WORKING U I _~~~~~~~~~~~~ I Wi Se OWApm Counry Ecnom'am DQaarnent The World Bank may 1990 WPS 422 Civil Service Reform and the World Bank Barbara Nunberg and John Nellis Some argue that the complexity and uncerainty of civil savice refonn place the field outside .he Bank's comparative advan- tage. But a retreat from civil service management reform is tantamount to denying the crucial importance of government adminisu'ative capacity to implement economic and social pro- grams. A more realistic approach is to ty to learn, through tir and error, how to make such programs work better. The Pa.b'. R. u- EaiAni Affasaan duvu_ PRE Wagi P,pL to *&vwmhUw dni_n a _ d Po _- _ ewd do 8 o f uG o U aN urn ,a pmumsn_ papm cam Of 0r a dirn MMM .iin. aW n aN.M b Ua ad A WGI,. Thm rEN._ Msrnin. A = , thu _m_*>> _ T%p dm _m b_m to d Wm d amL _ u lmNd4 OuW un W = W ? |dB M Pelimy, Nemwh si EztemZ Afthf. pub" S_at mOaement wed PrIvate Seega Oevei*meni This paper - a product of the Public Sector Management and Private Sector Development Division. Country Economics Department - - is part of a larger effort in PRE to assess the Bank's accomplishments. problems, and pmspects in the field of public sector management An earlier version of the paper was presented at a December 1989 conference on "lstitutional Development and the World Bank." Copies are available fre from the World Bank, 1818 H Sret NW, Washington DC 20433. P¶ease contact Rose Malcolm. rmom N9-055, extensionJ7495 (50 pages with tables). After rviewing civil service refomn work in the twice he present amount of staff supervision Bank. Nunberg and Nellis reach certain conclu- and specialized expertse. sions: Such technical assistance ans require mor The impact of Bank pgrams to contain the time to prepar and implement than do infrm cost and size of civil services thrugh emer- strcu projecs. Tey often get shor, shrift gency reform of pay and employment policies because of their dependence on the scheduling has so far been negligible. Reform efforts have and requirements of structual adjustnent not been ambitious enough; meaningful change lending. On the other hand, without SALs, wiU require more foreful reform. Middle-range many civil service reforms in TALs have no measurs such as voluntary departure schemes teeth and eardy rtirment prgrams are useful but ar not a substitute for biting the bullet. Most Bank activities have concentrated on shor-term cost-containment measues. More Whedtr mot aggressive rforn is feasible emphUsis must be given to longer-tern manage- is parly a technical but mainly a political issuc ment issues if sustained improvement in govem- But in the few countries where reform has been ment administrative capacity is to take place. carried out the political costs were lower than More atention must be placed on devising a most govemmnents (pettaps even the donors) cohernt overarching strategy and detailed expected. This may have been partly because of tactics for civil service reform. the surprising capacity of labor markets to absorb surplus government workers and panly Some argue thau the complex and uncertain because of the skillful handling of reforn. nature of civil service reforn places the field outside the Bank's comparaEive advantage. Functional reviews and competency testing They argue that the Bank should confine itsclf to prmvide symbolic assurance that the reform helping define economicolly rtional policies, prcess will be fair. Retraining. redeployment. such as the appmpriate. affordable size of the credit, and public works programs for rdundat wage bill. employees ar symbolically and politically effective but have limited practical impact and But the Bank cannot identify the need to are administraively diffiCUlt remove X thousand surplus personnel and assume that the job of removing them will be The Bank should no longer encourage or carried out by the government or a bilateral support mechanisms such as topping up execu- donor. The ctallenge for the Bank is to design tive-level salaries for key govermment posts projects that have measurable short- run cost- unless such incentive schemes are part of an containment goals but realize them in the acticn strategy for long-term structural reform. context of a strategy to solve the fundamental management problems in the long run. Technical assistance loans (TALs) for civil service management should probably provides The PRE Wortking Papa Senes dissaninmaes Lhe rlndings o( work under way in the Bank's Policy. Resex&ch. and Extcmraa Affas Comptel. An obcuve of he x ts X n sswet . eLndinp out quickly. even if premntaions e le than fuliy potishL fLT findings.niuprtaui and conclusions n thm papm do not nenarily r-pucsnt ozTicial 8aitk poitcy. Producd u the PRE DiOninasbn Ceaw Civil Service Reform and the World Bank by Barbara Nunberg and John Nellis Table of Contents Introduction 1 Part A. Pay and Employment Reform: -Thort.Term Issues 2 1. The Problem: Analytic Approaches 2 2. Recent Bank Operational Experience in Civil Service 5 Reform 3. Cost Containment: Operational Approaches to Pay 6 ane dVmployment Reform - A Progress Report 4. Impact of Cost Containment Refornm 21 Part B: The Long View: Civil Service Management 23 1.. Components of Rationalist Reform Efforts 25 2. Implementation Issues and Lessons in Development X Management 3. Observations 3 Part C: Conclusions 34 Titbles 37 Annex 49 * With the contribution of Louis de Merode (AF1PS) and research assistance by Meera Balasubramanian. Ci snvc 0UDNE AND S WOtN MM SODU?ON The emphasis placed by the World Bank in recent years on the major overhaul of devloping country economies has accentuated the importance of adequate public sector admistrative capacity, espe:ially within the central core 4.& government, that is, the civil service. Incr:esingly, the ability of civil services to carry out the critical -- mach less the routine -- functions of government has been found severely deficient. Purthermore, the size and cost of many civil services have been de4med excessive and have thus become key targets of adjustment. In a real sense, this growing concern vith containing the size and improving the performance of the civil service signifies nothing less than the redimensioning' of the state, reflecting a fundamental shift in the direction of the BDank's own policies. The new visdom is to 'ma=nage lose -- but better. This paper surveys recent Dank experience in civil service reform, and begins to assese the progress made. To the extent that conclusions about 'right" and 'wrogg strttegies can be reasonably drawn, soo guidance is offered, with the caveat that it. is too early to offer definitive policy prescriptions in this area. Most operational and evn policy work on these issues is still at an incipliat stage, of a duration of a few years or less. The papor focuses on two separate but related aspects of civil service refom work in the Bank. One deals with the shcrter-term, emergency steps to refora public pay and employment policies. These reforms usually focus on measurec to contain tho cost and size of the civil service, mostly in the context of structural adjustment lending. The other set of reforms are those dealing vith longer-range civil service strengthening efforts, some of which may support variouc of the nearer-teor cost containment measures, but most of which are directed toward ongoing, sustained maagement improvements. Many of these reforms have been included in technical assistance projects -- either those that stand alone as 'development management' operations or those that constitute direct institutional support for specific actions taken in SALs. Many of the issues discussed in the papor have been the subject of earlier World Bank reviews. 1 Material from these analyses is synthesized and updated where data permit. In particular, some of the papor's concepts and much of its findings stem from research currently being conducted by the Africa Technical Department (Public Sector Management Division) for a study, *Public Sector Pay Sm. Geoffrey Lamb, Institutioral fofarn: Som Lesons from Structural Adjustment Lnding,' 1984; and. Berbera unberg, 'Publc Sctor Pay and EmpIoymnt Rdefors,e World Bnk Disusslon PoPw' No. 68. 1989; add eRvlw of Public Sector Management issues In SUructural Adjustment LondIng, World Bank PRE Working Par, fortrio.ling 1990. and Imploymnt Reform in Africa, forthcomling in 1990. The organization of the paper is as follows: Part A focuses on the short- term pay and eployment reform issues. Section I reviews some current analytic approaches to the goveraennt pay and employment problem. Section II discusses the BDnk'- experience with these reforms through its lending operations and tries to assess *heir .ovetall impact to date. Part B of the paper examines the longer- term civil service management issues, analyzing the maln types of development management initiatives undertaken through recent Bank operations. Conclusions are then drawn about this experience. The final section of the paper attempts to drew the findings of part A and 3 together and offers recovmendations for future work. Part As Pay and Employment Reform: Short-Term Issues 1. The Problem: Analytic Approaches Recent analytic vork on public pay and employment issues has provided a general understanding of the broad outline of tho problem. Governments in many developing countries are unable to manage and finance thfir civil services. Evidence suggests '... declining moral. and work effort on the part of government workers, problems in staffing large number* of skilled positions, a paucity of compleontary inputs ln producing government services, and increasing imbalance botween the demand for government serrices, and the provision of public goods. .2 Civil services in many developing countries are too large, too expensive and insufficiently productive; and civil servants, especially thoso in managerial positions, are poorly motivated. They are too large in the broad sense that in many states the public sector is over-extended: i.e., it possesses too many agencies and organizations, charged with too broad a span of responsibilities; and in the narrower sense that too many of these agencies employ numbers of people excessive to requirements (in the sense of people and skills needed to fulfill off.cially assigned economic and administrative tasks). They are too expensive in the sense that public sector wage bills constitute too high a percentage of total government revenues, and account for too high a percentage of CDP. The agents within civil services tend to be poorly motivaced in that remsneration scales for upper and middle managers are low, often extremely low, in comparison to those of roughly equivalant posts in the private and frequently the parastatal sector; and wages are often severely compressed --the highest paid earning low multiples of the wages of the lowest paid. Many civil servants are insuffic±entlw productive in tho sense that they do not fulfill the tasks assigned to them (they are ineffective), or they carry out their assignments partially, with great delays, at high cost (they are inefficient). 208vid Undaucr et al., Governeent Wage Pollcy In Africa: Some Findings and Polcy lssues, World Bank Research Observer, Vol. 3, No. I (January. 1988), p. 1. Particularly acute for Afrlca, such problems are also Increasingly present In countries as gwgraohicallY dispersed as Srl Lanka. Bolivla. Morocco, and Argentina, to name but a few. 3 nalysis within the Bank on these issues has tended to focus on four problems 3 Excessive oublic sector wage bills. Excessiveness is a subjective quality, one for which no clear measurement methodology exists, and no standardized approach is applied throughout the Bank. The extreme cases are easy to identify; i.e., where public sector wage bills become so large that they begin to 'crowd out' other high priority ilems. In general, the wage bill is measured against one of the follovings the overall government budget; overall government revenues; GDP; or total recurrent expenditures. The latter measure is perhaps the most comon, and the degree to which non-personnel recurreut expenditures (such as those for supplies and maintenance) diminish in relation to personnel expenditures is taken as a key signal that the wage bill is inappropriately high. SurDlus numbers of civil servants. Although obviousl? related to the size. of the wage bill, statements about the "appropriate, nwALer of civil servants are often derived ad hoc from the proportion of public servants to the overall population, to the moder sector labor force, or from somewhat casual observations of 'too many people sitting around doing nothing. Sometimes the rate of expansion of public employmnt is taken as a warning signal of surplus numbers in government. Clearly, the definition of what c=nstitutes too many' in the public sector must consider the relative role of the state in the economy, the level of a country,* e,velopmant, andthe relative importance of the state as a primary source of 9'.itictl patronage and social welfare. To identify a personnel surplus, most -orld Bank reform operations rely upon the indicator of insufficient operating budgets for supplies and maintenance. Erosion of Public Service Salaries. In many countries the level of re.l public sector salaries -- and particularly civil service salaries --h .s eroded substantially over time, The result is remuneration that is too low either to sustain lower echelon workers above the poverty level or to attract and retain needed skilled personnel. Real remunerations are significantly affected by the rate of inflation and the regularity and nature of salary adjustments. Declines if real wages have often been cushioned by elaborate allcwances or non-wage benefit structures that in some countries have become an increasingly important pairt of the overall compensation package. A pervasive irony afflicts many ccuntriess i.e., the overall salary bill is too high while wages are too low. This dilemma is the product of years of trade-offs, given fiscal constraints, in favor of hiring growing numbers of employees at diminishing salaries. Wage erosion is particularly a problem where there exists an alternative market for public servants, either through a better paying parastatal sector, a domestic private sector with higher salaries and benefits or an international market to whicn highly skilled public servants may migrate. Waie Compression. In many cases, compression of the ratio of top to bottom salaries has increasingly been identified as a serious constraint on governmentsI ability to attract and retain qualified personnel at the middle and higher 3Th/s section draws heavily on Nunberg. t989. .4 loevls. Wage compression is partially a function of expansive employment policies as discussed above, but in some countries it may also derive from regime preferences for agilitarian salarr structures. This, in turn, may hinge on the ideological character of the regime or the degree to vhich its political support is drawn from the lower socioeconomic strata. Many of the general principles suggested above are derived from the only available data base; i.e., recent pirical analysis of wago and employment practices in a group of African countries.4 The general analy-!- and sector work carried'out in the Bank (and olsewhere) as well as observatior. gathered through operational experience confirm these findings for a broader range of cases. These findings represent a significant advance in our understanding of this pheonamnon. Indeed, the low wage/high employment scenario in depressed aacroeconoeLcs turns on its head previously accepted wisdom of the government as high wage-giver. ComparrAtive data that would permit procise measurement ur quantification of these phenomena remain to be collected, however. In short, the issue is difficnIlt to analyze. The reasons for this are several. Certainly, a standardized view of the proper functional span and size of the state remains elusive, despite the sweeping re-omrgence of liberal notions of minimalist government. For example, even in societies in broad ideological agreoemnt as to the appropriate role of the state, there is no usarguable criterion that deteraines the appropriate number of ministries, or sub-divisions of ministries, a government should posess; and relative prices (wage scales) in differing labor markets might provide sound economic reasons for a wide variation in employment levels in organizationi having roughly quivalent objectives. furthermore, the range of solutions to this quotion may vary by country, region, or level of industrialization. On the other hand, there is no clear, linear path to development in this regards large states with high numbers of civil servants per capita are prevalent in both the developed and underdeveloped world. Vexing measurement questions are unresolved. The methodology for determining appropriate wage levels for private and public sectors continues to be debated. Ways to measure productivity for central government agencies have been the subject of a vast but inconclusive literature, Questions remain about the potential consequences of public sector retranchment programs. The relative capacity of different types of labor markets (rural-urban, forml-informal) to absorb redundant labor from the public sector is poorly understood. And analytic techniques for determining the direct and indiroct, upstream and dawnstream costs and benefits of various reduction in force scenarios are still very primitive. Political economy features of public pay and employmnt practices have only begun to be analyzed. The real as opposed to the perceived political risks of government retrenchment programs are neither known aor at present calculable. Which are the key bureaucratic and societal coalitions that figure in employment and pay reform? How much reform is politically feasible for a given regime to undertake under what conditions? 4Lunoauer et al.. Elt There is, in sum, a lack of agreement on guiding analytic principles, a lack of fim measurement criteria, and a great regional variation in the intensity of tue syndrow. Noretholess, a growing body of empirical evidence on this topic -- accumulated mainly through Bank lending experience -- provides sufficieht information for at least a preliminary examination. The section that follows highlights the more striking features of this experience and then draws general lessons abovt the imple entation of these typos of reforms. 2. Recent Bank Operational txoerience in Civil Serice Reform Since 1981, civil service reform has featured prominently in 61 Bank lending operations. Of these, 38 were Structural Adjustment Loans or Credits (CUL or SACs), and 2Z were Technical Assistance Loans or Credits. (Table 1 lists lank operations with civil service reform components). 5 Four operations have been in the hKINA region (Europe, Hiddle East and North Africa); 5 in Asia; 11 in Latiu Amrica and the Caribbesan and 4* in Africa. Thus, two out of three of recent Bank-supported civil service reform programs have been in Africa. The magnitude aad intensity of the issue is clearly greatest in that region; and efforts discussed below are mainly, though not exclusivoly, taking place in Africa. (Table 2 shows the regional distribution of lank operations with civil service reform components by lending instrument.) Activity is quite new in this field; prior to 1981, the Bank was only tangentia,lly involved with civil service or administrative reform -- except for the occ&sional report, such as on Thailand and Indonesis, and some lnvolvement by educational specialists on civll service training instLtutes -- and what was -done ws *inaily on the project, not the policy, level. It has been the Bank's entry into policy-based lending, with its amphassL on demand management and improvlng the performance of adjustinS governments, that has led to the lncreased involvement with civil servLce issues. The most important set of reasons for the inclusion of civil service resorm in zecent adjustment operations center, tround the issue of budgetary burden, the recognition of the heavy, fiscal b-, 'en posed by large wage blls. For extreme exampl*, in the early 1980s, following rapid increases in the numbers employed, many African countri,s saw their wage bLils rise to account for more than half of total government revenues. Such increases were tolerable (though never wise) as long aS total revenues increased. But with the prolonged economic crisis and its attendant stagnation of growth ln government revenues, the difficult-to-roduce wage bill begins, ts discussed in the preceding section. to Ocrowd out' other critical current expenditurest maintenance and depreciation, the provision of essential supplies and equipment. This led to the lncreasingly coaon situation of teachers without books, doctors lacking medicine, postal workers having no stamps to sell, etc. The recognitton that this was taking place led some governmeats to initiate their owe rationallzation measures. It led the Bank increaslngly to lnclude .ivll service reform in adjustment -S,vern/ other tyDes of policy-based lending operations, such as Reconstruction Inport Credits and Econosic Recovery Credits (R/Ca and EfACs), were counted among the SAJs. They are Iteoized In Table 1. operations. Thus, the primary reason for the Bank's more recent, intense and direct approach to civil service :,form has been the need to help borrower governments manage demand, to cont'ln and reduce a major cost area. The second set of reasons centers around the perception of low effectiveness and efficiency levels in goverrun nt administrations, attested to ir. critical cases by the degradation of the state-provided (and supposedly state- maintained) physical infrastructureI and even in the less critical cases 1b i4.gh transuction costs, pervasive delays, and the prevalence of corruptlon. nprovement of effectiveness and efficiency levels is seen as the essential and ultimate goal of administrative reform. From the outset it was thought, and experience has confirmed, that raising those levels is a complicated, problematic and longer term effort. The relation of cost contsinment measures eo efficiency/effectiveness promotion is thus analogous to the relation of stabilization to the restoration of growth in adjusting economies: i.e., issues of demad managemnt take precedence and arc easier to effect than restarting of growth. For civil service reform one could easily argue that is should be the other vwy round; that assessing vhat the organizations of the ut¶te need to do, what resources, physical and human, they need to do it, and how to go about doing it, should precede the question of cost containment and reductions. Nonetheless, becauso of the magnitude and intensity of fiucal problems, the cost containment steps havo been soon as the first priority. 3. Cost-Contaien.ntv Operational Aporoaches to Pay and Employmnt Reform -- A Progress Report -This section eszmL'ies the most recent Bank experience in pay and employment reform, analyzing the short-term, emergency measures taken (mostly through adjustment operations and companion technical assistance projects) to contain the size and cost of civil services. The focus is on the period since 1987 when Bank-supported pay and employment reform wva last reviewed. The purpose here is to provide a progress report on approaches to the problem, concentrating on their implementation. New initiatives that have been devised in the intervening period are also analyzed. The following section evaluates, on the basis of available data, the overall impact of these measures by examining general trends in wage bill containment, employment increases, and rationalization of remuneration structures. Bank operations dealing with government pay and employment issues have supported a range of reform approaches. Many of these have been implemented on a trial and error basis, and we are only beginning to leazn about which are working and which are not. (The key measures taken by countries to reform pay and employment policies are summarized in Table 3). With regard to cost- containment, various of these actions have been ranked on a continuum of political difficulty ranging from the easiest -- undertaken first -- to the most difficult. The main steps taken have been the fillowing: -- the elimination of 'ghost' or non-existent names and workers; -- the elimination of officially sanctioned posts which are not currently filled; -- the retrenchment of temporary or easonal workers; the enforcement of retirement ago (or retirement efter z years of service) stipulations: the freosing of recruitment; -- the elimination of ;uaranteed entry to the civil service from the educat'onal or training system; -. the voluntary, or induced by incentives, retirement of surplus workers; and finally the dismissal of serving civil servants. A number of techniques were utilized to support the steps outlined above, including: -- Civil Sorvice Censuses -- Functional Reviews -- Payroll Computerization -- Competency Testing -a Public Informtion Campaigns -- Redeploymnt Training and Credit Programs '- Salary Supplemnts -- Skills Mobilization--Senior Executive Programs The implementation of these measures are dis:ussed below. Civil Service Censuses, Payroll Computerization and Clean-Up and Ghost Reduction Keasures The reduction of civil service cadres through the elimination of ghosts 6 -- the first and leaist politically sensitive approach to employment reform -- appears to have been an effective techniquo in emergency cost-containment programs, although comprehensive data on reductions accomplished through this method are not yet available. (Table 4 presents various measures taken by selected countries to reduce employment.) For Ghana and Uganda. at least, ghost elimination represented a useful instrument of employment reform. 7 Ghost removal was also claiMed to have been a significant feature of reforms in various other countries. After a staff audit in 1987 in Guinea, for example, approximately 1091 ghost names were stricken from the civil service 'payroll. In Cameroon, approzimately 5000 fict.tious nams were eliminated. In general, ghost reduction Gftsts are name on tth payroll, receiving a wage, who cannot be shown to ex/st physically. They are worAers who hove died, retired or otherwiso left the civil service but were never recordod as such. rhey are fictitious persons whose pay Is claimd by othes. rThey are variants on a name with one prson receivlng two or more salaries. 7Data on ghost removals In Ghans are very tentativo and remain to be verifled. Government claims that 11,O0 ghosts were eslminated through headcount and census exercises have not yet been validated. In Uganda. approxlately 30,000 ghosts have been Identifled, but there Is no evldence that those names have been removed from the payroll as yet. incurs low political costs because the only opposing constituency are the system abusers themselves. for whom a public admission of fraud would be nocessary to stake a claim to continuing payment. Indeed, in none of the above cases did ghost removal generate a public outcry. The first step in ghost removal is, normally, a civil service censuA to determine the numbor and type of government employees. In many countries, such a poll will be the first time in many years (or perhaps the first time ever) an attempt has been made to got an accurate picture of public employment. The need to establish who is, and who is not, legitimately enrolled on the civil service rosters and payroll is paramu.rAt. Civil soervice censuses and payroll sanitization havo been features of a number of pay and cemployment refoms as shown in Table 3. Few reforming governments have ignored the issue, though soa have proceeded without a clean- up exercise. Laos, for eample, is reported to hav shed an estimated 10 to 30 percent of government workers using a payroll listing as the sole form of personnel records. For most countries, though, a prior data collection and clean-up activity is -scessary. They serve so -ral useful purposes, including quantifying the often large ranks of temporary staff; providing information leading to the enforcement of the statutory retirement age, and, most important, regularizing payroll lists through the elimination of ghosts. The usefulness of .. census is related to its level of comprehensiveness and sophistication. The most elemetal census-takiug exercise may be little more than a 'head count' which simply establishes ths. number and the structure of government eployment (i.e. the number of employes* at various professional lovels or in different regional locations). Censuses of the head count variety are usually desiSned and carried out by governmezts themselves. (At early points in the reform process, Ghana, Gambia, Guinea, Bolivia and Uganda and others performed hoad counts oan their own without technical assistance.) The results of these counts have often beon disappointing. Findings have been incomplete and inaccurate, mainly because they have been based on existing civil service rolls rather than verification of physically present employees. While not expensive to undertake, head counts of this sort can still be a waste of scarce resources, since they must usually be followed by more rigorous censuses carried out by outside consultants. On the other hand, even flawed head counts can serve useful purposes. Tn Ghana, the initial civil service survey revealed the overall structure of employment which clearly indicated overstaffing at lower grades of tho bureaucracy, providing a starting point for policy reform discussions. Sven censuses that are more sophisticated in their validation methodologies may have only lIJ'ited applications. Successive censuses carried out by international consultL ..s in Ghana (there were three all told) used the payroll mehanism to count civil servants, but they did not actually verify the physical existence of employees -- the rationale being that just getting fraudulent names off the payroll nwas good enough. Neither did they establish durable links betwoen the computerized payroll system in the Ministry of Finance, the personnel records in the Office of the' Head of the Civil Service, and the annual budget. The censuses did, however, provide important, one-shot, baseline data to begin the employment reduction program, thus fulfilling an essential 9 function. &t least tWO pgoblems can be seen with the Ghanaian approach. One is that, vithout the institutional capacity to utilize the data in a computerized monitoring system, the quality of census data is subject to rapid erosion. Results becomA outdated quickly as new recruits enter the civil service (openly or clsndestinely) and as others leave. Personal data which might provide an overall profile of the administration also change, so that information about stpects such as the age structure and promotion pattevus is no longer valid. Moreover, it is important to be able to capitalize on the results of the census quickly because the patience of civil servants to endure successive surveys wears thin, as has been the case in Ghana after three such ezercisus in a period of a fev years. However, "doing it right* also imposes administrative and financial burdens. In Madagascar, the census was more rigorous; enumerators were hired and civil servants had to be present at the place of census. The penalty for ton-cooperation was severet suspension of paycheck. Thus, participation was high. The cost of this elaborate exercise was a non-trivial five percent of the wage bill. however. Furthermore, the organizational requiremonts for execution of the census -- i.e. organizing review committees, enumerators, census centers at both central and district levels -- wege also considerable. In general, recent experience vith ;ensus design and implementation suggeSts that such mechanisms are important first steps to getting toe reform process uavi,g; that their design should be kept simple, but strategic in the sense that they should be conceived as part of the establishment of an ongoing system of controls, and thab. their successful conceptualization and implementation generally requires external tecbnical assistance. Freezing Civil Service Recruitment Freezing or limiting civil service recruitment is an oiten applied mechanism for reducing government employment. It is only marginally more politically difficult than ghost elimination. (The difficulty is due to the potential disruptions that might be causwd by aspiring civil service candidates.) Hiring freezes have been widely used, with varying results, in a number of countries, including Costa Rica, Ghana, the Central African Republic, Congo. Gambia, Gabou and Mauritania, among others. A variety of mechanisms were used to administer these policies. Senegal, for example, instituted c hiring freeze in 1983, and created a high-level inter-ministerial commission to supervise the process. The commission, the secretariat of which met weekly. revievwd all departures from the civil service, with a view to determining (a) if it were necessary to fill the particular vicancy, (b) if so, with whom, and (c) if not whether the staff of the organization could be i.deu-d by one or whether the or3anization had some pressing personnel need in another area that justified a position. Between 1000 and 1500 reductions were reported to have been effected in the first two years of the commission's operation. This experience, and a similar but stronger scheme in the Central African Republic. are summarized in the 'Box' on cost containment in Annex 2. Variations on this theme include allowing hiring only if it does not result in a noet creation of civil service posts (Kenya has tried this); or limiting hiri'g to essential professional staff (in force in Malawi, Mali and Nigeria). 10 Automatic hiring has been discontinued in some African countries which had traditionally guaranteed government jobs to school graduates. (Examples are CAR, Congo, Gu4 *a, Mali, Somalia, Sudan, and Senegal.) This reform measure turns out to be somewhat complicated to implement, as the case of Senegal demonstrates. When the hiring freeze was instituted in Senegal, there were more than forty government schools and training centers. Prior to the freeze, entry to these centers equalled entry to the civil service. The freeze forced the Senegalese to curail severely the entry to and activities,of these centers; the result to date has been the continued existence of most of these institutions, but with ever-declining numbers of students or trainees. Still, the Senegalese government, as moot others faced with this problem, maintained its employment coitment to those already in the training pipeline, meaning that it may be some years before an effect from this particular lement of restructuring can be perceived.8 !A general, the record of hiring freezes appears to have been mixed. Where they have signified a haLt to automatic civil service recruitment from universities or public administration training institutions, they apply a significant brake on civil service expansion. But recruitment freezes may constrain government from achieving the necessary skill mix through renovation of its cadres with young entrants. Moreover, for some civil services, hiring freezes may be no -ore efficient in eliminating redundant civil servants than natural attrition. Indeed, in som countries (such as Madagascar), the height of civil service ezpansion took place in the early 1960s, when many new recruits vere hired, causing a persistent bubble in the civil service rolls. Hiring then leveled off. It could be argued that in these instances it makes sense to let natural attrition through normal retirement, which will occur over the next few years, take care of this surplus. Suspension of Automatic Increases and Advancements One cost containment measure which has come into increasing use in the most recent operations is the suspension of automatic increases and step advancements. This is of particular relevance in the francophone African countries where advances in grade and pay tended to be based solely on years of service. Thus, freezing recruitment or even modest reductions in the total number of the civil servants soomtimes has had but a limited effect. Senegal, for example, froze recruitment as early as 1983, and even succeeded over two years in reducing total numbers. Still, it found that automatic step and grade increases kept the wage bill on the increase; the fiscal result of what the Senegalese regarded as an heroic and politically dangerous effort was reduction in the annual average STho Sengal Case rases another ISsue: the Adfalll refoarA In the parapublic sector Is resulting In reduced tVansfers to and, It Is hoped, eventual reductlon In the numbers of public enterprises. But many of the staff now employed In Senegalese PEs re 'tenured civil servants who chose or were assigned to servlco In a PE. If they are dismissed from thO PE they ha'oe the legal right to a post In the civil servico proper. Legal rights can be ended, of course. but only at the expense of time, effort and the expenditure of some of the government's political base. 11 growth rate of the wage bill from fourteen to nine percent. The suspension of automatic advances is an element of recont reforms in five countries with Bank programs, including Dominica and Mauritania. In addition, Burundi and Cots d'lvoi:e temporarily suspended automatic pay increases to dampen drift factors. Both Burkina Faso and Cote d'Ivoire also restructured their pay systems so that promotions are now offered on a more selective basis rather than granted automatically. Strict enforcement of retirement for over-age civil *.rvants has boen a foature of a number of reform programs (including Cameroon, Gabon, Costa Rica, and Benin, Senogal, Ghana, and Guinea). In Guinea, for ezamplo, the Bank supported the enforcement of a retirement program in which all civil servants over 55, or all those who had completed thirty years of service, were to be placed in automatic and effective retirement. The result was a reduction of over 7000 cadres. While presenting an option for employment reduction that causes few political ripples, the impact of such programs has been minimal. This is partly due to the unreliability of most information about age -- even when gathored by modern survey techniques through civil service censuses -- and to the overall age profile of most civil services in most LDCs. This is especially the caso in Africa, where the average age of civil servants is bolow 40. Thus, the removal of over-aged employees usually targets a semll group. Early Retiremnt Proarams Increasingly, countrios find that implementing the more "easy options does not result in sufficient cost cutting. They thus begin to explore ways to actually cut back on the numbers of serving employees. Even at this stage, governments are loathe to take the most difficult political route; that of involuntary dismissal. A number of strategies have been tried with varying degreos of success. Usually, these options are offered as one among several in a mized approach to staff reductions. Early retirement is an option that has been proposed in several countries, including Senegal. This program targets civil servants within a few years of normal retirement age, either paying them a lump sum separation package with pension benefits to begin with normal retirement age, or by starting their pensions at the early departure date. Although systematic analysis has not yet been performed on these types of programs, their benefits (in terms of numbers reduced and wage bill decreases) would seem to be minimal and their costs significant. First, the net present value of the savings stream is likely to be low since the savings only represent three to four years of staff reductions instead of those that would occur if younger workers were to be removed. Furthermore, early retirement programs target the most experienced employees - - both at managerial and support levels. This poses a particularly acute cost for those civil services in which once high standards of professional performance and training have broken down. Early retirement means that the collective experience of the older employees, who might be the sole keepers of the efficient flam of an earlier era, is lost to the newer generation. Voluntary Departure Programs Voluntary departure schemes have been a feature of several civil service 12 reform programs, including those in Guinea, the CAR, Mali, Kenya and Somalia; one has boen proposed for Senegal. In general, voluntary departure schemes are viewed as politically palatable because they are not coercive. Voluntary departure programs are problematic, however, in that they often attract the best and brightest civil servants; that ls, those whom government would most wish to retain. Carried out on a large scale, voluntary departure schemes are expensive. In Guinea, voluntary departure with an associated premium was to be financed with an earmarked fund of GP 6.5 billion. The benefits involved a three-option plan from which departees might choose tos (a) drav a salary with accompanying rice rations for 60 months; (b) take 40 percent in cash and the remainder in regular salary payments; or (c) take a share of the severance allowance as a down payment on a new private sector business venture. In its proposed voluntary departure scheme, Senegal would offer 60 months of salaries to voluntarily departing civil 3ervants. The government has requested financial support for this scheme from the lank. Many countrils would prefer to carry out staff reductions exclusively through voluntary departure programs, particularly if they can find external financing for a portion of the costs. Nonetheless, it seems clear that the monetary incentives would have to be significant (and probably unaffordable) to encourage sufficient reductions in numbers. On a smaller scale, some civil servants might have a low reservation cost that would permit them to take advantage of voluntary departure incentives that were not exorbitantly expensive for government. In this way, voluntary departure, when combined vith retrenchment, could be a viable strategy to achieve significant employment reduction. Howevr, on its own it is probably not terribly effective. Retrenchment Retrenchment -- the direct and explicit dismissal of redundant civil servants -- is the most difficult and politically contentious measure used to reduce employment. Governments have strongly resisted these remedies, and have adopted them as last resort options. This reluctance is related to the mature of the state in many developing countries. An important purpose of governments is construed as the distribution of political patronage and social welfare through the provision of public posts to loyal followers and otherwise deserving clients. Overt employment reduction thus strikes at this basic definition of the role of government. Moreover, the fear of most political regimes that retrenchment will incits destabilizing social upheaval and political opposition serves to stiffen resistance to overt employment reduction masures. The task of opponents to staff cuts is made that much easier by fears that the labor absorptive capacity of the private sector is weak. It ls thus not surprising that involved Bank staff note difficulties and delays in complying with staff reduction conditionality in adjustment operations in the Central African Republic, Guinea Blissau, Guinea, Ghana, Sao Tome and Principe, Senegal, Costa Rica and Dominica, to name but a few recent oases. Supervision reports record acceptance of fwer dismissals and lengthier timeframes than originally negotiated; these reports note as well in several second adjustment operations the repetition, in somewhat stronger terms, of employment reduction conditions seen in the first operation (SAL 1- enforce retirement age regulations;' SAL 2-'strictly enforce retirement age 13 regulationos). Despite the considerable resistance of governments to undertake what they perceive to be politically risky measures, retrenchment is nonetheless taking place in several countries through Bank operations. Ghana, Guinea, Gambia, and CAR (countries for which some documented reduction figures are available) are cases in pointa as of 1989, approzimately 27,791 civil servants have been retrenched ln Ghana, in addition to those removed from the rolls by other mchanisms. In Guinea, an estimated 4245 have been retrenched. In the Gambia, fol'owing a census and staff audit, approximately S350 government employees (including 2600 temporary employees) were dismissed. And in the Central African Republic, 300 civil servants have been fired so far, with more reductions planned for coming years.9 (See Table 4). The data on these retrenchment exercises are still quite preliminary, and in-depth analysis of this experience is only nov being performed)10 It is thus possible to offer only tentative observations on impleen tation. One notable feature of these programs was the use of a variety of techniques to deflect political opposition, minimize the social cost of retrenchment, and provide technical support to redundancy programs. In Guinea and CAR (and planned for Senegal), for example, public information campaigns were designed, with the help of consultants, to inform governmnet employees through radio broadcasts and newspaper articles, of governmAnt intentions vith regard to administrative and financial aspects of the staff reduction programs Bank staff reported positive effects for these efforts. Teetin_ Cometency exams were used in at least one country. In Guinea, competency tests for sitting civil servants provided technical criteria to determine which personnel would be graded as surplus. Those who passed the test were to be retained, and those who failed would be entitled to a severance package related to the length of service. (They would not, however, be able to opt for the more 9In faCt, many of the Mechanisms described In this sectIon have been utilized In combnatlon with one nother, maklng It difficult to Isolate the relative utility or Impact of any singl one. In Ghana. for example, voluntary departure was not stixulated through additonal Incenttves: the 'voluntary ewaratlon package was the same as for redundant workers, thus providing little notivatlon for workers to leave of their own volitlon. At the saw time, workers theoretically had the optlon of early retirement, but would receive no extra payomnts over and above their regular ponslon rights. If they choe voluntary retirooent, however, they would roceive the lump sum separatlon payment In addition to thior pnsion. Thus, there were virtuaiiy no earry rorirews in unana, our som older voluntary departee. Clearly, It was not the nature of the Individual scheme, but Its relation to. the other parts of the staff reduction program that dot-rmined the outcome t0Systematic analysis of pay and eomloyment reform In three Bank cperatlons (Guine, Ghana and Gambia)ls the subject of the Africa Technical Doepartme,t, Publlc Sector Managent DivisIon study In progress, 'Pubilc Sector Pay and Employment Reform In Afrlca,' forthcoming, 1990. 14 lucritive package offered those in voluntary departure). Recruitment in subsequent years would be determined according to the new qualifications testing, and the promotion of employees retained would follow a logical procedure coupled vith training. Although the exams were first geared to university level standards in France for professional level positions, the requirements were later relaxed to fit local capabilities. The objectivity and credibility of the program was validated by the presence of external consultants, and this apparently enhanced acceptance of the practice. Despite this acceptance of the principle of testing, the exercise has proved to be cumbersome to administer, causing conflict and delays in the overall civil service reform program in Guinea. Personnel Banks Competency testing was couplad in Guinea with the creation of a special nersonnel bank vhere those employees who passed sxawv,, but were considered otherwiso redundant, were placed in a special catelory. These people remain on the payroll for siz months. If they do not find jobs, they are removed and dismissed. They also have the option of leaving of their own volition with a severance package somewhat less generous than that offered to voluntary departes. There are approximately 14,487 employees in this category as of this writing. (This figure may include some 3400 public enterprise employees, howevr.) The administration of this personnel bank has proved exceedingly complicated, and there is not yet clear indication that any of those placed in the pool have actually left government. There is concern that this category of employees will continue to impose an onerous burden on the wage bill without contributing to civil service productivity. Horeover, the presence in government of an (officially) idle class of disgruntled employees may serve to demoralize other government personnel. Severance Packages Severance payments -- financed by governments themselves --have accompanied retrenchment programs in most countries. A comparison of severance provisions is shown in Table 5. While the absolute value of these separation packages is impossible to compare, the formulae themse'ves appear to be relatively generous. These formulae are mostly embellishments upon the basic legal obligations of government to disaissed workers worked out through negotiations among various interested parties. The administration and the size of the severance packages appear to be among the most important determinants of the level of acceptance of the overall employment reduction program by affected employees. In Ghana, delays in the calculation and distribution of separation entitlement in the first year of the reform program stirred some disruptive opposition among affected civil servants, for example. Bank staff reported that protests were smoothed over in the second year when severance payments began to be awarded on time. Functional Reviews In some countries, functional reviews have provided an important technical rationale for employment cuts. (Table 3 shows a sample of countries that have carried out functional reviews). These reviews are discussed in greater depth in Part B of this paper. Their purpose is essentially to audit the functions 15 of component agencies of government and the number and type of staff presently carrying them out in an attempt to determine the optimal staffing arrangements for basic government tasks. Functional reviews are difficult to administer (they are usually carried out through technical assistance), and they are time- consuming. Indeed, as is discussed in detail below, in the two cases where they figured mobt prominently in the reform progr4m (Ghana and Gambia) many cvf the employment cuts actually took place before the functional reviews had been started. Nonetheless, it has been reported that the carrying out of these review. even well into the implementation of a retrenchment program provides an assurance that there is at least an intention to pursue staff reductions in a rational, technical,-and just manner. This assurance has served to dispel political and social discontent over the retrenchment programs as a whole. Redeplonmgnt, Retraining and Credit Proarams Bank operations have begun to address the issue of redeoloyment of redundant labor in their reform programs. Several countries have put together programe which offer training for jobs in the informal sector, credit schemes at favorable interest rates for small business or agriculture, and/or public works programs. (Ghana, Gambia, Guinea, and Bolivia offered such programs.) On a technical basis, these programs have been less than fully successful; in some instances thetir justification may be questioned. First, retraining for informal sector activities -- which is where many redundant workers tend to go, may be largely irrelevant; indeed, the number of poople benefiting from training for carpentry or dressmaking is probably quite small. In Ghana, for example, where this type of program was offered through the Prograe of Actions to Mitigate the Social Cost of Adjustment (PAMSCAD), the demand for such training was low. In most countries, the existent training institutions are not equipped to handle this kind of informal training, or if the demand were higher, to accomodate the number of trainees that potentially could result from massive lay-offs in the public sector. The administrative costs are high and the potential benefits unclear. Again in Ghana, an attempt vas also made to provide training through an apprenticeship program. This was difficult to organize, however, and only reached a limited group. Like the other redeployment schemes, this program was undersubscribed. Credit for small business and agricultural activities, another aspect of the program, was also in low demand. In the Gambia, the Institute of Business Advisory Services (IBAS) administered a similar program which received approximately 700 applications (out of more than 2000 workers laid off by the government) and provided credit and/or training services to only about 300 individuals. In Bolivia, a public works program was organized through the Emergency Social Fund to provide transitional employment for dismissed tin mine workers from COMIBOL, the state mining company Reports indicate that most miners did not take advantage of these programs, however. Instead, they sometimes used the cash payment provided by a generous separation package to relocate and start enterprises in the informal sector, or to find construction work elsewhere. In Guinea, the Government set up a special office, BARAF, (Bureau d'Aide a la Reconversion des Agents de la Fonction Publique) to assist departing civil servants in applying for credit to finance private businesses. Some Bank staff fear this credit program may have permanently damaged the country's credit system; indeed, there are now on the books large numbers of loans to departed civil servants that have almost no hope of being recovered. The Gambia small business credit scheme for redundant civil servants did not fare well in this respect either. Its loan recovery rate was 30 porcent (as compared to 65 percent, for example, for an SEC-financed credit scheme with more stringent eligibility requirements). This experience suggests that the technical rationale for these programs is weak. It would appear that many civil servants being retrenched nsither need nor expect to be retrained or redeployed. Most go into the informal sector for which therso ^eLw adapted credit schees. It would be ad-inistratively easier, and would probably havi a more beneficial economic impact, to simply award cash severance payments to departing staff. Another point is that redeployment program require considerable institutional capacity to ad-inister. In the Gambia, the program only got off the ground one and a half years after retrenchment had begun. In Ghana, tho lag tim was two and a half years. In Guinea, the program began on the anticipated date, but proved to be very difficult to manage. Typically, there are a number of institutions whose activities must bo coordinated and the preparation requireants are considerable. Moreover, because existing training and credit institutions are inadequate, new aechanisms often are required, imposing substantial additional costs. Finally, for some countries, the need for special mechanisms to rechannel labor into private markets may be less than anticipated. That is, there is some preliminary evidence to suggest that labor absorption has been easier than expocted -- or than governments typically contend. In Africa. for example, urban workers appear to have moved easily into agricultural activities. In Bolivia, the ability of the informal sector to absorb redundant public sector employees appears to havo been considerable. Thus, the technical rationale for retraining and credit is further eroded. These programs do perhaps have an important symbolic value to the extent that dismissed workers place value on access to this kind of assistance, whether they avail themselves of it or not. Such programs thus serve an important political function in defusing potential discontent among retrenched employees. In general, though, it may make more sense to concentrate on severance packages or generous golden handshakes in desigzing employment reductions programs, as this is where most eployee interest and demand are focused. Remuneration Issues Many of the cost-containmnt measures taken have focused on reducing asploymnt; but many steps have also been directed at containing the wage bill in the aggregate, and at removing distortions in the overall remuneration structure. These measures are discussed below. Wase Restraint Wane freezes are common measures. Full or partial wage freezes were 17 called for in Uruguay, Senegal, Gambia, Sao Tome and Principe. Cameroon, Gabon and Tunisia. (Prior to the Bank's increased involvement in wage bill restraint, wage froozes featured more prominently as conditions of D(F agreements.) Ways of instituting a wage freeze or at least restraining wages are numerous; i.*. by holding wages to existing levels in current terms;, by holding them in constant ternr; by allowing increases equal to a portion of, but less than the entirety of the rate of inflation; by setting a currency unit ceiling, or by agreeing that the wage bill cannot surpass a particular ratio (for example, percentaps of government expenditure). Rationalization of Remuneration IAcreasingly, pay and employment reform programs have also been addressing specific pay conditions for civil servants in an attempt to remove deaotivating distortions in government remuneration structures. A number of reform programs have aimed at improving pay conditions, rationalizing the overall system of romuneration, and building an institutional capacity in government to formulate *ad implement sound salary policies on an ongoing basis. Uaticnalization of the remueration structure has been an important reform objective. This has mainly consisted of attempts to reduce the proportion of rommnerations from non.wage benefits. In many countriec, these constitute an unacceptably large percentage of total compenation, as real wages have been steadily eroded by inflation and expanding employment. Most work on this issue has boen in the form of studies calling attention to the problem and recomending solutions. Such studies are being carried out in Senegal, Cameroon, Hauritania, Senegal, among others. In several countries, concrete steps have been taken to reduce allownsces. Bolivia, Laos, Guinea, Senegal, and Cameroon have reduced non-wage allowances (housing in Cameroon), in-kind benefits (rice rations in Guinea and special porformance prenia in Bolivia.) However, available data suggest that, in genoral, virtually no progress has been made on this front. As Table 6 indicates. non-wage benefits as a percentage of total compensation have increased on the average by nearly 8 percent for those countries that experienced a change. It is claimed that the Cameroon has also significantly reduced the number and extent of allowances, but data do not show any improvement between 1986 and 1988. Bolivia has officially taken actions to reduce benefits, but data on the results of those measures have not yet been collected. One reason for the slow movement on this issue is the difficulty in getting systematic information and disentangling the enormously intricate webs of benefit structures. For many coungries -- in particular, those in francophone Africa -- it may well be that refom will only result from a simple declaration nullifying the entirety of the present convoluted benefit system, and replacing it with a rational, salary- based reward system. Unfortunately, it is politically difficult to dispense with non-wage machanisas without redressing the inadequacies of the salary structure which stimulated the emergence of non-wage distortions in the first place. Bolivia's recent attempts to legislate a rational salary system have been circumvented, for example, by spontaneous non-wage bonus features to reward some employees outside of the strict wage system. Equalizins Pal Discregancies 18 Ineouities in pal amont various parts of governnent have also been the target for rationalization. In Mauritania, salary discrepancies among different central government ministries for comparable work, and the effects of inequities resulting from irregularities in employment status, were both the subject of studies financed by the Bank's Development Management Project. Closing the gay in pay and benefits among different parts of the public sector was the goal of some reform programs. In Jamaica, and many other countries, parastatal bodies were able to attract qualified candidates away from the civil service with substantially higher salaries. SAL III instituted conditionality that higher level management posts in the civil service be remunerated at 85 percent of equivalent grades in statutory bodies through a three-phase pay hike. Although two raises for managers did occur, the opposition to differential wage increases from unionized civil servants at lower skill levels prevented the third increase from being enacted. As a result of this delay and the continuing wage erosion through inflation, qualified professionals and managers continue to be in short supply in central government.11 Simplifting the Salary Structure In some countries, simplification of the salary grid has resulted from pay and grading study recoamundations. Studies have been carried out in Bank- financed programs in the Gambia, Ghana, Guinea, DoLinica, and Jamaica, for egample. In Dominica, the reuneration system, consisting of over 100 pay scales, was converted to a structure of 14 for middle and lower management; almost all jobs are now included in this new pay scale system. In Guinea, consultants undertook a job evaluation exercise and reduced 19 grades to 12. Wage Compression and Decompression For many countries, the erosion of average wages was the big problem to be addressed by pay reform. But, particularly in Africa, a principal objective of pay reform is the decompression of the wage structure. This is essential, given the difficulties in recruitment and retention of higher level staff whose salaries have sunk to very low multiples of the lowest ranked workers. (Table 7 shows some compression ratios for selected countries.) Decompression of the salary structure was a definite aim in Ghana, for example, where in 1984, before the SAC I program began, the compression ratio was approximately 2.5 to 1. It rapidly moved to 5.7:1 in that year and by 1989, it had decompressed to 7.8 to 1. The Government's stated objective of achieving a 13 to 1 ratio by 1991 may or may uot be ,eached, but it is clearly moving in the right direction. In the CAR, five years of reform produced no perceptible change in the compression it Mothodologies for determining the appropriate pay levels In the civIl service are only now being developed. In Jamaica, a comparative pay survey was carrfled out through the Administrative Reform Project, the Institutlonal counterpart for pay reform In SA. 1il. But the survey took so long to be cc -:leted that the AMP was terminateo before the results becm known. Slmilar surveys .-re carrled out In Ghana, the Gambla, and Tlhailand to determine the approprlate pay levels for clvil servants. 19 ratio. And in Laos, salaries decompressed from 3 to 1 to 6 to I under a prograa supported (in its later stages) by the Flnd and the Bank. For othor countries, progress is much less clear. During the reform period in the Gambia, for example, salaries appear to have become more compressed in the first years of adjustment, moving from 6.5 to 1 to 5.7 to 1.. -(This might havo been due, in part, to the introduction of the 'new, shorter twelve grade structure, however.) In guinea Bissau and Senegal, as voll, it would also appear that salaries compressed further during the reform period, from 5.2 to 1 to 4 to 1 in the former: to 7.8 to 1 to 6.8 to 1 in the latter. It must be underlined that all theso figures should be taken with a large grain of salt; they tend to be 'guess-timates', based on sketchy data. Salary SuDDlements Vhere reform of the salary structure has not been mminent, a number of countries have resorted to interim programs aimed at attracting qualified professionals into government. Salary suoolements are the most common mechanism used to compensate for low civil service pay at upper echelons. Although no systmatic docuentation exists, salary oupplements appear to be widely utilized. Exaples include Mozambique, Bolivia, Uganda, Guinea Bissau, Camroon and Niger, among others. In councries whore intrntational donor assistance provides an alternativo employment market for skilled professionals, salary supplements may be provided by eoternally financed projects in several different ways. Civil servants may accept supplements to top off their civil service pay by as much as 400 percent (as in Mozsambique or Bolivia), without taking leave from the government rolls. Indeed, in Moxambique (and elsewhere) Bank and other donor projects have hired individuals working in ministries to staff special project units at higher salaries while still retaining their government Jobs. The obvious conflict of interest potential here is enormous. Moreover, the demonstration effect for those government employees lacking access to topping- up benefits can be devastating.. Some c ril servants may take leave or actually separate from the civil service. Whi., this is a *cleaner' arrangement in principle, the cost may well be the loss of the skilled individual from government service over the short and long run. In countries such as Mozambique where human capital is so scarce, this brain drain could paralyze government functions. Moreover, on a large scale, such supplementation is affordable only with sustained external financing. The siphoning off and rewarding of the best and the brightest undercuts government motivation to improve conditions for the majority of individuals. On the other hand, salary supplements are viewed by many BanK Staff as unavoidable. They are seen as the only way of obtaining good staff to run projects and programs which, presumably, will enhance the functioning of the civil service, and restart the process of economic growth. The argument is that the loss of time and money that would occur with inferior staff is worth both the higher remuneration, and the demoralization of the unaffected majority. In some countries, salary supplements consist of those payments provided to hisher level civil servants over and above their civil service grade salazy. These payments -re offered largely because it is feared that, without topped up 20 salaries, the most capable personnel will be luned into internationally financed projects or, better-paying parastatal, or (where this is an option) into the private sector. There is little doubt that salary sLpplements have a corrosive and distorting effect on civil service moral& and management. Host important, they undermine the possibilities of meaningful structural reform in the longer term. The insidious aspect of this problem is that in many instances the offenders are tho donors, including the Baik. This is so despite the Bank's overt policy of not financing government salaries. Zn a few instances, innovations havo been proposed to deal vith this problm. In Bolivia, intornational donors contributed $ U.S. 6 million to form a foundation to finance 500 key high level government positions -- all performance evaluated -- outside of the budget and outside of the Da' targets for the wage bill. While this was not seen as a long-term solution to the salary problem, it was viewed as a means to bring som order to tho chaotic pay suppleeont situation. In Ghana, tho Bink assisted an innovative Skills Mobilization Scheme to finance local consultancies for key government positions essential to the economic recovery program. Fees for these coniultancies were to be at local private sector rates -- well over civil service levels -- and the positions were not to be filled by serving government employees. The idea was to attract skilled professionals from the private sector, and porhaps even repatriate skilled Ghanaians living overseas. After considerable debato over the tems of these contracts, the schem was made operational, but appears to have had little impact. At time of writing, forty-one individuals have been identified as eligiblo for the scheme and registered on a consultants roster, but only three government slots have actually been filled through this mechanism. Parastatals vs. Central Government There are various aspects to the salary supplemont problem. Attempts to approach parity betveen central sovernment salaries and parastatals were contemplated, if not alvays fully carried out, in Jamica, Ivory Coast and Bolivia. Such attempts sometimes result in a no-win situation. First, the equalization of state enterprise salaries with those of central government agencies is often not achievable; all that can be accomplished is to narrow the gap between the two in order to stem the flow of qualified staff from one to the other and to mitigate somewhat the resentmnt of civil servants at receiving relatively lower salaries. Second, even in cases where parity has been legislated, the incentives to performance of state enterprises themselves cannot help but be negatively affected. To the degree that parastatals behave more like central government agencies than private or competitive enterprises -- particularly vith regard to salary and personnel policies -- parastatal performance is likely to suffer. Institutional Arrangement Issues The Bank's interlocutor(s) in programs to reduce the size and contain the cost of the civil service have generally been the Hinistry of Finance and the 4inistry of the Civil Service. The tendency of the Bank to centralize its socus on these core agencies has occasionally oversimplified the complicated map of 21 institutions involved in various phases of cost-containment. In the Ghana case, the failure to perceive and bring on board the entirety of agencies involved in the process resulted in design mistakes and delays in impl¶mentation.12 In other instances, the failure to understand that institutional arrangoments in countries are more decentralizod and less controllable than donors would like (or are accustomed to) has led project designers to underestimate the potential for non- compliance with reform objectives. Again in Ghana, the exclusion of the Ghana Education Service from redeployment .discussions contributed to a misperception of the chances for post-freeze recruitment of teachers. Internal Bank nrocedures are also factors affecting the outcome of the reforms ezperiences discussed above. Onne lemnt increasingly recognized as crucial to the preparation of pay and employment programs is diagnostic sector work in anticipation of and in conjunction with SAL design. Much of the uneven progress noted at various points in this paper in the Ghana reform progrim might well have been mitigated by adequate diagnostic work that mapped out the strength. and weaknesses of the institutions that would need to be involved in the program. An 'institutional assessment' of this type might have resulted in a strategy which established the objectives, and the ways to reachx them, of each step of the reform program. 4. I?.act of Cost Containment Reform The above discussion has focused mainl" on aspects of implementation of government pay and employmnt reforms through Bank operations. Although a fair amount can be determined about how these programs are working from this implementational perspective, a sense of the real contribution of these reform programs can only be achieved by looking more closely at their outcomes and trying to assess the degree to which they have accomplished their stated objectives. (Note again that the usual warnings obtain about the limited extent and low quality of the data.) Wase Bill Imoact The primary objective of most reforms wag to reduce the aggregate wage bill. To what extent has this occurred in recent years? Is there any way to link changes (positive or negative) in the wage bill to Bank-sponsored government pay and employment reforms? (Table 8 shows wage bill trends for selected countries). For the fifteen countries with available data in which either the Bank --through a SAL or a TAL -- or an DMP program sponsored reform, ten (two- thirds) had wage bill increases for the last two years. Moreover, this group included those countries in which the reform program had progressed furthest, such as Ghana, Jamaica, Guinea, and the Gambia. In only four countries did the wage bill decline in absolute terms. In nine countries, wages and salaries rose 12A noted by J. rait oavis In -Revlew and Evaluation of Ghana's Clvil Servlce Refor. Programm," Consultant's Report, ATPS, World Bank, October, 1989. 22 salaries rose as a percentage of total expondituros. Moreover, the percentage of total expenditures on goods and services or materials and supplies, the indicator of the availability of necessary inputs for government staff, declined for seven out of the group and rose only for four. (For Togo, the percentage of current expenditures spent on goods and services declined for .he last year of available data). For the other four countries, data were not available. To repeat, these data are not very robust and it would not bo vise to overgeneralize from them. Indeed, there mia be any number of good reasons why the wage bill rose in these countries that are completely umrelated to civil service refou programs. It is not possible to explain reliably the idiosyncratic reasons for each vage bill increase. Even for cases where it is possible to probo data more deeply, the reasons for increases in the wage bill can be very ambiguous.13 Moreover, even at the aggregate levol, the time lag between reform and impact may bo longer than can bo captured in the available figures. It is possible that these results will only be visible 1-n future data sets. It should also be noted that personnel exponditure trends often do not show up in the aggregate wage bill figures. They can be hidden in the budget in myriad other (often non.itemized) categories. Thus, som of these wage bill figures may actually understate the total amount of personal emoluments for a given year. Taking all these caveats into account, however, the data suggest that the impact of these reform programs on aggregate wage bill reduction is negligible. The increase in overall wages and salaries expenditures may be due to general pay hikes awarded to civil servants to alleviate the erosion that had been affecting government salaries in many countries -- again, especially in Africa. As has been pointed out earlier in this paper, this general wage erosion affected different classes of civil servants differently, and higher level professionals salaries had fallen to dangerously low levels in many cases. Thus, the possible 'good newsv in a higher aggregate wage bill might be the decompression of the salary structure from top to bottom. Data on wage compression are poor. Table 7 shows changes in only six cases. Of these, half the wage structures showed decompression and half showed compression. Looking a bit more closely at individual reform programs it seems clear that Ghana has made progress in decompressing its salary structure, and Bank staff confirm improved morale among upper echelon personnel. In the CAR, it would appear that not much decompression has occurred. And the compression situations in Guinea ane tho Gambia are less encouraging still. Deepite reform programs, vages have actually become more compressed. In the Gambia, this is especially disturbing since a hotly debated salary increase of approximately 25 13 In 8u/lvla, for example, where ore detailed analytic work has recently been eoapleted on pubilc sector wage and employment trends, the Increase might be explalned by a rise In employment (perhaps with declining average wages) or an Increas'. In average real salarles. The data for that country are so ambiguous that It Is ilpossible to argue eIther case with confidene. See Barbara Munberg, 80oliv/C: A Review of Public Pay and EmJioyment Issues' 1990. 23 percent was awarded in 1988. Another possible explanation of wage bill incresses m3ght be the rationalization of the romuneration structure. As noted, non-wage benefits and allowances have traditionally constitu.ed an important component of the overall compensation package. These benefits have oftcen been unevenl distributed and have rarely boen linked to performance-based criteria -- or eve need. In most countries information about these benefit structures i unavailable to policy makers. And, of course, the more important the non-wag benefit contribution to the overall compensation package, the more i: is likel to distort the general distribution of remuneration, thus making culculation of compression ratios among various .evels of the civil service at. best onl indicative of real remuneration relationships. EuDlo7ment Reductions In the absence of clear improvements on the fiscal side, the overal objective of employment reduction retains importance. Has this been achieved under these programs? Table 9 shows the limited data available on employment trends for selected countries which have undergone some form of employment refosm. For the fifteen countries for which data are available, eight were able to reduce goverument employment during this period. Among these, out (Ghana) experiencid slippage through new recruitment in the Education Service. Two others - Jamaica and Cameroon -- experienced significant reductions in recent Years, but then reversed the trend slightly during the last year. For the other countriec, the rate ofIrowth remained relative4y constant during the reform period. In general, these reductions were of small magnitude. The limited impact on the wage bill for those few cases where results have been even modestly encouraging may be a function of the very low levels of employment cuts that. countries are besig asked and are willing to make. Indeed, it may well be that more serious impact can only be achieved through more drastic reductions. It is also important to note that the saving achieved through government cuts were generally not sufficient to pay for the subsequent salary increases avarded to correct previous civil service wage erosion. Indeed, preliminary evidence from Ghana and the Gambia suggest that aggregate pay increases vastly outstrip the savings accrued by the retrenchment exercise. rTe discrepancy between the relatively small fiscal savings of retrenchment and the generous pay hikes is partially explained by the extreme wage erosion that had taken place in many countries in recent years. It is not likely that significant redress of this erosion could be achieved through anything less than massive employee lay-offs, the likes of which have not been contemplated or carried out in any of the present cases. Finally, most employment reduction has taken place at the lower pay levels of the civil service. This has tended to mitigate the financial impact of the redundancies. Part 3: TEM LCNG VIEW: CIVIL SERVICE HANAGEMENT 24 Short-term efforts te correct glaring and burdensome distortions in government pay and employment practices have clearly dominated the Bank's operational agenda in civil service reform. However, with the advancement or completion of emergency surgary, many governments -- vith Bank support -- have begun to focus on longer term issues in civil service strengthening. The along view' transcends cost-containment to concentrate on detailed, slower-paced reforms aimed at improving levels of effectiveness and efficiency of government administrative systems. The grand objective of these 'rationalist' reforms is to create or strengthen a capacity for petsonnel policy formulation and for day- to-day management of the civil service. Underlying the longer-term approach is the recognition that shorter-term measures must bo supported by institutionalized systems that can sustain ongoing reform. To the extent that development management iaterventions are linked to policy-based reforms, they may be distinguished from the oublic administration assistance activities of other donors working in this field (particularly the United States Agency for International Development [USAID] and the United Nations Departmeat of Technology, Co-operation and Devolopment (UN-DTCj]). These latter have tended to engago ln eztensive, diffuse, training and improvement programs, often linked to particular projects, with little policy focus and few links to precise, measurablo, institutional outcomes. In essence, the Bank's long view, more rationalist' reform programs aim at: * AAatalling-in- civiLsorvices--personnel information and management systems, more tightly linked to payrolls, and including clear and appropriate career development schemes; * staff audits, to determine vhat personnel is on hand; • improved training systems; * revision, usually meaning simplification, of the legal framework governing the civil service; and * getting the right people into the administration, partly by stronger incentives to attract and retain them, partly by changing objectives and procedures in an effort to make the work situation more challenging and rewarding. Civil service management reform has been dealt with in a series of technical assistance projects of relatively recent origin. The main activit es of these development management operations are outlined bolow. The discussion then raises a set of important implementational issues, and concludes by drawing the first discernable lessons from these operations' results. Since 1981 approximately 23 TALs have supported civil service management reform programs. Most, though not all, were in Sub-Saharan Africa. Host of the operations have been specifically linked to institutional reforms contained 14somw of trese rALs were only partalaly devoted to developaent sanaegent Issues and thus cannot be strictly classifled as developnt can.gent projects. 25 in ALe (The remining have oecn supposedly freestanding projects, but even a fair number of these ha7e had an indirect relation to SAL reform programs.) There is no formsl model for reform that can be applied uniformly across countries, but there are identifiable sets of activities which most of these projects have in c,imon. These measures are presented below according to the phased order in wh.ch they usually occur in the reform process. 1. Comoonents of Rationalist Refor-uUfiorts Data Collection and Analysis In the least developed countries with weak reporting and records systems, the first need is to determine the facts; i.e. to determine the number, personal characteristics, skills, years of service, pay, etc., of the civil service. In some cases, there has been a coincidence of this data collection exercise and a civil service census (Mauritania, for instance). In others, such exercises may complement or supplant previous censuses which were methodologically flawed or did not go far enough in collecting pIrtinent information. This was the case for Whana and the Gambia, or example,. ' Vhilo tim-consuming and expensive, these data gathering and analysis exercises provide the data base for the establisheat of effective records managoomt practices. This means improving registry procedures and building systems for the ongoing collection, aggregation and analysis of personnel data. Even in the short run, a manpower data base for the civil service is essential to devising and implementing retrenchment strategies. They provide information that assists in the determination of priority areas or ministries to undergo functional reviews (discussed below), for example, or they help establish the scope for staff savings in various employment categories. Introduction of Computerized Payroll Functions Civil service management improvement programs financed by the Bank often have included support for the rationalization and computerization of the payroll mechanism, usually located in the Treasury in the Ministry of Finance. (Ghana, Senegal, Mauritania, Uganda are a few examples). In some cases this computerization has proceeded on its own; in others, the payroll system has been linked to the improved personnel management data base discussed above. Explicit links to the budget have also been forged. In several countries, implementation 1 IOGhana, for exampie. the original 1986 civil servicoe census and the manpower survey carried out by the Neadcount Committee In 1987 to Identify ghost workers were successful In Indicating In broad terms the overall numbers and distributlon of civil servants, but were technically flawed In other ways that linited their usefulness as a imangment tool. For example, the data were collected and stored manually, making agregate analysis difficult and time-consuming; the population" was not clearly defined so that It was not possible to tell exactly what the percentage coverage or response was; certain key pleces of summary personnel informatlon were not collected; and, the surveys were 'one-off exercises with no provision for maintaining or updating Informatlon. 26 of this integrated system has not been carried out, but the management TALs have provided support for studies to produce recommendations and the technical rationale for the implantation of such a network. The integration of the personnel data and the payroll and hudgetary information is somewhat cumbersome administratively and institutionally as it involves the coordination of various organs: ministries of finance and civil service, statistical institutes, computation centers and technical line ministries are usually all involved. But comprehensive and sustained control over personnel management functions cannot occur vithout a linked system. Furthermore, the greater the degree of decentralization of personn_Lmanagement, the more crucial an integrated network becomes. Most governments (and several donors) have been easily seduced by the appeal of computerized payroll systems, but there has been greater resistance to the notion of an integrated payroll-personnel management information system. Functional Reviews Functional reviews have been a common feature of Bank-sponsored civil service management reform efforts. Such reviews are in essence organizational audits on a sample or selected priority list of central government agencies. Their overall aim is to determine the appropriate match between types and numbers of staff, and tasks or functions of government. In adjusting countries, these review have sometimes been hurriedly devised and rapidly implemented. Thus, they have not always met the most rigorous standards of productivitv measurement. Nonetheless, functional reviews do represent a rough, often best available, effort to determine the appropriate number of perionnel and the type of skills necessar" to carry out the organizational objectives of various component parts of government. To some degree, the term, 'functional reviev' is a catchall for a range of related or similar activities, includingt 0Job Inspections -- A job inspection program reviews the structure and staffing of agencies in government to determine: what work needs to be done to fulfill the organization's objectives; whether the organizational structure is appropriate for its activities; whether the staffing numbers, grades, and levels of responsibility are appropriate to the needs of the work; and, whether some degree of consistency is maintained across agencies. * Orxanization and method studies constitute another instrument used in functional revievs. These are evaluations of the efficiency of a particular service provided in a similar way by several or many agencies, and where significant numbers of staff are employed. The aim is to identify main areas of weakness in performance, to find and remove bottlenecks, and, in sum, to simplify procedures. This process can yield significant staff savings as well as promote greater efficiency. * A third technique used in functional reviews is budgetary analysis. This is an examination of financial data in an effort to pinpoint sreas where manpower are being inefficiently employed. This, it is assumed, automatically identifies potential targets for budgetary savings. One approach is to undertake analysis by exceptions for example, to pinpoint agencies where the proportion of personal emoluments to total expenditure is particularly high, or where trends in 27 personnel emoluments have deviated significantly from observed trends in other program expenditure. * Functional reviews often use ratio analysis, a means of identifying potential problem areas requiring more detailed policy reviews. Ratio analysis is tailored to the precise functions and objectives of the organization or sector concerned. Typical examples include medical officer/bed ratios, teacher/non- teaching staff ratios, agricultural eztension officer/population ratios, and cost of tax revenues collected/tax collector. These ratios are then reviwed against government targets to assist in highlighting regional disparities and, in particular, flagging areas of apparent staff shortages or surpluses relative to the average. Ratio analysis is indicative rather than conclusive; it indicates needs for more specific policy reviews, and ultimately to more detailed plans for recruitment, training and staff developmnt. In the imediate, it helps identify functions which are obsolete, overlapping or otherwise redundant. The findings of functional reviews lead to the next phase in the sequence of administrative rationalization; the matchina of existing personnel to ideal needs. In principle, this should result in the identification and retention of those individuals with appropriate and needed skills. It should also lead to the reassignment (and porhaps retraining) of staff superfluous to the requiremnts of the agency under review, but who possess capacities and potential required elsewhere in the system. Finally, it should identify the individuals who are not needed, and should be dismissed or porsuaded to leave voluntarily. Functional reviews have been utilized in several countries (Ghana, Guinea, Benin, Gambia) to provide a technical basis for decisions concerning employment cuts. Although thQ heavily rationalist process described above suggests only an ideal type, functional reviews of the sort described have enhanced the credibility of redundancy programs, in Ghana and the Gambia, for example. In both countries, as part of the adjustment process, functional reviews were conducted over a period of a few months. They vere mainly designed and administered by international consultants. In Ghana, the consultants, in conjunction with British Overseas Development Assistance (ODA) -sponsored overseas training, succeeded in transferring personnel review techniques to local staff in the Management Servicos Division of the Office of the Read of the Civil Service (OHCS). The result was that a corps of competent job inspectors was built from scratch. This successful transfer of technology turned out to be an important and positive feature of the program. It not only boosted professional pride in the OHCS, but also extended a sense of Ghanaian ownership to the redeployment program. On th. other hand, job inspectors were not the highest priority among the manpower requirements of the Ghanaian civil service. Presumably, the demand for these skills will diminish considerably -- though not disappear -- once the initial, cost-cutting phases of the redeployment program have been completed. In the Gambia, by contrast, virtually no on-site training of local staff was accomplished by the implementing consultants in the execution of functional reviews. The consultants attributed this admitted deficiency to a number of factors, including intense pressure to produce the reviews very quickly. 2S It is worth noting that in both the Ghanaian and Gambian cases, a substantial number of employment cuts (and certainly those effected in the first months of tho reform programs) were made izn advance' of the findings of the functional reviews. However, these reviews did, reportedly, help in the justification and implementation of the second round of cuts. The reason is as followst the second-round targeted reductions fell on areas of less obvious 'fat, and tho data and recomendations of the functional reviews provided decision-makers vith 'technical' and 'objectivoe information with which to contest the various groups opposed to the staff cuts. In Senegal, a process akin to the functional reviws .- called organizational audits -- .vs carried out by the Office of Organization and Methods (BOM), a management training and trouble- shooting unit located in the prestigious Office of the President. The audits of DOM have progressed such more slowly than thoso performed by consultants. perhaps in part because of difficulties in winning the cooporation and support of key ministries in the eandeavor. While there is an advantage to the national control of the Senegalese exercise, one must note the cost of the suail?s pace at which policy actions stewLing from these review have materialized. Trainina In-service training of civil servants either on-the-job, or in short courses abroad have been a feature of a number of civil service management projects. ra several cases (Malaiw, Mauritania) the main focus of projects has beon on strengthening civil sorvice training institutes. Zn Mauritania, a "training needs assessment' study was financed through the Development Management Project, and a twinning arrangement was established between the Institute of Public Admi44stration in France and the National School of Administration (ENA) in Mauritania to carry out in-service training of higher level civil service personnel. The experience has been promising. EA -- once a stale institution focused exclusively on preoservice training for new civil' service recruits -- has turned its attention to improving the skills of government personnel already in service. Training in thus serving to reinforce government's commitment to improve quality and reduce quantity in the civil service. In Malawv, the sale focus of an Institutional Development Project is the training of higher and middle level managers in the civil service. Increasing the modest national capacity for management training has long been viwed by Government as an extremely high priority. This view has come to be shared by the Bank, which has attributed recent implmentation bottlenecks in adjustment programs to the severe shortage of trained personnel in the upper levels of the civil service. Thus, the project aims at the strengthening of the Malawian Institute of Management. Involved Bank staff view this project as a necessary component to civil service management improvement. Still, they are quick to point out that training and skills improvement are necessary but not sufficient conditions for reform. The crucial fact remains that there are few incentives, financial or otherwise, for newly trained and presumably improved managers to return to the civil service. Thus, the next and necessary step vill be to pursue improved personnel policies, incentive systems, pay and employment schemes, and performance 29 evalustion. Host of the work on the project to date has been done under a Canadian twinning oxercise which has worked well as a model for building training capabilities. Special Incentive Schemes for Biiher Level Civil Servants Weak incentives systems mean weak civil services. As discussed earlier, 'topping-up' mochanisms that circumvent civil service salary scales have been seen as a possible answer. They are used to attract and retain highly skilled professionals into government service, in particular so they may perform key economic recovery program tasks. Topping-up mechaniss were explicitly funded in two TALS: tho Bolivia Economic Management Strengthening Operation (EMSO) funded a non-govermental agency -- a 'foundation' -- to channel pooled extra- budgetary rosources for koy posts; and the Ghana Structural Adjustment Institutioal Support (SAIS) project created a Skills Hobilization Scheme to hire local consultants for government positions at salaries well over civil sorvice levels. These schemes were aimed at creating something akin to the establishod and enduring U.S. Sonior Executive Service. While they have succeeded in getting some skilled poople into government service, they have so far boon seen as short-term, incentivo-boosting mechanisms, not as enduring institutions. Such instruments bave been included in development management projects in an attempt to impose a modicum of order on the haphazard array of aechanisus creatod by govor iuts (and often suppetted by donors) to circumvent civil service remuneration constraints. Policy Studies Studies aimed at providing a technical data base for personnel policy formulation have been an important component of many civil service reform operations. Development management projects have financed studies to examine current practices and reoa mnd future actions on: salary policy, personnel management information system creation and maintenance, general employment policy, civil service training, and career development for civil servants. A numboer of studies have focussed on important but poorly understood topics, such as the system of fringe benefits and bonuses in Senegal. Host of these studies were intended to produce action programs specifying the needed and feasible steps of further reform. It has been argued that studies have sometimes substituted for or delayed action, as in Bolivia and Senegal for example. The idea is that the Bank and governments use studies to avoid bringing a conflict, or potential conflict, to a damaging head. This cynicism is not entirely justified, however; studies have often produced serious thought about important and contentious personnel issues, including what is the appropriate and affordable role for state agencies, and what is the proper size and cost of the civil service. Studies have supplied recommendations that form the concrete basis for Bank-country dialogue on these topics. 2. Implementation Issues and Lessons in Development Hanasement 30 Although many of the activities discussed in the preceding section have just begun to be impl-emented, several issues meriting attention have already emerged. The discussion that follows highlights t4v key points; those raised most frequently by Bank staff working on these projects. One perhaps unavoidable but nonotheloss worrisome feature of these development management operations is the short-term perspective on which they are based; i.e., tho quite limited eztent to which they have begun to address the longer-torm management Issues. As noted, the bulk of supported activities assists SAL cost-containment messures. The emphasis on the short-term is understandable but costly. It often comes at the expense of attention to long- term strategy and the need to build enduring, sustainable management systems. In Ghana, for example, with the exception of preparing job inspectors to carry out tasks in support of the retronchment exercise, little training to build up basic ad4mnistrative skills of the civil service has yet taken place. In Guinea, the prime goal of TAL I was to reduce numbers in the civil service. So far, there has been little support for the resolution of longer-term, structural wage issues. A census of the civil service was performed, but there has not yet been a follow-up to build an iafomation management system to install modern, impartial personnel review and control practices. In the Gambia, the follow-up staff inspoction program which was to be carried out by trained Gambians under the supervision of ODA long-term technical advisors in the post-retrenchment period has ground to a halt, possibly due in part to this failure to internalize these activities in the earlier phases of the reform process, and to the lack of comitment to long-term administrative reform by both government and the Bank. As noted, in several instances computerized data management systems have been installed for personnel records and these systems have been linked to the payroll and the budget. This is an accomplishment. But the. absence of a strategic framework for ongoing civil service management is troublesome. In the Central African Republic, for example, concerns have been expressed about inadequate attention to factors essential to civil service performance improvementst career development, training, incentive structures, performance evaluation sywtems, etc. The fear is that unless such changes are implemented, even a smaller civil service will remain unproductive. The fundamental questions are the long-term onese What kind and size of civil service does the country need? How should lt plan to get there? What should be the ultimate purpose of the reviews of the various legal texts regarding the wage system, the indmiity system, the rights and responsibilities of civil servants over the long haul? The absence of a strategy has not always been due to the overriding nature of short term SAL demands or the assumption that cost containment is everywhere and always the first priority. In somw Lnstances it has been due to a deficit of vision or coherence in the reform program itself. In Jamaica, for example, the Administrative Reform Project consisted of a series of largely uncoordinated activities -- including revision of classification systems and pay surveys -- the strategic value of which was not clearly defined. 31 There have been cases in which aversion to strategic considerations was the product of conscious choice. In Bolivia, creating a pool of comparatively skilled and well paid civil servants, to fill key government posts, was accomplished -- without enormous difficulties -- through the Economic Management Support Operation (EHSO). But government interest in larger policy questions (or even studies) on such questions as overall pay and employment levels, or career development policies, was conspicuously absent. According to observers, gove-ment reticence could be traced to the political implications of personnel policy reform: the rationalization of civil service incentives and employment policies would eliminate political patronage opportunities which are alternatively distributed to the 'ins' and 'outs' during electoral transitions. There is thus no clear constituency for restructuring the system. In some cases, strategic policy concerns have been crowded out by components the usefulness of which is more iimediately apparent to governments. In Mauritania, for example, computerization of the personnel management system, which enjoys considerable governmant comsitment, has been proceeding on schedule. But a number of delays have taken place in the preparation of a series of policy studies. It may be that changes in administrative technology are perceived as less threatening politically than overt examination of existing policies (even though the application of the technology may render transparent abuses and fraud by exposing 'ghost' workers and double salaries, etc.) Zfforts in these operations have concentrated on reforming systems in the core ministries of finance and civil service. The costs and benefits of establishina clear linkages between central ministries and decentralized orians have not been well addressed. In the Jamaica Administrative Reform Project, for e-aple, decentralized personnel units were to be set up in line ministries to coordinate with the central Ministry of the Public Service. Nowever, training of sectoral ainistry staff never occurred, as the central ministry was loathe to relinquish control of personnel administration. The question of sapropriate technoloi' and the capacity of borrowers to internalize new management systems and procedures has been an issue in several civil service management operations. Ghana and Mauritania provide two contrasting experiences in this regard. In Mauritania, computerization of the personnel management system occurred through the application of a technical package that was developed through elaborate fine-tuning by Tunisian consultants, contracted through a twinning arrangement between their government and the Ministry of the Civil Service. Intensive interaction over a period of many months resulted in a system neatly tailored to the Mauritanian context. Mauritanians were carefully trained, and by the end of a year, were ready to take over the system. Bank staff report that the twinning approach worked so well in Mauritania because of the cultural affinity between the Tunisians and the Hauritanians; and because the Tunisians had just lnstalled a similar system in their own country. From the outset, this system was designed to be comprehensive, connecting the payroll and the personnel records of the Ministry of the Civil Service. The design of a comprehensive system grow out of a detailed diagnostic study undertaken as part of Bank sector work on public sector managemnt in Mauritania. Adequate preparation, detailed diagnosis, appropriate technology, sensitively delivered: these are the building blocks of success. 32 In Ghana, by contrast, the link between tha computerized payroll system - - completed early in the adjustment program because of its importance to demand management -- and a proposed computerized personnel management system came as en afterthought. Where the Tunisians had carefully crafted a package for the particular needs of the Mauritanians, the expatriate consultants working in Ghana proposed an off-the-shelf-technology, estimated to take only six months to install. The establishment of the linkage is, at time of writing, underway; and it may well work. Still, it seems likely that there vill be A tradeoff between speed of installation and the degree of internalization among Ghanaians. Tho simple and obvious lesson to be drawn from these contrasting experiences is that dilferent circumstances dictate different approaches. Hauritanis was ready from the outset for a comprehensive approach, in part because of earlier diagnostic work and in part becauso of the good will towsad the Tunisian consultants. Although successive changes in the MLinister of the Civil Service created discontinuities, the degree of consensus-building necessary amog various parts of government to win support for tho project was relatively modest. Ghas presented a much more fractious and fragmented contest, in which each step of the reform program had to be sold to a videly diverse constituency. It is unlikely, given the initial resistance to reform, that the Ghanaians would have 'bought" a comprehensive personnel system as part of a grand civil service reform strategy so early in the reform program. Indeed, a good deal of the work of the consultants in Ghana -- which overall has been performed quite successfully .- has been gradually to gain the confidence of their counterparts and thereby build support for the reform program. The relationshin betwega develonment mnaszement efforts in TALs and SALs is an issue raised by Bank staff. Many complained that SALs pushed these developDnt management operations too fast. Most estimated that the time officially allocated for activities vas about half of what was actually required for implement4tion. The other horn of the dileia is that TALs not linked to SAL conditionality were viewed as 'toothless,' and implementation of these freestanding projects was made that much more difficult. Thus, in civil service reform the Bank seems faced with a difficult trade-off between realistic, longer- term timetables and leverage. In a number of instances, the sequencint of reforms throuth TALs was haphazard. Sometime, a component could not proceed because an unrelated aspect of the TAL or the companion SAL would be delayed, thus holding up loan effectiveness or disbursement. This sometimes seriously threw the phasing of the civil service reform items off-balance. In some instances, sequencing of activities had to be altered because the time required to build consensus among key actors had been seriously underestimated. In other cases, the Bank's understanding of tho network of principal institutional players was inadequate. in at least one instance support for a component was aistakenly negotiated vith the wrong agency. The relationship between the Bank and other donors has figured importantly in development management operations. Other donors such as the UNJDTCD and USAID have been active on these issues in a number of countries in which the Bank is supporting civil service reform. (Guinea, Ghana, Bolivia and Uganda are examples). The lack of coordination or a clear division of labor among these 33 efforts has resulted in overlapping or conflicting agendas, where there should be complementarity. Lack of agreement about terms-of-reference for consultants, operating schedules and styles has presented problems -- with varying degrees of seriousness -- in Jamaica, Gambia and Ghana, for example. Delays in project impl ementation have ensued. The familiar lament that staff weeks for supervision were inadeLuate to imolementation requirements was particularly poignant for theso operations. This was true for Bank staff time and for expert consultancies. For some aspects of dewelopment management, such as the computerization of payroll and personnel records schemes, specialized skills have to be sought outside the Bank; but in several cases, resources were not available and therifore generalist Bank staff were forced to superviso -- often inadequately, thel, felt -- highly technical components. 3. Observations What the above diseussion of experience suggests is that this group of doevlopment managemnt operations constitute a first generation of projects whose objectives are, for the most part, tied to the overall agenda of SALs. As such, they have only begun to scratch the surface of what is required to construct (and sometimes reconstruct) well-managed and well-performing civil services. Indeed, for the ost part, Bank lnterventions in this area have left untouched a wide range of longer-term, structural issues whose resolution is essential to sustainable improvements in government administrative capacity. These include the reform and institutionalization of systems, procedures and, most importantly. incentive structures. This study shows that as the shorter-term reforms are played out, the need to affect these underlying systemic problems through second and third generations of projects is becoming more evident, as is their complexity. The uncertain and sometime apparently intractable nature of these issues has already felled one successor administrative reform project in Jamaica, where tho Bank is now pursuing a more narrow approach to iastitutional development through a financial management project to follow upon the first Administrative Reform Project. A key conclusion is that avoiding civil service management problems will not make them go away. Some staff argue that the inherently complex and uncertain nature of civil service reform places the field outside the Bank's comparative advantage. They argue, or hope, that on this issue the Bank should confine itself to assisting in the definition of economically rational policies; i.e., tho appropriate and affordable size of the wage bill. But it is wishful thinking to believe that one can separate the formulation of policy from its implementation. For prime example, the Bank cannot simply identify X thousand surplus personnel, and then assume that the totality of the job of removing them will be carried out by the government or a bilateral donor. The challenge for the Bank is to design projects which have measurable short-run cost-containment outputs, but to do so in the context of a strategy to solve the more fundamental management problems in the long run. 34 Part Ct Conclusions Despite the oft-mentioned data constraints affecting the subject and the findings of this paper, some lessons to guide present and future civil service reform are emerging. These can be summarized as follovs: * The impact of Bank grograms to contain the cost and size of civil services through emergency pay and employment reforms has so far boen negligible. Zfforts in mest countries to reduce the wage bill and to decrease the number of civLl service employees have yielded minimal results. Moreover, attempts to correct distortions in the structure of pay and employment through the decompression of wages and the rationalization of the remuneration system have had but limited success. This disappointing record suggests that reforms to date have been insufficiently ambitious in scope to bring about the degree of change that is needed. Meaningful change is going to require more forceful reforms. - The question of whether more aggressive reforms are feasible is partly a technical but mainly a political issue. As mentioned in earlier sections of this paper, the political *conomy of pay and employment reforms needs further conceptual and analytical work. Nonetheless, it is possible to hypothesize from the few *xamples of countries where programs have been carried out that the political costs of implementing pay and employment reforms have been lower than most governments (and perhaps even the donors) had anticipated. Organized opposition to reforms has not resulted in regime destabilization, and social upheaval as a result of dismissals has not occurred. In part, this may have been a function of the surprising capacity of private sector labor markets -- particularly In agricultural and informal sectors, and particularly, but not exclusively in Africa -- to absorb surplus government workers. It may also have been a function of the unexpectedly good handling of political factors, including the skill with which regimes generated supporting coalitions and managed contesting groups, for example. What this suggests is that.quite possibly regimes can (for political reasons) and must (for economic reasons) make deeper cuts. How far any given government can push these reforms is, of course, unknown. But the relatively mild consequences of the minimal reforms undertaken so far can, it is hoped, influence governments' Derceptions of political risk and encourage them to take bolder actions in the future. * Most of the middle-range employment reduction mechanisms such as voluntary departure schemes and early retiremnt programs may be useful and politically astute, when applied in combination with more stringent retrenchment measures. But they have not yet proven to be effective in reducing employment in any significant manner. Thus, they do not providE a substitute for biting the bullet through explicit dismissals. * Technical analysis and support activities such as functional reviews and competency testing, for example, have been useful in providing a rational 35 basis for cost-containment measures. Their major contribution, however, !ay be the symbolic assurance they provide that the reform process has been undertaken with fair and equitable intantions. * Retraining, redeployment, credit and public works programs for redundant employees have certainly had a utility, but one more symbolic and political than economic. From a financial and technical perspective such programs have had limited impact and have proved administratively difficult. * Som Bank supported reform programs have promoted interim solutions to pay and employmnt problems through specialized incentive schemos for topping up executive level salaries for key goverment posts, or, more broadly, by widely supplementing civil service salaries through donor financed activities. Most observers familiar with the use of these mechanisms recoaize thoir liitations and costs. The problem is that noither hard-pressed governments nor operational staff have alternative means as their disposal. Still, what must be recognized is that these salary supplement methods do not provido enduring answers to the fundamntal problems of civil service incentives; indeed, they ultimatsly undermine the likelihood of devising a durable solution. The Bank should not encourage or support such mechanisms in the absence of an action strategy for long-term structural reformas in government pay and employment policies. * Most Bank activities in civil service reform have concentrated, understandably, on the short-term cost-containment masures. Considerably more ephasis will have to be given to longer-term management issues if sustained improvee nt in government administrative capacity is to take place. The first generation of development management projects now underway have taken an important step in this direction. More attention needs to be paid to devising a coherent, over-arching strategy for civil service reform, and detailing the set of tactics by which the strategic goals will be achieved. * Technical Assistance Loans dealing with civil service management issues require more staff supervision time and more resources for specialized expertise than has previously been allocated. Most estimsates suggest that these factors need to be more or less doubled to ensure successful implementation. * Technical Assistance Loans in Development Management require more time to prepare and implement than do infrastructure projects. While this is well- recognized, they are too often short-changed in this regard by their dependence on both the scheduling and substantive requirements of structural adjustment lending. On the other hand, such TALs profit from the association with both the policy focus and the increased leverage that SALs provide. Without SALs, many civil service reforms in TALs have no 'teeth'. * Many of the institutional obstacles to carrying out vitally needed civil service reforms can seem overwhelming and insuperable, especially when confronting the 'long-haul' problems. Indeed, the mixed record so far on many of these issues could discourage further efforts. A frequently observed response to this state of affairs has been to decry the Bank's lack of 'comparative advantage' in this field. But a retreat from civil service management reform is tantamount to a denial of the crucial importance of government administrative capacity to implement economic and social programs. A more realistic approach 36 is to try to learn through trial and error haow to make such programs vorli better. 37 ?AALZ L "a OIUATONS gm CIVIL SUZYCt UiOiU COCIONKWIS Coanc La/Cr La/Cc Appbov I La/Cr TY" 3Ymber Date (MS3M) CAR TAL Cr 1150 81/05 4.00 CAR TAL Cr 582. 85/04 8.00 CAR SAL I Cr 1732 86/09 14.00 CAR SAL 21 Cr 1916 88/06 40.00 CAR teao Mt Cr 1971 J8/12 13.20 proj ect MaL TAL Cr 1307 82/12 10.40 Uganda TAL 11 Cr 1434 83/12 14.20 Uganda a1 Cr 1844 87/09 65.00 Gulma TAL Cr 1259 83/03 9.50 CuL±ea SAL I Cr 1659 86/02 25.00 GuLnea SAL 11 Cr 1926 88/06 65.00 Cutnea I ICo Mgt Cc 1963 88/11 14.50 SuppOrt MauriLania SAL Cr 1812 87/06 15.00 Mauritania Dev Mgt Cr 1865 87/12 210.00 TolO SAL 11 Cr 1599 85/05 27.80 Togo TAL III Cr 1600 85/05 6.20 Malawi SAL IZI Cr 16W. 85/12 30.00 Kalawi 1D Cr 2036 89/06 11.30 Proj *ct Niger SAL Cr 1660 86/02 20.00 S'niga1 SAL 11 Cr 1656 86/02 20.00 Snegs.L SAL III Cr 1802 87/05 45.00 S4nfegal D0VC Mgt Cr 1910 88/05 17.00 Gaabia SAL I Cr 1730 86/08 5.00 Gambia SAL 1I Cr 2032 89/06 23.00 Ghana Pub Ada 81/12 Ghana SAL I Cr 1777 87/04 34.00 Chan SA Inat Cr 1778 87/04 10.80 Support Chna SAL 11 Cr 20015 89/04 120.00 Guinea Bissau SAL I Cr 1798 87/05 10.00 GUL3a±naissau TAL 11 Cc 1935 88/06 9.70 Guinea BLssau SAL 1S Cr 2019 89/05 23.40 Sud4n Pt & tEo Cr 1789 87/05 9.00 Mgt ProjOct Zaire SAL Cr 1831 87/06 55.00 Zaire ECaO gc Cr 1832 87/06 12.00 & ID 38 TABL 2 LAKOPMAT'IONS TI= CTT fyTC! COIEFNM.WS ST t ING ISTtS A o UCGIOE SAL& TALS TOTAL Africa 23 18 41 LAC 7 4 11 EMM 4 4 Asia 4 1 Total 38 23 61 TABLE 3 C tWOIBS 0 CIVIL uaVICg BUEZH .. le .allecisanl salacg Imes. .aladcalos adlege i Seghe e gele Wage Slkgauie Seal VaIual of dossr-a- Cscs,s iCb_Iaag SeadleeSuagwelUead eeiasg .1 leg Ugan tl Cidig. PseClag- U-se.sel s 1- IneSs.i_ CS Ceases UlSIWeca Ian#60* EBe_ -m. * gegema ageD 1 S-clSng r essl ram sltudes Umwgsmal Co. so refa. j Cutulag u.. easges If£saBevmee C4MW6&*om lezlee.. s.ngladoeb SIC CAB SAL Ba4ngldea IPA Bangladeh ItC Bolivia Bolivia SIC 8eAgiedeeb ft 64n4gsdol g Sr CAM SAL I CAM SAL Bagladash IPA CAB SAL CAR SAL CAB S&L CAR SAL Gaml SAL CM TAL It Malawi& C.e&a t& SL CAa VAL 1 CAR VAL a I Coasa Bitc SAL Cho" tAL CGhna SAL CAB SAL Ga-m SAL Cases, Ue- SAL Cmii. S&L GuaAs& VAL C"eae SAL Ghana TAL Cases Bite VAL t ha. SAL Gamble SAL C &aC SAL Bagel SAL Jsica SAL 1I Hale VAL GJmble S&L fbs" VAL Ghana SAL "han TAL JamaSic PA #&a&de Sc Hauscanas SAL Gb... SAL G".Aoa SAL CSha tAL GC.le.. SAL Jamaics SAL Sit USand VAL SI Hauaslgao SAL Gb.& tAL CG p." SAL Seal.. SiAL J_amIc. II Haeigaaia Pt C_-g. SAL Siamene SAL Guls. SAL 1.lg. SAL GuVa_ SAL Jamaica SAL il Hlaug&anA TAL Semeg&& SAL III Senegal TAL Nola& SAL Baig& SAL Badgl. SAL Hall tSL Bigot SAL WoHa&asiana SAL Haal& V AL Haseiag&eal SA HatirSaala SAL Hauglgeala SAL race SAL Del vi. Janaice PA Mitc SAL Sesegal SAL At regu SAL Sene&ga SAL All G"ealne aa.. SAL Jamect- SAL illg geegt SAL IS "la" Bisau SAL Sosagal tAL Slt Lank. U10 H"agtaenle SAL Susagep SAL 11 Laos SAL S-eg tAL III %0 tkastad SAL I ^aealgaa sism t e tekep SAL illt Comeae SAL tuSgeh SAL V Iailand SAL II Kiget SAL Ug"an EBC Ceag SAL Uganda VAL SO waga VAL fit lPet. TAL Ceage, SAL Bealas SAL Senegal SAL Gaboo SAL Siege& Leea" SAL Sea Tos 6 thailand SAL I luln0alpe SAL Shalsand SALII. Cmwome a. SAL Soo. SAL III Gaben SA Uganda £uBC Cae.eWasBt SAL See. Soms, a epal SAL rilecipo SAL Moal. SAL comag... SA Conge SAL Gab.., SAL Goimea Bissau SAL TSeanial SAL us"SPA- b Z 9 t Basinaa SAL Benlanc SAL Coage SAL lela.o SAL - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - lable I. ilLMOiWgu OL&OCTION a6ceaMuaa F SUSgCTEB C4MIMIUI I ------------- vue -t wWImAav iU0'LOUS couuia ca"I, auswaa" muI"C"L"iu: ""m I IEuua I ~ uma. Gb. ii I. l131 1 l1t - 5 1 C"-~~26 toX 4a} l@l *2 . 55 lb lip Due J24S 54459 Ic Ou CA iS9 %d lin se- ee4 Ugaad MO aSa carnage.. MM ,.M" 1 to tbs. ldipac boa bass iAGaMd bi Gvecamm.at but &&1 vwlldosed to sochileal amaolVal 5b eaeos s£ Lb. b.a is Gualm"| Caaaabc I. Osh' Nacbaos.eft. la Lb. ease .1 G.lessa Se e Lg . pbe Ca&." ba_b a" eke scale lead.ol Isg.s eke *9460-Sew eaul.apsaa.. d4 ftlsea so elaae.as ol- she &"s. *1 dolea. pa,maaia to l -eloss so 2bos. e4-asagaed Isoessce *.wie aod lv1uaasi. of Gb.... Cavil heiwic. Salem Uega. - sIn9. J. Wage "via. i*djuas.s- Ae *IIetesu5 at 4doe lest. de so leecilos publq.-S'. Guido *. -.egd 4-Cau1soA. l*^osevawi wish cemsgv esol4 "An abile &S glos rap &ad Lole"We.'. tlose I siudp. tlb Go"bl AdrAlesa.5.avo seom lftea..a. rJ-. Capon 41 rAaLE 5 . . 5[ n ~~~SYAIIC! PdR"U!A1 rct 5rtICT!D COCNT1!!S Ghana 4 inonths basic salary £nd of Service 2 months basic salary on avard termination cimos compleced years of service Guinoa Bissau 1 years salary paid on a monthly basis C.nCraJ African 40 monchs salary and all Republic cuualated employee contribution cowards Che pens on fund AoS 1 year's salary _ . Cu__- - Grade rolaced aaouncs Aaounc Index 400000 207.312 450000 332-560 500000 580-773 550000 7874!050 600000 1129- 347 650000 14064'525 700000 16044L980 750000 2139-2614 800000 Abovo 2164 Sources: Councry Scaff & Recul des Texces Scacucaires do la Fonc.ion PubLI4ue Culneonno 42 TABLE 6 C0) =1Tf!w 01 C]3TL SflYTC! Rl'N.TTom FOR I!eDcommL'its Country Year Benefits as a of Salary as % of cataL compensacion eoCal couponsacion Gambia 1982 11.93 88.07 1983 12.24 87.76 1984 12.63 87.37 1986 15.65 64.35 1987 20.57 79.43 1988 21.88. 78.12 Sonegal 1980/85 25.37 74.63 1988 44.86 55.14 1989 43.00 57.00 __CA&u--------- --i984 38.00 62.00 1985 40.00 60.00 1986 43.00 57.00 1987 41.00 59.00 Cameroon 1967 19.00 81.00. 1987/88 19.00 81.00 Bolivia 1982 70.00 30.00 Laos 90.00 10.00 Sources: Country staff, Scaff estimates, Coopers & Lybrand report 'Efficacsie de La Fonceion Fublique Pro5raiwe d. Reduction" (Senegal). Daca provided by the CGablan authorities. 43 Coemtr La/Cr La/Cr AprFvOl La/Cr TY" Nuber 3"eo Amount (US Sa) Sao Tome & Principe SAL Cr 1825 87/06 4.00 Congo SAL Ln 2866 87/06 70.00 Gabon SAL Ln 2933 88/04 50.00 Gabon TAL Ln 3114 89/08 5.00 Benin SAL Cr 2023 89/06 45.00 Cameroon Eco Mgt Ln 3110 89,07 9.00 Proj ect Carnroon SAL Ln 2866 89/07 70.00 Thailand SAL I Ln 2097 82/03 104.50 Thailand SAL 11 Ln 2256 83/03 175.50 Bangladesh PA Cr 1349 83/04 12.00 Bangladesh IPC 13 Cr 1655 86/02 200.00 Laos SAL Cc 2037 89/06 40.00 Tuzkey SAL 1S Ln 1987 81/05 300.00 Turkey SAL IV Ln 2321 83/06 300.80 Turkey SAL V Ln 2441 84/06 376.00 Tunisia SAL Lu 2962 88/06 150.00 Guyana SAL Ln 1948 81/02 14.00 Peru PS Ln 2204 82/09 1O.20 Jamaica SAL U1 Lu 2315 83/06 60.20 Jam&Lca PA La 2423 84/05 4.50 Jamaica SAL III Ln 2478 84/11 55.00 Cosca Rica SAL I Ln 2518 85/04 80.00 Cosca Rica TAL Ln 2519 85/04 3.50 Cosca Rica SAL rr La 3005 88/12 100.00 Bolivia RIC Cr 1703 86/05 55.00 Bolivia Eco M5t Cr 1977 88/12 9.70 scrcnpkleuing Dominica SAL Cr 1817 87/06 3.00 44 TABLE 7 VAC COP~S!ION AA 0OS 101 IK.C12D ccula=!! Country Year Co.pecssion RLtio Gambia 1985 8.5 I 1988 5.7 I CAR 1985 9.4' 1 1.968 9. O . Mauritania 1975 6.9 1 1985 3.0 1 GuLnea 1985 8.7 1 1987 3.8 l 1988 '.5 1 1989 4.6 1 CuiLnea BLsau 19J8 5.2 1 1969 X.O 1 Laos ft.-1988 3.0 1 1988 6.6 1 Ghana 19Jk 5.7 1 1989 7.8 1. Kali 1.95 16.5 ; 1LSer 1975 18.2 1 1985 14.8 I Senegal l19J0 7.8 1982 7.3 L 1983 6.8 1 1965 6.5 I Togo 1985 12. G 1 Ugana 1983/84 5.9 I Zaire 1.98 47.2 1 Burundi 1984 17.0 1 Caaeroon 1989 22.0 I Sources: Councry scAf. Dep-rcamenco de Inforuacics (Guinea oLssau) and dacs provided by the GambLan aucthoriCieos. Preiminry daca from OfRobinson: CLvil service pay in Africa (Geneva, ILO. forthcomLng) 99 6* 9w* 61 93 69 19 196**@v@* £5 99 90 (3)@"* -U 1ov*v**v**@@v@v@@@*@vX 990J. 13 "?c cc' rr'g go6 if' (99 LI 99 911).a:999 WsZ re' cc 1 of 6 cc It I t of &co "I* fet 01 I IK ¢ e¢ s - tl i6 s or It W eCV trsc IC- * e to 4 . *e %t go,9 cos e w*apnp c sDe *I D wc It .ei ot Se II : I t 'st t ) SaSi2 79sealm .4 DC oe, os os , ¢s 06 t cc nt vs 96 *2 Li tt I8s (s) *RMId _ alms"2 rc t c at" 90£: 6 t9 w06 *Ie t g to iIo' 339O (3) SaRSM dsu awa89*19 o09c 1oz OCICCOG 4139e 996t 36t1 .909 ct9?1 esMo 313 oeS I " teotl co906c 00t 69' oo o6ecot oe,"0ot o090I 3 ntos eI t De I UoVbe Ito9 901 rIt9"S ip-, ' e:occ Soe"C nes I ITo?et 1 0oe9 to 611 o o0999 0s I Oro 906,1 o . W :Cseu. D 00191 0016911 go seo1 WfOC'm s,0L K01 's "It 9001.06 90919!t "t"7'3n VW 9S (S s smin7Tmm& "m3 in' as" wm S"SoJitU" wms7v19mn ~wwej, IgYg I ts) tsd Twsez/ goau" PR 0t (2) in5U1 WS*JPI9 Ci) S __31 s UUS5mPgt C:) i8s_37 Stn:,I i tSinai vc Fe " l 0w CZ) 3ar#ftilped 7e90313 (3) aafiI7WS993. aws'maaY,3 OCTITO 06£ 914 f1 406fl£ em*.**3 79053 99es 901931 D9 0( 999t 90fls ua9U1t 3 _ *. 0ge' % 06166= 06619 06,992 66391i SMS, P I TSg.aYL (Sinaj va cS "-771s6) * w! 1! 19 91 19901 96691 SPY?I 341.? £31 (3) else"" 7"OSIm 69~~ ~ t 996£ 39£* 999 *s9 uU esMW es~W '*33l Di If01 9l? 999g 92-t 69n etI" I92) * ago"S T"9,191 go 9 Its I I 591i 99-91 9V61 0! 999its) *yeamm"oft n",ei3 916 39 9919 99t9 t 9?G9 391 1 Ct) U."8 9sX 955a8W`11 flet 9 i LI. 0109 9o9o @91n 0691ot 9 lei ". .St 790.3 oil Goli IS Itc os Itoro T-or ut) um,? _o& umm '3 999 0ll6t ",19 9919 SMISMASSS "Sol3 11 399 039 13 0051i'g I' &91 99 ITW¶P& 34M' 39 9 A 9s'9 966 96 00 999If 999 r9 *S'i&A9 VW '955 1 ~ o 6! 99'!? or9s 069!g 096!S 09t 0_1 *_1 sS 989 t 0~~~~~~2 t-@ Cgi MM U" VWtt} _S (138d WO U* "'lUll) in CZ) -1 1c 1.1* 79104,131 69C 3901 37I 99,01 03,9 139 lil$ oil$9~ CZ) 9S8Ra7"2 muts"31 6; ic eto e6 Al or," tCIP *t16 99" n Ju tv CZ t) a-_-am nmZt CIsC 909, I s~ 9199 969 £396 9999 9090 £936 C) SI&OWIves" 79903131K (!) *89937f3 *n8'uO oc _ D9 04'~ 09531 993930 00£193l1 *£99?IT 0619071 Wl£161t 099690(1 909599 08"107vow ?910 ?39969 6901 £ ts0303 01399£ 61169 90999 SC'Dv s 67196C' I 1O9010 too 91) 597," V 79"1 (s l79 7ns "nS cw f ) "pro r ' p*vi I .......................... ,......................f.. ....e.......... 936 5361 *e061 s J 9`94I 9 "31 199 n 941 . . . . a~~~~~~~~.. ............ a..ee......~aa.n nru ~ ~ ~ ~ s iniau u. ina"ons .~~~~~~~~ 3s)UVto.L 46 ... .. ................................................................................................................ 106 &9 M I1 L343 IN94.4 293 2. 940 t 97 944 W s^^Z. _" Waste :61 a? 3o o02 23e2 33 So 14.4e 9 .o it 1 4696in" am SWO_s 4.2o 8. o 9.10 L1.34 6.1o to-La Lo.o- .9 3o ,,s C .as _ u"iae *0e o .. 4.50 . S 5 o U.0 54.20 03.70 02.0 o: la o&L. 1*:'a. 92.7 92.2 tl 2.09.0 .10.4 2.43.4 M.. 2? a ?e ta.L Ugmmi.4 4 *9.9 4 4. 0 9 £9.2 63.0 ~~&8/C.aaw 3~1S~agw (1) 01.88 44.34 486 e39 43.9 613 3.2.4 $4.64 U8ICartam g lSht Sw (3) 30.43 25.34 .9.3 2U. " 2..4 13 1 2.0 .683 216A "SUIT" Lu&sgn (2 2 .47 29.41 27.71 2.32 203 23 93 204.6 24.40 uSITesl _1ti"oaw (22 6.73 9.32 6.3 9.*.n S.08 4.36 7 to0 2.4 WU&8S"t RGI4 (2) 32.4 24.71 sWAS 52.62 50.36 40.04 44.7 .5.54 NOUIteSal 6m (2) 7.67.0 2..2 20.3 2.5.2 L1.6.1 20.70 t2.7., U.2.7 SWege (Slum of CIA From" i UItes _" salauLes. 2*. 2U8.0 36.20 30.30 13 so co"ad 4wrt.66 42.4 46.20 147. 0 &9 '0 32 30 Ca.ain tue 70,.10 75.00 702. '6.20 9 It - Teal.' Z~~~~~~~.aiLu~~~~~ae. 2.~~~~23460 2.37.9So .24.64 2.3t.60 '.39 .0 Total Iin.e.. 09.90 " 742.0 75.70 4.30zo 72 :20 I O.SICMufuma 32.s~ax. (1) 33.32 24.64 27.2.9 34.75 .2.3' I 481C.aavms bp..e1tia,. (1^ 40.7? o 1.44 2. 43.2.7 *. t o- : VA4?.saL g_..gLp.ag (22 20.13 2.-?7 30.99 " U.a99 Z#ZZ G48UITSaIA ea4*at (2) 34.4 32.36 33.33 37 6 *" 4 s0 1ST*%a1 . e,.magm) 3*. 3 34.63 3736 2 ." 3 14.29 &81?eU a.3 3 ma ) . 6.9 43.38 43.28 72. '3 73.39 14m61 (StlamLa .1 CIA Froesa) we.. MA SaLinas. 78.30 62.30 92.70 2.00.44 2.04.64 LIL..60 t.19.60 2.22.10 2.23 . coo"i _ se qaes 3466 36.30 36.90 3tO 4 32.30 be.0. Cm""% 3nq.LSwm 272.77 2.79.1 292.23 .19654 2'I.2: reaMA gN13wt6 217.11 220.51 232.02 2.' 04 2.I 2. Total. * a 223.34 2.31.0 .7e 7 0 S440 2303.2 218.79 331.23 211 2 05 2 :. W4S/CuvW1G IMuiWLeo (2) 01.70 02.'4 *2.J0 01.40 50 .0 QUICMWOSU U__esue (2) V4IT8Sa toMitue (m) 9.10 3-4.70 51.30 50 co s0 . CASTsta,& ZotA.aua e WUiTstal I.v.mmm 52.30 5WIG0 .7 70 .640 .1 to SIT*CaaL les.om (3) roee (52.2.2.&o. of CIA ?CMOs) Wa5gg i4 Sta.LaL 26.70 37.60 37.6 26.40 34 2a 3 N4t.244*.e lAR & 42. 6.la 940 G .o . 00 2. 70 2220 17 .0I Cj svas ' Z poe"C.u 54. 54 2L0 56.40 43.10 72 30 '2 20 regal capmAlcs VWSCwgWmt 11 in.dIatmes (C!) 53.74 49.38 40.47 .4 47 50 s 9 2.7 MSSICwav 1_oitt.#o (2) *.'9 2.7..1 23.23 24 .06 29.27 Z. .7 W&SIT1ec&.& io.i&%gies (2) Res/teol, L _Z"Law. (z2 W&SIt?09 ROWmm (32 U40 MLLLo a lod " sLa^ W"eeg 4" 1alaelg 04.0 70.00 L2.2.00 444.20 533.06 2.239.00 3700.00 coo me sernise C.au.S 1awinLt9b 306.30 334.00 095.30 2.340 3314.46714.00 30744.00 Trsl £oagLuiss 5e33.3 720.9 2134.3 2391.0 4723.0 10931.0 .3049.00 toGal lews 243.9 524.2 929.4 2.420.9 2843.5 56704. 24670.00 wasi Cwfggg _o_L.uao (:2 30.77 2.3.00 M2. 26.07 .3.3 2. 0.7 L'.95 C"IChues tMomiaturv M WASITsA. Iasou"Lstes (S) Lt.1 9.72 L2.92 n 2..34 21.33 l.0 1.10 6"ltUsA& 5 _O atue.s (1* WASIT41464 3.vmme (XI 34.34 2.3.31 23.3.5 37.2.6 Is.56 U 29.74 2.4.1 04/Total. Uwawma (22 ...................................................................................................................... S sea *Uwt.u a" yrnL.AtLOa of -ia C.v-L S t.- S ale= FT.6g. it.s? 1949. :. &a Oawtas. qtLacte. 4i PLan6e. C4 e, a , _ Ls.IM _ G ah ILas.. ). aA Semik sclt *oeuao . OSts PU (Too. 0 t..aa. Scacsta.c 9at Itl44 *t JmAWN.. MILatuy Pm P2ag 4" t t SAI UmiA. _ *l. Oazta pt4do by tiC^o IL"@ege - 4sLaamS6LSLOO. 47 OUWSSSY :990 L S L Ut~~I% im L943 Li" Lt s ;9" ;§|t ;M ..................................................................................................................... ouu~~~t 930 2.961 2.953 1.962 2.964 1.953 1.9~~~~~~~~~64 .90s7s 1.868-. i'e*a u^c $4S-Laa1 3.'O 3.79 * 2' * 50 .6 20 2:3.0 o 1 *0 S.u~~~~~~~.e.a~~~~~^ .5.6 53.60 37 ': ?e.a2n Lww A.slAx *.37 70 385.90 :2' *o ea" tqswne t7 *0 32. 30 3.9 : w93/C6vvm%s £spemfLtut. (23 20.99 3t 9. 30 r7 44SICorro, Lo c:Atu 3 1t3. 2 3 8.30 l .1s W4SIT?%s,L t .4eLz1ue (33 '333-2 I 346 Zs &Si?esa2 ;I$/ttst1,3 22 23.24 27 .84 :. 56 ! w usai Rawini. 3 . 23.39 17.64 29 so I 8/?sTat lrves" (U) 58.57 3. 99 25 .1 02a :h (MUI.LL3. of CudLs Wj gea -- S3alaia 3237 2279 2501 52a53 2.952 26194 33920 .4'4^ Code& WA SereM.aaa 23 38449 '796 21449 25ss59 k23.2 60531 Co&*=&. 1a2ee4Ltuse 92.79 6330 6403 2.3901 35326 254932 60534 60553 31,11.0 T! S&L tZmaLsu.a 9464 7739 9530 o .975S 20090 52124 95075 '.2832 1.922. ewaL Daw.esi 2953 3~~~~~~ ~~~234b Lot"2.36 31727 34693 40979 2.05009 3..Z333 j 0161 Aawoo 1§sl 12 404 0l X2 |§ *t§ :so ;23 I 4S/C"aVOR% tZapu Ltuie (3 .9464 26. 35.4, 22.69 37.59 '3.06 "6.3I .. 54 44SiC.uWOeR 1AL2..a. (23 53.91 61.. 55.71 0.00 67.07 79.00 69.27 75 .3 : :0 I4S1s%e. LiVofLausia (3) 36.35 22. L6.92 37.58 27.55 26.71 25 is5 2 624 e4IT*sa& 1g.s",ttue c(U b4.26 M0I 50.33 0.00 52.05 $59.29 3.97 *3.35 3.30 w&SIts44 RWO <) 36.57 97,3 39.54 29.21 37.59 27.35 39 .2 3&,7g U48Tes.4 3evwm (23 74.35 lt." to. 0.o0 72.00 72.91 6. 57 9* 0 00 ;sL4 (TkmAas_ Of 4 . W o "A_ SaLaLea 53775 42919 "6"9 63271 r. 19 '6357 *0 WA S- 64Srgigs 36971 23280 34017 4084 M403 40L21 ?.24 CaIe s L"OSwd&v.e L39363 L2.53 I9' 35371 379954 1046&1 e&& Ivow"e 95730 753e0 57050 703100 =22970 t76040 215390 29'420 &SI/Cae.was seinSLuuru (2) .0.73 92.31 34.59 29.75 39 2' .9 3 UCmeC eas taZe ticute (2) 1.9.25 L2.67 2.34 22.95 L25.2 :- 84 iYU/tesa teainiuue (22 0&si?eal. Lup.a4twat (23 d9siSCAL l.vseonu(z) 603 63.01 23.35 29.21 IL $1 3.71TOC61 OR4 veuta ) 23.69 21.61 21.26 .9 .6 .9 so _Z."a &Las&" (2 of =1t3 J.6-s And4 S4 LeOS :.600 .9.7O 33.20 U2.30 II."9 6.70 1 s0 5 30 cga ADA S.wSi.c. 6.60 6.00 t 00 64.0 6.30 5 .0 5 .0 9 50 :4.,*gn9 taw.gOrAuCC 26.90 26.50 9.20 23.640 22.20 LI7 20 36 60 :2 l0 Ozuai, g nmtL".Wae 34.60 57.30 '9.50 19.20 59.00 O4.70 '24.LZ 52.30 .Ual lewepau. 27.20 21.0 26.20 21.30 23.04 2'.'4 32..0 39 l0 ! 4SIC48VOftI CAVGMLtUg* (sa 59.98 55..? 54.77 54.94 53.60 .5. 39 53 239 *. USI/Coesent LuVmWAkituae (2) 29.59 20.2.9 29.5 23.32 237.i4 219.7 32.2.9 36 4. V&SItOCAL 1fltoi.Ltuu (2) 39.30 Z3.5.7 26.67 20.73 20.27 21.35 12.2.4 30.3 ;I/?S*IT taa Lture (2) U2.09 3'.01 36.39 30.30 20.34 2..02 3L.23 12.i.. .i9SIate&L 2eveo(1) 58.82 4,636.9 02. 9 39.63 3 5.64 Li-.22 3.39S US/Tcas. Zeave (23 29.26 2U.94 26.62 39.37 2.8.26 23.72 2.9. L7 : A7 W4ea WA S_ A5sCLO 947.0 a 107.0 3062.0 L260.0 a 335.' 3503.5 z am*s Serwf... 2M1.9 '38.5 '90.3 419.2 ?46.9 *99.5 C..erez tlod"LcJrt 2394.0 2317.9 3233.0 2755.0 9L26 3 5134 ; Total gawyi3uu.e 3199.0 3335.3 393L.A .71.9.6 5503.6 1555 4 ?.saL Rvegww 3899.9 2339.1 2703.3 3200.5 W1LS1C.Aa.eu ew.ogLtg. (1) .0.39 34.04 33.29 33 54 23 8 22 ; C&5CIAgefoi Ia t rLtZ.ug (2) It.?$ 13.21 25.05 36.99 27 7I .5 .7 I SITeuah t pnd&ust (23 U2.33 123.3 27.53 26.S0 29.2. 2i 90 GCSI?T.aL ZapefCLcume (2) 9 *2 3t25 3.97 3.3.2 23.52 . 4 si?@SatJo 3.qgw.(2) 32.93 95.92 29.94 39.05 0&5I?t.sAh RfteWA (SU 13.24 13.34 2.8S. 14.73 .................................................................... .......................................... 48 TABLE 9 ~~~~~~~~~~~~~~................. ....................... .... ......................... .. .... .... ._........................................................... ................................................... II tGor 31n4 32X&0 32643 }s33X t c* ... ^t^ . ;.aboo L^L^eo 700 1: 430 :66330 4200 137300 isso00 Ch ~~~~~~~~~~~~~~~~~~~~~~~~~~~~300727 2844X: 1282|2 $ez@%&E §~~~~~~39t 63373 $7294 70249 "93 t0047 '1163 69407 69X43 * ts- 1X0X00~~~tioLo tl000d 11"Go L02000 t00s00 ZS0>a 79N00 72800 74400 c _~~~~~~~~~~~~~~~~~7S 14e20o 16295 4X L72600 191740 ;4150 L79110 C@ ~~~~~ ~ ~~~~~~~~~~~~~~~~~~~~~~~71364 ?1:80 C_i LX~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1209 :.26 U tS9000 t~~~~~~~~~~~~~~~~~~~39528 i&ti~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~62 23t1:03 CAA 21943 222XX 22791 1..90 .* ;a 3~~~~~~~~~~~4954 a8103 s. La9 |~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ti 3z 11 13a0 l &.2 Xtux 50§24 *;^sJ 5~~~~ ~~~~~~~~~~~~~~~~0970 52333 I2541 ........... ....................... I--s_ . ...................... commax SCAM CM£. . . S9&GLSJL Is titute ag j"a a. StCAI *ttz@Cd "S 4&z PN&A bT C 14CION &MUMELCO 4 Ilt s8X6 Mt *g wdia of Chasms Ct.lvtt Sorwt_ LOW - L9|§, * J. tots 0ata 49 Annex S: Sox" CIVIL SERVICE COSt CONAIMENt Containing :he concral governmencts wage bill ho- been a prar-y objeccive of civil service reform in che Central Aft tan Repubilc (CAR) and Chan&. :n both countries reform has been supported by structural adjustment lending and technical assistance from the World Bank Group. In the CAR, where the vage bill absorbs over 60% of domestic revenues, initial seeps focussed on disposing of Ighostm workers and then limiting the tocal numbors employed. However, this approach did nee vork well because agencies replaced low-paid workers by roc:uicing isaff ac higher levels, :c was then decided co Impose a limit on the wage bill In 1987. with further reductions ln 1988 and 1989. This was I to be done mainly by suspending regular annual incromencs. The wage bill has, In facc, declined only slighcly in each of the first tvo years (1987-1988). Atention is now also boginning to focus once more on reducing the size of tho civil service. Reform in Ghana began with tho eliminaicon of what the governoent has claimed to to be some 11000 'ghosts'from the government payroll. managemont appraisalsO were then undertaken by teams of officials from che ministries under review to identify categories of workers that vere overstaffed, without reference co individuals. The appraisals concentrated 4n the 30.40 lowest pay grades (out of 130) on che assumption that chis was where the main problem lay. and recommendations were made as co che number of posicions to be abolished. The recommendacions were reviewed bv a committee of sonior civil servants which decided how many, and who, should be recrenchod. Ministers chen reviewed the circu-sances of each one of che individuals recommended for cermination, taking ineo account noc only thair performance, but their years of service and number of dependencs, as well as whether ochor membors of the family were governoenc employees and were also being ctrminated. Only after this scrutiny were people actually dismissed. The noc result has been a reduction of contral governmenc employmenc In che range of 30,000 plus over four years. However, chis has noc yet led co anv budgecary savings because there has been a general (and badly needed) increase in civil service pay adjusted for inflation, cogecher wich a slighc widening of differentials betoven the highest and lowest paid, and because savings from staff reduccions have beon more chan offset in the first chroe years by che coscs of componsation. The racto of the central government wage bill co GOP (around 5%) has in facc Incroased slightly. 50 &.L civil servancs in the CAR were given che option of Leavin% govrrnment emplovuenc with A severance paymenc of 40 moncns' curr-n: salary, plus any personal concribucions they had made to che pension plan (to date these payments have averaged over US$10,000 per head). AS of Augusc 1989, about l.500 peoplo had applied for voluntary retiremont under chis scheo, and 300 had accually left che service. IMenwhile, plans for che dismissal of redundant employoes are procoeding very slowly. Those noe desinated as essential by their agoncy enter a cencral personnel pool, and may be hired by ocher agencies if they have appropriate qualifications. Those noc selected In either round are supposod to be dismissed from tho service on terms much Less favorable than for voluntary retirement, but wich some assiscance in obtaining alcornacive employment and access to credit to sec up in business or farming. However, no dismissals have yet taken place. In Ghana each dismissed civil servant receives four monchs base salay, plus cwo maonchs' pay for each year of service, and the average severance package for chose teorminaced through mid-1989 has boon estimated cc the equivalenc of 2.94 years of pay (1.96 years for coachers). The support program for chose who are re-deployed includes information, counselling and placement secvioes, training opportunities at existing vocational and tcechnical training centers and assistance ln agricultural production through provision of land, inputs, extension services and transicional food aid. Contact Date fcr naerr WPS416 World 9 ank Efforts at Povony Paui Glwwo Meaurement in he Third World: The Living Standards Massuremont Study WPS417 Modoing the Maiconombc W&" Esatrly May 1990 RF LUz Requiroments Picy Reform E. C. Hw 34303 Ptyabha Kongiamut Jan 3Zek WPS418 Does Devawation Hurt Pnrvate Ajay Chhber May 1990 M. Colin: Investment? Tho Indonsian Case Nemal Shafik 33490 WPS419 The Design and Sequencing of Btian Levy may 1990 S. Levy Trade and Investmo nt Policy Retorm: 37488 An Insfitutional Analys WPSA420 Meing Bank Imgaion Inveastments Gerald T. O0 May 199 C. Spoonur More Sustainable 30484 WPY421 Taxaton of FP dal Intermediaon: Christo0o Chamley may 1990 W. Pitayato. Mearurwment PvIocIe and Parick Honohan nakan Appecato to Five Afrcan Countries 37666 WPS42 CMil Sevn Reform and the Word Barbar Nuneg May 1 990 R. Macolm 1ank John Nedlis 37495 WPS423 Rlativ Price Changes and the M. Shahbaz Khan May 1990 WOR Cffice Growth of the Public Sctor 31393 WPS424 Mxico's Extenai Oebt Sweder van Wiinbergen Restructuring in 1989-1990: An Economic Analysis WPS425 The Earmiariting of Govrmmet William A. McCeary Reenues in Coloma Evamana Urbe Tobon WPS42G Growth.Orintnd Adjustment Riccao Faini - PrPgrams: A Statl Analysis Jaimo de Mob Abdei S.nhadjp-Senbii Julie Stanton WPS427 Exchange Reform, Parallel Markets Ajay Chhibber and Inflaton in Africa: The Case Nemat Shafik of Ghana WPS428 Pwrestroyka and lls Implications S" Satas May 1990 N. Camooell for Euron Socialist Countnes 33769 pPP Wnrttin P^.r S-,,g Contact WPS482 The GArr as International DisOpline J. Michael Finger Marel 1990 N. Arfts Over Trade Restrictions: A Public 38010 Choce Approach WPS403 Innovatvo Agricultura Extension S. Tjip Walker for Women: A Cas Study ot Caeo wn WPS404 Chile's LoMarketsin an Eraot Luis A.Riven Apnl 1990 R Luz Adjustv:it 3.4303 WPS40S Invstments in Solid Wate Ca1 Bana April 1990 S. Cumine Managment Opportunitis for Janb Bernstein 3735 Enmwonmeal Iprr vwent Frederick Wright WPS406 Townshi, Village, and Private Wilam Byrd April 1990 K Ch,en Industi rn China's Economic Alan GCib 38986 Reform WPS407 Puble Enterpris Relom: Ahied Gaal Apol 1990 G. Or. acaTetleh A Challngeor twe Wrl Bank 3764- WPS40 Seha Isum in Evaluating SI es y 1990 S. King-Watson Debt-Reducing Del ishac 01w 33730 WPS409 Fwanadl Policy and Corporate Manoor Daarmi Apnl 1990 U. Raggaroi lvtment in imperiect Capitai 37657 Markts: The Cas d Korea WPS410 The Cost of Capit and Investment Alan Auerbach Apni 1990 A. Bhaila in Developing Counries 37699 WPS41 Institutional Oimensons of Povey Lawrence F. Salmon May 1990 E Madrona Roduction 37496 WPS412 Exchange Rate Policy in Deveopin W. Max Corden May 1990 U. Carden Countrie 191 75 NPSA13 Supprting Safe Mothernood: L U Howard May 1990 S. Ainswonh A Revew of Finauoad Trens 31091 Fu Report WPS414 Supporting Safe Motherhood: L M. Howard May 1990 S. Ainswonh A Rviw of Ftnanal Trends 31091 Summary A VWPS415 How Good (or Bad) are Country Norman Hics Aonl 1990 M. Berg Projectons? Michel Vaugeo;s 31058