Report No. 7100-TA Parastatals in Tanzania Towards a Reform Program July 27,1988 Country VI Department Africa Region FOR OFFICIAL USE ONLY ! i: Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OMCIAL USE ONLY ABBREVIATIONS CCM - Chama cha Mapinduzi (Revolutionary Party) CPI - Office of the Commissioner for Public Investments CRDB - Cooperati7e Rural Development Bank DEVPLAN - Department of Development Planning, Ministry of Finance Economic Affairs and Planning DDCs - District Development Corporations ERP - Economic Recovery Program FAO - Food and Agriculture Organization of the United Nations FYP - Five Year Plan GDP - Gross Domestic Product GM - General Manager IDA - International Development Agency IFC - International Finance Corporation NBAM - National Board of Accountants & Auditors NBC - National Bank of Commerce NDC - National Development Corporation NMC - National Milling Corporation NPC - National Productivity Council NUWA - National Urban Water Authority OPEC - Organization of Petroleum Exporting Countries PIT - Presidential Implementation Team PLT - Permanent Labor Tribunal POC - Parastatal Organization Committee PPMB - Project Preparation and Monitoring Board RTC - Regional Trading Corporations SCOPO - Standing Committee on Parastatal Organisations TAC - Tanzania Auditing Corporation TANESCO - Tanzania Electric Supply Corporation TAZARA - Tanzania-Zambia Railway Authority TBS - Tanzania Bureau of Standards TIB - Tanzania Investment Bank TLC - Tanzania Legal Corporation This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authoriation. PARASTATA.S IN TANZANIA TOWARDS A REFORM PROGRAM Table of Contents Page No. Preface ...............***.................................... i Executive Summary ..... .... . .. .................................. ii CRAPTER ONE HISTORY, DESCRIPTION AND PERFORMANCE .................. 1 Evolution of Parastatal Sector ...................... 2 Principal Characteristics . .................................. 6 Parastatal Performance .................................... 11 CHAPTER TWO FACTORS DETERMINING PARASTAtAL PERFORMANCE .. .......... 21 Factors Affecting All Economic Activities ...................... 21 Factors Specific to the Parastatal Sector ...................... 22 CHAPTER THREE TOWARDS A REFORM PROGRAM .............. .............. 35 Past Efforts at Reform ............................................. 35 An Agenda for the Future ................... ......... 39 Institutional Arrangements .................................. 49 Figures 1.1 Fixed Capital Formation by Institution: Percent of Total 5..... 1.2 Fixed Capital Formations Million Shilling, 1976 Prices ....... 5 Tables 1.1 Government Shareholding in Parastatals, June 30, 1986 ........ 7 1.2 Parastatal Contribution to Sector Value Added and Wage Employment .................................. 8 1.3 The Parastatal Sector in Tanzania ................... 9 1.4 Supervision of Parastatals in Tanzania ....................... 10 1.5 Tanzaniak Resource Flows between Government & Parastatal Sector 1981/82 - 1985/86 . ......... ........ ................. 13 1.6 Index of Real Wages for Representative Groups ........ 15 1.7 Tanzania: Financial Performance of Parastatals Audited by Tanzania Audit Corporation, 1983-85 ..... ............... 17 1.8 Percentage of Production Lines Operating at Different Levels of Efficiency, by Ownership, 1987 .................. 18 1.9 Average Indicators of Performance Public and Private Industrial Firms, 1975-79 . ................................ 19 1.10 Tanzania - Average Indicators of Performance Sisal Sector, Public and Private Firms 1981-84 .... ........ 20 1.11 Indicators of Parastatal Performance by Selected Groupings, 1986 ............................. . 20 2.1 Tanzania: Tanzania Investment Bank (TIB) Sunmary of Arrears at 31 December 1986 .................... ...... 26 2.2 Tanzania: National Bank of Commerce Credit Plan Summarys Lending by Economic Sectors, at December 31, 1986 26 2.3 Tanzanias Parastatals with Accounts Overdue by More than Two Years .............. 31 2.4 Tpnzania Audit Cvrpcration: Parastatals Audit Results ........ 31 2.5 Tanzanias Accountancy Skills and Vacancies, 1986 ........... 32 APPENDIX A CHARACTERISTICS AND PERFORMANCE OF TANZANIA PARASTATALS ... .......... ..... . 56 Al PARASTATALS IN TANZANIA ..... ............. 56 A2 PARASTATAL VALUE ADDED BY SECTORS .............. 66 A3 ANALYSIS OF BUREAU OF STATISTICS SURVEY .. ............ 67 A4 INVESTMENT AND SOURCES OF SAVINGS ....... ....... 72 AS PARASTATALS REHITTING DIVIDENDS ..... .. ....... 74 A6 EXTERNAL FUNDS ONLENT TO PARASTATALS BY TREASURY ............. 76 A7 PARASTATALS WITH PROFITS OR LOSSES EXCEEDING SHS 5 MILLION ... 80 A8 FINANCIAL PERFORMANCE OF SELECTED PARASTATALS .............. 83 APPENDIX B EMPLOYMENT AND COMPENSATION IN TANZANIA'S PARASTATALS ... 96 Employment ....................................................... 96 Compensation Policy ........................ ..................... 105 Comparison with the Private Sector and Differentiation Among Parastatals ......... . ................................. 112 APPENDIX C ORGANIZATION, MANAGEMENT AND SUPERVISION OF TANZANIAN PARASTATALS .................... 116 APPENDIX D ANALYSIS OF QUALITATIVE AND SCOPO DATA ..... ............. 131 APPENDIX E EXAMPLES OF STATEMENTS OF GENERAL PRINCIPLES FOR PARASTATAL SECTORt BENIN, SENEGAL AND THE PHILIPPINES .. 134 PREFACE The Letter of Development Policy of 29 October 1986 calls for the preparation of an assessment of the parastatal sector which, along with other studies, should help greatly both in clarifying the future role and responsibilities of the public sector and in ensuring that thRse responsi- bilities are carried out more efficiently than in the past." This study constitutes the first major step toward achieving these ends. Its prin- cipal objectives were to develop a clear picture of the sector, how it performs and interacts with the rest of the economy, a diagnosis of the causes for its poor performance and a set of recommendations to improve the situation. In large part, the recommendations aim to initiate a process that will produce and implement a Parastatal Recovery Program that is con- sistent with and complementary to the Economic Recovery Program. The study was undertaken by a team of experts and consultantsl/ from the World Bank and the University of Dar Es Salaam with substantial assistance from the Ministry of Finance. The field work was undertaken in April, 1987 and a first draft of the report was revised after discussions with collaborators in Dar in October 1987. This report is divided into an executive summiary, three chapters and a series of Appendixes. Chapter 1 describes the parastatal sector and assesses its performance. Chapter 2 investigates the reasons for the poor performance observed, focusing on those factors that are amenable to im- provement through policy changes. Chapter 3 reviews past recommendations and efforts at reforming the parastatal sector and proposes steps to be taken from this point forward. The appendixes, among other things, present more detailed pictures of the parastatal sector, its economic and financial performance, employment and compensation practices and the institutional framework that manages and supervises the sector. 1/ Ronald G. Ridker, team leader, Sunita Kikeri, Russell Muir, John Nellis, Joseph Semboja and Mary Shirley. - ii - EXECUTIVE SUMMARY In recent years the Tanzanian Government has become increasingly concerned about the poor performance of the parastatal sector. This concern has been expressed in the Hamad Commission report of 1983 which identified gross inefficiencies and recommended remedies; the Annual Reports of the Tanzania Audit Corporation which have grown increasingly vociferous about the inadequacies of parastatal financial accounts and performance; a speech by President Mwinyi in November, 1985, which warned parastatals that they must get their financial affairs in order within the next two years or face serious disciplinary action; a 1987 report to Cabinet by the Presidential Implementation Team (formed in 1984 to implement the recommendations of the Hamad Co"ission); and the Nsekela Commission report of 1987 which was concerned vith remuneration of public sector employees, parastatal performance and related institutional issues. These expressions of concern are totally appropriate. While the parastatal sector has succeeded, at least partially, in accomplishing what was hoped for it at the time of the Arusha Declaration in 1967, it often did so only at high cost or at the expense of other goals. The nationalizations of the late sixties freed the country from dependence on foreign private investors; but the operation of the sector has resulted in increased dependence on foreign banks and aid donors and equal or greater dependence on imports (relative to GDP and to exports). An industrial base was established more rapidly than the private sector would have done; but after an initial spurt of growth, the pace of overall development has been very disappointing despite heavy investments. Indeed, growth in GDP per capita in constant prices has been negative during most years after 1976. Gross fixed capital formation and domestic savings as percentages of GDP have both declined, in large part (though not exclusively) as a consequence of parastatal performance. The composition of output has not improved. While substantial employment was generated, a class of indigenous managers and executives created and a variety of social services provided, the opportunity cost of doing so in this way was very high. And most important, far too many parastatals have been grossly inefficient in the sense that they have used more resources than they have produced for most of their lives. In the past, much of this inefficiency was hidden by the explicit or implicit subsidies and protection parastatal received. But with the discontinuance of subsidies, devaluation, tight budget and credit ceilings and reduced protection from foreign competition (because of the export retention and own import schemes), these inefficiencies have become more obvious and difficult to ignore. Until recently, efforts to correct these problems have focused on strengthening Government supervision and control. In the late 60's the Government established SCOPO, the Tanzanian Audit Corporation and sector holding companies with powers to advise, request reports, and sanction or otherwise intervene in a variety of parastatal decisions. In 1974 the post of Commissioner for Public Investment was established and combined with the functions of the Treasury Registrar, and in 1975 and again in 1983 the - iii - p,wers of this office were expanded. As a result of the Hamad Commission report, the all-purpose crop authorities were replaced by cooperatives and marketing boards, several parastatals were closed down or merged into other agencies, rules governing membership on boards of directors were changed, a large-scele effort to reduce public sector employment was undertaken and a series of programs initiated to reduce waste and increase cost-conscious- ness. The impact of all these efforts has be relatively small, in part because they have not gone far enough and in part because they have failed to attack more fundamental problems such as lack of competition and inadequate managerial accountability and autonomy. With the introduction of the Economic Recovery Program (ERP) in 1986, this situation began to change. By devaluing the shilling, raising interest rates, strictly limiting access to funding through the budget and credit institutions, introducing the export retention and own-funded import schemes, eliminating many price controls and lifting restrictions on trade and transport of related commodities, this program has placed the parastatals under considerable pressure to improve performance. Parastatals that can adapt to the new situation by improving internal efficiency or by shifting, for example, from lines heavily dependent on imports to export oriented activities, will prosper; those that cannot will be hurt with less prospect of being bailed out. While the ERP addresses some of the fundamental causes of parastatals' poor performance, it fails to address others and does not take into account special problems that parastatals face just because of the ERP. Increased financial pressure on parastatals to improve performance is not always translated into increased pressure on managers and senior staff. Even where it is. these individuals cannot respond properly if they are impeded by red tape, absence of technical knowledge, lack of resources or overwhelmingly large debt service obligations (made worse in some cases by recent devaluations). At a minimum, therefore, the following additional issues must be addressed to develop a complete program for parastatal reform. (1) Absence of overall stratest and clarity on Roals. Tanzania has no up-to-date statement of general principles that can serve as a guide to parastatal operations and decision-making; nor is there good coordina- tion between national goals and plans and parastatal goals and plans. The consequence is lack of clarity about performance criteria which allows poor commercial performance to be excused, ad hoc decisions of doubtful national benefit that are difficult to argue against, requests that parastatals perform functions they are not equipped to undertake, and expectations on the part of managers that their company will be bailed out rather than be allowed to go bankrupt if they fail to perform and that their job is rela- tively secure regardless of the fortunes of the company. To correct this problem the report recommends the appointment of a committee or task force to clearly lay out the criteria by which parastatal performance is to be judged, specify and explain the policies to be applied to the sector (and where necess&ry to individual parastatals) and lay this out in a comprehensive public policy statement. A considerable amoant of - iv - rethinking has already taken place (see first section of Chapter Three); what is necessary now is to codify these materials and fill in gaps. Examples of such statements developed for this purpose by other countries are included in Appendix E. It is also recommended that the President issue a statement announcing the appointment of this task force, indicating in general terms the new policies to be develced, announcing a serics of interim measures to apply until the policy statement is issued and period- ically reinforcing the new set of expectations (for example, that bailouts should not be expected). (2) Excessive number of parastatals given the availability of resources and managerial capacity. Tanzania has more parastatals than many much larger countries with greater administrative and managerial talents at their disposal. Most parastatals are underfunded--noncommercial parasta- tals because of a tendency to spread scarce budgetary resources evenly and commercial parastatals because of a failure to generate sufficient internal funds--and high quality managers and staff of both parastatals and super- visory bodies are overloaded. The consequence is that no agency, whether of high priority or not, is fully capable of getting on with its job. The only remedy for this situation is to reduce the size of the parastatal sector and concentrate the resources and managerial talent saved on those parastatals that remain. To accomplish this, the report recom- mends that parastatals be classified into four groups and special programs be established for each. First, noncommercial parastatals should be separated out and sub- jected to special review to determine whether their continued existence is still warranted: whether their functions are still needed and if so could be served better in other ways, for example by existing Governmental departments or private voluntary agencies. Those which are to remain parastatals should then be administered separately from commercial parastatals. Second, it is recommended that commercial parastatals be subdivided into those that must be maintained in public hands and made to work effectively, those that no longer have any rationale and are to be divested and the remainder that should be allowed to fend for themselves, prospering or failing without special help or intervention. Commercial parastatals to be maintained would work out with Government a plan or agreement that in the initial year would include equipment rehabilitation and financial restructuring if necessary and on an annual basis would include cost minimization targets, pricing formulas and investment plans. Such plans could have significant budgetary implications which need to be taken into account in forward budgetary exercises. For this reason, but also because the Government's ability to adequately develop and administer such plans is limited, the number of such agreements, and hence the number of firms in this category, must be strictly limited. Accordingly, this categor.' is likely to include firms typically considered public utilities or natural monopolies that must be regulated in any case, plus financial institutions, but few if any others. Parastatals to be divested would include those which no longer serve any useful purpose, which are clearly not capable of becoming economically viable in a reasonable period of time or which the Government decides it no longer wishes to manage itself. If a parastatal fiti one of these criteria, a plan for phasing it out and redeploying its useful assets would be worked out. It may also be necessary in some cases to develop a transition assistance program for labourers laid off in the process. Some candidates for this category could equally well be placed in the next category and allowed to fend for themselves; the choice should be made on pragmatic and political grounds, keeping in mind the benefits of rapid disposal of clear-cut cases. All other commercial parastatals would be allowed to fend for themselves without preferential treatment but with greater freedom to re- spond to competition and, if they find it useful, form joint ventures or enter into management contracts. Some may warrant temporary assistance to overcome past problems. Since Government resources available for such assistance are extremely limited, the report recommends the establishment of a fund from which awards would be made on a competitive basis to a limited number of firms best able to demonstrate their capacity to become economically viable if this assistance is provided. This is meant as an interim measure tG assist with transition problems; on a more regular basis the banking system should be prepared to play this role. (3) Weak ComPetitive Pressures and Discriminatory Treatment. Parastatals have been protected from domestic competition by enjoying exclusive rights to operate in certain fields and from international com- petition by the import licensing and foreign exchange allocatiin system. They have also enjoyed preferential treatment over the private sector, for example, in the allocation of credit and foreign exchange at low, official prices, in access to foreign technical assistance and capital, and in will- ingness to tolerate arrears in payment of bills and debt service. The consequence has been an absence of pressure for high-quality performance. At the same time, parastatals also have major disadvantages in facing competition. Appointment procedures sometimes result in weak boards and managerial teams; wage and compensation policies make it difficult to compete with the private sector for high-calibre, skilled personnel; parastatals are sometimes expected to perform services that are not in their best commercial interest; and they are saddled with problems from the past, including failure to develop the necessary skills to compete, managers with a civil service rather than an entrepreneurial mentality, and large foreign debts which now, because of devaluation, have become excessively burdensome. A variety of measures are required to ccrrect these problems. Among the more important are measures that impose greater financial discipline (by, for example, forcing the settlement or renegotiation of payment arrears and then, after a grace period, declaring any firm in arrears to be ineligible for credit or budgetary support regardless of - vi - priority) that make it clear that commercial parastatals are not required to perform social services unless these services are specifically contracted and raid for, and that eliminate all artificial barriers to entry into commercial activities. (4) Absence of a regulatory system that rewards and punishes on the basis of performance. In the private sector, the fortunes of senior managers are relatively closely linked to those of the company. In the public sector, where this linkage is much weaker, some administrative mechanism of rewards and punishments on the basis of performance must be substituted to force managers to be equally aggressive in seeking maximum profits. The present regulatory system does not accomplish this. To the extent it rewards and punishes managers at all, it is more for compliance with regulations and procedures than for commercial results. The report recommends inclusion of an incentive scheme in each contract plan negotiated with parastatals, performAnce to be judged by an expert coumittee. For the bulk of parastatals that do not have contract plans, a more general procedure is recommended. (5) Absence of an Authority to Implement Structural ChanRes. Present institutional arrangements include numerous agencies with overlapping responsibilities and no one with clear authority to implement changes in the system. To force this system to change, authority must be shifted upwards and downwards--upwards towards a central agency with the power to deve'op, implement and enforce a new set of policies and downward towards parastatal managers who, just because they have more authority can be held more accountable for results. The report recommends the establishment of a small agency headed by someone with the rank of Minister or Principal Secretary with the authority to develop and implement--or assign responsibility to other agencies to implement--the policy changes recommended. This agency, which to be effective may have to report directly to the President, should be considered temporary, operating only until the sector has been restructured and the line ministries' capacity to properly manage and supervise parastatals has been strengthened. The report also recommends a reduction in the number of ministries, holding companies and special agencies involved in parastatal affairs as the size of the perastatal sector is reduced and an increase in the freedom of managers and boards to establish independent policies for their parastatals, to be accomplished by shifting from reliance on uniform regulations to reliance on guidelines and negotiated arrangements. - 1 - CHAPTER ONE HISTORY. DESCRIPTION AND PERFORMANCE 1.01 The Tanzanian Government has become increasingly concerned about the performance of its parastatal sector and has taken a number of steps to deal with the situation. Wbile these steps have yielded some promising results, more stringent measures are needed and are being considered. The goal of this report is to assist with these deliberations by reviewing the problems of the sector and their causes, what the Government has done so far, how effective and appropriate these measures have been, and what else needs to be done to complete the process. The focus is on the overall policy environment within which parastatals operate rather than on indi- vidual institutions.l/ 1.02 Parastatals, as the term is used in Tanzania, include commercial enterprises in which the Government owns 50 percent or more of the shares plus regulatory bodies, marketing boards, holding companies and noncom- mercial agencies such as educational and research institutions set up under one of several acts of Parliament. These acts include the Companies Ordinance (Cap.212) which authorizes the establishment of limited liability companies for commercial purposes (most parastatals were established under tLis act); the Public Corporations Act, 1969 (which empowers the President to directly establish public corporations); the District Corporations Act, 1973 (corporations involved in trade and promotion of economic development at the district level); The Corporation Sole Act, 1975 (which established the Dar es Salaam Water Corporation Sole plus twc other organizations that are semi-autonomous and semi-commercial departments of sector ministries); and specific acts of Parliament (which established the Tanzania Harbours Authority, the Tanzania Railways Corporation, Air Tanzania, the National Urban Water Authority and the marketing boards). 1.03 This report is primarily concerned with commercial parastatals, those expected to generate a surplus through the sale (to the public or private sector) of the goods or services they produce. It will, however, review and comment on non-commercial parastatals as appropriate. Many of the comments and recommendations made about commercial parastatals will apply to other categories of organizations that operate within a similar policy and regulicory environment, for example, enterprises in which the Government has a minority but still a controlling interest and coop- eratives, which are considered part of the private sector but which receive, at times, preferential treatment from the Government (e.g., in the allocation of credit, donor finances, vehicles and equipment). 11 More detailed studies of parastatals in specific sectors are available in other World Bank reports. Most noteworthy, are Chapter III and Annex 3 of Tanzania: Agricultural Sector ReDort, 1983, and Appendix B of Tanzania: An Agenda for Industrial Recovers, 1986. - 2 - Evolution of Parastatal Sector 1.04 Up to the time of independence in 1961, the economic development that occurred in Tanganyika was largely the result of private initiative. The colonial Government provided administrative services and infra- structure, assisted some enterprises by purchasing minority shares and established crop marketing boards for major crops such as cotton, coffee and sisal. It had majority shareholdings in only three enterprises, a diamond mine, a salt mine and a meat packing plant. These shares were held by the Treasury Registrar, an office established in 1959 to manage the Government's portfolio. For the most part, the Government kept out of all management decisions and did little to regulate or even monitor the activ- ities of these enterprises. 1.05 With some modifications, this pattern continued until 1967. A number of additional parastatals were established, among them the National Housing Corporation, the National Insurance Corporation, the National Development Corporation (originally the Tanganyika Development Corpora- tion), a holding company to manage and promote the further development of state-owned enterprises, plus several additional marketing boards and financial and banking institutions. In 1966, there were about 43 para- statals; more than 2/3 of their assets were in the electricity and mining sectors, only 10 percent was in manufacturing and the remainder were spread over other sectors--corstruction, tourism, agriculture, commerce and finance. But overall the Government vas optimistic about its ability to attract foreign investors and donors and assumed that "the main thrust in industrial investment was to come from private investment" with the public sector 'filling gaps left by private investmentso.2/ To encourage private investors the Government inzroduced a number of incentives, for example, accelerated depreciation allowances, tariff protection and guarantees of profit repatriation. And the First Five Year Plan (1964-69), while adopting an ambitious public investment program, nevertheless assumed a private investment share of 75 percent in the industrial sector. 1.06 By the mid 60'., however, the Government had become disillusioned with reliance on private investment. The inflow of private capital from abroad was far less than expected and what little there was continued to go into traditional areas rather than broadening and diversifying the indus- trial base. At the same time, the political and social philosophy of the mid 60's called for independence, self-reliance and greater emphasis on social objectives such as employment generation and rural-urban equality; 2/ Figures from W. Edmund Clark, Socialist Development and Public Investment in Tanzania. 1964-73, University of Toronto Press, 1978. Quotes from the First Five Year Plan and a policy statement by then- President Nyerere introducing the Second Five Year Development Plan (1969-1974) included in J. V. Mwapachu, Manaxement of Public Enterorises in Developing Countries, Oxford & IBH Publishing Co, New Delhi, 1983, p.12. while the economic philosophy of the time called for the use of public ownership and central planning and control as the principal instruments to achieve these objectives. This philosophy was spelled out in the Arusha Declaration of February, 1967 which called for public ownership of all important economic activities. The goal of this new policy was to gain control of the economy so that it could be mobilized to serve a number of socioeconomic objectives, namelyt -- to promote self-reliance; -- to force the pace of economic development by mobilizing and generating surpluses for investment that might otherwise be used for consumption or leave the country; -- to change the pattern of development, placing more emphasis on rural and agricultural development and production to meet basic needs; -- to distribute the fruits of development more equitably by increasing productivity, employment and the standards of living of the masses, concentrating on the production of basic needs and eliminating profiteering and exploitation; and - to accomplish these ends efficiently, that is, at minimum cost to the economy. Other routes to controlling the ecor-my--reliance on regulation of the private sector and public investment alongside private investments--were generally rejected, the first, apparently, because of doubts about the ability of the Government to regulate the private (often foreign-owned) sector and the second because of shortages of skilled manpow-er and capital which, in the view of planners at the time, placed a premium on avoidance of duplication. 1.07 The call in the Arusha Declaration for greater emphasis on rur.' development and agriculture was not implemented during this period. Instead, the Second Five Year Plan (1969-74) continued the emphasis on industrial development--in contrast to the First FYP, however, the public sector was to account for 88 percent of total investment--and the Third FYP (1974-79) adopted what was called The Basic Industrial Strategy. The goals of this strategy were to reduce external dependence by import substitution, give industry priority over other sectors and, in particular, promote capital and consumer goods industries. In effect, this policy meant con- tinued large public investments in the manufacturing sector from the mid- 609 through most of the 709. 1.08 The result of the Arusha Declaration was rapid proliferation of public sector organizations and assets. The number of parastatals increased from about 43 in 1966 to 73 by the end of 1967, 380 in 1979 and over 410 at - 4 - the present time. Between 1964 and 1971, parastatal assets grew 5.5 times. About 30 percent of this increase occurred through nationalizations, most of which took place in 1967 and 1968. Forty percent of the assets acquired in this way were in the commercial sector (most involving the acquisition of real estate), 36 percent were in the manufacturing sector and the remainder were split roughly equally between agriculture and finance. The remaining 70 percent of the increase in parastatal assets during this 1967-71 period involved an actual increase in capacity, half because of the establishment of new firms and half because of the expansion of existing firms. Of the assets created by investing in new firms, half went into manufacturing, a quarter into transport (most of which went to develop the Zambian corridor in the wake of the Unilateral Declaration of Independence by the white controlled Government of Rhodesia in 1965) and 13 percent to tourism. Expansion of existing capacity occurred mostly in manufacturing, electricity, commerce and construction.31 1.09 Since 1971 parastatal growth has been much slower. Little was left to nationalize, fewer joint ventures were established and domestic and foreign savings became increasingly difficult to mobilize. A larger fraction of parastatals established after this period were holding com- panies, regulatory bodies, non-commercial organizations (such as educa- tional and health related agencies) and regionally oriented enterprises (regional transport companies, regional trading companies and district development corporations), most of which are quite small. While no compre- hensive estimates on growth in assets could be found for this later period, this picture is consistent with time series data indicating a fairly steady decline in parastatal fixed capital formation--in constant price shillings, as a percent of GDP and as a percent of total investment--from 1971 through 1985 (see Figures 1 and 2).4/ It is also consistent with information on parastatal assets in the manufacturing sector that indicates a dramatic slowdown in the rate of increase from the early 70s to the early 80s, a slowdown that would be even more dramatic if the figures had controlled for inflation over this time period.5/ 31 See Clark, op. cit. 41 See note to Figures 1.1 and 1.2. 5/ Parastatal fixed assets in the manufacturing sector increased from sh 21 million in 1964 to sh. 971 million in 1971, according to Clark (p.103), and to sh. 2,85.1 million in 1979 and sh. 3,278.4 million by 1981, according to unpublished data from the Bureau of Statistics presented in World Bank, Tanzania: An Agenda for Industrial Recovery, Vol. II, p.119. This implies an annual percent change (and absolute change) in the period 1964-71 of 73 percent (sh. 136 mill.), in 1971-79 of 15 percent (sh. 252 mill.) and in 1979-81 of 5 percent (sh. 146 mill.). Considering inflation rates during this period, parastatal fixed assets in manufacturing in constant prices may actually have decreased in the last few years. Fig. 1.1 FIXED CAPITAL FORMATION BY INSTITUTION: PERCENT OF TOTAL II'- Private To Other Public . . 40 E 30 -_S Z- Parastatals OS. * . I .. . ...... - 71 2 73 74 75 .7 77 7t 79 30 II 92 93 94 II Fig. 1.2 FIXED CAPITAL FORMATION: MILLION SHILLING, 1976 PRICES 4. | ~~Private Parastatals 7J 72 73 74 75 7J 77 7? 7O sO el 32 3 34 1S Notes The icrease from 1976 to 1977 Is misleading since a revised series of national accounts ws started In that year. A continuation of the olt series Indicates a continued decline, albeit at a *lower pace, from 1976 to 1982. - 6 - 1.10 This growth of the public sector was accompanied by the introduc- tion of a host of complex economic regulations and controls. The most important were central control of investments, administrative allocation of foreign exchange, imports and credit, price and wage controls, and 'confinementu--the assignment of exclusive rights to perform certain economic activities to various public sector enterprises. The prolifera- tion of such controls and confinement policies, many of which resulted in preferential treatment for public enterprises, has no doubt adversely affected the private business climate. While comprehensive figures are not available before 1971, private investment appears to have been low from the mid 60s to the mid 70s. Thereafter, however, it appears to have increased (see Figures 1.1 and 1.2 and Appendix Table A4.1). It is not clear why this might be the case. One explanation is that by the mid 70s parallel (un- regulated) markets had developed to the point where private investment once again became profitable, despite limited access to foreign exchange and credit at low official prices. Another is that these figures are misleading since the cost of private investment has been rising more rapidly than that of public investment (since private firms deal to a greater extent in parallel, uncontrolled, markets and parastatals more at official prices which have been rising less rapidly); if both could be deflated by more appropriate price indices, the picture might be quite different. Principal Characteristics 1.11 The net result of these developments is a very large, diverse and complex parastatal sector. Only in countries as large as Brazil (six times the population and 50 times the GDP of Tanzania) and Mexico (3.6 times the population and 35 times the GDP) does one find more than 425 parastatals. Pardstatal value added as a percent of GDP, which rose from about 2 percent in 1964 to 7 percent in 1967 and 9 percent in 1972 (or 12 percent of monetary GDP), now stands at about 13 percent.6/ This figure is about average for a sample of 34 developing countries that includes Zambia, Nicai-agua, Egypt, Venezuela and Tunisia but is 3 points above average if these special cases are excluded.7/ The share of nonagricultural wage employment by nonfinancial parastatals was 32 percent in 1978; only Zambia and Benin in a worldwide sample of 34 countries have higher percentages.8/ 6/ Data differ substantially depending on source. The figures for 1964-72 come from Clark, op. cit. p.63. The latest figure (for 1984) is computed from the Ministry of Finance and Economic Planning, The Economic Survey. 1985. A study by L. A. Msambichaka, S.M.H. Rugumisa and J.J. Semboja, The Role of the Public Sector in Development- -Tanzania, Economic Research Bureau, University of Dar es Salaam, June, 1985, places this figure at 19 to 20 percent of monetary GDP for the years 1972 to 1979 and 25 percent in 1980 71 Mary M. Shirley, "Managing State-Owned Enterprises", World Bank Staff Working Papers, Number 577, p.95. These cases are special because of publicly owned copper and oil companies that dominate these economies. S/ Employment by all parastatals as a percentage of total wage employment was estimated at 1.4 percent in 1964, 7.5 percent in 1967, 18.3 percent in 1972 and now stands as about 24 percent. Fixed capital formation by Tanzanian parastatals accounted for 50 percent of total investment in the early 70's and 35 percent in the early 80's, compared to a weighted average of 11 percent for industrial countries and 27 percent for developing countries.9/ 1.12 Parastatals are even more important in some sectors. More than a third of parastatal assets are concentrated in the energy and minerals sector, slightly less than a third are in industry and trade and 14 percent in agriculture (see Table 1.1). As Table 1.2 indicates, parastatals generated 47 percent of both value added and employment in the manufacturing sector and 37 percent of value added and 68 percent of employment in the transport sector. Among other things, they produce all of the cement, beer, fertilizer, and refined sugar; import all the petroleum, fertilizer and motor vehicles, and up to recently had exclusive rights to trade in certain commodities. Table 1.1 GOVERNMENT SHAREHOLDING IN PARASTATALS, JUNE 30, 1986 Paid up Share Capital Percent Sector (TSh. Million) of Total Financial Institutions 408.7 4.6 National Resources/Tourism 637.6 7.2 Communications and Works 624.8 7.1 Agriculture & Livestock 1228.5 14.0 Energy & Minerals 3249.7 36.9 Industries & Trade 2656.3 30.2 TOTAL SHARE CAPITAL 8,805.6 100.0 Sourcet Ministry of Finance, Treasury Registrar, Tanzania 9/ Shirley, 1983, pp. 96 and 97. - 8 - Table 1.2 PARASTATAL CONTRIBUTION TO SECTOR VALUE ADDED AND WAGE EMPLOYMENT Sector Value Wage Added Employment Agriculture 1.0 13.6 mining 21.3 55.1 Manufacturing 46.9 46.5 Electricity 21.2 4.5 Construction 28.4 36.4 Commerce 29.1 53.3 Transport 20.3 87.0 Finance and Real Estate (a) 0.2 0.7 TOTAL 12.8 24.4 (a) If real estate were excluded, these figures would be close to 100 percent. Sources Economic Survey, 1985 1.13 The sector is also very diverse. It includes large companies (for example, Aluminum Africa with over TSh. one billion in annual sales and over 1000 employees and Kilombero Sugar Company with over 6000 employees) as well as quite small ones (for example, THB Estates with 27 employees in 1985); natural monopolies (Tanzania Harbours Authority and Tanzania Electric Supply Company) and potentially competitive firms (Tanzania Sewing Thread Manufacturing Company); noncommercial institutions that are clearly important (University of Dar es Salaam) and others that are seemingly unimportant (Audio Visual Institute); and firms in every conceivable line of business (including consulting companies, marketing boards, trade associations, tourist promotion agencies, hotels, safari camps, institutes to promote local languages and culture). Appendix Al provides a relatively complete list with selected characteristics. 1.14 While a third of the parastatals are affiliated with the Ministry of Trade and Industry and 22 of these are supervised by the National Development Corporation, the largest holding company, virtually every other ministry as well as the offices of the Prime Minister and the President have some parastatals under their jurisdiction (see Tables 1.3 and 1.4 for a summary, and Appendix Al for a detailed picture). Table 1.3 THE PARASTATAL SECTOR IN TANZANIA Ministry No. of Of Which: Non- Affiliation Parastatals Holding Commercial Commercial Agriculture & Land Devt. 61 5 53 3 Communic.& Transp. 26 1 20 5 Fducation 7 0 1 6 Finance & Planning 20 0 15 5 Health 3 0 0 3 Mineral & Energy 22 2 20 0 Labor & Manpower 4 0 0 4 Nat. Res.& Tourism 36 0 31 5 PM's Office 85 0 76 9 President's Office 6 0 4 2 Trade & Industry 131 10 112 9 Water 1 0 1 0 Other a/ 11 0 6 5 TOTAL 413 18 339 56 a/ Includes Defense, Foreign Affairs, Judiciary and Land, Housing and Urban Sources Table Al. - 10 - Table 1.4 SUPERVISION OF PARASTATALS IN TANZANIA NUMBER OF SUPERVISING AGENCY PARASTATALS Agriculture and Land Cashewnut Authority of Tanzania 2 NAFCO 10 NMC 2 SUDECO 4 Tanzania Sisal Authority 5 Ministry 33 Total 56 Communications and Transport NTC 9 Ministry 16 Total 25 Finance and Planning Ministry 20 Total 20 Minerals and Energy STAMICO 14 Tanzania Petrol Devt. Corp. 4 Ministry 1 Total 19 Natural Resources and Tourism Ministry 36 Trade and Industry BIT 8 NCI 7 NDC 22 TEXCO 12 SIDO 1 SMC 13 TBL 3 TKAI 5 TLAI 6 SARUJI 11 Ministry 24 Total 112 Sources Table Al. - 11 - Parastatal Performance 1.15 The parastatal sector has a number of significant accomplishments to its credit. Through them the Government has eliminated private, especially foreign, economic power centers that might have controlled the economy. Parastatals have filled gaps left by the private sector, eliminated some sources of profite4ring and established a significant, albeit capital intensive, industrial base where little existed before. And the parastatals have served as a training ground for a new generation of indigenous business managers and executives; had the private sector been relied upon, fewer Africans would have received this training and it would have occurred more slowly. But the parastatal sector has not, or only partially, fulfilled other goals, and sometimes done so only at high cost. 1.16 Self reliance. Increased self reliance means reduced dependence on imported consumer and capital goods and on foreign experts and managers. Clearly, substantial progress has been made in reducing dependence on foreign expertise, although the education sector, not the perastatal sector, deserves principal credit. So far as the other two dimensions are concerned, progress has been far less than hoped for. Tanzania freed itself from dependence on foreign investors but has become more dependent on foreign banks and aid donors. It produces a much wider range of goods than in the 60's, but the import content of these commodities is high, sometimes higher than had the finished goods been imported. Statistics on imports reflect these trinds. As a percent of GDP, imports have remained above 25 percent from 1965 through 1980; they dropped below that level during the next few years (mainly because of severe foreign exchange shortages) but then rose again in 1986. Imports minus exports as a percent of GDP have been above 9 percent for 11 of the past 15 years (the only exceptions being 1972173 and 1976177); it was 10.5 percent in 1985 and 14.5 percent in 1986. In the manufacturing sector, imported input costs as a percentage of total input costs (excluding oil) increased from 15 to 53 percent between 1961 and 1984.10/ 1.17 The pace of development has not increased. From 1966 to 1976, GDP in constant prices grew at 4.6 percent per year, little more than 1 percent above population growth. Since then, it has grown at 1.7 percent per year with the result that GDP per capita is lower today than it was in the early to mid 70's. Value added and production in the manufacturing sector, where investments by parastatals have been concentrated, has actually declined between 1976 and 1986, from 3.1 to 1.0 billion shillings (in 1976 prices), a decline of 68 percent. 10 Government of Tanzania, Economic Survey, various years, and World Bank, Tanzania, An Agenda for Industrial Recovery, Volume I, p. 10 (henceforth cited as Agenda). These figures are underestimates since they are based on official prices; if imports and exports were valued at parallel market rates for foreign exchange, these percentages would be much higher,particularly in later years. - 12 - 1.18 Throughout most of this period, gross fixed capital formation has been decliiing, from 24 percent of GDF in 1976 to 16 percent in 1985 and 17 percent estimated for 1986. Most of this decline results from declines in parastatal and central Government investments; from 1977 to 1985 these components declined in constant prices by 65 and 60 percent respectively, whereas private investments declined by 22 percent. 1.19 Over the same period the amount of domestic savings mobilized from GDP to finance this investment also declined, from 20.8 percent in 1976 to 6.8 percent in 1985 and as low as 2.2 percent in 1986. Thus, the economy has become ever more dependent on foreign aid and loans to finance invest- ment, not less so as envisioned by the authors of the Arusha Declaration. A fall in retained profits out of value added generated by parastatals, from 39 percent in 1976 to 25 percent in 1984, accounts for some of this decline; but the bulk of it results from increased current account deficits by the central Government (see Appendix Table A4.2). Savings as a percent of value added generated by the private sector do not appear to have declined (but the evidence is weak since the only way to calculate private savings is as a residual which includes some public savings and many errors and omissions). 1.20 Dissavings by the central Government is significantly related to the failure of parastatals to contribute much on net to their owner. As Table 1.5 indicates, over the five year period ending in 1985/86, there was a net transfer on current account of only 991 million shillings. This figure includes corporate taxes which these firms would have paid to the Government even if they had been privately owned; if these taxes are excluded, there has been a net transfer from the Government to parastatals of 7.8 billion shillings over this period. Very few parastatals remit any dividends to the Government;ll/ indeed, more than three fourths are con- tributed by the Bank of Tanzania (see Appendix A5). Most subsidies are paid to crop authorities; in earlier years they were meant to cover differ- ences between purchasing and selling prices, but after 1984, when the decision was made to cease providing subsidies, they have been provided to cover past debts and interest obligations, principally of the National Milling Corporation.12/ Grants and subventions, defined as payments to non- commercial parastatals (e.g,, TSh. 318.5 million for the University of Dar es Salaam and 15.5 million for the Tropical Pesticides Research Institute 11/ Subsidiaries of holding companies are supposed to pay dividends to their holding companies which after taking care of necessary expenditures and capital projects, are supposed to remit dividends to the Government. In practice holding companies remit very little. 12/ However, these figures are probably underestimates since they do not include line items in parastatal accounts that appear to be in the nature of subsidies. For example, the National Transport Company accounts for 1985 show an income item of TSh. 3.9 million titled 'Trucking Project Funds from Treasury' and the Aluminum Africa Corporation had an 'Export Subsidy' of TSh. 4.3 million in its 1984 accounts. - 13 - in 1985/86), appear to be growing, though not in real terms. The major tax contributors are the brewery, cigarette and textile companies. The figures in Table 1.5 do not include interest payments and short term loans on current account and grants and loans on capital account which, taken together, further reduce parastatal contributiond. Table 1.6 TANZANIA: RESOURCE FLOWS BETWEEN GOVERNMENT A PARASTATAL SECTOR 1981/82 - 1986/86 (T nsnle Shillnings Millions) Not Transfer* to Government Dividends Grants Including Exeluding Number CorDorat . C -r"Fate Cor orat RitVtinc Amount Tex** Subsidies Subventions Tixiea Tix 1981/02 16 199.0 1186.0 965.1 S16.7 (89.9) (1874.8) 1062/68 9 888.6 1848.0 1,888.6 844.6 (496.5) (1844.5) 1988/84 9 862.1 1S16.0 1,822.1 696.0 (841.0) (1656.0) 1984/86 11 406.7 1e66.6 784.4 866.5 468.8 (1216.2) 1965/86 19 444.1 2959.0 714.6 1283.3 1455.2 (1508.6) Soure : Ministry of Finance, Treasury Registrar, Tanzania 1.21 The composition of output has not changed in the directions desired. Despite heavy investments to create capacity to produce capital and intermediate goods, particularly in non-traditional lines like chem- icals, fertilizers, rubber, plastics and metals, the composition of manu- facturing output judged in domestic prices has changed only modestly, capital goods increasing and consumer goods decreasing as percentages of value added in manufacturing between 1965 and 1983. Indeed. if judged in international prices, it has changed in perverse ways: in these terms, between 1965 and 1984 the share of value added contributed by the produc- tion of consumer goods increased from 56 to 85 percent while that of capital goods increased from 3 to 4 percent and that of intermediates decreased from 40 to 11 percent. The difference in outcome when judged by international prices is because the structure of protection is biased in favor of the most inefficient firms, to keep them alive; and such firms are concentrated in the intermediate and capital goods sectors.13/ 13/ Agenda, op. cit., Volume I, pp.3-4. - 14 - 1.22 The parastatalf employment--80,000 in 1974 and 176,000 in 1984 (24 percent of total wage employment in 1982)--but it has done so at considerable cost. Had the same amount of capital been invested in less capital-intensive and less foreign exchange-intensive ways, and had per capita output increased rather than declined during the last decade, employment op tunities would have been far greater. 1.23 Labor income, instead of increasing in real terms, has declined steadily since the early 70's; and parastatals appear to have led the way. Table 1.6, which covers all sectors up to 198C and the parastatal sector thereafter, indicates how dramatic that decline has been: more than a 70 percent decline for all categories of workers.141 During the 70's, para- statals in the industrial sector may have paid workers more than did the private sector, as one study of that period indicates.151 But all other evidence for the 80's suggests that parastatals paid workers less and that the differential videned.16/ 14/ It is difficult to believe that total income has fallen this much. One possibility Js that nonwage benefits have increased in real terms. Appendix B suggests that they have at least for higher salaried personnel for whom real wages have fallen the most. Another possibility, also suggested by Appendix B and probably of more significance, is that second jobs have become increasingly important. But it is difficult to believe that these two factors could have kept total income from declining significantly. 15/ See Msambichaka, et. al., The Role of Public Sector in Economic Develonment - Tanzania, Phase Two, September, 1985, Table 2.2, which surveyed 100 public and 30 private industrial firms and covered the period from 1975 to 1979. 16/ Unpublished data from the Bureau of Statistics covering medium and large manufacturing companies in 1981 indicate that parastatals paid operatives 7 percent less and non-operatives 27 percent less than private firms. See Agenda, op.cit., Vol.II, Table A 2.5. p. 103. See also Table A 2.10, p.107 for a sector breakdown. A sample of sisal estates (three private, two public and one joint) covering the period 1981-1984 also indicates that parastatals paid lower wages than private estates. See Semboja, et. al., The Role of Public Sector in Economic Development - Tanzania. Phase III. The Sisal Sector, November, 1986. Finally, two sets of data covering the period 1976 through 1984 indicate that parastatal labor costs per employee in constant prices declined substantially whereas per capita income for the whole economy remained roughly constant during this period. The first set involve responses to a Bureau of Statistics questionnaire sent to all para- statals indicating that remuneration per employee in constant prices declined by about 46 percent. See Bureau of Statistics, Analysis of Accounts of Parastatal Enterprises. 1973-1982, plus unpublished materials for 1983 and 1984 and the Economic Survey for price deflator and per capita national income. The second set, again from the Bureau of Statistics but covering only manufacturing parastatals, indicates that there was a decline of 63 percent in labor costs per employee over - 15 _ Table 1.6 INDEX OF REAL WAGES FOR REPRESENTATIVE GROUPS, 1969.100 Minimum Average Middle Top Sectors Year Wage Wage Salary Salary ------------------------------------------------------------- All 1968 90 97 95 102 1969 100 100 100 100 1970 96 103 103 93 1971 93 103 105 88 1972 119 99 106 75 1973 109 102 96 67 1974 135 133 91 56 1975 103 108 68 44 1976 83 104 64 39 1977 71 98 58 33 1978 61 86 51 30 1979 58 77 45 26 1980 I 63 j 65 37 21 Parast. 1981 60 56 24 18 only 1982 49 47 24 14 1983 40 39 17 12 1984 46 31 16 11 1985 31 HA NA 9 Notes The figure for average wage in 1979 was estimated by assuming that the average percent change in other categories applies to this case as well. Numbers for 1981-85, plus that for 1980 average wage, are for public sectors only, are based on 1977 prices, and may not include an adjustment for this. Sources For 1968-80, T.R. Valentines 'Wage Adjustments, Tax Rates, Accelerated Inflation in Tanzania." ERB, 1983. For 1981-85, plus 1980 average vage, National Productivity Council, Productivity and Operations Report for the Year ending 30 June 1986. 1.24 Parastatal service to the masses has also been disappointing. Publically owned shops in remote areas, while charging lower prices than private traders, have been virtually devoid of goods. Half the buses in Dar es Salaam and abo Footnote 16 Cont. the period. See Agenda, oy.cit., Vol.11, p.101. While these studies are supposed to take into account fringe and non-monetary benefits, they may not do so completely, and there is some evidence that parastatals provide higher level persontel with more of such benefits--for example, housing--than do private firm.. But this qualification does not apply to less senior staff who receive few such benefits. - 16 - Transport Corporation are out of service. Crop authorities were abolished because of their failure to provide farmers with the services they were created to deliver. The railroad can meet only 10 to 15 percent of the demand for freight services. Such examples can be multiplied manyfold. 1.25 Finally, the parastatal sector is inefficient, consuming more resources than necessary to produce the output it creates. Evidence for this statement comes from financial data, economic rates of return and partial indicators of efficiency. Financial data are the most widely available but only indicate ability of a firm to take advantage of the existing economic situation which includes price controls, subsidies, pro- tection from imports and so on; they do not indicate whether a firm is profitable to the nation as a whole--for example, how well it would do if it were not subsidized or protected at the expense of other parts of the economy. For that, one needs economic rates of return, that is, returns based on prices likely to exist in open, competitive markets. Only one recent study using such methods is available at the present time, and it is limited to data on firms in the manufacturing sector in 1984. rartial indicators of efficiency, for example, labor productivity and incremental capital-output ratios, can be used as rough indicators for other sectors and to support other findings, but can be misleading if used in isolation. All three sources of information support the general contention that the parastatal sector is inefficient, both by absolute standards and as com- pared to the private sector. 1.26. Financial Performance. Given the preferential treatment most parastatals enjoy, one would expect them to earn reasonably good financial returns. Some do. But many do not. The most comprehensive data on such returns comes from the Tanzania Audit Corporation (TAC) which produces aggregated data for its clients with acceptable accounts. As Table 1.7 indicates, aggregate profits for this group of companies was negative in 1984. Even in 1985 when profits were positive, though still small, nearly half of the 354 companies audited made losses. Highly unprofitable para- statals include not only the Crop Authorities, but important commercially oriented enterprises such as Steel Rolling Mills Ltd., Kilimanjaro Textile Corporation Ltd., Mwanza Textiles Ltd., and transport monopolies such as the Tanzania Railways Corporation. Enterprises with profits exceeding TSh 5 million a year include the National Bank of Commerce and the Bank of Tanzania, the Tanzania Cigarette Company and the Tanzania Breweries Ltd. A full list of firms with losses or profits over TSh. 5 million is shown in Appendix G. In its Annual Report for 1985, the TAC commented on this performance as follows: "...when consideration is given to total investment in the parastatal sector, the amount of profit earned is far from satisfactory. When it is further considered that the biggest loss makers are in the sectors of industries and trade, agriculture and livestock and communications, where the largest investments are made, the position is further tarnished. In the absence of set rates of return on investment, many parastatals seem content when they just barely break even.' - 17 - Table 1.7 TANZANIA: FINANCIAL PERFORMANCE OF PARASTATALS AUDITED BY TANZANIA AUDIT CORPORATION, 1988-05 198a 1984 1985 ProfIt Before Tax Frofit Before Tex Profit B fore Tax Numbr. (TSh. Million) Number tTSh. Million) Number (TSh. Million) Pereetetele with 196 8,876.5 21$ 2,911.9 180 8,811.4 Pereetels with 165 (910.2) 171 2!D%4.4. 165 (1,58.0) Net Profit So1 2966.8 a84 (82.5) 854 1958.4 Beurees Tanzania Audit Corporation Annual Report 1.27. While there are large variations in sector and individual company *zperiences, financial performance on average appears to have worsened during the past decade. Given the changes currently under way in the macroeconomic environment--in particular the elimination of subsidies, devaluation, increasing interest rates, reduction in the number of goods subject to price controls and deconfinement--this deterioration is likely to continue but at a faster pace for a while (see discussion in Chapter 2). The variance in experience, however, is also likely to increase: para- statals most heavily dependent on imports are likely to be hit the hardest while others more able to export, plus firms related to the agricultural sector, are likely to find their financial performance improving. Appendix AS provides details on financial performance over time for selected groups of, as well as individual, companies. 1.28. Economic Efficiency. The most up-to-date and comprehensive infor- mation on economic efficiency--and one that permits a comparison of the public and the private sectors--comes from a recently completed study of the industrial sector.17/ For present purposes, the most relevant findings from that study are reflected in Table 1.8. Fifty four percent of all the 171 Aaenda, on. cit., especially Vol. I, p.43ff and Table 16, Vol.II, p.139. This study surveyed 118 product lines in the industrial sector, 30 percent of which (on a weighted basis) are in the parastatal sector. - 18 - parastatal activities in the industrial sector are extremely unproductive in that they produce negative value added when all inputs are valued at world market prices. This compares to a figure of 11 percent for the private sector. On the same basis, only 18 percent of these parastatal activities are efficient in the sense that they use less inputs than the value added of their output, compared to 43 percent of the private sector industrial activities. Table 1.8 PERCENTAGE OF PRODUCTION LINES OPERATING AT DIFFERENT LEVELS OF EFFICIENCY, BY OWNERSHIP, 1985 A. Percent of All Activities Domestic Resource Cost /a Less Greater than Infinity Sector than One than One /b (Net. value added) Total Public 14 16 35 Private 10 21 4 Total 24 37 39 100 B. Percent of Sector Domestic Resource Cost /a Less Greater than Infinity Sector than One than One /b (Neg. value added) Total Public 22 24 54 100 Private 28 61 11 100 Total 23 27 50 100 la Short-run Domestic Resource Cost: cost of domestic factors (labor and capital assumed at sunk cost, shadow priced) for generating one unit of world priced value added calculated at actual levels of capacity utilization. Calculations made at capacity rates attainable if foreign exchange for recurrent inputs were not a constraint are not significantly different. lb Excluding infinity. Source: Agenda. OD. cit., Vol. II, Table 13, p. 136. The most likely explanation for these findings is that parastatals enjoy preferential treatment and bail-out" privileges that permit them to remain - 19 - in business longer than can private firms when they are operating ineffi- ciently. 1.29. Partial Indicators of Efficiencv. Tables 1.9 and 1.10 provide summary indicators permitting comparisons of public and private firms in the industrial sector during the period 1975-79 and on sisal estates over the period 1981-84. In both cases, output per worker (or per manhour) is roughly half that of private firms. In the industrial sector, this occurs despite the fact that output per unit of capital employed is about the same. Taken together, these two findings suggest the existence of substan- tial redundant labor in the parastatal sector. It is also interesting to note in Table 1.9 that manhours worked per employee are less even though labor costs per employee are more amongst public firms in the industrial sector.18/ Moreover, amongst sisal estates, parastatals receive a lower price for their output, probably because quality of their product is lower. Appendix A3 provides information suggesting that sales in constant prices per employee of parastatal firms in nearly every sector have declined over the period 1974-84, on average by 30 percent. Additional evidence to this effect is provided in Table 1.11 which separates out the district develop- ment corporations, which perform very poorly, and financial and other service parastatals, which perform reasonably well, from all others in a sample of over 100 public enterprises.19/ Table 1.9 AVERAGE INDICATORS OF PERFORMANCE PUBLIC AND PRIVATE INDUSTRIAL FIRMS, 1975-79 Public Private Manhours per Employee 2,160.2 3,009.0 Labor Cost per Employee 13,490.0 8,355.8 Labor Cost per Manhours 5.7 2.9 Output-Labor Ratio 41,944.4 90,467.0 Output-Manhour Ratio 19.0 31.2 Labor Cost/Total Cost (2) 32.8 15.9 Capital-Labor Ratio 2.8 23.8 Capacity Utilization (5) 56.2 50.2 Output-Capital Ratio 1.4 1.4 Size of Firm - Employee per Unit 972.4 81.4 Capital per Unit (at 1000 per unit) 25,782.0 16,248.1 Sources Semboja, op. cit. 18/ These figures could be misleading. Large firms typically pay more than small firms, and as Table 1.9 indicates, the public firms are on average much larger than the private firms in the sample. 19/ The District Development Corporations have apparently never performed well. See a paper by H. P. B. Moshi (Financial Performance of Public Corporations: the Case of District DeveloPment Corporations. Economic Research Bureau, 1980) which indicates that these parastatals incurred sizeable losses throughout the mid to late 70s. - 20 _ Table 1.10 TANZANIA - AVERAGE INDICATORS OF PERFORMANCE SISAL SECTOR, PUBLIC AND PRIVATE FIRMS 1981-84 Public Private 1. Labor Cost per Employee 6,958.3 14,089.5 2. Labor Cost per Manhours 2.8 5.9 3. Output Labor Ratio 16,049.3 35,785.3 4. (a) Physical Output per Manhour 2.4 2.8 (b) Output in Shillings per Manhour 6.8 15.3 5. Price of Fiber per Ton 4,674.3 5,152.3 6. Ratio of Variable Cost to Value of Output 1.1 0.7 7. Capacity Utilization 0.2 0.4 Source: Semboja, op. cit. Table 1.11 INDICATORS OF PARASTATAL PERFORMANCE BY SELECTED GROUPINGS, 1986 Industry Indicator 1 - 4 5 6 1. Value of Production / No. of Employees 874,000 83,000 1,208,000 Size of Sample, n 44 8 8 2. Wage Bill I No. of Employees 33,000 19,000 37,000 n 82 17 41 3. Profits/ Total Assets 0.307 0.175 0.435 n 35 5 5 4. Profits / Fixed Assets 0.304 0.175 0.435 n 33 5 5 5. Revenue / Total Assets 3.225 3.064 2.792 n 72 16 34 6. Revenue / Fixed Assets 51.539 40.781 6.657 n 70 17 37 7. Revenue / Total Expenses 2.289 1.466 1.670 n 78 17 44 Notes: Industries 1 - 6 refer to manufacturing, agriculture, trade, trans- port, District Development Cooperatives, and Services, respectively. Source: Joseph J. Semboja, The Parastatal Study: Analysis of the Qualitative and Scopo Data, University of Dar es Salaam, July 1987. - 21 - CHAPTER TWO FACTORS DETERMINING PARASTATAL PERFORMANCE 2.1 Chapter 1 indicates parastatals have contributed far less to the achievement of economic and social goals than was expected of them at the time of the Arusha Declaration and what they hpve accomplished was sometimes achieved at high cost. A wide variety of factors have been put forward to explain these disappointing results. Some are common to all economic activities in Tanzania and help explain the less than hoped for results obtained from private firms as well. Others, while specific to the parastatal sector, are at a level of generality that is not operationally useful. This chapter sorts through these various explanatory factors and focuses on those that appear to have the most implications for policy. Factors Affecting All Economic Activities 2.2 A variety of external factors have contributed to the generally poor performance of the economy during the last decade. These include deterioration in the terms of trade since the coffee boom of 1978, droughts that periodically force foreign exchange to be allocated for the importa- tion of food, the war with Uganda in 1979, the breakup of the East African Community which required extraordinary budgetary and foreign exchange expenditures for a time, and declining external assistance since 1978 in both nominal and real terms (though on a per capita basis it is still amongst the highest in the world). 2.3 But of at least equal significance has been the overall policy environment within which economic activities have taken place. The Government has relied primarily on centralized administrative regulations to control the economy. Foreign exchange, imports and credit have been rationed; the exchange rate, interest rates and most commodity prices have been set administratively; and a number of economic activities--banking and insurance, and trade, transport and production of certain commodities--have been reserved for and assigned to various parastatals, effectively eliminating competitive pressures for efficiency from these markets. These regulations were sometimes used to pursue inappropriate policies. For example, during much of this period, price controls and allocations of foreign exchange, credit and budgetary resources favored industry over agriculture to such a degree that (along with other factors, for example, confinement of trade to crop authorities that proved to be inefficient) the result was an actual decline in the production of export crops and hence in foreign exchange earnings on the one side and excess capacity in the industrial sector on the other. But even where the policies pursued were appropriate, the use of administrative regulations to achieve them has introduced its own problems. It has distorted incentives in such a way as to encourage investments in lines of business in which Tanzania has little or no comparative advantage and in plants that are too large, capital intensive and import dependent. It has provided protection to firms, private as well as public, that could not survive if they had to compete - 22 - with imports. It has encouraged managers to focus their efforts on seeking additional import allocations and permission to charge higher prices rather than on efforts to control costs. search for new markets and improved technologies. It has resulted in the growth of black markets, declining public revenues, and, in the end, less rather than more control over the economy as a whole. The alternative--using a minimum of direct controls (e.g., to regulate monopolies and provide temporary assistance to a few ,infant industries") and a maximum of indirect controls to guide the economy--is unlikely to have been worse; at least it would not have provided a premium for illegal activities. Factors Specific to the Parastatal Sector 2.4 In addition to these general factors, there are a wide variety of factors specific to the parastatal sector. This section discusses only those with particularly important policy implications. 2.5 Absence of Overall Strategy and Clarity on Goals. Tanzania has no up-to-date statement of general principles20/ that can serve as a guide to parastatal operations and decision-making. Nor is there good coordination between national goals and plans and parastatal goals and plans. One consequence is a series of ad hoc decisions that, while they benefit some special interests, are not necessarily in the national interest. All too often, for example, holding companies and ministries have conceived and planned new projects to satisfy the empire-building aspirations of their managers or the desires of foreign donors and equipment manufacturers. Such behavior is difficult to argue against in the absence of clear statements about national goals and priorities. Another consequence is that managers find it easier to justify and excuse inefficiency by pointing to noncommercial objectives that central Government decision-makers may never have sanctioned. For example, some managers appear reluctant to fire redundant workers or close uneconomic activities despite the fact that it is not the main purpose of the parastatal sector to provide employment, particularly unproductive employment. 2.6 Unclear goals and performance criteria also at times have resulted in parastatals being faced with excessive or inappropriate demands. For example, they have often been faced with interference in personnel decisions, asked to provide transport or other services without appropriate compensation, or asked to serve goals for which they were not established and equipped. An important example of the latter is the National Hilling 201 Examples of statements of such principles from other countries are included in Appendix E. - 24 - Corporation which has faced this problem since its inception.21/ Such pressures wi'l always be present; but they can be reduced, and parastatals' capacy to resist strengthened, by publically clarifying and restating the goals that parastatals are expected to perform. 2.7 Excessive Number of Parastatals Given the Availability of Resources and Mana erial Cavacity. As indicated in Chapter 1, Tanzania has more parastatals than many much larger countries with greater administrative and managerial talents at their disposal. The best evidence on this score, however, is derived from the commonly accepted observation that most parastatals are underfunded and understaffed--not in terms of numbers but in terms of skills and quality of staff, particularly at higher levels--and that many regulatory bodies, because they also are underfunded and understaffed, cannot handle the work-load required to do their job well. The tendency to spread limited resources evenly and thinly serves to keep inefficient parastatals alive but only at the expense of efficient, or potentially efficient, operations. Given the limited physical and human resources available to the Government, the general principle should be to fully fund and staff important operations and not to waste resources and quality staff on unimportant, unproductive activities. 2.8 Weak Competitive Pressures and Discriminatory Treatment. Evidence from many countries shows that a key factor in determining the efficiency and dynamism of any enterprise, public or private, is the degree of compe- tition it faces. Public enterprises that compete with one another, with private enterprises and with imports tend to be more cost conscious and more prone to seek new markets and improve service and quality than those that do not face such challenges. In the past, competitive pressures from all three sources have been weak. Individual parastatals have been pro- vided with exclusive markets or territories in which to operate; private firms have not been allowed to enter these areas; and imported commodities that might compete with domestic production have been deliberately restricted. Recent policy changes, in particular deconfinement of significant areas of production and trade, have permitted some changes in this situation (see Chapter 3); but the predominant characteristics of the 21/ At different times NHC has been used as the implementing agency for Government directives not only on foodgrain prices, but on famine relief, distribution of imported food, purchases of low value cassava, millet and sorghum, purchases from GAPEX, politically determined exports and Party directives on the size of village levies. In almost all cases, the financial implications or procedures for these actions were not clearly assessed before they were adopted.' Since then, considerable thought has been given to redefining NHC's role and agreement on a new reduced role consistent with financial realities may be emerging. - 24 - system are still protection and preferential treatment, especially for parastatals. 2.9 Monopolies or near-monopolies are present in most markets at all levels of business. At the national level, The National Bank of Commerce (NBC) has a virtual monopoly in providing current and savings accounts, and complaints about poor service are widespread.221 The Cooperative Rural Development Bank (CRDB) has recently been allowed to take deposits and provide short term credit, but its interest policy is ideitical and it deals almost exclusively with cooperatives. There is only one firm producing cigarettes. three plants operated by the same company and General Manager, one source of insurance, one maker of aluminum cookware, three cement plants owned and operated by the same company and two large shoe manufacturers both owned by the same holding company. While there are numerous areas in which the size of the domestic market is too small, given economies of scale, to support more than one or two firms, this is not the case in any of the markets mentioned above. At the local level, particularly in rural areas, the situation is undoubtedly worse: most farmers are faced with one source of fertilizer and seed, one purchasing agent for their crops, and sometimes only one or two retail shop in which to purchase consumer goods, and the possibilities for exploitation as well as poor service are great. 2.10 Competitive pressure is reduced not only by erecting barriers to entry of competing firms and imports but by preferential treatment. Chapter 1 has already indicated that few parastatals pay dividends and many, until recently, received subsidies and could realistically expect to be bailed out by the Government if they got into serious financial trouble. Other areas of preferential treatment include allocation of credit and foreign exchange, privileged access to technical assistance and donor funding and price controls, and frequently, exemptions from paying tariffs. 2.11 While there is no official policy discriminating agairst private credit applicants, the effect of the credit allocation process and access to guarantees by parastatals has been to provide parastatals with most of the credit that is made available at official interest rates, which have been negative in real terms. In addition, parastatals have been able to expand credit through arrears to the banking system with greater impunity 221 Not only are transactions slow and communication poor, savings policy discourages resource mobilization. For example, opening a savings account requires references, withdrawals can only be made twice a week and while there is a high nominal interest rate on savings (more than 27 percent a month but still lower than the inflation rate and only posted twice a year), no interest is paid on amounts in normal savings accounts over sh. 200,000; the customer must put funds over that amount in a series of short term certificates or into fixed term deposits at lower interest. NBC pays no interest on current accounts, yet parastatals must maintain their current accounts with them. Some of these practices have been inherited from colonial times and are similar Continued on next page - 25 - than can the private sector. At the end of 1986. 73 percent of Tanzania Investment Banks (TIB) loans outstanding were accounted for by parastatals and 54 percent of these loans were in arrears, compared to 27 percent for private loans (see Table 2.1). At the same time, 65 percent of the loan portfolio of the NBC was to parastatals, versus 13 percent to the private sector and 21 percent to cooperatives (see Table 2.2). Most of its loans are to the agricultural sector and within that sector parastatals (mostly marketing boards) account for 72 percent and cooperatives for 26 percent. Although statistics on arrears with the NBC are not available, there is no reason to believe the situation is much different. 2.12 Parastatals also have privileged access to technical assistance and donor funds at concessional rates. At the same time, the private sector has difficulty attracting expatriates since they cannot offer them an arrangement whereby they can convert a portion of their salary into foreign exchange, as can parastatals directly or indirectly receiving donor funds. 2.13 The foreign exchange allocation system has also favored para- statals, the private sector having to purchase a larger fraction of its Imports at parallel market rates.23/ Moreover, the fact that foreign exchange applications for dividends and interest payments overseas are given low priority makes it virtually impossible for private firms to raise foreign equity or loans. 2.14 In addition some of the goods subject to price controls are officially allocated, which typically works in favor of parastatals. Transport imports such as vehicles, tires and fuel, for example, are allocated by various groups and committees at the national and regional levels. Government and parastatals, such as the railway, TANESCO, the Tanzania Fertilizer Company, receive top priority. Thus, truck imports are allocated by the State Motor Corporation (with approval of the Prime Minister's Office) and about 65 percent of the trucks imported in 1984 and 1985 went to the public sector even though about 70 percent of trucking services are provided by the private sector. In addition, parastatals receive the bulk of vehicles imported under aid projects; for example, 100 trucks were given to the Tanzanian Cotton Authority by the Dutch Government in 1984. 2.15 Parastatals have major disadvantages in facing competition as well. Because of appointment proceedures and political considerations, their boards have less flexibility in selecting, or recommending the selection of, managers. Parastatal managers have less flexibility Continued from previous page to practices in other countries; they persist because of lack of competition and inherited regulations that have not been updated. 23/ The reason for this, however, is that parastrtals are more likely to be found in high priority areas (which the private sector cannot enter); Continued on next page - 26 - Tabl. 2.1 TMKM4TIA TAINZMA DOPif5 f GMU tAI ~~TA I SuKf OPF AW 5 AT 8t Dm61. 1966 rAARATATAL H3C jj; CTA != A T L Sa C T a n T L T AN mv AS MOF wV- MONET ra AS S OF wma AOW AS6O o A S 53OF TOTAL 57AM- oAS OF MAL ATDM- IN MNOEt 57A60236 MIl MONT Dpri VA4 A1VA MON0T ANAR P6 MEN6S MR 780 T80 SN tim '80 78' OR- tamO 780 78 SYTAI- ION '600 STMCU P7O.IT UL lcow 'ams STA.m PDNTFLIOr .0 saw Do 1. She.$fafetur 1676440 son"6 40.40 39.79 62767 1060 11.40 10.a 1 50486 an"" 3.6 S. Tw0rlm NW "LsuU SSU6 U04 V8.4 O.6 11 am 17.60 0.21 s0 MM O: 7 S0.20 8. Aggleultu#. - . $Vropvogolmf saw 486409 66.611 24.16 2471 - --- 8134 4116409 66.1 4. 1a 25641 sow8 15.72 8.111 6664 6866 64.66 8.86 s1o7" 11a2" .61 a. a NW Wood V4 0114 876 47.17 2.01 186m Q - 87f6 8O9.8 6. P1.11mg sa others 24611 1slow 76.62 10."7 1161 sow 44.91 2.66 3604 247410 67.69 7 0 T A L 140461 186154 a'.s 6.04 118641 811748 27.67 16.66 4600245 166681 46.112 Tabl 12.2 TNaM-MhZA:MTtSML SAW OP CO615 CmIT PLAN StoJ6:l Leave e 67 501C swT. AT oS6 U1. 11l6 CT 11hiIISgI0. Thmaau.as) ACINICLWUE L D