United Republic of Tanzania Privatization Impact Assessment 68869 Infrastructure July 21, 2005 2 Acknowledgements This report is the output of a technical assistance and dialogue mission financed by the Private Participation in Infrastructure Advisory Facility (PPIAF) at the request of the Government of the United Republic of Tanzania. The primary objective of the mission and the report is to support the review of infrastructure privatization in Tanzania commissioned by the President of the United Republic of Tanzania, H.E. Benjamin William Mkapa. Any representations made on the performance of organizations, enterprises and individuals are not meant to be authoritative and conclusive in the legal sense. In addition, please note that this report does not necessarily reflect the views of PPIAF and/or World Bank Management. The PPIAF funded team consisted of: Onno Rühl (Team Leader), John Nellis (Senior Privatization Adviser), Nilgün Gökgür (Privatization Adviser), Roger Christen (Privatization Adviser), Lucy Fye (Senior Private Sector Development Specialist), Vedasto Rwechungura (Program Officer), Aminetou Tidiani (Junior Professional Associate), Wendy Christen and Justin Schwartz (Interns), Yeshareg Dagne (Program Assistant) and Justina Kajange (Team Assistant). The team gratefully acknowledges the cordial and professional collaboration with the Privatization Review Commission consisting of Dr. Marcelina Chijoriga (Chairperson), Dr. Ken Kwaku (Coordinator), Dr. Heavenlight Kavishe (Secretary), Dr. Hawa Sinare, Dr. Ramadhani Dau, Mr. Ali Mufuruki, Col. Joseph Simbakalia. Main authors for the different parts of the report were as follows: Roger Christen TICTS, THA, TRC and TAZARA Nilgün Gökgür TANESCO and TTCL John Nellis DAWASA and Air Tanzania Onno Rühl Executive Summary and General Report 3 Table of Contents Executive Summary 4 General Report 6 I. Context and Methodology 6 II. Infrastructure: The Situation before Privatization 8 III. Country Preparedness for Privatization 9 IV. Capacity for Infrastructure Privatization 11 ` V. Consultations and Communication 13 VI. Consultants and (Development) Partners 14 VII. The Procurement Process 16 VIII. The Big Picture: Sector Policies 18 IX. Transaction Scorecard: A Mixed Picture 21 Case Study 1 The Lease Contract between DAWASA and City Water Services 23 Case Study 2 The Private Participation in Tanzania Electricity Supply Company Ltd. 41 Case Study 3 The Partial Divestiture of the Tanzania Telecommunications Company Ltd. 61 Case Study 4 The Partial Divestiture of Air Tanzania 80 Case Study 5 Tanzania Harbours Authority (THA) 90 Case Study 6 Tanzania International Container Terminal Services (TICTS) 95 Case Study 7 Tanzania Railways Corporation (TRC) 102 Case Study 8 Tanzania-Zambia Railway (TAZARA) 123 Annex 1: Presentations to PRC 131 4 United Republic of Tanzania Privatization Impact Assessment Infrastructure Executive Summary After the earlier and successful privatization of most parastatal enterprises in the industrial and commercial sectors, at the end of the 1990s Tanzania launched the privatization of its infrastructure enterprises. By 2003, five key infrastructure enterprises (TANESCO, power; DAWASA, water; TTCL, telecom; TICTS, the container terminal; and Air Tanzania) had some form of private participation. Largely due to the recent and well-publicized failure of the lease contract for water, but also due to doubts about the telecom and airline sectors, the program for infrastructure privatization is currently perceived as not having lived up to expectations. This led the President of the United Republic of Tanzania to appoint a Privatization Review Committee (PRC) in May 2005. The PRC’s mandate was to review the privatization program for infrastructure so as to learn from experience and recommend the way forward. The present report constitutes the submission of a PPIAF-sponsored team that assisted the PRC in its review. The report follows an outline given by the PRC. Country Preparedness. Around 2000, most infrastructure enterprises were in very bad financial shape and the physical state of their assets was poor. These enterprises therefore represented a marginal business proposition to most prospective private operators. This was exacerbated by the fact that Tanzania’s investment climate, although improving, is not yet internationally competitive; many elements require further work. At the same time, whilst the national privatization policy is sound and well tested, in most infrastructure areas sector policies are spotty, and regulatory agencies are either weak or absent. The formulation and implementation of strong, nationally owned sector policies as well as their implementation is therefore a key priority. Such policies should focus especially on solutions for providing affordable services to presently unconnected Tanzanians. Capacity. Since private participation in infrastructure was a novelty in Tanzania, none of the Government agencies were adequately prepared for the challenges of dealing with private operators. Worse, pre-privatization monitoring of parastatals was weak, resulting in very poor data being available on most enterprises. Capacity for the implementation of privatization transactions was in place due to the fact that the Presidential Parastatal Sector Reform Commission (PSRC) was an experienced and competent privatization agency. The major capacity weakness is post privatization, where line Ministries, the new regulatory agencies, and the new asset holding companies are all in dire need of improved capacity. Until this deficiency is addressed, the Government will always have trouble succeeding in the complex interactions required with private operators in infrastructure; i.e., the critical issues of drafting, negotiating, monitoring and enforcing contracts. 5 Communication. Despite the fact that there is a relatively strong consensus around the policy of greater private sector involvement in infrastructure, the level of communication and debate on privatization of infrastructure has been quite low. This has hampered management of expectations in the program as well as unnecessarily burdened labor relations in privatized firms. The present review creates a welcome platform for launching a much more open debate on infrastructure policy in Tanzania. Consultants and partners. There is evidence that Tanzania has not always got the best consultants and partners, with a special weakness observed in the case of legal advisers. It is important to be vigilant of this, as having the right advice is key to getting good results in negotiating with private partners. Equally, some of Tanzania’s private operators have got away with sending teams that may not have been their best. Greater due diligence in the choice of prospective partners is therefore called for. Tanzania might also review with its development partners the wisdom of specific privatization conditionality, as this can create pressure to complete transactions at the expense of quality. Procurement. Most infrastructure privatization transactions in Tanzania attract limited competition. The Government of Tanzania and its development partners should review whether the procurement procedures used —designed for highly competitive environments— might be revised to allow for greater flexibility. If greater flexibility could be introduced, stronger oversight and control would be necessary to avoid corruption. Preventing asset erosion. Even with greater flexibility in procedures, privatization transactions almost always take much more time than is first planned. In order to avoid further erosion of the already dilapidated assets in firms that are being privatized, one option might be to engage in an interim management contract with a private provider, accompanied by an interim investment plan financed by donors. Scorecard. Based on the eight case studies done for the present review, two transactions in infrastructure in Tanzania—the management contract for TANESCO and the concession for the container terminal—are rated as clear successes. One transaction — the lease for DAWASA— is a clear failure; whilst the partial divestitures of TTCL and Air Tanzania have elements of both success and failure. For the two railroads (TRC and TAZARA) and the harbor authority (TPA), transactions have not yet been completed. 6 United Republic of Tanzania Privatization Impact Assessment Infrastructure I. Context and methodology By the end of the 1980s, Tanzania had around 400 state owned enterprises (SOEs or parastatals), many inefficient and loss-making, sustained only through subsidies from the budget and soft or directed credit. Ultimately, many proved unsustainable1. This led the Government of the United Republic of Tanzania (the Government) to a strategic decision to privatize virtually all SOEs in a program managed by the Presidential Parastatal Sector Reform Commission (PSRC). This program was highly ambitious. Despite the fact that it followed the international drive towards privatization at the time, it was unprecedented in the Tanzanian context. All the same, the degree of consensus around the concept of privatization in Tanzania was and is remarkable, and testifies to the degree of dilapidation in the SOE sector prior to privatization. There was a scarcity of goods in many sectors—beer being perhaps the most startling example—and a generally very low level of services in many key sectors, particularly in infrastructure—meaning the public service commodities; i.e., electricity, water, telecommunications and transport2. The implementation of the privatization program in the non-infrastructure sectors has progressed well. By the early 2000s, the privatization of manufacturing and commercial parastatal enterprises was virtually complete. The program has been a solid success. It has made a significant contribution to Tanzania’s strong macroeconomic performance in the last decade, partly by severely limiting the flow of subsidies to loss making enterprises, and in their place obtaining tax revenues from privatized ones. Moreover, by reversing Tanzania’s previously negative image with international and regional investors, the privatization program has contributed to the dramatic positive turnaround in GDP growth rates over the past six or seven years. However, in the same period, infrastructure enterprises continued to perform poorly (see below) and little headway was made in their reform. Thus, in the early 2000s the Government, through PSRC, has focused its attention on the privatization3 of infrastructure. 1 “A Performance Assessment and Privatisation Impact Study in Tanzania� De Chazal Du Mee for the President’s Office for Planning and Privatization, March 2004 2 Throughout this report, the term infrastructure is used to refer to both infrastructure and utilities. 3 Throughout this report, when referring to infrastructure enterprises, the term privatization is used in reference to any form of public-private partnership ranging from complete divestiture to a management contract. This is to conform to the use of terminology in Tanzania. When an actual sale is described, the terms divestiture or sale are used. 7 Considerable progress was made with transactions completed for five key infrastructure enterprises4 by 2003 in the key sectors of water, telecommunications, power, ports and air transport. These transactions were carried out in a context of high expectations, fueled on the one hand by the severely disappointing performance of most enterprises before privatization, and on the other hand by the assumption that the successes of privatization of manufacturing/commercial firms would easily be repeated. There was a high degree of donor involvement and push behind the program and, surprisingly, a relatively low level of public debate. By 2005, it has become apparent that the infrastructure privatization program has not lived up to expectations. The poor performance of the privatized urban water utility is the most celebrated case of failure, and serious doubts have been cast over a number of other transactions, notably in telecom and air transport. This led the President of the United Republic of Tanzania to appoint a Privatization Review Commission5 and to request technical support of the World Bank in support of this Commission. The objective of the review was to take stock of transactions in infrastructure so far completed, and underway, and draw lessons learned from successes, failures, and the design of proposed transactions. Based on this, the Privatization Review Commission (PRC) is to make recommendations on the way forward, both at the level of the overall program and for individual enterprises. The President explicitly stated that the objective of the exercise was not to question the overall policy stance of greater private sector involvement, but to learn from accomplishments and mistakes in order to implement the stated policy more effectively. The World Bank team assisting the Commission was in Tanzania for the better part of June 2005. It was agreed that the World Bank team6 would conduct case studies on the eight enterprises listed by the President. These were the five enterprises for which transactions had been completed, plus the Tanzania Railway Corporation (TRC), the Tanzania Zambia Railway (TAZARA), and the Tanzania Harbour Authority (THA); the last three are in the process of privatization. Each case study was the subject of a presentation showing the team’s preliminary analysis and conclusions to the PRC, followed by an open discussion. Subsequently, the PRC presented to the World Bank team a framework for general conclusions derived from the debates on the case studies. Using this framework, the World Bank team presented to the PRC a set of preliminary general conclusions, again followed by an open debate7. The present report constitutes the World Bank team’s summary submission to the PRC based on its own research and the debates with the PRC. The general part of the report follows the PRC framework. 4 DAWASA (water), TTCL (telecommunications), TANESCO (power), TICTS (container terminal), and Air Tanzania). 5 Dr. Marcelina Chijoriga, Dr. Ken Kwaku, Dr. Heavenlight Kavishe, Dr. Hawa Sinare, Dr. Ramadhani Dau, Mr. Ali Mufuruki, Col. Joseph Simbakalia. 6 The team consisted of: Onno Rühl (Team Leader), John Nellis (Senior Privatization Adviser), Nilgün Gökgür (Privatization Adviser), Roger Christen (Privatization Adviser), Lucy Fye (Senior Private Sector Development Specialist), Vedasto Rwechungura (Program Officer), Aminetou Tidiani (Junior Professional Associate), Wendy Christen and Justin Schwartz (Interns), Yeshareg Dagne (Program Assistant) and Justina Kajange (Team Assistant). 7 All nine presentations as originally made to the PRC can be found in Annex 1 to this report. 8 The report is meant to serve as technical input to the PRC. Any representations made on the performance of organizations, enterprises and individuals are not meant to be authoritative and conclusive in the legal sense. In addition, please note that this report does not necessarily reflect the views of PPIAF and/or World Bank Management. II. Infrastructure: The situation before privatization By 2000, the state of infrastructure in Tanzania was bad, despite decades of Government and donor efforts to improve the situation through literally many hundreds of millions of dollars in investment. As shown in the table below, all major infrastructure enterprises made losses for a total of over Tsh 150 billion per year. In addition, operational and technical performance was weak. Connection rates were low compared to other countries in the region, leaving most Tanzanians without any services at all. Enterprise Operational performance Financial performance Container terminal Used 50% of capacity in 2000 THA Loss of 100 m Tsh in 2001 TTCL 173,591 subscribers in 2000 Loss of 17 billion Tsh in 2000 International call completion rate of 45% in 1999 TANESCO 431,700 customers in 2000 Net loss of 81 billion Tsh in 18,200 pending line backlog 2001, Net loss of 177 billion Tshs in 2002 DAWASA 48% demand serviced (in Dar Net loss of 3.4 billion Tsh in es Salaam) 2000/2001 90,000 connections for 2.2 Billing efficiency 70% in 1999 million people in 1999 Collection efficiency 50% in 1999 Losses 50% in 1999 3% of accounts metered TRC 670,000 passengers, Loss of 36.7 billion Tsh in 2002 15 deaths, 37 serious accidents, 10 collisions, 203 derailments and capsizements in 2000 Air Tanzania 197,749 passengers in 2000 Loss of 10.9 billion Tsh in Utilization 40% cargo 1999/2000 47% passenger seats TAZARA 630,000 passengers in 2000 Loss of 12 billion Tsh in 2000 Passenger traffic 71% of historical maximum (1992) Freight 61% of 1992 9 When reviewing the impact of the subsequent privatization program for infrastructure, it is important not to lose sight of the starting point. The most likely counterfactual to privatization would have been a continuation of a spiral of bad services, loss making and subsidies. Whilst it was and is fair to expect an improvement of the situation as a consequence of private involvement, realistic expectations would have indicated a long road towards recovery and sustained improvement. It is doubtful that many in 2000 knew the full extent of the crisis. With hindsight, the privatization program for infrastructure may well have been based on an overly optimistic set of expectations. III. Country Preparedness for Infrastructure Privatization Economic and Business Situation. Tanzania’s business environment has recently been assessed in a World Bank Investment Climate Assessment (ICA). The summary conclusion of the ICA is that the investment climate in Tanzania is improving but not yet competitive in a globalizing environment. Especially power compares less than favorably with neighboring countries (see box 1). Box 1: Tanzania Investment Climate Assessment 2004: main findings Enterprises surveyed cited tax rates (73%), electricity (59%), cost of finance (58%), tax administration, corruption, access to finance, and macroeconomic instability as leading constraints. It is worthwhile noting that in the World Bank’s 1999 World Business Environment Survey (WBES) respondents identified the same main areas of concern, although most problems were more severe at the time. Perceptions of constraints were not uniform. For instance, small firms were more likely to list access to finance as a major problem whereas large firms rated customs, trade regulations, workers’ skills and education as more serious obstacles. Both foreign and domestically owned enterprises viewed electricity, corruption and tax administration as major obstacles. However, cost of finance was viewed as a more serious problem by domestically owned companies. Finally, exporting companies cited customs regulations and transport as important problems. Regional comparison on the top three constraints (tax rates, electricity, and cost of finance) reveals that tax rates in Tanzania are not higher than taxes in neighboring Uganda and Kenya where corporate tax rates are at 40%. However, the VAT rate is a little higher, at 20%, than in Uganda and Kenya where it is between 14% and 18%. Also, tax evasion and informality are high in Tanzania compared to Uganda and Kenya. Power is a key concern in all three countries, but the problem is worst in Tanzania. The median estimate for losses in production due to power outages and losses was 5% of production, significantly higher than in any of the comparator countries. Enterprises in Tanzania were more likely to own generators (55% of enterprises) than in Uganda (35%). The cost of finance in Tanzania is similar to that of Uganda and Kenya. The median interest rate in Tanzania was 13% whereas in Kenya it was 15% and Uganda 18%, the median loan period was 3 years in all three countries. The government has already taken steps to address these constraints but the survey results indicate that more needs to be done. On taxes, the government should reduce evasion and informality in order to lessen the burden on formal enterprises without losing revenues. The power sector has improved since the government signed a management contract with Net Group Solutions from RSA in 2002 but the survey clearly underscores the limitations of this approach: More remains to be done. Finally, reforms in the financial system should focus on increasing access to finance for micro and small enterprises. 10 All but one of the case studies attached to this report note that prior to privatization the infrastructure assets in Tanzania were severely dilapidated, due to poor public sector management and consequent lack of adequate and efficient investment and maintenance. In addition, even in the rare cases where performance was effective, the infrastructure firms were incapable of expanding their services to meet the needs of the general population. Only a small percentage8 of the more than 30 million Tanzanians enjoyed connections to the power, water, sewerage and telecommunications networks (see table above). For most infrastructure sectors it would be fair to say that the existing firms in Tanzania in 2000 represented an unattractive business proposition. There were severe limits to the level of risk private operators could reasonably expected to take in these sectors, as well as to the overall appetite of private investors for infrastructure transactions in Tanzania. As shown below, this aspect was not sufficiently taken into account when preparing transactions. Going into its privatization program, Tanzania, as a formerly socialist country, had a poor reputation with international investors. The Government was well aware that in order for its pro-private sector policy stance to succeed, it was important to improve Tanzania’s reputation among private operators. In many cases, there is clear evidence of Government officials being guided by this consideration when deciding how to deal with a private operator. It could be that in some of these cases the desire not to offend investors went too far. Perhaps some Government officials feared that a conflict with a private operator might cause a broader problem with the donor agency of the operator’s home country. The latter impact could be particularly dangerous, given Tanzania’s high degree of donor dependency. On the other hand, the deportation of three British expatriate City Water executives in late May 2005, no matter how legally justified, and regardless of the depth of frustration on the part of Government, will certainly give potential investors in Tanzania second thoughts. The best advice on this matter would be not to over-focus on reputation issues but to enforce the Government’s rights under existing contracts consistently and diligently, seeking to avoid spectacular confrontations. Legal and Regulatory Framework for Privatization. The remarkable and steady progress in the privatization program since its inception shows that Tanzania’s national privatization policy is sound. Objectives are clear, enabling legislation and institutions are in place. A positive point for infrastructure privatization is that the policy and the process were extensively tested in the preceding privatization of the productive sectors. Another strong point is the involvement of key Government stakeholders in the decision on the privatization strategy for each enterprise, including the approval by the full Cabinet. Unfortunately, a sound overall privatization policy is not enough to deliver success in infrastructure privatization. A sound sector policy that embeds the privatization strategy for each enterprise, as well as the regulatory capacity to enforce the sector policy is needed in each sector. On this score, Tanzania was much less ready. As the cases show, 8 TTCL 3.1% - TANESCO 7.1% - DAWASA 23.8%; of households. ATC 0.6% - TRC 2.0% - TAZARA 1.9%; of population. 11 some aspects of policy, and, to a greater extent, regulatory capacity were lacking in all sectors. This fact in itself does not necessarily condemn the transactions (as eloquently demonstrated by the success of the container terminal concession), but it does lower the probability of a favorable outcome in each case. Communication with the General Public Privatization of infrastructure is a very sensitive matter. In the best case, the perception can be that Government is simply getting rid of activities where the public sector no longer needs to be involved. At the opposite extreme, in the worst case, privatization can be seen as the Government abdicating its clear responsibility for matters of life and death (water!). In order for any transaction, and indeed the program as a whole, to succeed, it is crucial that there be extensive public debate prior to and during the privatization process. This debate should center on the reason why the decision to privatize was taken in the first place. This will entail a debate about the performance of the parastatal prior to privatization. In addition, the debate should help set expectations on what can reasonably be achieved through the process of privatization. This is especially true in the previously prevailing Tanzanian environment of dilapidated infrastructure assets, and poor or missing financial and operational data in infrastructure parastatals. Although hard to assess after the fact, several representatives of civil society interviewed by the team stated that there was relatively little communication with the general public on infrastructure privatization, possibly because the policy thrust of privatization was not new and quite broadly accepted. One NGO representative summarized this apparent lack of communication as follows: “the majority of people are onlookers at the process of privatization (of infrastructure)�. A particularly weak point seems to have been communication on labor issues. IV. Capacity for Infrastructure Privatization Pre-privatization Unlike privatization in competitive sectors, infrastructure privatization almost always results in contracts that require continuing and sophisticated interaction between the Government and private operators. These contractual relationships can vary tremendously, but no contract allows the Government to let go completely of an infrastructure sector. It is always clear who gets the political blame in case of failure. Regulation, performance and contract monitoring and enforcement are therefore key issues. A logical starting point for building the capacity for these functions is the capacity in various institutions that were in place prior to privatization. In the case of Tanzania, this starting point was weak: Evidence from most of the cases suggests that line ministries were largely ineffective in monitoring parastatal enterprises before privatization. This contributed to the poor quality of operational and financial data in many sectors. Ministries also had no experience dealing with private operators; indeed, private management or ownership of infrastructure firms was a novel concept in Tanzania. At the same time, as shown by the case studies, the infrastructure enterprises themselves were in acute financial and operational distress, with the exception of the 12 Tanzania Harbour Authority. Overall, it is clear that the actors on the ground before privatization were ill equipped to deal with infrastructure privatization. The privatization process The privatization process is a distinct phase that requires highly specialized transaction skills that are rarely available in governments. Even high-income countries hire expensive advisors for key privatization transactions. In the case of Tanzania, a specialized body, PSRC, was set up to manage the task. Based on the evidence available and the results to date, PSRC seems to be effective. Its staff may be overloaded, but they are professional and have the means to hire outside advice where needed. The parts of the transaction process conducted by PSRC have been carried out in a professional and transparent manner. Parallel to this, however, several case studies have revealed that some ministries have engaged in side deals with the selected private partner and/or other players in the sector. Examples include the access rights granted to the Trans-African railway corporation and PANACHE, as well as the Management Assistance Contract for MSI-Celtel in the TTCL transaction. Whilst the railway side deals may well have been economical and sensible, PSRC should have been informed and even involved as this type of deal affects the privatization transaction. As to the Management Assistance Contract for MSI-Celtel, this appears to have happened after the privatization transaction and without any involvement of PSRC. There also appears to be no economic justification for the extra fee awarded to MSI-Celtel under the contract, especially because the fee is in no way linked to performance. This side deal is therefore a matter of grave concern for the integrity of the transaction process. Several of the cases considered show poor outcomes to the negotiations with the selected private operator after the transaction process. Before attributing these outcomes to weak negotiating capacity, it is important to consider that the object of these transactions often is an unattractive set of assets, as noted above. Unattractive assets lead to few bidders showing up; and those that do are not really sure what they are buying. It is therefore logical that once a bidder has his foot in the door, (s)he tries to negotiate down. To repeat, this effect is significantly exacerbated by the poor quality of the pre-privatization operational and financial data. However, even taking these effects into account, it appears that the legal capacity on the Tanzanian negotiating teams has been weaker than that of the negotiating partners. Unfortunately, legal skills are key to achieving good negotiating outcomes; a good legal team will be able to pinpoint weaknesses in the entire negotiating strategy. Post privatization As explained under pre-privatization above, the capacity for performance monitoring, contract monitoring, and enforcement and regulation are key to a successful outcome in infrastructure privatization. Evidence suggests that the more risk that rests with the Government, the more complex the interaction between the Government and the private operator becomes. As a recent study9 of the effects of privatization in African 9 PPI: Contracts to Improve Management and Performance, Report to The World Bank, June 2005, Castalia Advisory Group 13 infrastructure concluded: “The more limited the private sector involvement, the more complex is the interface with the government�. For Tanzania, where the risk appetite of private operators is limited, government administrative and regulatory capacity is particularly crucial. Unfortunately, a quick review of the relevant actors in Tanzania shows significant weakness in this area: PSRC may have some capacity based on its experience with monitoring pre-privatization and with contracts during the negotiation; however, it has no mandate to follow firms post-privatization. The Ministries are still ineffective in monitoring of parastatals; in fact PSRC’s taking over part of the responsibilities during the preparation for privatization may even have further weakened their capacity. In addition, the Ministries have still not acquired experience dealing with private operators as they had only arms-length involvement with the transaction process. The asset holding companies that are created out of the ashes of the original SOEs are young and weak, casting doubt over their ability to monitor and enforce the contracts 10. Finally, regulatory bodies are either absent (power and water) or young and still quite weak (telecom and transport). In short, the post privatization capacity picture is significantly weak. This area is one for priority action. When deciding on the appropriate course of action to take, it would be advisable not to come up with revolutionary reshuffling of roles or even less the creation of new bodies to solve the problems created by lack of capacity in existing bodies. By and large, the division of roles in the Tanzanian policy framework makes sense: line ministries are responsible for monitoring of parastatals, regulatory agencies for economic, environmental and safety regulation, asset holding companies for management of investment and liabilities. The only missing aspect is assessment of the impact of privatization, for which it would make sense to give PSRC a mandate. It would also be good to find a mechanism to avoid losing the experience gained by PSRC during the transaction and the negotiations with the winning bidder. Since PSRC is moving towards its last set of transactions this could be done by moving key staff involved in specific transactions to whichever body that will monitor the contract with the new private operator. In addition, the key lesson of the cases is that capacity building in all institutions involved is a high priority. V. Consultations and Communication “The majority of people are onlookers at the process of privatization� As stated earlier, the sentiments of non-Government actors are captured by the above quote. The perception of the review team is that the Government might consider key elements of infrastructure privatization too sensitive for public debate. Unfortunately, privatization is like plumbing: A problem does not go away if you hide it, it gets worse. A much more open debate on the state of infrastructure in Tanzania, and the scope for 10 This statement does not apply to every case. DAWASA performed well in monitoring and enforcing the City Water Services lease contract. 14 solutions, would be most helpful in building understanding for policy choices, finding creative approaches, and managing expectations. Since workers are the most directly affected group in privatization transactions, and because they can be highly effective in blocking the process, opening a dialogue with them should be priority number one. Around the world, the most frequent criticism from workers in privatized enterprises is that they were not consulted, or consulted too late. The lesson is therefore simple: The earlier the better! Regrettably, this lesson was not fully applied in Tanzania. In particular, one finds striking differences in the perception of workers in different firms concerning the process of retrenchment: In TTCL a proposed retrenchment, agreed to by Government and labor leaders, has been blocked for over a year by an injunction requested by a break-away union group. The dissident workers complain of never having been consulted. In TANESCO, on the other hand, the new management made a special effort to discuss with workers a voluntary departure process, which was carried out fully and peacefully. Labor leaders in TANESCO praise management for its frank and clear dialogue. One can even say that the degree of openness around the retrenchment process is a more important determinant of success than the level of benefits paid. Nevertheless, it is crucial that financing for retrenchment payments is assured up-front, and that promised payments be made fully, and in a timely manner. Where possible, this should be financed by the new private operator, as he has a stronger incentive to keep the costs reasonable and to select himself which workers stay on, which could improve productivity. On the other hand, having the new operator deal with this issue has the disadvantage of an increased prospect of labor disputes in the firm. Therefore in those cases where political sensitivity is high and/or it is economically not feasible for the private operator to pay (e.g. the railways), the Government needs to assure funding up-front. This solution will be less politically contentious, but usually more expensive, as governments tend to give in more readily to union demands. VI. Consultants and (Development) Partners As noted, the more complicated infrastructure deals require highly specialized preparation and transaction skills. Most transactions therefore make extensive use of technical, financial and legal advisors. Tanzania has amassed considerable experience in their use, and has learnt some valuable lessons for the use of such advisors: Selection and monitoring: Buyer beware! When bidding for transaction advisors contracts, investment banks and consulting firms tend to present their “A Team�. When the bait has been taken, they often try to send the “C Team.� It is therefore important to be precise when specifying the inputs required, in the bid but especially in the contract, and to insist on the quality that was promised. When called to keep their commitments, respectable firms usually respond. Note, 15 however, that if one specifies that certain high-placed individuals carry out the bulk of work, the fees will rise accordingly. The method of selecting the advisors is also important. Funding agencies like the World Bank usually apply procurement procedures such as quality and cost based selection (QCBS). QCBS is a good way of taking both price and quality into consideration …if you are buying a television set. But if you have cancer and are going into surgery, you want the best doctor you can find. Privatizing an infrastructure enterprise involves stakes that may well be too high to skimp on the cost of transaction advice, usually a small percentage of the value of the transaction. A method of selection based only on quality could well be more appropriate. Success fee for transaction advisers? It is standard practice is to hire transaction advisors on a success fee basis, i.e. to leave an important part of their payment conditional on completion and/or the value of the deal. In the case of Tanzania, we have already observed that the circumstances pre- privatization lead to weak investor appetite and thus weak competition. This has been strongly confirmed by the evidence in the case studies. Weak competition and the pressure to complete the deal—caused in part by the success fee structure—can be a dangerous cocktail. It is quite possible that in Tanzania’s circumstances, it would be in the national interest to pay the transaction advisors on a time spent basis so as to allow maximum emphasis on doing the right deal. Make sure you get the right partner The key lesson from the case studies is that the crux of the matter is to get the right partner. The lessons on transaction advisors trying to send the “C team� are equally applicable to prospective private operators11. In addition, it is very important to do thorough checks on the solidity of any financial commitments in the bid, and to have a full understanding of the structure of consortia and the corporate strategy of the different partners in each consortium. Such checks would have revealed quickly that MSI was a 100 percent mobile company; that Biwater’s team had significant weaknesses; and that South African Airways was not in the greatest financial shape, to name a few examples. Careful due diligence on bidders is always important, but when left with a single bidder, it is especially crucial. Get the best lawyers As stated above, the single most important skill for achieving a favorable negotiation outcome is legal advice. A good lawyer will not let a deal go through if (s)he does not understand every aspect of it, and therefore even non-legal aspects are covered by the questioning mind of a good lawyer. Private operators never skimp on legal advice. The most telling example was from the Air Tanzania case, where someone said “they brought 11 For example, the first management team sent to CWS by the British firm Biwater, did not seem to have the right experience and profile for the size and complexity of the utility in Dar es Salaam and had hardly worked for Biwater previously. The augmentation team from Biwater/Cascal that was sent after a year of engagement, had very little African experience. In fact neither of the two CWS CEOs over the two year period had any significant utility operations or African experience. 16 twenty; we were two� referring to the legal negotiations. The recommendation is therefore to hire the best lawyers and listen to their advice. Conditionality: Curse or necessity? A key feature of privatization for infrastructure is the fact that development partners often attach explicit or implicit conditionality to completion of a transaction. The classic example of explicit conditionality is to tie disbursement of a tranche of budget support to completion of a transaction. This has long been a favored instrument of the World Bank and the IMF. The softer implicit approach is to condition financing of investments in the relevant sector to the introduction of a private operator12. Both lead to strong pressure to complete a deal, and there is ample evidence that this has led key actors in more than a few transactions in Tanzania to believe that they had no choice but to proceed even if they had doubts about the deal. Unfortunately, this went straight against the acquired Swahili wisdom expressed in the proverb “haraka, haraka, haina baraka� (haste, haste, brings no blessing). There would therefore appear to be strong arguments for not using this type of conditionality at all. Before coming to this conclusion, one needs to consider the reason why this type of conditionality was invented in the first place: to help maintain momentum in politically sensitive processes by strengthening the hand of reformers. The need to maintain momentum is also acquired Swahili wisdom expressed in the proverb “haba na haba hujaza kibaba� (drop by drop fills the bucket). Based on the clearly perverse effect of conditionality in the low competition environment of Tanzania infrastructure, it would be best to work based on the knowledge that drop by drop does indeed fill the bucket, and work on strengthening sector strategies and their ownership through extensive public debate as the key instruments to assure momentum. Development partners offering to strengthen the hand of reformers through conditionality could then be politely declined. This does leave the dilemma of what to do in the interim to avoid the assets eroding further due to poor public management and no donor investment. This dilemma is treated below. VII. The Procurement Process Limited competition: Do standard approaches work? As noted, the number of investors willing to submit formal bids for privatization transactions in infrastructure is limited in Africa, and circumstances in Tanzania result in even more limited competition than elsewhere. At the same time, standard procurement procedures such as those of the World Bank, which are used in most privatization transactions in Tanzania, are designed based on the assumption of strong competition. Most notably, the procedures explicitly disallow parallel negotiations with bidders. This is artificial: In real life, anyone having to make a choice between two available contractors would get a bid from both and then hold a parallel negotiation using the two bids. When using standard procurement procedures, the Government is often forced to 12 Donor organizations use a variety of subtle expressions for this. Another favorite is “more efficient management practices�. 17 sign a contract knowing that its signature is only the first step in the process of renegotiation. The big difference is that renegotiation happens only with one partner. It would appear that it is time to review the existing process and better adapt negotiation procedures for the low competition environment of infrastructure privatization in low- income countries. A useful consideration when reviewing the existing procurement process is the degree to which transaction advisers contribute to limiting competition. Experience as well as logic indicates that only prospective private operators can really assess whether the risk in a proposed transaction is acceptable to them. As shown in most cases reviewed for this study, more bidders usually drop out at the last stage than submit bids. This is remarkable as in most cases last stage drop-outs have spent considerable amounts of money and effort on the selection procedure. It may well be that rather than asking transaction advisers to design the best deal for the Government; it would be in the Government’s interest to leave conditions a bit more attractive to bidders so as to ensure greater competition. Procurement procedures are designed primarily to promote economy and efficiency; however, increasingly they are also seen as an important tool to combat corruption. When thinking about redesigning procedures, it is essential not to lose sight of this aspect, as the key problem with the parallels in the paragraph above, citing examples of hiring contractors in real life, is that most people hire contractors with their own money. Public officials use tax payers’ money. The trade-off for making procedures more flexible, and especially for enhancing the scope for negotiation, would have to be much stronger stakeholder participation in and oversight of the procurement process. One of the key problems in all cases is the poor quality of pre-privatization data. This leads to increased uncertainty on all sides and creates clear opportunities for the type of behavior shown by MSI-Detecon in the TTCL deal: get a foot in the door and renegotiate immediately. The Government could learn from this lesson and the solution proposed by MSI-Detecon for future transactions: all bidders could be offered the possibility to propose a fixed payment up-front with a conditional payment based on first year audited results (or some other mutually acceptable formula). In the evaluation, a predetermined discount factor would be applied to the second payment, giving all bidders equal treatment while limiting their risk. Speed is important, but beware of wishful policy making A general lesson from all over the world is that as soon as privatization is announced the assets of the company in question start to erode. Management knows its time will soon be up and workers are anxious about upcoming retrenchment, creating an environment of poor motivation, and asset stripping. Development partners stop investing; they wait for more efficient management, as their money invested under public management often has delivered bad results. It is therefore crucial for the procurement process to be as fast as it responsibly can be and especially to avoid false starts. Unfortunately evidence, including in Tanzania, shows that the process is inevitably slow and false starts are the rule rather than the exception. It would therefore make more sense to adopt a policy based on the facts and accept that the privatization process will necessarily take a good deal of time. One should accept 18 that the public management has strong incentives for asset stripping as well as a generally poor record, and accept the need for continued investments to avoid further erosion of already dilapidated assets. One solution to consider would be to engage in an interim private management contract with an interim investment plan, financed by donors for the duration of the preparation for privatization. If the management contractor would have an equity option, this could actually work quite well in terms of incentives for future privatization. While the TANESCO case shows that this approach can produce positive results, it also underscores that this solution is only for the interim, as TANESCO is still insolvent today, and in desperate need of investment capital for network rehabilitation and expansion. VIII. The Big Picture: Sector Policies When reviewing privatization transactions in infrastructure as a basis for policy formulation, there is considerable danger of losing sight of the bigger picture in the sectors concerned. Selling, reforming or restructuring an enterprise are not objectives in their own right, they should be part of a coherent sector policy which aims at achieving the Government’s development objectives for the sector concerned. In most sectors, the main challenge for the future is how to provide services to the larger part of the population that is not connected to the network. Achieving more efficient management of the existing network will still remain an important part of the policy, as this will bring down the cost of providing services in the network, thus freeing up resources for expansion. In addition, it should be noted that the existing network generally does serve the productive enterprises, the competitiveness of which depends to a large extent on reliable infrastructure services, especially power. The key is to recognize that a sound and coherent sector policy for infrastructure is a cornerstone of the development strategy for any country, both in terms of poverty alleviation and competitiveness. This is especially important for a country like Tanzania, which still has a large infrastructure deficit, and needs to position itself in an East Africa region that is rapidly integrating. The review revealed that whilst sector policies exist on paper for most sectors, implementation is often timid, possibly due to lack of broad ownership of the policy. Energy Sector Policy: No broad ownership Although the Management Contract for TANESCO is a success, it is remarkable how little of the Government’s energy sector policy has been implemented. The policy was formulated as early as 1997, yet not even the new electricity law has been adopted. To a significant extent, this lack of implementation can be attributed to the diversions caused by a series of crises, as noted in the case study. Nevertheless, many interlocutors have referred to the sector policy as too complex or even moribund. Indeed it appears that a sophisticated unbundling scheme for a very small (500mW) power system may not be realistic. The key problem in the sector from a long term development perspective is surely not the structure of the existing power utility but the fact that 500mW is much too little to serve Tanzania’s growth needs in the medium to long term. 19 It would make sense to accept that the current energy sector policy will not be implemented as intended. The first consequence of that would have to be an extension of the current management contract for TANESCO to buy the time necessary to formulate a new policy and get ready for its implementation. This would allow a broad consultative process to formulate the new policy. It is recommended that this policy should be more focused on a long term vision for the growth of the sector, including rural electrification and connection to regional grids. This vision should be underpinned by a realistic investment plan based on the financing expected to be available from various sources (Government, development partners and the private sector). It would be reasonable for the new policy to include a much less complex plan for the restructuring and possible divestiture of TANESCO. Transport Master Plan: Positioning for regional integration The discussion on the big picture during the meetings between the PRC and the World Bank team was triggered by discussions on multimodal transport corridors when reviewing privatization transactions for enterprises that are part of such corridors, like the container terminal and TRC. The question discussed was whether it makes sense to proceed with the reform of one enterprise in the corridor when the others are not yet ready for reform. The large economic and fiscal contributions made by the container terminal since it was given in concession answer this question conclusively: It does make sense to proceed. That said, benefits would be much larger if all enterprises in the corridor were to achieve a similar level of efficiency to that of the container terminal. The impact of such an achievement on the competitive position of Tanzania in East Africa would be tremendous. In order to capture the enormous potential benefits of well functioning multimodal transport corridors, Tanzania should develop a Transport Master Plan. Such a plan should be based on a regional vision, both in terms of competition in the region as in terms of reinforcing synergies between countries. As all sector policies, the Master Plan should be underpinned by a realistic financing plan with clearly identified sources of financing. Telecommunications: Disappointing outcomes Like most countries in Africa, Tanzania has seen a boom in mobile communications that has generated tremendous economic benefits. The combined power of new technology and competition has delivered results beyond what anybody could have expected as recently as ten years ago. Nevertheless, upon closer examination, telecommunication service delivery in Tanzania does not compare favorably to other countries in the region. As shown in the case study, internet penetration is low, international call rates are high, and local service, even in the mobile network, is of lower quality and at higher rates than in neighboring countries. Especially disappointing in terms of Tanzania’s future competitiveness is the inadequacy of the IT backbone structure. The case study on TTCL reveals that the policies underpinning the privatization transaction may have been overtaken by the technology revolution in the sector. The performance targets focused heavily on voice teledensity and much less on data communication and IT backbone, the bread and butter of fixed line companies in the 21st 20 century. The high prices for many of the mobile services may well be the consequence of regulatory issues that stifle competition. In short, it would make sense for Tanzania to take a fresh look at its telecom policy so as to maximize the benefits of strong competition whilst at the same time making sure that there will be sufficient investment in a solid IT backbone. Water: Learn from the crisis; move on to the unconnected The Government has already taken a number of steps in response to the cancellation of the lease with City Water Services (CWS). It has acted to create DAWASCO, a state- owned entity that will perform the same role as CWS in what is intended to be a similar contractual relationship, i.e. a lease. It has also appointed a proven effective manager as CEO for DAWASCO. In addition, it has held intensive consultations with its development partners to ensure continuity in donor financing for the sector, in particular for the rehabilitation works. These steps have worked well and have avoided an acute crisis. The next priority is to apply the lessons of the failed lease contract with CWS in the immediate future. The first priority in this regard will be to corporatize DAWASCO so that it operates as an enterprise subject to commercial legislation. This is key for the success of any contractual relationship between DAWASCO and DAWASA. The second priority is to strengthen DAWASCO’s management immediately so as to get competent, results oriented performers in at least the three positions vacated by the expatriate managers. The appointment of the CEO deals with only one of the three positions. Subsequent to that, we would recommend that the contractual relationship between DAWASCO and DAWASA be revisited. Since a lease has proven too ambitious for a private operator, it would seem plausible that a public operator might face similar problems. It might therefore be best to consider the option of a management contract for DAWASCO. Finally, Tanzania might be well advised to seize this moment to initiate reflection and a debate on a national water sector policy that goes far beyond city water to the peri-urban and rural areas. In particular, lessons could be drawn from successful experiments in neighboring countries like Uganda, in providing water in provincial towns and peri-urban areas using local private operators. Air transport: Liberalization works, but ATCL in severe crisis The good news on air transport is that the liberalization of the sector has dramatically improved services to Tanzanian customers. Unfortunately, Air Tanzania (ATCL) has not been able to compete successfully in this new environment: With losses presently averaging US $600,000 a month, and increasingly acrimonious relations between the shareholders, ATCL is in crisis. Government has already injected an emergency US $2 million into the company, and it is a certainty that more public money will be requested soon, if that request has not already been made. We repeat that from an economic point of view, there is no justification for GOT to subsidize ATCL. The liberalization of air traffic across the region will ensure that adequate service will be provided to Tanzanian customers. The team recognizes the political and psychological issues involved, and the 21 unlikelihood that dramatic action will be taken in the period just before a general election. Nonetheless, the facts must be faced: If, within a very short time—six months?— ATCL does not receive an external reprieve, either from a change of heart and mind in SAA, or from some other external investor (Gulf Air?), then it should be shut down. No more public money should be spent to keep this non-strategic firm afloat. IX. Transaction Scorecard: A Mixed Picture As stated in the introduction, there is a perception in Tanzania that infrastructure privatization is in serious trouble. Having reviewed the cases, we are now in a position to present a scorecard of transactions as follows: Company Success ? Failure Not yet DAWASA X TANESCO X TTCL X THA X TICTS X ATC X TRC X TAZARA X For most companies, the score is relatively obvious: DAWASA is clearly a failure, the Container Terminal (TICTS) and the TANESCO management contract clearly successes. Also, it is clear that the two railway companies as well as the Tanzania Harbour Authority cannot yet be rated as the transactions for these companies are not yet completed. For the two companies in the question mark category, we would offer the following observations: The privatization of TTCL has been troublesome in many ways. At the same time, TTCL and Celtel have brought tremendous benefits to the Tanzanian economy and the sector overall has developed far beyond what was hoped for when the transaction was initiated. Given that nobody expected technology in the sector to develop as rapidly as it did, the fact that the privatization did not yield exactly the result that was aimed for is not necessarily a disqualifier. Hence, we come out in between success and failure. Air Tanzania was essentially dying when it was privatized. It was also a black hole for Government subsidies to keep it flying. The privatization transaction itself has to be rated a success, as it was clearly unexpected for the Government to get $20m for this company—even if $10m went back into the company—, allowing it to continue flying for three years at no expense to the Government. The most important achievement of the 22 Air Tanzania privatization was the fact that the company no longer represented a drain on the Government budget. Unfortunately, the Government has recently resumed bailing out Air Tanzania. Whilst the political motivation behind this decision is clear, from an economic perspective it puts the success of the Air Tanzania privatization transaction in jeopardy, as the Government could well find itself exactly where it started. The jury is still out, hence the question mark. 23 Case Study 1 The Lease Contract between DAWASA and City Water Services I. Introduction: The Situation Before Privatization The Dar es Salaam Water and Sewerage Authority (DAWASA) was created in 1997, as part of an effort to set up “independent and autonomous� water and sewerage agencies in all major Tanzanian cities. DAWASA was a quasi-commercial parastatal corporation, owned and operated by the Government of Tanzania. Prior to DAWASA’s creation, water and sewerage for Dar es Salaam had been supplied by a National Urban Water Authority and the Sewerage and Sanitation Department of the Dar es Salaam City Commission. Both before and after the creation of DAWASA there were severe problems in the production, transmission and distribution of water to inhabitants of Dar es Salaam. Over the years the water infrastructure had greatly deteriorated, because of low prices, poor management of billing and collections, and consequent lack of revenues for investment and maintenance. The price of water was held at relatively low levels. By 2002, DAWASA’s average tariff for domestic consumers was US $ 0.40 per m3; less than in many other African cities. Water delivered was often not safe for drinking; the city suffered periodic outbreaks of cholera and other water-borne diseases. By 2002, “unaccounted for water� (UFW) had reached an alarmingly high 55 percent. That is, because of leaks, non-metered connections, illegal connections, and some large-scale, organized theft, DAWASA management neither knew what happened to, nor collected revenues for more than half of the water entering the system at the reservoirs. An estimated 85 percent of the city’s population had some sort of access to water pipes; but this did not mean they regularly received water. In many areas, water was available only six hours a day and often much less. In some sections of the city customers complained that water was available for a few hours once or twice a week, or even more rarely. In some newly developed areas pipes were installed and bills were sent—but no water ever flowed out of the pipes. A 2001 household expenditure survey indicated that 45 percent of the city’s households bought water from neighbors, tanker trucks or vendors—with at least the last two charging unit prices that were a huge multiple of the DAWASA tariff. Since it was the poorer sections of the city that had the worst service. Inhabitants of these sections were the most reliant on non-network water; the least well- off were paying the highest prices for water. The financial performance of DAWASA, and its predecessors, was extremely weak, and had been so for many years. 24 Table 1: DAWASA Financial Performance Since, 1985 (TShs) Operating Government Total Income Total Surplus Income Subsidy ‘000,000’ Expenditure (Deficit) ‘000,000’ ‘000,000’ ‘000,000’ ‘000,000’ 1985 61 0 61 135 (74) 1986 98 0 98 164 (66) 1987 143 0 143 296 (153) 1988 199 0 199 409 (210) 1989 933 0 933 742 (191) 1990 765 0 765 5,464 (4,669) 1991 937 0 937 2,810 (1,873) 1992 1,395 0 1,395 3,712 (2,326) 1993 2,673 0 2,673 5,402 (2,729) 1994 2,784 5,223 8,007 7,980 27 1995 3,965 0 3,965 6,703 (2,738) 1996 2,660 1,502 4,162 6,248 (2,086) 1997 8,409 2,176 10,585 10,204 381 1998 10,432 1,246 11,678 11,159 519 1999 13,326 0 13,326 13,066 260 2000 13,963 0 13,963 13,835 128 2001 15,130 3,307 18,437 19,278 (838) 2002 15,323 0 15,323 19,376 (4,052) Source: Mugisha, Brown & Kiwanuka, “Water Reforms in Three East African Capital Cities,� World Bank Working Paper, March 2005, pp. 2-3. Table 1 shows that the Dar es Salaam water services ran a deficit in 13 of the 18 years prior to the signing of the lease contract in 2003. From 1990 onwards, the deficits grew greatly. In the eight deficit years from 1990 to 2002, outflows exceeded revenues by an average of 2.63 billion Tshs per year. Surpluses occurred in only five years (in three of which the Government provided substantial subsidies); these surpluses averaged less than a tenth of the average deficits. Even with generous subsidies from 1994 onward, DAWASA and its predecessors were unable to cover their operating costs, much less service their debts or manage depreciation charges. Repair of the network to reduce the shortages and improve water quality, and, most of all, expansion of the network to meet unmet demand, and increase access to piped water, was beyond their capacity. By 2002, DAWASA had about 100,000 customers for water and 22,000 for sewerage. Almost none of the water distributed was metered; bills to customers were estimates— often over-estimates—of consumption. The World Bank estimated that only about half of DAWASA customers actually paid their water bill regularly. In particular, many Government offices did not pay their water bills. In the 1990s various donors—from China, Britain, Germany, the European Community and the World Bank—had provided some resources for water network emergency repair and maintenance. However, the sums donated or lent were only a fraction of the investment required for DAWASA to offer reasonable standards of service to its 25 clientele. The needed investments were estimated as ranging from $150 million US (in 1991) to as much as $600 million US (in 2002). In light of the poor management record in Tanzanian water services, donors were unwilling to commit to massive investments. II. The Transaction The Strategy The situation became critical after 1997, as first floods and then drought further eroded the water system’s infrastructure and service. The Government of Tanzania reacted by creating DAWASA, passing (1999) a Water Law allowing for the privatization of DAWASA’s operational activities—but without specifying precisely the manner in which this would be done—and creating (2001) a multi-sectoral Energy and Water Regulatory Authority (EUWRA). With this strategy in place, and with donor assistance and prodding (from the World Bank in particular), in 1997 the Government began to look for a private operator to take over major responsibility for water production, transmission, distribution, billing and collection. The Process: Finding a private partner The search for a private partner was long and tortuous. It began in mid-1997 and took a full 6 years to conclude, going through two phases, with two rounds of bidding in the second phase. The process was at first handled directly by DAWASA and then, before the first phase bids, transferred to the Presidential Parastatal Sector Reform Commission (PSRC). The first phase (pre-2000) produced five expressions of interest and four actual bids. But since the preferred management option had not been made clear in the request for bids, the four submissions varied greatly: One proposed a concession, the second a lease, the third a joint venture, the fourth a management contract. The PSRC decided it could not decide between such different bids, and opted for a second phase of bidding. Prior to the second phase, the World Bank funded a consultant to suggest the optimal management system. Following discussion of this report, it was decided that a long-term concession, transferring most commercial risk and investment responsibility to the private partner, was unlikely to attract investor interest. A shorter-term, somewhat less risky lease contract was listed at the preferred option. This would transfer operational responsibility and some commercial risk to a private operator, but most of the needed investment capital would come from donor funds. This information was conveyed to all the bidders qualified in the first phase. Two then dropped out of the competition. (Reportedly, there were allegations from a withdrawing bidder that the transaction had been “rigged� in favor of the French firms.) The remaining two, Saur Internationale and Groupe Generale des Eaux, both from France, submitted formal bids (early 2000). The decision should have been simple: The bidder proposing to deliver water at the lowest average price, while meeting financial, quality and expansion requirements, should win the competition. But both bids contained additional qualifications and 26 requirements, all aimed at reducing the bidders’ commercial risk and exposure, and requesting guarantees and escape clauses in case of contingencies (such as non-payment of bills by Government agencies). Given the poor state of the assets, the commercial risks involved, and weaknesses in the financial and operational data within DAWASA, the World Bank staff involved in the process concluded that many of the points raised by the bidders were legitimate. They advised PSRC to negotiate the issues raised with both bidders and try to work out a mutually acceptable contract. However, the PSRC declined to evaluate or reopen discussion on the bids, ruling them non-compliant. The bidding documents were then revised. All bidders henceforth had to agree to: Contribute a set amount of equity in the management firm they would create; accept a specified amount of financial support to help them start up their companies; and accept a structure of fees they could earn by managing the donor-funded works program. The bid documents also specified the financial and operational performance standards the private provider would have to meet, including their payments and fees to DAWASA, which would become the Asset Holding Authority and manager of the 10 year lease contract (see Annex for a detailed listing of the operational requirements, and financial penalties for non-compliance). Some of these provisions, such as managing a part of the donor- funded works program, “sweetened� the deal for bidders, as this task should readily be profitable. It is by no means unusual for such incentives to be added when a first bidding round has not produced the desired results. The Final bid The final bid competition was reduced to a single item: the “operator tariff� 13 the private provider would charge. Since the aim of Government was to provide the best service to the public at the lowest price, then the lower the proposed “operator tariff,� the better the bid. To guard against a problem seen elsewhere—i.e., a bidder getting a foot in the door by committing to a very low tariff, and then calling quickly for renegotiation of the contract— the bid documents set a minimum operator tariff of 322 Tshs per m3. Three potential bidders responded to the new arrangements, the two French companies that had submitted bids in the previous phase, and a consortium led by the British firm, Biwater. To increase the likelihood of a successful conclusion, PSRC held two pre-bid meetings to answer questions and concerns of bidders. All three firms sent strong teams to these 13 The proposed tariff structure had three parts. The first part was the “operator tariff;� this would be the amount retained by the private operational partner. (Since this would be the main source of the operator’s revenue, it presumably had a strong incentive to reduce losses, improve billing and collection, and enhance efficiency.) Second, was the “DAWASA tariff;� this was the amount of each bill collected that the operator would pass on to DAWASA to cover its costs. Third was “a social connection tariff.� This would be a portion of collections deposited by the private operator into a “First Time Domestic Water Supply Connection Fund.� This fund would help pay for expansion of water service to unserved, low-income areas. In a further effort to ease the cost burden, Government specified that the first 5 m3 of water consumed each month by domestic customers be billed only the “operator tariff.� The lease contract specified how, when and why the tariff structure could be revised, calling for an annual review “to adjust the cost index formula or assumptions made when establishing the bid price;� an interim review, and a major review to “set the tariff that will apply for the second five year period of the Lease Contract.� (World Bank, Project Appraisal Document No. 25249-TA, 2003, 7) 27 meetings, and PSRC officials were hopeful that all three, and at least two, would submit final bids. In the event, only one formal bid was submitted (in the first half of 2002), from City Water Services, or CWS, a consortium composed of Biwater, a German engineering firm, Gauff Tanzania Ltd., and a Tanzanian investor, Superdoll. Economic theory and practical experience state that the more bidders the better, at least from the seller’s point of view. Multiple bids are certainly desirable, but there is nothing intrinsically wrong with a single bidder—if the bidding process is open, transparent and competitive, and especially if the party actually making a bid believes others are bidding as well. Moreover, as other Tanzanian privatization experience (e.g., the case of TTCL) shows, having multiple final bidders does not automatically lead to an optimal outcome. Still, there were reasons to be concerned about the CWS14 bid. First, some technicians on the World Bank team expressed concern over the quality of Biwater’s experience in operating water leases, in South Africa, Mexico and elsewhere, including in the U.K. Second, CWS had bid at the lowest allowed “operator tariff� of 322 Tshs per m3. Despite the safeguards that had been taken to assure that this minimum was not set too low, this was somewhat surprising. Third, to the pleased surprise of the PSRC, CWS offered to retain all DAWASA operational employees, even though this was not obligatory. (Only later was it asked why would CWS so readily give away one of its major cost-cutting opportunities?) Fourth, the management team that CWS proposed to place in Dar es Salaam was relatively inexperienced. Taken cumulatively, these factors led the donor team to call for a review of the operation by water sector experts in the World Bank. This was held in Washington in 2002. The review concluded that there were no obvious problems that could justify an objection from the World Bank. Later, in July 2003, a “quality at entry� assessment, undertaken by a panel convened by the World Bank’s Quality Assurance Group, looked at the design of the project. This panel praised the operation’s design and assigned the project a “highly satisfactory� score—the highest possible ranking. In light of CWS’s subsequent difficulties, the contract negotiators, and the donor staff and reviewers who assisted them, can be criticized for their inadequate due diligence concerning Biwater’s capabilities to manage in the Tanzanian setting. We shall return to this issue in section IV. For the moment, the purpose of recounting these events is to show that the Government in general and the PSRC in particular were not given any strong signals from their donor partners concerning the quality of the private operator selected. True, the World Bank project document rated the overall risks of the operation as “substantial;�15 but attributed this to likely problems in asset management and regulation in the involved Tanzanian Government agencies, DAWASA and EUWRA. Poor performance on the part of the private operator was considered, but the risk was assessed as “modest.� 14 Biwater/Gauff held 51 percent of the equity in the CWS operating company (note again this was a lease, not a divestiture: the private operators did not own the assets of the water company; these remained in the hands of Government); Superdoll held 49 percent. The principal negotiator of the contract, and the firm with operational management experience, was Biwater. 15 “Substantial� is the Bank’s second highest risk rating, after “High� and before “Modest� or “Negligible.� 28 The bid was accepted; final details were negotiated, and CWS began operations under the 10 year lease contract in August of 2003. The decisive factors leading to the acceptance of the bid were the pressing need to bring to a conclusion six long years of preparation, during which vital restructuring and rehabilitation of the city’s water services had been delayed, and to tap the US $125 + million in donor loans (from the AfDB, EIB and World Bank) and CWS equity associated with the lease contract. The primary assumption on the part of almost all involved, certainly from the donor side, was that it would be very hard if not impossible for the private operator to perform worse than DAWASA. But that is what happened. III. The Results There are four sources of information on CWS’s post-contract performance: 1. A partial set of operational and financial reports submitted by CWS to DAWASA. We have seen an annual report covering the period 08.03 to 06.04; and then monthly reports for 12.04, 01.05, and finally 03.05. 2. A report of the Government of Tanzania, compiled in 06.05, just after the cancellation of the contract, detailing the problem and describing the measures taken by DAWASA and Government to resolve the crisis. 3. A paper (cited above at the end of Table 1) on water sector reforms in East African cities, written at the end of 2004. (One of the author’s is the World Bank Task Team Leader of the Dar es Salaam Water Supply and Sanitation Project.) A report of a mediation team employed, in April and May of 2005, to determine CWS’s and DAWASA’s positions 4. and complaints with regard to the lease, and propose a constructive solution. There are discrepancies in the information given in these reports, particularly between nos. 1 and 2. For example, the GOT report states categorically that CWS failed to submit required reports to DAWASA, including “quarterly reports, financial statements, collection and cost summaries, annual reports, audit reports, budget, etc.� However, the existence of the CWS reports listed above show that at least some of this information was submitted. There are also discrepancies regarding the achievement of operational and quality targets. We present the GOT/DAWASA case first, then contrast it to the CWS information; and then add the relevant data from the other sources. Government/DAWASA Complaints The GOT/DAWASA argument is that in the entire time the lease contract was in force, from August 1, 2003, to May 25, 2005, CWS never fully met its contractual obligations. Government alleges that CWS: 1. Failed to inject the required amount of equity or meet its borrowing requirement. CWS had committed to putting in $5.5 million US in equity 29 and borrowing $4 million in year one of the lease; by the end of year one, it had put in $3.9 million in equity and borrowed $3.1 million. By the time the lease was terminated, CWS had invested a total of $4 million in equity and had borrowed $4.8 million. 2. Failed to pay lease and other fees due to DAWASA (“which relies entirely on lease fees from CWS� for its income). Government calculated that by termination, CWS owed DAWASA Tshs 4.3 billion in lease and equipment fees and other services. 3. Failed to deposit the “social connection tariff� into the “1st Time New Domestic Water Supply Connection Fund.� Government claimed that CWS was using these monies to pay operating expenses. The total amount that should have been deposited to the fund was Tshs 900 million. 4. Failed to submit required financial and operational reports (see above). 5. Failed to meet contractually stipulated performance targets. For example, in lease year one CWS was supposed to install 16,500 new meters, add 1,000 new water connections, and reduce greatly repair time on burst pipes. According to the GOT, in year one CWS actually installed 8,751 new meters (47 percent shortfall), added 400 new connections (60 percent shortfall), and did not reduce repair times. Most damaging of all, there was 6. “a decline in the availability of water in many parts of Dar es Salaam.� Despite a 15 percent increase in water production (mostly coming from the associated works repair and investment program funded by donor loans, a portion of this project that seems to be working well), Government claimed that parts of the city that previously had water service twice a week were now reduced to getting water once a month.16 CWS Reports The existence of the CWS reports calls into question the blanket assertion of the GOT that CWS failed to submit required any information. However, the data submitted by CWS that we were able to see were partial and do not address (and were not intended to address) many of the issues raised by GOT in its June 2005 statement. What the CWS information does confirm is that the company was in very serious trouble. First and foremost, CWS reports show that collections were consistently less than the targeted amount. For the whole of the period 08.03 through 03.05, CWS managed to collect, on monthly average, Tshs 975 million, against a monthly target of Tshs 1.3 billion—a 25 percent shortfall, or about $295,000 a month. Operational costs were also much higher than anticipated. CWS’s Annual Report for 2003-04, submitted at the end of June 2004, reported a net loss of the operating company of Tshs 6.9 billion. There was a further accumulated loss of Tshs 6.6 billion in the period 07.04 through 03.05, for a total 20 month loss of about US $12.3 million. The reports do not give details on CWS’s injections of equity, its rate of meter installation, and its payments, or not, to DAWASA 16 In one meeting a Tanzanian said that after CWS took over the service he almost never received water at his home. On one occasion the water began to flow suddenly out of a hose lying in his yard. His young son was so unused to this that “ he thought it was a snake.� 30 and the Supply Connection Fund. The reports do claim that CWS is meeting or exceeding performance targets regarding water quality and repair times. A Semi-independent View Mugisha, Brown and Kiwanuka looked at CWS performance under the lease from August 2003 to November 2004. They stated that CWS had not, in this period, fulfilled the important obligation to establish an accurate and timely data base on all aspects of operations, from production to collection (“not much progress has been registered in accomplishing this task.� 2005, 11) This complicated performance assessment. Nonetheless, on the basis of limited information, they concluded that under CWS: Water quantity was “lower than the average during DAWASA by about 20 percent,� collection efficiency had declined, and the 270 per month new connection rate “does not show improvement from the pre-lease era.� (12) Moreover, in the four years prior to the lease, the authors note that DAWASA annual (not monthly) collections ranged from around Tshs 13 to 17 billion. Assuming that these pre-lease collection data are accurate, this means that DAWASA’s average monthly collection rate ranged from Tshs 1.1 to 1.4 billion—all much higher than that achieved by CWS. CWS not only failed to meet its collection targets; it failed to do as well on collections as DAWASA had done. Table 2: CWS vs. DAWASA Collections Monthly Collections 1600 No payments by government institutions during DAWASA period 1400 1200 1000 DAWASA TShs M CWS 800 3 Month MAvg Linear (DAWASA) 600 Increases in average tariff and 400 15% - Dec 2002 7% - Aug 2003 6% - Aug 2004 200 0 May-03 May-04 Nov-02 Nov-03 Nov-04 Aug-02 Sep-02 Apr-03 Oct-02 Dec-02 Jan-03 Mar-03 Jun-03 Jul-03 Aug-03 Sep-03 Feb-03 Oct-03 Dec-03 Jan-04 Feb-04 Mar-04 Apr-04 Jun-04 Jul-04 Aug-04 Sep-04 Oct-04 Dec-04 Jan-05 Feb-05 Mar-05 31 Information from the Mediators During an intense process to resolve the conflict, in April and May 2005, mediators tried to establish the positions of both sides and construct a compromise. The data they uncovered are the same as reported above. However, the mediator’s report does not indicate that CWS contested the GOT/DAWASA claims that it failed to meet financial and operational targets in the lease. Rather, CWS’s position, as reported by the mediator, is that changed circumstances made it impossible for those targets to be met, and justified alterations in the contract. CWS Position According to the mediator, for CWS, the crux of the matter was that the collection targets in the lease contract were far too ambitious. The assumptions on which they had based their business plan were based on inaccurate or out-of-date information. Non-payment by Government agencies was greater than expected. Drought conditions reduced water quantities. Staff costs were higher than anticipated,17 as were transmission losses. These factors combined and resulted in lower than planned collections; and this changed all aspects of the business plan and decreased the internal rate of return. Up to the moment the lease was terminated, CWS claimed that it was committed to the project. Deepening of the Conflict; Efforts to Resolve DAWASA, as the asset and contract manager, had three mechanisms at its disposal to correct the situation. It could levy financial penalties under the terms of the lease (see Annex); it could “call� the performance bond;18 and it could launch a mediation process. Assisted by the World Bank—which very much wanted the lease to work well— DAWASA vigorously pursued these three avenues. It levied penalties on CWS for non- compliance of lease obligations, first in April 2004 (Tshs 70 million) and again in July 2004 (Tshs 50.3 million). DAWASA says CWS did not pay these penalties. In April 2004 the supervising Minister of Water and Livestock held a meeting to resolve the problems, at which Government pledged to push public agencies to pay their water bills, and CWS committed to improve performance. The situation did not improve. In September 2004 the Minister ordered CWS to pay its fees to DAWASA, and appointed an independent auditor to assess contract performance and judge whether an interim tariff increase was justified. CWS changed its CEO and CFO. Disputes arose between the various shareholders of CWS, with Superdoll, the local Tanzanian investor, refusing to inject more equity unless it received a larger voice in the management of the firm. In November 2004 the appointed auditors gave the opinion that 17 As they had promised, CWS maintained almost all former DAWASA employees. Pre-lease, DAWASA had about 1,500 employees. A few stayed with DAWASA to work on asset and contract management; most went to CWS. At the end of December 2004 CWS had 1358 employees; this declined to 1338 by the end of March 2005. Taking into account the workers who stayed with DAWASA, employment losses were miniscule during the CWS period. 18 The contract required CWS to post a $5 million US performance bond that could be forfeited to DAWASA in case of flagrant non-compliance with the contract. DAWASA initially thought this bond could be called in part, thus inducing the operator to comply with the contract. This proved legally impossible; the bond had to be called entirely or not at all. 32 there were “no grounds for an interim review of tariffs.� In December 2004 and again in January 2005 CWS proposed a revision of the lease terms, asking for, inter alia:  a reduction in the amount of equity required from the investors,  an increase of tariffs,  a reduction of lease and other fees,  a write off of existing obligations to DAWASA, and  a free hand in reduction of staff. At about this moment, DAWASA learned that it could not call the performance bond in part; this had to be an all-or-nothing step. The mediators were called in from London to reconcile the two sides. In May 2005 the mediators proposed a workout based on revision of a number of items in the contract, only two of which seemed truly critical: The collection rate, and the length of the lease. CWS asked for an extension of the lease from 10 to 15 years, to give it more time to improve the system and recover its investments. (It also asked for additional external financing to effect repairs and expand the network.) In short, it asked for a reduction of risks and a relaxation of contract terms. GOT/DAWASA was willing to extend the lease for five additional years, but only if CWS could commit to raising and keeping collections, within five months, to a level of Tshs 1.75 billion a month. CWS said the best it could do would be to collect Tshs 1 billion a month for the rest of 2005, Tshs 1.39 a month for 2006, and Tshs 2.0 billion a month in 2007. They asked that the lease be lengthened in advance of improved collections. Evidently, these positions could not be reconciled. On May 25, 2005, the Government of Tanzania cancelled the lease contract, called the performance bond in its entirety—and deported from Tanzania three expatriate managers of CWS. At last report, Biwater/CWS was consulting its lawyers in the U.K. to consider how to deal with what it termed an “unlawful…..repudiatory breach of contract.� (Financial Times, 6.29.05, p. 7) IV. Lessons Learned & Policy Implications CWS’s performance, and the subsequent cancellation of the lease, are costly, not simply to the development of water services, but to efforts to reform Tanzanian infrastructure as a whole. While GOT, DAWASA and three involved donors have acted rapidly to maintain water service in Dar es Salaam and put an institution in place to manage and rehabilitate the sector, the fact is that management of water is now back in public hands, hands that historically failed to manage the service well. As one Tanzanian professor stated; “I fear it will go back to the same old bad parastatal performance.� Moreover, “the summary deportation of three British expatriate executives from Tanzania� (Financial Times) certainly does not help Tanzania’s image among international investors. Nor does this saga encourage local investors, who see the extent to which Superdoll and Gauff TZ Ltd. got burned. Thus, the financial, administrative and opportunity costs to Tanzania are high, and could increase greatly depending on the nature and outcome of any legal process started by CWS/Biwater. In addition, the 33 opponents of private involvement in infrastructure provision—many NGOs, unreconstructed socialists—are now saying “we told you so,� and trumpeting their past opposition to this contract. And any proposed transaction that resembles even slightly the CWS contract, such as the concession in TRC, where there is a single final bidder, is coming under great suspicion. How did this transaction go so wrong? 1. A first factor is Government’s poor past management of the water sector. DAWASA’s dilapidated assets, insufficient quantity and poor quality of information raised commercial risks, and reduced investor interest. The few investors prepared to bid regarded the proposed lease contract as a flexible, first- step measure that could and should be changed as gaps in the information were filled, and errors corrected. The bidders were not sure of precisely what they would get, and how they would be allowed to operate if they won the competition. Thus, they tried to push the risks off on to the other contracting party. Government, having to answer to a broader constituency than a set of single- minded shareholders, viewed the lease contract as immutable; a deal is a deal. Conclusion: The chances of concluding a credible, enduring lease or concession contract in infrastructure, and particularly in water, improve with the quality of the firm being privatized, and the quality of the information on the assets and liabilities investors will acquire or manage. In Africa, this means—to put it bluntly—the chances of concluding a credible lease contract, particularly in the water sector, are low. 2. Second, and relatedly, the lease approach was not appropriate for the Tanzanian water situation. Given the poor condition of the water sector and its information base, it would have been better to opt for a management contract, in combination with donor financing for rehabilitation and expansion. The information gaps, collection uncertainties, the complexities of the tariff and fees structure, the details of the performance requirements, the inevitable politicization of a commodity essential for life—all proved too much for the private operator, which found that it could not make money under the terms of the lease it had signed. Indeed, shifting away from the lease towards a management contract is what CWS tried to achieve in its renegotiation efforts. CWS asked Government to assume most risks and, in essence, guarantee payment to CWS for its management services. The earlier bids from the French private providers were also closer to management contracts than to true leases (indeed, that fact led to their rejection). Other African experience with management contracts in water, in Uganda and Guinea for example, show that they too have their problems; several have not been renewed. But non-renewal (in the case of Guinea, after 11 years of fairly successful service) is far less disruptive than contract cancellation 22 months after signature. Opting for a management contract at the outset would have been more suitable for DAWASA’s situation, the GOT’s degree of tolerance for ambiguity, and the private sector’s assessment of risk. Note, however, that this conclusion applies only to the water sector. Leases and concessions have been applied successfully in other, less contentious sectors in Tanzania; e.g., the lease in the container terminal in the port. 3. Third, there should have been better due diligence on the capacities and competencies of the single bidder in the final round. The World Bank Task 34 Team Leader for this operation had enough doubts that he called for a review within the institution. Members of the review group interviewed said they could not see a glaring reason to recommend stopping the project. This is far from a ringing endorsement of the operation; one suspects the presence of a common World Bank problem—pressure to lend. It is likely that all of the negotiators and advisers, Tanzanian and foreign, felt on their shoulders the pressures of time, the mounting costs of inaction, the need to bring this long process to a close. Moreover, had their been a recommendation that Biwater’s bid be rejected, what then? Who would have come back for a third phase of bids if not the same small group, perhaps even including Biwater that had bid before? What would have been different in the next round to avoid the outcomes of the previous ones? How would the service be maintained in the interim? The difficulty of answering these questions, and the incorrect assumption (noted above) that the private operator could not do worse than DAWASA, led to an acceptance of the bidder. 4. One good piece of news is that DAWASA and the GOT handled well the asset and contract management tasks. There is great and often justified concern that African asset management and regulatory agencies are too new and inexperienced to handle the contract monitoring and enforcement roles (e.g., TCRA’s inability to change the behavior of TTCL re the fixed-line quota in its contract). In this instance, DAWASA was vigilant in monitoring CWS performance, in taking the corrective measures allowed in a forceful and timely manner, and in bringing in other actors—the Minister, the World Bank, the mediators—in an effort to solve the dispute. DAWASA and Government showed greater flexibility during the mediation process than CWS. If GOT goes ahead with its plan to conclude a lease between DAWASA and DAWASCO ( the new public operating company set up to replace CWS), DAWASA’s past asset and contract management experience will help. 5. A second positive note is that the associated works program seems to have gone well; these improved facilities are presently coming on line and should lay the base for future service expansion and improvements. V. Conclusion Shortly after the lease was cancelled, the presiding Minister termed these events a case of “contract failure, not policy failure.� However, the fact is that private provision of water services has proven problematic the world over, and especially in Africa, where there have been problems in Mozambique, Guinea, Mali, Uganda, and now Tanzania—though there have also been a few African success stories in privately provided water, in Cote d’Ivoire and Senegal, for example. In this sector, GOT would have been better off had it tried a management contract, not a lease. The way forward: priorities for action The GOT has already taken a number of steps in response to the cancellation of the lease with CWS. It has created DAWASCO, a state-owned entity that will perform the same 35 role as CWS, eventually under a similar contractual relationship, i.e. a lease. It has appointed an experienced CEO for DAWASCO. In addition, it has ensured continuity in donor financing for the sector, in particular for the rehabilitation works. These steps have kept water flowing and avoided an acute crisis. The next priority is to apply the lessons of this case. The first step will be to corporatize DAWASCO so that it operates as an enterprise subject to commercial legislation. This is key for the success of any contractual relationship between DAWASCO and DAWASA. The second priority is to immediately strengthen DAWASCO’s management team to put top performers in at least the three positions vacated by the expatriate managers. The appointment of the CEO deals with only one of the three positions. Subsequently, we recommend that the contractual relationship between DAWASCO and DAWASA be revisited. Since a lease proved too ambitious for a private operator, it would seem plausible that a public operator might face similar problems. The option of a management contract should be considered for DAWASCO. Finally, Tanzania might be well advised to seize this moment to initiate reflection and a debate on a national water sector policy that goes far beyond city water to the peri-urban and rural areas. In particular, lessons could be drawn from successful experiments in neighboring Uganda in providing water in provincial towns and peri- urban areas using local private operators. 36 Appendix A DAWASA/CWS Lease Contract: Financial & Operational Requirements, Incentives & Penalties Conditions: If the operator:  Fails to pay the Lessor Tariff to the Lessor within 2 weeks of the due date, it shall pay to the Lessor a penalty equal to 0.5% of the Weekly Revenue (WR) for each week of delay after the due date.  Fails to pay the Rental Fee or any Additional Rental Fees to the Lessor as agreed within 2 weeks of the due date, it shall pay to the Lessor a penalty equal to 0.5% of the WR for each week of delay after the due date.  Fails to produce documents for financial and accounting provisions within 15 days of being required to do so, it shall pay to the Lessor a penalty equal to 0.5% of the WR for each week of delay after the due date.  Fails to indemnify the Lessor, it shall pay to the Lessor a penalty equal to 0.5% of the WR for each week of delay after the request to produce the documents was made.  Fails to meet the key performance indicators it shall be liable for the penalties contained in Annex Table 2 (below) and contract can be terminated if targets are not met.  Continuously misapplies the First Time New Domestic Water Supply Connection Fund, it shall, upon the first written notification by the Lessor and upon each subsequent written notification by the Lessor of an application to the Fund pay to the Lessor a penalty equal 0.5% of the WR.  The maximum liability for the Operator under any year should be limited to 7.5% of its income.  If a sum due from the Operator to the Lessor is not paid before the due date then the Lessor is entitled to deduct such sum by way of set-off from any sums due from the Lessor until the Lessor has recovered the full amount. Limit of Liability: The Liability of the Operator or the Lessor should not exceed 10% of the Operator’s income in any Contract Year. 37 Appendix B: Key Performance Indicators Key Annual Performance Targets Performance Targets Component Yr1 Yr2 Yr3 Yr4 Yr5 Notes Drinking 95% 95% 95% 95% 100%** Water Quality Leaving Water Treatment Plant/ Borehole source Drinking 95% 95% 95% 95% 100%** Water Quality in Distribution Effluent Transition Transition Transition 90% 95%** Quality Value Value Value Customer 16,500 25,000 30,000 30,000 31,000 Meter Installation New 1,000 4,000 10,000 10,000 10,000 Domestic water supply connections Transmission Transition Transition Transition Transition 7%** main losses Value( which Value Value Value shall not be worse than base value) Water Transition Transition Transition Transition 28%** Base value distribution Value( which Value Value Value to be losses shall be not determined worse that base for each value) separate district meter zone Collection 60% 70% 80% 85% 90% Efficiency Repair time for reported bursts on water supply pipes: 38 Service Pipe Repair (in 75% 85% 90% 95% Target Repairs and accordance time for mains up to with the repair is and including obligations of 12hrs 100mm the contract) all diameter reported bursts Mains above as exit or after 60% 70% 80% 90% Target 100mm commencement time for diameter and date repair is up to and 18hrs including 300mm diameter Main above Set up system 50% 60% 70% 80% Target 300mm for undertaking time for diameter and and mentoring report is up to and burst repairs 24hrs including 600mm Mains above 40% 50% 60% 70% Target 600mm time for diameter repair is 48hrs Repair time for reported background losses on water supply pipes Service pipe Repair (in 40% 50% 60% 70% Target repairs and accordance time for mains up to with the repair is and including obligations of 48hrs. the 100mm the contract) all breach of diameter reported integrity of background pipe sizes losses as greater existed at or than after the 100mm commencement diameter date. Set up shall be system for considered undertaking to be a and monitoring “burst� background losses and repair Data Acquire and Acquire Acquire Acquire Acquire Collection report data in and report and report and report and order to annual annual annual report determine base data data data annual 39 values data transition values, year five targets % of No worse than Transition Transition Transition 30% Customers base value Value Value Value receiving less than 5m pressure at the tap %of No worse than Transition Transition Transition 50% Customers base value Value Value Value receiving less than 10m pressure at the tap 40 Appendix C: Financial Penalties on Non-compliance with performance targets: Target Targets Contract Contract Contract Contract Contract Category not Yr1 Yr2 Yr3 Yr4 Yr5 complied (Tsh (Tsh (Tsh (Tsh (Tsh with Million) Million) Million) Million) Million) Drinking NA 32 64 97 129 water leaving from plant “A� Water 32 64 97 129 Targets quality in distribution Effluent 32 64 97 129 quality Data 257 32 64 97 129 collection One target NA 0 0 32 64 failed Two 0 32 64 129 targets failed Three 32 64 129 257 targets “B� failed Targets Four 64 129 257 515 targets failed Five 129 257 515 751 targets failed Six targets 257 515 751 923 failed Seven or 515 751 923 1,094 more targets failed 41 Case Study 2 The Private Participation in Tanzania Electricity Supply Company Ltd. I. Private Participation In 1997, the Tanzania Electric Supply Company Ltd. (TANESCO) became a “specified public corporation� when it was selected by the Government of Tanzania (GOT) for restructuring and privatization. At the same time, it was recognized as a highly indebted state-owned utility, and the Presidential Parastatal Sector Reform Commission (PSRC) became not only the manager of any future privatization transaction for TANESCO, but its official receiver as well. PSRC henceforth supervised routine transactions of TANESCO and monitored the restructuring process in line with the objectives set forth in the National Energy Policy. However, by law, PSRC cannot value, sell or dispose of TANESCO shares or assets without approval from the Treasury, the line minister and the Attorney General. TANESCO will remain under the supervision of the receiver manager, PSRC, until it is fully restructured and privatized. Power Sector Policy Reform (1999) In 1999, the GOT approved in principle a new electricity industry policy and restructuring framework, the Power Sector Reform Policy. Its objectives were to increase the efficiency of the electricity sector to meet electricity demand and provide for a sufficient reserve margin; to accelerate electrification so as to ensure access to the broadest cross-section of the population and centers of economic activities; to ensure the long term economic viability and sustainability of the industry so that it can meet the challenges of economic development; and reduce public sector expenditure and the debt of the electricity sector. This general policy contained the following specific proposals:  The vertically integrated state-owned electricity monopoly would be un-bundled vertically into functional units of generation, transmission and distribution;  Competition in generation and distribution would be introduced via horizontal un- bundling, (where applicable), while seeking to safeguard stakeholders and consumers’ interests through regulation;  New legislation would be enacted to capture structural changes in the electricity sector (and replace the Electricity Ordinance of 1931);  A regulatory agency would be established prior to the privatization of major state- owned assets in the electricity sector. 42 The GOT’s intention was to provide necessary incentives to encourage private investments, to apply the least cost basis principle for investment programs and to access concessionary funds to foster rural electrification program. However, all of what was proposed remains to date on paper. The GOT has not yet un-bundled the sector nor passed the new electricity law (a draft bill has long been prepared). Nor is a fully functioning regulatory agency yet in place. The GOT has taken a gradual, step by step approach, and so far only allowed private management in TANESCO while contemplating about how to implement its overall Power Sector Policy. Pre-Private Participation Performance (2000) TANESCO’s long-standing technical and financial problems spurred the GOT to reform the power sector and seek private participation. By 2000, the vertically integrated utility was facing serious liquidity problems with major system losses (25%), low bills recovery (80%), high payroll expenses and the absence of effective control systems on critical expenditures, particularly fuels. Moreover, it was suffering from deferred maintenance and past underinvestment. Data was sparse and of poor quality; it was difficult to assess TANESCO’s technical and financial performance. KPMG (Forensic and Investigative Accounting Group) of South Africa (with funding from the Swedish International Development Agency [SIDA]) was hired in 2001 to get a baseline data prior to private participation. The consultants had difficulties collating the fixed and variable costs by plant per month and measuring system losses. They detected major discrepancies among data provided by the generation and transmission departments. Forced outages were recorded manually at the feeder stations and they were not always accurate. The consulting team concluded that one of the major threats to the financial position of TANESCO were weaknesses in the metering, billing and collection process. Another worry was that the aggregated salary expenses had almost doubled between 1998 and 2000. While the aggregate basic pay increased three and a half times, overtime and pension fund contributions doubled within two years. Medical expenses decreased substantially yet the “safari allowances� almost doubled. These allowances were given to TANESCO employees for their out of town expenses if and when they were on official business. They included even expenditures incurred by non-company officials attending meetings outside of Tanzania including the World Bank Economic Development Conferences.19 It also appears that high payroll costs included some “ghost� employees and unexplained “promotions.� KPMG pointed out that there was no proper budget control system; this resulted in over expenditure on capital work orders. There were misallocations between capital and recurrent expenditures and “major projects� and a so-called “project dummy account.�20 There were similar reported irregularities such as unbilled items, account forging, under billing, defective metering, feeding to suspended or to discontinued accounts, accounts not in billing system, excessive credits, frauds, unaccounted cash, cash misappropriation, 19 “The Presidential Parastatal Sector Reform Commission Report� on handover audit services rendered for Tanzania Electricity Supply Company Limited, Final Report, KPMG, February 2002, p. 210. 20 Ibid., p. 225 43 delayed banking, staff payments, petty cash, inflated suppliers’ prices, non-existent suppliers, unaccounted fuels, and other unaccounted materials. All these irregularities within the four fiscal years prior to privatization, that is, between 1998 and 2001, cost the firm and the GOT an estimated TShs 5.2 billion or US$ 5 million.21 There were also some external factors outside the control of public management at the time. The need to reduce TANESCO’s hydro dependency in generation (with accompanying shift to thermal) contributed to its deteriorating finances. The fluctuations in dam water levels prompted by droughts, combined with increased demand, necessitated the use of expensive to operate22 thermal stations as well as isolated thermal plants set up with very expensive to operate diesel motors or gas turbines. Table 1. Change in Hydro-Thermal Ratio in Generation (MW) Power Stations 1998 2000 Hydro Stations 96% 86% Grid Stations 1% 11% Isolated Thermal 3% 3% TOTAL 100% 100% Source: Calculated from data in KPMG Report to PSRC on TANESCO, February 2002, p. 192 Thermal plants had more consistent but more expensive to operate generating capacity compared to hydro plants. As a result, the change in hydro-thermal ratio increased fuel consumption. TANESCO’s fuel expenses jumped from TShs. 9 billion in 1998, to TShs. 15.9 billion in 1999 and again to TShs. 39.2 billion in 2000--almost a three-and-a half fold increase between 1998 and 2000.23 Nonetheless, the critical factor leading to TANESCO’s near financial crisis in 2002 was the unsustainable financial burden of the Independent Power Producers (IPPs). This, in turn, led to the enterprise’s increased dependency on GOT subsidies and its growing indebtedness to international financial institutions. It is worth summarizing the background and the crippling financial burden of the IPPs. Financial Burden of Independent Power Producers (IPPs) Starting in early 1990s, the GOT allowed TANESCO to enter into co-generation and power purchase transactions. TANESCO became increasingly obliged to meet its 21 Explanatory Memorandum prepared by KPMG to PSRC on December 13, 2001, Annex B. p. 1-2 22 Whereas hydro-electric generating stations are expensive to build; since they involve large civil works, once built they are inexpensive to run (there is no cost for water as an energy source). Whereas thermal plants are relatively inexpensive to build, the energy source, fuel, is rather expensive especially diesel fuels as compared to gas or heavy oil. 23 Gulam Dhalla, Financial Consultant, “Financial Review of TANESCO,� report to the World Bank, July 22, 2003, p. 4 44 financial obligations signed in the Power Purchase Agreements (PPAs). 24 The IPPs vary from simple contracts such as Tanwat PPA (1993) and Kiwira PPA (1995) to more complex ones, namely the Independent Power Tanzania Ltd. (IPTL) (1997), and later Songo Songo Gas-to-Electricity Plant, known as Songas, (2001). IPTL and Songas became operational respectively in 2002 and in 2004 but preparations for both had begun several years before. The previous administration had started negotiations with IPTL before the present government assumed power in 1995. On July 9th, 1996, the GOT and the TANESCO Board agreed to a conditional electricity license to IPTL without having undertaken the necessary legal reforms to allow private entry into generation sector—and also in the absence of a regulatory authority to issue such a license. Despite some initial reluctance, the GOT decided to honor the agreement, partly because of the urgent need to substitute Tanzania’s hydro-dependency with diesel fired thermal plants. IPTL was to design, construct, own, operate and maintain an electricity generation facility with a nominal capacity of 110 MW in Tegeta (approximately 25 kilometers north of Dar es Salaam). IPTL’s initial ownership was held by Merchmar Corporation of Malaysia (70%) and VIP Engineering and Marketing Ltd. of Tanzania (30%).25 On behalf of IPTL, Stork Warstilla Diesel BV, started the construction in February 1997, at an agreed total cost of US$ 163 million.26 In its Power Purchase Agreement, the GOT committed itself to pay both the capacity charges (capital recovery fees, fixed operating fees and service fees for two elements—capital and debt component) and energy charges (energy purchase price for two elements—fuel cost component and variable operations and maintenance component), both determined in US$ and in kWh. This agreement was later subject to investigation. The issues raised included the charges calculated in US$; TANESCO’s funding the capital expenditure for the erection of a power plant without any ownership; the apparent high cost of purchasing energy as opposed to other PPAs; and the absence of a decrease in the capacity price commensurate with the decrease in the debt component thereof.27 The World Bank was not and has not been involved with this deal at all, with the exception of the arbitration (and later discussing a loan to finance the retrofit of IPTL from oil to gas). The World Bank agreed 24 Power Purchase Agreements (PPAs) are the contractual agreements to purchase power from an IPP and the same as off-take agreements whereas a long term agreement to purchase minimum amounts of the product of the project at an agreed price and it is often entered into by one of project sponsors on a take-or- pay basis. 25 It is reported that Merchmar Corporation recently transferred its shares in IPTL to another Malaysian firm, Ranhill Power Bhd (a group of boiler manufacturing companies in the UK, Indonesia and Malaysia with 62% of its gross revenues coming from power generation in Africa). Merchmar Executive Chairman, Datun Tan Keen Wan, reportedly said that his firm wanted to sell its stake in IPTL at a “fair� price although it contributed 90% of the company’s earnings in 2004. 26 Some still argue that the project cost was inflated and it should have cost only US$ 80 million, not US$ 163 million. 27 “The Presidential Parastatal Sector Reform Commission Report� on handover audit services rendered for Tanzania Electricity Supply Company Limited, Final Report, KPMG, February 2002, p. 89-147 45 to cover the cost of the dispute settlement that lowered IPTL’s construction cost while reducing its capacity charges proportionally.28 The GOT also entered into a somewhat similar Power Purchase Agreement with the Songo Songo Gas-to-Electricity Project (Songas) in late 1990s. The goal was to develop natural gas from the Songo Songo gas field in Kilwa District to provide Tanzanians with a reliable source of low cost electricity.29 The project’s present major sponsors are CDC Globeleq and Pan African Energy, with minor sponsorship from the Ministry of Energy and Minerals (MEM), TANESCO, TPDC and TDFL, together with international financial institutions such as the European Investment Bank and the World Bank. The project costs were estimated at US$ 350 million. The construction of Songas to supply gas to fire the 100 MW Ubungu Power Plant suffered considerable delays; the first gas reached Dar es Salaam only in July 2004. In sum, both IPTL and Songas (combined installed capacity of 210 MW) now generate electricity for TANESCO, which pays both capacity and energy charges to both companies. In May 2005, the financial burden to TANESCO for these two alone reached US$ 12.5 million per month, more than half of TANESCO’s gross monthly revenue of US$ 22 million. This constitutes a serious financial burden on the utility. Table 2. Monthly Capacity and Energy Charges of IPTL and Songas in 2005 (US $ millions) IPTL Songas Capacity Charges 3 4.5 Energy Charges 4 1 TOTAL 7 5.5 Source: “TANESCO 5 Year Development Plan,� presented to Parliamentarian Investment Committee, May 26, 2005, p. 3530; After some delays, the construction for Songas began in 2001. TANESCO’s present private management had very little insight into the initial agreement at the time it bid for the management contract. It only learned about the details after taking up office in May 2002. Similar to IPTL, the private management soon found that the initial Power 28 The arbitration process carried out by the International Centre for Settlement of Disputes at the World Bank took four years between 1998 and 2001. 29 The original ownership of the company was shared among the two major sponsors, TransCanada Pipeline Limited (TCPL Tanzania Inc.) and Ocelot Tanzania and the minor sponsors were TANESCO and Tanzania Petroleum Development Corporation (TPDC). The major sponsors were replaced in 2002 by AES Sirroco of USA (who later was again replaced in 2003 by CDC Globeleq [100% owned by Commonwealth Development Corporation Group]) and Pan African Energy, a gas developing company with operations in several African countries, taking over the shares of Ocelot International. 30 It should be noted that the interviews with TANESCO’s private management on June 16, 2005 revealed that the total monthly capacity and energy charges for both IPTL and Songas has even increased to US$ 14 million (of which US$ 8 million is for IPTL and US$ 6 for Songas) by June 2005. 46 Purchase Agreement signed between TANESCO and Songas had not properly researched the future cash flow implications of these transactions on TANESCO’s expenditures and its subsequent impact on the company’s tariffs. The Transaction (2000-2002) Strategy The GOT decided to bring private participation to TANESCO in 2000 in advance of a consensus being reached among the stakeholders as to how this participation should or would be done. The sector was not yet un-bundled and the enterprise was certainly not yet ready for a lease or a concession contract, much less for divestiture. Furthermore, it would take a great deal of time to design and implement complicated transactions of this nature. Thus, as an interim solution, all parties—PSRC, the Ministry of Energy and Minerals and TANESCO’s Board (as well as the international donors)--agreed that a “performance-based management contract� was the best available form of private participation. It was decided that it would be put in place for a period of 24 months with a possible extension of an additional year. On a parallel track, PSRC continued to work with international consultants for eventual corporate restructuring of TANESCO and to allow for its vertical and horizontal un- bundling. Since the World Bank and SIDA were intimately involved with TANESCO’s financial problems, they assisted in preparing the terms of reference for the bid documents of the Management Support Services Contract (MSSC). Process Competitive Tender. PSRC compiled a long list of possible contractors. The merits of each potential bidder were assessed according to agreed criteria aiming at a short list of about six contractors. Each potential contractor was contacted in writing to make sure that it was willing to bid. Bidding documents were issued using a prescribed World Bank “Request for Proposals� format; selection was made in terms of World Bank technical and financial adjudication procedures. Originally eleven firms expressed interest and 6 were short-listed. These included Eskom Enterprises Ltd., ESBI International Ltd., Iberafrica Systems Ltd., NetGroup Solutions Ltd., Vattenfall International and NRECA International.31 Only four of this group attended the pre-bid conference. One asked for a time extension which was not granted. The GOT was not willing to lose momentum and accept any delays. Three bids were received and evaluated on technical merits by an experienced team drawn from various ministries. The final three bidders were: 31 “Tanzania Electricity Management Contract,� in: Experience with Private Participation Initiatives in Sub-Saharan African Infrastructure: What at the Lessons for Future Policy?, Castalia Report for the World Bank, June 2005, p. 105 47  NetGroup Solutions (NGS) Ltd. (South Africa)  Eskom Enterprises (South Africa)  ESB International (Ireland) The bids were evaluated in a way that the technical competence of the private operator got 80 percent, with the remaining 20 based on the fee demanded. The Winning Bid. One bid was judged incomplete, and another was not compliant with Tanzanian law. In the end, there was only one technically acceptable bid, that of NetGroup Solutions (NGS) Ltd. (The NetGroup). The Management Services Support Contract (MSSC) was awarded to the Net Group. Although originally intended to commence on January 1, 2002, NetGroup’s takeover was delayed. The contract was eventually signed to cover the period between May 2002 and May 2004, with the option to extend it for one year. PSRC and TANESCO’s Board jointly became the supervising contractual parties of NetGroup. Management Contract Structure The Management Services Support Contract (MSSC) was designed to turn around TANESCO both financially and technically. Financial turnaround was to be achieved by increasing monthly revenues. This would be done by a well-orchestrated campaign to cut-off non-payers, and, at the same time, cut costs by introducing strict financial discipline within the organization to reduce unnecessary spending and waste. Technical turnaround aimed at upgrading capacity, refurbishing deteriorating systems and conducting routine maintenance to keep the system in serviceable condition. The contract included support to TANESCO’s Board of Directors and PSRC to prepare and implement the next stages of privatization activities. The new management agreed to “ring-fencing� (separating the accounts of) generation, transmission and distribution as a first step towards unbundling. Concerning remuneration, the first management contract set up a lump-sum fee with two components: A fixed retainer fee of US$ 2,649,010 and a relief projects fee up to a maximum of US$ 1,309,380. For the second portion of the fee, the private management had to submit a plan for each of the proposed projects for consideration by TANESCO’s Board. The MSSC specified the terms of payment for both fees. A second, very important feature of the remuneration for the private management was a success fee, paid on the basis of achievement of measurable performance targets. The success fee was calculated on (a) the monthly percentage of power loss reduction (technical efficiency); and (b) the monthly improvement in the ratio of operating costs to revenue collected (operating efficiency). The latter gave an incentive to the private managers (i) to cut costs (via expenditure control, improved management information reporting, reduction of generating costs, timely implementation of disconnection policy and other measurable performance indicators), and (ii) to increase revenue (via improved customer management practices including accuracy of billing, hooking up new customers improved cash collection and). 48 The management contract clearly spelled out the obligations of both parties including dispute resolution mechanisms. The NetGroup referred to its own team as “utility doctors� and their good understanding of TANESCO’s financial and technical difficulties was well presented in their bid. On May 17, 2002 the NetGoup took complete charge and began to manage all aspects of the operations, maintenance, and expansion of TANESCO for 24 months applying commercial electric utility management principles. This first contract ran for two years and ended on May 17 2004. It was first extended on an emergency basis for three months, and later and more formally for another 24 months. At the moment, the NetGroup is implementing its second management contract, which is scheduled to end on December 31, 2006. The NetGroup has developed an extensive computer model for calculation of its own success fee on a quarterly basis. Table 3. NetGroup’s Total Remuneration during the Nine Quarters (From June 2002 to September 2004) Total US$ Average US$/month I. Fixed Retainer Fee 2,649,010 110,375 II. Success Fee: 4,399,290 162,937 a. Power Loss Success Fee 21,995 815 b. Operational Efficiency 3,784,295 140,184 c. Adjustments 4,377,295 162,122 TOTAL REMUNERATION 10,833,279 413,496 Source: Management Monitoring Report, Negotiations on New Management Supply Service Contract, TANESCO, June 9, 2005, p. 4. The on-going second management contract has an agreed fixed retainer fee of US$ 5,547,182 for the period between August 1, 2004 and December 31, 2006. It also has a turnaround activity fee, subject to approval by TANESCO’s Board, in the value of US$ 5,000,000. The second MSSC has a slightly different incentive scheme for the success fee compared to that of the first management contract. It is no longer based only on power loss and operational efficiency. The full payment of the success fee for operating efficiency is also conditional upon the private management meeting minimum performance levels for quality of service and supply. The second management contract now includes three elements with specific rewards and penalties:  A financial bonus of 4% of the operating profit (EBITA) above US$ 6 million per quarter in 2004, US$ 7 million per quarter in 2005 and US$ 8 million per quarter in 2006; 49  Quarterly improvements in a quality of electricity supply measure, known as the Customer Average Interruption Duration Index (CAIDI), where the contractors can gain or lose US$ 1,000 per minute over or under the agreed amount;  Electrification by number of new connections above an agreed target, paid/penalized at a rate of US$ 30.32 Initially it was agreed that the contract would be monitored through TANESCO’s Board of Directors, PSRC, and the Ministry of Energy and Minerals, and the Audit Committee of the Financial Affairs of the company. A Board appointed (and SIDA funded) Monitoring Consultant (MC) assists the Board to monitor the performance of the NetGroup, assessing performance and outcomes, and recommending the value of the success fee. The Monitoring Consultant recommended changes in the incentive structure of the first contract; these were adopted and the second contract was revised accordingly. III. Results and Outcomes (2002-2004) Operational and Financial Performance The remuneration arrangement correctly gave a very strong incentive to NetGroup to decrease power losses and increase revenue collections. Substantial improvements were achieved during the first contract. However, Tanzania (and TANESCO) was hit by a severe drought in late 2002-early 2003, putting the company back into a catastrophic financial situation. Management efforts—supported by the relief projects (Financial Modeling, Tariff Analysis, Network Performance Improvement, Utility Information Systems Improvement, Revenue Management and Meter Audit and Distribution System Rehabilitation)—helped to achieve the necessary technical turnaround. In addition to the drought, which reduced water behind dams to very low levels and decreased electricity generation, TANESCO was hit by the heavy financial burden of the two IPPs (IPTL and Songas) as they became operational in 2002 and 2004 respectively. Absent these factors, TANESCO’s performance would have improved even more than it did. Operational Performance The first management contract achieved significant results in operational performance:  Installation of 30,000 new connections, increasing the total connections from 500,000 to 530,000 in 2003. Expansion later slowed as a result of the shortage of materials and negative effects of the drought on TANESCO’s finances;  System losses were reduced from 28% to 23% (less than expected because faulty meters were installed);  Network reliability improved and the number of partial grid failures was reduced; 32 Management Monitoring Report, Negotiations On New Management Supply Contract, TANESCO, June 9, 2005,p. 13 50  Increase in bill collection from 67% to 93% while bringing in extra revenue of US$ 5 million per month. Within a year and a half after taking over private management, TANESCO collected substantial government arrears amounting to US$ 16 million by December 2003;  Hydro-thermal ratio fell from 86:17 to almost 50:50 by the end of 2004. Financial Performance The first management contract also achieved significant financial results (again, despite the drought):  Annual revenue increased by 35% from TShs 139 billion in 2001 to TShs 189 billion in 2004 (monthly revenues further increased to TShs 22 billion during the first half of 2005);  Annual operating expenses were reduced almost by 30% from TShs 62 billion in 2001 to TShs to 45 billion in 2004; this was made possible by enforcing authority limits, applying strict procurement practices in line with GOT guidelines and by promptly settling suppliers’ invoices;  While TANESCO owed approximately US$ 33.8 million plus a bank overdraft of US$ 9.2 million in 2002, after two years the debts have been paid and liquidity improved;  Net cash balances increased and net liquidity (cash balances less current liabilities) improved, thus enabling TANESCO to secure two commercial loans from private banks. This has led to some additional investments. Despite these significant improvements, TANESCO still remains a loss-making enterprise with its cost of sales above its revenues. Moreover, its equity is shrinking, and it has outstanding loans and accrued interest in the order of US$ 800 million to the GOT; this debt to government has almost tripled (from U$ 281 million in 1998) over the last six years.33 33 “Policy Solutions for Parastatal Debt Study,� Draft Report, prepared by Parastatal Sector Reform Commission, Vol. II., November 1998, p. 112 51 Figure 1. Change in Financial Performance (2001-2004) TANESCO - Financials 800.0 600.0 400.0 2001 Tsh bn. 2002 200.0 2003 0.0 2004 Sales Cost of Net profit Taxation Net profit Equity -200.0 sales (loss) (loss) before after tax -400.0 taxation Source: Information gathered from the TANESCO Annual Reports, 2001-2004. Impact on Stakeholders Consumers Consumers expect to benefit from TANESCO’s private participation through increased access and coverage, more affordable tariffs, and improved quality and reliability of service. When the private management arrived in 2002, network expansion has come to a halt as a result of institutional limitations and severe lack of funding. The Net Group restarted the electrification process once the financial turnaround was achieved by increasing new connections and therefore access and coverage. The NetGroup has increased connections around 30,000 per year—better than the past but still slow for a country with only 10% of its population having electricity access and 90% un-connected. Accelerated electrification is achievable provided institutional capacity could be created and a clear mandate can be given to solicit funding for the process.34 34 NetGroup, TANESCO Management Support Services Contract, Stakeholders Report, No. 4, 2004, p.14 52 Figure 2. Increase in Electricity Coverage (2000-2004) 600000 500000 400000 300000 200000 100000 0 2000 2001 2002 2003 2004 Customers New customers Source: Data provided by TANESCO management, June 2004. The private management introduced a simplified domestic tariff structure by designing a “poverty tariff (D1)� for customers in low-income groups and a “general usage tariff (T1)� for customers with higher energy consumption. The former subsidized lifeline tariffs were also reduced from 100 kWh to 50 kWh per meter reading period. The “poverty tariff� is for domestic users consuming up to 50 kWh at a rate of TShs 38 per kWh, and there is no basic monthly service charge. This tariff is 40% below cost and is the lowest in East Africa. If poverty tariff users increase their consumption above 50 kWh, they pay a rate of TShs 115 per kWh (instead of TShs 38 per kWh) up to a consumption level of 236kWh. However, the cost above 50 kWh is phased up to 236 kWh and customers are not penalized for the full amount if they exceed the subsidized part. Customers pay only for units used and without any additional monthly service charge.35 The “general usage tariff� starts at consumption levels above 236 kWh. Customers in this category pay TShs 95 per kWh, with a basic monthly service charge of TShs 1,700. Half of TANESCO’s electricity revenue is generated by its domestic users (paying domestic low tariffs and general usage tariffs) whose consumption remains below 7500 kWh. The rest of its revenue comes from low voltage supply to large domestic customers and small businesses, which pay a low voltage maximum tariff (T2) of TShs 63 per kWh. High voltage customers pay the high voltage maximum demand tariff (T3) of TShs 58.50 per kWh. Both T2 and T3 users, however, pay a higher basic monthly charge. Additional revenue is raised by supplying electricity for public lighting and to Zanzibar. Under the private management, TANESCO customers experienced only two modest tariff increases of 4.8% in June 2004 and the same in January 2005 out of three tariff 35 “TANESCO 5 Year Development Plan� presented to Parliamentarian Investment Committee, May 26, 2005, p. 26 53 increases proposed.36 The tariff increase of 4.3%, which the NetGroup proposed in May 2003, was sanctioned by TANESCO Board and did not take effect. The tariff increase of 4.8% in June 2004 was accompanied by the new domestic tariff (D1). As the old domestic tariff (T1) remained but excluded any poverty subsidy, the domestic customers experienced a real tariff increase of approximately 10%. The private management introduced another average tariff increase of 4.8% in January 2005 in an attempt to bring the large customer tariffs in line with other countries in East Africa so that TANESCO’s rates are more competitive. The current average tariffs of about US 7 cents in Tanzania still remain low: Kenya is charging US 8.2 cents per unit while Uganda is charging US 8 cents.37 TANESCO’s domestic and industrial customers have benefited from improvements in quality and reliability of electricity. Despite unanticipated problems, the private management has been thus far careful to avoid load shedding. Yet the system remains vulnerable. The very old and overloaded electricity infrastructure is not coping very well under high demand conditions. Rapid strengthening and renewal of the system is imperative. Employees Initially TANESCO employees were extremely opposed to private management. Access to TANESCO by the private management team was denied; the start of the contract was delayed from January 1, 2002 until May 17, 2002 because labor protested, fearing that the new managers would fire large numbers of the workforce. Reportedly, the out-going public management instilled this fear. Finally, the NetGroup management team arrived on May 17, 2002, under the protection of the Field Force Unit (special police). The Chairman of the TANESCO Board welcomed the new management team, and the transition was conducted peacefully. In the past there was not a good communication between the public management and the unions. The employees still complain that the previous management never solicited their opinion regarding privatization. In contrast, the NetGroup has managed to build a very healthy relationship with the workforce. In 2003, the unions agreed to a generous Voluntary Retrenchment Agreement. A total of 1319 employees, representing 22% of the labor force, were paid a total severance package of TShs. 21,559, 905,000. This suggests an average pay-out per employee of TShs 16,345,644 (US$ 15,605) far exceeding the sums paid out in most other privatization programs.38 36 At any one stage only a tariff increase of below 5% is allowed and approved by TANESCO Board and any increase above 5% needs to be approved by the GOT. 37 Additional tariff increases are highly likely. The private management projects that the average tariff will increase from the current TShs 70/kWh to TShs 130/kWh, if the GOT fails to write-off its long term debt, subsidize IPTL by 50%, buy down Songas, or relax Songas on on-lending terms and conditions, relax taxes and duties, in: “NetGroup Management Support Service Contract: Stakeholders Report,� No.4, 2004, p.9 38 Reportedly the average employee payout for US$ 12,500 in Argentina was considered one of the highest. It is interesting that TANESCO then surpasses Argentina in its generous severance packages with June 2003 exchange rate of TShs 1,047.4 to US$ 1. 54 Table 4: Trends in Employment between 1995 and 2005 1995 1999 2000 2001 2002 2003 2004 Employment 7440 6899 6662 6540 6433 4996 4857 % Reduction 7.3% 3.4% 2% 1.6% 22% 2.7% Source: Information gathered during interviews with TANESCO management, June 2005. TANESCO’s employees benefited from privatization in a number of ways. While the Voluntary Retrenchment Agreement led to a generous golden handshake, the remaining employees received a flat salary increase of 20% across the board salary in early 2005. The private management has also recently engaged the services of a consultant to conduct a review of existing salary scales, with significant input from the trade unions. Union representatives reported that the private management created a “Trust Fund� where the NetGroup puts aside an amount commensurate to their own bonus. The Fund is intended to pay out the workers at the time of any future retrenchment or retirement, and is monitored by the union representatives themselves. The NetGroup has also been keen on bringing TANESCO into line with world standards and trends from which it had been isolated in recent years. It offered training to TANESCO employees on the use of a testing device of its own design known as the Portable Energy Meter Testing Instrument (PEMTI). The instrument which has been successfully used in South Africa allows a single person to test up to 60 meters a day and small teams can audit a city within a couple of months. In addition, the NetGroup also streamlined and modernized the management practices within TANESCO with a particular focus on introducing information technology to replace a multitude of outmoded and cumbersome manual systems. Since the employees benefited both directly and indirectly from the presence of private management, it is therefore understandable that they have become supporters of the private management team, and in particular the Managing Director, Rudy Huysen. Government Since we did not have the trend in past government subsidies for recurrent and development expenditures, we could not accurately assess the fiscal impact of private participation. (The data received thus far from the Ministry of Finance on subsidies was sparse and did not show the trend over the past 10 years.) However, the rough data from the Ministry of Finance illustrates that subsidies for recurrent expenditures rose in 2004/2005 to TShs 99.7 billion from TShs 18 billion in 2003/2004 and TShs 13 billion in 2002/2003. This substantial jump in 2004/2005 compared to the previous two years is related to the capacity charges for Songas (TShs. 28.5 billion) and to the fuel cost allowance for operating the Ubungu Power Plant during the delayed construction of Songas (TShs. 53.2 billion)39 39 Information on subsidies is received by the Ministry of Finance, July 11, 2005. TShs 28.5 billion subsidy for Songas hardly covers TANESCO’s total expenses of TShs 31.6 billion incurred for late commissioning of Songas turbines (TShs 18 billion), bad performance of the plant (TShs 2 billion) and not 55 TANESCO also received modest subsidies for development expenditures. They were, however, dedicated exclusively to rural electrification amounting to modest TShs 1.48 billion in 2004/2005, TSh 1.30 billion in 2003/2004 and TShs 1.21 billion in 2003/2002. The government did not necessarily benefit from TANESCO’s improved efficiency via increased taxes. The total taxes paid by TANESCO remained more or less the same after private management took over. Taxes deferred (import duty, excise duty, and VAT) also increased in 2003 and 2004. As a state-owned utility, TANESCO even under private management still enjoyed the preferential tax treatments from its owner, the GOT. Moreover, TANESCO continues to owe a handsome sum of US$ 800 million to GOT which has not been serviced for many years. In the past the GOT has obtained grants or concessionary loans (typically on IDA terms) from donors and on-lent them to TANESCO on quasi-commercial terms with TANESCO bearing the foreign exchange risk. This arrangement was intended to make TANESCO face a market cost of capital and therefore ensure efficient allocation decisions. But several factors imposed constraints on TANESCO’s ability to promptly service its debt to GOT. They were (i) poor enforcement and oversight by GOT of its loan agreements with TANESCO, while donors kept lending to TANESCO despite the fact it was not servicing its debt, and (ii) the practice of forcing TANESCO to borrow from GOT at near commercial terms in order to finance GOT’s social obligations for rural electrification with no prospect of recovering their costs40. This level of debt is unsustainable. A massive debt restructuring measure awaits the Ministry of Energy and Minerals and the Ministry of Finance. The NetGroup (as well as the international donors and lenders) has frequently pointed out the urgency of debt restructuring so that TANESCO can continue to borrow for its most needed investments. Overall Economy The overall Tanzanian economy appears to have benefited from private management considering the utility was on the verge of a collapse when the private management team took over. The poor performance of the power sector has been a serious problem and businesses were severely complaining about power shortages. The median enterprise reported that it suffered power outages on 48 days in 2002—more than any comparator countries. Median generation losses due to power shortages were also higher in Tanzania than in the comparator countries. Most large businesses, if they could afford it, were using independent generators because of TANESCO’s unreliability of service. Enterprises in the tourism sector were more concerned about power with close to two- thirds ranking it as a serious constraint.41 The operational and financial performance improvements brought along with the NetGroup within a short period of two years is hardly sufficient to have a fundamental sharing in the grace period (TShs 13.6 billion), as stated in: “TANESCO 5 Year Development Plan� presented to Parliamentarian Investment Committee, May 26, 2005, p. 36 40 Gulam Dhalla, “Financial Review of TANESCO,� prepared for the World Bank, July 22, 2003, p. 7 41 Investment Climate Assessment: Improving Enterprise Performance and Growth in Tanzania, The World Bank, November 2004, p. 79 56 impact on overall economy. However, private management’s ability not to load shed and to reduce previous outages definitely has a positive impact on overall economy. III. Explanations and Lessons Learned In terms of performance post-contract compared to performance pre-contract, the measure is a great success. As noted, however, much remains to be done to make TANESCO fully successful and sustainable. Despite the competence and commitment of private management under a well-designed and executed management contract, there are factors not under the control of the NetGroup that hinder the company’s growth and expansion. The utility still requires government subsidies to cover at least partially the capacity and energy charges imposed by the two major IPPs (IPTL and Songas). It also needs to turn to lenders (international and domestic) to finance its investments. Unfavorable South-South Investment In the absence of local capital and the reluctance of many foreign investors to commit to the region, South-South investments have recently become a favored alternative in several African (as well as Asian and Latin American) countries. However, some private investors from other developing countries in Asia took advantage of the liberalization in energy sectors. For example, Westmont Corporation of Malaysia brokered a crucial stake in power generation through Merchmar Corporation of Malaysia in IPTL. 42 Merchmar Corporation was one of the two unsolicited bids received for the construction of an electricity generating plant. It needs to be noted that many IPPs were entering into electricity markets in emerging markets in the mid-1990s in the absence of competitive bidding without much transparency. The poorly constructed IPTL deal (in terms of the national interest) increased the financial burden of TANESCO to the point of bankruptcy. It turns out that the Merchmar Corporation’s investments in Tanzania proceeded in spite of the warnings by the anti- corruption agency that officials had been bribed. Press reports alleged at vast kickbacks at very high levels. The risk analysis of IPTL was poorly done, according to the critics, and it imposed unnecessary costs by stalling development of much cheaper power projects. Those projects were more likely to have exploited the country’s abundant natural gas reserves. The IPTL deal has forced Tanzania to use scarce foreign exchange to import diesel as agreed in the Fuel Supply Agreement (FSA).43 The IPTL transaction is now widely recognized as a deviation from the least cost generation expansion, caused in part in the absence of an appropriate legal and regulatory framework. TANESCO has been continuously seeking relief from this unfavorable 42 It is reported that Merchmar Corporation recently transferred its shares in IPTL to another Malaysian form, Ranhill Power BHD (a group of boiler manufacturing companies in the UK, Indonesia and Malaysia with 62% of its gross revenues coming from power generation in Africa). 43 Gitau Warigi, “East and East-Central Africa,� in: Global Corruption Report 2001, Transparency International, p. 78-79 57 commercial transaction into which it was forced. The deal has been subject to international arbitration proceedings in 2001, resulting in a reduction of plant tariff by 25%.44 At the same time, a new World Bank loan is extended to convert IPTL to gas- firing. There is, however, an on-going dispute between IPTL’s shareholders that might force the company into liquidation, possibly providing TANESCO with an opportunity to restructure an even commercially more favorable new deal.45 Need for Government Subsidies TANESCO’s increased need for GOT subsidies is a result of poorly structured PPA deals, and the GOT’s policy on not passing on to end-users the concessionary terms it receives from donors and lenders. After Songas became operational in 2004, the subsidies rose to TShs. 99.7 billion in 2004/2005.46 When broken down, TShs 81.7 billion was dedicated exclusively for Songas (TShs 28.5 billion of which for its capacity charges, and TShs 53.2 billion of which for the fuel allowances) and the remaining TShs. 18 billion for the capacity charges of IPTL. On a monthly basis in 2004/2005, the GOT subsidies for capacity charges average (excluding the fuel allowance) to TShs 3.8 billion. This suggests that they constitute only 31 % of total monthly capacity, energy and fuel expenses TANESCO is obliged to pay combined for IPTL and Songas (a total of US$ 12.5 million per month in 2005). (See section under 1.3. Financial Burden of IPPs). On May 9, 2005, TANESCO requested that the GOT share the grace period for Songas so that it can survive in the short term. It estimates that if the loan and interest payments are differed for 3 years, the total monthly capacity payments to Songas will be reduced by 40% to US$ 2.35 million. This will substantially reduce the utility’s cash flow burden. Reliance on International Donors and Lenders The success of the management contract was made possible by the generous grants from the Swedish International Development Agency (SIDA). It has paid for the fixed management fee of the NetGroup. A portion of the increased TANESCO revenue covers the success fee. As of June 2005, the total amount of funding received via PSRC to the NetGroup reached US$ 3,616,897 (or US$ 3.6 million) in 2004 and 2005. In addition, the IDA credit also enabled PSRC to pay for the consultancies on design and restructuring and trading arrangements in the amount of US$ 2.5 million.47 Including the relief projects during the first contract, it is expected that both management contracts will cost SIDA approximately a total of US$ 8 million by the end of December 2006. Besides grants, TANESCO needs the support of the international lenders as it tries to attract new investments in order to survive commercially. New investments are required for generation, transmission and distribution expansions and upgrading on a continuous 44 “Memorandum and Recommendation of the President, IDA, to the Executive Directors on a Proposed Credit of US$ 43.8 million to the Government of Tanzania for an Emergency Power Supply Project,� The World Bank, Report No. P7623-TA, May 14, 2004, p. iv. 45 NetGroup, “TANESCO Management Support Services Contract,� Stakeholders Report, No. 4, p. 3 46 Information received from the Ministry of Finance, July 11, 2005 47 Data provided by the Accounting Office at PSRC, July 14 , 2005 58 basis. TANESCO’s next five year plan projects its investment needs for US$ 614 million (mostly in transmission and distribution) and it expects to borrow half of this amount commercially while the rest is projected to come from grants, capital contributions and TANESCO’s own funds. Competence and Dedication of Private Management It was both the competence and the dedication of the private management team that allowed them to turn around TANESCO technically and financially at a time of crisis. The professional team of five utility doctors was proposed in NetGroup’s original bid. After winning the contract, the same team, without any replacements, took over TANESCO’s management. All team members have excellent professional qualifications with intensive experiences in managing electricity utilities. The Managing Director, Mr. Rudy Huysen, has started from scratch three distribution utilities, one in South Africa and two in Namibia. He worked as managing director of eight electricity enterprises in Southern Africa. All of these have been success stories. Huysen’s strength is his ability to turn a badly performing electric utility into a success. His former experience includes working as Financial Director of Utilities on numerous occasions in addition to working directly on generation, transmission, and maintenance and job control. Similarly the rest of the management team has extensive qualifications and expertise in managing electricity utilities. All have worked at one point or another in their careers with Eskom Enterprises in South Africa. They have management and business skills in addition to their specific knowledge and experience in metering systems, data management and geographical information systems and revenue management. Well-Designed and Executed Management Contract Last but not least among the factors accounting for this success story was the well- designed and executed management contract with a performance-based incentive structure. This made all the difference for TANESCO’s turnaround and delivered good results. The architects of the management contract need to be congratulated and PSRC should consider taking the elements of this contract design into account when initiating other lease and concession contracts in the future. A two-year contract is too short, however, for any private management to achieve a complete technical and financial turnaround of a large utility such as TANESCO. Now that the management contract has been extended for another two years, the private managers might be able to achieve even better results. The general public appears not well informed about the importance of the incentive bonus given to the private management team. The success fee received by the private managers is only a fraction of the savings they bring to the enterprise. Some Tanzanians argue that government policy, not the private management, has been responsible for achieving the positive results. They point out to government allowing TANESCO to disconnect electricity to non-payers, including police stations, hospitals and even to Zanzibar. What is important to remember is that similar government policies in the past did not apparently produce the same outcomes under public management. 59 IV. Conclusion TANESCO’s first-ever private operator is responsible for significant gains to TANESCO and to the sector. Yet there remain critical issues: TANESCO can not finance its investment needs of US$ 614 million (and this amount even excludes additional generation and rural electrification). It can not even finance its own operations. The enterprise is dependent on the international donors and lenders for its investments and on the GOT to help pay the capacity and energy charges owed to IPPs. More importantly, TANESCO remains technically insolvent with its accumulated government debt of US$ 800 million. Urgent action is required to restructure TANESCO’s debt. Furthermore, a more aggressive GOT intervention might even buy out the IPTL plant and relieve TANESCO of its crippling capacity charges. Once TANESCO has completed ring fencing, it would need clear policy guidance on the next reform steps, including privatization (if that is what the policy makers agree upon). The Way Forward The GOT would be well advised to maintain the private management team for a total of 7 years (that is, three years beyond the existing second contract). However, the GOT needs to start planning now before the second contract ends at the end of 2006. It can either decide to extend it for another three years or, if not, plan what to do next. The time gained by extending the management contract could well be used to develop a sound and coherent sector policy including a realistic investment plan that is broadly owned. Since the private management agreed in its second contract that it “will accept notification of willingness to extend the contract before July 1, 2006.� 48 This means that PSRC, TANESCO Board and the Ministry of Energy and Minerals have literally one single year to decide. As we have seen in all our cases, a well-done privatization takes a long time and requires good planning by all parties involved. Policies implemented in response to crisis generally produce inferior outcomes. 48 Management Monitoring Report, Negotiations on New Management Supply Services Contract, TANESCO, June 9, 2005, p. 6 60 APPENDIX A. SUMMARY INFORMAITON ON TANESCO’S PROFIT AND LOSS AND BALANCE SHEET (2001-2004) December December December December YEAR 31 2004 31 2003 31 2002 31 2001 Profit/Loss Statement Sales S 188,475 165014 150015 139209 Cost of sales OC -266240 -189968 -138739 -106480 Gross Profit OI -77765 -24954 11276 32729 Other operating income 138772 37957 24975 8859 Operating expenses -45793 -78268 -37205 -62081 Net finance expense -77969 -157089 -120747 -59989 Net profit (loss) before taxation NI -62755 -222354 -121701 -80482 Taxation NT 53152 45556 32857 88340 Net profit (loss) after tax NAT -603 -176798 -88844 7858 Balance Sheet Fixed assets PPE 1198521 1223966 1227753 1241460 Current assets CA 149808 138653 154970 130689 Current liabilities CL 556444 504625 360068 280361 Deferred taxation 3527 4076 429 1475 Net Assets A 453388 449718 575786 631525 Long Term Debt LTD 331799 373106 370893 332826 Equity E 453388 449718 575786 731525 Stated capital I 293912 293912 293912 293912 Ratios PROFITABILITY Operating Income/Sales (OI/S) OI/S -41.3% -15.1% 7.5% 23.5% Operating Income/Property, OI/PPE -6.5% -2.0% 0.9% 2.6% Plant and Equipment (OI/PPE) Net Income/Sales (NI/S) NI/S -33.3% -134.7% -81.1% -57.8% ROA NI/A -13.8% -49.4% -21.1% -12.7% ROE NI/E -13.8% -49.4% -21.1% -11.0% 61 Case Study 3 The Partial Divestiture of the Tanzania Telecommunications Company Ltd. I. Partial Privatization In 1997 the Government of Tanzania (GOT) decided to allow private companies to invest and help operate the country’s major utility and transport sectors, specifically its power, water, port, railways and telecommunication enterprises. Only with respect to telecommunications (and the airline) did this policy call for partial privatization (via divestiture) of what had previously been a full-owned state entity; in other sectors the tactics chosen were concessions, leases and management contracts. Legal and Regulatory Reforms (1993-1997) The Tanzania Telecommunications Company Limited (TTCL) had been subjected to a series of legal and regulatory reforms beginning in 1993. The Tanzania Communications Act of 1993 enabled the state-owned single entity Tanzania Post and Telecommunications Corporation (TPTC) to separate its telecommunications and postal services as well as its savings bank. This led to the formation in 1994 of the Tanzania Telecommunications Company Limited (TTCL), Tanzanian Postal Bank (TPB) and the Tanzania Postal Corporation (TPC). The Act also provided for the establishment of the Tanzania Communications Commission (TCC) as an independent regulator for telecommunications (except for broadcasting and television) and postal services. TTC was entrusted with licensing telecommunications network operators as well as the licensing of postal services. Donors were eager to support the GOT’s attempt to restructure and ready TTCL for its eventual divestiture; a number of them agreed to provide a total of US$ 250 million in loans and grants to the state-owned utility.49 Between 1994 and 1998 TTCL embarked on its ambitious Telecommunications Restructuring Program (TRP) with the objective of raising Tanzania’s low telephone density from 0.3 to 0.7 per 100 inhabitants by the end of 1998. This would be accomplished by rehabilitating the existing network to meet the fast-rising demand for reliable services in urban and rural areas. Next in 1997 the GOT developed and published its National Telecommunications Policy (NTP) calling for further expansion of telecommunications networks and services through private sector participation. The NTP liberalized mobile and value added services except for fixed line services. The GOT granted an exclusive right to the 49 The donors included the World Bank, European Union (EU), Swedish International Development Agency (SIDA), African Development Bank (ADB), Danish International Development Agency (DANIDA), Japanese Development Agency (JICA) and Kuwait Fund for Arab Economic Development (KFAED). 62 incumbent operator, TTCL, to operate the international gateway, leased and fixed lines in mainland Tanzania. In Zanzibar, however, TTCL and ZANTEL were granted a duopoly of operations. TTCL at the time also held 29% of Mobitel’s shares, the first mobile license granted to a private operator (in 1994). Pre-Privatization Performance (1997) Despite the ambitious restructuring program, TTCL has not greatly improved performance by 1997. Challenges ranged from unsatisfied demand and low productivity (despite reduction in labor force from 4671 to 3650), to high international and even domestic tariffs, poor management and weak operational systems, inexperience of marketing in a competitive and liberalized environment, and lack of clear regulatory regime for commercial operations. Its financial challenges were equally dramatic. It had a large debt burden and negative future cash flow. Even though the exchange capacity has grown annually at almost 10% during the restructuring program, connections had not kept up. TTCL’s quality of service indicators remained poor with low call completion and call drop rates, high faults per line and with low level of digitalization. In 1996, two years after launching the restructuring effort, Tanzania still had the lowest percentage of digital lines in all Sub-Saharan Africa. Furthermore, much of the network infrastructure required replacement and the maintenance burden was high. This implied that TTCL had not invested the US$ 250 million of loans and grants effectively. It was short of achieving the telephone density level of 0.7 per 100 inhabitants as planned. Tanzania still maintained one of the lowest telephone density levels with less than 0.4 in the region compared to South Africa (11.35), Botswana (9.27), Kenya (1) and Zambia (0.8) per 100 inhabitants.50 In 1997, the GOT (and the World Bank) was ready to develop a divestiture strategy and launch the necessary process to divest its shares in TTCL. 51 The Transaction (1997-2001) Strategy The Presidential Parastatal Sector Reform Commission (PSRC) was entrusted with developing TTCL’s divestiture strategy. In 1997, PSRC engaged the services of a respected international financial advisor, NM Rothschild & Son, to review TTCL’s as well as the sector’s overall performance and recommend the most appropriate divestiture strategy. NM Rotschild & Son was retained on a success fee basis. Rotschild reviewed TTCL’s audited financial accounts of 1996 and 1997, and submitted its “Privatization Strategy Report� in 1998. It roughly calculated the value of TTCL’s enterprise value between US$ 250 and 300 million and estimated the enterprise’s debt to 50 “A Performance Assessment and Privatization Impact Study,� prepared by the President’s Office Planning and Privatization, United Republic of Tanzania, March 2004, p. 104 51 For the period prior to 1997, financial data is not available but only operational performance data. 63 the GOT at US$ 241 million. Since this debt level was not sustainable, it recommended financial restructuring by way of converting a major portion of the debt into equity.52 The GOT took over TTCL’s loans worth US$ 183 million, and waived all outstanding liability of income and sales tax for the period up to the end of 1998. The decision to sell all TTCL’s non-core assets also improved TTCL’s balances prior to privatization. While by 2000 TTCL was not debt free (as some argued),53 its debt had been reduced to a manageable US$ 15.8 million. However, major issues still needed to be addressed including whether or not to grant a period of exclusivity to the privatized TTCL, and whether to transfer either a minority or majority stake to a private strategic investor. Exclusivity or Non-Exclusivity The GOT was keen on privatizing TTCL on an exclusivity basis, fearing they would not attract a decent investor without such an offer. The financial advisor did not advocate bringing in a Second National Operator (SNO) either. This decision was prompted by the developments in the region. Uganda and Kenya were already having a hard time attracting strategic partners into their incumbent telephone companies. Uganda had opted to bring in a SNO even prior to privatizing Uganda Telecommunications Ltd. (UTL). It succeeded only after two failed attempts. Similarly, Kenya struggled to find a credible investor for its fixed line operations and had gone through a few failed attempts. Rotschild proposed, however, “limiting TTCL’s exclusivity to urban areas only (with a population of 50,000 and above) so that it can overcome the problem of under-provision of telecom services in rural areas.� Nevertheless, the strategy report did not articulate how to separate urban and rural provision of telecom services.54 It has also been reported that the financial advisor was totally opposed to competition in fixed lines and dismissed ZANTEL’s (the fixed line service provider in Zanzibar) willingness to compete in mainland if non-exclusivity was a possibility.55 Once the decision was made to grant an exclusivity period to the privatized monopoly, the next debate shifted on determining the number of years for the exclusivity period. Rotschild recommended a period of at least seven years with a possible extension. But the GOT did not agree and was willing to grant it only for four years. Rotschild (as well as the World Bank) recommended that TTCL sells its 29% stake in Mobitel prior to its divestiture. The intention was to create a level playing field for the private investor if and when TTCL were eventually granted both fixed and cellular licenses. Accordingly, the GOT agreed and TTCL sold its shares in Mobitel. Divesting Minority or Majority Stake To increase the chances of attracting a credible strategic investor, Rotschild recommended that GOT sell a clear majority of TTCL shares. Nonetheless, the GOT felt 52 “Privatization Strategy Report,� prepared by NM Rotschild & Son, London, UK, 1998 53 A Performance Assessment and Privatization Impact Study,� prepared by the President’s Office Planning and Privatization, March 2004, p. 113 54 Memorandum, sent to PSRC by N.M. Rothschild & Son on December 16, 1998, p.2 55 Interviews conducted by the World Bank telecommunications specialists, July 7, 2005 64 comfortable selling only a minority share and decided to divest, at least initially, 35%. The financial advisor accordingly proposed a two-phased divestiture strategy. Stage I—Initial Sale of Shares to a Strategic Investor The strategic partner would be sold 35% of the shares (payments to go to the Treasury Registrar as TTCL’s owner at the Ministry of Finance), but the sum would subsequently be transferred into TTCL as an equity injection. The remaining 65% of the shares were to be retained by the GOT. Stage II—Planned Transfer of Remaining Shares The GOT would decide at a later, unspecified date to transfer the remaining shares as follows:  Up to maximum of 10% to Tanzanian financial investors  Up to maximum of 14% to overseas financial investors  Up to maximum of 5% to TTCL staff  36% (more of the remaining shares) to general public via IPO The Process It took almost four years—from 1997 to 2001—to complete the first phase of the transaction and divest 35% of TTCL’s shares. The second phase was never implemented. To handle the first phase, and following the World Bank procurement rules, a two-stage competitive bidding process was launched. Competitive Tender PSRC sent out at the end of June 1999 letters of invitation to over 120 organizations to participate in the privatization of TTCL through an issue of new shares representing 35% of enlarged share capital. Twenty-nine interested parties requested a copy of the Information Memorandum. Nine of the 29 submitted pre-qualification documents. Of these, six were declared “pre-qualified.� These included the consortium of Mobile Services International Cellular Investment Holdings B.V. (MSICI) of Netherlands and Detecon Gmbh of Germany; Vodacom of South Africa, Sasktel of Canada; MTN of South Africa, Mauritius Telecom; and an Indian consortium of Mahanagar Telephone Nigam Ltd. (MTLN), Telecommunications Consultants India Ltd. (TCIL) and Videsh Sanchar Nigam Ltd. (VSNL). Subsequently, only 3 agreed to submit final proposals.  Consortium of Mobile Services International Cellular Investment Holdings B.V. (MSICI) of Netherlands and Detecon Gmbh of Germany;  Vodacom of South Africa;  Sasktel of Canada. 65 The Winning Bid The three bids are compared below in terms of purchase price offered for 35% of the shares and the connection commitments to expand TTCL’s existing network. Table 1. Final Three Bids for 35% of TTCL’s Shares, June 2000 Bidder Amount Bid Connection Commitments MSICI/Detecon US$ 120 million 800,100 (450,000 fixed) Vodacom US$ 66 million 850,000 Sasktel US$ 60.5 million 575,000 Source: Compiled from various documents provided in the data room at the PSRC, Tanzania. PSRC declared the Consortium of MSICI and Detecon the winner, judging that the consortium was willing to offer what they specified in the proposal (even after some naturally expected negotiations). The proposal looked and sounded promising with: …..an exceptional roll-out plan with total connected line reaching 800,100 (450, 000 in fixed lines, 131, 250 in mobile lines and the remaining 218,850 in alternative technologies to be determined) by the end of December, 2003; an investment infrastructure of over US$ 900 million in 10 years, with US$ 543 million invested in first few years; over 4,200 new jobs immediately and rising to 12,500 employees over 10 years; a pay-out to GOT including taxes and dividends which will be in excess of US$ 1.5 million over 10 years; an increase in telephone density from 0.66 by the end of 1999 to 3.49 per 100 inhabitants by the end of 2003; over 50% of the budget of approximately US$ 27 million to be spent on training in the first 4 years.56 The winning proposal also stated its plans for 30% of the roll-out in the rural areas arguing that “every community should have some kind of communication center (tele- boutique, tele-centers or call centers with internet access, printing, copying, faxing in addition to local, national and international calling.� Altering Payment Terms of Purchase Price Immediately after it was declared the winning bidder, the consortium of MSICI and Detecon entered into negotiations with the GOT in an attempt to change the terms of the bid. It declared that it was not willing to sign the Subscription and Shareholder’s Agreement for the full bid price of US$ 120 million without receiving the audited accounts for 1999 and 2000. The Information Memorandum sent out to all the potential bidders in June 1999 contained TTCL’s audited accounts only through 1998. For reasons that are not fully clear to the consultants, but probably because it feared it might lose what seemed like an attractive offer, the GOT agreed to vary the payment terms. The consortium proposed an initial payment of US$ 60 million (as capital injection to TTCL), with a second payment of an additional US$ 60 million after verification of the accounts for the year 1999 and 2000——provided that the latter showed an improvement 56 Bidding Document, prepared by MSICI and Detecon Consortium, May 5, 2005, p. 8 66 over 1999 accounts. If audited 2000 accounts showed a position worse than the accounts of 1999, the second payment would be reduced in accordance with an agreed formula. The Shareholder’s Agreement had specified this formula but the initial and the projected second payment were only handwritten and signed by one of the consortium members on behalf of the MSICI of Tanzania.57 The second payment became dependent on the performance of TTCL as reported in the accounts for the year 2000. When the 2000 accounts were audited and officially published, a dispute arose. The investor interpreted the accounts as showing a worse financial position compared to the accounts of 1999, and said this justified a reduction in the second payment. The GOT, on the other hand, argued that the 2000 accounts showed that the private investor had to make its second capital injection of US$ 60 million to TTCL. But the consortium insisted to subtract US$ 55 million from the promised second payment of US$ 60 million and commit to a final payment of only US$ 5 million instead. A long and bitter dispute ensued. The expert opinion of an outside consultant, KPMG of London, was solicited. It was eventually determined that the minority partner was correct and it was obliged to pay only US$ 4.96 million with an accrued interest of US$ 321,000, a total of US$ 5.281 million. This dispute over the second payment took over three years to resolve (in January 2004). Full Management Control with a Management Contract It was understood at the outset that the strategic partner would insist on, and should be given, full management control. The GOT’s dual privatization objective for TTCL was to ensure an equity injection for the development of the fixed line communications network and infrastructure and, at the same time, to obtain management expertise— which TTCL desperately needed. On February 23, 2001 the minority partner signed the Subscription and Shareholding Agreement and took over TTCL with full management control. In the meantime, the private partner managed to secure an additional management contract from the Treasury Registrar (as the owner of TTCL’s majority shares) at the Ministry of Finance, apparently without the knowledge of PSRC. This “Management Services Agreement� specified the obligations of the private investor, MSICI of Tanzania, and TTCL to one another58:  MSICI’s obligations to TTCL included providing management and financial services (arrangement of capital financing and provision, expertise of technical services, and standards of care and performance);  TTCL’s obligations to MSICI included compliance for monthly reporting, payment for expenses to all assistance from MSICI, payment for software and hardware; and compensation for MSICI’s services in the form of a “management assistance fee� calculated as 3.5% of monthly gross turnover of TTCL;  TTCL agreed to cover all reasonable expenses incurred by MSICI of Tanzania and its employees in the fulfillment of its obligations; 57 Signed copy of the Subscription and Shareholder’s Agreement, February 23, 2001, p. 27 58 A copy of the “Management Services Agreement� is included in TTCL/Celtel Performance Capital Investment Structure and Funding (date not specified) in the data room at PSRC. 67  This management service agreement was to be renewed automatically after 4 years unless the parties agreed in writing to terminate the agreement three months prior to the end of the current four year period. The private investor also signed the License Contract acquiring five licenses (an exclusive license for fixed lines and non-exclusive for cellular, radio paging, internet and data services). It paid a combined license fee of US$ 1.5 million to Tanzanian Communications Commission (TCC). Immediately after taking over TTCL, the private investor established Celtel of Tanzania as the fully-owned subsidiary of TTCL on May 7th, 2001. License Contract A first draft license contract was sent to TCC for comments in 1998 even prior to sending out the Information Memorandum to potential investors in 1999. It took nearly two years for TCC to review the draft license contract, to comment and propose any changes. In the consultants’ view, insufficient consideration was given as to what extent the terms and conditions in the license contract was advantageous or disadvantageous to TCC (and ultimately to the GOT). All bidders were offered at the end a package of five licenses: fixed lines on exclusive basis for 4 years plus mobile, radio paging, data and internet licenses on non-exclusive bases. TTCL’s license contract for the package covered a period of 25 years and was subject to renewal for every five years. The final license stated that the licensee must ensure a total of 450,000 Fixed Direct Exchange Lines (Fixed DELs)--whether by means of fixed or wireless local loop lines (including existing DELs)--within four full years of the effective date. It clearly specified that the total new connected fixed DELs would reach 294,800 at the end of the fourth year of the exclusivity period. In addition to system expansion through a roll-out plan for fixed lines, the licensee was expected to provide public payphones, international services; meet customer obligations and quality of service requirements, and to interconnect operators, and not to practice unfair cross-subsidies and guarantee training to local personnel. II. Results and Outcomes (2001-2004) Operational Performance In the first three years after privatization, TTCL’s operational performance in fixed lines did not improve. The new private management did not expand the existing fiber optic backbone infrastructure.59 In contrast to the steady increase in fixed network exchange capacity (or the installed capacity) during the restructuring period prior to privatization, the new connections remained far below what the private investor promised in its bid and agreed in the license contract. Since an almost equal number (or higher) of fixed direct 59 TTCL has not calculated its bill collection efficiency comparing the billed DELs to the paid for DELs. Such statistics have not been analyzed and reported. 68 exchange lines (DELs) were disconnected for non-payment right after privatization, the net increase in fixed DELs declined by the end of 2001 and increased only gradually later. Figure 1. Trend in Exchange Capacity and Fixed Line Connections 300000 250000 200000 150000 100000 50000 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Network Exchange Capacity Connected DELs Billed DELs Source: Calculated from data provided by the legal department at TTCL, June 2005 TTCL has recently estimated that Tanzania’s demand for fixed telephone services is 747, 881 DELs. The existing network capacity of 243,734 lines, therefore, is a third of this demand, and it is even lower for connected DELs.60 This suggests that TTCL did not exploit the market for fixed lines. TTCL invested relatively little in fixed lines. Its subsidiary, Celtel Tanzania, on the other hand, enjoyed generous capital investments, and began to compete successfully with other mobile operators. Several TTCL employees argue (as Table 2 shows) that the bulk of US$ 60 (and later additional US$ 5.281 million) which the GOT received as privatization proceeds in 2001 was injected into Celtel Tanzania, and hardly any went to TTCL’s fixed line operations during the first three years. The low level of investments in fixed lines is in direct contrast to US$ 185 million TTCL invested before privatization during the restructuring period. This amounts to 75% of the total funding of US$ 250 million received from the international community. In addition, TTCL also spent an additional 12.5% or US$ 23.1 million in capacity building (training and consultants).61 After privatization, TTCL’s fixed line operations received a total US$ 60 Group Financial Statements for the year ended 31 December 2004, Tanzania Telecommunications Company Ltd., p. 5 61 Data provide by the legal department of TTCL, June 2005. 69 40.25 million for capital investments between 2000 and 2005, 84% of which only in 2004 and in early 2005. It appears that the new management did not spend much on training either (again in contrast to its stated promise in the bid). Table 2. Capital Investments into Fixed and Mobile Services (in US$ millions) Investments 2001 2002 2003 2004 2005 Total (May) TTCL/Fixed 2.28 (7%) 2.56 (28%) 1.70 (7%) 25.96(75%) 7.75 40.25 (36%) Celtel/Mobile 32.70 (94%) 6.70 (82%) 22.30 (93%) 8.90 (25%) N.A. 70.60 (64%) TOTAL 34.98(100%) 9.26(100%) 24.00(100%) 34.86(100%) N.A. 110.85(100%) Source: Compiled from TTCL/Fixed data provided by the legal department of TTCL, June 2005 and Celtel/Mobile data from the report, “TTCL/Celtel Performance Capital Investment Structure and Funding,� in the data room at PSRC, Tanzania. Financial Performance TTCL’s financial performance did not improve after privatization. Its turnover remained more or less the same in nominal terms, approximately US$ 100 million, and declined in real terms between 2001 and 2004. Its operating expenses also remained high resulting in negative net profit after taxes with the exception of the year 2002. Annex A. gives a summary of TTCL’s Profit and Loss Statements and Balance Sheet between 1997 and 2004. Under new private management, TTCL remained a loss making enterprise, while its fully owned subsidiary, Celtel’s operations, proved profitable. Celtel’s aggressive marketing strategy and network expansion led to significant increase in customers and regional presence covering 26 regions. During the last four years its total number of subscribers reached 740,000. Celtel’s annual revenues are now about US$ 75 million. III. Impact on Stakeholders Consumers Consumers expect to benefit from privatization via increased access and coverage (resulting from additional investments), competitive and affordable tariffs, and improved quality and reliability of services. As he announced TTCL’s winning bid at the deal closing ceremony right after private investors took over, Professor Mark Mwandosya, the Minister of Communications and Transport, emphasized too that the ultimate beneficiary would be the consumer of telecommunications services in Tanzania and the economy as a whole.62 In reality as a result of this transaction, Tanzanian consumers lost from limited 62 Professor Mark Mwandosy’s speech at TTCL’s Deal Closing Ceremony, immediately after privatization in February 23, 2001, p.6 70 expansion in fixed lines but benefited from competition and increased access through Celtel’s entry into the mobile market. The slow expansion in fixed lines did not increase access and coverage in this segment of the market and left many potential fixed line customers un-served during the exclusivity period. By May, 2005, a four full years after privatization, the total number of connected fixed lines is only 152,776. TTCL’s new investors and operators failed to meet the system expansion requirements under the License Contract, as shown in Table 3. Table 3. Shortfall between Targeted and Actual New Fixed DELs 1st Year 2nd Year 3rd Year 4th Year Total Target 41,000 60,500 78,900 114,400 294,800 Actual 22,906 34,907 22,511 21,670 101,994 Shortfall 18,094 25,593 56,389 92,730 192,806 % Un-Met 44% 42% 71% 81% 65% Source: Calculated from data provided by TTCL’s Legal Department, June 2005 Moreover, the bulk of the new fixed line connections was limited to urban areas and the new operators did not make much of an attempt to increase coverage in rural areas (as they proposed in their bid). At the same time, some customers lost their connections because of non-payment as the new management tried to improve bill collections rates. The cessations far exceeded the new connections in fixed DELs thus reducing further the number of net new connections at the end. It appears, however, that TTCL decided to introduce a pre-paid service. Pre-paid amount and pre-paid calling cards countrywide will target the middle and low end user customer base, residential and small and medium enterprises. Poor or rural consumers have not so far benefited much from privatization. While the number of public pay phones almost doubled from 2400 in 2001 to 4438 in 2004, most of these phones were installed in urban centers in Dar es Salaam; hardly any went to rural or semi-urban areas. Contrary to what was proposed in the bid and entered in the license contract, the private investor did not install “at least two public pay phones for every 3,000 inhabitants in any centre of population of more than 3,000 inhabitants.� Still, Tanzanian customers as a whole benefited from increased competition through Celtel’s entry into the mobile market. By the end of 2004, Celtel has captured 22 % of 1.7 million total telephone subscribers (mobile and fixed combined) following Vodacom with 49% and Mobitel 17%. Yet, TTCL’s fixed line subscribers’ share constitutes only 8.5% (while Zantel is another 4.3%) of the total market. It is constantly argued, however, that all fixed line operators are presently facing a tough market in Africa while demand for mobile lines far exceed the demand for fixed lines. TTCL’s new operators encountered the same changing context when they took over. Nonetheless, as mentioned earlier, the new management apparently ignored a substantial market demand for fixed lines. 71 The healthy competition among three major mobile telephone providers, Celtel, Vodacom and Mobitel, in the mainland and Zantel in Zanzibar, led to lowered tariffs for both mobile and fixed telephone services. The telephone tariffs became more affordable for 1.7 million Tanzanian telephone subscribers. Fixed line tariffs, on the other hand, for local, national and international calls remain higher than what the mobile operators offer and also in comparison to rates offered in neighboring countries.63 Recently, TTCL has started to provide broadband service such as leased lines and ISDN as well as intelligent PABX. However, some TTCL staff maintains that it is now too late, and that the firm missed a major market opportunity between 2001 and 2003. During this time, businesses and residential customers chose connections to the internet via satellite hook-up and would now be reluctant to switch to fixed line internet services, even if TTCL should become eager to serve this market. On the other hand, the improvements in quality and reliability of service in fixed lines brought significant gains to TTCL’s fixed line customers. The call completion rates for international calls increased from 37% in 2001 to 90% in 2004, while call completion rates for long distance calls jumped from 41% in 2001 to 97%. The percentage of faults cleared within 24 hours remained more or less the same after privatization having already reached 98% by 2001. Digitalization levels of the existing exchange capacity increased from 93% in 2000 to 99% in 2005. However, only the already connected fixed line customers enjoyed these quality service improvements. Employees Many of TTCL’s employees have borne the costs of privatization. A total of 1021 employees representing 22% of the work force lost their jobs prior to privatization, during the restructuring period. At the time of privatization, the private partner was not asked to, and did not promise to maintain TTCL’s labor force. On the contrary, it had even stated in its winning proposal its intention to increase employment significantly. But as the market shifted and the technological requirements changed, the private management initiated a retrenchment program to lay off 2400 employees in an attempt to reduce the number of staff from 3445 to 1045. Table 4. Trend in Employment and Labor Productivity 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Total # 4,671 4,775 4,798 4,719 4,575 3,650 3,650 3,571 3,517 3,448 3,445 Staff/1,000 53 52 51 45 39 25 24 21 25 24 23 lines Source: Data provided by the legal department at TTCL., June 2005 63 See “Telephone Tariffs� in US$ and exclusive of VAT which is 20% as of May 31, 2005, on Tanzania Communications Regulatory Agency (TFA) website, www.tcra.go.tz. 72 When the private management first announced its intention to retrench the workforce in June 2003, the Communications and Transport Workers’ Union of Tanzania (COTWU) and TTCL management negotiated a Voluntary Retrenchment Agreement. Both parties acknowledged the necessity to reduce substantially the number of employees. The Retrenchment Agreement was approved by the Ministry of Finance (which raised concerns on the high level of payments to be made) and by the Ministry of Labor. But many of the employees were unwilling to accept a golden handshake. They insisted that they were still expecting to receive 5% of the shares which the GOT set aside to divest in the second phase of divestiture. They maintained that the management was trying to reduce the workforce because they had not invested or productively utilized the existing labor force. They also argued that they were never consulted neither during or prior to privatization. They took their case to the Industrial Court of Tanzania. In March 2004 this court issued an injunction restraining TTCL management from implementing the retrenchment program. The union representatives now claim that the Voluntary Agreement was procured fraudulently and should thus be nullified, and that payment of 5% to TTCL workers should be affected before the implementation of retrenchment exercise. Management claims that it is ready to offer an improved retrenchment package, but talks with the union representatives have not been resumed, at the time of writing this report. Delayed retrenchment imposes additional operating cost in excess of TShs. 680 million per month for 2400 presumably excess staff. This translates into additional annual staff cost of TShs 8.2 billion or US$ 8 million. More importantly, TTCL soon might not even have the financial resources to pay for the agreed retrenchment package.64 While TTCL’s employees working on fixed line operations were facing retrenchment, Celtel’s workforce almost doubled from 127 in 2001 (the year it started operations) to 242 in early 2005.65 Government For GOT the impact of privatizing TTCL seems mixed. While it can benefit from transferring its shares in profitable Celtel to the private investor, it has suffered some losses from the TTCL’s partial divestiture. The Treasury did not receive the privatization proceeds of US$ 65 million instead they were injected into TTCL. More precisely, the bulk of it went into investments in Celtel. Contrary to the private investor’s rosy forecasts in its bid, the Treasury has not been collecting any dividends as the majority shareholder of TTCL. And total taxes paid by TTCL (fixed and mobile combined) to the Treasury remained more or less the same after privatization. However, the composition of taxes changed: Corporate taxes declined because TTCL Group has experienced low profitability; and it remained below pre- privatization levels. Employment taxes increased mostly as a result of salary consolidations. Prior to privatization, there were significant allowances paid outside payroll which were not taxed. 64 Gilder Kibola, “Brief Report on TTCL Retrenchment Programme,� Company Secretary and Head, Legal and Regulatory Affairs, TTCL, June 2005 65 Data provided by the sales and distribution department at TTCL, June 2005 73 After privatization, TTCL benefited from tax exemptions granted by the Tanzanian Investment Centre (TIC). The new investor obtained a Certificate of Incentive (# 110023) on April 11, 2001 entitling Celtel to tax exemptions amounting to US$ 4.7 million for the last four years on procurement of equipment for its major investments (import duties, excise taxes and deferment of VAT on imports until after commissioning of a project).66 At the same time, TTCL as a majority state-owned utility continued to benefit from its former tax exempt status with another US$ 3.1 million between 2001 and 2004.67 A total of US$ 7.8 million tax exemptions constitute a form of indirect subsidy or the revenues forgone for the Treasury. There were also other on-going disputes between Tanzania Revenue Authority (TRA) and TTCL. TRA issued tax assessments on the 1998, 1999, 2000, 2001 and 2002 years of income amounting to TShs. 28.5 million as a result of reclassification of previously allowable expenses against tax on pension contributions, capital allowances available on machinery and bad debt. TTCL had strongly objected to the assessments made and appealed to the National Tax Appeals Board.68 In addition, the GOT or the TCC (now TCRA since 2004) did not receive the fines it had levied on TTCL for non-compliance with its roll-out obligations specified in the license contract. According to the TCRA, these accumulated fines amount to US$ 44 million to date.69 This is in addition to accumulated US$ 40 million owed by Celtel Tanzania to TTCL as interconnection charges. Furthermore, the GOT as TTCL’s owner have been paying 3.5% of TTCL’s gross revenue to new management in compliance with the management assistance agreement (as discussed under section 1.3.2. Process: Full Management Control with a Management Contract). This fee has increased gradually from US$ 3million in 2001 to US$ 6 million in 2004. Nonetheless, the GOT is about to benefit as it prepares to sell 25%, or almost half of its 65% stake, in a very profitable mobile company, Celtel, to the private investor. This will bring considerable revenue to the Treasury if it decides not to inject it into TTCL. The GOT might also decide buy out the shares of its minority partner in TTCL. Nonetheless, if it negotiates a good price for TTCL’s shares (as a loss-making enterprise), it can benefit from Celtel’s high priced shares. Competitors and Overall Economy Among the mobile operators, Celtel is perceived as unfairly reaping the benefits of not paying the interconnection fees it legally owes to TTCL. This has led to several interconnection disputes brought by other mobile telephone providers, who complain about the lack of a level playing field. 66 Information provided by the legal department of TTCL, June 2005 67 Tax data provided by the Tanzania Revenue Authority (TRA). 68 Group Financial Statements for the year ended December 31, 2004, Tanzania Telecommunications Company Ltd., p.2 69 Interview with Professor John Nkoma, Director General, Tanzania Communications Regulatory Authority (TCRA), June 20, 2005 74 The recently established Tanzania Communications Regulatory Agency (TCRA) has tried to reduce the interconnection fees by introducing cost-based interconnection rates starting October 2005.70 When Minister Mark Mwandosya announced the GOT’s decision on June 30, 2005 to separate Celtel from TTCL (by selling 25% of its stake and thus reducing its shares from 65% to 40% in Celtel), he noted that “Celtel Tanzania used TTCL facilities such as booster stations to which other mobile phone companies had no access.�71 Despite the significant increase in mobile telephones, Tanzania still ranks as the lowest telephone density in Africa combined with poor and inadequate backbone infrastructure. Had TTCL’s new management invested strongly in fixed lines and exploited the market demand in fixed and value added services, the situation might today be very different. Low internet use now has to be tackled; Tanzania needs to catch up with information technology and build a knowledge-based economy and generate economic growth and jobs. Broad band internet connection via fixed lines is generally regarded as the most cost effective way to launch this process. Even though wireless technology can bring similar services to customers, it is likely to take time since it has just begun to emerge only in industrialized countries. IV. Explanations and Lessons Learned There were external and internal factors that might explain the above results and outcomes. An attempt is made below to summarize external factors that were outside the control of the policy-makers, while the internal factors emphasize what the policy-makers could have done differently in order to achieve better results and outcomes. External Factors Financial and Legal Advisors PSRC, correctly, selected its financial and legal advisors through competitive tendering. The financial advisor, N.M. Rothschild & Son, and the legal advisor, Clifford Chance, generally prefer to take such assignments on a success fee basis, which amounts to 2-3% of the sale price of the shares divested. As they act as brokers between the government and the potential investors, they are eager to sell the shares slated for divestiture at an acceptable price to the government and to the investor, in a reasonable period of time. As they receive their success fee on completion of the transaction, they are motivated to get the deal done. We have not seen the agreed and signed contracts between the advisors and PSRC, but reportedly, the advisors received their success fee on the winning bid price, namely US$ 70 Interconnection Determination No. 1 of 2004, prepared by Tanzania Communications Regulatory Authority, p. 10. 71 From The Guardian, July 1, 2005 75 120 million which the private investor offered. As noted, at the end of the day the GOT actually received only US$ 65 million (more accurately US$ 65,281,000). Presumably, the advisors were able to argue that their contracts calculated the success fee on the basis of the amount specified in the winning bid, and if the GOT accepted, post-bid, to revise the price, that was not their concern. The PSRC might in the future draft and sign tighter contracts, tying the fees to the amount actually received by the GOT. World Bank’s Procurement Procedures The GOT was confined to operate within World Bank’s procurement procedures. The procedures allow only the winning bidder to negotiate with the GOT. In the TTCL case the winner, MSICI/Detecon Consortium, managed to negotiate a two-stage payment after it was declared the winner. The problem is not necessarily with the phased payment system; this might be a good way to reduce the uncertainties inherent in weak data and poor accounts. The problem is one of transparency. Apparently, the opportunity to suggest a phased payment system was not given to other bidders. The World Bank, and those using its procedures, might consider the pros and cons of a two-phase payment system so that other bidders can have the same chance to make similar offers and negotiate. Internal Factors Unreliable Financial and Operational Data The bidders were constrained by unreliable financial and operational data while preparing their bid and conducting their due diligence. Without judging the actions of any one in the TTCL process, one can say that poor quality data provides an opportunity for unscrupulous bidders to offer inflated prices, knowing that it is likely that if they win, they will uncover information at a later date which, they can claim, negates the assumptions of their earlier offer. In the TTCL case, it is not surprising that the winning bidder chose to alter the payment terms after having won the bid on the basis of a high price. Even PSRC seemed to sympathize with private investor’s desire and willingness to re-negotiate the terms and conditions of its original bid under the circumstances. Interviewees at PSRC acknowledged the “inflated financial data� that TTCL provided to Price Waterhouse & Coopers in its 1999 financial reports. They pointed out that the GOT presumed to sell TTCL “without knowing what it was actually selling because there were problems with the accounts and TTCL had not gone through adequate preparation for privatization.� Similar to financial data, both operational and bill collection data had their own deficiencies. The private investor found out about poor collection rates only after it took over management in 2001, not prior to bidding. It disconnected a number of fixed lines for non-payment even risking serious vandalism and theft as a result. 76 Private Investor’s Strength and Interest in Mobile The leader of the winning consortium, Mobile Services International Cellular Investments Holdings B.V. (MSICI) (and now Celtel International) had expertise only in mobile networks. Established first in 1989 in Netherlands it has become the leading specialist wireless telecom company focusing exclusively on owning and operating cellular telecom networks in Africa. At the time of privatization, it partnered with Detecon Gmbh, which did have experience in fixed line, and created a new Tanzanian company, called MSICI of Tanzania. MSICI had only once ventured into fixed line operations in Sudan. However, it did not have any prior knowledge of managing a state- owned enterprises or infrastructure utility. Within the consortium MSICI held 51% of the shares (and in the event no financial investor was found for the planned 30%, it had the possibility of eventually increasing its shares to 81%). In contrast, Detecon Gmbh, held only 19% of the shares in the consortium. Detecon was an engineering consulting firm founded in 1977. Deutsche Telecom AG owned it 49% and the remaining 51% of its shares was divided among Deutsche Bank (18.2%), Dresdner Bank (18.2%) and Bau und Handels Bank (14.6%). Detecon had expertise in fixed lines and the bidding document included a letter of recommendation from the Deutsche Telecom attesting to that. Unfavorable License Contract and Weak Regulation Annex A of the TTCL license specified the system expansion requirements and the rollout obligations on a yearly basis. The Licensee was required to present to the Regulatory Authority a rollout plan. At the end of the every year of the exclusivity period it had to demonstrate how the actual rollout has been conducted explaining any divergence from the original plan. The Annex presented a formula for assessing the “economically deficient quarters� during the year so that the Licensee that would reduce, if not eliminate, penalties in respect to failure to meet the required total DELs. 72 The Tanzanian Communications Commission (TCC) apparently objected to this protective measure, but did not request revisions. The TCC was at the time a new organization; it did not have sufficient capacity to deal with issues regarding the flaws in the License Contract prepared by Clifford Chance. Later, in 2003, the GOT passed a law creating the regulatory framework for the new Tanzanian Communications Regulatory Agency (TRA). The new agency has been in operation as a quasi-government agency only since March 2004. PSRC now regrets not selling the fixed and mobile licenses separately. However, it is conceivable that it would have been difficult to attract any investor into fixed lines if the GOT had not offered the mobile license in conjunction with exclusivity in fixed lines. Similarly, TTCL might not have attracted any credible investors at all in the event the GOT had brought competition into fixed lines by bringing in a Second National Operator 72 License Granted Under the Tanzania Communications Act 1993 by the Tanzania Communications Commission to Tanzania Telecommunications Company Ltd for the Installation, Operation, Maintenance and Exploitation of a Public Telecommunications System in the United Republic of Tanzania,� Execution Copy, Annex A., February 23, 2001 77 (SNO) or simply by not giving exclusivity to TTCL. In early 2000, international investor appetite in fixed line telecom markets had declined considerably compared to the period in mid-1990s. Minority Stake with Government Majority on Board The GOT could have chosen to sell not a minority but rather the majority stake in TTCL. Had it agreed to do so, it might have attracted more credible buyers with a serious interest in all aspects of the company’s business. A majority stake of equity would have meant a majority control on the Board as well. Some observers claim that the GOT’s majority representation on the TTCL Board has successfully blocked the minority investor’s willingness to invest and expand fixed lines. Despite full management control with a management contract, the private investor has often complained that the GOT Board Directors have managed to block major investment decisions during the last four years. The explanation for this blocking of investments is not clear. Perhaps self-interest is involved: Reportedly, the Board Directors representing the GOT have insisted on the use of certain equipment suppliers, and have blocked investment decisions unless they get their way. V. Conclusion The impact of TTCL’s partial privatization has led to mixed results and outcomes. The main short-term losers from privatization appear to be TTCL’s employees—depending on the outcome of the on-going dispute between labor and management. Consumers have lost from limited investments and lack of expansion and access in fixed lines. But they gained from Celtel’s entry into the mobile market, increased competition and ensuing lower tariffs. Only the existing and (relatively few) new fixed line customers benefited from the quality of service improvements in fixed lines. On the other hand, the GOT as the majority owner of TTCL has been left with a loss- making enterprise at the end of the four year exclusivity period in 23 February 2005. It has just sold 30% of its shares in Celtel, hopefully at a good price, and it will be able to receive dividends and increased taxes in the near future and TTCL will collect new (and accumulated) interconnection fees. The Way Forward The Tanzanian economy as a whole has lost from TTCL’s reluctance to invest in fixed line backbone infrastructure, not meeting its license obligations and not even meeting the existing demand for fixed lines. The un-met demand for fixed lines remains three times the number of connected DELs. Internet penetration is low. International rates are high. Local service is still of lower quality with higher prices than in many other countries in the region. All this is critically important at a time when Tanzania needs to participate in the knowledge economy and relies increasingly on information technology and internet services. Now it is GOT’s turn to negotiate a good price for private investor’s shares in 78 TTCL and either turn the enterprise around itself or seek another international investor with substantial majority stake. 79 APPENDIX A. Summary Information on TTCL’s Profit and Loss Statement And Balance Sheet (1997-2004) Dec 31 Dec 31 Dec 31 Dec 31 Dec 31 Dec 31 Dec 31 Dec 31 YEAR 2004 2003 2002 2001 2000 1999 1998 1997 Profit/Loss Statement Turnover S 106804 116477 106364 111017 86732 110218 86055 66006 Operating Expenses incl depre. OC -113341 -125818 -106265 -121432 -103763 -97289 -73937 -73800 Gross Profit OI -6537 -9341 99 -10415 -17031 12929 12118 -7794 Net profit before taxation NI -7166 -1686 2958 -11222 -17047 13270 12118 -8064 Taxation NT -1238 -55 734 -1241 -9751 4744 3717 0 Net profit after tax NAT -8404 -1741 2224 -9981 -7296 8526 8401 -8064 Balance Sheet Non current assets PPE 195930 189402 172821 168224 157108 142288 167216 121116 Current assets CA 105513 129406 115393 113673 73339 126857 64912 45826 Current liabilities CL 73729 86291 54181 50196 36779 63114 35281 29409 Non current liabilities LTD 11569 13474 13249 13141 13855 12736 1069 1027 Equity E 216145 219043 220784 218560 179813 193295 195778 136506 Stated capital I 245246 239740 239740 239740 191012 191012 Ratios PROFITABILITY Operating Income/Sales (OI/S) OI/S -6.1% -8.0% 0.1% -9.4% -19.6% 11.7% 14.1% -11.8% Operating Income/Property, OI/PPE -3.3% -4.9% 0.1% -6.2% -10.8% 9.1% 7.2% -6.4% Plant and Equipment (OI/PPE) Net Income/Sales (NI/S) NI/S -6.7% -1.4% 2.8% -10.1% -19.7% 12.0% 14.1% -12.2% 80 Case Study 4 The Partial Divestiture of Air Tanzania I. The Pre-Privatization Situation Formed in 1977, following the demise of the regional East African Airlines, Air Tanzania was a wholly state-owned and operated company, serving major cities inside the country, and a number of regional destinations (that varied over time): Nairobi, Mauritius, Comoros, Kigali, Bujumbura, Lubumbashi, Entebbe, Lusaka and Harare. Following the liberation of South Africa, Air Tanzania started a Dar-Johannesburg flight, which quickly came to account for a full quarter of its passenger revenues. In the mid-1990s, Air Tanzania owned two and leased one Boeing 737 aircraft, and leased one Dornier 228. It had a workforce of about 500, or 125 people per plane. According to a 2002 study by the consulting firm, Simat, Helliesen and Eichner (SH&E), the airline’s expenses were consistently higher than its cash revenues, with fuel, maintenance, leasing and administration costs all above industry averages. Through the 1980s and 1990s, Air Tanzania struggled with low load factors, increasing domestic competition, a money-losing and frequently changing regional route structure, an inappropriate and expensive fleet, no focus on the potential lucrative cargo business, overstaffing, poor cash and credit management, high operating expenses, and an increasing debt load. The consultants estimated that even if cut its labor force to 250, the company would be “overstaffed, relative to its fleet.� Air Tanzania’s accounts show that by 2000 the company was in dire financial condition. Lacking virtually any working capital, and technically insolvent, the company survived apparently because of subsidies and assistance by the GOT Treasury. The accounts for the years 1995-1996 through 1999-2000, carried out by the Government-operated Tanzania Audit Corporation, show net losses before taxation in only the last two years of the period. But these accounts show “non-operating revenue� averaging Tshs 1.8 billion per year for the five year period. It is likely these were direct subsidies, and it is likely that these Government injections of assistance had started well before the 1990s. Second, the accounts indicate that Air Tanzania paid no taxes.73 Most important, accumulated losses mounted dangerously in the period, reaching Tshs 11.2 billion in 1999-2000. And when the accounts for the following year were carried out by Price Waterhouse Coopers (in preparation for privatization), they calculated the accumulated losses at Tshs 19.8 billion—about $US 20 million. In short, before privatization, Air Tanzania was just about defunct. 73 The Tanzania Revenue Authority (TRA) confirms that in the period 1998-2000 Air Tanzania was granted tax exemptions in the amount of Tshs 34.4 million. Note that this sum is minuscule compared to the billions in exemptions granted to each of TANESCO, TTCL, DAWASA, TRC, TAZARA and THA. 81 II. The Transaction The Strategy Tiring of subsidizing a money-losing business, the Government of Tanzania (GOT) revised its air sector policy at the end of the 1990s. It decided “to disengage from operational activities and to allow private sector participation and market competition wherever possible.� (PSRC information sheet on Air Tanzania, 1999) To “promote international trade and tourism,� GOT liberalized the air transportation sector and licensed a number of private air companies to serve domestic and regional routes. An autonomous Tanzanian Civil Aviation Authority (TCAA) was created in 1999, to regulate the safety and public service aspects of all airlines operating in the country. The IFC, which became the lead advisor in the privatization process, stated in its Information Memorandum on Air Tanzania (2002) that “ATC is not considered ‘a strategic asset,’� and “GOT has no intention of maintaining any residual shareholding, ‘Golden Share,’ nor other forms of equity interest in the privatized….� company. The decision was made to sell a stake in Air Tanzania to a strategic partner, in the hope of bringing in capital, expertise, and access to larger markets. The initial sale was to be 49 percent of shares, with Government retaining the majority stake. Note that in the Information Memorandum to potential bidders, IFC stated that following the initial sale, GOT “plans to continue the divestment process� by sales to employees, the public and local institutional investors. The Memorandum also held out the prospect that the GOT would consider the eventual sale of more shares to the chosen strategic partner, up to 75 percent of equity—but that was for the future, and dependent on legal issues related to “bilateral air service restrictions.� As a start, only the minority stake would be placed on the market. The Process In March of 2001, the GOT employed the IFC as the lead advisors in the Air Tanzania privatization. SH&E were hired as technical consultants, Clifford Chance engaged as legal advisors, and PWC taken on to handle accounting issues. An “indicative timetable� for the transaction was established: Table 1: Air Tanzania Privatization Timetable No. Activity Planned Actual 1 Recruitment transaction consultants By May 2001 Done 2 Commencement of strategy analysis May 2001 Done 3 Strategy report submitted September 2001 10.01 4 Government approval of strategy Oct.-Nov. 2001 ½ 5 Pre-qualification of bidders December 2001 02.02 6 Launch of tender with transaction documents February 2002 05.02 7 Deadline for bid submission September 2002 Done 82 8 Announcement of preferred bidder November 2002 Done 9 Contract closure and handover November 2002 12.02 The timetable was reasonably tight and the deadlines were substantially met, with only slight slippages in a few steps, including the final handover date (December 2, 2002). The sales strategy included the following elements: To attract a bidder, GOT undertook to remove most liabilities from Air Tanzania and only transfer to the new Air Tanzania Corporation Ltd. (henceforth ATCL) some current commercial trade accounts, and sums due to the International Air Transport Authority (IATA, the body that regulates international air business). The bulk of non-operating assets and about $ 6millionin liabilities would be transferred to a new Air Tanzania Holding Company (ATHCO). Its job would be to manage or sell the non-core assets, and recover as much as possible of the outstanding debts. Thus, ATCL would start its new life with a very clean balance sheet. GOT would retain a majority stake in the firm, but any strategic investor would insist on corporate control. IFC thus structured the draft shareholders’ agreement to give management control to the investor. To enhance further the chances of getting a good strategic partner, the ATCL Evaluation Panel in the PSRC announced it would award 20 percent of ranking points on the basis of the completeness of the bidders’ proposed business plans, and “the extent to which they meet Government objectives for ATCL.� (PRSC information sheet, 2001) On the negative side, from the investor’s viewpoint, the GOT was a signer of the Yamoussoukro Decision on regional air liberalization, meaning no protection would be offered to ATCL; it would have to operate in a competitive environment. The bidding documents proposed that the sum paid be split between the Government and the ATCL. The reasoning was that the company needed funds for restructuring while the GOT needed to recover the costs of sale and offset the substantial liabilities taken on. Expressions of interest were sought; eight potential bidders responded: Aero Asia (Pakistan), Air Consult (Ireland), Comair (South Africa), Gulf Air Falcon (UAE), Kenya Airways, Nationwide (South Africa), Precision Air (Tanzania) and South African Airways. At subsequent bidders’ conferences held to answer investors’ questions, strong and persistent interest was noted in the representatives of both Kenya Airways and South Africa Airways. There then occurred the September 11, 2001, attacks in the US. At once, the international air travel and tourism picture darkened. Air Tanzania’s precarious finances slipped further. GOT and IFC feared that the company might not survive until a strategic partner was found. In consequence, an interim management contractor was brought in; i.e., Speedwing, an affiliate of British Airways.74 To cut costs, the previously reduced fleet was cut further, to a single plane, and more routes were eliminated. Emergency funds were raised by the sale of one B-737 and some houses owned by the company. In 74 Speedwing had to commit not to pass on any information regarding Air Tanzania to British Airways, in case BA decided to bid in the forthcoming privatization. 83 addition, 243 of the company’s 493 employees were retrenched 75 (leaving Air Tanzania with 260 workers for a single aircraft). Despite all the difficulties, PSRC hoped that the keen interest of several bidders meant that two or more good bids would be submitted. However, in September of 2002, Kenya Airways and Nationwide submitted letters “declining to bid, but stating they were still interested, should the transaction fail.� (IFC) South African Airways (henceforth SAA) made the only formal bid—but it seemed a very good one. The Bid, and Post-bid Discussions SAA, clearly unaware that it was the sole formal bidder, made an aggressive financial offer: US $20 million, half to go to GOT, half to recapitalize the airline. The Evaluation Panel in the PSRC had already announced that SAA’s business plan was compliant. Though other bids would have been welcome, all were delighted that the regional powerhouse airline, SAA, wished to acquire ATCL. The bid was quickly approved by Cabinet on October 9, 2002. By all accounts, SAA was shocked to learn it had been the only formal bidder. It asked to reopen discussions on some aspects of the bid, including the price. The PSRC and GOT, quite correctly, maintained that the bidder had to respect the “fundamental nature of the transaction.� However, PSRC was willing to discuss SAA’s second main concern; the nature of corporate and financial control in ATCL. It was agreed that both GOT and SAA would appoint four Board members each, with a ninth member to be an impartial, jointly-appointed director. In addition, GOT committed to selling a further 10 percent of shares to a private Tanzanian investor within six months. Agreement was reached and the transaction concluded on 12.02.02. On that day, SAA deposited US$ 20million(Tshs 19.8billionat the then prevailing exchange rate) in a special account, of which $10millionwas immediately transferred to the GOT. The First Business Plan Many of the subsequent debates on the pros and cons of this transaction rest upon the question of what SAA did or did not forecast76 it could do once it assumed control of the firm. Thus, we have to review the business plan originally proposed by SAA for ATCL. This business plan forecast a continuing loss in year one, but a profit both at the operational level and at the bottom line in year 2, and steadily increasing profits thereafter. 75 On average, each retrenched ATC employee received the equivalent of $8,642. 76 Business Plan forecasts are not hard and fast promises. They are only estimates of potential market response to actions within the control of management. This was acknowledged in the ATCL business plan submitted by SAA. It notes: “At this point it is virtually impossible to determine with any level of certainty what the requirements of ATCL will be over the first two years. There are numerous unquantifiable liabilities, such as forward ticket sales and more importantly operating liabilities that may materialize once proper management controls are implemented and ATCL operates for a number of months.� 84 Table 2: Forecast ATCL Income Statement (in SAA’s business plan, 2002) Partial Income Statement (Tshs billions) Year 1 Year 2 Year 3 Year 4 Year 5 Net revenues 40.51 81.52 97.60 113.68 113.68 Total expenses 43.91 80.00 93.30 106.61 106.88 Operating income (3.40) 1.52 4.30 7.07 6.08 Net income (4.40) 0.53 3.56 6.67 6.85 This business plan had some unusual financial features. First, with regard to the $10 million injected into ATCL by SAA: The plan said this was to be “a loan repayable over five years on normal banking commercial terms….this amount will not be treated as a shareholder’s in the new ATCL.� This seemed to be an attempt to minimize SAA’s financial risk in ATCL. Moreover, the plan stated that the projected working capital requirement of $10 million “is an estimate and not necessarily the amount provided for in the capital and training account.� The meaning is not clear; it may have been a further hedging by SAA of its commitments. Second, the forecasted cash flow statement made it clear that “the company, during the first two years is relying on its forward sales monies for survival. This is against all airline standards, however since the ticket stock to be used is SAA’s the risk is in-effect being carried by SAA.� The issue of ticket stocks later became a subject of dispute. The business plan forecast large start-up costs ($6.2 million) in the first year of operation, for: “…..renovation and total revamp of ATC House, the purchase and installation of a complete IT solution, additional provision, if needed, for the transportation of passengers holding existing tickets, costs relating to a large advertising and re-launch campaign, costs relating to the revamp of certain passenger facilities such as lounges etc., and a possible shortfall, if any, in working capital requirements as given in the Price Waterhouse balance sheet as of 1 Nov 2002.� (SAA Business Plan, 2002) The operational essence of the business plan was an ambitious program of fleet and route expansion to make Dar es Salaam a regional hub in its own right and a feeder into SAA’s global program. This was to be led by an equally ambitious training and rehabilitation plan inside the company. 85 III. Results The first acquisition was new planes. Two B-737s were “wet leased�77 from parent SAA, and one prop-jet F-28 leased from AirQuarius (RSA), bringing the fleet back to four aircraft.78 Staff remained at 250, reducing greatly the employee/aircraft ratio. New management from South Africa entered, ostensibly on an interim basis, and was seen as eager to implement the expansionist business plan. From 12.02 through 03.04 (16 months) As a result of fleet expansion, internal and regional flight frequencies increased; from 10 and 8 respectively, to 49 and 19. On time performance improved from 71 to 80 percent. Flight cancellations, which had plagued the old ATC, were greatly reduced. Improvements were registered in revenue management, accounting, maintenance, safety and training. (Air Tanzania Annual Report, 2003-2004) Technically, the airline underwent a positive transformation. But there was bad news on the financial front: In the 16 month period covered in the first Annual Report, ATCL lost Tshs 7.7 billion, or about US $ 7 million. This was a substantial percentage of its available resources, and half again as much as the loss forecasted in year one in the business plan.79 Still, losses had been expected at the outset, and many of the turnaround expenses were thought to be one-time, non-recurrent costs. The real issue was what would happen next. On this, there were some worrying signs on the cost front: The income statement in the first Annual Report showed that the largest three items in ATCL’s cost structure were flight costs (Tshs 9.7 billion), fuel (Tshs 9.7 billion), and lease charges (Tshs 9.3 billion). The Annual Report clarified the status of the $10million(Tshs 9.9 billion) paid by SAA at handover: It was considered as “cash injected by SAA to acquire 49% shareholding;� it was not a loan. To the end of 03.04, Tshs 3.9billionof this sum had been spent on “training, working capital and capital expenditure.� From 03.04 to 12.04 ATCL’s situation deteriorated rapidly through the remainder of 2004. Most of the regional routes failed to make money. Destination reorganizations were carried out, first in 04.04, and again in 11.04, resulting most dramatically in an exit from the Nairobi route. Fuel, airport fees, salaries and operating costs remained high. In its best months in this period the company barely broke even; the normal pattern was that of loss. 77 Wet-leasing is the hiring of an aircraft complete with crew, maintenance and insurance by one airline (the lessor) to fly on the routes of another airline (the lessee). A dry-lease means the lessor runs the leased aircraft with its own resources. 78 According to the CEO, in the first year ATCL also leased two DASH-8 turbo-props. These were gone by the time of the first Annual Report. 79 As ATCL made a loss, no taxes were due. There was “an estimated tax loss� of Tshs 2.25 billion. The total estimated accumulated tax loss was Tshs 8.6 billion, “available for set off against future taxable profits.� (ATCL Annual Report, 25) 86 Moreover, in reaction to its own very poor financial position, new management was appointed in parent SAA—which began to search vigorously for ways to cut costs. In trouble at home, SAA began to lose interest in sustaining the partnership with loss- making Air Tanzania. As the two sides shifted in their perceptions from that of partners to that of competitors, this led first to concern, then tension, and then outright disputes between ATCL’s two shareholders. Representatives of GOT on ATCL’s Board felt that SAA was not living up to the commitments it had made to restructure the company, and that it was advancing its own interests over those of ATCL. They began to question the size of fees SAA charged for its equipment and services.80 The TCAA regulator questioned the financial propriety of the continuing use of SAA ticket stock and demanded ATCL issue and use its own ticket stock. For its part, SAA noted that because the company was in much worse shape than it had anticipated at time of share purchase, costs were higher than forecast. SAA now saw the original business plan as too ambitious and in need of revision. Moreover, GOT never had fulfilled the commitment to sell the additional 10 percent of ATCL equity to a Tanzanian investor, which would have made SAA the clearly dominant shareholder. A sign of the growing problems was that ATCL had three different CEOs in its first two years of operation. Second Business Plan — 12.04 By 12.04 ATCL was down to four domestic and three regional destinations, and a reduced flight frequency rate. Despite these and other efforts to cut costs, and despite steady growth in total passenger numbers, monthly losses increased. The company faced stiff competition that limited its ability to raise prices, and it could not handle the high fuel, operation and leasing costs. A revised Business Plan issued in 12.04 noted that losses in the previous month had exceeded $600,000, and that the company was “severely short of cash.� More broadly, and rather unusually, it asserted that “SAA has not focused on ATCL nor grasped the vision in its entirety where much of SAA’s management and the sales force view ATCL as a competitor.� The airline was “barely sustainable� and faced only two choices: substantial recapitalization and change of strategy, or going out of business. Not surprisingly, the Plan recommended the first option, centered on the wet-leasing of a B-737 800, the re-opening of regional routes, and inaugurating flights to Dubai and Mumbai, expanding poles of regional business. The Plan called for $4.4millionin new funding, with most of the large working capital requirements “to be obtained from third party sources.� Detailed financial forecasts estimated that losses would persist under the new strategy for about 12 months after adoption, after which earnings would turn and stay positive. ATCL’s Board accepted this plan “in principle.� From 12.04 to the present At present, ATCL’s financial situation is dire. The balance on the original $10 million investment is basically nil. The profit/loss statement for the period 04.04 through 03.05 tells a grim story. 80 ATCL made monthly lease payments to SAA of $1.8 million, and monthly maintenance and other fees of a bit over $800,000, for a total monthly transfer of $2.6 million. 87 Table 3. ATCL Summary Profit/Loss Data 04.40 through 03.05 (unaudited) Item Amount (Tshs billions) Total Gross Revenues 43.813 Total Operating Revenue 38.728 Direct Operating Costs 40.125 Operational Loss (1.397) Non-operational Overheads 9.470 Net Loss (10.866) In this period, fuel costs alone accounted for more than a third of direct operating costs, while staff costs made up about half of non-operational overheads. ATCL’s net loss in this period was close to $10 million, pushing total losses since divestiture over $17 million. In an interview in June 2005, the CEO acknowledged that:: “We are not a sustainable entity now.� The firm is “eating up capital;� losses in the period 04.04 through 03.05 averaged more than $800,000 a month. The CEO estimated that turnaround funding requirements were now in the range of US $8 million, and that two years (not one) of restructuring would be required before ATCL could be expected to return to positive earnings. The proposed second business plan has been put on hold, pending resolution of the growing dispute between the shareholders, which has become more open and acrimonious. To weather the crisis, on May 27, 2005, GOT injected Tshs 2 billion (US $ 1.78 million) into ATCL. As of the date of writing (07.14.05) the balance remaining of the injection is about $498,000. With losses at their present rate this sum will disappear quickly. 81 The airline is once again on the verge of collapse. IV. Lessons Learned & Policy Implications A transaction that promised so much, and was greeted with great enthusiasm and hope, has now gone sour. The shareholders have fallen out and there are rumors of both parties threatening legal action. GOT, which launched the sale back in 2000 largely to stop its throwing of good money after bad, is once again subsidizing the airline. It is highly 81 This suggests that ATCL is eating through this injection at a rate of about $31,000 a day. At that pace, the injection will be exhausted at the end of July, 2005. 88 likely, given the hemorrhage of money in ATCL, and the fact that national elections will be held in a few months, that the first GOT payment of T shs 2 billion will be followed by others. What is absolutely clear is that a large amount of fresh capital is needed if ATCL is to survive and make a renewed effort to become competitive and profitable. If that needed capital should come from SAA or some other private or external investor (Gulf Air? previously interested parties?), well and good. Reportedly, Tanzanian decision-makers are talking to South African leaders in hopes of finding a political solution to the crisis that will keep South African interest, and money, in ATCL. And ATCL’s Board is searching for other investors. The fact that air passenger traffic is increasing in Tanzania and the region, and the profitable operations of other East African airlines—e.g., Kenya Airlines and Precision Air—suggest that a firm with the right business model and deep capital resources might make a go of it.82 The consultants’ view is that this is a private business matter and sector. There is no economic or indeed social justification for the needed capital to come from the Government of Tanzania, either as a loan or as equity. Why? Because governments always have objectives in addition to profits and efficiency. Experience has shown how difficult it is for governments to run any business, much less the ruthlessly competitive present-day airline business, with the hard-headedness and agility needed to succeed. ATCL’s present situation is disastrous. And its proposed business plan is very risky and expensive and would subject the company to compete head-on with Emirates, one of the richest and most successful airlines in the world. (Not to mention that Precision Air has stolen a march on ATCL and is already running flights to Dubai.) The chances of finding an investor are slim; the chances of overall success are slight. Moreover, why should GOT invest in a sector that touches the lives of so few Tanzanians? To illustrate, in the period 12.02 – 03.04 ATCL carried 217,261 passengers. Many of these, of course, will be repeat passengers and a number, surely, were not Tanzanian citizens. Still, even assuming that all these people were citizens, the total number represents but ½ of 1 percent of the population of the country. What is the economic rationale of subsidizing air service for such a small number of people—who, most certainly, do not come from the poorer sections of society? What are the alternative uses of public resources invested in ATCL? The point is not to deny the social or economic utility of air service; rather it is to argue that—as the GOT stated back in 2002—air service is not a “strategic� industry that provides an essential commodity, or that is provided by suppliers operating in monopoly markets and thus in need of strict supervision on matters other than safety. If ATCL were to fold, air service would 82 In the best of circumstances it will be difficult. The June, 2005, “East Africa Air Transport Survey� (World Bank) expresses doubt that ATCL can compete without subsidy. It states: “Tanzania air transport market is still small but steadily growing and Air Tanzania posts ambitious objectives for expansion with new routes to London, Dubai and regional destinations. However, in view of the outcome of the first two years of operations, it is doubtful that these objectives can be achieved without a major new influx of equity capital which the private shareholder may be reluctant to provide for the time being. Air Tanzania’s problems illustrate how difficult it can be to make a company competitive after downsizing it and changing its status from a state-owned protected enterprise to a commercially-oriented operation.� 89 continue in Tanzania—though it is quite likely Precision Air would seize the opportunity to raise prices until a new entrant stepped in and competition were restored. The consultants’ are aware of the political and psychological issues attached to air service, especially one carrying the national name. We are also aware (to repeat) of the difficulties of taking dramatic action in the period just before a national election. Nonetheless, the facts must be faced: If, within a very short time—six months?—ATCL does not receive a reprieve, either from a change of heart and mind in SAA, or from some other external investor, then it should be let go to sink or swim entirely on its own. If it should fail it would be a business, not a social failure. True, in industrialized countries airlines can usually obtain some easy credit or legal rulings that tilt the playing field in their favor and allow them to survive—for a time. But the sad truth for developing countries is that they cannot afford as many non-economic luxuries; they are forced to be more rational than their rich neighbors. In the consultants’ view, no more public money should be spent to keep this non-strategic firm afloat. V. Conclusion Was this a “good� privatization? The first best outcome would certainly have been for ATCL to become a profitable, competitive, expanding private firm, along the lines of Kenya Airways. Had that happened, or if somehow the present crisis should be resolved and ATCL performance became positive, then there would be no complaints. If, as seems much more likely, ATCL should fold, then the conclusion of most observers would be that this transaction was a failure. But there is another way to look at it. At the time of sale, Air Tanzania was a steady and large loss-maker, providing little service at a high price. Its large accumulated losses had cost the GOT dearly costs that came from subsidizing the annual losses. The share purchase by SAA pro, both in terms of servicing the liabilities ($6 million worth), and bearing the direct and opportunity vided the GOT with sufficient resources to cover the liabilities, and to recapitalize the company. Tanzania obtained 29 additional months of national air carrier service, at no cost to the GOT (until the end of May 2005), which remained the majority shareholder. It is regrettable that the strategy did not work. It is regrettable that financial performance has been so poor. It is regrettable that the shareholders are falling out; and it would be a regrettable business failure if ATCL were to disappear (as so many other African national airlines have disappeared). But none of this is as regrettable as the restart of subsidy flows from the GOT to the airline. In sum, it looks like this privatization will not end in a “first best� outcome. But a drastic downscaling without subsidization, or even an orderly liquidation, could be a second best outcome. Among the worse possible outcomes would be the company’s return to an existence dependent on government subsidies. 90 Case Study 5 Tanzania Harbours Authority (THA) I. The situation pre-privatization In late 1996 the Government of Tanzania (GOT) took the decision to expand its privatization program to divest all Public Enterprises involved in major utilities and infrastructure (ports, railways, electricity, telecommunications, etc.). In transport, the reasons behind the expansion of the program were to (i) lower transportation costs and achieve a higher degree of reliability (rail & port), (ii) to give the domestic private sector an opportunity to participate more fully in the economy by opening up areas that were previously restricted to or dominated by the public sector. This allowed greater scope both for entering into partnerships with international firms and for significant technological and managerial skills transfers. It would also (iii) lower prices to consumers through increased productivity (and therefore lower unit costs) on imports coming through the container terminal as well as products transported by train. The Tanzania Harbors Authority (THA), a public corporation 100% owned by the GOT and incorporated by the THA Act of 1977, was governed by the Public Corporations Act of 1992. THA activities concerned mainly the Dar es Salaam service port, the gateway to the north and south transport corridors into and through Tanzania to the inland countries of Burundi, Rwanda, Democratic Republic of Congo, Zambia, Uganda and Malawi. In this context GOT felt THA could help generate greater economic growth if it could better compete and play a major role within the regional transport scene. THA core business units include the container terminal (now leased), general cargo berths, grain terminal, passenger terminal and lighterage quay, oil terminals, marine services and dockyard. THA operates other lesser ports, Tanga and Mtwara as well as the minor ports of Lindi, Mafia and Kilwa. Ancillary activities include Bandari College, medical services, and some real estate. THA has a workforce of about 4,000. The Government’s decision to privatize THA followed four years of recession in port traffic, from 1997 through 2001. 91 Traffic through Tanzania’s main Sea Ports (1993-2003) tons DAR ES 1993/4 1994/5 1995/6 1996/7 1997/8 1998/9 1999/0 2000/01 2001/02 2002/03 SALAAM IMPORTS 3130725 3393560 3073447 2964083 3217702 3420527 3272752 3218586 3431432 4247337 EXPORTS 901129 994685 858334 744618 683821 678978 740634 624018 715092 762262 TOTAL 4031854 4388245 3931781 3708701 3901523 4099505 4013386 3842604 4146524 5009599 TANGA IMPORTS 55596 42151 28396 36093 68362 94362 82158 96105 97742 114869 EXPORTS 99363 58085 78401 70064 96304 81903 92192 105767 152335 187089 TOTAL 154959 100236 106797 106157 164666 176265 174350 201872 250077 301958 MTWARA IMPORTS 47345 37000 42264 53381 68865 74049 82101 74244 65555 77898 EXPORTS 47790 57926 64355 63609 92519 111083 118387 109694 72264 91667 TOTAL 95135 94926 106619 116990 161384 185132 200488 183938 137819 169565 GRAND TOTAL 4281948 4583407 4145197 3931848 4227573 4460902 4388224 4228414 4534420 5481122 The combination of improving economic conditions in Tanzania and renewed CARGO AND PASSENGER TRAFFIC AT MAJOR PORTS economic growth in inland countries following the stabilization of political 6000000 5000000 situations there have led to significant 4000000 Tons increases in port traffic since 2002 in all 3000000 2000000 types of traffic, except passenger traffic. 1000000 Container traffic has recorded significant 0 yearly increases since the Cargo Terminal 94 95 96 97 98 99 00 01 02 03 19 19 19 19 19 19 20 20 20 20 was leased in late 2000. CARGO PASSENGERS DAR ES SALAAM PORT UNCONTAINERIZED CONTAINERIZED PERFORMANCE 2003 LIQUID BULK The port utilized 48% of its annual capacity of 10.5 m tons in 2003, up by 21% over 2002 with imports representing 85% of traffic. The port’s annual general cargo capacity of 4.5 m tons inclusive of containerized cargo was 70% in 2003, an increase of 26% over 2002. For containerized traffic, the port of Dar es Salaam has a capacity to handle 250,000 TEU83s per annum. The port recorded a remarkable increase in containerized traffic when it handled 178,154 TEUs in 2003, compared to 144,902 TEUs handled the previous year. The port handles bulk liquid at two installations; the Single Buoy Mooring which is an exclusive facility for handling crude oil tankers of up to 100,000 tons, and the Kurasini 83 Twenty foot equivalent 92 Oil Jetty (KOJ) for handling white products ships with capacity of up to 40,000 tons. The port is rated to handle 6 million tons of bulk liquids per annum. During 2003 it handled 1.9 million tons of bulk liquid of which 98% was offloaded, 15% more than 2002. Dar es Salaam port serves the neighboring countries of Zambia, Malawi, Rwanda, Burundi, DRC and Uganda. Transit traffic inclusive of general cargo, containerized and bulk liquid products, constituted 26% of the total port traffic during 2003, up 22% compared to the previous year. The port registered a big increase (39%) in ship calls to 1,459 ships in 2003. The increase is attributed mainly to coastal ships, which registered an increase of 306% (541 dry general cargo ships and 15 tankers). Deep-sea ship calls remained overall at about the same level as the previous year with 291 dry general cargo ships, 103 tankers and 509 container vessels calling at the port. The ships’ stay time in port improved for deep-sea vessels to 2.9 days for dry general cargo ships, 1.3 days for container vessels and 4.4 days for tankers. Productivity at the Container Terminal improved further to an average of 434.4 moves per 24 hours in 2003, against 396 the previous year. The Dar es Salaam Container Terminal is now rated among the best in the region in terms of productivity. At the General Cargo Terminal productivity per shift per gang leveled at 221.5 tons, 446.0 tons and 176 units for break bulk, dry bulk and vehicles respectively. When compared to the previous year’s performance, major improvement was registered in off loading of car carriers. OTHER PORTS Operating at a much lower level than Dar es Salam and at about half capacity, the other two major ports in Tanzania, Tanga and Mtwara, will require much investment to rectify the lack of depth and of a functioning railway for the former, and to dredge the port, for the latter. Both ports have shown a similar recent increase in volume. THA FINANCIAL STATUS In 2003, THA generated operating revenue of Tshs 69.6 billion, an increase of 24% THA - Financials mainly due to an increase in cargo throughput by 16% while expenditure 120.0 100.0 growth was kept at 2% which helped 80.0 generate an operating surplus of Tshs. 11,8 Tsh bn. 60.0 billion. (see appendix 1) 40.0 20.0 Over the past five year’s the Harbour 0.0 Authority has shown little revenue growth -20.0 1998/1999 1999/2000 2000/2001 2001/2002 2002/2003 but has regularly generated a return of Op. Revenue Op. Expense about 10% of assets with a marked jump in Op. Surplus/(Deficit) Fixed Assets 2002 following the privatization of the Container Terminal. The Operating deficit of 2002 stemmed from civil unrest in neighboring countries, as well as negative developments in Zambian Copper Mining. 93 II. Impetus for Privatizing THA The impetus for privatizing THA therefore stems mainly from the objective to improve T.H.A’s competitiveness against ports strung along the East Africa coast from Mombassa to Durban; and to take advantage of changing economic conditions of accessible countries in southern Africa such as Burundi, Rwanda, Democratic Republic of Congo, Zambia, Uganda and Malawi. Tapping these markets will not be an easy task since the Port’s competitiveness is linked to the efficiency of road and rail infrastructure within Tanzania, and to some extent is impacted by the political stability and the economies of the land locked countries. Government’s objectives in privatizing THA were to improve efficiency of the port operations and to enhance the quality of services offered to customers so as to improve the port’s competitiveness. The original reform strategy for THA, commercialization, was changed in 1997 to that of unbundling. The Government engaged external consultants/advisers, Rotterdam Maritime Group of the Netherlands, to assist in the privatization and liberalization of the Coastal Ports and the Maritime Sector. The primary objectives of the consultancy services were: 1. To develop an enabling environment to bring about the privatization and liberalization of the sector; 2. To review and finalize the strategy for privatizing THA ; 3. To provide legal, commercial, port management and investment banking advisory and support services to the Government through the PSRC in the implementation of the approved privatization option for the core activities of THA. The Consultants evaluated four options for privatising THA: 1. Multiple concessions for and disposals of the individual business units within THA and the restructuring of THA as a landlord port enterprise; 2. One single master concession to cover core business activities and disposal of non core activities and assets ; 3. A master concession and sub-concessioning by the master concessionaire; 4. Creation of free ports and export processing zones. Strategy The Landlord enterprise model of port management, the most widespread model in use internationally, was selected. In the Landlord model the Government retains ownership of the major port assets and the Harbour Authority is converted into an asset holding entity (the Landlord), while port activities are given out to private sector management through long-term agreements such as leases or concession contracts. The approach unbundled 94 THA into different business units each with its own strategy. The Container Terminal was selected to go first. The transaction was completed in 2000 (see the next case study). Activity Privatization Method Single Buoy Mooring BOT Concession Oil Jetty, Privatization Multi-purpose terminal + workshop; Marine services + dockyard; Container Terminal Port-city development; Public Private Partnership Passenger terminal Lease to Private Operators Bandari College Commercialization Medical Services; Separate Cost Centers Other Ports Subsidies required In April 2000 the Government made the decision to create two independent multi-sectoral regulatory agencies, one to regulate the utilities, the other to regulate all forms of transportation covering rail, land and marine transport. This led, in 2004, to the passage of a new Ports and Maritime Sector Law, and the establishment of the new regulatory agency SUMATRA (Surface and Marine Transport Regulatory Authority). NEXT PHASES RMG proposed its services to carry out all the remaining unbundling activities. The World Bank objected to this and the next phase of THA privatization will now only involve privatizing the Single Buoy Mooring, the Oil Jetty and the Multi-purpose terminal. RMG have started drafting the call for tender documents for these operations. No Activity Indicative Timeframe Status Consultancy firm/consortium to Done 1 April 2001 start phase I 2 Strategy report September 2002 Done Government approval on strategy Done 3 By October 2003 for the maritime sector Three months after On-going Implementation of strategy - phase Government approval with 4 II an elapsed time of 24 months 5 Legislation presented to Parliament October 2004 Establishment of independent 6 multi-sector regulatory agency for By March 2005 transportation 95 Case Study 6 Tanzania International Container Terminal Services (TICTS) I. PRIVATIZATION PROCESS The Government successfully completed the privatization of the Container Terminal in 2000 through an international competitive tender process. Eleven bidders expressed interest of which five were pre-qualified. Two final bids were received and the winning bidder was the consortium of International Container Terminal Services Inc., Manila, Philippines, and Vertex Financial Services Ltd of Dar es Salaam. The consortium formed and registered a local company at the end of April 2000 called Tanzania International Container Terminal Services Ltd (TICTS) with International Container Terminal Inc-Philippine (ICTSI) holding 70 % and Vertex Financial Services 30%. The ten-year lease contract was signed on 5 May 2000. Following a four-month delay due to maintenance and labor issues, the responsibility for running and operating the Container Terminal was handed over to TICTS on 10 September 2000. In 2001, Hutchinson Port Holdings (HPH) Hong Kong purchased the ICTSI shares in TICTS and Vertex transferred its shares to Harbour Investments Limited (HIL) following a takeover of ICTSI. This transition has been seamless. LEASING ARRANGEMENTS The lease is for a ten-year period comprising a fixed annual payment of US$3,680,000 and a throughput fee of US$ 13.00 per TEU. THA is responsible for major defects related to design and construction. A Performance Bond of US$ 5.0 million was issued in favor of the Lessor. Performance targets were established as part of the lease at : - Years 2 – 4 : 20 Crane Moves per hour - Years 5 – 10 : 25 Crane Moves per hour After the first year, tariffs were to be reduced by a fixed amount equal to 3% of the starting level in each year of the lease for the first five years. By the 5th year, expatriate staff would have to be reduced by 50%. TICTS started operations free of any liabilities and benefiting from a recent major infrastructure rehabilitation program (although major maintenance of equipment was sorely needed). Their main strategy was to refurbish and upgrade all major handling equipment, to supplement or replace Container Handling Equipment as required, to introduce 96 Computerized Processes for Container Handling and Control, Equipment, Maintenance, Financial and Human Resource Management Activities, to implement intensive and on- going training programs for Tanzanians, to work closely with Customs and Inland Transport Providers (Rail and Road) to ensure seamless inland and cross-border transport logistics and finally to launch an international marketing effort to attract business to the Container Terminal. OUTCOMES TICTS financed the retrenchment of some 50-60% of staff to reduce to about 380 employees today. However, overall employment has not suffered; some additional 500 people have been employed on a sub-contractual basis in areas such as security. Salaries increased on average 300% and employees share in the profits of the corporation (5% of dividends are distributed as an annual bonus representing about one month of salary). Extensive training programs have been Container Terminal - throughput implemented leading to the port now being the fastest in sub-Saharan Africa, faster 250000 than some ports in Europe and Australia. 200000 Expatriate staff has been reduced from 17 150000 TEU to 4; there are 5 Tanzanian managers in 100000 senior management positions today. 50000 TICTS doubled the container throughput in 0 5 years, a 106% increase from 2000 to 1997 1998 1999 2000 2001 2002 2003 2004 2004 including a 1025% increase in the Imports Exports Transhipment volume of transshipment containers (from 5000 TEU in 2000, to 56249 in 2004). TICTS improved the Gross Crane Rate by Container Terminal - efficiency 168% from 8 containers per hour per crane in 1997 to 21.4 in March 2005. 40 35 Container moves per TICTS also: 30 hour 25 Container Dwell Time - Reduced Container Dwell-time by 20 (Days) 15 70%. 10 Gross Crane Rate (containers per hour) - Improved Gate Productivity and 5 0 Road Truck Turnaround Time. 5 97 98 99 00 01 02 03 04 - Improved Customer Relations -0 19 19 19 20 20 20 20 20 ar M through better communication and speedy reaction to requests for information or action. - Improved availability of operational equipment. - Completed various Training Programs for Operational, Engineering and Finance Personnel. - Introduced new Computer Technology, and latest container handling equipment. Further container handling equipment is being ordered. Introduced a truck booking system. - Complies with the International Ship and Port Security Regulations. - Has had a positive impact in the market place through joint marketing efforts with other stakeholders, THA and Railways. 97 - Invested some US $6 million to date. FINANCIAL PERFORMANCE At the financial level performance, TICTS is reported in the press 84 as the most profitable unit within the Hutchinson Port Holdings; this due to high tariffs85. CONTAINER TERM. Financials 40 35 30 25 Tsh bn 2002 20 2003 15 10 5 0 Turnover Direct costs Profit before tax Tax Dividends TICTS has also contributed to increasing the level of Corporate, Employment, Value Added and Withholding taxes for the overall Port operations. TICTS has collected and paid to GOT over $136 million during the first five years of the lease (appendix 2). THA POINT OF VIEW The lessor has met or exceeded all stipulated targets. Yet THA management feels strongly that the privatization process was embarked upon without a proper policy being in place. The Container Terminal was the most profitable entity in the THA operations and privatizing it first deprived THA of sorely needed revenues. THA management feels it has lost money on the transaction value itself and that returns to THA on the value of the assets could have been better. THA claims to have lost money over the first four years 84 Per MD of THA 85 TICTS MD 98 of the lease and has only very recently shown a positive return on its investment in Container Terminal assets. THA did not like the way in which the privatization process was handled. THA had been doing well financially for the years prior to privatization and feels it should have been recognized as and considered a valuable contributor to the process. THA felt that PSRC attributed too great importance to the proposals of the appointed Privatization Agent instead of using the THA staff’s knowledge and understanding of the organization. THA felt the Privatization Agent relaxed certain contractual conditions more than was necessary so as to ensure a successful transaction. THA feel that the PSRC should not be paramount in the privatization process, but should act in an advisory capacity only. THA claims that GOT rushed into the leasing of the Container Terminal to meet the conditionalities of a large credit from the World Bank86. TICTS IMPEDIMENTS With the rapid expansion of volume at the Container terminal, new problems are being experienced due mainly to the lack of performance of road and rail services, the lack of space at the port and customs clearance delays, related to pre-shipment and destination inspections. Discussions are underway to allocate extra space to the Container Operations. TICTS considers the regulator tariff of 1% of revenue to be too high. II. CONCLUSIONS The Container Terminal lease has been a resounding success. This experience confirms the GOT and PSRC vision that private management in port operations can lead to major productivity increases. All productivity performance targets have been exceeded. The Container Terminal is seen as the fastest in sub-Saharan Africa. The private operator has invested in the operation and has earned his return. Although TICTS does make money on this operation, and apparently a lot, so does GOT through major collections of charges and taxes to the tune of $136 million over 5 years. This is not a bad deal for GOT. And customers and consumers have benefited from regularly decreasing tariffs. Employees have shared in the benefits of this process, through increased salaries and profit sharing. Although some employees lost their jobs at the start, today more people are employed than before. Staff has been trained, local managers promoted, salaries increased, and safety and security have been markedly improved. 86 Unconfirmed 99 Lessons learned - Privatization does work when the conditions are right. Although the sudden change of shareholders soon after privatization was disconcerting, a strong and experienced shareholder/operator took over operations and proved his worth. The lesson here is that shareholder agreements need to be scrutinized at the time of privatization to ensure that takeovers of privatized companies are carried out without jeopardizing the objectives of the process. - The Container terminal did have a good history of profitability, market prospects and potential for increased productivity. The amount of investment was low in relation to revenue, and tariffs were high in relation to costs. Not all privatizations occur in such good circumstances. - The bidding process was transparent and a real competition took place. As more and more transactions are carried out and experience and confidence is gained by PSRC, the criticisms addressed to the process for lack of communication with the holding company or for too lenient asking conditions should diminish. The issue here is whether the procurement procedures get the best advisors for Government? Is the quality and cost based system of selection appropriate for privatization when the consequences of a bad deal can be catastrophic for the country? - Let us not look upon a moneymaking operation as a bad deal for GOT. In any event the lease will have to be renegotiated in 5 years and this will provide an opportunity to significantly reduce costs and make the Port even more competitive in the region, and reap greater returns for Government and Tanzania. 100 Appendix A. THA Consolidated Financial Statements 2000-200387 (Tshillings 000) 2000 2001 2002 2003 Fixed Rental Income 2,938,664 1,452,501 3,413,739 Royalty 1,160,908 3,277,040 2,240,613 Services to Ships 7,870,235 7,199,760 8,775,398 10,071,425 Handling Overside 11,243,396 7,044,736 5,248,453 6,910,664 Shore Handling 12,440,139 5,128,381 4,479,177 5,668,219 Other Shore Handling 118,498 1,268 9,685 95,567 Wharfage 17,869,143 18,772,558 23,095,203 30,106,945 Storage 7,684,723 4,102,414 2,196,031 2,720,973 Complementary Services 2,577,260 1,805,278 1,443,994 1,252,869 TOTAL OPERATING REVENUE 59,803,394 48,153,966 49,977,481 62,481,014 Non-Operating Revenue 126,626 191,330 239,245 119,734 TOTAL REVENUE 59,930,019 48,345,296 50,216,727 62,600,748 Sales and Benefits 14,928,245 12,448,682 13,475,882 13,693,640 Maintenance Expenses 4,108,216 2,250,440 2,763,531 2,722,290 Operating Expenses 6,451,891 4,570,442 4,724,537 4,747,329 General and Admin Expenses 2,720,974 2,132,089 2,208,833 2,929,158 Provision for Audit Fees 50,179 36,957 42,845 45,416 Depreciation Expense 7,095,585 7,566,530 10,898,066 10,297,139 Financial Expenses 6,250,420 4,740,606 5,057,160 4,939,454 TOTAL OPERATING EXPENSES 41,605,511 33,745,747 39,170,854 39,374,427 Non-Operating Expenditure (382,042) (772,937) (371,501) (755,053) Retrenchment Costs 2,202,206 TOTAL EXPENDITURE 41,223,469 35,175,016 38,799,353 38,619,374 SURPLUS/DEFICIT 18,706,551 13,170,280 11,417,373 23,981,375 87 THA annual report 101 Appendix B. Amounts collected by TICTS on behalf of GOT or paid to GOT by TICTS. 2000-2004 Amounts collected for GOT and payments made by TICTS (Mill US $) TOTAL Rental 18.4 Royalties 9.7 Wharfage 67.3 PAYE 2.6 Corporation Tax 14.5 VAT 12.7 Withholding Tax (Rentals) 4.4 Withholding Tax (Others) 2.5 Withholding Tax (Dividends) 2.0 Skills Levy 0.8 Municipal Levies 0.5 Licenses 0.0 Import Duties 0.7 Total 136.0 Source : TICTS 102 Case Study 7 Tanzania Railways Corporation (TRC) I. The context: Railroads in Tanzania The German Colonial Government started the first rail line in what was then Tanganyika in 1893. The railroad network changed hands when the British succeeded the Germans at the end of WW I. The rail network expanded as financing became available between the world wars, and it linked with neighbouring railroads to reach its pinnacle in 1968 as part of the East African Railways Corporation (EARC), combining the railroads of Tanzania, Kenya and Uganda. Tanzania Railways Corporation (TRC) came into being after the collapse of the East African Community in 1977 and the dissolution of EAC. TRC is state owned and governed by the TRC Act of 1977. The network consists of two main lines, namely the Central line and the Tanga line. Since 1976, Dar es Salaam is also linked to Southern Africa via the Tanzania Zambia Railway (TAZARA). TAZARA, financed and built by the People’s Republic of China, was opened to provide a route from Zambia to the sea bypassing Rhodesia, now Zimbabwe. Pressures on the railway sector Railway companies in Tanzania held strategic significance for a substantial period but their unique role has been overtaken by events. Today most customers have the freedom to choose their route and method; road transport has increasingly eroded TRC’s and TAZARA’s once dominant positions. Both railways have found it difficult to compete effectively. Railroads being forced deeper and deeper into the red is a global phenomena caused by fierce competition from roads. The road hauler does not pay the full economic cost of road infrastructure nor does he recover the financial cost. Still, the rapid growth of road traffic cannot be attributed merely to the lower cost structure of the road hauler. The extreme flexibility of road transport and use of better technology to economically provide on-call services, door to door deliveries at precisely scheduled times, point to point deliveries, documentation for shipments across several countries, and a host of other user friendly features make road transport the choice of many customers—even when freight rates are higher by road. Flexibility pricing makes the road hauler a formidable competitor. Legal constraints on lower tariffs rarely improved railways’ competitive position. 103 Both Tanzanian railways have benefited from aid-supported projects, but there is substantial disappointment amongst customers, governments, and donors concerning the results of such investments. Studies have concluded that managerial freedom and entrepreneurship, as well as resources needed to manage change, is unlikely to be found in public ownership.88 Despite the acute performance problems faced by both railways (see below), it is clear a great deal of freight, both actual and potential, is available, that it can be carried profitably, and that railways still have a comparative advantage in this business. The same may be true for passenger traffic. Tanzanian transport policy over the past decade has been characterized by steady deregulation and the consequent entry of the private sector. Within Tanzania the removal of monopoly power and the deregulation of road transport allowed the entry of a vigorous and competitive private sector, made up of both large and small firms. This has had to a substantial loss of market share by TRC and TAZARA for both domestic and import traffic. The transit traffic to inland countries of Zambia, Malawi, DR Congo, the Great Lakes region, South Africa and Zimbabwe has also been affected by the emergence of competition from alternative routings; e.g., through the ports of Durban, Mombassa, Nacala, Beira and Maputo. TRC is therefore exposed to considerable competition. TRC’s Situation The Tanzania route network comprises approximately 2722 km of single-track meter gauge for a total length of 3083 track km. The Central line runs from Dar es Salaam to Tabora (850 km) and from there, one branch goes to Kigoma (453 km) and another to Mwanza (386 km). The Tanga line starts from Tanga to Moshi and Arusha with a total length of 430 km. To connect these two lines there is a 186 km long link line between Ruvu Junction Station on the Central Line and Mruazi Junction on the Tanga line. There are three other branch lines i.e. Kilosa- Kidatu – 102km; Kaliua-Mpanda – 212 km and Manyoni-Singida – 115 km. The Central and Tanga lines were built at the beginning of the 20th century whereas the others were built later, with the last constructed between 1985 and 1997 i.e. the Singida-Manyoni line. TRC is connected to Uganda Railways Corporation by wagon ferries over Lake Victoria, and to Kenya Railways Corporation through a link from the Tanga line to the Kenyan rail network at Taveta. 88 Report on Privatization Options for Railways in Tanzania, Bevin Wade, Oxford Policy Management, October 1997. 104 In 1992, a donor funded Railway Restructuring Project (RRP) resulted in (i) the renewal of 200km of track, the provision of rails and sleepers for casual renewals on 1000km of line (1/3 of the network) and of equipment and machinery for track maintenance and welding, (ii) rehabilitation, repair and purchase of a total of more than 2600 wagons (almost the entire fleet), of 24 locomotives out of 90 and of 50 passenger coaches (with another 27 purchased) out of 150. Notwithstanding that effort, some of the network is still laid with the original German track, circa 1910, and most of the locomotives and rolling stock are more than 30 years old. On issuance of the Call for Tender for the Concession : - The diesel locomotive fleet consisted of 9 types of elderly problem-prone locomotives. 49 main line locomotives, 20 branch line locomotives, 24 shunting locomotives, 1 operational steam locomotive, and 23 non-operational locomotives available for spare parts or rehabilitation. TRC had hired 10 locomotives from RITES (an Indian railway firm whose name we shall see again) on leases expiring in 2004. - TRC operated 166 vehicles of which 134 are coaching stock and 1,534 wagons (50% covered). 423 are non-operational. Wagons are liable to be interchanged with Uganda Railways on the Dar es Salaam corridor. The net balance is of about 300 in favour of Uganda. - TRC’s manpower was of the order of 7,550 composed of: o 177 Managers, engineers, accountants and other professionals, o 1,222 technicians, station masters, permanent way inspectors o 1,399 personnel stores traffic accounts clerks, mechanics, electricians, o 4,752 laborers, messengers, point men, gangmen, etc. As of 2003, peak volume exceeded 1.6 million tons per year. Up-traffic, mainly imports, invariably exceeds exports-oriented down traffic. Transit trade is usually the most profitable. Passenger traffic and seasonal and long haul traffic accounts for 15% of revenues but 17-18% of costs. The numbers of passengers carried historically exceeded 1 million but this fell to 586,000 in 1996, and has now stabilized at a level of about 700,000 in 2001/2002. TRC has long been capacity constrained on most services, most of the time. In particular, wagons are generally in shortage, especially covered wagons. TRC Financial Situation TRC’s situation today is the result of its history of non-investment89, repeated interruptions of expansion programs (during WW1 and WW2, the great depression), and the imposition on management of non-commercial objectives. More recently, the 1977 break-up of the East African Community and Pan-African railways left some of TRC’s principal assets, such as railway workshops, outside Tanzania. In addition, political and then financial disruption in Uganda, Rwanda, Burundi and the Great Lakes in the 1970s and 1980’s, deprived TRC of much lucrative transit trade. Combined with limited investments, this made for relatively few profitable years. 89 As an example, cost savings in 1909 for the central line involved a 2% grade for part of the Rift Valley wall climb requiring banking locomotives and special braking operations. 105 Over the period 1979 to 1991, rate increases were, for political reasons, held well below the inflation rate, further depressing the capacity of TRC to maintain and invest. A surge of investment from 1993 onward (RRP) helped alleviate the investment backlog of TRC’s fundamental track weaknesses (two thirds of TRC’s track was laid before WW1), but constraints on axle load, the need for better subgrade ballasting and culverts and a persistently high accident rate continue to plague the network. Major accidents have increased from 85 in 1995 to 278 by 2000, and there are some 2000 minor accidents per year. There were 93 deaths, 169 serious injuries and 29 collisions from 1998 to 2002. The rail line was closed due to various accidents for over 3 months in total in 2001 (112 days) and 2002 (111 days) and respectively 22 and 23 days for wash-outs and floods. TRC’s financial situation over recent TRC Financials years has been one of revenues barely 80 keeping pace with expenditures. 60 Although revenues are increasing, 40 operational costs are increasing even Tsh bn 20 more, resulting in steadily decreasing net 0 1998 1999 2000 2001 2002 profits, leaving the railroad in ever worse -20 -40 condition. Over the period 1992 to 1996 -60 the drain on TRC cash flow varied REVENUE EXPENDITURES SURPLUS AEI between Tshs10billion and Tshs18billion per year. In 2002 GOT restructured TRC finances to keep it afloat90. TRC o swapped some Tshs 52billion debt with GOT into equity; o converted long term loans to capital advances for some Tshs 44billion; o recorded exceptional costs and deferred foreign exchange losses for Tshs 13billion and Tshs 21.billion respectively; o received a GOT subsidy of Tsh 40billion for operating non-commercial rail networks, past fuel levy and road toll; o but still managed to post a net loss of Tsh 37.0 billion. The lengthy and drawn out process of divestiture is also further straining TRC’s financial crisis. Towards the end of 2003, the delay of the concessioning process had a negative impact on the performance of TRC, especially in respect to locomotive and wagon availability and rail breakages, and TRC could no longer cope with the rising demand for freight transport. TRC is caught in a low performance trap in which problems lead to low performance which in turn increases the problems. Managerial abilities and resources necessary to escape have not been non available, although a market for TRC services seems to exist. Operational Situation There is an urgent need to increase throughput and speed of goods transported by rail. There is also a definite need to reduce manpower and to improve service levels. The condition of the railways at present is extremely poor. Approximately 200km of TRC 90 The liquidity ratio in 2001 was 0.7:1. 106 track has to be renewed, as it is past its useful life and suffering an escalating frequency of rail fractures. Technological Challenges: Safety is a concern for TRC. The main body responsible for regulating and maintaining safety standards is SUMATRA (Surface and Marine Transport Regulatory Authority); however this Authority is still new and is not yet fully operational or ready to take on this responsibility. Down but Not Out Traffic has changed little over the Tanzania Railw ays Corporation - 1991 - 2002 past ten years. Passenger traffic has steadily declined, and freight 2500 traffic has not shown much growth. 2000 TRC’s current performance is 1500 1000 clearly not contributing to the 500 expansion of the economy. In 0 comparison, in recent years major 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 increases in the shipment of all Passengers 000 Freight 000t goods and commodities have been recorded at the Port of Dar es Salaam. The general consensus is that there is plenty of potential for the railways. Tanzania is well situated vis-a-vis neighboring Burundi, Uganda, Kenya and Democratic Republic of Congo, and is well located to attract traffic from other countries. II. Privatizing TRC In late 1996, the Government of Tanzania (GOT) took the decision to expand its privatization program to divest all public enterprises involved in major utilities and infrastructure (ports, railways, electricity, telecommunications, etc.). In transport, the reasons behind the expansion of the program were to (i) lower transportation costs and achieve a higher degree of reliability (rail & port), and (ii) to give the domestic private sector an opportunity to participate more fully in the economy by opening up areas that were previously restricted to or dominated by the public sector. This allowed greater scope both for entering into partnerships with international firms and for significant technological and managerial skills transfers. It would also (iii) lower prices to consumers through increased productivity (and therefore lower unit costs) on imports coming through the container terminal as well as products transported by train. Since 1996, TRC has since undergone internal restructuring. Its railway hotels have been leased out, its catering services have been contracted out, and the marine services have been hived off and corporatized as the Marine Services Company. TRC has discontinued certain non-viable services such as Tanga—Arusha passenger service, without 107 Government objection. It entered into a performance contract with the Government (GOT) in 1997. Reform Options Since 1996, different privatization options have been considered and evaluated: - Public flotation - Joint venture - Management contract - Short term lease - Long term Concession Various options for concessions contracts are: - Fully integrated concession - Functionally separated concession - Concession for specific service - Joint venture concession The experience with railway concessions in Latin American and African countries (Cote d’Ivoire, Burkina Faso, Gabon, Cameroon and Malawi) has, with few exceptions, proven to be very encouraging, and is now accepted as the most promising approach for sustainable railways in developing countries. (More information on international experience and lessons learned with other rail concessions in Africa is provided in Appendix 2.) A concession is a privatizing option that retains the ownership of assets by the state or a public entity while providing many of the benefits of efficiency and profitability of a privately run business. The concession contract is limited to a period of 10-30 years in the case of railways. The concessionaire pays a fee for the use of the assets, and makes a commitment to maintain the assets and invest in additions and improvements as per the agreed terms of the contract. Normally, the concessionaire has the freedom to set tariffs on commercial conditions, and retain profits made (subject to normal taxation). The concessionaire furnishes a bond to support its commitment to provide the investment. One of the features of the concession as a privatization technique is that the initial cost of taking over the railroad is much reduced as compared to an outright purchase. The market for the sale of a railway is normally quite limited, thereby reducing competition. Implementing a concession normally takes less time than a sale of a railway. Concessions have proven to be flexible arrangements that can be designed to suit a variety of specific conditions while leaving the formal ownership of the existing assets in state hands. Matters of national pride and concerns for the sale of a high visibility asset, not to mention the right pricing, are issues that arise in an outright sale. These concerns can be adequately addressed through the concessioning option, if the contract is negotiated carefully, monitored well, and enforced. 108 Concessioning a railway should be considered when sale is out of the question and when: - The railway is in difficulty and internal reform approaches to achieve commercial viability have not been successful to turn the railway around on a sustainable basis; - Investment is required to improve efficiency and/or increase capacity. Concessioning represents a means for tapping private sector capacity to induce private investment. It can lead to increased manageability of the enterprise, resulting in improved efficiency, and a better competitive environment for and business orientation of management. Concession of railways is designed to improve managerial autonomy and increase freedom from political pressures. It can also improve a firm’s access to capital, business orientation, private sector management culture, strongly motivated managers and workforce, turnaround in a minimum time and the ability to meet the competitive challenges of a changing market. Concessioning, it is argued, allows a transaction through a transparent and contestable process, and a short contract preparation and finalisation period. The following table presents concerns governments frequently express about concessions, and how concessioning advocates argue they can be addressed: Loss of control over a strategic institutionFuture conditions can be set such as traffic levels and train schedules during periods of emergency. Service deterioration or discontinuance Penalties can be applied. Deterioration of quality of infrastructure Dedicated depreciation funds can be implemented, periodic inspection can be made by independent authority with penalties for non-compliance. High cost or transport monopoly or abuse Open access can be provided for; built in competition or periodic review checks for excess profits can be made by the regulator. Public service obligation Can be costed-out and subsidised. Safety or freedom of access, enforcement Can be regulated or randomly checked by of standards regulatory entity Traffic growth beyond projection, windfall The concession fee can vary related to profits traffic levels; taxation. Staff redundancy Can be specified in the agreement Low priority given to transport business or Careful screening of bidders; only those failure of a concession with good track record are allowed to bid; government to periodically review and take over concession if it should fail; mechanism is put in place for quick resolution of disputes. Failure for concessionaire to return assets Remedy to be built into the agreement for resolution of disputes; appropriate bonds. Minimum service Include in the agreement for low priority transport business 109 The Concession Framework Selected In May 2001 the GOT decided to privatise TRC as a vertically integrated concession, without open access, and to be granted exclusive right (except for the pre-existing access rights of the Tanzania African Railway Co., TARC, discussed below) to a 25 year concession to operate railway transport services on the railway network. Tanzania African Railway co. (TARC) GOT is concessioning TRC in a In 1997 TARC was given a 20 year exclusive competitive commercial sector. The access to operate certain services in TRC’s objective is to create a financially network. The agreement was redrafted in 2002. viable railway and provide sufficient TARC has exclusive right till 2017 to operate cost effective and efficient transport international transit traffic originating from service while maintaining the assets South Africa from Kidatu (the Tazara-TRC and safeguarding national interests. interchange), to Mwanza (the Lake Victoria The GOT has decided to establish a port serving Uganda) from which freight can Reli Asset Holding Company Ltd. be transported by road to Rwanda and Burundi. (RAHCO) and empower RAHCO to TARC commenced operations in 1999. It was delegate concession of the assets for a further agreed that TARC could also haul unit period of 25 years with the right to train traffic from Tanga Cement Company on a use, manage, maintain and develop non-exclusive basis until August 2008 immoveable and moveable assets and throughout the TRC network subject to 12 provide rail service transport as mos. notice by either TRC or TARC . undertaken by TRC. This arrangement was made known to all bidders for TRC; the details are declared in the The wining bidder will be required to Request for Proposal along with all other major arrange financing and establish a TRC contracts. concessioning company. It will sign the concession agreement with RAHCO to be incorporated in Tanzania, and satisfy the requirements of the licence issued by SUMATRA, as a regulated service provider. The licence will be to provide rail transport services conditional to the acceptance of the safety plans by the Chief Inspector of Railway Safety. The concessionaire will be responsible for the maintenance of concession assets and provision of train planning and control for freight and passenger service. The concessionaire will make regular concession payments to RAHCO for rights to the concession. The concessionaire will provide rail freight and passenger service on the railway network; operate and develop the assets, have the option to purchase the inventory and stores of TRC; acquire the commercial contracts of customers and suppliers, and be obliged to manage and develop the network over the life of the concession. Rail transport services have to be managed and operated by the concessionaire as a commercial activity. The concessionaire will have the freedom to define service configuration, set tariffs, and contract with specific customers. 110 The Government will not provide direct or indirect subsidy to the concessionaire for provision of freight or passenger transportation. The concessionaire will be required to meet base passenger service requirements at a level determined by the Government as stipulated in the concession agreement. The concessionaire will have the option to lease moveable assets, all TRC rolling stock, track maintenance equipment and miscellaneous equipment. RAHCO will retain ownership of locomotives, wagons and carriages. Assets and rolling stock related to the operations will be handed over as is, where is. The Concessionaire can contract, lease or purchase rolling stock from third parties without authorization from RAHCO. Operation and maintenance of moveable assets is the responsibility of the concessionaire who will bear all corresponding costs. Other assets The Marine Rail Ferry concession will be for 5 years and the Concessionaire can thereafter negotiate an extension or best alternative to provide lake transport. A Fibre Optic Cable is laid by some of TRC’s tracks; for this, a 5 yr concession is provided with the concessionaire encouraged to commercialize unused capacity. The Concessionaire is responsible for all renewals and maintenance of moveable assets. RAHCO is to maintain ownership of all railway land and buildings, rail ferry and fibre optic cable. RAHCO is responsible for providing financing for the Capital Investment Program for some $24.35 million to upgrade priority sections east and west of Tabora. The concessionaire will be responsible for insuring that railway infrastructure satisfies minimum conditions during concession and minimum conditions at hand-back. In addition railway infrastructure renewals and upgrades must conform to specified minimum standards. The concessionaire has the option to apply to RAHCO for handback of moveable assets for which he has no use before the end of the concession term. Staff The concessionaire will hire required staff, determine conditions of employment and deal with buildings and questions of employee housing. The Government will take responsibility for retrenchment of workers not hired by the concessionnaire including payment of severance and accrued pension payments during the concession of 25 years. Track access The Concessionaire is free to provide access to other potential railway operators at terms that are mutually acceptable at a price that can be negotiated. Service to other operators will be provided on exactly the same basis as for TARC. The concessionaire collects haulage and wheelage fees from TARC. Other contracts of this type are in existence, notably with the company PANACHE which is starting hauling to Rwanda. These contracts have been declared in the bidding documents and were available for inspection in the PSRC Data Room for the transaction. 111 Reli Assets Holding Company (RAHCO) RAHCO will grant the concession activities and sign the concession agreement. It will serve as landlord for moveable and immoveable assets on behalf of GOT. It will monitor the concession contract and the conditions of the concession agreement. RAHCO will receive concessions payments from the concessionaire as well as proceeds that accrue from the operation of the fibre optic cable that TRC installed along part of their line. GOT has requested that IDA (the World Bank) provide a Partial Risk Guarantee in support of the TRC Concession and has received approval in principle for providing TRC support. The guarantee will support RAHCO co-payment obligations under the concession agreement. A Credit of $33 million for civil works for some 160km of track east and 39 km west of Tabora has been approved by the World Bank as support to the concessioning of TRC. Concession payments will have two components: - A variable concession fee, calculated as a percentage of the gross revenues of the concessionaire and any affiliates as defined by the concession agreement, payable in Tshs. - A periodic fixed concession fee (based on the winning bidder’s financial proposal) in USD indexed to inflation. The Concession Company will be incorporated as a limited liability company under the laws of Tanzania and be fully subscribed by a Strategic Shareholder Company with a minimum capitalization of $10million as a condition precedent to the concession. The Strategic Shareholding Company will hold a minimum of 51% of the outstanding shares of the Concession Company and offer a minimum of 20% of shares and not more than 49% of shares to Tanzanian investors. The Bidders management team must have minimum qualifications of 3 continuous years of experience running minimum sized railroad, with a minimum traffic volume, and minimum number of passengers per year. The Bidders had to show proof of ability to raise financing at the rate of $6 million per year. The bidders also had to meet specified financial ratios of liquidity, debt equity ratio, leverage, and credit rating for each member of the Consortium. The preferred (winning) Bidder must : - Establish the Strategic Shareholding Company not more than 60 days after notification of the Preferred Bidder, (providing $10million or other acceptable method of insuring that the Preferred bidder will maintain control of the company); - Establish the Concession Company not more than 90 days after notification of the Preferred Bidder; - Sign the Concession Agreement not more than 30 days after creation of the Concession Company; 112 Policy Reforms By improving the operating systems, physical resources and staffing levels, a competent concessionaire should be able to make the railway financially viable, but will not be in a position to finance substantial immediate renewal of the basic infrastructure (which remains owned by GOT). In these circumstances, there needs to be clear demarcation of responsibilities between GOT and the concessionaire. GOT will take responsibility for the initial renewal of tracks (financed through the IDA credit), retrenchment of surplus staff and past environmental and social liabilities (to be financed), and the concessionaire would take care of all operational investments and future maintenance and upgrading of the track. It is estimated that the renewal will cost $33 million. It is unlikely that a concessionaire will be willing to finance such immediate heavy investment requirements in track infrastructure owned by GOT, in addition to the other essential investments in operating assets (locomotives and rolling stock) and operating systems (particularly communications). Financing of the track renewal by GOT is a prerequisite for the successful concessioning of the railway. GOT has therefore obtained the World Bank’s assistance for the track renewal, and is looking for assistance to liquidate some outstanding social and environmental liabilities in order to ensure the success of the concessioning process. III. Objectives of Concessionning TRC The concessioning of TRC is expected to result in a substantial improvement in the railways’ capacity, operating efficiency and level of service. The specific benefits for stakeholders are expected to be: (i) for freight customers: reduced transport costs; higher quality and predictability of services; improved national and international competitiveness; reduced damage; (ii) for passengers: lower transit times, fewer delays, reduced accident rates (low income passengers, travelling third class, are protected against real fare increases or reductions in service levels); (iii) for GOT: positive returns from the railways through lease payments and concession fees, as well as taxes on the concessionaire’s profits and reduced pressure on the road infrastructure leading to lower road rehabilitation and maintenance costs; (iv) for the concessionaire: improved physical infrastructure providing the opportunity for attracting more rail traffic and thus the potential for higher profits; and (v) for neighbouring countries of Uganda, Burundi, DRC and Rwanda: reduced transit costs and better access to the sea. 113 Employee Impact Retrenchment is in the future plans; however, the full salary burden and cost of retrenchment will be on TRC, which does not have the money for this. This is a very serious problem, for 4000 employees out of the 7000 employees at TRC will be up for retrenchment. These retrenched employees will also have limited prospects for the future, due to their skills being so specialized. Other Reform Measures Taken Several reforms have already been put into place. A new Railway Act has been created. The asset holding company, RAHCO (Railway Asset Holding Company) was formed in 2004. In April 2000 the Government made the decision to create two independent multi-sectoral regulatory agencies, one to regulate the utilities, the other to regulate three forms of transportation covering rail, land and maritime. The legislation enabling the creation of these regulatory agencies was tabled and passed in the April 2001 parliamentary session, and assented and printed in June 2001 as the Surface and Marine Transport Regulatory Authority Act 2001 (SUMATRA). However, SUMATRA is still busy recruiting. The Government also passed a new Railway Act in February 2002. IV. The Bidding Process The Government of Tanzania advertised the Concession opportunity in December 2001 and January 2002, and short-listed four rail operators who were issued with the bidding documents. However, only one bid was received by the bid submission deadline of March 2003, and it was judged unresponsive. The Government therefore decided to restart the transaction process. The consultants hired to oversee this transaction process were CPCS Transcom. In 2002 a second round of bidding commenced. This re-bid was done in two steps, a Pre-Qualification Round followed by a bidding round with a Two Envelope System (Technical and Financial Proposals); this was designed to ensure the most likely participation of fully and conditionally qualified bidders. The results were that of the 7 bidders, 3 were fully qualified:- - Great Lakes Railway Company Consortium (Comazar/Bolloré Consortium of South Africa); - NLPI/SPOORNET Consortium Of South Africa; - RITES Consortium of India; 114 The following four applicants partially met the pre-qualification criteria and the Government therefore decided to conditionally pre-qualify them to allow them further time to meet the pre-qualification criteria in full: - CANAC of Canada; - Dynamic Rail of Southern Africa; - East Africa Rail Consortium of Southern Africa; - Sheltam/Mvela Consortium of Southern Africa The bidding documents were issued on 24 June 2004 and the bidders’ conference was scheduled to start on 20 August 2004 in line with the timetable below: 1. Issue RFP – 24/06/2004 2. Data Room available from 24/06/2004 – 11/11/2004 3. Deadline for bidders to submit comments on draft Concession Agreement – 13/08/2004 4. Bidders Conference and Official TRC site visit – 20/08/2004 – 28/08/2004 5. Issue Addendum to RFP, including revised Concession Agreement – 16/09/2004 6. Deadline for conditionally pre-qualified bidders to submit pre-qualification forms – 30/09/2004 7. Deadline for completing or amending consortia – 30/09/2004 8. Deadline for bidders to submit comments on revised Concession Agreement – 30/09/2004 9. Notification of fully pre-qualified bidders – 13/10/2004 10. Issue Addendum to RFP, including final version of the Concession Agreement on 21/10/2004 11. Final date for bidders to submit questions and seek clarification – 28/10/2004 12. Final responses to bidders’ requests for clarifications – 04/11/2004 13. Proposal submission date – 11/11/2004 @ 10h00a.m. 14. Notification of technical evaluation and invitation to attend opening of Financial Proposals - 09/12/2004 15. Date of Public Opening of Financial Proposals 16/12/2004 16. Date of Notification of Award of the Concession and designation of Preferred Bidder - 20/01/2005 17. Establishing Strategic Shareholding Company not more than 60 days after notification of the Preferred Bidder 18. Establishing Concession Company not more than 90 days after notification of the Preferred Bidder 19. Sign the Concession Agreement not more than 30 days after creation of the Concession Company 20. Mobilization Period - Mobilization plan to be proposed by Bidder in Proposal This schedule was delayed by about six months with the opening of the financial proposals occurring in June 2005 115 Results and Outcomes Only two bids were eventually received, from the Great Lakes Company Consortium (regrouped with Sheltam), and Rites Consortium of India. The Great Lakes Technical bid from the Great Lakes Railways Company Consortium was found91 to be substantially non-responsive92 since it did not accept the terms and conditions of the concession agreement that was finalized during the tender period. This left the Rites Consortium as the only bidder. RITES are associated with India Railways, the second biggest public railway company in the world. Great Lakes have taken legal action regarding their disqualification.93 Evaluation of technical bids was based on determination of proposal completeness and responsiveness as well as an assessment of detailed business plans. Proposal completeness and responsiveness was judged on the provision of the required mandatory documents, including acceptance of the terms and conditions provided by the Tanzania government in the concession agreement, together with submission of a separate financial proposal. Assessment of the technical proposal was also based on an assessment of the bidder’s capacity to undertake the concession, Rail India Technical & Economic Services Ltd especially the plans regarding business, (RITES) draft safety, mobilization, investment, RITES, with roots in Indian Railways is a immovable and movable assets Government of India Enterprise and a multi- maintenance together with the financing disciplinary ISO 9001 consultancy organization plan. that has been in business for more than a quarter of a century. RITES started consultancy first with The Rites Consortium of India offered a railways, mostly outside India, and later in other concession fee, worth over $100million94, transportation sectors both in India and outside. It to the Railway Asset Holding Company has over the decades graduated to other (RAHCO) for running Tanzania Railways infrastructure sectors. Corporation (TRC) operations. Presently, RITES has more than 35 on-going The winning bidder was to be officially projects in 13 countries worldwide. RITES undertakes investigations and feasibility announced in July 2005, and the studies, integrated design services, institutional concession agreement was to be signed by management and technical support for new railway November 2005 between the Operating projects and rehabilitation and modernization of Company, which will be formed by the existing railways systems. winning bidder, and RAHCO, the asset RITES have contracts for the “Maintenance holding authority. Mobilization is management of rolling stock� for Atlantic Railway expected to take about six months after the network - Fenoco SA, Colombia, and recently won announcement of the winning bidder to the Mozambique Rail Concession. enable the private operator to pull together (Source : RITES website) 91 Dr. Heavenlight Kevishe as reported by the Guardian. 92 Great lakes would not undertake to provide the required passenger service and required financial support from GOT for un-profitable services. 93 An injunction was in place until the 27th of June 2005, not to declare the winner while the judge examined the case. The judge may only render judgment on the process, as to whether it was followed or not, and not on the quality of the bidder. 94 The Guardian. 116 his team, and also involve the purchase of inventories and preparation for taking over operations at the commencement date of the concession agreement. The private operator is expected to take over TRC operations by the end of December 2005. V. Conclusions It is not fully clear just why only one investor was willing to make a formal compliant offer for the TRC Concession. Reasons might include: - Relatively long distances and very small tonnage – Colombia with the same length of track has 60 times more freight; Argentina NCA has the same length but transports 90 times more freight; - The condition of track, wagons and locomotives are poor which translates into heavier investment needs with much longer return periods due to low traffic; - Heavy investment in road network offers stiff competition to the rail; - The existing fuel levy means railways subsidize competing roads Whatever the reason, most potential bidders declined to make an offer, leaving RITES alone in the field.. Repeat of City Water? The fear, of course, is that a poor outcome similar to that experienced with City Water Services (the DAWASA lease) might be in the making, as the TRC concession process shows some similarities to what happened in water. Similarities In both cases the first call for bids was unsuccessful. In both cases, the second call for bids produced but a single formal bidder. Both infrastructure firms were in a crisis situation prior to the transaction. DAWASA was in financial and physical disarray; TRC reportedly requires Tshs 50 billion that GOT simply doesn’t have. In both cases, donor investment funds for badly needed rehabilitation will only start to flow once the transaction is complete, raising pressure on decision-makers. In the water case, Tanzania ended up with an apparently inexperienced set of managers that underestimated the difficult market they were entering; there is some concern this may the case with the rail sector as well. Recall that the more experienced French water firms did not bid for the water lease in the second round, considering the conditions unfavorable. In the case of TRC, the experienced South African railway firms could not convince themselves to submit a fully responsive bid; this led to their disqualification. In short, there are enough similarities to raise concern. 117 Differences But there are significant differences between the two privatizations. In DAWASA credible financial and operating information was considered lacking, this is not the case for TRC. Railroads keep an inordinate amount of data. BIWATER, the selected bidder in for DAWASA’s operations, did not have a strong reputation or wide experience in Africa. RITES, on the other hand, are the experienced technical arm of a large and public corporation which most likely has, or has had, problems similar to those of TRC. The crucial question is: Does RITES have the commercial and especially the financial ability required to turn around TRC? RITES do have experience in operating railways in Africa and South America. It is presently running Mozambique Railways and results after six months are apparently good. But six months is too short a time to reach a conclusive judgment, particularly with regard to the issue of raising investment capital. In Columbia, RITES’ only other operations contract, the service provided is maintenance and management of rolling stock95, not a full concession. Moreover, BIWATER stated at the outset that it would not retrench staff in DAWASA, and then reversed themselves within a few months. In TRC, in contrast, a labor retrenchment program is underway (although it does need financing.) Thus, there are some differences that provide some confidence in RITES vis-a-vis BIWATER. The Way Forward Based on these pros and cons, our recommendations on TRC are: • Immediately strengthen RAHCO’s and SUMATRA’s contract monitoring and regulation capabilities by recruiting experienced technical assistance professionals. We recommend that a rail expert be hired to review the situation and give one final round of advice before concluding the concession agreement and closing the transaction; • Perhaps it would be possible to delay the final decision on RITES’ bid until the GOT reviews its performance in Mozambique (and its reported involvement in bids for rail concessions in neighboring territories); and • Finalize plans for dealing with the retrenchments, unpaid pensions and creditors. 95 RITES website 118 APPENDIX A – OPERATIONAL AND FINANCIAL PERFORMANCE OF TRC TRC Summary Operating Statistics Traffic 1999 2000 2001 2002 2003 Train-km (000) Passenger 722 789 784 640 531 Goods/trip 3061 3208 3359 3282 3312 Mixed 179 183 183 181 182 Trans Africa (1) 246 537 575 413 Other 793 598 718 721 701 Subtotal 4755 5072 5581 5399 5139 Loco-km (000) 5329 5820 6226 5939 6042 Average train Speed Passenger 31 30 30 30 30.2 Freight 18 16 16 16 16 Vehicle-km (000) Freight (4-axle equivalents) 56 65 71 69 74 Passenger Cars 12 13 13 13 12 Gross tonne-km (mill) (excl loco) 2672 2996 3358 2554 2782 Gross tonne-km (mill) (incl loco) 3096 3476 3871 3104 3628 Average train size (trailing tonnes) Passenger 489 473 476 475 457 Freight 649 665 686 682 704 TRC Passenger Load Factors 2003 First Second Second Third TTE’s Sleeping Sitting Passenger km (000) 9,630 42,579 10,558 363,108 18,154 Vehicle km (000) 1,294 2,236 223 5,731 Average Capacity 16 36 60 80 Seat/Birth-km (000) 20,704 80,496 13,380 458,480 Load factor (%) 47% 38% 44% 63% 119 TRC Locomotive Utilisation (sample week in July 2003) Hours Class Fleet Avail. Hours Worked Km Worked Per Operating Loco Ave. Speed Total Oper’g Train Other Train-hrs Total-hrs Dist. (km/h) 73 15 8 3592 1464 2128 30612 183 449 3827 21 89 9 8 3753 3705 48 55822 463 469 6978 15 73R 10 10 5592 5328 264 76525 533 559 7652 14 88 35 31 18105 17745 360 259995 591 604 8666 15 87 2 2 1032 - 1032 8174 - 516 4087 8 64 21 16 9017 3249 5768 78968 203 564 4935 24 65 4 4 2192 669 1523 20712 223 731 6904 31 35 4 2 1272 - 1272 - - - - 636 36 20 17 5424 0 5424 0 - 417 - - 37 5 5 3120 0 3120 0 - 624 - - Total 125 103 53099 32160 20939 530808 332 547 5472 17 TRC FINANCIAL STATEMENTS Item (TShs.Millions) 1998 1999 2000 2001 2002 Operating Revenue 29,601 44,703 48,467 56,609 59,120 Non operating Revenue 4,988 9,890 5,141 2,141 795 Total Revenue 34,589 54,593 53,608 58,750 59,915 Operating Expenditure 24,632 34,802 37,666 42,242 45,556 Non operating Expenditure 10,162 7,190 15,055 15,497 16,615 Total Expenditure 34,794 41,992 52,721 57,739 62,171 Surplus before Exceptional Items -205 2,601 887 1,011 -2,255 Exceptional Items 1,573 650 147 3,738 -34,462 Net Surplus -1,778 1,951 740 -2727 -36,717 (Source : TRC Annual Report 2002) 120 APPENDIX B LESSONS TO BE LEARNED FROM OTHER CONCESSIONS • Lack of clarity in concession contracts can give loopholes for private investors to shift all the operational, business and investment risk to the Government; • The railway concessions which are working in Africa are doing so because the concessions have been modified subsequently to resemble Management Contracts with an ownership option; • There is always a tendency of trying to drag the Government into the operation and have it take on the business risk. Concessioning Railways in West Africa: Lessons Learned: I. SITARAIL: 1. Following two decades of decline and failed attempts to improve the state railways in the public sector, the Abidjan-Ouagadougou Railway has begun to establish a sustainable “track record�. This is especially noteworthy, as it was the first private rail concession in West Africa and served as a “starting point� for later concessions in Cameroon, Gabon, and Congo. 2. The achievement of profitability for the railway is a remarkable achievement after decades of decline. The performance is not only the result of a favorable macroeconomic environment, but also of a sustained program of investment, maintenance, cost controls, and a more commercial orientation. It also suggests how difficult it can be and how long it may take for a concession to be successful for all parties. This reinforces the need for long-term financing, investment support, and a strong sponsor in such projects. 3. The bi-national nature of the concession shows that regional approaches to transport problems are feasible, and in many ways essential for the financial and operational viability of transport networks in Africa. Concession Design and Award 1. One of the most important lessons is the need to devote a great deal of attention to the bidding and selection process for the concession. There is a great need to prospect for potential operators, to strengthen the bidding between candidates. To increase the pool of potential bidders, there is a need to design in the greatest possible detail the “rules of the game� of the concession before requesting bids. In this case, the lack of clarity may have served to reduce potential interest, and created a perception (which may not have been true) that some parties with close ties to the government would have an advantage in working out details of the concession. 2. The lack of specificity in the bidding documents led to an extended period of negotiations before the concession contract was signed. While some negotiation will always be present, the Abidjan-Ouagadougou concession confronted extra difficulties 121 because it involved the concessionaire, two Governments and two landlord corporations, with sometimes divergent interests. 3. The participation of the railway workers’ unions in the negotiation process, while adding to complexity, eased the implementation of the initial layoffs, and was key to the concession getting off to a good start. 4. Flexibility in concession terms may be helpful in the short-term, but may prove more difficult over time. The rolling duration of the concession enables long-term planning and investment. In contrast, the three-year horizon for the payment of the “usage fee� is helpful in the early transition/rehabilitation phase, but may introduce short-term incentives and opportunism in the longer term. The Abidjan-Ouagadougou concession has come to realize this, as it was working toward a longer-horizon payment structure as of 2000. Concession Operation 1. The investment scheme, in which the concessionaire proposes a capital program and bears its costs through servicing of the debt, but relies on the Governments to mobilize the funds, has both good and bad points. The separation of investment planning from finance creates the risk that needed funds will not be forthcoming if the Government is unable to raise external funds or use government revenues. On the positive side, the requirement that the concessionaire service the debt creates incentives to make only investments that are commercially viable. Overall, given the difficulty of raising capital for an untested concession experiment, this investment planning and financing scheme represents a clever and workable compromise in the prevailing risk environment. 2. The creation of railway landlord corporations by each country led to misunderstandings and conflicts between the corporations and the concessionaire. These problems led to delays in the implementation of the rehabilitation and investment programs. Over time, as traffic grows and investment needs increase, these conflicts may worsen. Overall, the role of railway landlord corporations should be limited; once financial and operational performance has proved sustainable, investment responsibilities should devolve to the concession. Regulatory Issues 1. Given strong inter modal competition from trucking and the relatively diverse traffic mix, regulation of tariffs was not a major issue. Regulation thus needs to emphasize safety and environmental issues. 2. In the future, regulatory issues may arise with respect to access rights and charges, once the concessionaire’s right to exclusive use ends after seven years. In practice, the better the service provided, the more likely shippers are to continue to utilize the concession services rather than operate as third party operators paying a fee to SITARAIL. If third party operators do arise, the negotiation of the infrastructure access fee is likely to be quite contentious - just as it is in many other regions of the world. 122 3. Passenger services, especially intra -urban and short-haul operations have continued to lose money. Long-term continuation of such services will likely require public service obligation (PSO) agreements and/or explicit subsidies. These issues will require regulatory oversight with respect to the monitoring of the quality and performance of such services. II. CAMRAIL: General Lessons 1. Even with a clear concession agreement, mobilization of financing remains perhaps the most difficult aspect of private participation in transport. The CAMRAIL concession gives a more extensive financing and investment role to the private sector than the SITARAIL concession in Côte d’Ivoire/Burkina Faso, but had been unable to reach financial close by mid -2000. 2. Multilateral agencies placed a major role in providing technical assistance, on-lending support, and especially through the funding of an innovative pension scheme for redundant workers. 3. Theft, corruption, and illegal access to the network and equipment are severe problems. The concessionaire has a much stronger incentive to address these problems than did the state railway, yet there remains an important role for public agencies to support enhanced enforcement. 4. Overly optimistic forecasts make the mobilization of financing more difficult, result in delayed financial close, and increase the risks that funds may not be provided at all. Concession Design 1. The agreement shows that it is possible to design and award a railway concession in Africa that makes the concessionaire responsible for investment, financing, and commercial risks. In fact, the relatively complete nature of this contract may make such concessions more attractive, as they may reduce the degree of political risk or the effects of non-performance of government obligations. 2. In contrast to the Abidjan-Ouagadougou concession, Cameroon sought a more competitive process of bidding through a very detailed concession contract and award criteria. However, the complexity of the bidding process, involving both technical and financial criteria, made it extremely difficult to evaluate. 123 Case Study 8 Tanzania-Zambia Railway (TAZARA) I. Introduction Railway companies in Tanzania were established to insure strategic purposes and so served for substantial periods. However their unique strategic role has been overtaken by events. Virtually all customers have freedom of choice of route and the increasing role of road transport has eroded TRC’s and TAZARA’s dominant position. Both railways have found it difficult to compete effectively. Both railways have benefited from aid supported projects but there is substantial disappointment amongst customers, governments, donors on the results of such investments. Studies have concluded that managerial freedom and entrepreneurship as well as resources needed to manage change to survive is unlikely to be found in public ownership.96 Despite the acute performance problems faced by both railways, it is clear that substantial increasing quantities of freight are available to be moved for which railways still have a comparative advantage and can be carried profitably. The same may be true for passenger traffic. Transport policy over the past decade has been characterized by steady deregulation and the consequent entry of the private sector. Within Tanzania the removal of monopoly power and the deregulation of road transport have allowed private entry into the transport sector, with both large and small firms competing. Consumers are free to choose their transport mode. The consequent competition to the railways has been fuelled by increased efficiency improvements as well as by public investments in roads and their maintenance. All this has resulted in a substantial loss of market share by TRC and TAZARA for both domestic and import traffic. The transit traffic has also been affected by the (re)emergence of competition from other African ports, in particular Durban and Beira, as well as the continued strong competition from Mombasa. Historical Background of TAZARA The Tanzania Zambia Railway Authority (TAZARA) was opened to provide a route for Zambia copper ore to the sea bypassing Rhodesia, now Zimbabwe. TAZARA was constructed as a turn key project between 1970 and 1975 and financed through an interest free loan of RMBY 988 Million (equivalent to US Dollars 500 Million) from the Peoples Republic of China. It started commercial operations in July 1976. 96 Report on Privatization options for Railways in Tanzania, Besin Wade Oxford Policy Management October 1997 124 TAZARA is an integral part of the southern Africa Regional Rail Transport Network. It is a rail link of approximately 1,860 km. long rising at sea level from Dar es Salaam to New Kapiri Mposhi at 1400m above sea level in the Central African country of Zambia. The railroad is designed with a 1067 mm gauge that permits through traffic operations with contiguous railways of Southern Africa. Equipped with adequate facilities, TAZARA has a designed capacity of 5 million tonnes of freight a year. TAZARA currently handles exports/imports of both Tanzania and Zambia, as well as Malawi, DR Congo, the great lakes region, South Africa and Zimbabwe. The construction of Kidatu Transshipment Depot has opened and linked new markets of Kenya, Uganda, Rwanda and Burundi. The containerized traffic arriving from South Africa as well as imports through Dar es Salaam port can now be easily transshipped at Kidatu (by TAZARA or TRC) to other destinations in the SADC countries and East & Central Africa. Spiral of Decline The ending of the Apartheid era in South Africa ten years ago, has resulted in the opening up of competing routes southwards, and an increase in competition from road transport following the liberalization of that sector and public and private investment in it. Also, the decline in the production of Zambian Copper has directly resulted in TAZARA’s loss of freight traffic and a resulting reduction in the Zambian economy with the effect of even more reduced export and import traffic. TAZARA has been unable to recapture its market. From over 1 million tons transported in 1992, today TAZARA has difficulty exceeding 600,000 tons of traffic. Passenger traffic has fallen from its peak of nearly 2.3 million in 1992 to half that number today. TAZARA - Evolution of traffic 1991-2002 Transportation 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 TAZARA Import/Export & Local Performance TAZARA Share of Zam bia Copper (000 tonnes) 450000 700 400000 600 350000 500 300000 Zam bia Copper 400 250000 Production Total Exports 200000 300 TAZARA Traffic Total Im ports 150000 200 100000 100 50000 0 0 2001 Est 2001 1996 2000 2001 2002 2003 1980 1985 1990 Act 2002 Est 2002 Act 2003 Est 2003 Act 125 Passengers (000) 1570 2280 2220 1824 1733 1269 1326 1634 1388 1609 1262 1212 Freight (000 tons) 825 1049 889 635 639 614 586 639 619 638 558 677 Financial Situation Even after loosing half its TAZARA Total traffic business volume TAZARA has been unable to adjust its resource 1400 1200 base, business revenues have not 1000 been able to cover its operating 800 000t Taffic (000t) 600 costs for some time. 400 200 This has led to shortages of 0 working capital, which in turn 87 89 91 93 95 97 99 01 03 19 19 19 19 19 19 19 20 20 has negatively impacted TAZARA’s ability to maintain its asset base and to provide a quality service to its customers, resulting in further pressure on traffic volumes. TAZARA now find itself in a position where net assets have been TAZARA Profit and Loss (USD m illions) entirely wiped out by accumulated 80 losses. TAZARA is unable to 60 service its long-term loans and its Turnover 40 current liabilities exceed its current Operating Expenses 20 assets. Loss for the year 0 Net Assets China continues to assist TAZARA -20 1999 2000 2001 2002 2003 in commodities and spare parts, to -40 the tune of some 12 agreements, each valued at approximately US $10 million97. TAZARA has a built up huge amount of accruals, and is carrying large amounts of debt. The company’s pension fund is under funded by some $4 million and accruals to employees stemming from leave payments, medical allowances and overtime, are huge. The company faces regular cash shortages, and suppliers now demand that the railroad pays in cash for key necessities such as fuel. TAZARA’s profit and loss statement and balance sheet from 1998 – 2003 are appended and show a history of significant losses year after year without any credible corrective action being put into place by management. Hopeful Signs The recent improvement in the Zambian Economy ($1billion FDI in the mining sector), combined with an promising signs from the Democratic Republic of Congo’s economy, are positive for the struggling TAZARA. The good relationship that exists between Tanzania and Zambia also further promotes a positive investment climate. 97 MD TAZARA 126 TAZARA is preparing its employees for privatization, and this involves talks with Union representatives, who then in turn discuss the privatization process with the workers. It was said that the employees feel no apprehension at the idea of privatization. Some retrenchments have occurred, the workforce being scaled down from an approximate total of 5500 employees to 4000. Conditions for Success TAZARA could possibly be viable with its current levels of traffic and resources, but needs a significant amount of support given its existing financial obligations; in the order of over $100million. Consultants98 have concluded that TAZARA can be a viable and profitable business if the company is able to improve its operating performance sufficient to capture business up to a level of 800,000 tons per annum and is able to restructure it’s workforce to a level of around 1300. For this to happen, financial support is required to : - deal with TAZARA’s long term debt of approximately US$50 million; - reduce its levels of external creditors to sustainable levels. This includes trade creditors, unpaid pension contributions and contingent liabilities litigation, which is estimated to be $40 million; - finance any deficit that may arise from its pension fund, estimated at US $ 4 million; - finance the retrenchment costs associated with the staff reduction, estimated at US $25 million. Legal and Regulatory Reforms The Consultant has recommended that TAZARA be privatised as a vertically integrated bi-national railway under a concession arrangement, whereby the concession holder has the obligation to maintain TAZARA’s infrastructure assets, and to operate a specified minimum passenger service. The concessionaire will also have the exclusive right to operate freight trains on the rail line. TAZARA’s rolling stock could be sold or leased to the concessionaire. The market size is insufficient to allow open access at this stage. But the concession could eventually allow for open access in the future if certain targets were achieved. These trigger points could relate to traffic/revenue levels, the operational performance of the concessionaire, and other factors which would be detailed during the concession definitions phase. Level Playing Field In January 2004 Railway Systems of Zambia Limited (RSZ) signed a 20-year concession agreement to provide rail and freight transportation services on the former Zambia Railways Limited network. RSZ—a consortium led by New Limpopo Bridge Projects 98 Nick Allen, PricewaterhouseCoopers, Workshop on TAZARA, 8 September 2004 127 Investments Ltd., with the Spoornet division of Transnet as a major shareholder—also was granted a seven-year concession to operate rail passenger services between Livingstone and Kitwe. The transaction99 was part of the Zambia Railways Restructuring Project, sponsored by the World Bank100. There is clearly a very important relationship between RSZ and TAZARA, since TAZARA’s core traffic flows to and from the Copper Belt and the DRC Border. TAZARA relies on efficient and economic access to RSZ’s system to offer a competitive service. At the same time RSZ itself is competing for this traffic that has a choice of South African or Tanzanian ports for entry or exit. It is essential that a level playing field be created for competition to exist in the regional market. This can only be done if access by the TAZARA operator to the Copper Belt and the DRC border can be ensured on transparent, fair and stable terms. As a precondition of RSZ being allowed to bid for TAZARA, RSZ should agree to enter into a contractually enforceable service level agreement with a fall-back of the right to own haulage by the TAZARA operator, prior to privatization. TAZARA should also be allowed operational access to the track to the copper Belt and to the DRC Border after privatization. If such an agreement cannot be reached, then RSZ should be excluded from the competition, and the regulatory arrangements should be strengthened to provide additional comfort for other prospective bidders, that fair access conditions will be enforced through the regulatory process. This would be best achieved by strengthening the role of Zambia Railway Authority in monitoring RSZ’s behaviour towards TAZARA. Reform Structure Legislative change will be required for the implementation of any structural option for TAZARA. The following regulatory and institutional reforms should be made to support the restructuring of Private Sector Participation in TAZARA:- - A new bi-national asset holding entity should be set up to wind up the infrastructure assets and possibly rolling stock (if these are not sold), and to manage the concession contract; - TAZARA should remain as is until such time as it has wound up its residual assets101 (if any), liabilities and responsibilities, after which time TAZARA itself should be wound up. TAZARA will probably require underwriting from the two governments for creditors to agree to this. If the Task Force does not favour this 99 RSZ estimates that it will pay $253 million to the Zambian government over the life of the concession, and invest $40 million over the next five years in infrastructure and rolling stock improvements. (RailwayAge magazine, July 2005) 100 Ottawa, Canada-based CPCS Transcom Limited acted as transaction advisor, assisting the Zambia Privatization Agency to successfully concession the Zambia Railways’ assets and operations. (RailwayAge magazine, July 2005) 101 None of the company’s non-core assets are significant enough to warrant the active involvement of the Zambian Privatization Authority (ZPA) or PSRC in any disposals. TAZARA management however should continue to look for opportunities to rationalize the company’s non-core asset base, through asset disposals and outsourcing on closures. 128 option then having TAZARA also as the asset owner and contract manager is possible, although this is not our preferred recommendation. - SUMATRA in Tanzania and Zambia’s Inspection of Railroads Authority would regulate safety in each of the two jurisdictions, and collaborate in doing so to ensure consistency for the operator. - The Environmental Council of Zambia and National Environment Management Council in Tanzania should regulate environmental matters in their respective jurisdictions. - As TAZARA operates in a competitive environment, no permanent economic regulator would be required. However the contract manager should monitor the economic behaviour of the concessionaire to identify any anti-competitive behaviour. Next Steps - Approval of recommendations (the above proposals are before Government); - Establishing institutional arrangements for implementation; - Securing funding for the costs of privatization, in particular for retrenchments; - Interim changes of TAZARA governance and management; - Financial and corporate restructuring at TAZARA; - Legislative changes; - Transaction preparations; - Transaction implementation. PSRC and the Zambian Privatization Authority will need to appoint a designated leader to oversee the divestiture. At the moment both Authorities are supposed to be overseeing the divestiture process, but in truth neither Authority has really taken the lead other than to Request funding for the Options study from the Public Private Infrastructure Facility (PPIAF, managed by the World Bank). Staff reductions are already underway. Starting from a peak of 5191 employees in 1999, staff numbered 3966 in 2004. A further reduction in staff of 2666 workers is planned by the year 2007, leaving a total TAZARA workforce of 1300. Transition Measures It is estimated that the divestiture process could take three years to be concluded. An estimated investment amount of US$13m will be needed to see TAZARA through this transitional phase in terms of maintenance and minor investments. The Consultant to the transaction has recommended that an Interim Management Contractor be put in place for the three-year transitional period because: - Management teams succeeding each other have failed time and time again to have any impact whatsoever on TAZARA’s financial situation. - A performance based management contract has more chance of having some effect on the situation than another employment contract. The consultants also recommended that a Turnaround Board be put in place: 129 - To give the Management team the necessary autonomy to carry out its mandate. Over the next few months Government will need to make a decision on the Consultant’s Option Study proposals. In the meantime meetings are planned in China to obtain China’s point of view on the issue of privatization, privatization strategy and debt reduction. There is a possibility that China, as a major creditor to TAZARA, may wish to be involved in some capacity in the privatization process. Or that China’s involvement 102 in Zambian Copper could present other opportunities for transport for TAZARA. The Government of Tanzania should, in the meantime, actively seek means to reduce TAZARA’s debt, seek financing for the retrenchment packages and an interim management contract as well as funding for critical investments during the transition period to privatization. When comes time to design the transaction conditions Government should be aware that transaction advisers have a tendency to design the best deal for the Government with the result that bidders, unfortunately the most experienced in some cases, drop out of the competition in the end. It would be in the Government’s interest to leave conditions a bit more attractive to bidders so as to ensure greater competition. 102 NFC Africa announced in August 2000 that it would start a US$150 million rehabilitation programme at the Chambishi Copper Mine. NFC Africa is the Zambian subsidiary of China Non-Ferrous Metals and owns 85 % of the Chambishi near Kitwe. NFC Africa declared it would not be sending its copper concentrates from the Chambishi mine in Zambia to China for processing since “Zambia is a landlocked country and transport will be too high for us to send the concentrate to China�. (NFC executive Wang Peng as reported in the Platt’s Commodity News) 130 Appendix A - TAZARA financial information 1998-2003 (Tshs billion.) 1999 2000 2001 2002 2003 Turnover 27 34 28 26 25 Operating expenses 40 42 43 40 33 Loss -15 -12 -18 -19 -12 Fixed assets 126 115 99 92 79 Current assets 14 11 11 11 11 Current liabilities -35 -39 -44 -39 -38 Long Term Debt -31 -36 -36 -44 -47 Equity 74 51 29 20 4 Stated capital 54 50 45 43 38 Grants 33 34 32 30 27 Reserves -14 -33 -48 -53 -62 Source: PWC options study report