Report No: ACS15142 Middle East and North Africa Governance Reforms of State-Owned Enterprises (SOEs) Lessons from four case studies (Egypt, Iraq, Morocco and Tunisia) August 2015 GGO23 MIDDLE EAST AND NORTH AFRICA Document of the World Bank Standard Disclaimer: The main authors of this report are Gaël Raballand, Gilles Veuillot, Lydia Habhab, and Phillipe de Meneval. They benefited from the inputs of Fahmi Ben-Abdelkader, Anne Brockmeyer, Ahmed Darwish, Jarl Hansstein, and Magued Osman, from comments and suggestions from Henri Fortin, John Speakman and Maria Vagliasindi and from the guidance of Shantayanan Devarajan, Guenter Heidenhof and Hisham Waly. Contents Chapter 1 Introduction ............................................................................................................................ 12 Defining SOEs, a Legal Perspective ......................................................................................................... 12 The Rationale for SOEs ............................................................................................................................ 14 The Principles of Good Governance of SOEs .......................................................................................... 16 Chapter 2 Overview of SOEs in the MENA Region................................................................................... 21 A Brief History of SOE Reforms in the MENA Region.............................................................................. 21 Where Does MENA Stand Now? Some Stylized Facts on SOEs in the Selected MENA Countries .......... 22 A Brief Look at the SOE Landscape in the Selected MENA Countries ..................................................... 24 Numerous Reforms Attempts in the Region ........................................................................................... 30 Chapter 3 An Assessment of SOE Reforms in the Selected MENA Countries ......................................... 32 Growing Complexity of SOEs................................................................................................................... 32 An Assessment of SOE Reforms in the Selected MENA Countries ......................................................... 33 Chapter 4 The Common Challenges Facing SOEs across the Region and Their Impact........................... 37 Transforming the State’s Role from Management to Ownership .......................................................... 37 SOE Autonomy ........................................................................................................................................ 38 Corporatization and Independence of Boards ........................................................................................ 43 Transparency, Reporting, and Accountability......................................................................................... 45 The Impacts of Poor Governance of SOEs: Corruption, Misuse of Public Funds, and Overemployment ................................................................................................................................................................ 47 Chapter 5 What New Reform Priorities Could Change SOE Governance in the MENA Region? ............. 51 The Limits of Ambitious SOE Governance Reforms in the MENA Region............................................... 52 A Gradual Path toward Success: Lessons from the Corporatization of the OCP in Morocco ................. 54 Reforming SOE Governance through Partnerships with the Private Sector ........................................... 57 Which role for the World Bank on SOEs reforms?.................................................................................. 58 Appendix The Legal Definition of SOEs in Egypt, Iraq, and Tunisia ......................................................... 59 References ................................................................................................................................................... 62 Executive Summary In the early 2000s, governments began to recognize the uneven implementation and mixed success of past privatizations. Meanwhile, the critical voices denouncing the adverse social effects of privatization were gaining traction in a context of rising unemployment and sluggish investment. Privatization efforts began to stall or even see a partial reversal. In the Arab Republic of Egypt, for example, a highly public squabble over the department store Omar Effendi ended in the re-nationalization of the chain.1 The governance of state-owned enterprises (SOEs) is a crucial issue for economic and social development in Middle East and North Africa (MENA) region for four main reasons. First, these enterprises play an important role in key sectors of the region’s economies, including banking, transport, and industry. Second, SOEs have benefited from connections to the ruling elite, at the cost of citizens, as debated in post-revolution Tunisia. Third, as the room for fiscal maneuverability becomes increasingly narrow in many MENA countries such as Tunisia and Morocco, it is important to assess the budgetary risks associated with state-owned enterprises and institutions. Finally, although the participation of the state in the productive sector has developed in many countries in the MENA region using various instruments, new governance models are being developed.2 The main objective of this report is to identify the main governance problems related to SOEs and their main impacts on economic development in the selected countries. This report builds on four background papers, each on a MENA country: Egypt, Iraq, Morocco, and Tunisia. Morocco was selected because it carried out a well-known reform in the early 2000s and has a much more diverse and sophisticated SOE portfolio than most countries in the MENA region. Egypt and Tunisia were selected because they have been lagging behind in reforms, and SOEs have been used to generate employment since the Arab Spring (2011–12). Finally, Iraq was chosen because of its problems related to its post-conflict situation and the fact that SOEs have traditionally been major economic players in this country3. Each background paper presented the same aspects: descriptive statistics on the size of the “sector�, legal definition of SOEs, main governance problems, main reform attempts and their outcomes and examples of governance problems/reforms for some emblematic SOEs. Good governance of SOEs in MENA region has always been a stated goal in countries in which their economic and social roles are important. It has been difficult empirically to identify 1 By matching sample firms (privatized) with control firms (SOEs) over 1994 –98, Omran (2004) demonstrated that privatized firms did not exhibit significant improvements in their performance relative to SOEs and then questioned the benefits of Egyptian privatization during this period. 2 For more details on the growing complexity of different types of SOEs around the world, see Musacchio and Lazarrini (2012). 3 World Bank engagement has also been important regarding SOEs reforms in the four selected countries and provided us useful information. universally the most economically and socially efficient type of governance for SOEs. Because SOEs are likely to remain, in the medium term at least, state-owned in MENA countries, governments are seeking to ensure that they perform as well as possible. Over the last two decades, MENA countries went through a corporatization4 process, along with some privatizations. The main stated objective was usually to reduce subsidies and public funding to SOEs. In several countries of the region, some PPP (public-private partnership) arrangements with private capital were developed. Despite such major reforms, the results have usually been mixed at best, with the remaining governance problems characterized by overemployment (exacerbated since the Arab Spring) and recruitment of cronies, fiscal risks, and pervasive opacity. SOE reforms were congruent with the implementation in the 1990s of new public management (NPM) ideas in countries such as New Zealand, the United Kingdom, and the United States. The main underlying idea of the NPM approach was that, to be more efficient and effective, the public sector should behave much more like the private sector. In fact, SOEs are usually constrained in the following areas: appointment of managers, employment level, wages and benefits for staff, procurement, selling prices, and investment levels (Nabli 2001). The NPM approach was adapted to SOEs at the end of the 1990s (Wettenhall and Thynne 2002). Based on the idea that private-like management of SOEs would improve their efficiency, several tenets were adopted for SOE reforms such as the need to corporatize SOEs (because the approach implied a new governance structure in which the board would serve as the main management institution), the need to give more autonomy to the SOE management and board (to reduce political interference), and the need to have the state act as an owner, looking for profits and efficiency. As noted by Wettenhall and Thynne (2002), “devolution� and “let managers manage� have been the mottoes of the principles of governance in SOE reforms. The guidelines of the Organisation for Economic Co-operation and Development (OECCD) on the governance of SOEs (OECD 2005) identify the main principles of good governance that would lead to better performance (mostly from a profit maximization perspective). Table ES.1 sums up the main expected changes to SOE governance structures if the OECD guidelines were applied in a reform process.5 Table ES.1 Changes Expected from SOE Reforms: OECD Guidelines6 Before After State as a manager 7 , managing day-to-day State as owner (profit maker, contractual 4 Corporatization is the process of transforming state assets into joint stock companies. The state retains majority ownership of the SOE but strives to introduce it to corporate and business management techniques. 5 Rigorous empirical evidence of such impacts is difficult to reveal and is not plentiful. (Vagliasindi 2008a). 6 The report focuses on supply-side aspects of SOEs reforms since demand-side has been usually weak in the region and, even in the OECD guidelines, relations with stakeholders are not necessarily with services users. 7 There may be an overlap between the principle 1 and 2. We kept both principles due to the fact they are distinct principles in the OECD guidelines. business of the SOE, including decisions on relations with SOEs)establishment of production volumes, procurement, and staffing professional monitoring structures of SOEs No autonomy of SOEs (management does not Autonomy of SOEs (relaxed public controls ultimately decide on employment, staff status, based on risk, financial autonomy, independent procurement decisions, investments, selling management responsible for results and liable prices, etc.) to sanctions in case of poor SOE performance) Statutory SOEs but not corporatized, Corporatization in most cases with a board considered part of civil service making the main management decisions No financial reporting, no public disclosure, Financial reporting, public disclosure, opacity transparency Table ES.2 outlines how the four selected countries perform along the four principles defined by the OECD guidelines (OECD 2005): (1) transforming the role of the state from manager to owner; (2) strengthening the autonomy of SOEs; (3) corporatizing SOEs and ensuring the independence of boards; and (4) increasing transparency and accountability. Table ES.2: Main Characteristics of SOEs in Selected MENA Countries, as Derived from OECD Guidelines - Strong department - Weak SOE - Fragmented - Fragmented in charge of monitoring ownership, ownership, monitoring SOEs - No transparency - Major role of - Major role of (DEPP) about the Ministries, Ministries, - Uneven quality constraints on - Poor - Very poor and monitoring of SOEs monitoring of monitoring of performance - Weak contracting SOEs SOEs performance contracts performance - DEPP facing “subsidiarization� of some SOEs and lack of controls - Depends on - Lack of - Strong political - Very limited companies/sectors separation of interference and autonomy, - Strong vehicle for government limited - Very strongly public investments responsibility as autonomy, dependent on (infrastructure) and owner and policy - Increasingly subisidies, therefore highly and regulatory dependent on - Political power, subsidy-dependent responsibilities state subsidies. setting employment for many of them - Strongly numbers for key subsidy-dependent SOEs - Political power, setting employment numbers for key SOEs - Massive - Boards composed - Boards - Boards almost corporatization mainly of civil composed exclusively - Board directors servants mainly of civil composed of civil most often - By law, approval servants servants, politicians and of board decisions - Weak power of - Very weak power civil servants by minister boards for most of boards appointed by the SOEs government or king - Annual aggregate - Last report in report (over 100 2007 pages, available on - No discussion at web) sent to National Assembly Parliament - Financial transparency by law Note: DEPP = Department of Public Enterprises and Privatization. Morocco has achieved some steps on transparency, professionalizing SOEs, monitoring, and even giving more autonomy to some SOE management. By contrast, the three remaining countries, despite some reforms on paper, have not improved much and mainly continue with a governance style of SOEs characterized by strong political interference and opacity. Morocco excepted, none of the selected countries have an explicit and well-articulated policy for State ownership, which is often considered an element of a sound SOE governance environment. Based on the four case studies, the four main messages of the report are the following: 1. Most SOEs of the selected countries are in great need of reform in light of their economic and social weight and the governance issues they are encountering. 2. However, most MENA countries are suffering from a low governance environment that prevents ambitious reforms, and can explain the relatively weak outcomes of past reforms. 3. Golden standards usually fail in MENA and may even be counter-productive in the MENA political environment by creating “false reforms�. 4. There is room for a more modest and progressive approach for SOE reform in MENA, adapted to the specific environment of each country. There is a real disconnect between window dressing attitudes and actual behavior8 toward SOE reforms in the MENA region. As a result, reality is hidden, and reforms do not materialize and 8 Anthropologists and organizational sociologists began to document �institutional mimicry� in the 1970s and 1980s, demonstrating that beyond reforms on paper there is strong tendency for organizations to continue with the prevalent practices (see Bugnicourt 1973). remain largely on paper. Institutional mimicry prevails, which explains why, despite reforms on paper and governance arrangements closer to those of the private sector, an administrative approach to SOE management continues and administrative practices remain largely intact. Indeed, the disconnect between the stated ambitions and the reality of the day-to-day functioning of SOEs remains strong. Organizational sociologists have pointed out that “organizations often arrange their core activities according to accepted models, or templates, in their field� (D’Aunno, Succi, and Alexander 2000) or “tend to model themselves after similar organizations in their field that they perceive to be more legitimate or successful� (DiMaggio et al. 1983). It is not surprising, then, that politicians and bureaucrats strive to copy the OECD model of SOEs because OECD countries are the ones rightly perceived as the most successful ones. Their organizational characteristics are mimicked in the region by undertaking corporatization, placing independent members on boards, strengthening units in charge of monitoring SOEs, strengthening financial audits, introducing more transparency, developing special vehicles with PPP frameworks, and so on. This development has also been promoted by the fact that most donors and international financial institutions (IFIs) have supported the same type of model of SOE reform. However, as noted by Bugnicourt (1973), “mimicry exempts from thinking about possible alternatives� and the likelihood that such reforms will materialize on the ground is extremely low (as we will present in this report) for two main reasons: first, an “organization is highly sensitive to the cultural environment within which it lives� (Seiznick 1996) and, second, “weak competition makes [organizational] change unnecessary� (D’Aunno, Succi, and Alexander 2000). Mimicking private sector practices in the public sector in the MENA region is unlikely to succeed because political interference remains widespread. 9 For example, in most MENA countries, on paper boards of directors are supposed to be the major decision-making bodies of SOEs, but in reality that power remains in the hands of politicians. Therefore, as long as politicians continue to interfere in SOE management to generate employment, subsidize parts of the population directly (or indirectly) to win votes and popular support, or interfere in public procurement, reforms will not have a strong impact on the ground. Moreover, in the MENA region most SOEs operate in a noncompetitive environment and derive part of their revenues from a monopoly situation. Therefore, there is no pressure to innovate, compete, and increase performance.10 That does not mean that the institutions and governance of SOEs should remain as they are but that the governance of SOEs in the MENA region (as in any region of the world) is a deeply political process. Formal SOE reforms across the board in the region (especially in countries 9 This was an underlying message of Benhassine (2009) on the main constraints of private sector growth in MENA. 10 This was also demonstrated empirically for samples of SOEs in China by Su and He (2012) and Kang and Kim (2012). There, SOEs perform more poorly than private companies, all other things being equal. such as Egypt, Iraq, and Tunisia) are unlikely to achieve tangible results, especially if reforms simply consists of transplanting a “universal� governance model of SOEs. The report is structured as the following: the second chapter is an overview of the SOEs reforms in the MENA region, then a detailed assessment of the four case studies is presented; the fourth chapter presents the common challenges facing SOEs and their impact and the final chapter presents the main policy recommendations. Chapter 1 Introduction This chapter begins by defining a state-owned enterprise (SOE) from a legal perspective, then presents briefly what has been and is the rationale for SOEs in the region. It concludes by describing the trends in SOE reform over the last 50 years. Defining SOEs, a Legal Perspective The SOE landscape has become increasingly diverse. There used to be some relatively well-defined criteria, but with the growing complexity of state participation in the economy, there is no longer a uniform definition, and especially because the definition of a SOE has always been country-specific. Nor is there a uniform legal regime. Lack of a Uniform Definition   Lack of a Uniform Legal Regime   Non-uniform Application of Criteria among Countriesand Sometimes within Countries 11 For more details on SOE definitions in the selected countries in the Middle East and North Africa region, see the appendix. Box 1.1 Legal Form, Legal Regime: Key Distinctions between a Statutory SOE and an Incorporated SOE Statutory SOEs. 12 Statutory entities are established by a law or a decree, and they are not incorporated as commercial companies. As a result, statutory entities tend to be legally hybrid entities because their governance framework often borrows features from both the public sector and the private sector. Having a distinct legal personality from the state, they are autonomous from the state and benefit from rules of operation that are somewhat more flexible than purely governmental institutions, although they remain in essence an emanation of the government with a high level of political interference in management. Statutory SOEs typically include research institutions and regulatory or public domain management agencies, although denominations are rarely a reliable indicator of the nature or regime of a SOE. Corporatized SOEs. These SOEs are registered under the legal framework applicable to commercial companies and therefore operate according to rules comparable to those governing the private sector.13 The enterprise will thus generally have capital and shareholders, a board, a management with clear delineation of responsibilities, and control mechanisms. In theory, their main objective becomes making a profit. In practice, however, SOE regimes may differ significantly based on (1) the law by which they abide (often there are several exceptions to corporate law), (2) the extent of the ownership, and (3) the modality of the ownership. The Rationale for SOEs The traditional rationale for state ownership relies on various notions of market failure such as natural monopolies, public goods, externalities, and information asymmetries, combined with imperfect contracting.14 A major problem is that market failures may sometimes be perceived but not real because they may be difficult to measure. In the case of a natural monopoly, 15 economies of scale prevent the achievement of cost minimization unless output is produced by a single monopolistic supplier. This is most common in sectors that require a network infrastructure, such as utilities and transportation, or that require high initial capital investments, such as natural resource extraction. After the Great Depression, 12 Often referred to as public establishments or établissements publics in civil law countries. 13 This is what is generally referred to as “corporatization�that is, transformation into a private business–type of corporation. This term, however, is not used consistently. In some cases, it simply refers to the use of a corporate structure rather than a clear application of private sector rules. 14 For a summary of these arguments, see Christiansen (2013) and OECD (2005). For a more detailed account of the historic thinking on state ownership, see Shleifer (1998). 15 A type of monopoly that exists as a result of the high fixed or start-up costs of operating in a particular industry. It is usually considered for utilities such as power and water because the cost of creating a network to produce electricity or water for all households is very large and can deter investment if there is competition. economists thus advocated the nationalization of railroads and utilities (Simons 1948) and of land, mineral extraction, and telecommunications (Lewis 1949). The thinking was that a monopolistic state-owned enterprise could ensure wider provision of services at lower prices than a private monopoly or a competitive sector. The government’s public service obligationfor example, to connect remote areas of the country to the road network or to deliver mail to all householdswas an additional reason for state ownership of these businesses. Failures in the market provision of public goods also provide a rationale for state intervention. Public goods such as defense and national security services are not excludable, meaning that their consumption cannot be limited to paying individuals. A private company, which can collect returns only from paying customers but not from free riders, would thus underprovide these goods and services. Another market failure that motivates state provision is externalities. In a pure market economy, goods causing negative externalities (e.g., pollution) are overprovided and goods with positive externalities (e.g., health and education) are underprovided. Vaccinations, for example, reduce the risk of infection not only for the vaccinated population but also for those not vaccinated. The social returns thus exceed the private returns to vaccination, and state intervention is desirable to achieve the socially efficient level of consumption. Based on a view of the negative externalities, Meade (1948) advocated the nationalization not only of iron and steel production but also of the chemical industries. Finally, information asymmetries in the form of moral hazard and adverse selection problems may warrant state ownership of businesses such as insurance (Lewis 1949) because only the state has the power to mandate the purchase of insurance and thus guarantee a certain degree of risk heterogeneity in the insured population (Lewis 1949; Akerlof 1970; Stiglitz and Weiss 1981). Grossman and Hart (1986), Hart and Moore (1990), and Hart (1995) developed a theory of incomplete contracts in which contractual incompleteness gives the owner of an asset greater control and bargaining power and thus stronger incentives to improve quality or reduce production costs. Sappington and Stiglitz (1987) showed that market failures can be addressed through state ownership if the government is benevolent. In this case, government ownership avoids the information, contracting, bargaining, and renegotiation costs incurred when providing a contracting service to the private sector.16 In addition to mitigating market failures, governments may value state ownership for strategic reasons. First, governments use state ownership as an industrial policy tool to support nascent industries that may have economies of scale in the long run (the “infant industry� argument), preserve declining sectors (e.g., shipbuilding), prevent takeover by foreign companies (e.g., mineral resources), and accelerate technology transfer. Industrial development in Japan and the Republic of Korea and the post–World War II reconstruction of Europe were largely based on 16 Hart, Shleifer, and Vishny (1997) consider the choice between private and public ownership through the lens of the Grossman-Hart-Moore theory, with an application to prison management. They conclude that government ownership is desirable when there is a strong trade-off between cost reduction and quality enhancement, innovation is unimportant, and competition and the reputational mechanism are weak. state ownership of strategic companies. Second, governments also employ state ownership to achieve broader economic development goals such as supporting economic growth through state-owned banks, guaranteeing employment, and developing remote areas (Burgess and Pande 2003). Third, state ownership is a political economy instrument. It allows the government to control the prices of certain goods and thus helps to distribute subsidies in a context of underdeveloped welfare systems. State enterprises are also used to reward political allies through employment creation, above-market wages, and transfers (Shleifer and Vishny 1994; Shleifer 1998). For example, it was arguably for political reasons that the British government continued to run loss-incurring coal mines for three decades after World War II. The Principles of Good Governance of SOEs According to Nabli (2001), if implemented according to plans, SOE reforms could have major positive impacts not only by reducing fiscal risks by decreasing hidden subsidies, direct transfers, and overstaffing, but also by strengthening competition and developing capital markets. As noted by Megginson and Netter (2001) and Gómez-Ibáñez (2007), SOE reforms in developing countries began in the 1960s because of the poor performance of many of the SOEs. The reform movement sought to strengthen the internal capacity of SOEs. The second wave of reforms in the 1980s emphasized the lack of managerial accountability and autonomy and led mainly to performance contracting and corporatization. The latter implied revising the governance structures of boards of directors, who became the main decision-making institution. These SOE reforms were encapsulated in several documents published in the early 2000s. To enrich the discussion about possible avenues for performance-enhancing SOE reforms, the rest of this chapter presents the main principles of good governance of SOEs with references to the OECD Guidelines on Corporate Governance of State-Owned Enterprises (OECD 2005). This theoretical framework, mainly based on those guidelines, will be useful in the next chapter, which assesses the governance of SOEs in the four selected countries in the Middle East and North Africa (MENA) region. In its guidelines, the Organization for Economic Co-operation and Development (OECD) codified in the area of SOE management policy recommendations to improve the governance of SOEs. This OECD document has been used by several countries undertaking SOE reforms in recent years, and the OECD guidelines seem to have influenced the thinking and design of some SOE reforms in the MENA region. This document is divided into six parts: (1) an effective legal and regulatory framework for state-owned enterprises; (2) the state as an owner; (3) equitable treatment of shareholders; (4) relations with stakeholders; (5) transparency and disclosure 17; and (6) the responsibilities of the boards of state-owned enterprises. In the selected countries in the MENA region (except Morocco), most of the SOEs do not have 17 A specific document on transparency for SOEs was published by OECD (2012). private shareholders (or those with a small minority of shares). Therefore, the third part is marginally relevant. Moreover, because this study focuses more on economic, financial, and governance issues, the fourth part seems to be less interesting for this purpose. Therefore, four main principles have been selected to present governance-related issues linked to SOEs: 1. Transforming the state’s role from management to ownership (part 2 of the OECD guidelines) 2. Increasing SOE autonomy vis-à-vis the state in the areas of staffing, public controls, and procurement (part 1) 3. Increasing corporatization and strengthening the role of boards of directors (part 6) 4. Increasing transparency with financial reporting/disclosure (part 5). Principle 1. Transforming the State’s Role from Management to Ownership State ownership creates two potential problems. First, the state as owner may not pursue pure profit maximization; it may have many more objectives (social or even political). This may result in setting specific objectives for the SOE that divert the day-to-day management of the company away from the profit-maximizing path such as through targeted hiring, subsidization of some public institutions or individuals at noncommercial rates, and pressure on investment levels and locations. The second problem is that the state is not only the owner but also the manager, and the combination of these two roles creates a conflict of interest and thus inefficiency.   Principle 2. Increasing SOE Autonomy vis-à-vis the State In this regard, financial autonomy is especially important because only when financial autonomy is granted and companies are allowed to keep a share of their profits can they actually base decisions on profit maximization, make more rational economic and financial decisions, and take an interest in their performance. When SOEs operate only for the government and remit all their returns to the government, their incentives to increase profits for the government are usually minimal. 18 For more information on the debate, see Vagliasindi (2008b). The centralized model (in an agency or not) has often resulted from the implementation of privatization programs. 19 This can be somewhat difficult to define, but a commonly accepted definition is that such activities produce or deliver specific goods or services that identified individual clients are willing to purchase - and SOEs derive the bulk of their income from such activities. 20 Including, for example, for the management of donor-funded projects. Principle 3. Corporatization and Strengthening the Role of Boards   21 “Corporatization is the act of reorganizing a SOE into a legal entity with corporate structures similar to other companies, including a board of directors, management, and shareholders. The main goal of corporatization is to allow the government to retain ownership but still enable it to run SOEs efficiently and on a more commercial basis like other companies� (Kikeri et al. 2014). 22 Vagliasindi (2008b) has shown that firms with higher proportions of outside directors and smaller boards tend to make better (or at least different) decisions about replacing chief executive officers (CEOs), all other things being equal.   Principle 4. Increasing Transparency with Financial Reporting/Disclosure   Transparency is crucial to limiting the misappropriation or waste of public resources in SOEs and to limiting poor decisions such as in public investments. Better financial reporting and disclosure reduce the information asymmetry between the SOE manager and politicians or civil servants in charge of monitoring SOEs (Gómez-Ibáñez 2007). Finally, and not least, transparency and good financial reporting and disclosure enable better scrutiny by civil society of the allocation of spending in SOEs in order to limit waste of public resources, thereby improving firm performance. Chapter 2 Overview of SOEs in the MENA Region The independence of the Arab Republic of Egypt (1954), Morocco (1955), and Tunisia (1956), as well as the Ba’athist revolution in Iraq (1968), led to the creation of nation -states whose development strategies were built on socialist values, inward-looking economics, and nationalism. The creation of state-owned enterprises (SOEs) thus reflected these values and served purposes beyond simple economics. SOEs were (and continue to be) utilized as a social safety net and to garner the loyalty of citizens. Social objectives such as job creation, subsidies, and income redistribution rivaled economic ones. Nationalization of strategic sectors and natural resources made way for monopolies that allowed the government to use SOEs to wield power and control, putting politics above productivity, profit and loss, and management procedures. Some of the benefits were that public investments were made in infrastructure, health, and education, which boosted industrialization and translated into high growth. The economic growth of the Middle East and North Africa (MENA) region at this time was among the highest in the world, averaging 7.1 percent per year in the 1960s and 5.8 percent on average in the 1970s (World Bank 2004). During the late 1950s and 1960s, governments moved from regulating private sectors to direct control of production through the nationalization of private assets. In response, public sectors became the dominant employers in many MENA states. Wide-scale public provision of social services rapidly expanded state bureaucracies, absorbing many new entrants into the public sector. In a number of states, public commitments to the development of human capital took the form of populist education policies that provided free, universal access to higher education and guaranteed public sector employment to secondary and university graduates. A Brief History of SOE Reforms in the MENA Region 1970s–1980s As early as the 1970s, some of the MENA countries such as Egypt and Tunisia began to reevaluate social policies and take tentative steps toward economic liberalization. By the 1980s, the triggers of an economic crisis were well under way. Worldwide, falling oil prices, high interest rates, and general economic decline ensued. High government debt from inefficiencies, poor performance, high costs, and little to no revenue generation from public enterprises further deteriorated the economies of Egypt, Iraq, Morocco, and Tunisia. Governments struggled to meet their public wage bills, which ranged from 6 percent to almost 20 percent of their gross domestic product (GDP). Inward development policies coupled with weak regulatory environments discouraged private investment, reduced opportunities for trade and industrial development, and increased public debt. Unemployment levels increased as the gap widened between policies and the capacity of governments to sustain their social commitments. Governments thus faced growing pressure—both domestic and international—for economic reform. 1990sPrivatization The early 1990s was marked by sluggish economic growth, weak labor productivity, and high unemployment. GDP growth averaged only about 1 percent over the decade. Rising debt levels and fiscal deficits created budgetary pressures and made the maintenance of loss-making SOEs unsustainable. In this context, several MENA countries, including Egypt and Tunisia, embarked on structural adjustment programs (SAPs) under the guidance of the International Monetary Fund (IMF) and World Bank. 24 Seeking to restore the macroeconomic balance through liberalization and deregulation, these programs advocated SOE privatization to increase government revenue and competition. The results of the privatization process in the 1990s were mixed. In some countries, the proceeds from privatization were significant. In Tunisia, for example, the total sale proceeds of SOEs from 1987 to 1993 amounted to US$134 million.25 Thirty-three public sector enterprises were entirely or partially privatized. In Morocco, privatization proceeds also provided considerable revenue for the state and were used to fund public investment projects under the Hassan II Fund for Social and Economic Development.26 However, the privatization process also encountered significant popular resistance, leading to large demonstrations and strikes, and faced bureaucratic and administrative challenges. In Egypt, for example, the process of evaluating SOEs and tendering lacked transparency and consistency. The combination of weak institutions and widespread corruption led to a preferential treatment of those connected in business, who disproportionately benefited from privatization. In Iraq, only limited privatization efforts were undertaken in 1987, after the cease-fire with Iran, and the size of the state portfolio did not shrink further in the 1990s. Where Does MENA Stand Now? Some Stylized Facts on SOEs in the Selected MENA Countries27 It is always difficult to generalize about SOEs because there are country and sector specificities. However, we sought to create some macro proxies based on employment and losses data. Figures 2.1 to 2.4 mainly portray two categories of countries (of the four selected) in reflecting SOE characteristics. The first category consists of Egypt, Iraq, and Tunisia, and in the other is Morocco. In the first category, SOEs are large companies (over 1,400 employees on average), and they remain relatively large employers (employing from over 1.5 percent up to 3.2 percent of countries’ populations aged 15–64). They are also major loss makers, with annual transfers equal to up to 6 percent of GDP. Morocco has a multitude of SOEs, over 700, 28 with numerous small subsidiaries, but overall the sector is indebted, mainly because public investment is largely carried by the country’s SOEs. The sector does, however, generate a large cash flow (over $5 billion in 2011). Figures 2.1 to 2.4 are based on government sources. Data are for 2011, or 2007 in the case of Tunisia. 24 The IMF recommended privatization and SAPs for Egypt and Tunisia for instance. 25 All dollar amounts are U.S. dollars unless otherwise indicated. 26 The Hassan II Fund for Social and Economic Development is a public financial institution created in 2000. It grants financial assistance for investment projects in some industrial sectors (e.g., to acquire professional buildings or new equipment). 27 For more details on some case studies, see OECD (2012). 28 This high figure largely stems from a much wider definition of SOEs (discussed in chapter 1). Figure 2.1 Number of SOEs per Million Inhabitants: Selected MENA Countries Source: Authors’ calculations. Note: Figures for Egypt, Iraq, and Morocco are for 2011; for Tunisia, 2007. Figure 2.2 Average Size of SOEs (Number of Employees), Selected MENA Countries Source: Authors’ calculations. Note: Figures for Egypt, Iraq, and Morocco are for 2011; for Tunisia, 2007. Figure 2.3 Share of Employment in SOEs as a Share of Population (Aged 15–64 Years)(in percentage): Selected MENA Countries Source: Authors’ calculations. Note: Figures for Egypt, Iraq, and Morocco are for 2011; for Tunisia, 2007. Figure 2.4 Share of Total SOE Losses as a Share of GDP: Selected MENA Countries Source: Authors’ calculations. Note: Figures for Egypt, Iraq, and Morocco are for 2011; for Tunisia, 2007. A Brief Look at the SOE Landscape in the Selected MENA Countries29 29 More information is given on Morocco since it is the country were, by far, among the selected countries, transparency on SOEs and SOEs reforms is the most widespread. Egypt In 1952 Egypt’s private sector accounted for 76 percent of economic investment. Following the nationalization plans carried out by President Gamal Abdel Nasser, this percentage drastically shifted within a few decades to government investment accounting for over 80 percent of economic investment (Carana Corporation 2002). This figure included all banking, insurance, foreign trade, medium and heavy industry air transport, and public utilities, as well as many retail stores, newspapers, maritime transport, construction companies, and large infrastructure assets. Most laws governing the SOE sector in the 1970s, 1980s, and early 1990s focused on the organization and reorganization of the sector, yielding little impact on overall company performance, management, or efficiency. Because of the worsening macroeconomic condition of the Egyptian economy, external debt burden, and inflation, privatization was legalized through Law 230/1989 and was part of the recommendations of the Economic Reform and Structural Adjustment Program (ERSAP) in 1991 between the government of Egypt and the International Monetary Fund. Actual privatization of SOEs began slowly. Between 1990 and 1993, only 11 SOEs were liquidated, but over the next 15 years over 400 companies were privatized. In Egypt, there are now 52 economic authorities, 102 service authorities, and 146 state-owned enterprises that are affiliated with nine holding companies. 30 Service authorities depend on recapitalization or operate on a subsidized business model. SOEs either are under specific ministries or are companies 100 percent controlled by the state and operating under the authority of line ministries. These are mainly companies considered “strategic� in sectors such as electricity, ing, aviation, banks, housing, and petroleum. They also may be under the Ministry of Investment, a category that includes most agricultural, textile, chemical industry, mining industry, transport, construction, tourism, pharmaceutical, and food processing holdings. Iraq Since the 1968 Ba’athist revolution, Iraq’s economy has been heavily centralized and dominated by SOEs. Substantial income from the oil boom of the 1970s, the prohibition of non-Arab foreign direct investments, and very few “assistance� efforts contributed to a large state role in controlling and directing production and distributive activities, diminishing the private sector and increasing the importance of SOEs in the economy. A short-lived wave of privatization took place in the late 1980s and early 1990s under the auspices of restructuring and streamlining enterprises as well as trying to control bloated SOE employment. About 40 SOEs were privatized in the cement, construction materials, food 30 The number of holding companies fell from 27 in 1991 to 9 in 2012, and the number of affiliated companies fell from 314 to 146 during the same period. 31 For more information on the legal regime, seethe appendix to this report. canning, and distribution sectors, among others. However, in 1992 the state reversed its decision, bringing most of them back under state control. Iraq became essentially a closed economy and suffered further breakdown and isolation during the years of the United Nations embargo (1991–2003) in which SOEs were denied raw and intermediate materials, leading to significant production decreases as well as the spread of a large informal economy. After the fall of the Saddam regime in 2003, the Coalition Provisional Authority (CPA) 32 pursued a privatization effort and began to shut down some SOEs. This effort caused significant problems, including job losses for well over half a million Iraqis in less than six weeks.33 The U.S. Department of Defense later acknowledged the importance of SOEs to Iraq’s economy and created the Task Force on Business Stabilization Operations, whose objective was to restart many of the SOEs that the CPA had closed. According to the latest estimates, Iraq’s SOEs number about 170 (down from 192 in 2003–2004 and 176 in 2010–2011). They are distributed among various ministries. The Ministry of Industry and Minerals holds the largest share (71), followed by the Ministry of Oil (16), Finance (13), and Construction and Housing (11). SOEs are organized in various holdings, regional branches, and factories. Tunisia In Tunisia, state-owned enterprises have had a major impact on the economy. At independence, the Tunisian government inherited the control of key sectors of the economy such as mining, transportation, and banking. At the time, state-owned enterprises and institutions were regarded as essential to developing projects whose interests went beyond those of the private sector such as phosphate mining, oil exploration, cement production, and oil refining. In addition, these enterprises were quickly assigned socioeconomic objectives, such as employment and income redistribution, for the benefit of the community, and the sector thus grew. In the mid-1980s, SOEs represented 30 percent of jobs in the formal sector, distributed among some 190 SOEs. However, public policies dealing with the creation of state-owned enterprises have “lacked any comprehensive vision [because] they have been created in an ad-hoc manner over time, mainly in response to economic and social pressures� (Ben Letaïef 1998, 70). Transportation, industry, and banking are the three main sectors with the highest number of state-owned enterprises. Because of poor performance, high costs, and structural adjustment programs, a wave of privatizations took place in the late 1980s and early 1990s.34 Morocco After two decades of privatization, the Moroccan government continues to maintain substantial equity shares in a large number of major Moroccan firms. Among the top 10 largest Moroccan 32 The CPA was established as a transitional government formed after the ousting of the Saddam Hussein regime in 2003. Citing UN Security Council Resolution 1483 (2003), and the laws of war, the CPA vested itself with executive, legislative, and judicial authority over the Iraqi government from the CPA's inception on April 21, 2003, until its dissolution on June 28, 2004. 33 http://cco.dodlive.mil/files/2014/02/14_Prism_160-164_Bowen.pdf. 34 For more details on the privatization process, see Saghir (1993). companies in 2012, four were controlled entirely or partially by the Moroccan state. It is the controlling shareholder (with 96 percent of total shares) in the Office Chérifien des Phosphates (OCP), the largest exporter with almost 30 percent of the country’s total exports. The Moroccan state is also a major shareholder in three listed companies 35 that are among the 10 largest capitalizations in the national stock market. In December 2012, these companies accounted for almost 30 percent of the total market capitalization of the Casablanca stock exchange, the second largest stock market in Africa. Public investment is mainly carried out by the SOEs. Investments undertaken by Moroccan SOEs quadrupled from 2001 to 2011, growing at a 14 percent annual average rate. These investments, covering all sectors of economic and social activity, constituted 8.6 percent of GDP in 2011 versus 4.6 percent in 2001. Over the last decade, Morocco has undertaken an extensive reform program aimed at modernizing its legal and regulatory framework and adapting its institutions to new dynamics, particularly those driven by privatization and liberalization. These trends have led to profound changes in both the private and SOE sectors. The corporatization of public operators and the development of state shareholdings in limited liability companies have drastically changed the SOE sector and have had far-reaching institutional implications. Morocco has developed some new forms of state participation in the economy. Several companies have been established for the purpose of acquiring new participation in the economy a minority or majority stake in SOEs or in private businesses whether or not they are publicly listed. Several companies have also been established to promote investment in a sector (e.g., tourism, public housing), and they may play an active role in private financing mechanisms. Indeed, the growth of both domestic and foreign investment has become a stated objective of the Moroccan government. Sector strategies have identified specific investment needs, often based on the identification of specific projects. Several agencies have since been put in place to facilitate the implementation of sector strategies and their investment objectives. The roles of these agencies vary, depending on the needs of each sector. Table 2.1 summarizes these various roles. One role is project engineeringfor example, Agence pour le Développement Agricole (ADA) identifies agricultural projects in all regions and markets them to investors, and Société Marocaine d’Ingénierie Touristique (SMIT) is in charge of engineering turnkey tourism projects aligned with the objectives of the tourism development strategy, Plan 2020. Another role is sector coordination. Some agencies are in charge of coordinating major transactions relating to a sector. For example, the ADA is responsible for coordinating and managing the leasing of state-owned land. The investment promotion agency Agence Marocaine de Développement des Investissements (AMDI) primarily focuses on promoting investment in the industrial sector, whereas investment opportunities in other sectors tend to be promoted by the specialized agencies in the area such as ADA and the Fonds Marocain de Développement du Tourisme (FMDT)). However, AMDI cannot itself contribute to the structuring or financing of projects. Yet another role is participating in sector-specific investment. Some agencies such as Société d’Investissements Energétiques (SIE), SMIT, and FMDT are mandated to invest capital 35 The Moroccan state owns 30 percent of the total issued shares of Maroc Telecom (IAM), 25 percent of Banque Centrale Populaire (BCP), and 79 percent of Compagnie Générale Immobilière (CGI). IAM is the largest capitalization in the national stock exchange market with almost 20 percent of the total capitalization. in large projects they develop. Finally, the sector-related agencies provide financing for major capital investments supporting the implementation of sector strategies. For example, the Fonds Hassan II pour le Développement Ecoomique et Social (FDSHII) and Caisse de Dépôts et de Gestion (CDG) play the role of developers, sponsors, or shareholders in many large projects, particularly in infrastructure and territorial development. Table 2.1 Public Entities Involved in Investment Support and Implementation, Morocco SOE as strategic SOE as project implementer of sectoral strategies financer (loans, SOE as special project (project engineering) equity, or guarantees) SOE as project sponsor vehicle CDGa Al Omrane AAVBR Housing and MedZ Marchica Med territorial SONARGES development Agriculture ADAb Credit Agricole and agribusiness Fonds Special ADM TMSA Transport, Routier ONCF infrastructure, SNTL and logistics ONDA Marsa Maroc ADEREE SIE ONEE (renewables) OCP Energy, water, MASENb ( solar MASENb and mining energy) ONHYM (oil and gas exploration) AMDI CDG Industry and FH2 new technologies SMITb FMDT SMIT Tourism Source: Authors’ representation Note: AAVBR = Agence pour L'Aménagement de La Vallée du Bou Regreg ; ADA = Agence pour le Développement Agricole; ADEREE = Agence pour le Développement des Energies Renouvelables et de l'Efficacité Energétique; ADM = Société Nationale des Autoroutes du Maroc ; AMDI = Agence Marocaine de Développement des Investissements; CDG = Caisse de Dépôts et de Gestion; FMDT = Fonds Marocain de Développement du Tourisme; MASEN = Moroccan Agency For Solar Energy; ONCF = Office National des Chemins de Fer ; ONDA = Office National Des Aéroports; SIE = Société d’Investissements Energétiques ; SMIT = Société Marocaine d’Ingénierie Touristique; SNTL = Société Nationale des Transports et de la Logistique; SONARGES = Société Nationale de Réalisation et de Gestion des Stades . a. Invests mostly in tourism, housing, and urban and territorial development. b. Institution in charge of strategy implementation generally but also promotes its own projects. The development of public-private partnerships (PPPs) for infrastructure or utilities, which often involves complex legal structures (e.g., that distinguish between asset holding companies and operating companies or that specify the use of special-purpose vehicles under project finance schemes) is yet another example. The recent PPP trend may accelerate with the adoption of the draft PPP law, which would provide a uniform framework for PPPs and therefore challenge the traditional definition of SOEs and raise new governance questions. Numerous Reforms Attempts in the Region In Egypt, Law 203 (1991) corporatized many SOEs. Egypt has adopted some SOE laws of member countries of the Organisation for Economic Co-operation and Development (OECD) and adjusted them to suit its legal context. Another wave of reforms occurred during the period 2004–11 when the Egyptian cabinet took measures to reduce SOE debt from LE 33 billion ($5.6 billion) to LE 9 billion ($1.5 billion) and sought to be more transparent on subsidies and transfers. The same period also saw the establishment and empowerment of regulatory bodies in all sectors. One of the most notable reforms during that period was lifting all restrictions on foreign ownership of privatized SOEs. However, some strategic and national security–related sectors did not allow more than 49 percent ownership by foreigners. Finally, the government adopted a PPP strategy in 2006 with the establishment of a PPP Central Unit in charge of planning and managing PPP projects. In Iraq, action plans and strategies have called for reforms as well as in Iraq’s cooperation with the international donor community. Examples are the International Compact with Iraq, 2007; Task Force for Economic Reforms (TFER), 2010; National Development Strategy, 2010; and Governance Action Plan (“supporting ongoing SOE reform and PPP�), 2013. However, the reform strategies, which focused on corporatizing SOEs, were characterized by a lack of coordination, recurrent neglect of gaps in local absorption and institutional capacity, and absence of execution. In Tunisia, state-owned enterprises are legally defined in Law 89-9 of February 1989. This document was amended in 1996 and 2001, but it has not been substantially modified. The legal framework for SOE accounting was established by Law 89-9, supplemented by the 1997 decree on external audits and, for some SOEs, by corporate law. Over the last decade, Morocco has undertaken an extensive reform program aimed at modernizing its legal and regulatory framework and adapting its institutions to new dynamics, particularly those driven by privatization and liberalization. Contrary to the efforts of other countries, these reforms have led to profound changes in both the private and SOE sectors. The corporatization of public operators and the development of state shareholdings in limited liability companies have drastically changed the SOE sector. In particular, the Moroccan government has taken measures that define a new legislative framework governing SOEs. Adopted in 2003, Law 69-00 on the financial control of SOEs is the cornerstone of corporate governance systems within the SOE sector. Creation of the Department of Public Enterprises and Privatization (DEPP), which was designed to play the role of the state as shareholder, was a leading measure in the monitoring and governance process of SOEs. The contractualization of relations between the state and SOEs is also a flagship program of the Moroccan government, established by Law 69-00. In 2008 Morocco adopted a Both the code of 2012 and a proposed revision to Law 69-00 call for a systemic conclusion of such programmatic contracts. In 2013 a directive by the Prime Minister’s Office also called for the preparation of multiple annual plans to later be converted into programmatic contracts. A guide has also been issued by DEPP on the preparation of such contracts. 36 Though not binding, those codes put forth useful recommendations along the lines of the OECD principles of 2005. The role of the state is three-pronged: (1) as a strategic leader in order to define orientations and policies likely to create the conditions for economic and social progress and regional development (the main instrument to this end is programmatic contracts); (2) as a controller as per the provisions of Law 69-00; and (3) as a shareholder with the main objective of promoting an open economy, maximizing the contributions of SOEs to growth, improving the valorization of state participation, and giving SOEs the incentive to commit to social goals and defend the country’s interest. Such objectives should be transparent, including through an annual report on SOEs submitted to Parliament. Other recommendations include a transparent nomination process for SOE boards with due consideration of professional ability (technical, economic, and financial) and nomination of external, independent directors; the promotion of fair competition; and the adoption of dividend policies to allow for both remuneration of the state and development of the SOE. It is also suggested that an annual report on governance and compliance with the Code of Good Governance be prepared. Chapter 3 An Assessment of SOE Reforms in the Selected MENA Countries Before presenting in more detail the various weaknesses of state-owned enterprises (SOEs) and SOE-state relations, this chapter describes the growing complexity of the SOE landscape (especially in Morocco), which has important consequences for the transparency and management of SOEs. Growing Complexity of SOEs The term SOE refers to neither a unified legal regime nor a uniform reality. In recent years, with the development of state capitalism and the resurgence of SOEs in the Middle East and North Africa (MENA) region, the SOE landscape has grown more complex.37 Legal forms (statutory SOEs or corporations), legal regimes (subject to public procurement rules or not, to civil service rules or not), the extent of public ownership, and the modality of ownership vary drastically. This situation may be understandable in view of the variety of SOEs and country specifics, but it can create confusion when trying to summarize suggestions for better governance and may lead to the implementation of ill-suited constraints for some companies. The following sections illustrate this phenomenon in the MENA region based on three criteria commonly used: extent of ownership, modality of ownership, and competition. Extent of Ownership In some countries, state enterprises are defined as 100 percent state-owned enterprises (enterprises with state participation of less than 100 percent can also be found in some countries but are called something other than state enterprises). Partial ownership (and therefore partial private participation) is also possible. It is also possible that shares of the public corporation will be offered publicly on stock exchanges. Another form of mixed public-private participation is when a company is established jointly by a public party and one or several private partners (joint ventures), or when a public institution or a SOE acquires participation in a private corporation. Modality of Ownership Ownership can be either direct or indirect. Directly owned SOEs establish subsidiaries, or they can participate in other enterprises. Indirect ownership is a method of participation that has expanded over recent years. Competition Some SOEs operate in a competitive environment, vying with privately owned corporations offering similar goods or services. Other SOEs might benefit from a monopoly situation (guaranteed by law or de facto). The decision to have a SOE operate within a monopoly situation 37 Musacchio and Lazzarini (2012) present the current trends on a growing multiplicity of various forms of state capitalism. touches on a wider economic governance debate that remains relevant, including as part of the new forms of state capitalism. This multitude of criteria makes it difficult to come up with a simple classification of SOEs, even within a given country. For example, Morocco has a sophisticated SOE landscape in which more than 700 companies have state participation. In parallel with the significant efforts to privatize a number of classical SOEs, the number of institutions controlled by the state has reportedly increased over the last decade, mainly in the form of subsidiaries (with at least 50 percent of shares) established by public corporations or statutory SOEs such as Caisse de Dépôts et de Gestion (CDG), a financial institution that manages significant amounts of regulated savings. Overall, it may be more difficult to define a SOE in Morocco than in other countries because of the multiplicity of legal texts governing state participation in the economy (see box 2.1 for the legal definition of a SOE in Morocco). Box 3.1 The Legal Definition of a SOE in Morocco An Assessment of SOE Reforms in the Selected MENA Countries principles for SOE reform developed in chapter 1 using the OECD guidelines: (1) transform the state’s role from management to ownership; (2) increase SOE autonomy vis-à-vis the state; (3) pursue 38 For such enterprises, the law says that control mechanisms will be provided for under the concession agreement. corporatization and strengthen the role of boards; and (4) increase transparency with financial reporting/disclosure. Based on these four principles, 10 criteria have been selected, using worldwide experiences with SOE reforms (such as Gómez-Ibáñez 2007 and Vagliasindi 2008b) and the OECD guidelines to describe concretely the four main principles. This classification obviously generalizes excessively the situation for some SOEs in some countries, which may be in a better situation than most other SOEs. Therefore, some criteria may appear to be “partially� met. This typology could prove useful in representing as concretely as possible SOE reforms and their implementation in the selected countries:  The role of the state as owner is usually carried out through a permanent structure monitoring SOEs (criterion 1), frequent reporting on SOEs (criterion 2), performance contracting between the state and SOEs (criterion 3), and monitoring of contracts (criterion 4).  SOE autonomy is difficult to measure, especially the absence of strong political interference in SOE management. The difference with state ownership may also be problematic because when the state plays an ownership role, it usually gives more autonomy to SOEs. For that reason, we use some proxies of SOE autonomy and reduced political interference: the extent of interference in SOE employment (criterion 5) and the frequency of recapitalization (criterion 6).  For corporatization and the power of boards, the presence of independent board members is usually considered as a proxy for SOE reform (criterion 7) as well as the fact that boards are not legally subject to a minister’s approvals (criterion 8).  Finally, to assess transparency of SOE management it is crucial that financial statements be prepared, audited, and published regularly without delays (criterion 9) and reports on SOEs published annually (criteria 10). The subsequent assessments were based on background papers that included legal and document reviews and interviews with SOEs manager and key stakeholders. Table 3.1 Comparison of 10 Criteria for SOE Reforms in Selected MENA Countries: Four OECD Principles for SOE Reform Egypt, Arab Iraq Morocco Tunisia Rep. 1. State as owner (1) Central permanent No No Yes Yes structure monitoring SOEs (2) Frequent reporting on Partially No Yes No SOEs (3) Performance contracting Partially No Yes Yes between state and SOEs No No Yes No (4) Monitoring of contracts 2. SOE autonomy vis-à-vis the state (5) Did not increase No No No No employment substantially in the last two years (6) No frequent No No Partially No recapitalizations or cross-subsidies 3. Corporatization and boards (7) Independent board No No No No (except members banks) (8) Board decisions not Yes No Yes No legally subject to minister’s approval 4. Transparency and disclosure (9) Financial statements No No Partially No published regularly without delays (10) Reports on SOEs Partially No Yes No. Published published annually in Sept. 2014 2.5/10 0/10 7/10 3/10 Note : partially gives ½ point. Based on the table 2.1, it appears that Egypt and Iraq are in a similarly difficult position in which what is considered to be good practice for governance of SOEs has not been undertaken in the areas of SOE monitoring, transparency, or boards, for example. In its study of the Egyptian business climate in 2010, the OECD reached a number of conclusions that signal why Egypt’s SOEs may face some problems. The enforcement of a legal framework when assessed against OECD principles was at an implementation rate of 46 percent. Some reforms were undertaken in the early 2000s, but they remain largely on paper and are not being fully implemented. However, progress is currently being made on financial disclosure in the broader public sector in Egypt. Ministry of Finance Circular No. 3 of 2012 mandates the disclosure and publication on the ministry’s website of key budgetary and financial information such as approved annual disaggregated budgets, financial statements, and disaggregated final accounts for 46 of the 52 economic authorities and more than 160 public sector companies (both individual and holding), including petroleum, electricity, gas, water, tourism, chemical, agriculture, mining, food processing, and transport. If implemented, this mandate should tremendously increase transparency in Egypt regarding SOE financial performance. Tunisia is in a relatively better position, but it lacks some basic reforms in transparency, SOE autonomy, and board composition and functioning. Among the four selected countries, only Morocco has undertaken some major reforms and implemented them. However, some progress remains to be made on political interference and the autonomy bestowed on SOE management. The issue of political interference is the core problem in SOE governance in the region. In countries such as Egypt, Iraq, and Tunisia, politicians usually interfere in SOE management, creating serious constraints in many areas (appointments, employment conditions, pricing of goods or services, procurement, or investment decisions). This situation is costly for SOEs, and therefore controls are usually multiplied and the accountability of managers is reduced, all of which leads to frequent recapitalization and higher subsidies. Chapter 4 The Common Challenges Facing SOEs across the Region and Their Impact This chapter describes the current reforms of state-owned enterprises (SOEs) in the selected Middle East and North Africa (MENA) countries. Despite the more advanced reforms in Morocco, SOEs suffer from strong constraints and the weaknesses common in the selected countries:  States intrinsically continue to manage and monitor SOEs without permanent and well-staffed structure.  Most SOEs suffer from limited autonomy from the state.  The powers of boards of directors remain limited despite corporatization and ample board mandates on paper.  Transparency and disclosure of information remain limited. Transforming the State’s Role from Management to Ownership With the exception of Morocco, the state continues to manage and not act as owner in the selected MENA countries. On the one hand, at best there is a central permanent structure to monitor SOEs (an example of this structure is the Department of Public Enterprises and Privatization, DEPP, in Morocco see box 4.1). In most cases, however, various structures are in charge of monitoring SOEs, and yet reporting on SOEs is highly erratic, and performance contracts are mainly rhetorical exercises. As explained in the report by the Organization for Economic Co-operation and Development (OECD) on SOEs in the MENA region: “ambiguous objectives and fragmented oversight can make SOEs easy prey for non-commercial and sometimes overtly political operatives� (OECD 2013, 24). For example, in Tunisia three structures (under the Prime Minister’s Office and the Ministry of Finance) are in charge of (theoretically) monitoring SOEs. Overlapping mandates and fragmentation of oversight weaken state ownership. On the other hand, the state usually continues to manage SOEs, sometimes on a daily basis with all the problems it creates. Over the last decade, in efforts to improve the performance of SOEs using the model inherited from the new public management approach to public service, performance contracts were prepared and signed in some countries such as Tunisia or Morocco. The original idea was to let SOE managers have more leeway when leading companies in exchange for generating revenues and services for the state as a shareholder. However, the expected impact seems to have been much higher than the actual impact. Indeed, several ministers were not committed to these contracts, or they expected the state to refuse to comply with the contracts and therefore did not sign them. Moreover, most contracts in Tunisia did not have quantitative indicators with baselines, and no sanction mechanisms were defined. Therefore, the contracts were more a declaration of intention without any binding effect, which explains why many heads of SOEs assumed this process was simply a rhetorical one. Indeed, most contracts were discussed very briefly at the board level, perhaps reflecting the fact that political interference in the management of SOEs remains strong. As noted by Gómez-Ibáñez (2007), achieving positive results from performance contracts is usually difficult because of the information asymmetry between ministers and SOE managers and then complexity of negotiating them that usually leads to easily achievable targets. Box.4.1 The Example of DEPP in Morocco Law 69-00 provides for the possibility of programmatic contracts ( contrats de programme) that would be concluded between the state and SOEs. Those contracts, which are signed on behalf of the state by both the Ministry of Finance and the sectoral line ministry, define over a period of several years mutual undertakings as well as technical, economic, and financial targets, together with the means to attain them and the modalities to monitor implementation. Contracts can also be concluded with subsidiaries, either directly by the state or by the SOE holding the participation. SOE Autonomy In practice, SOEs usually suffer from political interference that takes various forms: appointing politically affiliated managers, setting employment levels and forcing managers or boards to recruit staff, placing pressure on managers to grant some public contracts to some companies and utilities, and interfering in investment policies. In this context, boards are largely dysfunctional and the important decisions bypass them. Despite wide mandates on paper, they are mainly rubber stamps. The following subsections present the key constraints on SOEs related to employment and boards. The Constraints on Employment and Human Resource Management In many countries worldwide, SOEs are used as political tools of employment for social purposes, resulting in both overemployment and politically motivated recruitments. As explained by the OECD, “maintaining employment through SOEs is not an uncommon strategy across the region, regardless whether this employment is productive or not� (OECD 2013, 60). The poor organization of SOEs is also a factor that contributes to overemployment or inadequate employment, especially the lack of a properly organized human resources department. “Employing enterprise staff under the public service systems of owning states was soon recognized as a primitive arrangement; mostly, the corporations were separated from those systems and thus enabled to develop staffing arrangements appropriate to their particular functions� (Wettenhall and Thynne 2002). In Tunisia, the number of non-administrative public establishments (Etablissements publics non-administratifs, EPNAs) continues to grow because they offer the management teams of these institutions more latitude, which can be explained in part by the greater wage flexibility provided by this form of SOE. The growth in the number of SOEs makes the task of monitoring them more difficult and is questionable from the perspective of the criteria for defining a SOE. This may explain the fact that EPNAs are sometimes arbitrarily created by a supervisory ministry. Moreover, the “reform of large SOEs, especially those located in remote locations, can be contentious since it may result in layoffs which may be socially and hence politically unacceptable, particularly when alternative employment opportunities are difficult to find� (OECD 2013, 60). In remote and impoverished regions, SOEs are usually used for employment generation. According to the Tunisian press, Groupe Chimique Tunisien (GCT) recorded losses of some TD 2 billion over the last two years, which can be explained in particular by the fact that the headcount of its main subsidiary, Compagnie des Phosphates de Gaza, tripled between 2010 and 2012, rising from 5,000 to 16,000 by 2012, while production fell from 8 million tons in 2010 to 2.5 million tons in 2012. GCT is by far the main employer in a remote region of Tunisia where unemployment is very high as well as social pressure. This situation stands in sharp contrast to the sound health of the Office Chérifien des Phosphates (OCP) in Morocco over the same period.39 Too Many or Too Few Controls? SOEs suffer either from a proliferation of multilayered controls without a clear delineation of mandates or from minimal controls, depending on the political economy or the size or strategic interest of the SOE. In Egypt, SOEs that are 25% or more owned by the public sector are required by Law 203 to be externally audited by government agencies. Law 203 companies are not required to hire private external auditors (and over 90% of the companies surveyed in the Corporate Governance Report (2008) did not use private audit services). Companies are required by Law 203 to submit quarterly performance reports and an annual report to the holding company. In addition most holding companies have developed a range of other internal (but informal) monitoring mechanisms but implementation of such measures depend on the holdings. 39 http://www.leaders.com.tn/article/11294/print. In Iraq, the external audit system is essentially formed by three governmental entities: Board of Supreme Audits (BSA)(Law 31/2011), Office of the Inspector General (OIG) (CPA Order 57), and The Commission of Integrity (CoI) (Law 30/2011). It is characterized by (i) actual lack of independence and transparency, (ii) overlapping and confusion of roles and tasks of the BSA and the OIG, deriving also from misinterpretation of the original CPAs and subsequent Laws, (iii) lack of capacity and human resources to undertake tasks such as management and operational performance assessments), (iv) disputes between Ministers and IGs, (v) consequent downgrading of the role of the OIG limited de facto to external monitoring. Moreover, internal controls of SOEs are executed by an “ad hoc� committee of employees depending directly on the Director Generals of the Ministries in charge of the SOEs. Internal controls seem rather limited in scope, focusing more on administrative and financial procedures than on actual performance of SOEs. In Tunisia, “to meet this requirement [of controls], the public authorities’ response has often been to create additional structures responsible for that function. By increasing the number of bodies involved, these attempts at reorganization have instead generated increased complexity, overlapping, and duplication of work� (Ben Letaïef 1998, 441). There is a paradox in control systems for state-owned enterprises. In theory, SOEs should be weighed down by controls from the supervisory authority, the regulatory body, the state auditor, and the statutory auditors. In practice, however, although some enterprises are inspected, others seem more able to act at their leaders’ discretion.40 The proliferation of controls usually results in lack of transparency, incoherent decisions, and sometimes deadlocks. Control takes multiple forms and is more an issue of regulatory compliance (of procedures and processes) than real and fundamental dysfunctions (in the outcomes or efficiency of processes or procedures). There is excessive polarization among the regulatory bodies, emphasizing form over substance and procedures over results. This attitude toward controls may explain why SOE results and effectiveness are overlooked in favor of their procedures. Moreover, no matter the size or sector, all SOEs are subject to the same controls and regulations. Unlike in Morocco, the concept of risk is almost completely absent in Tunisia. There, audits often cover up to four or five years. Under these conditions, it is difficult to verify both the accuracy of the accounts and SOEs’ economic and financial management, or their financial practices, efficiency, effectiveness, or good governance. Because of the scale of Tunisia’s extended public sector,41 the number of staff devoted to SOE monitoring appears to be very limited, with fewer than 120 people working in the horizontal regulatory bodies and 80– 90 40 In Tunisia, in control bodies such as the corps de contrôle, which are subject to the hierarchical authority of their respective ministers (prime minister or minister of finance), inspectors receive off-program assignments that are often dictated by ad hoc considerations. 41 The sectors subject to controls include the state’s administrations, local governments, and any enterprises in which the state is a majority or minority stockholder or that receive public funding. The list of SOEs, even if limited to those in which the state has majority holdings, includes over 150 entities. magistrates in the Court of Accounts, whereas the list of SOEs, even if limited to those in which the state has majority holdings, includes over 150 entities. As a result, only the largest companies are subject to in-depth audits with any degree of frequency. Between 1981 and 2008, all 160 SOEs were audited by the Court of Accounts at least once. The problem this poses is all the more acute because the SOEs’ internal audit departments and state controls are often deemed insufficient, if not nonexistent. In addition, inspectors may be faced with asymmetrical information and may not have the skills or the information they need to perform their work effectively. Moreover, some politically sensitive companies benefit from limited controls but with subsequent fiscal risks. In Iraq, the control system of SOEs is characterized by uneven enforcement of current legislation, overlapping competencies of implementing agencies, and lack of institutional capacity in procurement and audit. In Morocco, according to Article 23 of Law 69-00, the Caisse de Dépôts et de Gestion (CDG) is exempt from the SOE monitoring mechanisms established by this law. The CDG is overseen by a supervisory committee composed of the governor of Bank Al-Maghrib, a representative of the head of the government, a representative of the Ministry of Finance, and two representatives of the Supreme Court. This committee, however, has only an advisory function and lacks de jure or de facto authority to review management decisions made by the head of the CDG. With the exception of the CDG, Morocco has a fundamentally different approach to control of SOEs and can also be viewed as a relatively good example of giving greater autonomy to SOEs. In Morocco, the philosophy of control has been fundamentally altered by adjusting the level of control to the level of risk (see box 4.2 for more details). Thus the main criterion for defining the different types of control is the quality of the SOE’s information and management systemin other words, its ability to anticipate and manage various operational and financial risks. The control type is not linked to the SOE’s type of activity (commercial or otherwise) or its legal status. Law 69–00 distinguishes between two main categories of control: ex ante controls and ex post controls. Ex ante controls apply to SOEs whose information, management, and monitoring systems do not offer the levels of security needed to anticipate risks. The aim is to regularly monitor the management of these organizations and to ensure that their economic and financial operations comply with the laws, regulations, and charters that apply to them. Control is exercised through three main agencies: the Ministry of Finance, the state comptroller, and the paymaster responsible for the compliance of expenditures, who may be authorized by order of the minister of finance to audit revenues. Instead of ex ante controls, supporting controls apply to state-owned enterprises and institutions (établissements et entreprises publics, EEPs) that can show they have effectively implemented an information, management, and internal monitoring system, duly approved by the minister of finance and comprising a number of management tools. This type of control aims to reduce the level of control, including the elimination of prior approvals, and bring enterprises up to standard by putting a management system in place that can guarantee sound risk management. Box 4.2 Financial Controls in Morocco Morocco has an unusual approach to financial controls in the MENA region because Law 69-00 provides for two regimesone for ex ante control and the other one for ex post control. The selection of the former or latter depends on the financial risks the state faces and the quality and reliability of information systems. In the other countries, SOEs are, in theory, subject to the same ex ante and ex post controls. State-owned enterprises with direct control by the state are, in principle, subject to the ex post control regime. The main features of Law 69-00 include the following:42  Approval by the minister of finance of budgets, draft financial estimates, and decisions pertaining to the allocation of profits  The presence of a state controller who sits on the board with a consultative role. Specific control attributions can be assigned to the controller by the minister of finance over a number of operations.  The preparation of a series of instruments, including a transparent regime for employees (e.g., on recruitment conditions, pay policies); a company chart setting organizational structures and their attributions, including management and audit; a manual describing operating modes for those structures and internal control mechanisms; a policy governing procurement and control ; accounting principles allowing for the establishment of transparent financial statements to be audited by certified external auditors; a three-year activity program with economic and financial projections; and a yearly management report  The establishment of an audit committee to include the state controller and two to four members nominated by the board or the deliberating body. Its mission is to control the proper operation and performance of the enterprise and to supervise internal and external audit missions.  The mandatory transmittal to the minister of finance of a number of documents, including yearly financial statements and a management report, a report on participations, an audit report from the external auditors, and consolidated financial statements. The Problem of Subsidiaries to Circumvent Controls In a country like Morocco, the most spectacular development related to corporatization has been the proliferation of subsidiaries. More than 350 SOEs were created over the last decade (Semmar 2012), and most of them are limited liabilities and are entirely or partially controlled by public operators (see table 4.1). Table 4.1 Number and Percentage of the Various Legal Forms of SOEs in Morocco % of total Number SOEs Statutory public establishments 239 33.4  Companies with state shareholdings 476 66.6 Limited liabilities with direct ownership of the Treasury 42 5.8 Subsidiaries and other state shareholdings 434 60.7 Total state-controlled companies 715 100 Source: MINEFI du Maroc 2013. According to the SOE directors interviewed in Morocco, the underlying reason for this trend is to seek greater flexibility in the management of certain specific activities, to build partnerships with private companies, and to attract new private investors. Achieving such goals would obviously produce more flexibility because some SOEs use private sector rules of management43. However, it could also create some potential problems because of the absence of oversight that can lead to circumventing regulations on public employment and wrongdoing, as well as possible losses of state capital. For example, the 182 subsidiaries of Caisse des Dépôts et de Gestion, which account for 38 percent of total subsidiaries in the SOE portfolio, are not subject to review by the Ministry of Finance and DEPP. This absence of oversight is even more problematic because most of newly created firms are involved in sectors highly exposed to risk such as tourism and real estate, and in more than half of all cases the CDG is the only capital provider. Corporatization and Independence of Boards In Egypt, SOEs were corporatized, but the government remained in the driver’s seat in the chain of command. The central role was awarded to the prime minister, who appoints the board of directors of the holding companies and their general assemblies, and the minister of investment, who heads the General Assembly. This clear lack of independence increased the criticism directed at the privatization program and, above all, created an overall social and public cynicism about the reform program. 43 A large SOE may be expected to have subsidiaries for a number of legitimate reasons (operational, legal, tax for instance). 44 It is worth noting again that one of the key issues with corporatization is employment status. One of the first questions usually raised is whether labor contracts should be transferred to the new corporation and their legal status, including whether employees should, as the case may be, retain a civil servant status. This explains why corporatization usually raises concernsor oppositionamong the workforce and trade unions. What is in general implemented is a mix of civil servants and contract staff with widely differing benefits, which creates tensions and demotivation among some staff. The preponderance of politicians on SOE boards underscores the primacy of the political agenda over the economic performance of the company, which may work in favor of political short-term objectives. “Despite SOEs’ de jure autonomy, boards staffed with administrative and political elites that are not nominated through a structured process by the state can arbitrarily intervene in management decisions due to high centralization of bureaucratic power� (OECD 2013, 24). In Egypt, the appointment of executives in holding companies falls within the norms of the findings of the Hawkamah–International Finance Corporation (IFC) survey of 2008, which concluded “that being a high-profile public officer remains a primary criterion for nominating a director to the board of a SOE and that competency and skills are secondary requirements.� Therefore, in a number of SOEs the directors do not seem to have a strong command of the strategic issues in the sector in which the company operates, mainly because of the absence of a clear and transparent nomination process. In practice, nomination processes for board posts in the MENA region are often based on criteria other than competence, and board membership is sometimes used as a compensation award for retirees. In Egypt, a director's length of service as a board member and tenure on various board committees is not properly defined for most companies, and the attendance records of board members at board and committee meetings are not standardized as is the practice in most companies. Meanwhile, the qualifications of board members are not disclosed, the links between remuneration and company performance and commercial and noncommercial objectives are not well-defined, and the remuneration policies for executive and non-executive directors are not developed. In theory, boards of directors have substantial powers in Tunisia. However, in practice these powers are considerably reduced for SOEs because of two factors: (1) the board of directors does not in fact designate the enterprise’s chief executive officer (CEO), and (2) the board’s decisions are only confirmed after “approval by the supervisory authority� (Law 96 -74 of July 29, 1996, Article 10, new). 45 As for EPNAs, the state appoints the CEO directly by decree (Law 89-9 amended, Article 33.8). The board of directors acts on a consultative basis only. Finally, the supervisory authority approves the SOE’s budget forecasts, balance statements, operating accounts, and income statements. In Morocco, a prime example is the composition of the OCP board. Of the 10 directors, there are three ministers, five secretaries general of different ministries, the head of the Treasury, and the CEO of Banque Centrale Populaire (BCP). 46 Moreover, despite the relaxation of the rule of 45 In practice, this approval is often considered tacit because the supervisory ministries do not formally approve these decisions. 46 Other than the Treasury, BCP is the only shareholder in OCP; it holds 5.88 percent of total shares. appointment of heads of SOEs foreseen in the constitution of 2011, it remains a significant obstacle to boards of directors performing their function of monitoring and, specifically, exercising the right to remove the company manager.47 The lack of independent directors is another major limitation on the role of a board of directors. The creation in 2009 of the Moroccan Institute of Board Members (IMA) has not yet remedied the virtual absence of professional directors.48 A limited number of ministers and senior government officials sit on several boards simultaneously. For example, the director and the secretary general of the Caisse de Dépôts et de Gestion sit, as a director or a member, on more than 30 boards of the different subsidiaries of the group. Transparency, Reporting, and Accountability In the selected countries, transparency is usually dysfunctional at the SOE level and at the “sector� level (“aggregate reporting�), the only exception being Morocco. In Morocco, DEPP produces a comprehensive report annually 49 , whereas Egypt, Iraq, and Tunisia are characterized by opacity. In Egypt and Iraq, the number of SOEs is not easily available, and one needs to compute manually, for example, the aggregate losses or transfers to SOEs. It is even more difficult to have accurate data on employment levels and benefits: “The Egyptian Ministry of Investment does not possess any information about companies overseen by the defense, transport or other ministries or the military which is estimated to control a large number of companies in a variety of sectors� (OECD 2013, 18). However, that situation should change in the near future. In Tunisia, the situation is not necessarily better because the available information is often fragmented and sometimes uncertain, with the last full report on state-owned enterprises dating back to 2007. The “Report on the Results of State-Owned Enterprises and Public Institutions� (“Rapport sur les Résultats des Entreprises et Etablissements Publics�) was published by the Prime Minister’s Office (2007) and prepared by the unit assigned to monitor the productivity of state-owned enterprises and institutions. The report merged the results of 181 state-owned enterprises and institutions (out of 189) for 2006.50 47 Law 02.12 governing the appointment of senior government officials has expanded the prerogatives of the Government Council, bringing the total number of nominations by the council to 1,181 senior positions compared with 17 in the past. It has also empowered the council, chaired by the head of government, to suggest candidates for the management of the 39 “strategic SOEs� established by this law. Nevertheless, the final hiring decisions on the appointment of managers remain at the discretion of the king, by decree ( dahir). 48 The IMA, Institut Marocain des Administrateurs, was created in 2009 to train directors and to professionalize the practice of corporate directorship. 49 This report is presented to the National Assembly, is available online and presents detailed analysis and is approximately 100-page long and more. 50 A shorter version of this report (20–30 pages long) was published in September 2014 with the main aggregated figures. The quantity and quality of the financial information published by state-owned enterprises in Tunisia is very disparate and largely short of the minimum required for adequate transparency and accountability to the Tunisian community. Most SOEs in Tunisia do not publish their financial statements on their websites. Although they are required to publish them in the Official Gazette (Journal Officiel), this method is both costly and impractical for users compared with electronic publication. For three out of five SOEs studied in details in Banque Mondiale (2014), the external auditors’ or legal auditors’ reports on SOEs’ financial statements contain reservationsthat is, they do not certify that the annual accounts portray an accurate image of the SOEs’ financial positions and results. Without transparency of any kind on employment levels, the extent of transfers or subsidies, and even the number of SOEs, the accountability of SOE managers and politicians is minimal, which opens the door for misuse of public funds and corruption (as illustrated shortly) without talking about poor performance. Indeed, a vicious cycle of inefficiency and embezzlement seems to have taken hold in several MENA countries (see figure 4.1). A system of governance that is poorly constructed from the start results in poor public resource management, which leads the authorities to strengthen controls, but at the same time those controls further divest the managers of these enterprises of responsibilities. For example, in recent years in Egypt, Iraq, and Tunisia thousands of jobs have been created in SOEs mainly based on political decisions rather than on needs assessments. As a result, many SOEs in those countries are virtually bankrupt. For example, in Tunisia, a SOE doubled its wage bill because of the recruitment of hundreds of employees in 2012, and so it cannot cover its expenses (after having been profitable). In this context and because of the importance of recapitalization and the increased drain of public resources, financial controls are numerous. In fact, in Tunisia no less than three a posteriori control bodies can potentially control any SOE. Because of this, managers feel less and less accountable to their boards or to the population. They feel that they are only accountable to the minister or prime minister who appointed them. Because of the financial impact of political decisions on SOE management that usually are made opaquely, managers do not feel responsible, and they simply request regular recapitalization and more subsidies to cover the higher expenses. Therefore, politicians feel they can request more employment and more investments because they are the ones able to grant SOE managers more public resources. The process, then, comes full circle. Managers are only accountable to politicians, and so they do not have any incentive to enhance the performance of their companies (because such a goal would not align with the goals of politicians), and politicians are not really constrained from using SOEs for political purposes to please some individuals, regions, or stakeholders. Figure 4.1: The Vicious Circle of Poor Governance of SOEs Source: Authors’ representation based on interviews. The Impacts of Poor Governance of SOEs: Corruption, Misuse of Public Funds, and Overemployment There is limited empirical evidence of how the poor internal governance of SOEs affects their performance except for some recent work on SOEs in China. Su and He (2012), using a sample of 744 publicly listed manufacturing firms in China between 1999 and 2006, demonstrated that firm efficiency, as estimated using stochastic frontier analysis and data envelopment analysis, is negatively related to state ownership. In addition, they found out that firms with more independent boards are more efficient, supporting the argument that a board of directors can be an effective internal governance mechanism. In the MENA region, the literature on the governance of SOEs is relatively limited, and it rarely touches on the economic and financial impacts of poor governance. One obviously needs to be cautious when generalizing about the impact of poor governance on the misuse of public funds. Underlying factors, such as cronyism, may explain the misuse of public funds despite the good internal governance structure on paper. However, in Tunisia the bad practices and misuse of public funds in SOEs were scrupulously documented. According to a report published in November 2011 by the National Commission for the Investigation of Cases of Corruption and Embezzlement (Commission Nationale d'Enquête sur la Corruption et les Malversations), there were recurrent poor practices at several SOEs:  Access to state-owned land at prices completely out of step with market prices was a highly lucrative practice, especially in a climate of booming property development.  Public services and SOEs were used for personal purposes in order to give companies in the Ben Ali family significant competitive advantages.  Abuse of state assets for private purposes was evident, such as the use of Tunisair services by Karthago Airlines without paying for them.  The regime's state banks were used for special access to credit under advantageous terms. Moreover, a state can find a multitude of ways to keep SOEs afloat: direct support in the form of subsidies or indirect support such as through access to finance under preferential conditions (e.g., via public banks), state guarantees, public contracts that would not be subject to regular procurement procedures, preferential tax and regulatory treatment, or schemes to benefit from various state facilities. In the literature, it is what Kornai (1986) calls the soft budget constraint. In Egypt, SOEs drain public resources because the state has to invest public resources in assets, but SOEs do not generate profits and depend on subsidies, despite the fact that some of them benefit from monopolies. Although the assets of the SOEs under the Ministry of Investment, including the assets of their affiliated companies, accounted for almost 7.5 percent of Egypt’s gross domestic product (GDP) in 2011 and their profit did not exceed 0.1 percent of GDP. The companies not falling under the Ministry of Investment realized profit of about 0.2 percent of GDP, although their asset share of GDP was almost 2 percent. Moreover, the so-called economic authorities amassed an aggregate operating loss of more than $15 billion (LE 114.5 billion) in 2013. Only the economic authorities of three sectors (out of nine) had a net profit without subsidies (see table 4.2).51 In Iraq, the federal budget increasingly subsidizes SOEs by means of loans through state banks, capital allocations (state bank loans for implementing and rehabilitating projects in SOEs), and special funds for emergencies (unforeseen expenditures and such but also misused to finance salaries). Table 4.2 State funding for SOEs: Iraq, 2007–12 US$, millions 2007 2008 2009 2010 2011 2012 51 The three profitable sectors were transportation, communication, and aviation with a net profit of LE 14.4 billion. The economic authorities in the industrial and mineral resources sector recorded the highest losses, LE 86.6 billion. 3,000 Total SOE subsidies 905 2,560 2,906 2,564 2,411 (est.) Source: Iraqi authorities. In Iraq, according to 2010 data, about 30–50 percent of SOE employees were not working (excess labor or non-active employees) because of overemployment that greatly increased SOEs in 2005–07. Subsidies were thus created to generate employment and rents because wages and salaries are 30–50 percent higher in the private sector than in the public sector. Within the energy sector in Tunisia, the existence of several state-owned enterprises allow cross-subsidization, which can conceal the ineffectiveness of some of them. The absence of good practices in this sector in Tunisia is illustrated here by referring to the main criteria defined earlier in this volume:  State acting as an owner. The energy sector has not been under scrutiny because the units in charge of monitoring SOEs (Direction Générale des Participations et Unité de Suivi de la Productivité des Entreprises Publiques) have had difficulty gaining access to economic and financial data and are staffed with only a few people.  Autonomy. Some of these companies, such as STIR, Tunisia’s refinery, and STEG, the enterprise in charge of power generation, have been constrained by recruitment campaigns in recent years, which partially explains their recent extremely difficult financial circumstances.  Boards of directors. The boards of energy sector companies consist of civil servants; these companies do not have independent boards of directors.  Transparency and disclosure. Opacity prevails because the cross-subsidization has not been made public (at least with aggregate figures). Moreover, some companies such as STIR have delayed publishing their audited accounts. STIR receives a major subsidy from ETAP, the national petroleum company. Thus in an opaque environment cross-subsidization 52 enables companies such as STIR and STEG to appear in a much better situation than they really are (see figure 3.2 for an indication of the cross-subsidies among ETAP, STIR, and STEG). Tunisia’s energy SOEs operate in an environment that is not conducive to financial discipline (whereas this is crucial to have performing SOEs 53). Moreover, this subsidy system is not transparent (in terms of its policy objectives, the amounts involved and how they are allocated) and fair. In the aggregate, if both STIR imported crude oil and STEG imported natural gas based on prices 52 ETAP imports oil and gas on behalf of STIR and STEG. In doing so, ETAP imports crude oil and sells it at 30 percent of its cost to STIR (TD 50 per barrel), or less than half the price on the international market. 53 For more details, see Kikeri et al. (2014) in chapter 5. in the international markets, over TD 1.2 billion in additional funds would have to be transferred from the Tunisian state budget to the two companies. Furthermore, because ETAP’s profits are currently being used to reduce a portion of budget transfers (approximately TD 300 million), STIR and STEG would benefit from TD 1.5 billion in indirect subsidies, or over 2 percent of GDP. If this amount were incorporated within the budgeted energy subsidies, the total would be 50 percent higher than 5 percent of GDP, for a total in excess of 7 percent of GDP. The state transfers the price subsidy on petroleum products to STIR and regularly allows that SOE to import such products. As a result, this model appears to be very costly (through the loss of revenue for ETAP and little incentive for STIR to increase its efficiency) as well as opaque (over procurement unit prices, etc.). In 2013, over TD 200 million was transferred from ETAP to STIR and over TD 1 billion from ETAP to STEG. Figure 4.2 Cross-subsidization in Tunisia among ETAP, STIR, and STEG Oil (for the state) STIR STIR STIR ETAP Meeting refinery’s Imports of finished National production of needs at TD 50 per Refinery products of 2.4 0.6 MT + imports of ton production MT 1.2 MT (crude) at TD 170 of 1.6 MT per ton Subsidization at TND 120/ton Distribution to (70%) distribution companies, Natural gasETAP including SNDP National production of 2.9 MTOE + imports of Subsidization at TD 810 1.6 MTOE + 1.1 MTOE per TOE (90%) for via Algerian pipeline (transit) at TD 900 per imports Compensation and TOE STEG (5.3 budget transfers (when MTOE) necessary) Power generation + Natural gas distribution Source: Banque mondiale 2014. Note: MT = metric tons; MTOE = million tons of oil equivalent; TOE = tons of oil equivalent; ETAP = ; SNDP = Société Nationale de Distribution de Pétrole; STEG = Société Tunisienne de l'Electricité et du Gaz ; STIR = Société Tunisienne des Industries de Raffinage. Chapter 5 What New Reform Priorities Could Change SOE Governance in the MENA Region? In his fair assessment of reforms of state-owned enterprises (SOEs). Gómez-Ibáñez (2007) points out that “it should be no surprise that the efforts to reform SOEs in the 1960s, 1970s and 1980s were disappointing. . . . Given the record, . . . it is difficult to recommend specific reform strategies with great confidence.� Vagliasindi (2008a) also notes that “we still know very little about the effects of board structure or corporate governance more generally on firm value or performance.� The Limits of Ambitious SOE Governance Reforms in the MENA Region Lack of a good record of reforms and little empirical knowledge should make countries and donors cautious as they move forward in this area. SOE Governance Reforms in a Low Governance Environment As noted earlier, it is difficult to directly link governance and firm performance, and benchmarks are rare and usually subject to criticism.54 However, what has been demonstrated empirically and theoretically is that the characteristics of a country are an important determinant of firm-level governanceits costs and benefits. Firms benefit from good governance because it allows them to access external markets on better terms, but that benefit is not of much value in countries with weak and inefficient capital markets (Doidge, Karolyi, and Stulz 2006). Therefore, in low governance countrieswhich include most countries in the Middle East and North Africa (MENA) regionSOE governance reforms are unlikely to bring the results that would be expected in a country with good governance. For example, in countries with extremely thin capital markets, what benefits might a SOE derive from disclosing financial statements on time? In a country where civil society does not scrutinize the use of public resources and collective action is difficult to achieve, what could be the impact of more transparency on subsidies to SOEs or on waste of public resources? SOEs operate in a governance context, and, even though governance structures could be amended, it does not mean that practices would radically change. The Success or Failure of SOE Reforms and the Political Process As one observer puts it, “Opportunities for reform arise when events temporarily alter the political economy of the firm� (Gómez-Ibáñez 2007). And Nabli (2001) said the following about SOE reforms in the MENA region: “The feasibility of SOE reform is related to the costs of reform and to the political pressures opposing it.� In most countries, SOE reforms are essentially designed to include corporatizing SOEs, changing boards and board compositions, and increasing transparency and disclosure. 54 Despite those problems, Megginson and Netter (2001) quote some studies on the impact of privatization on firm performance and explain the usual rationale for privatization of SOEs. On one side, the losers in such reforms are usually quite numerous and politically powerful such as the employees of utilities, some high-level civil servants, and politicians. Indeed, reform in a SOE usually  Overreliance on Form over Results It is unclear whether implementation of the four principles of SOE reform described earlier would prevent the effects of cronyism and patronage. 55 Reforming the labor code, however, is a task that goes well beyond restructuring a SOE. 56 See Brockmeyer et al. (2015) for more details. This is an element mentioned by Krueger (1990) in favor of privatization. There is a disconnect between window dressing attitudes and the behavior57 surrounding SOE reforms in the MENA region. As a result, the reality is hidden, and reforms do not materialize, remaining largely on paper. Institutional mimicry prevails and therefore explains why, despite reforms on paper and governance arrangements closer to those in the private sector, an administrative approach to SOE management continues to prevail and administrative practices remain largely intact. In short, there is a strong disconnect between stated ambitions and the reality of the day-to-day functioning of SOEs. According to organizational sociologists, “organizations often arrange their core activities according to accepted models, or templates, in their field� (D’Aunno, Succi, and Alexander 2000) or “tend to model themselves after similar organizations in their field that they perceive to be more legitimate or successful� (DiMaggio et al. 1983). Therefore, it is not surprising that politicians and bureaucrats strive to copy the SOE models followed by member countries of the Organization for Economic Co-operation and Development (OECD) because they are the ones rightly perceived as the most successful ones. Their organizational characteristics are mimicked in the region: corporatizing, placing independent members on boards, strengthening the units in charge of monitoring SOEs, strengthening financial audits, introducing more transparency, developing special vehicles with public-private partnership (PPP) frameworks, and so on. In conclusion, where political control over SOEs remains quite high, such as in the MENA region, the usual reforms are unlikely to succeed and a different approach should be used. Because this objective is usually has a medium-term perspective, a push for transparency is most probably needed on the cost of SOEs, employment levels and wages, productivity (but combined with training for the media), and the actual state of SOE management and its impact on taxpayers and on the economy in general. A Gradual Path toward Success: Lessons from the Corporatization of the OCP in Morocco he Office Chérifien des Phosphates 57 Anthropologists and organizational sociologists have documented �institutional mimicry� since the 1970s, demonstrating that beyond reforms on paper there is strong tendency for organizations to continue with the prevalent practices (an important article is that by Bugnicourt 1973).  Alongside its commercial mandate, the OCP assumes a vast developmental mandate in a variety of areas. This includes the promotion of agriculture (such as mapping fertile lands in Morocco, promoting the sustainable use of fertilizers by small farmers, or promoting innovative agricultural projects through its innovative agricultural fund), the promotion of employment (such as a large training programs for youth known as OCP Skills and an initiative to fund small and medium enterprise development), the promotion of the chemical industry sector (with the proposal that some OCP facilities be made available to the industry such as laboratories for 59 The monopoly applies to exploitation but also to the distribution and commercialization of phosphates and their by-products. research and development), and its overall participation in a debate on how to promote food security worldwide (conceived as a public policy mandate). It is extremely difficult to link governance reforms with specific performance changes. However, from interviews of various stakeholders, the recent reforms of OCP management had enabled better performance. For example, the OCP’s contribution to the Moroccan state is increasing. As of 2012, the OCP’s total financial contribution to the state (taxes and dividends) amounted to DH 4.72 billion (or over $500 million), making it the main contributor among SOEs to the state budget. Today, the OCP remains the world’s largest exporter of phosphate products, and it holds stakes in the capital of a number of subsidiaries, mostly in the chemical industry but not exclusively (e.g., real estate), including a series of joint ventures in emerging markets such as Brazil, India, and Turkey. What could be inferred from the OCP changes? The OCP does not follow the first best governance model, and accountability to the board of directors may not be the most important factor explaining the success of OCP reforms. However, the main lesson could be that although the de jure institutional architecture may not be the most important factor in the better performance of SOEs, it can nevertheless provide a cornerstone for in-depth reform. Indeed, the OCP management has definitely gained managerial autonomy and the financial means to invest and recruit, while the company continues to have a large developmental mandate for the country. Partnerships with the private sector are important to enable SOEs reforms. Reforming SOE Governance through Partnerships with the Private Sector External pressure and innovation from private players can prompt SOEs to increase their competitiveness and adapt to new market rules. This has been proven in the banking sector and in logistics and transport services. But when competitive forces are not sufficient to push SOEs to reform, mixed ownership can be an alternative approach to create the necessary reforms from within. Mixed ownership brings talent, ideas and processes right to the heart of a company's decision-making process and challenges the core functioning of the organization. Faced with tighter fiscal space and the need to accelerate their growth rapidly, most non-oil MENA countries need to shift their public investment objectives toward sectors with high potential for productivity gains, which will be critical to achieving sustained high rates of growth in their gross domestic products (GDPs), in particular through exports. These non-oil MENA states would benefit from encouraging more private sector participation in the sourcing, preparation, financing, and implementation of investment projects within the framework of public-private partnerships, where SOEs could play a proactive role. From this perspective, SOEs will need to shift their role from a direct one to that of provider of investment opportunities to be achieved by local or foreign private investors. This new role should notably be framed through partnerships with the private sector. Which role for the World Bank on SOEs reforms? Table 5.1: Do’s and don’ts of SOEs reforms in a low governance environment Appendix The Legal Definition of SOEs in Egypt, Iraq, and Tunisia Egypt Egypt has about 260 public enterprises operating under different regulatory regimes. They fall under four broad categories:  Economic authorities (52) and service authorities (102). Economic authorities are.... Service authorities depend on recapitalization or operate on a subsidized business model.  Public business sector companies under specific ministries. These companies, which include the Egypt Electricity Holding Company (EGELEC) and the Egyptian Company for Water and Waste Water, are 100 percent controlled by the state and operate under the authority of line ministries. These are mainly companies considered “strategic� in sectors such as electricity, telecom, aviation, banking, housing, and petroleum.  Public business sector companies under the Ministry of Investment. This category includes most agricultural, textile, chemical industry, mining industry, transport, construction, tourism, pharmaceutical, and food processing holdings as well as the Egyptian Natural Gas Holding Company (EGAS). These holdings fall under Law 203 on corporations. A relatively large series of what were formerly “government units� have been corporatized under a new regime established in 1991 (Law 203) as part of the privatization effort. Law 203 puts forth a regime that is close to private sector corporations in their day-to-day operations, including the establishment of a board of directors as well as control and reporting mechanisms.  Private business corporations with state participation. Both the Ministry of Investment and line ministries may hold stakes in the capital of private sector corporationsfor example, in the form of joint ventures.  For some ministries, holding companies have been established grouping 10–25 state-owned enterprises (SOEs) or public companies. The shareholders’ meetings of holding companies are chaired by the minister and attended by the board of directors, which is composed of political appointees. In affiliated SOEs, the shareholders’ meeting is headed by the chair of the holding company board and attended by board members of the holding company. Up to four board members are selected by the holding company shareholders at their meeting. Shareholders’ meetings approve board reports, distribution of profits, and continuation or dismissal of the chair and directors. As a result of privatization, a certain number of Law 203 SOEs have private sector shareholders. Iraq Iraq’s economy is still largely controlled by the state and, compared with other countries in the region, is relatively unsophisticated when it comes to the methods for state participation in the business sector:  State enterprises. These enterprises are defined by Law 22 of 1997 as “financially self-sustainable and fully (100 percent) state owned economical units with moral personality and financial and administrative autonomy and operating under economic principles.� Law 22 provides for a regime governing their establishment, operation, and liquidation. This regime, however, is relatively limited in its provisions. Although companies have capital, no reference is made to shares or to shareholders. The main corporate organ is the board of directors, comprised of eight members, of which six are designated by the minister to which the enterprise is attached. The general manager is nominated by the cabinet. In practice, in Iraq’s largely centralized and state-controlled economy, state enterprises reportedly operate as ministerial departments.  Mixed companiesand other participations. The state can also participate in companies jointly with the private sector. Mixed companies have state participation of above 25 percent and are subject to a specific regime, defined under Law 21 of 1997 (amended in 2004). Law 1997 also defines the regime of regular private corporations. Under this 25 percent threshold, the company is subject to general corporate law. Shares of companies established under Law 21 can be publicly listed. Tunisia A series of texts govern state participation in entities active in the productive sector in Tunisia. Law 89-9 of 1989, pertaining to statutory SOEs, enterprises, and participations, provides a definition of SOEs (Article 8). SOEs include:  Statutory SOEs of a non-administrative nature fixed by a decree  Corporations whose capital 100 percent owned by the state  Corporations whose capital is at least 50 percent owned by the state, local governments, statutory SOEs, or corporations 100 percent owned by the state. These SOEs are governed by a mix of specific laws (statutory SOEs) or general corporate law (corporations), together with the provisions of Law 89-9. Issues covered by Law 89-9 for SOEs include representation by the state at meetings of the board of directors and control modalities over board decisions, rules pertaining to the recruitment of personnel, communication of information to public authorities, audits and control mechanisms, and procurement. It should be noted that, although the terms corporation and statutory SOEs of a non-administrative nature may seem to refer to productive activities, these criteria are not clearly set by Law 89-9. In practice, it is reported that the list of statutory SOEs of a non-administrative nature includes a number of establishments whose activities cannot be considered productive. Law 89-9 also provides rules pertaining to issues such as state participations, defined as participations held by statutory SOEs or 100 percent-owned corporationsparticipations whose exact nature is not defined, nor are the thresholdsand the restructuring of “enterprises with a public participation.� It appears that Tunisia recognizes all of the forms of state participation: statutory SOEs and corporations, direct and indirect participation, total and partial ownership, monopolistic or subject to competition. It is, however, very difficult (if at all possible) to summarize the regime applicable to these various forms of participation. References Akerlof, G. A. 1970. ‘The Market for Lemons: Quality Uncertainty and the Market Mechanism.� Quarterly Journal of Economics 84 (3). Banque Mondiale. 2014. Pour une Meilleure Gouvernance des Entreprises Publiques en Tunisie, Rapport 78675-TN, Washington DC : World Bank. Benhassine, Najy. 2009. From Privilege to Competition. Washington, DC: World Bank. Ben Letaïef, Mustapha. 1998. L’État et les entreprises publiques en Tunisie. Paris: L’Harmattan. Brockmeyer, Anne, Khatrouch, Maha and Raballand, Gael, 2015. "Public sector size and performance management : a case-study of post-revolution Tunisia," Policy Research Working Paper Series 7159. Bugnicourt, Jacques. 1973. “Le mimétisme administratif en Afrique: obstacle majeur au développement.� Revue française de science politique 6: 1239–67. Burgess, Robin, and Rohini Pande. 2003. “Do Rural Banks Matters? Evidence from the Indian Social Banking Experiment.’� American Economic Review 95 (3). Carana Corporation. 2002. “The Results and Impacts of Egypt’s Privatization Program: Special Study.� http://www1.aucegypt.edu/src/wsite1/Pdfs/Results%20and%20Impacts%20of%20Privatization% 20in%20Egypt.pdf. Christiansen, Hans. 2013. “Balancing Commercial and Non-Commercial Priorities of State Owned Enterprises.� OECD Corporate Governance Working Paper No. 6. Organisation for Economic Co-operation and Development, Paris. DiMaggio, Paul J. and Powell, Walter W. 1983. “The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields�, American Sociological Review, Vol. 48, No. 2, pp. 147-160. D’Aunno, Thomas, Melissa Succi, and Jeffrey A. Alexander. 2000. "The Role of Institutional and Market Forces in Divergent Organizational Change." Administrative Science Quarterly 45: 679–703. Doidge, Craig, G. Andrew Karolyi, and René M. Stulz. 2006. “Why Do Countries Matter So Much for Corporate Governance?� Journal of Financial Economics 86 (1): 1–39. Gómez-Ibáñez, José A. 2007. “Alternatives to Infrastructure Privatization Revisited: Public Enterprise Reform from the 1960s to the 1980s.� Policy Research Working Paper 4391, World Bank, Washington, DC. Grossman, Sanford, and Oliver Hart. 1986. “The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration.� Journal of Political Economy 94: 175–202. Hart, Oliver. 1995. “Corporate Governance: Some Theory and Implications.� Economic Journal 105 (430): 678–89. Hart, Oliver, and John Moore. 1990. “Property Rights and the Nature of the Firm.� Journal of Political Economy 98 (6): 1119–58. Hart, Oliver, Andrei Shleifer, and Robert W. Vishny. 1997. “The Proper Scope of Government: Theory and an Application to Prisons.� Quarterly Journal of Economics 112 (4): 1127–61. Kang, Young-Sam, and Byung-Yeon Kim. 2012. “Ownership Structure and Firm Performance: Evidence from the Chinese Corporate Reform.� China Economic Review 23: 471–81. Kikeri, Sunita, Brumby, Jim, Arrobbio, Alexandre, Nunez, Alejandra, Berg, Alexander, Barros, Ana Cristina Hirata, Moreno-Dodson, Blanca, Robinett, David, Fortin, Henri, Steinhilper, Immanuel, Garrido, Jose, Zoratta, Laura, Verhoeven, Marijn, Beauchard, Renaud, Vani, Sanjay. 2014. Corporate governance of state-owned enterprises : a toolkit. Washington, DC : World Bank Group, available at http://documents.worldbank.org/curated/en/2014/10/20286791/corporate-governance-state-owne d-enterprises-toolkit Kornai, Janos. 1986. “The Soft Budget Constraint.� Kyklos 39 (1): 3–30. Krueger, Anne. 1990. "Government Failures in Development." Journal of Economic Perspectives 4 (3): 9–23. Lewis, W. Arthur. 1949. The Principles of Economic Planning. London: George Allen and Unwin. MacCarthaigh, Muiris. 2010. “Managing State-Owned Enterprises in an Age of Crisis: An Analysis of Irish Experience.� Policy Studies 32 (3): 215–30. Meade, James Edward. 1948. Planning and the Prince Mechanism: The Liberal Socialist Solution. London: George Allen and Unwin. Megginson, William L., and Jeffry M. Netter. 2001. “From State to Market: A Survey of Empirical Studies on Privatization.� Journal of Economic Literature 39 (2): 321–89. MINEFI du Maroc (Ministère de l’Economie et des Finances du Royaume du Maroc). 2013. Rapport sur le secteur des Etablissements et Entreprises Publics. Rabat: MINEFI. Musacchio, Aldo, and Sergio Lazzarini. 2012. “Leviathan in Business: Varieties of State Capitalism and Their Implications for Economic Performance.� Unpublished manuscript. Nabli, Mustapha. 2001. “Institutional Analysis of State-Owned Enterprises Reform and the MENA Region.� In State-Owned Enterprises in the Middle East and North Africa, edited by Merih Celasun. London and New York: Routledge. National Commission for the Investigation of Cases of Corruption and Embezzlement (Commission Nationale d'Enquête sur la Corruption et les Malversations). 2011. Tunis, November. Nguyen, Thuy Thu, and Mathijs A. van Dijk. 2012. "Corruption, Growth, and Governance: Private vs. State-owned Firms in Vietnam." Journal of Banking and Finance 36 (11): 2935–48. OECD (Organisation for Economic Co-operation and Development). 2005. OECD Guidelines on Corporate Governance of State-Owned Enterprises. Paris: OECD Publishing. http://www.oecd.org/corporate/ca/corporategovernanceofstate-ownedenterprises/34803211.pdf. ______. 2010. Accountability and Transparency: A Guide for State Ownership. Paris: OECD Publishing. ______. 2012. Towards New Arrangements for State Ownership in the Middle East and North Africa, Paris: OECD Publishing. ______. 2013. State-Owned Enterprises in the Middle East and North Africa Paris: OECD Publishing. Omran, Mohammed. 2004. “The Performance of State-Owned Enterprises and Newly Privatized Firms: Does Privatization Really Matter?� World Development 32 (6): 1019–41. Prime Minister’s Office, Tunisia. 2007. “Report on the Results of State-Owned Enterprises and Public Institutions� (Rapport sur les Résultats des Entreprises et Etablissements Publics). Tunis. Saghir, Jamal. 1993. “Privatization in Tunisia.� CFS Discussion Paper 101, World Bank, Washington, DC. Sappington, David. E. M., and Joseph E. Stiglitz. 1987. “Privatization, Information, and Incentives.� NBER Working Paper 196, National Bureau of Economic Research, Cambridge, MA. Seiznick, Philip. 1996. “Institutionalism ’Old’ and ‘New.’� Administrative Science Quarterly 41 (2): 270–77. Semmar, Abderrahmane. 2012. “Corporate Governance of State-Owned Enterprises in Morocco: Evolution and Perspectives.� In Towards New Arrangements for State Ownership in the Middle East and North Africa. Paris: OECD publishing. Shleifer, Andrei. 1998. “State versus Private Ownership.� Journal of Economic Perspectives 12 (4): 133–50. Shleifer, Andrei, and Robert. W. Vishny. 1994. “Politicians and Firms.� Quarterly Journal of Economics 109 (4): 995–1025. Simons, Henry C. 1948. “A Positive Program for Laissez-Faire.� In Economic Policy for a Free Society. Chicago: University Press of Chicago. Stiglitz, Joseph E., and Andrew Weiss. 1981. “Credit Rationing with Imperfect Information.� American Economic Review 71 (3). Su, Dongwei, and Xingxing He. 2012. “Ownership Structure, Corporate Governance and Productive Efficiency in China.� Journal of Productivity Analysis 38 (3): 303–18. Vagliasindi, Maria. 2008a. “The Effectiveness of Boards of Directors of State Owned Enterprises in Developing Countries.� Policy Research Working Paper 4579, World Bank, Washington, DC. ______. 2008b. “Governance Arrangements for State Owned Enterprises.� Policy Research Working Paper 4542, World Bank, Washington, DC. Wettenhall, Roger, and Ian Thynne. 2002. “Public Enterprise and Privatization in a New Century: Evolving Patterns of Governance and Public Management.� Public Finance and Management 2 (1): 1–29. World Bank. 2004. Unlocking the Employment Potential in the Middle East and North Africa: Toward a New Social Contract. Washington, DC: World Bank.