted by the public service.

67.    It was initially envisaged that the SASE scheme would be
operationalizing a was as follows:

        a. The government would adopt and adhere to a medium term pay
           target, which would (initially) cover the period FY00/01 – FY04/05.

        b. A fund would be created to provide the government with extra
           budgetary support to supplement salaries of SASE scheme
           beneficiaries over the course of the medium-term.

        c.    All personnel of a particular job grade and step/increment, whether
             SASE-scheme beneficiaries or non-beneficiaries, would be entitled
             to the same base salary, as is currently applicable. The base salary
             for a given fiscal year will be set out in the MTPRS. The difference
             in compensation between SASE scheme beneficiaries and non-
             beneficiaries would be the positions that they hold and whether
             salary supplementation is afforded to these positions. Within the
             funding limits, positions deemed key/critical may be included under
             the SASE scheme and afforded salary supplementation. The
             amount of the supplement in a particular fiscal year will be the
             difference between the target monthly salary and the actual monthly
             base salary for the given fiscal year.

        d. It is the position that would be determined as key and thus selected
           for supplementation, not the incumbent in the position. Hence, the
           salary supplement would be attached to the position, not to its

        e. The SASE scheme target salary scales would serve as the
           benchmark for the payment of any salary supplements, as this will
           allow the government to internalize such supplementation as and
           when extra budgetary support is phased out.

        f. Each fiscal year, as the government makes salary adjustments in
           line with MTRP, the gap between actual government pay levels and
           target pay levels would narrow, thus, reducing the need for extra
           financial commitment, as the government‟s ability to pay
           competitive compensation out of the general wage bill rises.

        g. Donors would agree to phase-out LCC arrangements. The SASE
           scheme target salary scales would serve as the benchmark for the
           payment of any salary supplements, as this would allow the
           government to internalize such supplementation as and when donor
           support is phased out.

Budget Analysis to Facilitate Pay Reform                                 Final Report – 08 May 2004

68.    The majority of stakeholders consulted during the process of designing
and developing the scheme at the CG level and discussions of the potential
benefits of extending it to the Local Government (LG) level, were of the view
that inducing the SASE scheme at the would serve as a significant motivator
towards improved work performance and service delivery, since better pay
would enable them to spend more time and pay more attention to their work. 11
Of course, the selection of beneficiaries of the scheme would have to be
perceived as fair and objective, otherwise implementation would be

Major issues in SASE implementation

69.    There are four major issues of SASE implementation that require
attention at this juncture:

        a. The divergence between the medium term pay targets and annual
           pay levels may have contributed to the reduced willingness of
           donors to provide financial support the SASE scheme as:

             i.   there is no clear commitment of the CG to its own scheme;

             ii. without the convergence of the actual salary levels with the
                 target salary levels, there is no clear time horizon for phasing
                 out the salary supplement and no exit strategy for donor funding;

             iii. with the gap between the actual salary levels and target salary
                  levels persisting the cost of funding the scheme will be
                  increased significantly.

        b. There was slow pace of rolling-out the SASE scheme, where only
           four CG ministries, independent departments and agencies (MDAs)
           were benefiting from the scheme after the four years after the
           scheme‟s inception, as compared to the 11 MDAs that were
           targeted for SASE scheme support during the first phase of roll-out.
           Major problems with the slow pace of roll-out being:

             i.   By the time that all MDAs are in a position to benefit from the
                  scheme, its salary supplements would likely be inadequate to
                  induce any change in work behaviour on the part of civil
                  servants and not be worth the transaction costs (the effort
                  required to process all the proposed beneficiaries) on the part of
                  senior personnel within the MDAs and at the Inter-Ministerial
                  Working Group (IMWG);

             ii. The SASE scheme was conceived as an approach to
                 rationalizing the current ad hoc LCC arrangement and
           For studies on explorations of the potential implications of extending the SASE scheme to
the Local Government level, see: Valentine (2001a) and Valentine (2003b).

Budget Analysis to Facilitate Pay Reform                 Final Report – 08 May 2004

                 overcoming the attraction, retention and motivation problems
                 related to the more gradual approach of pay reform. The very
                 gradual phasing-in of SASE means that many donors may still
                 feel obliged to pay LCC until all MDAs are under SASE. In this
                 case the shift of funds from the payment of LCC to SASE
                 scheme support would be unlikely to occur.

        c. The SASE-scheme funding arrangement and the possible
           temptation to projectise the scheme, as Individual donor
           organizations were providing funding for SASE scheme support to
           individual MDAs rather than pooling funds, raising the possibility of
           replacing LCC associated with projects and programmes in MDAs
           with projectising the whole MDA through SASE.

        d. Fourth is the time horizon for implementing the scheme, where the
           original proposal for the phasing-out SASE salary supplementation
           over the period FY00/01 – FY04/05, because unrealistic as a result
           of (a) and (b) above.

Need for an objective assessment of the Scheme

70.    In light of the current difficulties in implementing the SASE scheme, the
Government has made a decision to restrict its coverage to those MDAs
currently benefiting from the scheme pending the outcome of an assessment
of the scheme‟s impact on performance. Such an assessment is well
warranted. As pay comprises the largest element of recurrent expenditures,
there are clear reasons, for the need for pay to drive better performance and
service delivery. Operating under a severe financial constraint and given the
potential sensitivities of a selective approach, the Government is keen to
ensure that increases in public sector spending are not swallowed up by
increases in public service pay – unless these have proven links to
improvements in work performance and/or improved service delivery and

71.     The scheme‟s implementation problems are attributable, in part, to the
lack of dialogue and coordination at the conceptualization stage. The SASE
scheme went from a consultancy report to a government policy document in
record time.      The consultancy report was submitted to the relevant
government department on 7th February 2000. The report was transformed
into a government policy document and in circulation within a week. The
document is dated 15th February 2000. It is doubtful whether appropriate
consultations with MDAs and relevant stakeholders transpired during this
period. Hence, the current lack of consensus on the role that SASE within the
context of the overall implementation of the MTPRS. While the major of
stakeholders perceive the scheme as an essential element of the MTPRS,
others see it as a side/separate issue.

Budget Analysis to Facilitate Pay Reform                                   Final Report – 08 May 2004

72.     The proposed assessment would allow the Government to learn for the
actual experience of implementing SASE and have a clearer understanding of
its likely impact on public service performance and service delivery. This is
consistent with the original proposal for the operationalization of the scheme
where it:

         a. Recognized that there was a large element of learning-by-doing in
            implementing such schemes; and

         b. Proposed that the scheme be piloted before being extended to the
            public service as a whole.

73.      In the view of the SASE scheme operationalization study report:

         “Adopting the pilot program approach would allow more
         learning-by-doing and provide scope for some experimentation
         to learn what works and does not before the scheme is
         extended to other MDAs.”12

74.   The outcome of a SASE scheme assessment may prove useful in fine-
tuning the operational aspects of the scheme as well as broadening the
consensus on the applicability of the scheme at both the CG and LG levels.

75.    Still even while SASE-scheme implementation remains limited to
current beneficiary MDAs, the lack of adherence to the annual salary
adjustment consistent with achieving the targets remains a major issue to be

VI.      MTPRS and the Determination of the Annual Wage Bill Ceiling

76.    As stated earlier the significant and growing divergence between actual
salary levels and medium term target salary levels raises issues about the
Government‟s commitment to the MTPRS and/or the implementation process.
As the wage bill ceiling poses a major constraint on the capacity to implement
pay reform, the current wage-bill posture within the MTEF and the wage-bill
determination process are major policy issues.

77.   While the major focus has been on the adequacy of the wage bill for
purposes of MTPRS implementation, the issue is two-fold:

         a. The global amount of the wage-bill allocation; and

         b. Optimizing the use of a given wage bill to obtain the desired level of
            salary enhancement and MTPRS implementation.

           For discussions on the need to pilot the SASE scheme before its roll-out across the entire
public service, see: Valentine 2000a.

Budget Analysis to Facilitate Pay Reform                 Final Report – 08 May 2004

78.     Viewing the issue in this manner allows us to considers scope of
improvement in the implementation process among the various actors with
portfolio responsibility over the wage bill and hence, who impact on MTPRS
implementation. The portfolio responsibility for addressing wage bill issues is
divided between Ministry of Finance (MoF) and Department of Public Service
Management (PSM):

        a. MoF has portfolio responsibility for preparing the MTEF, setting
           wage bill ceilings which are consistent with the fundamentals of the
           fiscal framework, and administering the budget.

        b. PSM bears responsibility on issues related to the size and grade
           distribution of the public service and the salary structure. These
           combined yields a certain wage-bill outcome.

MoF, the MTEF and the adequacy of Wage Bill allocations

79.     Given that the WB represents the largest non-debt service related
element of GRE, a well-functioning MTEF should take into consideration the
pay reform priorities in terms of proposed salary and public service
employment adjustments.         Progress of sorts have been made in this
direction, after years of being urged on the need to address issues of pay
reform within the context of the MTEF, over the past two years pay reform has
been identified as one of the major policy objectives.

80.   However, pay-reform priorities appear to be given little consideration in
the determination of the annual wage bill ceiling and nor does the MTEF
makes clear reference to the resource required of implementing the MTPRS.

81.       MoF sets a global wage bill figure that fits within the macroeconomic
framework, irrespective of MTPRS for that salary trajectory and other pay-
reform priorities for that year. Given the overly cautious wage-bill posture,
even by IMF standards, where there has been a reluctance to raise the wage
bill in line with the agreed MTPRS and target salary trajectory. As indicated in
Section II, whereas the wage bill-to-GDP ratio consistent with the MTPRS
implementation is 4.9% for the current fiscal year, the actual ratio is 4.4%.
Hence, from the perspective of MTPRS implementation, the wage bill
allocation is inadequate. As discussed in Section III, the under-funding of the
wage bill can be addressed, at least in part, by reallocating expenditures from
total employment allowances within the OC budget to the PE budget, without
negatively affecting work performance or service delivery.

PSM, Establishment Controls and the Optimization Challenge

82.    PSM has the portfolio responsibility to determine the optimal use the
wage bill allocation. Part of the optimization challenge is to decide whether to
use the wage bill allocation for salary enhancement (and MTPRS
implementation) or to finance an increase in the establishment. While
challenging the MoF to increase the wage bill allocation to facilitate MTPRS

Budget Analysis to Facilitate Pay Reform                 Final Report – 08 May 2004

implementation, the choices made by PSM in addressing the optimization
challenge don‟t necessary reflect the priority that it argues should be accorded
to MTPRS implementation. In its revealed preference in the pay-employment
trade-off, in recent years PSM has shown a willingness to trade-off real wage
levels against higher levels of public service employment, or at least has
loosened the reigns on establishment controls in a manner that in inconsistent
with MTPRS.

83.    Perhaps loosening the establishment-control reigns was necessary
after many years instituting an employment freeze. However, there should be
an explicit recognition that given the limited wage bill allocation, increased
employment comes at the expense of the desired MTPRS implementation. At
present, employment expansion is winning out over pay enhancement.

84.    The employment expansion is in part the result of a conscious policy
decision made by the GoT when it adopted the PEDP, which lead to the
expansion of the system primary school and an increase in the number of
teaching personnel. As the teaching service comprises the largest share of
public service personnel, implementing the PEDP has and will have a more
impact on employment levels.

85.      However, the most rapid growth of employment has occurred in non-
priority areas and without planning. For example, the number of auxiliary
personnel (TGOS grade personnel) has increased sharply over the past three
years: from 37,273 in July 2001 to 40,358 in February 2004. While the
number of auxiliary personnel has increased by 8.3% increase, the total
growth in the number of personnel paid through the CG payroll has been 5%.
Hence, the employment trends do not necessarily reflect government policy
priorities or MTPRS implementation requirements.

86.   Closer adherence to the MTPRS requires that PSM accept its portfolio
responsibilities and better control employment growth.

The Way Forward

87.    The way forward in strengthening the MTPRS implementation process
requires a two-track approach:

        a. Strengthening the policy process and dialogue between the PSM
           and MoF; and

        b. Establishing an independent advisory body to oversee pay reform
           implementation in the broader public service.

Strengthening the Policy Process

88.     Consistent implementation of the MTPRS require that:

Budget Analysis to Facilitate Pay Reform                Final Report – 08 May 2004

        a. A consensus should be sought between PSM and MoF on the
           appropriate salary adjustments on a year-to-year basis in line with
           implementing the MTPRS within the context of the MTEF.

        b. The pay-reform priorities to be achieved over the MTEF period be
           clearly identified and articulated to avoid a major disconnect
           between the MTEF and the MTPP and the allocation of resources in
           the Budget.

        c. Annual reviews of the medium term pay targets and the
           corresponding wage bill ceilings to be reviewed as part of the MTEF
           process, with the intent of facilitating the implementation of the

89.    With respect to pay reform, the policy process itself appears to need
strengthening. The major stakeholders on pay-reform matters, particularly the
PSM and MoF should thoroughly proposed employment levels, salary
adjustments and the underlying wage bill implications and reach consensus
on these.       Having achieved an agreement or at least a common
understanding, these should form the basis of input into the MTEF. Pay
reform priorities should be taken into consideration in the MTEF and in setting
guidelines for annual wage bill determination.

90.    The onus is not just on the MoF to seek more consultation with PSM in
the wage-bill determination process. PSM must objectively consider the trade-
off between establishment growth and salary enhancement in its proposals of
pay reform priorities and target salary levels. An analysis of pay-reform
implementation and pay-reform (and employment) priorities should be
undertaken by PSM on an annual basis. The report containing this analysis
should be discussed within the context of the public expenditure review
(PER). Just as in the case of other major cross-cutting issues, the results of
this analysis should be incorporated into the MTEF. The framework for
considering pay reform implementation and priorities within the MTEF
timeframe should be similar to that accorded sectoral issues or other cross-
cutting issues, where the structure is as follows:

        a. Review of major issues and implementation matters;

        b. Objectives and focus in the Medium Term;

        c. Priority activities for resource allocation – recognizing funding
           requirement, budgeted ceiling and likely ceiling as a percent of
           funding requirement.

91.     Currently the capability of the PSM to undertake such analysis to meet
the requirements of the PER and MTEF process are limited. Its capacity to
articulate its case and to interface with the MoF and the relevant PER Group
is constrained by the lack of an incumbent with adequate background and
experience. The capacity of the PSM will need to be strengthened to facilitate

Budget Analysis to Facilitate Pay Reform                    Final Report – 08 May 2004

dialogue between themselves and MoF. In addition to this, the program
coordination, monitoring and evaluation component of the PSRP needs to be
strengthened to enhance its capacity to address MTPRS implementation and
monitoring matters.

Establishment of an independent Compensation Commission

92.      Appropriate implementation of the MTPRS is so crucial to the
facilitation of public service reforms, improved performance and service
delivery that a more concerted effort is required to ensure its success. In
addition to this, it is important to recognize that the MTPRS was conceived as
a systematic and comprehensive approach to rationalize compensation of
personnel pay through the CG payroll. As compensation for the diverse
groups are drawn from the same Treasury (rather from PE, government
subventions, or otherwise) issues are raised regarding the relationship
between compensation, the extent of harmonization, and where their
conditions of service are regulated by different acts, how to ensure
consistency and harmonization.

93.      Different organizations (for example, executive agencies) may function
under different circumstances than do the core public service. As such they
may require different conditions of service and more flexibility in setting its
compensation structure. Still, while these agencies may require some
flexibility in determining their remuneration structures and remuneration mix,
efforts should be made by the subventing authority to strive to adhere to the
principle of equal compensation for equal work.

94.    One possible solution that have been adopted by a broad range of
countries to the challenges of ensuring effective implementation and
monitoring of government pay policy within the core public service and across
the broader public is to create an independent body with this mandate.

95.     The following are examples of countries within the region were
governments have established independent bodies to oversee implementation
of their respective pay policies and implementation strategies:

        a. In South Africa public service compensation matters are addressed
           by The South African Independent Commission on the
           Remuneration of Public Office Bearers;

        b. In Botswana The National Employment, Manpower and Incomes
           Council (NEMIC), bears responsibility for overseeing the
           implementation of the government pay policy, including:

            i.   Determining the scope for and magnitude of annual salary
                 adjustments taking into consideration:

                    the government‟s ability to afford the adjustment; and

Budget Analysis to Facilitate Pay Reform                       Final Report – 08 May 2004

                    proven increases in the cost-of-living;

            ii. Undertaking a major review of public service salaries and
                conditions of service very four or five years;

            iii. Assessing the competitiveness of public service compensation
                 to those in the private and state corporations.

        c. In Kenya and Malawi the governments have recently established
           independent public service remuneration boards to oversee the
           implementation of their respective medium term pay policies. In the
           case of Malawi, the board is comprised of seven eminent citizens
           who are recognized for their independent views.

96.    A Public Service Compensation Commission should be established by
the GoT to implement and monitor the performance of the MTPRS and to
harmonize compensation in the various institutions that comprise the public
service. The Commission should be responsible for the following:

        a. Oversee the implementation the MTPRS and to make
           recommendations to Government on compensation issues across
           the broader public service;

        b. Make proposals on harmonizing compensation across the public
           service for public institutions which draw funds from the
           Consolidated Fund, while taking the need for diverge organizations
           to have a degree of flexibility in determining of compensation levels
           and structure;

        c. Undertake regular reviews of remuneration and benefits in line with
           the MTPRS in the public service; and

        d. Coordination and monitoring the implementation of remuneration

97.    All remuneration proposals of the Commission would be required to fall
within budget ceilings and be approved by the GoT.

VII.    Conclusions and Implications of the Report

98.    The MTPP and MTPRS adopted by the GoT clear articulate the need
to systematically raise compensation for the core professional, technical and
management personnel in public service, which is recognized as a
prerequisite for improving public service performance and service delivery.
Pay reform is seen as central to the concerted effort to restore human and
motivational capital and goodwill (embodying honesty, integrity and
commitment to performance) within the public service.

Budget Analysis to Facilitate Pay Reform                  Final Report – 08 May 2004

99.     However, the major findings of the report point to a growing divergence
between the actual and target salary levels. The major explanatory factor for
this divergence lies in the reveal preferences of the Government, which are
contrary to the requirements for effective implement the MTPRS:

        a. Allowance proliferation is again favored over systematic and
           comprehensive salary adjustments, distorting the incentive structure
           and reducing transparency;

        b. Employment expansion proceeds at an rapid pace and in a manner
           which is contrary to Government policy and the principles set out in
           the MTPP;

        c. Increases in the wage-bill share for auxiliary personnel (through
           employment growth and salary compression) is favored over
           improved capacity to attract, retain and adequately motivate middle
           and upper-middle level professional, qualified technical and
           managerial personnel.

100. Current implementation practices are now consistent with and do not
augur well for improved work performance, service delivery and governance.

101. There appears to be a growing tolerance for ad hoc and non-
transparent measures, and lax establishment control which lead to unplanned
expansion of the budget, with the results being inconsistent with stated
government policy.

102. The wrong incentive structure is again gaining prevalence within the
public service. Only when the problems in the incentive regime are fully
addressed can we realistically expect the process towards the evolvement of
an effective, efficient and motivated public service. If implementation of the
MTPRS continues to stall, public servants will increasing perceive the MTPP
and MTPRS as defunct, and government once again as having breached its
social contract with the public service. Once this perception threshold is
reached, it would be hard to restore goodwill towards public service and
improved work performance. This would limit the country‟s efforts to improve:

        a. Governance;

        b. The production and distribution of essential public goods and

        c. The formulation and implementation of economic policy; and

        d. The management of public expenditures.

103. From the standpoint of improved public expenditure management,
improved transparency and accountability, and improved delivery and
accessibility of public services, it is more efficient to reallocate the funds use

Budget Analysis to Facilitate Pay Reform                      Final Report – 08 May 2004

to pay excessive employment allowances from the OC budget to the PE
budget and then adopt a systematic and comprehensive approach to salary

104. Getting MTPRS implementation back on track is crucial to the
(potential) success and sustainability of the public service reform,
performance improvement and service delivery efforts. The success and
sustainability of these efforts is essential to the realization poverty reduction
strategy objectives.


105.    The major recommendations of the report are as follows:

        a. Freeze the growth of total employment allowances;

        b. Rationalize current total employment allowances payments,
           reallocating the difference from OC to PE for the purpose of salary
           enhancement. (As indicated in paragraph 37 of this report, by
           shifting these funds from OC to PE, the wage bill could have
           been 13.0% higher in FY03/04 within the same overall budget

        c. The number of workshops should be drastically reduced to allow
           public servants to test the knowledge gained from recent workshops
           and apply lessons learned;

        d. To ensure that revised MTPRS salary targets are met:

            i.   CG wage bill should be adjusted on an annual basis in a manner
                 consistent with effective MTPRS implementation;

            ii. Employment growth should be restricted to agreed government

                    for improved accessibility and quality of public services in the
                     social sector;

                    to recruit young professional and qualified technical
                     personnel to address attrition and succession issues in the
                     public service.

        e. Undertake an assessment of SASE implementation in the 4 MDAs
           currently receiving SASE salary supplementation.

        f. Ensure that MTPRS implementation is:

            i.   taken into consideration in the MTEF process in a manner
                 similar to that for sector and other cross-cutting issue;

Budget Analysis to Facilitate Pay Reform               Final Report – 08 May 2004

            ii. monitored through the PER process.

        g. Establish a Public Service Compensation Commission to implement
           and monitor the performance of the MTPRS and to harmonize
           compensation in the various institutions that comprise the public

Budget Analysis to Facilitate Pay Reform                 Final Report – 08 May 2004


    1.      Background:

            Four years ago the Tanzania Government committed itself to
            undertake Pay Reform as part of the Public Service Reform
            Programme and adopted the Medium Term Pay Policy and
            Medium Term Pay targets. While efforts to improve pay levels
            have been taken, the pace of pay enhancements is not
            consistent with the envisioned targets. There is a need to
            accelerate that pace.

            In that regard the President’s Office, Public Service
            Management (PO:PSM) is exploring possible measures for
            accelerating the pace of pay increases. As part of the
            exploration it intends to undertake an analysis of the budget
            for the period up to 2003/2004 with a view to examining
            possible options for speeding up the pace of pay reform.

    2.      Objectives:

            It is envisaged that the budget analysis will enable the
            government to:

                        Assess the ability of the Public Service to increase
                         public servants’ salaries to levels that will enhance
                         motivation and productivity;

                        Ensure that the recommended salaries are
                         competitive for attracting and retaining competent
                         staff; and

                        Review the structure of the Government budget to
                         identify some scope for allocating more resources to
                         salary payments.

    3.      Scope of the Services (Tasks):

                   Review the distribution of budget resources between
                   PE/OC over the past five years to determine whether the
                   OC budget has any PE related elements, which could be
                   shifted to PE;

Budget Analysis to Facilitate Pay Reform                Final Report – 08 May 2004

                   Examine trends in the PE and OC expenditure ratios in
                   relation to GDP over the five years;
                   Following the analysis of the above trends provide
                   options and recommendations on appropriate wage
                   bill/GDP ratios in the medium term;
                   Review the existing arrangements and mechanisms for
                   decision-making on Yearly Salary adjustments and
                   recommend appropriate adjustments with a view to
                   ensure the timeframes are aligned to the budget decision
                   making process; and
                   Focus on how the Selected Accelerated Salary
                   Enhancement (SASE) can best play its role as a
                   temporary mechanism to attract and retain key
                   personnel in MDAs while government has yet to reach
                   the Medium Term Pay Targets, without creating too
                   much distortion in the pay structure.

    4.      Expected Outputs:

                   Options and recommendations on appropriate Wage
                   Bill/GDP ratio in medium term.
                   Recommended suitable annual salary adjustment that
                   are competitive to attract and retain competitive staff
                   with the view to ensure the timeframes are aligned to
                   the budget decision making process.

    5.      Input:

            The Public Service Management (PSM) will provide the
            following documents:

                   Medium term pay policy document
                   Budget framework document
                   Wage bill/GDP ratios for the past five years
                   Personnel data.

    6.      Reporting Requirement:
                   The assignment will be carried out within 15 working
                   days as follows:

Budget Analysis to Facilitate Pay Reform                 Final Report – 08 May 2004

                      Following contracting, the analysis of data should be
                       undertaken in March 2004 with the preliminary
                       results being available in draft form in five (5) hand
                       copies at the middle of March 2004.
                      The draft report will be reviewed by PSM for
                       Completeness and internal consistency by 18th
                       March, 2004; and
                      Final Report to be presented to Permanent Secretary
                       (PSM) on 24th March 2004. The report should be in
                       five (5) hard copies and an electronic copy.

    7.      Qualifications:

            The Consultant should have a post-graduate degree in
            Economics,       Business     administration       or Public
            Administration and a track record of undertaking
            consultancies/studies on Pay/Salary/Improvements in
            public service organizations for at least 7 years.

    8.      Institutional Framework:

            The consultant will report to the Permanent Secretary,
            President’s Office, Public Service Management (PO:PSM).
            However, She/He will work closely with the Director, Policy
            Development Division and the Chief Technical Adviser.
            She/He will need to consult closely with senior officials of the
            Ministry of Finance.

Budget Analysis to Facilitate Pay Reform                         Final Report – 08 May 2004

Appendix 2. List of Persons Interviewed
Name                         Post                                                Organizatio
Representatives of the Public Service                                            n
 Joseph Rugumyamheto              Permanent Secretary                           PO – PSM
 George Yambesi                   Director of Policy Analysis                   PO – PSM
 Emmanuel Mlay                    Assistant Director                            PO – PSM
 Agnes Meena                      Assistant Director                            PO – PSM
 Gelase Mutahaba                  Chief Technical Advisor                       PSRP
 Peniel M. Lyimo                  Permanent Secretary                           MoF
 R.H. Khijjah                     Deputy Permanent Secretary                    MoF
 K. Kamugisa                      Commissioner for Macroeconomic                MoF
                                  Analysis and Research
 N. Magambo                       Commissioner for the Budget                   MoF
Representatives of the Development Partners
 Jack Titsworth                   (Acting) Governance Adviser                   DFID, Tanzania
 Annabel Gerry                    Governance Adviser                            DFID, Tanzania
 Jytte Laursen                    Economic Advisor, Counsellor                  Royal Danish
                                  (Development)                                 Embassy

Budget Analysis to Facilitate Pay Reform                Final Report – 08 May 2004

Bevan, S. and L. Horner. “Reward and reform,” Public Service Unit at The
Work Foundation, Public Service Unit Briefing No. 2, June 2003.

Cherikovsky, D. and C. Bayulken. (1995) “A pay-for-performance system for
civil service doctors: the Indonesian experiment,” Social Science and
Medicine, 41(2): 155-161.

Chomitz, K.M., with G. Setiadi, A. Azwar, N. Ismail, and Widiyarti. (1998)
"What Do Doctors Want? - Developing Incentives for Doctors to Serve in
Indonesia's Rural and Remote Areas." Policy Research Working Paper, World
Bank, Development Research Group, Washington, D.C.

Civil Service Department. The President‟s Office. 2000. “Operationalisation of
the Selective Accelerated Salary Enhancement Scheme,” Public Service
Reform Programme, the United Republic of Tanzania, Dar es Salaam,

Greenberg, J. 1990. “Employee theft as a reaction to underpayment inequity:
The hidden cost of pay cuts,” Journal of Applied Psychology, 75, 561-568.

Haque, N.U. and J. Aziz. 1999. “The Quality of Governance: „Second-
Generation‟ Civil Service Reform in Africa,” Journal of African Economies,
8, AERC Supplement: 68-106.

International Monetary Fund. 2001. “Tanzania: Staff Report for the 2001
Article IV Consultation and Third Review Under the Poverty Reduction and
Growth Facility,” Prepared by the African Department, International Monetary
Fund, Washington, D.C.

Israr, S.M., O. Razum, V. Ndiforchu and M. Martiny. (2000) “Coping strategies
of health personnel during economic crisis: A case study from Cameroon,”
Tropical Medicine and International Health, 5(4): 288-292.

McPake, B., D. Asiimwe, F. Mwesigye, M. Ofumbi, L. Ortenblad, P.
Streefland, and A. Turinde. (1999) “Informal economic activities of public
health workers in Uganda: implications for quality and accessibility of care,”
Social Science and Medicine, 49(7): 849-865.

Mukherjee,     R.   2001.     “Total    Rewards      for  Civil  Servants,”

Niehoff, B.P. and R.J. Paul. 2000. “Causes of Employee theft and Strategies
that HR Managers can use for Prevention,” Human Resource Management,
39(1): 51 – 64.

Robinson, S.L. 1995. “A Typology of Deviant Workplace Behavior,” Academy
of Management Journal, 38(2): 555-572.

Budget Analysis to Facilitate Pay Reform               Final Report – 08 May 2004

Robinson, S.L. 1996. “Trust and Breach of the Psychological Contract,”
Administrative Science Quarterly, 41(4): 574-599.

Roenen, C., P. Ferrinho, M. van Dormael, M.C. Conceição, W. van
Lerberghe. (1997) “How African Doctors make ends meet,” Tropical
Medicine and International Health, 2(2): 127-135.

Schiavo-Campo, S. (1998) “Government Employment and Pay: the Global
and Regional evidence,” Public Administration and Development, 18(5):

Schiavo-Campo, S. (1996) “Reforming the Civil Service,” Finance &
Development, September: 10-13.

United Republic of Tanzania. 2002. Guidelines for the Preparation of the
Medium Term Plan and Budget Framework for 2003/04 – 2005/06, Draft,
issued by The President‟s Office, Planning and Privatization and the Ministry
of Finance, Dar es Salaam, Tanzania.

United Republic of Tanzania. 2003a. Volume II Estimates of Public
Expenditure Consolidated Fund Services (Section I) and Supply Votes
(Ministerial) for the year from 1st July, 2003 to 30th June, 2004, As Passed
by the National Assembly, Government Printer, Dar es Salaam, Tanzania.

United Republic of Tanzania. 2003b. Guidelines for the Preparation of the
Medium Term Plan and Budget Framework for 2004/05 – 2006/07, Part
One, issued by The President‟s Office, Planning and Privatization and the
Ministry of Finance, Dar es Salaam, Tanzania.

United Republic of Tanzania. Education Sector Development Programme.
2001. Primary Education Development Plan (2002-2006). Basic Education
Development Committee (BEDC), Dar es Salaam, Tanzania.

United Republic of Tanzania. President‟s Office, Civil Service Department.
1998. Public Service Medium Term Pay Policy, Dar es Salaam, Tanzania.
Valentine,   T.R.    1999.    Pay    Reform    Implementation Study:
Operationalising the Medium Term Public Service Pay Policy, Final
Report, presented to the Public Service Reform Programme, The United
Republic of Tanzania, Dar es Salaam, Tanzania.

Valentine, T.R. 2000a. Operationalisation of the Selective Accelerated
Salary Enhancement Scheme, Final Report, presented to the Public Service
Reform Programme, The United Republic of Tanzania, funded by the
Department for International Development (DFID) East Africa, Dar es Salaam,

Valentine, T.R. 2000b. A Pilot Scheme for Operationalising Selective
Accelerated Salary Enhancement (SASE) for the Health Sector: Study 1 -
Issues to be addressed, a report presented to The United Republic of

Budget Analysis to Facilitate Pay Reform              Final Report – 08 May 2004

Tanzania, Ministry of Health, Health Sector Programme Support, Dar es
Salaam, Tanzania.

Valentine, T.R. 2001a. Pay Reform within the Context of Local
Government Reform, Final Report, a report presented to The President‟s
Office - Regional Administration and Local Government, Local Government
Reform Programme, Dar es Salaam, Tanzania.

Valentine, T.R. 2001b. A Pilot Scheme for Operationalising Selective
Accelerated Salary Enhancement (SASE) for the Health Sector: Study 2
– Operational Details, a report presented to The United Republic of
Tanzania, Ministry of Health, Health Sector Programme Support, Dar es
Salaam, Tanzania.

Valentine, T.R. 2002. Revisiting and Revising Tanzania’s Medium Term
Pay Reform Strategy: Final Report, prepared for The United Republic of
Tanzania, Civil Service Department, Public Service Reform Programme, Dar
es Salaam, Tanzania.

Valentine, T.R. 2003a. A Medium Term Pay Reform Strategy for the New
Integrated Public Service Reform Program for Tanzania, prepared for The
United Republic of Tanzania, Civil Service Department, Public Service Reform
Programme, Dar es Salaam, Tanzania.

Valentine, T.R. 2003b. Operationalising A Selective Accelerated Salary
Enhancement Scheme At The Local Government Level: Final Report, a
report presented to The President‟s Office - Regional Administration and
Local Government, Local Government Reform Programme, Dar es Salaam,
Valentine, T.R. and J.R. Wheeler. 2003. Consultancy on Wage Bill
Management and Civil Service Performance Enhancement for the
Government of Kenya: Final Report, a report presented to The Office of the
President, Directorate of Personnel Management, Nairobi, Kenya.


To top