Communication from the Commission to the Council and the European Parliament - The Reform of State-Owned Enterprises in Developing Countries with focus on public utilities: the Need to Assess All the Options
/* COM/2003/0326 final */
COMMUNICATION FROM THE COMMISSION TO THE COUNCIL AND THE EUROPEAN PARLIAMENT - The Reform of State-Owned Enterprises in Developing Countries with focus on public utilities: The Need to Assess All the Options
1. Poverty reduction is the over-arching objective of the EU's development co-operation  - measured in the medium-term by progress towards the Millennium Development Goals. Improving the performance of State-Owned Enterprises (SOEs), and SOE reform, can have a major impact on developing countries' prospects of achieving those goals. The EU furthermore aims to foster the integration of developing countries into the world economy, for which the development of the business/private sector plays a key role.
 Joint Declaration of Commission and Council on Development Policy, November 2000; Resolution of the UN General Assembly n. 55/2 'United Nations Millennium Declaration', September 2000.
2. For some years there has been heavy pressure on developing countries to reform SOEs. This pressure has been prompted by the heavy burden of SOE losses and economic inefficiency on macro-economic and fiscal balances and growth. Though the Commission has often linked its budget support programmes to adjustment conditionality of the Bretton Woods Institutions (BWIs)  and has been engaged in close and constructive talks with the BWIs, it has not always been explicit on the specifics of SOE reform included in adjustment programmes.
 The International Monetary Fund and the World Bank.
3. The objectives of SOEs reform have not always been made sufficiently clear. Some possible goals - e.g. reducing fiscal deficits by reducing subsidies to SOEs or cashing privatisation receipts - may conflict with others - e.g. ensuring low-cost access to services. The Communication aims at clarifying key issues in this respect and argues for the importance of making the objectives of reform explicit and clear. Only thus can trade-offs be made transparently, and the achievements of each reform assessed.
4. There will usually be several possible ways to achieve these goals. The Communication does not attempt to settle the debate on the advantages of different forms of ownership of enterprises - public, private or PPP. Rather it argues for the essential importance of looking objectively at all the options and their sequencing and selecting the one that best meets the needs of the particular country and field. This needs to pay careful attention to the capacity and resource constraints of the country.
5. Once objectives have been set and a reform option chosen, it will be essential to ensure that progress is monitored. The lack of adequate monitoring in the past makes it impossible to answer many questions about what has and has not worked in general. But one key general lesson that emerges from all the experience to date is that an adequate regulatory framework is essential for successful reform. Beyond that, solid conclusions on growth, employment, service delivery, and fiscal impact of reform are elusive.
6. Public utilities  are something of a special case, and of growing importance in the debate. They are frequently characterised by natural monopolies, and each field has specific characteristics (and specialised language) that make the debate complex. The importance of the regulatory framework is, however, particularly evident for all public utilities. In general, the key issues are about the access, affordability and quality of each service  - with realistic target levels varying greatly according to the characteristics of a country, particularly its level of development - and quality and efficiency of management.
 Used here in the restricted sense of water, electricity, postal and telecommunications services and transport infrastructure.
 Using these words in their popular sense and not in the technical senses which they have taken on in specialised debate on public services of general interest (e.g. COM(1996)281/03; COM(2000)580
7. To sum up, the objective of this Communication is to propose a more active role for the EC in the debate on and the design of SOE reform in developing countries, with a special emphasis on public utilities, in order to ensure that all the possible options are assessed before choosing one and that reform is implemented in a transparent manner, with adequate sequencing and regulatory frameworks and monitoring mechanisms. This Communication is complementary to the one on private sector development in third countries being developed in parallel.
2. Introduction and Background to the Communication
8. For the last two decades, the failures of SOEs and the resulting burden on fiscal balances, economic efficiency as well as on access to basic services, have been at the centre of structural reform programmes all over the world.  The main responses to these problems in both developed and developing countries have been market liberalisation and changes in incentives and control - often through changed ownership. Though the impact of the introduction of competition and a change in ownership cannot in practice be completely separated, the Communication deals mainly with the latter.
 Privatisation in competitive sectors: the record to date by Sunita Kikeri and John Nellis.
9. For changes in incentives and control, the options ranged from commercialisation, through alternative forms of management such as Public Private Partnerships (PPP), to partial or full privatisation. Details are found in Sections 5-7 below. During the 1980s and 90s there was a marked increase in the volume of divestiture of SOEs in virtually all the regions of the world. Accurate figures are difficult to come by, but conservative estimates put the proceeds of these sales at US$850 billion between 1990 and 1999. In developing countries, US$250 billion were raised during this period and there was a 15% decline in the relative share of the value added of state-owned enterprises in GDP  due to privatisation and the slow growth rate of the public sector.
 Global Development Finance 2001 - The World Bank.
10. The growing number of studies on the reform of SOEs suggest that it is impossible to say that one option or another is unconditionally "the best" . The purpose of this Communication is therefore not to suggest a "preferred" form of SOE reform - privatisation, restructuring, PPP, or other. An unconditional public or private prone approach may not be fully consistent with the EU position expressed in Article 295 of the EC Treaty relating to its internal market, which establishes the EU neutrality as regards the issue of the ownership of any enterprise. In this context, the importance of the private sector as an "engine for growth and source of employment and revenue should not be undermined." (COM(2000) 212 final : The European Community Development Policy)
 Public versus Private Ownership: the current state of the debate by Mary Shirley and Patrick Walsh
11. This Communication proposes a more active involvement of the EC on the issue. There are compelling reasons to do so. First, the impact of SOE reform on the economies of developing countries can affect the achievement of the Millennium Development Goals, to which the EU is firmly committed. This effect could be positive, through reduced fiscal deficits, increased economic growth, improved provision of essential services and building of management capacity, if reform is effective, or negative, through increased unemployment, reduced access to basic services, loss of managerial and administrative know-how and increased corruption if it is not successful .
12. Second, Member States have broad expertise in SOE reform. The EU has unique experience of combining economic liberalisation with common "rules of the game", and an unequalled record of reconciling competitive markets with the provision of services of general interest . By recommending to developing countries examples based on its own experience, the EU shows that it is not applying double standards.
 e.g. in rail, road, and air transport, telecommunications, postal services, public broadcasting, water distribution, sanitation, gas and electricity supply etc.
13. Third, the EC already supports many SOE reform programmes - whether privatisation or other forms of change - directly or, more often, indirectly within the framework of adjustment programmes led by the Bretton Woods Institutions (BWIs).
14. There has been a growing dialogue with the BWIs during the last years on a large number of issues. However, if the EC wants to enhance the discussion with the BWIs on the design of SOEs reforms, there is the need first to strengthen its own analysis and define guidelines. The Communication therefore reviews the reasons behind the build-up of a large public sector in developing countries. It notes the political nature of some of the key choices involved. It highlights the reasons for reforming SOEs, discusses the risks and benefits of different types of reform, and suggests lessons for best practice for both recipient governments and donors. Finally, it makes some practical recommendations for enhanced involvement by the EC in the reform process, notably the need to define and monitor the objectives of reform and to examine all the options (including privatisation).
15. For the sake of clarity, the analysis will put special emphasis on the case of public utilities, but given the range of sectors concerned, will come up with broader conclusions that can be applied more generally (notably to the financial and competitive sectors). . The liberalisation of agriculture and the reform of crop marketing boards have been excluded from the analysis - not because they are unimportant but because they have too many specific characteristics. In contrast, public utilities deserve special attention both because of market structure (most are natural monopolies) and because of the substantial impact that the provision of basic services such as transport, electricity and water has on the competitiveness of the economy and on public health, and thus on poverty reduction. Within its own internal market, the EU recognises the notion of general interest attached to the universal provision of such services, and in a number of areas provides for specific obligations to be met by any public or private provider. While the minimum standards might be very different in developing countries, many of the principles can be valuable in providing policy insights.
3. The Build-Up of Large Public Sectors
16. In many developing countries the build-up of a strong public sector dates back to the colonial period and the first initial years of subsequent independence. With the surge in de-colonisation, political leaders sought greater national control of the economy, especially the activities on which their countries depended most for foreign exchange earnings. This led to the nationalisation of many foreign companies during the 1960s and 70s and to the acquisition of companies in strategic sectors such as petroleum and mining in order to influence corporate policies on production, pricing and investment. Import substitution and accelerated industrialisation were often the justification for the creation of SOEs in manufacturing. Similarly, banks were nationalised or state-owned banks established to contribute to rapid, state-led development or to meet challenges that the private sector was seen as ignoring, e.g. providing rural credit.
17. In developed economies, in particular in Europe, state ownership also experienced a period of popularity just after the second world war and later in the 60s and 70s when it was seen as a way of protecting employment levels (especially in Europe). As recently as in the 1980s there was a wave of nationalisation in France, and a number of calls elsewhere for the state to bail out ailing industries. State ownership was also considered a solution to market failures in sectors of natural monopoly. By entrusting SOEs in these sectors with general interest objectives rather than profits, they were expected to be able to adjust prices, quality and output towards more socially efficient levels.
18. In the 1970s the criticisms of this model of state economic management gained momentum, particularly in developed countries. In developing countries, while some SOEs operated well, many experienced severe problems, creating heavy fiscal burdens. The low productivity of investments in SOEs was a burden on the economy as a whole. Others suffered from a lack of resources for essential investments. Insufficient provision of basic services hindered competitiveness and particularly harmed the poorest. . Essentially political choices about the level of subsidy for basic services became entangled with technical debates about the quality of SOE management.
4. The Need for SOE Reform
19. In financial terms, SOEs in developing countries often performed very poorly. Many firms operated at a loss, entailing heavy opportunity costs to the national economy and a heavy fiscal burden for the State through explicit subsidies, the assumption of operating losses, or re-capitalisation costs made necessary by the continued erosion of the capital structure of enterprises.
20. In non-financial terms, the results of SOEs in developing countries were also frequently disappointing. Their provision of goods and services has often been deficient. Water and electricity supplies, in particular, are often limited and unreliable, hindering the country's competitiveness and harming public health, while very few SOEs in manufacturing are able to compete internationally .
 The politics of patronage in Africa, parastatals, privatization & private enterprise, by Roger Tangri
21. The main factors determining the performance of enterprises, whether private or public sector, can be summarised as incentives and capacity. Among the key influences on the performance of SOEs have been two important aspects of incentives, political interference and unclear objectives (often reflected in the absence of a genuine budget constraint). Similarly there have been two important capacity constraints, skills (including technology) and funds (whether for investment or for working capital), which have also limited their ability to deliver.
22. Politics played an important role in the expansion of the public sector in developing countries and is at the root of these management problems. The level of political interference in SOE management has been high in areas such as investment, tendering, pricing, employment levels and appointments. There are many examples of public utilities being prevented - chiefly by political considerations - from raising charges to cover marginal costs, without an explicit political decision on the level and incidence of such implicit subsidies. A further widespread problem has been extensive overstaffing, often linked to political patronage.
23. Capacity and financial constraints also contributed significantly to management weakness and played a decisive role in the poor performance of many SOEs. The impact of all these problems is further aggravated by the lack of competition, especially in less developed countries where regulation and the rule of law are weaker. The lack of an appropriate legal framework made it harder for these countries to mobilise the foreign and local financial resources that might have allowed them to maintain an appropriate level of investment.
24. This does not mean that SOEs are necessarily mismanaged. Nor are they unique in their problems: the private sector in most developing countries has not always been immune from political interference (although the channels of influence were more direct with SOEs). Private sector enterprises also suffered from a number of governance problems and management deficiencies that led to progressive and generalised losses of competitiveness. However, the management problems of the public sector in developing countries have monopolised the debate. The reform of SOEs became a priority issue after the debt crisis of the 80s. But plus attention tended to focus on the means - usually ownership changes - rather than the ends - reduced fiscal deficits, higher quality or lower cost of outputs, employment generation, even economic growth - and hence the key political issues, such as the level and distribution of subsidies, were all too often lost from view in "technical" debates on the ownership structure of enterprises .
 Exploration of the Impact of Privatisation: Overview and Case Studies by Oxford Analytica
5. SOE Reform Options Within the Public Sector
25. Answers to these problems in both developed and developing countries have usually been sought in market liberalisation and ownership changes. It can be difficult to separate the effects of the two, since price liberalisation or removal of barriers to the entry of private competitors often accompany ownership changes. On the other hand, in several cases, privatisation without liberalisation has shown mixed results for societies. However, this Communication will focus mainly on the question of SOE reform, considering the objectives of reform and the options available to Governments to attain them.
26. The preferred choice of many governments was commercialisation, i.e. subjecting the operation of public firms to market-related management disciplines through a range of different contractual forms, hard budget constraints and where possible competition, with the state retaining ownership. This was intended to capture the benefits of private ownership without the loss of public assets.
Reform of the Water Sector in Burkina Faso: a successful tale of commercialisation.
During the 90s the national water provider ONEA underwent extensive technical and management restructuring, including the separation of rural and urban water supply. This aimed to make the public utility commercially viable and extending both water and sanitation access to low-income areas for an affordable price. The government identified its objectives clearly through a series of policy documents setting time-bound targets for reform. At the end of the process, in 1999, ONEA had achieved 95% cost recovery and 80-85% of the population served with clean water, regardless of their income - better than most of Africa.
Sustainability of service provision and affordability were secured through higher tariffs for large consumers in order to cross-subsidise social tariffs for lifeline consumption, allowing tariff revenue to cover operational and capital costs. Improvement in access to water was mainly obtained by connecting existing small-scale systems to the network and placing commercially- operated kiosks in informal settlements. This allowed for a minimum service affordable to all, and tackled the problem of unregulated mobile vendors. Quality was improved through government investment programs and subsidies for installing sanitation.
There was heavy pressure for privatisation, but the strong political commitment of the Burkina Faso Government and the support of some EU Member States, and the provision of technical and financial assistance by the whole donor community, made reform within the public sector a success.
27. The range of management reform options within the public sector also includes price liberalisation, the removal of barriers to the entry of private competitors, the introduction of performance contracts, and greater autonomy given to the managers of state-owned enterprises to determine prices, hire and fire employees and make investment decisions. Market testing of particular functions, to secure value for money by comparison with private sector options, has also been used to improve internal efficiency, though government capacity to manage these mechanisms is a constraint in many cases.
28. On the other hand, it is often argued that SOEs still remain vulnerable to political interference in decision making and subject to weak financial discipline (partly because of state-owned or politically influenced bank lending) . This interference also applied to staff matters. Many appointments to senior management positions were made on the basis of political connections. SOE reform within the public sector sometimes did little to lessen the burden on public finances, in a context of declining international assistance. In such cases, the privatisation of public firms would depoliticise economic decisions, put an end to the mismanagement and corruption seen as endemic in SOEs, and lead to improvements in productivity, fiscal performance, service delivery and profitability. In fact, bilateral and multilateral donors encouraged governments in developing economies to embark on privatisation for many reasons, with insufficient assessment of the objectives involved.
 Roger Tangri op cit
29. More recently, the peculiar market structure of public utilities (e.g. natural monopoly, large sunk costs) lent itself to another variant of reform: Public Private Partnership - a transfer of operational control to private providers while maintaining (majority) ownership in public hands. Contracts can greatly vary in terms of duration and obligations for both actors and usually seek to introduce competition not within the market but for the market. A more detailed description of PPPs is found in Section 7, with a subsequent discussion of the range of issues raised by the peculiarities of public utilities (Section 8). Nevertheless it is too soon to evaluate the efficiency of these new instruments that are in most cases still evolving.
30. Finally, the recent poor financial, operating and safety performance of some privatised firms (particularly in public utilities) in developed countries has led to experiments with other forms of ownership, such as mutualisation. In this case the assets are vested in a non-profit company funded by bond issues rather than shares, and any profit is then reinvested in the company. The company is usually controlled by relevant stakeholders - service users, the local community, national government, workers in the enterprise, and the commercial private sector are often represented. The advantages are firstly that the firm is run commercially but not owned by private shareholders, and secondly that it can pursue public interest concerns. It is not yet clear whether such an approach can be applied in developing economies, given its reliance on suitable legal, administrative and financial infrastructures, as well as on customers with the ability to pay and on sufficient incentives for effective management and safeguards against capture. In view of its limited track record, it is not covered in further detail in this Communication. However, it seems to merit more exploration than it has so far received.
6. Privatisation , successes and failures
31. Though exact figures are not available, the empirical evidence shows a growing number of privatisations in developing countries. Proceeds increased particularly in the 90s and in 1999 they were four times the level of 1990, following a peak of six times in 1997.
Figure 1. Developing Countries Privatisation Proceeds 1990-2001
>REFERENCE TO A GRAPHIC>
32. In recent years there have been many studies and reports on the outcome of privatisation. From the abundant literature available there emerges a mixed message.
33. In commercial sectors, empirical evidence confirms large efficiency gains from the joint implementation of privatisation and market liberalisation . There is consensus that the profitability, productivity and output of firms improve substantially after reforms. New capital investments are sometimes made to upgrade equipment . Such gains are particularly evident for high and middle-income countries, while low-income economies indicate more mixed results. Privatisation and liberalisation in sub-Saharan Africa and transition economies of the former Soviet Union, for instance, have not always achieved the results expected in terms of efficiency. In these cases, ineffective public administrations, fragile legal systems, weak regulatory capacities and poorly-developed financial markets seem to be at the root of the limited success.
 Birdsall N. and Nellis J. (2002). Winners and Losers: Assessing the distributional impact of privatisation. WP n.6 Center for Global Development; Kikeri S. and Nellis J. (2002) op cit.
 The divestiture of the Polish cement plants in the mid-90s with the contractual obligation for the new owners to invest in their development is a case in point.
Examples of privatisation with a positive impact 
 From Privatisation in Africa by Oliver Campbell White and Anita Bhatia.
Privatisation can have a positive impact on firms financial performance. In Kenya, profits of the Housing Finance Company, privatised in 1992, rose by more than 100% between 1992 and 1993. Privatisation has also enabled firms to diversify product lines and upgrade and rehabilitate facilities. In Uganda, Shell International reported that privatisation allowed the firm to concentrate on core activities and to contract out noncore activities such as canteen services and gardening. Investment far exceeded the requirement under the privatisation deal to invest $10 million over three years: in just two years $13 million was invested in new and rehabilitated filling stations.
Management changes and new investments associated with privatisation have resulted in improvements in performance. In Ghana, the acquisition of a majority stake by Lever International in Unilever Ghana Limited, which increased its shareholding from 45% to 70%, brought about significant changes in the company. It gave the investor the necessary flexibility on matters relating to strategic direction, capital investment, mergers and acquisitions. The capital investment made has enabled the company to reduce unit costs and to report a 50 % increase in production without increasing energy consumption.
34. Financial sector reform deserves further attention, since the creation of a competitive and commercially-oriented banking system is not only important per se, but can enhance the success of SOE reform in all sectors. Bank divestitures have produced substantial savings (e.g. Argentina) and, in transition economies like Estonia, Hungary and Poland, privatisation has led to better outcomes because of bank restructuring early in the transition. State-owned banks tend to hinder stock exchange activity, have often high ratios of administrative costs:revenue and, above all, concentrate their lending on other SOEs, allowing them "soft" budget constraints. However, banking privatisation in Chile (70s), Mexico (90s) and South East Asia (late 90s) was accompanied by severe financial crisis because of weak regulatory frameworks and lack of monitoring and enforcement procedures.
35. The outcome of privatisation has varied widely between countries and sectors, and as a function of the sale process, but there is evidence that it enhanced efficiency more than equity. Many SOEs were overstaffed and retrenchments had a significant impact on income distribution. Where applied, severance schemes were expensive, but without them job losses pushed many ex-employees towards poverty. Evidence for gains in overall economic growth is not certain and recent studies indicate that poor regulation and the absence of broad institutional reforms may be the reasons undermining the positive impact of privatisation on growth . Fiscal proceeds and the end of subsidies to large loss-makers substantially contributed to the reduction of public deficits in developing countries. However, there is still room for improvement since proceeds have often fallen well short of those expected, for reasons including delayed payments, lack of transparency, and limited government capacity to monitor and enforce contracts.
 Sachs J. (2000). The gains of privatisation in transition economies: is change of ownership enough?. CAER Discussion Paper 63, Harvard.
Zambia Consolidated Copper Mines (ZCCM) : the sad saga of a flawed reform process
ZCCM, with its own farms, schools and hospitals, was more than a company, accounting for more than 10% of Zambia's GDP and at one stage 90% of total exports and 70% of foreign currency earnings. After nationalisation in 1972, lack of maintenance and capital re-investment, and wide-spread mismanagement, reduced ZCCM's copper output from a peak of 720,000 t in 1969 to just over 286,000 t in 1999. At the time of its privatisation it was losing US$1-1.5 million a day (9% of GDP in 2000) and had a total indebtedness of US$1 billion.
The Privatisation Act entrusted the Zambia Privatisation Agency with the responsibility for all divestiture, a task it had been performing in recognised transparency and honesty. Instead, the privatisation of ZCCM was carried out in an opaque manner by a special negotiating committee appointed by the President, amid wide-spread allegations of corruption and asset stripping. Attractive bids were declined and the process stalled for several years, while the Asian financial crisis led to a sharp fall in copper prices. Potential buyers either lost interest or sought to renegotiate their deals.
Finally in 2000, after the World Bank withheld a $530 million credit until ZCCM was sold, the divestiture process was completed. Many analysts contended that the deal compared very unfavourably with the June 1997 bid by the Kafue consortium - especially when incentives to Anglo, such as a cut in corporation tax, a 20- year exemption from profit tax and a reduced power tariff, are taken into account. This rather generous treatment did not however, prevent Anglo from deciding to halt its operations in Zambia in 2002 after it had spent over $350 million, leaving the government with the daunting task of finding another operator.
36. The most frequent criticism of privatisation processes is their lack of transparency. Public assets are transferred to private hands by non-competitive methods, fuelling the system of political patronage. This makes for suspicions of corruption. Lack of transparency, and consequent corruption, also often reduce privatisation proceeds. A good example is the Russian "loans for shares" privatisation scheme  between 1995 and 1997. Privatisation of copper mining in Zambia is a further example of how lack of transparency and adverse market trends can lead to serious shortfalls in divestiture proceeds (see box above).
 Sachs J. (2000). The gains of privatisation in transition economies: is change of ownership enough?. CAER Discussion Paper 63, Harvard.
37. Sales revenue can be also be reduced because of the absence of real competition in bidding, or a lack of bargaining power on the part of the government. For important sectors like network industries, the level of investment and technical expertise required of the private provider limit the number of companies able to bid. Cartel arrangements are highly possible. Local businesses play very little part in such SOE reforms. In addition, limited capacity prevents governments from carrying out complex price evaluations of SOEs and properly managing the selection and negotiation stages. This often leads to inadequate prices and unfavourable conditions attached to the sale.
SENELEC - two unsuccessful attempts at privatisation
The privatisation of the Senegalese electricity company, SENELEC, has been attempted on two occasions without success.
On the first occasion in March 1999 a consortium bought 34% of the shares of SENELEC under contractual obligations including service and investments levels. However, 18 months after handing over the operations, the Senegalese government decided to end the contract, leading to an agreement in January 2001 by which the government repurchased all the shares held by the consortium. It wanted a larger investment programme and accused the consortium of not honouring its investment obligations, while the consortium felt that SENELEC had been in a worse condition than they had been told.
A second attempt at privatisation was initiated in July 2001. It was decided that 51% of the shares would be allocated to a "strategic partner" on the basis of the purchase of existing shares and an increase in capital, thereby making available funds for investment as well as increasing the capacity of the company to seek further debt finance. In addition the obligation to carry our out investments would be more clearly defined. This process too was unsuccessful: two bids were received but neither bidder completed a deal. The government is now re-examining the options for privatisation.
38. Privatisation may also be unsuccessful in the absence of an appropriate regulatory framework. This is particularly important for sectors where it is practically impossible to introduce competition, creating a risk of going from a public monopoly to a private one, with serious consequences for social welfare. In developing countries, the absence of proper regulation is often linked with the pressure to sell off SOEs as quickly as possible in order to free resources to reduce the fiscal deficit or to meet donor conditions, as well as to lack of the technical and financial resources it requires. It can also be connected with the fact that the divestiture of an unregulated monopoly is much more profitable. As a result, enforcing ex-post regulations, to ensure competition or to control monopolistic situations, becomes very difficult.
39. Finally many privatisation processes have suffered from inadequate monitoring mechanisms during and after the transfer of public companies to private ownership. Clear contracts, followed by transparent monitoring and enforcement, are crucial to devising solutions, overseeing and reporting on the smooth working of the enterprises themselves, and ensuring that the new private owners honour their commitments and that the state fulfils its obligation vis-à-vis local communities and any workers made redundant.
7. Public-Private Partnerships: an alternative to the outright sale of public assets
40. The difficulties encountered by privatisation operations in developing countries have often reduced the pace of entire programmes and led the donor community to consider alternative options to outright divestiture in their support to public sector reforms. One of these options has been the creation of PPPs, enabling the injection of private resources and expertise into state-owned enterprises.
41. Public-Private Partnerships are basically a contract between a private investor and the state for the provision of a service. There are different types of contracts, of which the table above  gives a few examples. In exchange for the provision of a service, the private promoter receives fees, rights or receipts depending on the type of agreement.
 Source: African Development Bank 2002.
42. It is worth noting that any of these forms can be focused on a range of objectives. Their choice is a political one, of course within a range given by e.g. availability of resources as well as other demands. Thus the emphasis can be on access, affordability and quality of service, or on maximum fiscal gain, or on employment protection, or any other policy priority. What is important is that objectives be specified clearly in the contract, and then compliance be monitored. Ideally the level of subsidy should be transparent, so that the political choices associated with different subsidies - the crucial questions of who gets how much subsidy - can be explicitly addressed.
43. Investment in private infrastructure projects began from a relatively low base in the early 1990s, reached a peak of $123.3 billion in 1997, declined in 1998 and 1999 in response to crises in emerging markets, and resumed their upward trajectory in 2000 (see Figure 2).
>REFERENCE TO A GRAPHIC>
44. By 2000 some 154 developing countries had concluded this kind of deal to develop their infrastructure. Between 1990 and 1998, PPPs were estimated to have had a total value of some US$496 billion: $14 billion representing Africa's share, $237 billion for Latin America and the Caribbean, $147 billion for Eastern Asia and the Pacific and $38 billion for Southern Asia.
45. By sector, telecommunications has dominated investment, with some $292 billion during 1990-2000. Electricity accounted for $197 billion, transport $125 billion, water and sewerage $37 billion, and natural gas transmission and distribution $33 billion.
46. PPPs have come in for increasing political criticism, partly because of the loss of national control implied by foreign multi-national companies operating essential services and partly because of the loss of jobs often associated with reform. The process of contracting has also been open to the same criticisms as that of privatisation, with concerns over transparency and the risk of corruption.
47. Developing the local private/business sector is an important goal of co-operation, particularly for economic growth. The private sector may also have an important role, depending on particular national arrangements and at the behest of the state, in the provision of public services. However, market failures mean that there remains an important role for public action in these areas. Public utilities are, as noted above, perhaps the most common form of PPP. They have a number of special characteristics , whatever the form of ownership under which they operate, and hence merit more in-depth analysis.
 Notably the scale of their fiscal impact, their market structure - frequently a natural monopoly - and their direct impact on poverty. They also form a large part of the SOEs remaining after previous rounds of reform.
8. Public Utilities
48. Public utilities are a sub-set of the broader category of services of general economic interest. This Communication focuses on the subset of greatest significance for development, which always includes water supply and sanitation, basic transport infrastructure and generally electricity and postal services. Telecommunications also fall within this category in almost all developing countries, and in most cases some public transport services are also included. However, the aspirations to universal provision which would apply within the EU are often unattainable in the short to medium term in many developing countries.
49. The European Union takes a neutral stance on the ownership of public utilities, and recognises that an issue of general interest could be attached to the universal provision of basic services by these utilities. The concept of universal service "refers to a set of general interest requirements that guarantee that everyone has access to certain essential services of high quality at prices they can afford, in industries where these services are provided through a network" . In developing countries, tackling issues such as access, affordability and quality is particularly important, given the extremely narrow reach of existing networks, the low potential and immaturity of markets in specific areas (e.g. rural) and sectors (e.g. sanitation) and the poor quality of service, particularly in low-income neighbourhoods. Thus, the impact of reform on poverty is expected to be far-reaching.
 The Services of General Interest in Europe. Communication of the European Commission, OJ C 281, 26.9.1996, p. 3.
50. "Basic needs" has been variously interpreted in development literature and in further different ways in the debate on services of general interest within the EU. Similarly, "access" has been used in the development literature to cover a range of issues differentiated in the public services debate (access, coverage etc.), while "quality" also encompasses issues such as absence or interruption of service as well as the technical usage of the term in public service debate.
Services of general economic interest in the EU
Services of general economic interest (SGEI) are a key element in the European model of society and the new Article 16 of the EC Treaty confirms their place among the shared values of the Union. Public authorities consider the provision of such services (e.g. water and electricity supply, postal service) to the population as necessary.
States are free to choose whether to provide them directly or entrust a private operator with their supply.
When market forces alone cannot ensure a satisfactory provision of these services, public authorities can lay down a number of specific service obligations, dealing mainly with access, affordability and quality issues. The fulfilment of such obligations may trigger the granting of exclusive rights to private or public providers and compensatory payments for the deficits incurred in the provision of those services (within a limited amount in order not to distort trade and competition).
At the heart of this policy, lies the interest of the European citizens who have come to expect high quality services at affordable prices.
51. Concerns have been raised that reforms in state-run public utilities may increase tariffs and connection fees, worsening the conditions of low-income households. Whatever the validity of such concerns, it is important to recognise that in most cases low-income households in developing countries have not been adequately served by state-run public utilities. In many cases, the great majority of poor people does not have access to the existing subsidised or price-controlled services at all, but have to pay much higher prices (often well above even unsubsidised prices to existing customers) for "informal" supplies of dubious quality. Prices below cost recovery and lenient attitudes towards non-payers have also frequently undermined the ability of utilities to expand their networks and maintain an acceptable quality of service. In the water and sewage sector the impact of such situations on the spread of water-related diseases has been widely documented  while electrical fires due to illegal connections are common among low-income households in developing countries.
 e.g. Ghana, see Vaz L. and Jha P. (2001). Note on the Health Impact of Water and Sanitation Services. CMH Working Paper Series n. WG5:23. World Health Organisation.
52. PPPs in public utilities have contributed so far to the lessening of the financial burden of loss-making SOEs and have generated resources for the state. Between 1990 and 1998 PPP investments are estimated to have totalled nearly US$500 billion.
53. Access to water or electricity services has a mixed record in privatised SOEs. Among the positive examples are electricity privatisation in Chile and Guatemala, and water privatisation in Ivory Coast, Bolivia (La Paz) and Philippines (Manila). On the other hand, water privatisation in Guinea, Argentina (Buenos Aires) and Bolivia (Cochabamba) failed to achieve significant increases in connections, mainly because of excessively high connection fees. Improvement in access is particularly evident in telecommunications. High rates of service expansion have been experienced, for instance, in African and Latin American countries. In this case the introduction of competition, made possible by mobile technology, seems to be a decisive factor. In general, significant differences in access between urban and rural or sub-urban areas remain across all sectors.
Bolivia water PPPs : a tale of two cities
In 1997 the Bolivian Government issued a 30-year concession for the private provision of water and sanitation to La Paz and El Alto. Competing bids were judged on the number of new connections offered for a pre-determined tariff: the winner contracted for almost 100% coverage for water and 90% for sewerage. Complementary support through micro-finance and community work helped to achieve these levels, together with substantial cross-subsidisation - domestic fees were set at about half the cost of provision.
In Cochabamba, in contrast, the water company was privatised without competition. The new owners decided to source water differently from the plans made by the public utility. They trebled tariffs, but did nothing to improve quality or extend access to low-income households. Following public protests, the company was re-nationalised and the assets given to the municipality to run.
54. Affordability of services is ultimately a political decision, since a government can decide what level of subsidy it wishes to provide, given other commitments in relation to resources. It has usually been true that tariffs rise after reform, since prices were previously held below cost recovery levels. This happened for example in Peru, Argentina, Bolivia and Guinea. However, cost reductions due to increased efficiency can also lead to lower tariffs for consumers, as happened in the water sector in the Ivory Coast, in electricity in Chile or telecommunications in the United Kingdom. Sectors where it has been possible to introduce some limited competition, such as electricity generating and telecommunications, are the ones to show the best results in terms of affordability.
55. Quality of service generally improves after reform, as witnessed by a cross-section of utilities and countries. Moving from the extremely low levels of quality provided by many SOEs represents an important benefit, especially for sectors where public health is at stake. However, the issue of raising quality still remains after reform. In perfectly competitive markets there may be no need to regulate service quality, but in sectors characterised by a natural monopoly it has been shown that providers tend to set low standards in relation to price (especially in sectors with low price elasticity like water and electricity).
Morocco Telecommunications: a lucrative PPP in the form of a concession
In 1999, for instance, Morocco awarded a US$1.1 billion mobile phone concession for 15 years to Medi Telecom, a consortium of Spain Telefonica, Portugal Telecom and Moroccan investors. The government used the proceeds, which represented half of its total annual capital inflows, to repay some 6% of the country's debts as well as make fresh investments in the public sector. (source: African Development Bank) This example illustrates the positive impact that Foreign Direct Investment can have in improving the quality of services and economic performance of developing countries and in mobilising local and regional financial resources.
56. The magnitude and incidence of these effects depend on three main factors: competition, the regulatory framework, and monitoring and enforcement mechanisms.
57. The introduction of competition has positive results in terms of prices, service expansion and quality (as shown by reform of the telecommunication sector in a number of countries) which benefit the consumer. However, it has to be borne in mind that technological constraints represent a barrier to competition in the majority of infrastructure sectors, particularly electricity distribution, water supply and sewerage.
58. A sound regulatory framework is essential. When competition is not a feasible option, regulation acts as a substitute for the market, taking on some of the functions of a competitor and addressing the monopoly power of the provider. Obviously there is no single approach to the regulation of public utilities, and the outcome of regulation depends on the framework chosen.
59. For the water and electricity sectors, there is evidence that poor people are often the losers unless reform agreements are carefully constructed. However, if poverty-focused conditions are built into an agreement it can have a positive impact on service provision. In the case of water supply in Bolivia (La Paz), service expansion was obtained through connection targets specified in the agreement with a private investor. In both Guatemala and Bolivia the net proceeds of the sales of electricity companies were invested in network expansion. Targeted subsidy schemes in Chile allowed the government to mitigate the impact of tariff increases and improved the attractiveness of low-income customers to the provider. Where such conditions were set the fiscal benefits of reform often decreased because of implementation costs or lower prices offered by new managers. Thus the trade-off between revenue raising for the public sector and consumer protection is difficult to escape. The choice remains ultimately a political one.
60. Monitoring and enforcement mechanisms after reform are also vital. There is little evidence of the extent to which the commitments made by the new managers in terms of investment, labour conditions, or service delivery, have been met. There also seems to be little sign of monitoring of the impact of privatisation or PPPs by governments: most reports are of academic, donor, or civil society origin. The main reasons for this lack of monitoring stem from weak legislative frameworks and a lack of resources for regulatory authorities. The result is that the design of contracts is biased towards the provider, for lack of government bargaining power and enforcement is weak, allowing opportunistic behaviour such as delays in payment and maintenance and low quality standards.
9. Guidelines and key issues for Successful SOE Reform
61. Governments need to define their objectives clearly, in ways that can be monitored transparently. They must decide what is the most important issue for their citizens: service quality or accessibility, fiscal gains or employment generation, taking account of benefits, costs, resources and alternative uses. Trade-offs may occur, for example between productive efficiency and employment levels, revenue raising and consumer protection or quality and prices, as well as between needs and resources. It is therefore crucial for governments to set priorities. Decisions like these, and those on subsidy level and incidence, are fundamentally political decisions and need to be made explicit and transparent if governments are to be held properly accountable.
62. Before deciding on a formula for reform, governments would need to examine all the options, including restructuring within the public sector, PPP, privatisation, and mutualisation, analysing the impact they are likely to have on access, affordability, and quality of services and goods, on public finances and on the country's economy. This assessment should also take into account the employment and social consequences. The reforms should be dealt with in an open and transparent process involving all relevant stakeholders, especially since employment issues are often central to their acceptability.
63.Transparency in both design and implementation is critical to the outcome of reform. There are two main reasons for this: first, a lack of transparency implies higher risk of corruption; and second, poor information during the process can undermine the confidence of public opinion and oblige the government to reverse the reform. Donor support to improve transparency can be crucial and it is important that their intervention be clearly separated from the commercial interests of bidders from donor countries. Detailed and regular information campaigns targeted to the whole population should be an integral part of this, which should help to secure acceptability of the type of reform envisaged and of its consequences.
64. The design of the regulatory framework and monitoring of post-reform performance are key issues, which have an influence on all the impacts analysed so far. There is no one regulatory approach that fits all utilities, but rights to access, affordability and minimum quality of the service should be guaranteed as a public interest, taking account of the availability of resources. The benefits of a proper regulation and control range from direct reduction of poverty to consumer protection, from better quality of service to less corruption. Since developing countries may lack the capacity to tackle such issues successfully, donors should pay careful attention to ensuring that they are addressed. Technical assistance may need to be provided for the design of the legal framework, and selection and negotiation with new managers. Financial assistance may also be needed, at least during the phase immediately following reform.
65. SOE reform in developing countries requires a marked increase in the level of foreign direct investment flows. For this to happen governments must create a conducive climate based on peace, democracy and stability, a stable macroeconomic environment, a transparent and predictable fiscal system, enforcement of the rule of law by an independent judiciary, and respect for human rights. Indeed, Foreign Direct Investment goes primarily to countries that have a good business climate, and developing countries must strive to improve theirs, with the help of the donors. Similar conditions are needed to promote domestic private investment as well. This is true for any form of change of ownership, whether privatisation, PPP, or mutualisation; and indeed remains substantially true for successful commercialisation and reform within the public sector as well.
66. As noted with regard to regulation, sequencing is crucial. An important sequencing issue is the inclusion of the financial sector reform, given the impact, which unreformed banking institutions, can have on the incentives confronting enterprises during the reform process. Furthermore, financial sector reform has other benefits in its own right, notably in reducing the scope for corruption and abuse of office and for increasing economic efficiency overall.
67. Gains in access, affordability, quality, greater efficiency, and improvement in the fiscal position are the expected outcome of well-designed reforms in the medium and long run. However, negative consequences at least in the short-term must not be underestimated, for example in terms of job losses, reduced access to services, or increases in poverty levels with negative consequences for stability and growth. Therefore any reform should include an appropriate social protection strategy, not only in order to mitigate such negative consequences, but more importantly to ensure its overall success. Any isolated reform effort is unlikely to achieve its objective if it not embedded in a consistent package of mutually reinforcing economic employment and social policies.
10. The Role of the EC in the Reform of SOEs in Developing Countries: Recommendations and Implications
68. As one of the main development aid providers, the EC has long been involved in funding and reforming SOEs in developing and transition countries of the former Soviet Union and CEECs. It has done so in several ways. It has, through the EIB and ERBD, provided loans to this sector and, sometimes, refunded severance costs in the course of industrial restructuring. In a number of cases, it has financed the import of spare parts to allow an ailing industry to resume production, and provided technical assistance to establish restructuring programmes.
69. The EC has also supported SOE reform directly in some Mediterranean and Tacis countries. It has also supported them indirectly, as in the ACP region, through macro-economic support programmes. In the latter case, it linked the budgetary aid it provided for structural adjustment programmes supported by BWIs to the implementation of SOE reform processes. Though the EC collaborates constructively with BWIs in many areas, it has not generally been involved in the details of these reforms. However, given the importance of successful SOE reforms for the attainment of the Millennium Development Goals, the EC should now pursue a closer dialogue with the governments and the BWIs on the matter. The policy conclusions above form the basis of this stance
70. It must where appropriate voice its concerns about the shortcomings of current approaches, to the countries concerned and enhance its dialogue with the BWIs in this respect. By using the broad range of expertise existing within its Member States the EC should become a more active partner in trying to design such reforms to take more fully into account its objectives of poverty reduction and enforcement of sound macroeconomic frameworks in developing countries.
Discussions with the developing countries concerned, and with the BWI, should take place prior to any SOE reform which directly or indirectly involves the EC . European Commission Delegations entrusted with increased decentralised responsibilities would need to play an active role in the discussions to be held at country level during PRSP processes or during BWI missions, concentrating on the objectives of reform, identification and assessment of reform options, including sequencing issues, the requirements of post-reform regulation and monitoring and the need to address employment and social implications in a consistent and integrated way. Member States should also be involved in these talks.
71. When a decision is taken on this basis, the EC should mobilise the appropriate development policy instruments to back its implementation. Where a role for the private/business sector (e.g. in privatisation or PPP) has been chosen as the best option, it should include its instruments for private sector development in third countries. It may also be able to help through its macroeconomic support programmes. The former can make financing and investment guarantees available, particularly through the European Investment Bank. Both can help to promote a favourable business climate, essential for attracting foreign and domestic private investment, and assist in building the capacity to implement social protection strategies.
72. On the other hand, this approach should avoid direct EC involvement in the implementation of a SOE reform programme if there has been no assessment of the needs and conditions for its success. Special arrangements should be made to review programmes of support for SOE reform that have already been approved and are being implemented.
73. Where SOE reform is already in hand, the EC should, when ever possible, make an assessment of the conditions for its successful contribution to the overall objective of poverty reduction and its specific concerns over access, affordability, quality, economic growth, employment and fiscal impact. This, obviously, may take time to attain. Meanwhile, if a country falls off track with the BWIs for failing to successfully carry out a SOE reform programme, the Commission should aim to make its own assessment of the specific case and the implications for its continued support - albeit initially relatively informally until a more structured methodology can be developed and resourced.
74. The Commission should enhance its capacity to provide high-quality and timely technical assistance to governments for the key stages of SOE reform. This should cover the dialogue outlined above on the objectives of reform, identification and assessment of reform options, including sequencing issues, and the requirements of post-reform regulatory framework and monitoring, as well as the processes of contract design and the identification of and negotiation with new managers (whether within the public sector or through ownership change).
75. Some additional assistance might be provided in international promotion, to attract investors interested in enterprise take-overs or in building partnerships with the state, where those options are identified as optimal. If governments wish to divest after restructuring, they might benefit from the expertise of high-level civil servants in the Member States who have played a similar role in their own countries.
76. This new commitment should play a part in ensuring that SOE reform helps to deliver the reduction in poverty that is the central goal of development policy.