3. Crowding Private Firms In—or Out? Introduction Policies and regulations that typically apply to most firms and markets—and that determine market outcomes—often do not apply in the same way to businesses of the state (BOSs). BOSs benefit when principles, policies, or regulations give them explicit advantages that unlevel the playing field over private firms. They also have advantages implicitly when policies and regulations that apply to all firms favor them in those markets. The differences in the treatment and market conditions of BOSs and private firms can lead to two adverse economic outcomes. They can reduce the market-based incentives to improve technical efficiency (productive efficiency effects). They can also distort market outcomes—including prices, quantities, and the ability of rivals to compete—and resource allocations between firms and activities (allocative efficiency effects). State-owned enterprises (SOEs), in particular, are often responsible for implement- ing a policy mandate.1 In some cases, this approach may be a strategy of the state. In oth- ers, it may be due to a blurring of responsibilities between the SOE and its supervising entity. The way that SOEs operate in markets—from hiring to production decisions and pricing of products and services—can therefore reflect the implementation of these policy mandates. In many cases, this relationship will involve implicit subsidies to the SOE. For example, if an SOE is used to provide the population with essential goods at below-market prices, it may earn a below-commercial rate of return and will need support to fund its activities. Or it may involve the creation of a monopoly to control prices, quantities, or other market parameters directly. As a secondary effect, the policy mandates of the SOE could affect the incentives it faces for achieving technical effi- ciency as it moves away from a profit-maximizing objective. This situation can also have implications for performance and therefore create the need for further advantages. SOEs have greater potential for political capture and a greater likelihood of being used for political patronage. The close relationship between SOEs, governments, and their supervising entities allows the state to use them to achieve political objectives. The  resources of SOEs can be used to fund political parties, support electoral cam- paigns, or gain the support of politically influential individuals. Politicians can win 45 votes by employing workers in an SOE. Politicians can personally profit from arranging patronage jobs in an SOE for their political backers, relatives, and friends. Politicians or local officials can also benefit from influencing the selection of suppliers and contrac- tors or marketing goods in short supply. As with the policy mandate, political capture has the secondary effect of dampening SOE incentives for commercial performance. The exposure of SOEs to state influence tends to be greatest when ownership rights ­ xercised by a line ministry and lowest when ownership rights are exercised by a are e specialized agency that operates at arm’s length from the government (Vitale, Moiso, and Wanner 2020). But it is rare to find arm’s length specialized agencies responsible for SOE ownership (figure 3.1). In 67 countries with this information, only 4 have this ownership setup. Instead, in more than half of high- and middle-income countries, line ministries most commonly ­ exercise SOE ownership rights. But high-income countries are more likely to have safeguards to ensure that the chief executive officers of SOEs are appointed by board members rather than by public authorities, which reduces the propensity for SOE decision-making to be influenced by government (fig- ure 3.2). Where these safeguards are not in place, the SOE will be more likely to oper- ate with noncommercial objectives (either implicitly or explicitly) and be more prone to political capture. FIGURE 3.1  Who Exercises the Ownership Rights in SOEs? 100 8 4 15 80 25 % of respondents 30 60 13 40 25 15 20 37 30 0 High-income countries Middle-income countries (40) (27) Degree of Specialized agency at arm’s length from government exposure Specialized agency NOT at arm’s length from government to Treasury or ministry of finance or economy political Line ministries coordinated by a specialized agency influence Line ministries Sources: OECD product market regulation (PMR) questionnaires, 2018; OECD and World Bank PMR questionnaires, 2013–22; and data collected by the World Bank on selected PMR questions for eight Middle East and North Africa countries. Note: High-income countries are Australia, Austria, Belgium, Canada, Chile, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, the Republic of Korea, Kuwait, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, New Zealand, Norway, Poland, Portugal, Romania, Saudi Arabia, the Slovak Republic, Slovenia, Spain, Sweden, Switzerland, the United Arab Emirates, and the United Kingdom. Middle-income countries include both upper-middle-income and lower-middle-income countries. Upper-middle-income countries are Albania, Argentina, Belarus, Brazil, Bulgaria, China, Colombia, Costa Rica, Ecuador, Jordan, Kazakhstan, Kosovo, Malaysia, Mexico, Moldova, Montenegro, Peru, the Russian Federation, Serbia, South Africa, and Türkiye. Lower-middle-income countries are Côte d’Ivoire, the Arab Republic of Egypt, Indonesia, Morocco, Tunisia, and Viet Nam. OECD = Organisation for Economic Co-operation and Development; SOEs = state-owned enterprises. 46 The Business of the State FIGURE 3.2  Who Appoints the Chief Executive Officers of SOEs? 100 21 80 4 53 % of respondents 18 60 2 40 35 57 20 10 0 High-income countries Middle-income countries (40) (28) Degree of Board of the firm exposure Organization exercising ownership rights to political Combination of board and public authorities influence Public authorities Sources: OECD product market regulation (PMR) questionnaires, 2018; OECD and World Bank PMR questionnaires, 2013–22; and data collected by the World Bank on selected PMR questions for eight Middle East and North Africa countries. Note: High-income countries are Australia, Austria, Belgium, Canada, Chile, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, the Republic of Korea, Kuwait, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, New Zealand, Norway, Poland, Portugal, Romania, Saudi Arabia, the Slovak Republic, Slovenia, Spain, Sweden, Switzerland, the United Arab Emirates, and the United Kingdom. Middle-income countries include both upper-middle-income and lower-middle-income countries. Upper-middle-income countries are Albania, Argentina, Belarus, Brazil, Bulgaria, China, Colombia, Costa Rica, Ecuador, Jordan, Kazakhstan, Kosovo, Malaysia, Mexico, Moldova, Montenegro, North Macedonia, Peru, the Russian Federation, Serbia, South Africa, and Türkiye. Lower-middle-income countries are Côte d’Ivoire, the Arab Republic of Egypt, Indonesia, Morocco, Tunisia, and Viet Nam. OECD = Organisation for Economic Co-operation and Development; SOEs = state-owned enterprises. Channels for Granting Advantages to BOSs As a market regulator, a policy maker, or an enforcer of market rules and laws, the state can provide advantages to BOSs. It can do this through the way that market rules, regulations, and policies are specified “on the books” or in the way those rules, regu- lations, and policies are implemented or enforced in practice (figure 3.3). There are three main channels through which the state can grant policy and regulatory advan- tages to BOSs: ■■ Channel 1. The state provides explicit advantages to the SOE through ­ preferential treatment. Principles, policies, or regulations can be applied to an SOE in a way that gives it advantages over private firms, creating an unlevel playing field. These advantages are most relevant in competitive markets. ■■ Channel 2. The state provides implicit advantages to the BOS firm through policies and regulations that apply to all firms in the market in which a BOS firm operates. These rules and policies protect the BOS firm’s position in a market or dampen competition in markets with BOSs. These advantages are most relevant in ­ partially contestable markets. Crowding Private Firms In—or Out?47 FIGURE 3.3  Framework of Channels for Granting Advantages to BOSs, with Examples and Primary Effects Channel 1. Explicit advantages to SOEs The state provides direct advantages to the SOE through preferential treatment • Raises the risk of cross-subsidization between noncommercial and commercial activities, Requirements for operations undermining the price mechanism of SOEs • Alters the incentives to operate efficiently and commercially, distorting pricing mechanisms Debt advantages • Artificially lowers costs for the SOE relative to the private sector, leading to resource misallocation • Creates a risk of moral hazard or lower incentives for efficiency Procurement advantages • Possibly forecloses the private sector from procurement markets • Increases the price of goods and services for the state • Creates artificially low barriers to entry and lower ongoing costs of regulatory compliance for Exemptions from or weak SOEs than for the private sector, leading to resource misallocation enforcement of laws • Competition law exemptions: higher risk of anticompetitive behavior by firms • Financial disclosure, bankruptcy exemptions: lower transparency and accountability, leading to moral hazard and weaker incentives for performance • Environmental law exemptions: greater environmental risks Channel 2. Implicit advantages to BOSs The state provides indirect advantages to the BOSs through policies and regulations that apply to all firms in the market in which a BOSs operates • Protects BOSs from competition from foreign producers Import restrictions • Dampens competition and thus raises prices and reduces quantities • Protects BOSs from the threat of new, more efficient firms entering Limits on entry • Dampens competition and thus raises prices and reduces quantities • Dampens price competition in the BOS market; for example, prevents more efficient firms from Price or margin regulation undercutting the BOS firm SOE involvement in • Creates the potential for the BOS firm or the regulator to apply regulation in such a way that restricts regulatory decisions the private sector from entering the market or that raises the operating costs of competitors Channel 3. Policy and regulatory distortions affecting BOSs The state does not adequately regulate and implement pro-competition rules that would prevent anticompetitive conduct, either sector-specific or economywide No separation of natural • Competition cannot take hold in competitive segments monopoly segments from • Distortions spill over between segments competitive sectors Lack of access regulation or • SOEs prioritize short-term policy goals over long-term commercial considerations and investment appropriate and independent • SOEs prioritize profit objectives, or regulation is insufficient to deliver equity or access objectives price regulation in natural • The private sector is foreclosed from upstream or downstream segments. SOEs leverage market monopoly sectors power between segments Differential, insufficient, or inadequate implementation • Risk of anticompetitive behavior by firms is higher of competition regulatory frameworks to BOSs Source: Based on World Bank, forthcoming. Note: In this figure, SOEs is used when the information is based on laws that provide explicit advantages to SOEs as defined by coun- tries. BOSs is used when all firms in the market are subject to policies and regulations. BOSs = businesses of the state; SOEs = state- owned enterprises. 48 The Business of the State ■■ Channel 3. The state does not adequately regulate and implement pro-competition rules that would prevent anticompetitive conduct, either sector-specific or econo- mywide. This channel is the most relevant for sectors with natural monopolies. The advantages affecting the market conditions and incentives faced by a BOS firm differ based on the degree of contestability of the economic sector in which the firm operates. ■■ In competitive or partially contestable markets, actual and potential private competitors can be affected by the advantages provided to the BOS firm. The priority here is to understand whether explicit and implicit advantages crowd out, create an unlevel playing field, or dampen competition with actual and potential competitors (channels 1 and 2). It is important to know whether poli- cies or regulations restrict competition and whether pro-competition policies and regulations will help to boost competition (channel 3). This knowledge is particularly relevant in partially contestable markets, such as network sectors. ■■ In natural monopoly markets where social welfare is pursued by concentrating pro- duction in a single firm, the main policy and regulatory objective of the government should be to prevent the firm from unduly exercising significant market power and to ensure that it invests sufficiently in the good or service. To do so, additional regu- lations and policies need to be implemented to avoid distortions. The priority here is to know whether the firm is adequately regulated to prevent undue exercise of market power (channel 3). It is also possible to assess whether the BOS firm ­operates under the same conditions and incentives as a hypothetical private firm in the same position—that is, whether the BOS firm receives explicit advantages (channel 1). SOEs, in particular, clearly have advantages in middle-income countries (box 3.1). Explicit advantages, implicit advantages, and inadequate protection against anticom- petitive firm behavior are evident in 58 countries. Implicit advantages seem to be more common, particularly limits on entry. The most common explicit advantages for SOEs are debt or tax advantages (see chapter 4). Channel 1: Explicit Advantages to BOSs Explicit advantages can result from the absence of firm-level principles and policies that should discipline SOEs, as well as from general policies and regulations applied differentially to SOEs and to fully private firms. The first type focuses on operational requirements for SOEs that, if absent, would tilt the playing field against private firms. The second type includes regulatory advantages and economic advantages provided to SOEs. Economic advantages include debt, taxes, procurement processes, and preferen- tial access to finance, land, infrastructure, or other inputs. When SOEs deliver both commercial and noncommercial services, they should separate those activities to ensure that funds provided to the SOE to fulfill its public Crowding Private Firms In—or Out?49 service obligations do not cross-subsidize goods or services in another market. This separation is important because cross-subsidizing services can distort market mecha- nisms in commercial segments and create an unlevel playing field for private firms that do not receive subsidies for noncommercial services. To minimize the risks, countries can put in place frameworks or rules to ensure that these services are separate and thus minimize the risk of cross-subsidies.2 But most middle-income countries impose BOX 3.1 Advantages of SOEs Identified in a Systematic Review of Core Analytics across the World Bank A systemic review of core analytics covering 58 middle-income countries finds that almost percent of state-owned enterprises (SOEs) enjoy some type of advantage. Implicit advantages 90 ­ are the most common in the sample. Within this group, limits on entry are prevalent. Explicit advantages are observed in a third of the sample. SOEs commonly face lower tax burdens and can  receive financing (such as government guarantees) not available to private companies. Figure B3.1.1 describes the implicit and explicit advantages for SOEs. Explicit advantages SOEs can benefit from financial and procurement advantages. For example, in Jamaica SOEs may benefit from public guarantees with the approval of parliament (Public Bodies Management and Accountability Act, Article 5A), and the government regularly subsidizes the debt of SOEs and absorbs their losses (IFC 2022). A joint stock company in Kazakhstan has special rights not afforded to other companies, such as the preemptive right to buy strategic facilities and bankrupt assets, and is exempt from government procurement procedures (IFC 2017). Implicit advantages As examples of limits on entry, many SOEs enjoy monopoly rights in network industries.a Firms typically acquire many of their inputs—transportation, energy, telecommunications, financial ser- vices (that is, network industries)—in local markets. If these upstream markets lack competition, goods and services needed for production are not priced competitively. As a result, firms may be less competitive than their foreign rivals and GDP growth may suffer. Reforms to open key markets to competition have boosted productivity and growth (Kitzmuller and Licetti 2013). Governments typically determine “strategic sectors” that are fully reserved either for SOEs or for joint ventures with limited private ownership of at most 49 percent. In Indonesia, SOEs hold monopoly or quasi- monopoly positions in sectors where competition would be viable, such as energy generation and distribution, seaports, toll roads, and imports of agricultural inputs and cereal grains (IFC 2019). These positions expand to public ­ procurement contracts, as is the case in Bolivia, where only  majority-owned SOEs can obtain contracts with the Ministry of Public Works, Services, and  Housing for the execution of social inclusion projects in the telecommunications sector (IFC 2021b). Inadequate protection against anticompetitive firm behavior Key sectors of the economy are excluded from the Malaysia Competition Commission’s antitrust scrutiny, including specific sectors such as telecommunications and energy that are subject to sectoral rules, as well as all enterprises entrusted with operating services of general economic (Box continues on the following page.) 50 The Business of the State BOX 3.1 Advantages of SOEs Identified in a Systematic Review of Core Analytics across the World Bank (continued) FIGURE B3.1.1 Advantages Identified in a Systematic Review of Core Analytics across the World Bank a. Type of advantages b. Implicit advantages Inadequate SOE is involved in protection against regulatory decisions 37 27 anticompetitive firm behavior Price or quota regulation 37 Implicit advantages 40 Limits on entry 62 Explicit advantages 32 Import restrictions 25 0 10 20 30 40 50 0 20 40 60 80 % of countries % of countries c. Explicit advantages d. Inadequate protection against anticompetitive firm behavior Exemptions from or Differential, insufficient, or weak enforcement 28 inadequate implementation of laws of competition regulatory 40 frameworks to SOEs Other state support 20 Lack of access regulation or Procurement appropriate and independent 23 price regulation in natural 18 advantages monopoly sectors Debt and tax advantages 53 No separation of natural Requirements for monopoly segments from 22 operations of SOEs 33 competitive sectors 0 10 20 30 40 50 60 0 10 20 30 40 50 % of countries % of countries Source: Analysis of 57 World Bank Country Private Sector Diagnostics, 2 Country Economic Memorandums, and 1 Integrated SOE Framework published between 2017 and 2023. Note: SOE = state-owned enterprise. interest or having the character of a revenue-producing monopoly.b Mozambique’s Council of Ministers establishes that SOEs must abide by the competition law; however, the implementation rules are still unclear (IFC 2021a). In Moldova, as in many client countries, the competition author- ity still lacks sufficient human and material resources to ensure all its attributions, including fully fledged enforcement of competition rules (IFC 2023). a. Network industries are defined as those industries in which a fixed infrastructure is needed to deliver the goods or services to end users (OECD 2000). b. Competition Act of Malaysia, Article 3 on application, which excludes activities included in schedules 1 and 2. Crowding Private Firms In—or Out?51 no  requirement for SOEs to separate their noncommercial from their commercial activities (figure 3.4). Of 28 middle-income and 40 high-income countries with data on separating commercial from noncommercial activities in all sectors, only very few (4 percent) are middle-income countries; full separation is more frequent in high- income countries (23 percent). The separation of activities is more frequent in some sectors and more prevalent in high-income countries (53 percent) than in middle- income countries (21 percent).3 For differential application of policies and regulations, product market regulation (PMR) data for 27 middle-income countries and 39 high-income countries suggest that SOEs  commonly receive economic and regulatory advantages, particularly in middle-income countries (figure 3.5). SOEs frequently receive economic advantages, such as access to finance on better terms than private operators. In Belarus, SOEs receive preferential tariffs for electricity and other public utility services. In Ukraine, direct subsidies to SOEs represented about 1.3 percent of gross domestic product in 2015, with direct subsidies to the coal and energy sector alone accounting for about 1  percent of gross domestic product (Pop et al. 2019). Public national air ­ carriers benefit from preferential treatment for fuel prices in Kuwait and time slots in the Arab Republic of Egypt (Arezki et al. 2019, 55). FIGURE 3.4  Requirements for Separating Commercial and Noncommercial Activities, by Country Income Level 100 21 80 4 52.5 % of countries 60 40 75 22.5 20 25 0 High-income countries Middle-income countries (40) (28) No Yes (in all sectors) Yes (only in some sectors) Sources: OECD product market regulation (PMR) questionnaires, 2018; OECD and World Bank PMR questionnaires, 2013–22; and data collected by the World Bank on selected PMR questions for eight Middle East and North Africa countries. Note: High-income countries are Australia, Austria, Belgium, Canada, Chile, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, the Republic of Korea, Kuwait, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, New Zealand, Norway, Poland, Portugal, Romania, Saudi Arabia, the Slovak Republic, Slovenia, Spain, Sweden, Switzerland, the United Arab Emirates, and the United Kingdom. Middle-income countries include both upper-middle-income and lower-middle-income countries. Upper-middle-income countries are Albania, Argentina, Belarus, Brazil, Bulgaria, China, Colombia, Costa Rica, Ecuador, Jordan, Kazakhstan, Kosovo, Malaysia, Mexico, Moldova, Montenegro, North Macedonia, Peru, the Russian Federation, Serbia, South Africa, and Türkiye. Lower-middle-income countries are Côte d’Ivoire, the  Arab Republic of Egypt, Indonesia, Morocco, Tunisia, and Viet Nam. OECD = Organisation for Economic Co-operation and Development. 52 The Business of the State FIGURE 3.5  Advantages Available to SOEs over Private Firms, by Country Income Level a. Access to finance on better terms b. Receipt of regulatory advantages 100 18 19 20 22 80 8 7.5 % of countries 26 60 48 40 74 72.5 52 20 33 0 High-income countries Middle-income countries High-income countries Middle-income countries (38) (27) (39) (27) No Yes (in all sectors) Yes (only in some sectors) Sources: OECD product market regulation (PMR) questionnaires, 2018; OECD and World Bank PMR questionnaires, 2013–22; and data collected by the World Bank on selected PMR questions for eight Middle East and North Africa countries. Note: High-income countries are Australia, Austria, Belgium, Canada, Chile, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, the Republic of Korea, Kuwait, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, New Zealand, Norway, Poland, Portugal, Romania, Saudi Arabia, the Slovak Republic, Slovenia, Spain, Switzerland, the United Arab Emirates, and the United Kingdom. Middle-income countries include both upper-middle-income and lower-middle-income countries. Upper-middle-income countries are Albania, Argentina, Belarus, Brazil, Bulgaria, China, Colombia, Costa Rica, Ecuador, Jordan, Kazakhstan, Kosovo, Malaysia, Mexico, Moldova, Montenegro, North Macedonia, Peru, the Russian Federation, Serbia, South Africa, and Türkiye. Lower-middle-income countries are Côte d’Ivoire, the Arab Republic of Egypt, Indonesia, Morocco, Tunisia, and Viet Nam. Panel a does not include Kuwait or Tunisia. Panel b does not include Albania. OECD = Organisation for Economic Co-operation and Development; SOEs = state-owned enterprises. Regulatory advantages for SOEs are less common than economic advantages, but they still prevail in many middle-income countries, frequently exempting SOEs from all or some laws and regulations that apply to private firms. Examples include exemptions from economywide commercial and corporate laws in Tunisia, bank- ruptcy laws in Kuwait, and electricity regulation in Morocco. Egypt, Kuwait, and Tunisia also exempt or exclude SOEs from the competition law. For instance, the competition regulatory frameworks include exemptions granted to certain catego- ries of SOEs, such as utilities. In most countries, SOEs are not incorporated as lim- ited liability companies. In addition, corporate laws themselves might provide advantages to SOEs or open the door to conflicts of interest, even if unintended. SOEs might be ruled under a sepa- rate corporate framework or under the general commercial framework, especially for private firms with a minority state shareholding. Questions remain about the applica- tion of bankruptcy law due to the priority of public credits, the nonseizure of certain public assets, and the difficulties of executing decisions against the state. And conflicts of interest for the administrator or even the board might pertain to formal or informal instructions from the state or to their limited responsibility in decision-making. Crowding Private Firms In—or Out?53 Channel 2: Implicit Advantages to BOSs Implicit advantages for BOSs include policies or regulations that restrict entry and rein- force dominance, such as trade and foreign direct investment restrictions, quotas, or market outright limits to entry. They also include regulations that dampen rivalry in the ­ (such as price controls) and policies that raise the risk that BOSs or the regulator will restrict competition, particularly when the market is partially contestable. Systematic evidence on the presence of implicit advantages for BOSs is scarce. But PMR qualitative data for 24 middle-income and 37 high-income countries suggest that, in middle-income countries, entry is more often restricted through laws and regulations in network industries, where SOEs are fairly common. Middle-income countries, much more often than high-income countries, limit the number of opera- tors in various network industries where competition is viable, including by creating local or national monopolies. Moreover, ownership rights and regulations are not always separated, increasing the likelihood of an uneven playing field. PMR data for 24 middle-income and 34 high-income countries show that the body that regulates a sector with SOEs is, in almost 40 percent of middle-income countries, not separated from the one that e ­xercises ownership rights in the SOEs. In Egypt, for example, the National Telecommunications Regulatory Authority falls under the authority of the Ministry of Communications and Information Technologies, which also owns 80 percent of the public telecommunications operator. In some cases, the SOE regulates itself. In South Africa, Transnet acts as both the port regulator and the owner-operator of all major commercial ports. Channel 3: Policy and Regulatory Distortions Favoring BOSs Some policies and regulations are typically required in natural monopoly sectors to prevent the undue exercise of market power and ensure that the monopoly has incen- situation tives to invest. When these regulations are missing or poorly implemented, this ­ is likely to lead to high prices, lower access to goods or services, or underinvestment. Although these policies and regulations would be required regardless of firm owner- ship, key regulatory safeguards may be enacted less frequently or implemented less sufficiently when the natural monopoly is an SOE, partly because of conflicts of interest and soft budget constraints.4 The lack of structural separation paired with weak implementation of third-party access regulation can insulate SOE incumbents from competition. Separating natural monopolies from partially contestable markets would allow for competition and restrain the natural monopoly from leveraging its market power in adjacent markets. But, in the electricity and gas sectors, many countries do not have vertical separation between transmission and generation or supply. This is more prevalent in middle- income countries than in high-income ones. 54 The Business of the State Advantages to BOSs in Competitive Sectors Combining data on BOS presence with sector information on government policy makes it possible to assess systematically whether governments tilt policies in favor of BOSs. Governments generally intervene indirectly in markets by regulating ­ economic activity. When governments own enterprises or engage in public pro- curement, they intervene directly in the market, by selling or buying goods or services. Given this dual role, governments may have incentives to tilt policies in ­ favor of the firms they own. Trade-Related Barriers to Competition The increased prominence of BOSs in global markets has drawn attention to the poten- tial protection of BOSs through policies and import restrictions that advantage BOSs in exporting activities. Governments may protect BOSs through tariffs, import quotas, and other nontariff barriers. Some countries have higher ad valorem equivalent tariffs or nontariff measures in sectors with BOSs than in sectors without a BOS presence (maps 3.1 and 3.2). Across all sectors, more than half of the sample countries have higher tariffs in BOS sectors than in private sectors. In Kenya, Namibia, the Seychelles, Tanzania, and Türkiye, sectors with a BOS presence have more than 5 percent higher tariffs, a ­ ­ significant dif- ­ ectors with a BOS ference. In Ecuador, Paraguay, the Philippines, Tunisia, and Türkiye, s presence have higher ad valorem equivalent nontariff measures. MAP 3.1  Tariffs in Sectors with and without BOSs Di erence in tari s in sector with BOS presence –5 5 IBRD 47467 | AUGUST 2023 Sources: Data from the United Nations Conference on Trade and Development Trade Analysis Information Systems (TRAINS) data set and the World Bank Global Businesses of the State (BOS) database. TRAINS data are for 2012–20; BOS data are for 2022. Note: BOSs = businesses of the state. Crowding Private Firms In—or Out?55 MAP 3.2 Ad Valorem Equivalent of Nontariff Barriers in Sectors with and without BOSs Di erence in nontari measures in sector with BOS presence –1 1 IBRD 47468 | AUGUST 2023 Sources: Data from Kee and Nicita 2022; World Bank Global Businesses of the State (BOS) database. Data cover from the early 1980s until August 2022. Note: BOSs = businesses of the state. Some sectors with BOSs have more trade restrictions. The Organisation for Economic Co-operation and Development (OECD) Services Trade Restrictiveness Index (STRI)5 for telecommunications shows that state ownership in telecommunica- tions is not linked to more service trade restrictions.6 However, in air transportation, service trade restrictions are higher in countries with a state-owned carrier. Further, in countries with a state-owned carrier that has been privatized, the year of its privatiza- tion is correlated with trade restrictiveness: the more recent the privatization, the more restrictive the current policies. There is also evidence that advantages are conferred to BOSs in export markets.7 Some trade remedy cases brought against BOSs in the World Bank Trade Remedies Data Portal show that government preferential treatment directly or indirectly sup- ports exports. These policy measures, often referred to as trade remedy actions, are implemented by government authorities against imports likely to have an adverse effect on national production, either through dumping or subsidies. Domestic Competition Barriers World Bank Global Businesses of the State (BOS) and Anticompetitive Laws and Policies (ALP) data show that regulatory restrictions on competition are linked to the presence of BOSs in competitive markets. The ALP data collected for this report cover major barriers to competition that relate to regulations restricting entry, reinforcing dominance, facilitating collusion, or protecting vested interests.8 Overall, regulatory 56 The Business of the State barriers to competition are present in 75 percent of the country-sector pairs in the ALP data. Restrictions are slightly more frequent in sectors with BOSs, with 78 percent of them in BOS ­sectors. But the presence of BOSs in a sector does not automatically imply regulatory barriers to competition. In 25 percent of the sectors with BOSs, market reg- ulations do not appear to restrict competition in major ways. The findings underline the importance of regulatory reforms to accompany ownership reforms. The findings also suggest that regulatory reforms may be more difficult in BOS sectors, potentially because of strong vested interests and direct links between state enterprises and regulators. The ALP data demonstrate that BOSs and SOEs benefit from both explicit and implicit advantages in competitive sectors (box 3.2). For example, SOEs are granted exclusive rights, are involved in regulating the sector, or are exempt from economywide laws—all explicit advantages that are not available to private competitors. Implicit advantages that apply to BOSs and private competitors alike are even more prevalent. These  advantages affect entry or regulate the conduct of all market players in competitive sectors, including BOSs. In a few cases, explicit and implicit advantages are combined, giving certain SOEs full control over the market. BOX 3.2 Advantages Identified by Anticompetitive Laws and Policies Data Explicit advantages ■ Exclusive rights. Vietnamese law requires the signing of production-sharing contracts for oil and gas exploration and extraction with a state-owned enterprise (SOE) that is also engaged in downstream sectors. Although the participation of SOEs in oil and gas sectors is some- times required upstream, risks are often borne disproportionately by private investors, giv- ing SOEs involved in downstream sectors an explicit advantage over competitors. In electricity distribution, the provision of services is also reserved for SOEs in some countries.a ■ SOE involvement in regulation. In Ethiopia, an SOE is involved in price regulation and market allocation, while being the sole importer of fertilizer in the country. In Viet Nam, an SOE is one of several market participants in the oil and gas and cement sectors, but it is authorized by law to supervise its competitors. In Angola (cement) and Serbia (intercity buses), SOEs are authorized by law to participate in sector regulation together with private competitors.b In these cases, the public authority exercising ownership rights is also the regulator, giving it an outsized influence over policy.c ■ Exemptions. In several countries, SOEs are exempt from economywide laws. In Tunisia, SOEs operating in monopoly sectors are exempt from the competition law. The exemption also covers sectors in which competition is viable, such as the supply of electricity. In Angola, public procurement rules are waived for contracts with SOEs. And, in Pakistan, direct government-to-SOE contracting is permitted. There are also SOE-specific exemp- tions from bankruptcy laws in several countries, such as Serbia. (Box continues on the following page.) Crowding Private Firms In—or Out?57 BOX 3.2 Advantages Identified by Anticompetitive Laws and Policies Data (continued) Implicit advantages ■ Entry regulations. In Angola and Ethiopia, firm entry into the distribution of cement and fertilizer, respectively, is limited. In Argentina and Nepal, firms can start to manufacture petroleum products by official initiative only.d In air transportation, foreign direct invest- ment (FDI) restrictions are ubiquitous, and most apply in sectors with state-owned domestic carriers. Of the 10 countries covered by the Anticompetitive Laws and Policies data set, FDI restrictions exist in 9 countries, and state-owned carriers serve domestic routes in 8 countries.e ■ Conduct regulations. The presence of SOEs coincides with price controls in some cases. In most countries, the wholesale and retail prices of gasoline are regulated (either capped or fixed), although indexing mechanisms are common (Baig et al. 2007; Gillingham and Keen 2012; Kpodar and Imam 2020). Where prices “at the pump” are regulated, SOEs are usually present in extraction, refining, or distribution.f Cement prices are regulated in Angola and Ethiopia, which have strong, direct government participation in the sector.g In Angola and Argentina, state ownership in transportation sectors also coincides with price controls. a  . In Ethiopia, for instance, investment laws restrict the transmission and distribution of electrical energy to the government through the integrated national grid system. The service is provided by the Ethiopian Electric Utility. In Mexico, only the Comisión Federal de Electricidad, a state-owned utility, can distribute electricity. b. In coach transportation in Serbia, the SOE and private incumbents are authorized by law to participate in decisions over  route allocation. In Angola, the SOE is one of several cement manufacturers represented in the industry association. The association has the right to advise on import authorizations. c. This risk is heightened by the “revolving door” between SOEs and government.  d. Additionally, in Argentina entry is by official initiative only in the cement, electricity, and air transportation sectors. In  Nepal, imports of petroleum products are also restricted.  e. The eight countries that restrict SOEs and FDI in air transportation are Angola, Argentina, Ethiopia, Nepal, Pakistan, Serbia, Türkiye, and Viet Nam. Peru restricts FDI through equity caps, but has no state-owned carrier. f. Nine countries have BOSs and price controls in the oil and gas sector: Angola, Argentina, Ethiopia, Nepal, Pakistan, Peru,  Serbia, Türkiye, and Viet Nam. Price controls include permanent and temporary measures. The Philippines has an SOE in the oil and gas sector, but no price controls.  g. Four countries (Argentina, Nepal, Pakistan, and Türkiye) have cement BOSs and no price controls. In the remaining three countries (Peru, the Philippines, and Serbia), cement prices are not regulated, and the government does not participate in the sector directly. Advantages in Network Sectors Network industries are different from competitive sectors in that firms are inherently more likely to outgrow their rivals. Incumbents in network sectors have strategic advantages over their competitors because of characteristics specific to these indus- tries. Economies of scale, technological bottlenecks, network effects, and scarcities cause firms to be larger and more dominant. Such firms also have incentives to exploit their advantages in adjacent markets and thus become multisectoral. 58 The Business of the State States can apply a range of policies to contain the market power of network operators and have intervened in network sectors primarily as owners. Policies in network sectors range from participation in markets through BOSs or public pro- curement to intervention as regulators or referees. In natural monopoly sectors, for instance, firm ownership allows the government to set prices below the monopoly level, thus increasing output and market efficiency. The state can also leave the operation of infrastructure to the private sector and regulate the prices that domi- nant operators charge. And the state can intervene to unbundle natural monopoly activities from competitive activities, thus opening some market segments to competition. State ownership and restrictive regulations are correlated in middle-income coun- tries. Whether state ownership, specifically SOE presence, is associated with (the lack of) pro-competition regulation in network sectors can be studied with two measures in the PMR indicators. The first is an index of government involvement in network sec- tors, which includes information about state ownership, divestment procedures, and special voting rights. The second is an index of the degree to which market regulations inhibit or promote competition. Overall, the link between direct and indirect state intervention in markets is stronger in middle-income countries than in high-income countries (figure 3.6). This difference is strongest in telecommunications, where nearly FIGURE 3.6  State Involvement and Restrictive Regulations, by Country Income Level, 2018 6 5 Government involvement index R² = 0.2423 4 3 R² = 0.0242 2 1 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Regulatory restrictiveness index High-income countries Middle-income countries Sources: OECD product market regulation (PMR) questionnaires, 2018; OECD and World Bank PMR questionnaires, 2013–22; and data collected by the World Bank on selected PMR questions for eight Middle East and North Africa countries. Note: The data cover energy, communications, and transportation in 39 high-income economies and 23 middle-income economies. The maximum value for the regulatory restrictiveness index is 6. The scores on regulatory restrictiveness do not contain information about state ownership. The value for the United States was calculated as an average of the values for the states of New York and Texas. OECD = Organisation for Economic Co-operation and Development. Crowding Private Firms In—or Out?59 all middle-income countries cluster around weak state involvement and strong pro- competition regulation or around strong state involvement and weak pro-competition regulation. State involvement and market regulations are also correlated in the energy and transportation sectors, but the differences between high-income and middle- income countries are smaller in these sectors. Disciplining Anticompetitive Behavior Authorities have been applying competition laws to BOSs; however, when it comes to SOEs, competition enforcement often focuses on strategic sectors in which governments have an interest in protecting efficient market outcomes. Case studies ­ can help explain how SOEs are treated when they ■■ Conduct exclusionary behavior that prevents new entry, forces existing rivals from the market, or otherwise diminishes competition through the abuse of a firm’s dominant position in the market; ■■ Engage in anticompetitive agreements with rival or vertically linked firms through collusion or vertical restrains; or ■■ Attempt to implement mergers or acquisitions that may significantly hinder ­ competition in the market. To this end, the World Bank collected 224 decisions on each of these categories involving at least one SOE.9 The majority of cases are about abuses of dominance, and two-thirds of the decisions collected pertain to only four sectors: energy, telecommu- nications, transportation, and natural resources (figure 3.7). In other words, they cover practices in sectors in which the state would have a special interest in maintain- protect consumer welfare because ing competitive pressures that foster efficiency and ­ they are key inputs for other markets. Firms typically acquire important inputs— transportation, energy, telecommunications, and financial ­ services—in local markets. Thus, weak competition in upstream markets may cause domestic firms to be less competitive than their foreign rivals. The case studies confirm that authorities have enforced competition to discipline the behavior of SOEs for collusion, abuses of dominance, and potential negative market effects of mergers and acquisitions or privatizations. They also uncover enforcement and political economy challenges related to the nature of SOEs—including exemptions from competition law, interference from public authorities, and difficulties in collect- ing fines. Abuses of Dominance Abuse of dominance cases involving SOEs often aim at disciplining attempts by vertically integrated public incumbents to limit entry into market segments in which competition is possible. Given the position of SOEs as a monopoly or an incumbent 60 The Business of the State FIGURE 3.7  Anticompetitive Conduct, by Type of Conduct, Sector, and Country Income Level a. By type of conduct b. By sector 100 21 22 4 80 6 6 12 2 24 16 15 10 % of cases 60 13 21 40 26 54 62 15 20 20 23 0 High-income countries Middle-income countries High-income countries Middle-income countries (94) (130) (94) (130) Abuse of dominance Energy Banking and financial services Anticompetitive agreements Telecommunications Postal and communication services Mergers and acquisitions Transportation Manufacturing and industrial markets Natural resources Other Source: Based on 224 case studies collected from the early 1980s until August 2022 on competition and SOEs. Note: Numbers in parenthese are number of case studies conducted. In panel b, “Other” includes health, waste management, and lottery and gaming sectors. SOEs = state owned enterprises. with political support and access to resources, abuse of dominance is the main area of enforcement for SOEs (Moroccan Competition Council 2014). For instance, of 21  abuses of dominance referred to the Competition Tribunal between 1999 and 2016 in South Africa, 13 involved former or current SOEs, with 10 initiated ex officio by the Competition Commission (Healey 2018, 6; Roberts 2017). Denying access to  essential facilities and limiting interconnection are common practices across countries and sectors. Such practices include denying access to the wholesale market for broadband internet in France10 and South Africa,11 refusing access to the national network of gas pipelines in Italy (AGCM 1999),12 and limiting access by the railway company to underground or aerial pipelines to crossing railways in Chile.13 Moreover, price-based abuses, which amount to only about a third of the cases in middle- income countries, focus mostly on abusive or discriminatory pricing and less on predatory pricing and margin squeezes. These practices are often facilitated by regulation, thus confirming the risks of conflicts of interest when it comes to the relationships between SOEs and the authorities. Anticompetitive Agreements A comparison of cartels involving SOEs with those involving only private operators shows that, on average, the former are more stable and require longer investigations for competition authorities to issue a decision. Of the almost 939 cartel cases in the World Crowding Private Firms In—or Out?61 Bank Anti-Cartel Enforcement database at the global level in the past five years, only 18 involved SOEs.14 This result is consistent with the case studies analyzed in this chapter, which confirm that SOEs seldom participate in cartels—or that fewer cartels involving SOEs have been investigated than those involving private operators only. Why might this be? SOEs need not obtain market power through a cartel because they often have prominent market positions. And public operators might be reluctant to violate the law, which could explain why a third of the cartel cases involve leniency applications.15 But nine of these cartels involved bid rigging when SOEs participated in public tenders (building roads or pipelines for gas transmission or providing steel), thus directly hitting the public budget. Mergers and Acquisitions That May Significantly Hinder Competition in the Market Exemptions for merger control for SOEs, together with potential interference of pub- lic authorities in decision-making, can limit the benefits of market consolidation. PMR data confirm that, in most countries—high-income and middle-income coun- tries alike (85 percent in a sample of 39 high-income countries and 27 middle-income countries)—mergers, equity issues, and restructuring plans of SOEs are cleared by national or subnational governments. The application of public interest clauses may also distort the level playing field in merger control if they prohibit mergers that harm the economic activity of the state-owned domestic players (OECD 2016). In addition, authorities can interfere in decisions, leaving technical considerations on the side. For example, in Colombia in 2001, the president overruled the decision by the Competition Commission blocking the merger between the then-public national airline and a smaller company (Healey 2018). International experience confirms several outstanding challenges when applying antitrust and merger control to SOEs. Five challenges need to be addressed to leverage competition regulatory and institutional frameworks to level the playing field: 1. Potential conflicts of interest with regulators 2. Difficulties distinguishing BOS decisions in relation to instructions from public authorities and actions based on regulation 3. Difficulties disciplining the anticompetitive behavior of foreign BOSs 4. Need for adapting antitrust tests and analytics to SOEs, given public mandates 5. Prominent use of remedies rather than fines, because half of the antitrust cases against SOEs established violations but did not include fines, especially in middle-income countries, or the fines were low Complementary mandates to traditional antitrust tools—such as the ability of authorities to advocate for pro-competition principles (that is, competition advocacy)— can help to shape BOSs’ behavior. Advocacy can address BOS conduct outside the scope of competition law or when enforcement has been proven unsuccessful or unlikely. Advocacy can also help to identify markets in which the presence of BOSs might be 62 The Business of the State distortive and to design pro-competition processes to phase out BOS presence or bring in private sector participation, through privatization or public-private partnerships. The International Competition Network—World Bank Competition Advocacy Contest data set confirms the existence of advocacy contest measures targeting SOEs, but they remain limited in number.16 Of 265 cases gathered from 62 jurisdictions by 2022, only 31 cases focused on competitive neutrality, and only 10 focused specifically on SOEs. Examples of advocacy targeting SOEs, beyond the contest itself, include monitoring recently liberalized markets and complementing enforcement. In  Moldova, the electricity market was monitored continuously after privatization of  the distribution system operator, and competition authorities eventually recommended unbundling the vertically integrated incumbent to increase competition in partially contestable markets.17 In China, the competition authorities created a fair competition review mechanism to identify restrictions on competition in regulation that could potentially affect many markets with an SOE presence. Systematic mechanisms to analyze whether BOSs in commercial markets fulfill a subsidiary role to that of the private sector can focus BOS participation in markets where they are needed. According to the subsidiarity principle, if private agents are interested and capable of supplying enough goods and services to meet demand, direct participation of the state might not be needed. Instead, the state may intervene in those markets only by supervising and controlling the behavior of private agents. Although few countries have an explicit mandate to perform this assessment, adopting the sub- sidiarity principle offers complementarities with competition law. Notes 1. This chapter differentiates between businesses of the state (BOSs) and state-owned enterprises (SOEs), in line with definitions detailed in box 1.1 of chapter 1. The term BOSs is used conceptually to discuss the participation of the state in markets as an important stakeholder, ­ including firms with minority state ownership starting at 10 percent, as well as participation in firms that are directly as well as indirectly owned by the state through another company. The term SOE is used when referring to existing literature and empirical work that use the term and when referring to country-specific SOE policies and reform agendas that are aligned to a country’s own definition. These definitions are often limited to firms owned by the central government, with direct state ownership of 50 percent or more. 2. Such rules should ideally be complemented by a requirement to show a positive rate of return. However, previous competitive neutrality assessments by the World Bank suggest that such a requirement is rare in lower-income countries. And, even when some requirements exist, they are often not laid out in regulations. In Morocco, for example, some contract programs negotiated on a case-by-case basis require a positive net present value (NPV), and a positive NPV is required for a BOS when creating a new subsidiary. But BOSs are not systematically required to achieve a commercial rate of return. In Sri Lanka, the state directs BOSs to make a positive NPV in their investments, but this requirement is not set out in the regulations. 3. Requirements for separation are more common in European countries, although some countries such as Albania, Moldova, North Macedonia, and Norway have no such requirement. In Latin America and the Caribbean, only Mexican BOSs in select sectors are required to separate Crowding Private Firms In—or Out?63 commercial from noncommercial activities; in the Middle East and North Africa, many BOSs deliver both commercial and noncommercial services, but no country requires separation between these services. In East Asia and Pacific, China, Indonesia, Japan, the Republic of Korea, and Malaysia have no requirements in any sector. 4. Maskin (1996) provides a detailed explanation of soft budget constraints. 5. The OECD STRI reports on services trade restrictions in 19 major services sectors in 50 countries (38 OECD and 12 non-OECD). The index covers restrictions on foreign entry, movement of people, other discriminatory measures, barriers to competition, and regulatory transparency. The OECD STRI is matched with state ownership data from the OECD PMR and the International Civil Aviation Organization. 6. However, the most restrictive countries have state ownership in the telecommunications ­ sector, suggesting that, although state ownership is not a sufficient condition for protections in the sector, it may be a necessary condition. Indonesia, Kazakhstan, and the Russian Federation have the highest level of restriction in the telecommunications sector, and all have state ownership in the sector. 7. The World Bank Trade Remedy data set covers countervailing and antidumping cases in more than 30 countries. The data set reports on cases from the 1980s through 2019 and includes all investigation actions initiated or measures implemented by countries, including a significant amount of information on the initiating cases (and the timing of their progression), the product or product group involved, the national petitioners and foreign entities targeted, and the result- ing measures applied. The time series in the database account for measures that may have been suspended over time. 8. The ALP data cover 10 countries: Angola, Argentina, Ethiopia, Nepal, Pakistan, Peru, the Philippines, Serbia, Türkiye, and Viet Nam. In each of the countries, data have been collected for 14 competitive sectors at the Statistical Classification of Economic Activities in the European Community (NACE) four-digit level based on a review of publicly available primary and second- ary regulations. The data have been aggregated into seven sector groups: upstream oil and gas, downstream oil and gas, electricity, cement, fertilizer, air transportation, and road transportation. The data cover laws on the books (de jure) but not their implementation (de facto). 9. The World Bank data set of case studies on competition and SOEs includes 224 decisions of com- petition authorities in which an SOE was involved in both anticompetitive conduct and economic concentration from the early 1980s until August 2022. 10. Decision 05-D-59, dated November 7, 2005. 11. Case 11/CR/Feb04. In South Africa in 2002, the Competition Commission received a complaint against Telkom from the South African Vans Association and 20 other internet service providers alleging that Telkom was abusing its dominance to exclude (existing and potential) competitors in the downstream market for internet services. The commission and later the Supreme Court of Appeal found that Telkom had abused its dominance by refusing to grant competitors access to an essential facility (in contravention of section 8(b) of the Act) and that it had induced customers not to deal with competitors (contravening 8(d)(i) of the Act). 12. The Italian Competition Authority also found that the method of calculating the charge used allowed the BOS firm to fix the price independently of effective demand for the transport of third parties’ gas and was likely to lead to the imposition of unjustifiably burdensome contractual conditions. 13. Ruling No. 76/2008, dated October 14, 2008. 14. The World Bank Anti-Cartel Enforcement Database covers all decisions from competition agen- cies related to hard-core cartels around 75 jurisdictions worldwide. The database covers the universe of cartels investigated and closed with an official decision by the competition agen- cies in first instance for both privately owned and state-owned enterprises. In total, 939 cases were mapped between 2017 and 2022 (March) and closed by the competition agencies, including 64 The Business of the State 18 investigations involving SOEs. Out of these cases, 6 involved leniency applications. In 90 per- cent of the cases, SOEs were sanctioned with fines, which amounted to US$566 million. 15. Leniency programs allow the first cartel member approaching the competition authority with information on a cartel to be exempted from the fine. A well-functioning leniency program can destabilize and deter cartels by creating a permanent threat that any of its members may come forward to the authority in order to avoid the fine. The programs that have proven to be the most successful give complete amnesty to the first cartel member to come forward and reveal the inner workings of the cartel to competition law enforcers. Thus, most competition agencies around the world, including in Brazil, Canada, Chile, China, Colombia, the European Union, India, Korea, Mexico, Türkiye, and the United States, have the ability either to offer total immunity or to reduce significantly the fines for violators that cooperate in their investigations. Miller (2009) finds that the leniency program in the United States reduced the rate of cartel formation by 59 percent and increased the rate of cartel detection by 62 percent. 16. The International Competition Network—World Bank Competition Advocacy Contest database consists of submissions of successful advocacy initiatives from competition authorities, ­ regulators, and public bodies to the International Competition Network—World Bank Group Competition Advocacy Contest from 2014 onward. 17. In addition, the Competition Commission recommended expediting the process of revising the tariff customer agreement and setting obligations for the public retail supplier in order to protect consumers through observance of service quality indicators (supply vs. interruption of electricity, accurate billing). References AGCM (Italian Competition Authority). 1999. “Provvedimento n 6926 (A221) SNAM-Tarriffe DI Vettoriamento.” Official Bullettin 8/1999, February 25, 1999. Arezki, R, A. Barone, K. Decker, D. Detter, R. Yuting Fan, H. Nguyen, and G. Murciego. 2019. Reaching New Heights: Promoting Fair Competition in the Middle East and North Africa. Middle East and North Africa Economic Update, October 2019. Washington, DC: World Bank. https://elibrary​ .worldbank.org/action/showCitFormats?doi=10.1596%2F978-1-4648-1504-1&mobileUi=0. Baig, T., A. Mati, D. Coady, and J. Ntamatungiro. 2007. “Domestic Petroleum Product Prices and Subsidies; Recent Developments and Reform Strategies.” Working Paper 07/71, International Monetary Fund, Washington DC. Gillingham, R., and M. Keen. 2012. “Mitigation and Fuel Pricing in Developing Economies.” In Fiscal Policy to Mitigate Climate Change. A Guide for Policymakers, edited by I. W. H. Parry, R. de Mooij, and M. Keen, ch. 6, 103–32. Washington, DC: International Monetary Fund. Healey, D. 2018. “Competition Law and State-Owned Enterprises: Enforcement.” DAF/COMP/GF​ (2018)11. Paper prepared in support of a presentation for Session V at the 17th Global Forum on Competition, OECD, Paris, November 29–30, 2018. IFC (International Finance Corporation). 2017. “Creating Markets in Kazakhstan.” IFC, Washington, DC. https://www.ifc.org/en/insights-reports/2017/creating-markets-in-kazakhstan. IFC (International Finance Corporation). 2019. “Creating Markets in Indonesia: Unlocking the Dynamism of the Indonesian Private Sector.” Country Private Sector Diagnostic. IFC, Washington, DC. https://www.ifc.org/content/dam/ifc/doc/mgrt/201910-cpsd-indonesia-v2.pdf. IFC (International Finance Corporation). 2021a. “Creating Markets in Mozambique: A Study Conducted in Partnership with SIDA.” Country Private Sector Diagnostic, IFC, Washington, DC. https://www.ifc.org/en/insights-reports/2021/cpsd-mozambique. IFC (International Finance Corporation). 2021b. “The Plurinational State of Bolivia: Unlocking Private Sector Potential to Achieve a Sustainable and Inclusive Recovery.” IFC, Washington, DC. https://www.ifc.org/en/insights-reports/2021/cpsd-bolivia. Crowding Private Firms In—or Out?65 IFC (International Finance Corporation). 2022. “Creating Markets in Jamaica: Country Private Sector Diagnostic.” IFC, Washington, DC. https://www.ifc.org/en/insights-reports/2022/cpsd-jamaica. IFC (International Finance Corporation). 2023. “Creating Markets in Moldova: From a Remittances- Driven Economy to Private Sector–Led Sustainable Growth.” Country Private Sector Diagnostic, IFC, Washington, DC. https://www.ifc.org/en/insights-reports/2023/cpsd-moldova. Kee, H. L., and A. Nicita. 2022. “Trade Fraud and Non-Tariff Measures.” Journal of International Economics 139 (November): Art. 103637. Kitzmuller, M., and M. M. Licetti. 2013. “Competition Policy: Encouraging Thriving Markets for Development.” ViewPoint Public Policy for the Private Sector Series, No. 331, Finance and Private Sector Development Vice Presidency, World Bank, Washington, DC. Kpodar, K. R., and P. A. Imam. 2020. “To Pass (or Not to Pass) Through International Fuel Price Changes to Domestic Fuel Prices in Developing Countries: What Are the Drivers?” Working Paper 2020/194, International Monetary Fund, Washington, DC. Maskin, E. 1996. “Theories of the Soft Budget Constraint.” Japan and the World Economy 8 (2): 125–33. Miller, N. H. 2009. “Strategic Leniency and Cartel Enforcement.” American Economic Review 99 (3): 750–68. https://doi.org/10.1257/aer.99.3.750. Moroccan Competition Council. 2014. “Special Project: State-Owned Enterprises and Competition.” Paper prepared for the International Competition Network Annual Conference, Marrakech, April 23–25, 2014. https://centrocedec.files.wordpress.com/2015/07/soe-and-competition2014​ .pdf. OECD (Organisation for Economic Co-operation and Development). 2000. “IV. Regulatory Reform in Network Industries: Past Experience and Current Issues.” OECD Economic Outlook 67 (1): 151–71. OECD (Organisation for Economic Co-operation and Development). 2016. “Public Interest Considerations in Merger Control.” Background paper prepared by the OECD Secretariat, Paris. DAF/COMP/WP3(2016)3. Pop, G., M. Licetti, S. F. Gramegna Mesa, and S. Dauda. 2019. “Reducing Market Distortions for a More Prosperous Ukraine: Proposals for Market Regulation, Competition Policy, and Institutional Reform.” World Bank, Washington, DC. https://documents1.worldbank.org/curated​ /en​ /368301553112891891/pdf/135463-WP-P169603-PUBLIC.pdf. Roberts, S. 2017. “Assessing the Record on Competition Enforcement against Anti-Competitive Practices and Implications for Inclusive Growth.” Working paper, University of Johannesburg. https://doi.org/10.13140/RG.2.2.26699.59686. Vitale, C., C. Moiso, and I. Wanner. “A Detailed Explanation of the Methodology Used to Build the OECD PMR Indicators.” Organisation for Economic Co-operation and Development, Paris. World Bank. Forthcoming. Markets and Competition Policy Assessment Toolkit. Washington, DC: World Bank. 66 The Business of the State