Wage policy and the structure of wages and employment in Zambia

Document of The World Bank FOR OFFICwAL USE ONLY Rept N. 5727-Z& WAGE POLICY AN3D THE STMUCTJE OF WAGES AND EUNLOYMNE1 ifay 7, L986 This document has a restricted distribution nd may be used by recipients oniy in tbe performance Of !heir official duties. Its centents maw not otberwise be disclosed without World lank authorizaton.  FOR OFFICAUL USE ONLY Preface This report is based on a mission which visited Zambia in June 1984, consisting of Oey Astra Neesook (Head of Mission), David Lindauer, and Parita Suebsaenz. The study was carried out in collaboration with the Prices and Incomes Commission of Zambia, and formed part of a larger Bank-sponsored study of government wage policy in Africa. This document has a restricted dtrdbudon and way be used by rciplents only in the peformance of their official duties Its matents may cot othewise be dsosed without World Dank authorization.  Contents PAGE I. Introduction ............................... I _- aL. Naroecoonaic Backgrona ...........................................3 It-. Wages and Employment in the Formal Economy---..----..-.....-8 IV. Real Pay Adjustments and the Evolution of the Wage Structure ...... 18 V. Managing Wages and Employment in the Governacent Sector ......... .34 VI. Conclusion ..... ...o.....ewoo...,...... 48 Annex J. Trends in Real Wages ................................... .....50 Annex II. Total Compensation: Adjustments to Basic Salaries .... .. 62 Anner III. Starting Salaries Versus Progression.. .................... 71 Annex IV. Adjustment of Earnings for Skill DifferentiaLs ... _. 76 Bibliography ............. 78 Appendix Tables . ........... ... .. .. ............. 82  I. Introduction Over the past decade, the Zambian economy has been subjected to some adverse externaL circumstances, the outcome of which has been a falL in its real per capita income and a rapidly deteriorating external debt situation. Along with a number of other African countries in similar situations, the chatlenge for Zambia has been for it to undergo some fundamental adjustments in the structure of its economy while protecting as far as possible the standards of living of different groups in the population. In order to undertake difficuLt structural reforms, the government will have to rely in Large measure on its civil service whose efficiency in the performance of its duties will be crucial to the success of the adjustment program. At the same time, severe financial constraints limit the government's ability to maintain a civil service suited to such a task. Government policies concerning wages and employment are part and parcel of the process of adjustment. In the face of severe external difficulties of the sort that have befallen Zambia, the manner in which wage policies respond to these probLems can have far-reaching effects on the distribution of the burden of adjustment to the falL in the nation's real incomes. Moreover, it will directly affect the ability of the government to maintsin a competent civil service able to meet the chalLenges confronting Zambia in its efforts to reshape the economy and put it on a firm footing again. Policies related to prices and wages in fact have been activeLy pursued by the government of Zambia since Indepedence. Although such policies are properly viewed as integral parts of any system of national economic management, a number of special economic and institutional features provide further rationale for a comprehensive prices and incomes policy in Zambia. First, the country is heavily dependent an foreign trade, with a concentration of exports in copper and other metals. This situation entails both great potential benefits and substantial vulnerability to international economic fluctuations. Given the recent deterioration in the external terms of trade facing Zambia, a critical question is how government actions, market forces and institutional arrangements determine who will absorb the resulting income Losses. Second, the Zambian domestic market is small and many industries are characterized by monopolistic situations. The government may therefore have a role through pricing policies to intervene in an attempt to protect consumers from the higher prices and lower output levels monopolies tend to offer. Third, because of the dominance of the government and parastatal sectors in the formal sector labor market, government decisions concerning public employment and wages have a major effect on the formal sector as a whole, as well as on the go'rernment budget. Rather than being able to use private sector wage levels as a yardstick, the government finds itself having a major impact on wages and salaries throughout the formal sector whenever it takes action concerning wages or employment of its own civil servants. This fact needs to be taken into account by the government in the formulation of an overall prices and incomes policy. Finally, given the size and strength of the trade unions in Zambia, the formulation of wage and price policies may provide the opportunity for government, management and Labor to reach a consensus on wage and price adjustments. Without some such mechanism for reaching a consensus, the resolution of competing interests may prove to be more difficult.  -2- Against this background, the major wage and price policy concerns facing the 7ambian government include short-run stabilization and inflation control; market structure issues of efficiency and equity; the management of the public sector, that is to say employment and wage issues in the government and parastatal companies; poverty aLleviation and distributive issues; and industrial relations and dispute settlement. This paper discusses recent developments in wages and employment in the formal sector in Zambia in relation to these concerns, with special attention on the government sector itself. More specifically, it looks at the distribution of real wage losses among different groups of workers; the effectiveness and consequences of the government's policy of "narrowing the gap'; and the problems related to the management of wages and employment in the government sector. The following section provides a brief background of recent oacro- economic developments which have had a profound effect on real incomes and on the governmentIs budgetary position. Section III gives some information on the government sector in relation to the formal sector as a whole, the process of wage setting in the formal sector and some historicaL background of government wage policy in Zambia. In Section IV trends in real wages and the changing wage structure and their consequences are discussed. Problems spe- cifically related to the management of the government wage biLl are raised in Section V while some policy issues and conclusions are offered in Section VI. Detailed discussions on a number of subjects reLevant to the overall discussion on wage and employment policies can be found in the annexes. A-nex I examines the trends in reAl wages both at the aggregated and disaggregated levels; Annex II the composition of pay packages and the effects of tax policy on the wage structure and wage trends; Annex III the problem of using starting salaries as opposed to allowing for progression up the salary scales; and Annex IV the adjustment of -earnings to take account of skill differentials.  -3- II. Macroeconolic Background The present wage and employment situation in Zambia must be viewed within the context of the overill macroeconomic developments of the past decade. Policies pertaining to different economic sectors have implications for the levels and movements of wages and employment and, conversely, direct decisions concerning wages and empLoyment will have repercussions on the Lev -s of production in the different sectors. The foLLowing briefly sumr- marizes past trends of major economic indicators, and discusses various steps taken by the government to alleviate recent economic and financial problems in the economy and to manage wages and employment in the government sector I- Some relevant macroeconomic indicators are presented in Table 2.1. Zambia's total dependence on the copper industry for foreign exchange earnings has resulted in the present economic crisis. Uhen real copper prices started to fall in 1975 owing to an international oversupply of the commodity, the destabilizing effect on the Zambian economy was feLt imuediateLy. By early 1984, real copper prices were 57 percent lower than they had been in the period 1970-74. In the years following the initial setback, the deteriorating export situation, coupled with the rise in import prices due primariLy to world-wide inflation, had devastating effects on the domestic economy. With copper still constituting 90 percent of merchandise exports, the external terms of trade facing Zambia has become progressiveLy worse since the mid-1970s. While import prices have continued to climb, by an average of 12 percent per annum between 1974 and 1983 in terms of current U.S. dollars, export prices have been falLing in line with copper prices, resulting in the terms-of-trade index for 1983 being only 31 percent of that in 1974. Consequently, the economy has been plagued by huge current account deficits in the balance of paymencs; declines in real imports which in turn have caused high underutilization of productive capacity; declines in national savings and real investments: a deteriorating government financial situation characterized by large budget deficits, reduced public expenditure, and rising debt obliga- tions. In addition, high rates of inflation aggravated by government borrow- ing from the banking system have persisted throughout the late 1970s up to the present. The result has been a general contraction in the Zambian economy during the past decade. Gross domestic product per capita in real terms has been declining since 1975; by 1983, it was 20 percent Lower than in 1975. However, performance has varied among the different sectors. Agricultural production stagnated between 1978-83 because of droughts in 1979, 1980 and 1982. Growth per annum during this period amounted to onLy 0.4 percent, a substantial decline from the 1970-78 period when agricuLtural growth averaged 2.8 percent per annum. Production in the mining sector has declined by approximately one percent per annum since 1974. Lack of imported inputs has exacerbated production problems in this sector. Manufacturing production fell sharply in 1975 and stagnated until 1981 when it began to pick up again. although many firms still operate at low capacity. Industries which were able to rely on domestic sources of supply such as food, beverages and textiles have grown, while other industries which had to depend on imported I/ This section is drawn from World Bank Report No. 5000-ZA, Issues and Options for Economic Diversification, April 16, 1984.  Table 2.1: SELECTED MACROECONOMIC INDICATOERS, ZAMBIA, 1970-1983 1970 1975 1976 1977 1978 1979 1980 1981 1982 1983 Copper price In U.S. cenis per pound 64 56 64 59 62 90 99 79 67 72 Copper production in millions current K 648l0 455,2 633.7 557,8 557,4 825,4 955,4 781,5 716,8 Fxport of goods and non-laclor services in millions current K 685.4 575.0 832.3 781,5 755.3 1,209.3 1,270,8 982.0 976,6 In millions current U.S. $ 929.8 1,526.1 1,611,4 1,126.4 1,049.8 torms of trade (current U,S. 49 1974 100 100 40 49 36 28 31 Copper erms of trade (current U.S. S) 1970-74 100 125 54 60 51 46 59 60 50 44 48 Current account balance in millions current K 73,4 -469,6 -99,0 -224,6 -252.9 -8.8 -511.9 -743.9 -666,4 as S of GOi 5.8 -29,7 -S.3 -11.5 -11.3 -0,3 -17,0 -21,6 -18.7 In millions current U.S. S -295 - -649 -853 -716 -306' as A of GOiP -12 -17 -22 -19 -9g Debt service payment (millions current U.S. 1) 59.0 96.9 122,4 196,3 241.,9 257.2 293,2 284,7 190.2 126.2 1 Gi)P/capi ta (1970 K) 305.0 296.8 300.2 277.2 279.3 250,2 2511 254.0 240.5 236.S Consumer prIce Index (low-income) 1975 - 100 70,8 100,0 110,8 142,3 165,6 181.6 202,9 231,3 260,2 311,2 Government expenditure 3/ (millions current K) 454,3 855,0 771,5 821,0 815,4 956,4 1,302,0 1,J80.6 1,643.2 1,364,3 U Government expenditures by Ministrles 4/ (millions current K) 523,9 455.0 468.5 469.4 514,2 672.2 723.2 1,046.7 917.6 5/ Government expenditure on wages 6/ (Imn'lions current K) 80,5 152.8 176,9 184,5 189,0 203.8 256,6 352,4 370,9 306,7' 'Estimate I/ '974 figure. 2/ Average for 1975-1979. 3/ Figures for 1970, 1976 and 1980 exclude certain book entries representing conversion of outstanding government loans to various public enterprises Into equity and/or assumption of parastatal debt by the governments 4/ these exclude rural beer surtax fund, pensions, subsidles, and constitutional and statutory expenditures, 5/ Budgeted figures, 6/ Total compensation to government employees comprising personal emoluments for established staff, wages for classified dally employees, other employee benefits, and wage adjustments, Sources World Bank Report No, 5000-ZA, April 16, 1984, , . . .. 6  inputs such as fabricated metals, chemicals and petroleum have contracted. The construction industry has fallen by 9 percent per annum resulting from the Lack of fixed investments. The service sector has been constrained by shortages as in other sectors; the value added in trade and transport has declined, while the modest growth in this sector has been led by an expansion of government empLoyment and general business services. The decline in output of the copper industry and the performance of individual sectors have significantly altered the structure of the Zambian economy. The mining sector which accounted for 36 percent of GDP in 19-0 made up only 15 percent of GDP in 1983. The construction industry, historically tied to mining, also saw its share in total output fall from 7 percent of GDP in 1970 to 2 percent in 1983. As a result, percentage shares of other sectors rose, particularly in manufacturing and services. Consistent with the above shift is the change in the structure of formal sector employment. Employment shares in agricuLture remained virtually unchanged between 1970-83; those in mining and construction feLl, indicating movement of labor from these industries into the manufacturing and services sectors whose employment shares ross (see Appendix Table 1). Employment growth rates of the various sectors substantiate these trends. During this period of economic depression, inflation in the economy was very high. Consumer price increases after 1975 have remained above 10 percent per year, accelerating to 16-20 percenc per annum between 1976-78. The slowing of consumer price increases in 1979 and the early 1980s reflected the IMF standby agreement in April 1978 which placed restrictions on credit expansion. However, the latest figures for 1983 indicate higher price increases once again. These reflect in part the generaL decontrol of wholesale and retail prices announced by the government in December 1982. The sharp deterioration in the international terms of trade during the past decade has placed Zambia's balance of payments situation in chronic disequilibrium. Current account deficits increased from an average of 12 percent of GDP in 1975-79 to an average of 20 percent in 1980-82. However, in 1983 the current account deficit was reduced to only 9 percent of GDP through the rMF-assisted stabiLization program and a sharp decline in imports resulting from lack of commercial credit. The government resorted to heavy external borrowing to finance current account deficits and, by the end of 1982, Zambia's external liabilities were at about USS4.5 billion, of which US$1 billion were for overdue payments. The increase in borrowing has escalated debt service obligations. In 1974, service payments on medium- and long-term public debt were equivalent to less than 8 percent of export earnings; this increased to an average of 20 percent of exports between 1978-80 and to 25 percent (35 percent including rMF repurchases and charges) in 1981. By 1982, with large arrears accumulated, tocal debt service was 38 percent of exports. Debt service obligations in 1983 were estimated at US$550 million, representing 52 percent of export earnings. In addition, these obligations do not include approximately US$1 billion accumulated external payment arrears. Unable to pay these obliga- tions, the government has sought debt rescheduling, starting with an agreement with members of the Paris Club in May 1983. Although these arrangements may alleviate the problem in the short run, longer-term solutions must be found if Zambia's debt situation is to improve.  Balance-of-payments problems and the overall economic deurerssion have adversely affected the government's fiscal performance. Public sector deficits increased from 6 percent to 15 percent of GDP between the first and second half of the 1970s. During 1980-82, deficits averaged 18 percent of GDP. The increase in deficits has resulted from both a decline in revenues (maialy from the faLl in mineral revenue) and an increase in governmenc expeaditure reLative to GDP. Total government expenditure increased from 37 percent of GDP between 1970-74 to 43 percent during 1980-82. In 1983, expenditures dropped to 33 percent of GDP as a resuLt of the implementation of the government's financial stabilization program. However, total expenditures have been declining in real terms, and in 1983, public expenditures were 24 percent lower than in 1974. There has also been a shift in the structure of expenditures. The share of expenditure for wages increased from 18 percent in 1975 to 28 percent in 1983, reflecting salary awards and increased government employment. In addition, an increasing share of expenditures was devoted to operating expenses and debt service. Consequently, capital expenditure has been drastically cut back, as reflected in the faLL in its expenditure share from 29 percent in 1975 to only 13 percent in 1983. Given Zambia's external circumstances, the challenge for the government has been not onLy to manage the economy as a whole, but also to manage the public sector. The government is dependent on qualified civil servants to help put the economy on the growth path again. Moreover, with rapid price inflation there is considerable pressure for it to maintain or even raise the level of real wages- The economic contraction has also led to pressure for the government to expand government empLoyment, given the natural growth of the labor force in the face of contracting labor demand from the private sector of the formaL economy. However, increases in the government wage bill compete with -_ priorities for government expenditures which have been declining in real terms. ALl this puts the government in the unenviable position of deciding on the appropriate strategy for managing wages and employment in its civil service in such a way as to have a smoothly functioning bureaucracy without putting an additionaL burden on government expenditures and the budget deficit. The government has taken major steps starting in December 1982- January 1983 to alleviate problems of the recent past, including a 20 percent devaluation, general price decontrol, increases in the interest rates and budgetary measures in order to curb budget deficits. Furtherwre, in April 1983, the government and the IMF agre-ed on a one-year standby arrangement, sought debt rescheduling, proposed a freeze on government salaries and established a ceiling on other wage increases at 10 percent. In July 1983, a flexible exchange rate system was adopted. The government also negotiated another standby arrangement with the IMF for 1984. It has launched a program to rehabilitate the miniag industry by strengthening management, improving efficiency and cutting costs. It has sought to improve the efficiency of public enterprises in the ZIMCO (Zambia Industrial and Mining Corporation) group by granting them greater autonomy in management and encouraging a more commercial outlook.  -1 So far, benefits of these measures include increases in agriculturaL producer prices resulting from exchange rate adjustments Which, together with concessionaL tax rates for agricultural incomes, have Led to a signiEicant. increase in the area under cuLtivation. The decontroL of industrial prices has had a restraining cEfect on domancic damand while increasing the profitability of firms as weLl As their investmenc capabilities. These meAsures are indications from the government of its intention to move awny from direct controL sad allow market Eorce., price signals and private sector initiacives to play a greater role. 1i 6  III. Wages and EmpLoyment in the Formal Economy To serve as an introduction to the discussion of wage poLicy in Zambia, this section provides a brief description of the formal economy and the process of wage setting in the government, parastatal and private sectors. A more detailed discussion of wages and employment in the government sector appears in Section V. The Formal Sector: Size and Composition iormal sector employment in Zambia accounts fgy onLy a small fraction of the labor force, approximately Z2 percent in 1980. - Estimates for earLier years suggest that recenj formal sector growth has not kept pace with the growth of the labor force. - Table 3.1 gives a distribution of formal sector employment for Zambian empLoyees by employer group (government, parastatal and private) and industry.- The table is based on the 1983 Nanpower Survey which covered some 100,000 individuals in four provinces (Central, Copperbelt, Lusaka and Southern) which together account for two-thirds of total formal sector empLoyment. The formal sector is dominated by the pubLic sector which consists of the government and parastatal companies; these accounted for 37 percent and 38 percent of total formal sector employment in 1980 respectively, with the private sector accountiag for only 25 percent. 3/ The composition of the three employer groups by industry groups is very different. The government is dominated by community services and construction; together these account for 87 percent of total government employment. Practically all of mining employment is in the parastatal group where it accounts for 44 percent of total parastatal empLoyment. Manufacturing, transport and communications, and distribution make up the major part of the remaining empLoyment in the parastatal group. The private sector is not in the mining and electricity sectors at all but otherwise is more spread out across the industries than the other two sectors; it provides for a significant share of employment in manufacturing. Each industry is typically dominated by only one or two of the employer groups, with agriculture being the exception. 11 The 1980 Population Census gives a preliminary estimate of the labor force of 1,756,262. An estimate of formal sector employment of 379,300 in 1980 is taken from Table 1 of the Employment and Earnings Survey, Supplement to the Monthly Digest of Statistics, Volume XIX, Nos. 4 to 9, April/September 1983, Central Statistical Office, Lusaka. 21 See International Labour Office, Zambia: Basic Needs in an Economy Under Pressure, ILO, Addis Ababa, 1981, Table 4.1, which gives an estimate of formal sector employment as a percentage of the labor force of 252 in 1965 and 27% in 1969. 3/ The official pubLications of the Central Statistical Office refer to the government sector, consisting of the central and local governments, as the public sector.  -9- Table 3.1: DISTRIBUTION OF ZMIA<B EMPLOYEES IN THE FORMAL SECTOR BY EMpLOYER GROUP AND INDUSTRY, ZAMBIA, 1983 Percentage in Government Parastatal Private AgricuLture, forestry and fisheries 7.9 7.7 11.0 Mining ad quarrying - 43.5 0.3 Manufacturing 0.5 14.8 27.4 ELectricity and water 2.1 3.5 - Construction and aLLied repairs 18.3 - 19.0 Distribution, restaurants and hotels 1.5 9.2 17.1 Transport and coummunications 0.7 14.2 2.7 Finance, insurance, reaL estate and business services 0.2 7.0 13.0 Commmity, social and personal services 68.7 0.2 9.4 TOTAL 100.0 100.0 100.0 Percentage of employer group in formal sector in 1980 1/ 37.3 38.0 24.7 1/ All employees, i.e., Zambians and non-Zambians. For 7ambian employees, the proportions in the government, parastatal and private sectors in 1983 were 37.8, 38.0 and 24.3 percent respectively. Source: UnpubLished data from the Manpower Survey, 1983; and Employment and Earnings Survey, Tables 1, 2 and 3 for the quarter ending December 1980, Supplement to Monthly Digest of Statistics, Vol. I, Nos. 4 to 9, April/September 1983, Central Statistical Office, Lusaka.  - 10 - The very different distributions of employment in the three employer groups across industries are underscored in Table 3.2 which gives a distribution of different occupations across employer groups. Over tw-o-thirds of all professionaL workers are in the government and only 7 percent in the private sector. On the other hand, 65 percent of production workers are in the parastatal companies, with only 12 percent in the government. To the extent that the accupationaL breakdown roughly corresponds to the different labor markets in which the three employer groups compete for workers, the public sector taken as a whole is able to exert considerabLe influence over aLl wage and employment Levels throughout the wage economy. However, the government may find itseLf in vastly different situations depending on its share of employment in different occupational groups. It clearLy has a big effect in the market for professional workers but far less in the market for production workers because of its relatively smaLL share of total empLoyment there and in view of the pressures exerted by nonwage urban workers who constitute part of the totaL supply of such workers. iage Setting in the Formal Sector In order to understand how wage and empLoyment decisions are made in the forml sector in Zambia, it is usefuL to classify formaL sector workers not simpLy by empLoyer group but also according to whether or not they are unionized. In the government, the latter distinction is unimportant since all wages are set by the government ahether or not the workers belong to a union. All firms in the parastatal sector are unionized, in the sense that the majority of their workers belong to one or more unions which negotiate with the employers over their workers' terms and conditions of service. The distinction within a parastatal company between the unionized and nonunionized workers corresponds roughly to that between blue-collar and white-collar workers. For private firms, some of them, including the larger ones, are unionized, while the remainder are not. Within the unionized private firms, there is again the distinction between the unionized and nonunionized workers. The process of wage setting in the formal sector in Z7abia is by no means independent between the government, parastataL and private sectors. To take first the government sector, salary adjustments in the civil service since Independence have generally been based on the recommendations of salaries review commissions appointed specially for the purpose at roughly four-year intervals. The adjustments have involved changes in both the general leveL and the structure of salaries. For the years between reviews, civil servants have generaLly had to forego any adjustments in their p j, even when the government's general wage guidelines have allowed increases. - This means that when adjustments are finally made, they tend to be Large and consequently have a disruptive effect oa the formal sector as a whole by setting off a series of pay adiustmpnts in the parastatal and private companies in response to government action. 4/ The 1980 salaries review (.(uchangwe) was an exception to this in specifying additional increases for lowly-paid workers for subsequent years as well.  - Ii - Table 3.2: DISTRIBUTION OF DIFFERENT OCCUPAIIONAL GROUPS ACROSS EZPLOYER GROUPS, 1983 (percent) a~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Occupation Government Parastatal Private TotaL ProfessionaL 68.2 25.2 6.6 100.0 Administration 28.4 40.3 31.3 100.0 ClericaL 29.6 49.0 21.3 100.0 Sales 35.0 29.4 35.6 100.0 Service 35.3 34.7 29.9 100.0 Agriculture 18.0 44.6 37.4 100.0 Production 12.0 65.3 22.7 100.0 Labourers 39.3 33.9 26.8 100.0 Source: Preliminary tabulations from the Manpower Survey, 1983, Central Statistical Office, Lusaka.  - 12 - The approach of the salaries review commissions in the past has been to compare the salary scales in the civiL service with those r the parastatal companies, especially for the more senior levels. Recomendations for salary increases were intended to close the gap between the civil service and the parastatal sector on the grounds that the government needed to be able to attract qualified people and, if it could not compete with the parastatal company salaries, government officials would resign and join the parastatals. Thus government wage decisions are not independent of the wage level and structure in the parastatal sector. Determination of pay leveLs within parastatal enterprises is characterized by systems which are both flexibLe and rigid. In the prevailing system of parastatal pay administration, parastatal employees are either unionized and covered by a collective bargaining agreement, or nonunionized and come under the ZINCO (Zambia Industrial and Mining Corporation) pay scales (currently referred to as the Z scales). The only exception is nonunionized ZCCN (Zambia Consolidated Copper Mines) employees who are covered by a separate saLary scale which was not made available to us and which is generally believed to be considerably more generous than the Z scale. Nonunionized personnel generally refers to white-coLLar and professional workers, including expatriates, while unionized workers refer primarily to blue-collar workers. However, there can be considerabLe overlap with respect to both salary levels and occupations between the nonunionized and unionized coverage. In fact, a strategy for raising compensation Levels of particular yositions is to move these jobs out of the unionized domain into the Z scale. 6 The Kwanakatwe Commission in 1975 made recommendations which led to a uni-ified salary scale for the civil service and the nonunionized workers in the parastatal companies of ZINCO, thereby acrhieving the desired comparability between their salaries. Since then, ZIMCO has found this arrangement to be ausatisfactory for a number of reasons. First, it feels that the unification has hurt its subsidiaries which have to compete with the private sector, not the government, for their workers. Private companies are at an advantage in having the freedom to fix and adjust their salary scales whenever necessary. Second, the unification has affected the wage relationship between the unionized and nonunionized workers. The wage scale and fringe benefits for unionized workers are obtained through collective bargaining. The nonunionized workers can be expected to make similar demands in order to retain the existing pay differentials. In fact, it would be in their own self-interest to acquiesce to union demands in order to be in a position to justify similar increases for themselves subsequently. In the private sector there is no restraint on this catching up, but in the parastatals where there S/ See, for example, Report of the Commission of Inquiry into the Salaries, Salary Structures and Conditions of Service, Volume I, 1975, p. 19; and Report of the Administrative Committee of Inquiry into the Salaries, Salary Structures and Conditions of Service, Volume I, November L980, pp. 23-24. 6/ Nitrogen Chemicals of Zambia has apparently used this strategy.  - 13 - is pressure to keep the salaries of nonunionized workers in line with the government scale which is adjusted only periodically, the level of pay of unionized workers has been catching up with that of nonunionized workers and, not surprisingly, creating dissatisfaction within the latter group. Since the implementation of the recommendations of the Hwanakatwe Commission, the system of the unified salary scale has become ineffective. In practice, ZINCO has a great deal more flexibility than the government through its extended scale for professional and technical workers which allows it to increase their salaries by up to 40 percent of their basic leveLs. Moreover, there is further flexibility because the effective salary scale for all the workers in a company is determined by which of four possible salary grades the chief executive of a ZIMCO subsidiary is ranked. It is also generally beLieved that the parastatals offer superior fringe benefits compared with the government. Thus the parastatals are not as constrained as the civil service in terms of their abiLity to compete with the private sector for workers. Furthermore, although aLL nonunionized personnel in ZIKCO are in theory covered by the same set of salary scales and conditions of service, there is flexibility at the enterprise levei. First, there is considerable leeway in classifying the entry salary of any given job amidcst the 19 salary grades. While there is a move within ZIMCO to have a more formal inter-firm system of job evaluation and salary determination, at present the flexibility remains. Second, many nonuage perquisites, especially the provision of transport, access to training courses (especialLy overseas) and the character of housing benefits, vary across firms and resuLt in inter-firm variations in total compensation for similar jobs and workers. Third, variations in the effective salary scales for firms occur depending on She leveL at which a firm's managing director is paid as mentioned above. Middle management salaries are, therefore, a "top-down" function of where a firm has been classified by ZIMCO. However, there has probabLy been progressively less variation in inter-parastatal pay differentials for nonunionized employees over time, as a resuLt of the move away fro independent wage fixing in 1975 and the evo_tion of a progressively more unified ZI4CO corporate saLary structure. - Turning to unionized workers, each parastatal negotiates its own collective bargaining agreement with union representation- at the firm Level. Only in rare cases will an enterprise have to negotiate with more than one union, although one agreement may cover more than one enterprise. Recent agreements have tended to be negotiated for a duration of 2 years, down from 3-year contracts which were prevalent in the early to mid-1970s. Amendments to existing contracts have been employed as a means of adjusting basic salary levels during the life of a contract. SaLary negotiations have tended to be for uniform absolute pay increases, i.e. for so many ngvee/hour, regardless of occupation or pay leveL (see Appendix Table 22). Such a policy is favored by che trade unions and contribute to salary compression in percentage terms. Wage negotiations often set wages for each salary grade in a contract. A 7/ The resulting inflexibility was seen as a severe constraint in Kafue Textiles where technical staff could not be attracted and retained in sufficiently large numbers.  - 14 - schedule of annuaL increments or, more generalLy, of an intra-grade salary scale, whiLe comon to nonunionized staff, is less common for unionized workers and also contributes to a reLatively more compressed wage structure for tht' when compared to the nonunionized workers. k"jr private firms which have unionized workers, the process of wage setting is Amilar to that in the parastataL companies. In determining the structure of unionized collective agreements in both employer groups, employers' .ssociations pLay an important role. At present, the - istry of Labour recognizes 13 employers' associations and 18 trade unions. _ Employers' associations consist of both parastatal and private firms which negotiate wages and conditions of service with a trade unjon representing a given industry through Joint Industrial Councils (JIC). 9 The resulting JIC agreements act as guidelines for members of an employers' association which negotiate firm-specific agreements. go" strictly these guidelines are adhered to is difficult to determine. By law alL members of an association must not pay less than stipulated in the JIC agreement. However, by withdrawing from the association, an employer is no longer bound by the conventions of the association's agreemnct. This loophole has allegedly been exercised increasingly during the last few years, especially in the construction industry. The Zambia Federation of Employers also confirmed that this practice was becoming more comon. Only scant evidence exists for a few parastataL companies of how individual collective agreements compare with the JIC agreements and no evidence has been obtained on how such a differential has changed over time. To conclude, the process of wage setting in the formal sector is quite complex. It is worth emphasizing here that wage decisions taken by the different empLoyer groups within the formaL sector are not independent of one another. Salaries -review commissions of the government take inco account parastataL and private sector pay in making their recomnendations concerning civil service pay adjustments. The parastatal companies of ZIKCO in turn react to government wage decisions and adjust their salary scales accordingly. Recent Formal Sector Wage Developments Although Zambia had no central agency which was formally charged with the responsibility of formulating an overaLL prices and incomes policy until the recent establishment of the Prices and Incomes Cormission (PIC), the government has nevertheless had a great deaL of influence over rates of nom- inal wage increases in the formal sector. First, it is directly responsible for setting the salary scales in the civil service which alone accounts for over one-third of total wage employment and which dominates the labor market for certain categories of workers as already noted. Second, it has from time to time attempted to infl-ence wage decisions in the parastatal sector. The frustration of the competition offered by the parastatal companies and the 8/ As reported in its Annual Report as at December 31, 1983 and December 31, 1982 respectively. 91 Wage minima resulting from these agreements are reported annually in the Ministry of Labour Annual Reports. (See Appendix Table 23).  perceived need for the civil service to catch up with it periodicaLLy even- tuaLLy led the government to forge a unified salary scale for the government and parastatal sectors in the late L970s. FinaLly, the government has attempted to influence the wages of ,mionized workers in both the parastataL and private companies through the periodic issuance of wage guidelines. In addition to its influence over wages, the government is of course responsible for determining the leveL of employment in the civiL service. Between 1967 and 1978 the totaL number of estabLished posts in the civiL service grew at an average annual rate of 4 percent, from 42,870 to 71,74L1 W Since then the number of estabLished posts has risen much more sLowLy. TabLe 3.3 summarizes government attempts to set general wage guideLines, government wage decisions regarding the civiL service, and key wage settlements in the mining sector over the past two decades or so. Although government preoccupation in the area of wage poLicy has been increasingly with wage restraint, this has not always been the case. The wage scales for Africans and non-Africans were compLeteLy distinct up to L961, while the period 1961-1967 saw a graduaL movement towards a unified nonraciaL pay scale which was marked by large pay increases in all sectors, the unification tak;ng the form of a rise in African wages to meet those of non- Africans. The impetus to this deveLopment came from the unification of pay scales in the mining sector foLLowing the recommendation of the Brown Cotmuission in 1966 to raise the wages of Africans by 22 percent. Thus the objective of reducing the gap between the leveLs of pay for Africans and non- Africans was achieved LargeLy through wage increases. A somewhat different officiaL wage policy started to emerge aft .r the first Turner report in 1969 when a temporary voluntary wage freeze was instituted. Since then government wage poLicy has consisted of two major components. The first is wage restraint, as evidenced by the various generaL wage guidelines: a 5 percent per annum ceiLing on nominal wage increases in 1970; a government pay freeze in L977; a limit to K6 per monch nominal wage increase in L978 after the second Turner report, foLLowed by a K156 per year nominaL wage increase in 1979; and, finaLLy, a 10 percent ceiling on nominal wage increases in general, coupled with no increases in the civil service, foLlowing the standby agreement with the International Monetary Fund (IMF) in 1983. The second component of government poLicy concerning wage adjustments is that known as "narrowing the gap." This policy goes back a long way and has manifested itself through government salary adjustments which grant greater percentage increases for the lower-paid workers. In L966 the Whelan Commission granted a 22 percent average salary increase, but the increases were tapered at the top of the scale. Since then the O'Riordan (1971), lwanakatwe (1975) and Muchangwe (1980) Comnissions have all recommended differential percentage increases designed to reduce the gap between the higher- and Lower-paid workers. The poLicy has been adopted by other sectors of the formaL economy as well. For example, in the mining sector, although the Brown Commission granted a uniform 20 percent increase to all African 10/ Second Report to the Governent of Zambia on Incomes, Wages and Prices in Zambia: Policy and Machinery, H. A. Turner, et al, I.L.O., Geneva, 1978, Part III, Appendix K; and Establishment Register, 1978.  Tabld ),IL I3NICAL WACE iJU1iMLIldiS GCViOCMIkI IALAA ADJUSTCTI