3 The Privatization of Public Enterprises

Saíd El-Naggar
Published Date:
June 1989
  • ShareShare
Show Summary Details
John Nellis and Sunita Kikeri1

The objectives of this paper are

  • to review the importance and performance, in both financial and efficiency terms, of public enterprises in developing countries;
  • to examine the questions of whether and why performance improvements should result from privatization; and
  • to discuss, in light of the preceding analysis, the approach used by the World Bank concerning privatization.

The central argument of the paper is that public enterprises were created to meet a mixed set of economic, financial, and political objectives; that they have done poorly in fulfilling the first two goals and too well in fulfilling the last; that privatization is an attractive, potentially constructive, but difficult to implement reform mechanism; and that the best means of promoting efficiency gains—the primary concern of the World Bank—lies first and foremost in the enhancement of competitive forces, and secondarily in a change of ownership.

In this study, the term “public enterprise” refers to government-owned (more than 50 percent) and controlled entities which are supposed to “earn most of their revenue from the sale of goods and services, are self-accounting, and have a separate legal identity”2 which distinguishes them from ministerial or purely administrative bodies. Many developing countries include hospitals, universities, training schools, research institutions, and administrative agencies in their lists of public enterprises. This paper excludes such undertakings and deals more narrowly with revenue-generating enterprises.

Public Enterprises

Extent of Sector

Public enterprises meeting the above definition are important financial and economic actors. Worldwide, at the beginning of the 1980s, they were estimated to account, on average, for 10 percent of gross domestic product (GDP) at factor cost.3 Public enterprises have been important in industrial nonsocialist countries as well as in centrally planned and developing countries. For example, the United Kingdom’s public enterprise sector—prior to 1979—accounted for exactly 10 percent of GDP, had £55 billion in turnover, and employed 1.75 million people. In France, after the 1981 nationalizations, public enterprises employed 16.6 percent of all French salaried workers (excluding the agricultural sector), and contributed 17.2 percent of value added (again, outside agriculture), and 33.5 percent of gross fixed capital formation.4 Many other industrial countries, including Austria, Canada, and Japan, have or had at one time substantial public enterprise sectors.

In developing countries, the reliance on public enterprises to achieve socioeconomic goals has been even greater than in industrial countries. Their contributions are significant; for example:

  • in value added in manufacturing, in a great many cases exceeding 50 percent of national totals;
  • in total investment, averaging 25 percent in what appears to be a representative sample of 14 developing countries;5
  • in shares of utilization of developing countries credit systems (at a rough estimate, averaging 30 percent of domestic and 20 percent of total foreign debt); and
  • in nonagricultural employment, averaging about 15 percent in developing countries versus 4 percent in nations of the Organization for Economic Cooperation and Development (OECD).

These attempts to estimate averages mask great regional variations in the size and financial and economic importance of public enterprise (PE) sectors. For example, in a sample of 13 African countries, based on data from the early 1980s, PEs accounted on average for 17 percent of GDP.6 Sketchy evidence indicates that public enterprise sectors in Latin America and Asia fall in between the world average and the higher African figures; and all of the known instances, in the non-centrally planned economies, where the public enterprise share of GDP is above 17 percent, are developing.

Origins in Developing Countries

How did these large PE sectors come about in developing countries? A common path was for them to inherit a public enterprise sector at independence—as colonial administrations were often more economically intrusive and interventionist than governments in the metropole—and then add substantially to it, for reasons both ideological and pragmatic. The former led government officials to mistrust or outrightly oppose the activities of the private sector (especially the foreign private sector, though in many instances the domestic private sector did not escape suspicion). The argument was that to develop, governments needed to hold and lead from “the commanding heights of the economy.” Following this reasoning, socialists regularly argued that public enterprises would generate surpluses which government planners could then invest in high-priority areas. This, it was reasoned, would lead to more rapid and rational development of the economy than would occur if major investment and resource allocation decisions were left to private owners. More pragmatic considerations were numerous: they included the assessment that many countries had no alternative, for the moment, to reliance on PEs. This was because there was no local private sector; or the local private sector was insufficiently developed; or there was no politically acceptable private sector (referring to the widespread developing country problem of the domination of trade and commerce by an ethnic minority). Both the ideological and the pragmatic sets of reasons coincided with the interests of political elites, who often used PEs to generate employment for followers and managerial positions for loyal supporters.

Both ideologists and pragmatists were (and to a large though decreasing extent remain) united in their distaste for direct foreign investment. For these, and many other7 reasons, a very large number to PEs was brought into being in the period 1960–80. For another African example, it is estimated that over half of existing PEs in the sub-Saharan region were created between 1967 and 1980.

Around the world, many of these PEs were direct government initiatives. A fair number of commercial and industrial PEs had been failing private firms, nationalized to prevent their closure and a subsequent loss of employment. And it should be acknowledged that some few of these PEs were created with the support or at the insistence of aid and funding agencies, including the World Bank.8

Performance: Finances

With regard to the crucial issue of performance, there is widespread agreement that PEs have not generally lived up to the expectations of creators and funders. There is of course variation in performance from region to region (and from sector to sector), with Asian countries showing fewer problems than Latin American public enterprise sectors, which, in turn, tend to perform somewhat better than PEs in sub-Saharan Africa. Nonetheless, some generalizations are admissible. Two basic shortcomings are perceived: too many PEs cost rather than make money; and too many operate at low levels of efficiency.

The first standard—money-making or money-losing—is a firm and quantifiable measure of financial performance (assuming there is agreement on accounting standards and calculations). It assesses whether the investor—the owner of the equity—is getting more out of the venture than he or she is putting in. The second is an inherently more relative and more complex measure of economic performance. There are a number of issues related to efficiency, ranging from whether the resources invested in a firm are being used well or optimally to measurements of the net benefits flowing to (or from) the economy as a whole as a result of the investments in an enterprise. Most developing country governments have been far more troubled by financial losses than by low levels of economic efficiency. We deal first with the issue of financial profitability.

Let it be noted and underlined that many public enterprises, in many countries, are making money. A recent survey of 48 PEs in North and sub-Saharan Africa showed that in 1984, 12 of the 48 reported net profit margins in excess of 4 percent, and 5 reported net profit margins of more than 10 percent. Ten of the 48 PEs had returns on shareholders equity at 25 percent or higher.9 The point is not that these are outstanding levels of performance or excellent rates of return; it is rather to show that not all public enterprises, even in a region as difficult as Africa, are loss-makers (and 1984 was a very hard year in Africa). In every developing country one will find one or several PEs, or even whole subsectors which, despite commonly perceived problems, still manage to run at a profit.

In many of these cases, profitable performance will be due to competent, hardworking managers, using their resources in a shrewd manner, capable of resisting production-reducing or cost-increasing demands of the government. But in other public enterprises, the reasons for profitable performance may not rest with the efforts of management. Sometimes profits will be made, but will be due to the exploitation of a monopoly position or government pricing policy which allows an inefficient operation to make money. Some public enterprises make a short-run profit, obtained at the expense of an inappropriate long-term investment or depreciation policy, which goes unnoticed or uncorrected by government reviewers. In a number of countries, government-imposed macroeconomic distortions, such as an overvalued exchange rate or a real negative interest rate, allow firms (public and private, but usually favoring the public) to generate profits because they are paying for inputs or capital at what amount to subsidized prices. Throughout Africa and Latin America, and to a lesser extent in Asia, PEs receive direct and indirect subsidies in the form of direct government transfers for investments or working capital. In some countries such transfers are listed on the books as ordinary revenue, thus further distorting the operational picture. Other practices which cloud the transparency of financial operations of public enterprises include exoneration from import duties, taxes, etc. On the other hand, many loss-making public enterprises would be profitable if they had been properly structured financially at their creation, if their prices had been rationally adjusted in a timely manner, and if they were not obliged to fulfill costly social, noncommercial objectives, for which they are not properly, if at all, compensated.

What this rapid review suggests is that financial profitability is never a complete nor often an accurate measure of public enterprise performance. The issues of monopoly pricing and noncommercial objectives must always be taken into account. Problems are compounded in developing country economies, where macroeconomic distortions are often severe and prevalent. In brief, one can say that there are some PEs which generate profits; there are many which do not but should; and there are many which generate profits which would not if their activities and resources were properly accounted for.

What is particularly alarming for many developing country governments is that so many of their PEs are financial loss-makers, large loss-makers, even with the concessions and special privileges they often enjoy. In Kenya, for example, despite the awarding of generous concessions, the rate of return on the estimated $1.4 billion invested in the PE sector between 1963 and 1984 was a meager 0.4 percent, which implies that a fair number of the enterprises have consistently been loss-makers. This is indeed the case; a 1986 study of 16 major Kenyan agricultural and agroindustrial parastatals showed that aggregate losses for the years 1977–84 totaled K Sh 2.934 billion (or $183 million at 1986 exchange rates). A 1985 study of PEs in 12 West African countries revealed that 62 percent had net losses and 36 percent had negative net worth. Even in countries where only a few firms are losing money, the size of the losses in one or two firms alone may be fiscally significant, for example, in Thailand and Mexico.10 Outside of Africa, one finds high numbers of loss-making PEs in Argentina, Brazil, and elsewhere in Latin America, and in Sri Lanka, the Philippines, and Jamaica, to name but a few other countries where the problem is significant.11

As individual enterprises, and then significant percentages of whole PE sectors, begin to run up losses, more and more governments find themselves faced with a limited number of hard choices. They can allow the PEs to adjust their prices and cut their costs, in an effort to improve their profitability levels and self-financing ratios, and reduce the burden on the budget. Many governments have been reluctant to raise prices or cut costs, at least to the required degree or with the required frequency, fearing popular reaction to increased prices for the goods or services sold by the PE, and fearing equally the political ramifications of cost-cutting by means of shedding labor or closing uneconomic plants or lines.

They can cover the losses through direct budgetary transfers or indirect concessions. Many governments have long followed this path, claiming that the social objectives fulfilled by PEs justify their subsidization, at least for a time. The questions this policy raises are many: how well are the PEs fulfilling their noncommercial objectives; how long can and should a poor country continue to subsidize a loss-making enterprise or sector before expecting the investments to yield a reasonable return? As time passes, if (as is so often the case) financial performance does not improve, and resources become ever more scarce, then the fiscal pressure increases and the ability to sustain this option decreases—regardless of how well the noncommercial objectives are being achieved. For example, in 1988, net Senegalese Government transfers—a high percentage of them to cover operating losses—to its wholly owned and operated PEs equal 96.5 percent of the national fiscal deficit. The situation is similar, if perhaps not quite so acute, in many other countries. The point is simple: unless the sector is covering its costs, depreciation, and finance charges, or unless some other segment of the economy is producing surplus resources which can be used to maintain the public enterprises subsidies (we ignore for the moment the issues of cross-subsidization and opportunity costs), or unless such resources can be generated externally, the flow of resources to PEs will eventually dry up.

As direct budgetary support has declined (owing to increased information on the persistent poor performance of the enterprise or the sector, or sheer exhaustion of funds, or IMF/World Bank adjustment programs, or a combination of factors), governments have often allowed PEs access to the national and international credit systems. Large, state-guaranteed loans have been granted, and extensive overdraft facilities accorded. In Thailand, for example, PEs account for 63 percent of all foreign borrowing. This is not necessarily bad; the question is to what extent are the borrowings applied to productive investments? The answer is that in too many countries borrowed resources have often been imprudently invested, and yielded low or negative rates of return. The result has been that PE borrowing has contributed to both the international debt crisis and the state of paralysis so often seen in developing country domestic banking and finance systems (and to huge interest charges and further working capital shortages in the enterprises themselves). To put it more simply, too many PE loans are nonperforming; they are not being paid back.

Theoretically, a solution other than privatization is available. Governments could simply end their subsidization and let the industrial and commercial PEs sink or swim on their own. Presumably, this would result in some PEs performing well according to normal business practice, but it would undoubtedly, in many countries, result in large numbers of the nonperforming enterprises going under, dying a “natural” commercial death.12 But almost no government, in industrial or developing countries, has taken such a coldblooded economic view, primarily because of the importance of the labor and regional distribution of resource issues in political-economic affairs generally, and in developing countries in particular. Many governments have thus avoided or postponed or approached cautiously fundamental, sector-reducing reforms. (There are exceptions: Mexico, Chile, Togo, Thailand, and Bangladesh, for significant examples.) They have tended to hope that relatively small-scale changes in the firms—changed procedures, altered supervision structures, internal organization and management modifications, and financial restructuring—would find a way to raise revenues, to cut costs and indebtedness, and to bring about reform without a change in the ownership of the assets.

Progress with these sorts of incremental reform has been both modest and slow. Governments find it easier to state good intentions than to alienate powerful interests. They are frequently (not always) willing to change managers, but rarely able to cast the PEs loose into competitive environments, and provide them with simple goals (“make profits”) and the resources, including managerial autonomy, to accomplish these goals. Governments have been particularly reluctant to allow failed enterprises to disappear, to “exit” in the economist’s term. For their part, donor agencies have tended to be overoptimistic with regard to the time necessary to effect reform. They sometimes exhibit excessive hope that changes in behavior will be brought about by changes in technique. Thus, over the years, excessive faith has been placed on the impact to be achieved in the short run by the introduction of PE reform tools, such as analytical accounting, management information systems, and performance agreements.13 This is not to say that these techniques do not have a positive effect. They do. They show the appropriate paths to the maximization of financial and economic objectives, or they illustrate the costs of sociopolitical deviations from those paths. They do not, in and of themselves, instill a willingness to subordinate either self-interest or political objectives to economic rationality.

Thus, governments and donors have undertaken many adjustments at the margin; but all too often enterprise losses continue, and the net and negative impact on the budget mounts. The recent study by Nair and Filippides showed that in 21 of 25 developing countries reviewed, the median PE share of “overall public sector deficits during 1980–85 was 48.3 percent….” (In the remaining 4, all Latin American cases, PEs contributed positively to public sector balances in the period under review).14 It is this large budgetary burden of PEs which most worries government decision-makers. But Nair and Filippides also show that the issue of budgetary burden and net financial flows between governments and PE sectors is very complex. A non-obvious point is that neither a high budgetary burden nor a net outflow from government to PEs is a conclusive indicator of poor performance by the sector. But where the subsidized firms are persistently unprofitable, where their investment programs are persistently weak and nonproductive, and where they are not fulfilling their noncommercial objectives in terms of employment generation, technology transfer, or contributions to regional development and equity, then it is safe to characterize their performance as very poor. And the point is that this persistent lack of goal fulfillment on all defensible fronts is increasingly evident in an increasing number of countries. (Less defensible goals would include continued high-paid employment for political loyalists and the provision of rent-extraction opportunities. These goals, regrettably, are being widely achieved.) But it is the generally financial line of reasoning, and not any dramatic change of ideological viewpoint, which is leading developing country officials to consider fundamental alternatives to past PE practice.

Efficiency Considerations

On the surface, the economic conception of efficiency is simple; it is a measure of “output that is derived from a given input.” Efficiency is increased “if output rises without an increase in inputs;” or if fewer inputs produce the same output.15 In practical terms, an enterprise is efficient or inefficient not in some absolute sense but only in relation to a standard of measurement, that is, to levels of output previously attained, or to levels of output seemingly easily attainable, or to levels obtained by some alternative user of the same set of resources, or in relation to industry or regional norms, etc. In an enterprise (public or private) efficiency gains are achieved by increasing the productivity of the factors of production, such as labor, technology, or capital. If labor works harder, or is provided with the means to produce more with the same level of work, efficiency rises. If output is maintained at past levels while inputs are reduced, efficiency rises.16 This is elementary.

What is not so simple is that many developing countries have experienced great difficulties in effecting efficiency gains in enterprises which are state-owned. The common plight of many developing country PE managers is that even when they know what they should do to maximize the returns on the resources at their disposal, they do not possess the autonomy which would allow them to take both revenue-enhancement and cost-reduction measures. They frequently cannot adjust—usually meaning raise—their prices to reflect changes in input costs and market conditions. They have capital structures and investment fund shortages which force them to use worn-out or obsolete production techniques. PEs often have excessively large workforces, which cannot be laid off in times of reduced production or demand. (This is perhaps the most common complaint of managers, from Thailand to Tanzania, from Argentina to Zambia.) PE managers are constrained for political reasons to continue using uneconomic product lines, unprofitable routes, and loss-making plants or subsidiaries. Necessary government approvals for budgets, investments, procurement, and hiring take time and add to administrative, transaction, and supply costs. Managers are often recruited on criteria other than technical competence. Boards of directors are composed of civil servants who defend ministerial interests rather than promote the welfare of the firm.

These constraints on efficient operation are well known.17 They are widespread, from region to region and sector to sector. Given their prevalence, and given the modest results of incremental reform (that is, rehabilitation of the PE under continued state ownership), a fair number of vocal and influential observers now argue that one must go further. Their view is that to effect the needed financial and efficiency improvements one must change the ownership of the assets; one must privatize the PEs.

The Relationship Between Improved Performance and Private Ownership

Why is it thought that changing ownership will improve performance? As shown above, unsatisfactory performance under public ownership is apparent or at least decreasingly disputed. But leaping from this point to a conclusion that private owners could not do worse is not a serious argument. Unsatisfactory performance under public ownership does not automatically prove that private owners will do better. And just what is “better” in the particular circumstances of the developing countries?

The last question is relevant when considering efficiency gains. That is, there is considerable evidence that private owners tend to outperform public enterprises in financial terms, and this is important in and of itself. But the efficiency issue is more complicated than straight financial returns. With what standard of efficiency performance is the PE being compared? Local public sector competitors? Local private sector competitors? Local private sector competitors managed or owned by foreigners? Some regional or international standard? Many PEs have no local competitors, public or private; this is the case with most public utilities (except, increasingly, in the transport sector). Physical and financial performance indicators, establishing comparative standards, are difficult to devise. Employee costs per unit of service delivered, for example, may legitimately and necessarily vary greatly from country to country depending on the nature of the labor market, technology employed, etc. Thus, differences in physical and financial ratios do not necessarily reflect differences in efficiency (though in some instances efficiency distinctions seem impossible to avoid: private Singapore Airlines reportedly runs twice the number of aircraft with half of the employees of public Air India). Most difficult to quantify and compare is the value the owner, government, places on the noncommercial operations of the enterprise. What is the value to society of maintaining employment levels in region x, of retaining scheduled transport service to city y, of having enterprise z provide housing and health benefits to its employees? Economists argue that these noncommercial objectives can be better achieved, in a more cost-effective and transparent manner, by making them the responsibility of agencies other than PEs, or at least by isolating these activities from the main operations of the firms, that is, “costing them out,” and offering special compensation. Many developing country observers do not believe that their government agencies can perform equally well, much less better, the noncommercial activities now so frequently entrusted to PEs. And the “costing out” notion appears complicated and difficult to implement. The different demands and expectations placed on PEs pose obstacles to measuring adequate performance levels.

Partly because of the difficulty of devising an absolute standard of efficiency, it has become a commonplace in the emerging literature on privatization to note that economic theory provides neither “arguments for global condemnation nor for global preference of public production.”18 Some go further:

The basic difficulty of arguments for divestiture on the grounds of an alleged superior allocative efficiency is that, while mainstream micro-economic theory does point to the allocative superiority of competition, it is actually silent on the ownership issue.19

In other words, neoclassical economic theory associates efficiency outcomes with market structures in general, and with the extent of competition in particular. Ownership is a secondary consideration.

Despite the lack of a causal theoretical linkage between private ownership and efficiency, there exists nonetheless a set of plausible arguments associating private ownership with increased productivity and efficiency. These arguments have a common core; they identify factors accounting for poor PE performance and reason that private owners would avoid or evade the noted constraints. The reasoning is as follows:

  • Under private ownership, there would be less political interference in the decision making of the firm.
  • Managers (and perhaps workers as well) in private firms would (or at least could) receive higher salaries, more clearly linked to productivity and profitability norms.
  • Privatization would impose on firms the discipline of commercial financial markets (as opposed to the “soft budget constraint” so often enjoyed by PEs).
  • Privatization would replace supervision by disinterested government bureaucrats with that of self-interested shareholders; they would impose commercial profitability as the main objective of the firm, and judge managers on their success or failure to achieve this goal.20

The fundamental contention is that private sector managers are given a simpler but at the same time more demanding set of signals and incentives than PE managers. With a change to private ownership, assets become tradable, the discipline of hard budgets and commercial capital markets is imposed, a market develops for managers, and the managers thus respond to new signals and incentives. Efficiency will improve as managers maximize their profits. And if efficiency does not improve, the enterprise will go out of business, which is an equally tolerable outcome, since the resources previously tied up in low-return uses could then be applied to more productive pursuits.

These arguments are simple, vigorous, and plausible. And there is considerable empirical evidence to support the contention that private sector operators outproduce public enterprise managers. For prime example, one normally finds higher rates of return on total assets employed in the private sector than those prevailing in the public enterprise sector. In Thailand, for example, in the period 1983–87, the PE rate of return was between 2.75 and 3 percent, the private rate 9 percent. In Korea, in the early 1980s, the public rate was 7.8 percent, the private 27.5 percent. In India, from the late 1960s to the mid-1980s, the public rate was between 2 and 3 percent, the private between 9 and 12 percent. These figures lend weight to the arguments of the privatizers. It must be noted, however, that the return on assets comparison is a very crude tool. Many of the PEs examined no doubt had huge and extraordinary asset bases in line with government-ordered infrastructure or expansion programs. In such cases, low rates of return on assets would not be an accurate short-run indicator of performance. It appears then that like is not being compared with like, because many of the public enterprises in all three countries cited are social service providers, or working under other politically imposed constraints which increase their asset base or reduce their profits. More rigorous tests of the question in developing countries are lacking; in the few that do exist—a study of total factor productivity in PEs and private firms in Tunisia, for example, and a comparison of jute and textile mills in Bangladesh under public and private control21—one can find few efficiency differences between public and private sector performance. (In both cases cited, the private sector firms outperformed the PEs in financial terms.)

Many officials and observers of PEs from the developing countries are resistant to the arguments in favor of privatization. They seldom dispute that there is political interference in the running of PEs, that PE managers tend to be underpaid, that there is a “soft budget constraint,” etc. But, with the significant exception of substituting shareholder for civil servant supervision, they do not see why reforms require a change of ownership. Nor are they at all convinced that their domestic private sectors will prove capable of better performance than their public sectors. They tend to believe that the same factors which account for inefficient public sector performance will influence private sector performance.22 Nor, as noted, do they readily accept one potential solution: free entry on the part of foreign investors and managers. Governments should, they reason, be able to cajole or force PEs to apply for investment and operating capital as private firms do, with no special privileges. Governments should be able to create appropriate incentive systems, to attract and reward good managers. Governments should be able to isolate PEs from excessive or inappropriate supervision, through performance contracts or similar mechanisms. Governments, in short, should be able to provide commercial and industrial PEs with the necessary set of goals and proper managerial autonomy which would allow them to attain good performance levels under state ownership. Privatization, runs their counterargument, is either unproven or unnecessary.

The problem, of course, is that while governments should indeed be able to effect these reforms, and many have committed themselves—formally, strongly, and repeatedly—to installing such reforms, the record of implementation is deplorable. The record generally shows—admittedly, with exceptions—that regardless of intentions and stated promises, governments tend to continue to interfere, almost always in efficiency-decreasing ways. This being the case, say the proponents of privatization, it is too costly to attempt further incremental reforms. When and where governments cannot resist the temptation to interfere, and when and where that interference is for reasons other than market failure,23 then one could improve the situation by removing the ownership of assets from public hands. This could contribute significantly to the creation of an appropriate environment in which managers can concentrate on profit maximization, and be judged accordingly.

Actions of the World Bank

The poor performance of PEs has prompted many governments to seek the assistance of the World Bank in PE reform, including privatization. This assistance has been provided mainly in the context of the structural adjustment effort and, more recently, through specific Bank attention to the development of the private sector in its borrower countries. The objective of Bank involvement is the promotion of economic efficiency. To this end, the Bank views privatization “not as an end in itself, but as one of many means to help governments increase the efficiency of both government and business.”24 Where it can be shown or reasonably expected, on the basis of enterprise-level diagnosis, that privatization will contribute to efficiency promotion and deficit reduction, and do more than other available actions, then the Bank has supported and will continue to support divestiture.

For a number of years, well in advance of the current interest in privatization, the World Bank indirectly supported private sector development through its traditional lending programs. More recently, the adjustment process has supported borrower country efforts to devise an appropriate policy and regulatory framework, aiming at the enhancement of competitive forces and allowing more freedom for, and less regulation of, the private sector. The elimination of price controls, import restrictions, monopolies, preferential treatment for the public sector, and obstacles to the entry and exit of firms, helps to create a level playing field between the public and private sector. These measures build an environment which encourages private sector interest in the purchase of PEs and offers the necessary incentives for privatized firms to flourish.

Since the early 1980s, the Bank has provided more direct assistance for the privatization process to governments in approximately thirty countries in all regions, the majority of which are in sub-Saharan Africa. The Bank has also been involved in privatization efforts in Brazil, Argentina, the Philippines, Morocco, Panama, and Jamaica.25 The nature and extent of Bank assistance vary from case to case. In countries with a relatively recent history of PE reform (Sao Tome and Principe, Guinea Bissau, and Nepal, for example) efforts center on laying the foundations, that is, establishing a dialogue for a rethinking of state policy on PEs and developing plans for potential future Bank lending. In more advanced cases, support has been provided for specific activities related to privatization, including (i) formulating a framework, methodology, and strategy for privatization (usually in the context of the rationalization of the entire PE sector); (ii) developing data and studies on enterprise performance and viability to select candidates for privatization; and (iii) establishing action plans and, in some cases, timetables for privatization. In Niger, for example, funds provided in a Bank credit paid consultants to review all 54 Nigerien PEs with a view to recommending which could be kept, sold, or closed. On this basis, the Government decided to divest 22 of the 54 PEs. An action plan was then prepared, defining specific steps and a timetable for implementation. The Government subsequently received International Development Association assistance to undertake the reforms. In Togo, a similar classification exercise led to action plans for the divestment of 9 larger PEs under a Bank structural adjustment loan. Such sector-wide exercises have also been carried out in other countries: Benin, Mauritania, Madagascar, Senegal, and Zaire, for example.

A number of countries have avoided this broad study, or “master plan” approach, as being too time-consuming, too likely to generate internal political conflict, or both. They have instead chosen to proceed with privatization on a case-by-case basis, particularly where overall PE sector rationalization has not been a central thrust of the government’s adjustment program, where sensitive political factors require a more low-key approach, and where privatization is being tackled on a subsectoral basis (manufacturing PEs, for instance). For example, the case-by-case approach has been adopted in Mexico, Jamaica, and Panama, partly because of the desire to focus on short-run issues—the attainment of fiscal savings by divesting a certain amount of PE assets, and achieving quick privatization successes to demonstrate the feasibility of continuing the reform—and partly because the discreet (and nontransparent) approach helps to avoid political problems.

The Bank has also assisted governments in undertaking enterprise-level restructuring measures prior to privatization. These steps vary according to the nature and performance of the enterprise, availability of financial resources, and the proposed technique of privatization.26 Some financial restructuring, including settlement of liabilities, equity injections, and promotion of financial discipline through elimination of direct subsidies and cheap credit, may be necessary to enhance privatization prospects. There is always the risk, however, that wrong actions may lead to inflated price expectations rather than increase the value of the PE, or, on the other hand, that the restructured PE may look so good that government loses its commitment to the sale. Generally, a government which has mismanaged a firm will probably not restructure it well; it is most likely wiser to take a lower sale price than to conduct a restructuring that a private owner would undertake—better—himself.

The International Finance Corporation (IFC), the arm of the World Bank lending to the private sector, has provided equity investments in privatized companies; the World Bank has never done this.27 The Bank has often financed studies to determine the extent of debts and arrears between government and PEs and between the PEs themselves. These studies prepare action programs to resolve outstanding liabilities, and establish systems to assure that the arrears will not reoccur. These types of action “clear the brush” for both privatization and PE rehabilitation schemes. Such support has been provided in the Congo, Niger, the Central African Republic, Gabon, and Mali, and are central to the Moroccan PE rehabilitation loan (the second phase of which will focus directly on privatization). Restructuring of government debts with possible recourse to debt-equity swaps are a recent feature in privatization programs. In Zaire, support is presently being provided for the establishment of legal and technical mechanisms for a debt-equity swap program. Holders of Zairian external debt obligations will have the opportunity to exchange these for equity in privatized enterprises. And in Togo, assistance will be provided to strengthen financial institutions serving as posting houses for debt swapping agreements.

Nonfinancial restructuring measures prior to sale are sometimes necessary. A change in status from a government corporation to a limited liability company may be legally required before one can modify management structures, accounting practices, financial dependence, personnel regimes, etc. Management contracts and leases (both forms of partial privatization, or privatization of management) may be a preliminary step toward improving performance and enhancing prospects for sale. Staff reductions may be necessary prior to sale, especially where government employment policies have led to overstaffing. This is a, indeed, the, issue in many countries. Purchasers see labor-shedding as an important factor in cutting costs and boosting returns, while governments are most reluctant to alienate workers and add to unemployment. The Bank has devised ways to address these issues. A restructuring fund in Ghana, for example, finances advisory services to determine restructuring needs and to assist with management contracts. In Morocco and the Philippines, advisory services are provided for legal matters pertaining to privatization.

Bank loans attempt to minimize the social impact of privatization (that is, unemployment). In Togo, alternative employment opportunities for former PE employees are being examined with funds from the Bank credit. Actions being reviewed include a staff reconversion/ retraining program and lines of credit to laid-off workers to assist in the start-up of small enterprises. In Senegal, a “reinsertion fund” has been established to ease the transition of workers laid off as a result of privatizations. This fund refinances loans made by participating commercial banks to provide resources for small-scale projects in different sectors. A similar program is under consideration in Ghana. A training and redeployment program is presently working in Benin, which screens laid-off workers and assigns them to retraining programs, or provides them with credit for agricultural investments. And in Mali, Sudan, and Madagascar, funds have been set up to provide direct compensation—that is, severance pay—for laid-off staff.

Finally, the Bank has also provided assistance to strengthen institutional capacity to manage privatization. Normally, a number of actors and agencies are involved in the process. These include central privatization commissions, asset valuation bodies, ministries of finance, ministries of public enterprises (or some equivalent focal supervisory point), sector ministries, and PEs themselves. In the absence of well-defined roles for each actor, these arrangements can result in confusion, overlap, and conflict. Moreover, severe shortages of skilled staff (auditors, lawyers, financial analysts, and investment bankers) pose obstacles to the implementation of the privatization process. To overcome these deficiencies, Bank-financed technical assistance, in several countries, delineates and rationalizes the functions and responsibilities of each institution. Clear lines of authority are established. Short- and long-term advisors assist in a variety of tasks, including defining strategies, undertaking viability studies, assisting in PE classification, setting the appropriate sale price, giving legal advice, identifying private partners, reviewing bids, negotiating deals, and coordinating overall reform efforts. The provision of an experienced private businessman to the Ministry of State Enterprises in Togo is an example of where this process has worked well.

The Bank has provided support to existing institutions such as ministries of state enterprises (as mentioned in Togo, and in Niger as well), ministries of finance (Mali and Morocco), and technical units in the offices of heads of state (Congo and Senegal). In other cases, the Bank supports the creation of “focal points” for divestiture. A recent paper suggests the general desirability of a central administrative unit to manage the privatization program.28 The creation of such a unit is suggested because line ministries have a vested interest in maintaining a public portfolio under their control. The Bank is assisting in the creation of divestiture units in several countries. In Ghana, a Divestiture Implementation Committee, aided by a technical team, serves as the focal point for privatization and receives substantial Bank support. In the Philippines, the Asset Privatization Trust has been given assistance to evaluate companies, design specific privatization arrangements, and strengthen internal policies, procedures, and organization prior to sale.

Implementation Experience

In sum, many countries are now looking to privatization as an efficiency-enhancing measure; and the World Bank is providing assistance for this process. It is not yet possible to assess definitively what has been accomplished. The problem is the lack of data and the recent genesis of the activity. Where data are available, they focus on numbers of enterprises privatized rather than on the quality and economic and fiscal impact of the transactions. These are crucial issues, since, from the Bank’s perspective, the justification for privatization can only be that the private owner is using more efficiently the resources previously wasted (or used in a less than optimal manner) in the PE. Despite these data scarcities, some preliminary findings have been gathered.

Tentatively, one can say that the overall record of divestiture is mixed, but generally positive. Privatizations with Bank support and assistance have occurred in a number of countries. Jamaica, Panama, Guinea, Togo, and Niger offer cases in point. There is scattered but mounting evidence that many privatizations—in Malaysia, Jamaica, Mexico, Panama, Niger, Thailand, and elsewhere—have resulted in significantly improved financial performance in the privatized firms. In these countries and programs, headway has been made despite the technical and political complexity of privatization. (Complicating factors and obstacles to sales have been reviewed in a number of the studies previously cited; they include inadequate policy frameworks, poorly developed capital markets, outdated or poor-quality accounts, heavy liabilities, labor union opposition, bureaucratic resistance, and reluctance to sell to foreigners.) Commitment from the highest levels of government, well thought-out publicity campaigns, the presence of foreign participation, and World Bank and other donor assistance are some of the factors contributing to completed transactions. (Donor involvement is not a necessary condition; Mexico has, quietly and without any such involvement, sold over 200 PEs in the last five years, and through liquidation and merger eliminated about 500 others. More than 400 remain.) On the other hand, many other countries with fairly long-standing privatization programs such as Ghana, Senegal, Zambia, and Turkey are facing difficulties in carrying out their stated divestiture intentions. Delays stem partly from the fact that many governments commit themselves in the abstract to over-ambitious and unrealistic privatization programs because of disappointment with PE performance and the obvious need to stem the hemorrhage of financial resources. But the concrete implementation of the process may be slow and tortuous,29 as governments attempt to avoid or counteract the less palatable (usually political) potential ramifications of divestiture: increased unemployment, increased foreign ownership, increased concentration of wealth or property in the “wrong” hands, plant closures, pricing problems, etc.

Some initial trends are discernible in actual privatizations. Both large and small PEs have been sold, though in most countries it has been the relatively marginal, or least politically significant PEs which have been put up for sale. Small and medium-sized PEs in manufacturing and services are the most likely candidates for privatization. A range of techniques has been utilized. Public offering of shares is usually the most desirable and commonly proposed method of privatization, because it achieves both widespread and generally domestic ownership; but it is also the most difficult to implement. Sale of assets is relatively easy, and has been the chosen method in countries lacking the necessary conditions for successful public flotations (see below). Private placements, reviewed discretely by a few highly placed officials, have been used in Mexico and elsewhere. Leases and management contracts have been utilized in hotels and other service activities where partial privatization—of management—is thought sufficient. Technically complex methods such as management and/or employee buy-outs, despite their apparent political attractiveness, have only recently been tried in some countries (the Philippines, Mexico, and Chile, for example). In the sales which have occurred, foreign participation has often been involved, sometimes in a minority participation.

In Jamaica, nearly J$500 million of assets of both big and small PEs have been privatized by public share offerings (National Commercial Bank, cement company), sale of assets, leasing (Jamaica Broadcasting Company), and management contracts (hotels). In Guinea, the partial privatization of 15 industrial enterprises of all sizes was conducted with Guinean shareholding remaining at 50 percent. In Togo, 4 PEs have been leased to foreigners, 1 PE has a management contract with a Togolese national, 5 have been fully privatized through sale of assets, and 2 through partial opening of share capital to private shareholders. In Panama, 5 out of a designated list of 7 PEs have been privatized. And in Niger, as of December 1987, 3 PEs had been fully privatized and 8 partially privatized (only 2 were sold, and only in part, to foreigners).

These numbers are encouraging. They are also only partial: these are but a few cases in the recent wave of privatization in which the World Bank happens to have been involved (or, as with Mexico, where the Government passed on information to the Bank). The total number of PEs privatized in the developing countries in the past five years must rapidly be approaching a thousand. Despite this, indeed because of the rapid increase, some issues are surfacing as cause for concern. First, the selection of candidates for sale and the choice of divesting technique have posed difficulties. A prime issue is the specification of criteria for sale, with governments often weighing “strategic importance” more heavily than economic and financial viability in the categorization of PEs into those to be sold, closed, and retained in state hands. The definition of “strategic” tends to be highly sociopolitical in nature. On this basis, relatively marginal PEs may be selected for sale, though they are often the least attractive to buyers. Socially important (though loss-making and nonviable in some cases) or more profitable commercial PEs are retained in public hands. Agreement on which PEs to sell can be a protracted process and has led to long delays. In Niger, for example, it took roughly fourteen months of intermittent negotiations to agree on the broad classification of PEs. In Senegal, a long set of negotiations and discussions resulted in the Government producing a list of ten PEs to be sold. As the implementation deadline neared, two larger PEs were removed from the Government’s original list of candidates. As of April 1988, no acceptable bids had been made for the remaining PEs. It is instructive to note that in early 1988 the Senegalese Government did receive five separate offers for a food-distributing PE, but rejected all five as unsuitable. A World Bank mission concurred that the five offers received all had major drawbacks—demand for monopoly position, other forms of protection, concessionary financing, etc. The mission did however suggest that the Government might have made a more strenuous effort to negotiate modifications to the offers.

As noted, governments often fear that privatization could lead to the concentration of wealth in the hands of “undesirable” domestic or foreign groups. They thus would strongly prefer to find some way of distributing ownership widely. It is a widely held belief, both in and outside developing countries, that capital markets are so thin and embryonic that public stock offerings cannot often be used to privatize. The Jamaican experience with the National Commercial Bank offering, and a similar experience in the stock exchange in the Philippines, calls this view into question. Still, many developing countries have no equities market of any sort. In Senegal, for example, part of the reason for the slow pace of privatization has been the persistent government emphasis on wide share ownership despite the lack of a stock exchange and a weak banking system (the vehicle proposed by the Government to serve as a temporary and embryonic equities market). Add to this inadequate institutional capacity and the existence on the sale block of loss-making PEs with minimal public appeal, and the difficulties of privatization are evident.

Another common issue is the lack of transparency in making specific deals. This is a serious but not insurmountable problem. Positive, though less than fully transparent transactions have been concluded in many countries, for example, Togo and Mexico. But in some instances, competitive bidding procedures have been ignored and prices have been set without clear valuation methodology. Sales, at unstated prices, have sometimes been made to dubious purchasers, such as ruling party politicians and others lacking in business experience. In addition, special privileges such as monopoly rights, favorable financing terms, and protection from imports have been granted to newly privatized enterprises. In one African country, for example, the new private cigarette manufacturer received heavy protection, with confiscatory taxes on competing production and a monopoly on imports. An eleven-year monopoly on the sale of Coca Cola and other soft drinks was obtained by a privatized distributing firm, and production limits were imposed on competitors. High rates of protection have been granted to a leased (and thus partially privatized) steel mill in another country, including government obligations to re-purchase at book value any new investment made by the lessor during the lease period.

A serious and growing impediment to implementation has been the lack of budgetary resources to finance the contingent liabilities of the divested firm, such as provision of severance pay for laid-off workers. It has come as a shock to some countries to realize that a program designed to ease the fiscal burden can, in the short to medium run, cost money. This has been a major problem in Ghana, for example, where past employment policy and rewards to political loyalists led to significant overstaffing in PEs, while pay policy and collective bargaining agreements granted exceedingly generous severance pay. The lack of money to pay severance has brought part of the privatization program to a halt; government does not have the money to meet its contingent liabilities, and does not have the political force (nor the desire) to repudiate its ill-conceived generosity. In other countries, labor opposition—potential or manifest—has affected the pace of privatization, although this depends to a large extent on the prevailing political system, the extent to which labor is organized and dependent on public jobs for a living, and the degree to which labor has been involved in the design and implementation of the privatization program. To take the Senegalese case once more, a parliamentary system of government coupled with strong trade unions has made privatization a particularly contentious and difficult proposition. In Niger, on the other hand, the absence of well-organized labor groups and a general climate of declining public wages have posed fewer obstacles to completing privatizations.

Lessons from Experience

A few lessons emerge from this review. First, while privatization assists overall sector rationalization and efficiency—at a minimum by reducing the number of companies to be supervised by government oversight personnel—reform programs must be tailored to individual country circumstances. Many sub-Saharan African countries, for example, are severely inhibited by environmental weaknesses in their efforts to privatize: capital markets are nascent, financial resources are scarce, the local private sector is weak (and foreign investors are not always acceptable), adequate institutional arrangements are lacking, and PEs on the block are often those least fit for sale. In these cases, attention must be given to developing the necessary policy and regulatory framework and sequencing these reforms to fit in with divestiture programs.

Second, there is in many countries a need for greater transparency in the privatization deal-making process. This will become increasingly important as the small and medium privatizations are completed, and governments move on to tackle the larger, more important, more sensitive sales, in which full transparency will be essential. Some countries have, as noted, sacrificed transparency for speed and political tranquility. But in the absence of open valuation and sales procedures, the potential for irregularities, corruption, and collusion—not to mention just plain bad business deals—is greatly increased. Sales may be made to parties lacking in experience, managerial capacity, and technical expertise, thereby negating the intended benefits of better management. (If, however, the poorly managed firm then failed, it would be a social pity but a private, and no longer a public, loss.) Equally important, the granting of special privileges to the new private owner has led in some documented cases to implicit subsidization of newly privatized enterprises.

Governments sometimes and purchasers always argue that special privileges are necessary to demonstrate the desirability of privatization to potential investors, and to generate quick successes that provide favorable publicity to the program. Privatizing governments must be ready to deal with purchasers insisting on low prices, and more often, protection for the new business, on the grounds that business conditions are difficult, and the assets they are buying are in very poor condition. Indeed, some adjustments may be necessary to attract buyers. For example, governments cannot reasonably expect to be paid historic values for assets which are, all too often, eroded. Lowering of sales prices may be in order. But one must guard against awarding the new private owner protective barriers, a monopoly position, concessionary financing, special access to inputs, etc. The general point is that the case for privileges must be made openly, so that the concerned public and qualified observers can weigh and assess the nature of the deal being struck.

The benefits of concessions are that they bring in dynamic investors and entrepreneurs; the costs are that distortionary concessions create new sources of inefficiency in the economy and set dangerous precedents for future privatizations. The guiding principle must be to assure that privatization does not take place at the expense of liberalization. That is, changes in ownership must not become the end and only goal of the process. This could lead to a small group of private owners using their position—and the political influence that flows from such a position—to bar further competition. The goal, to repeat, is an increase in the general level of economic efficiency.

Bank operations attempt to tackle this issue by calling for government commitment not to extend any subsidies, privileges, or loan guarantees to any privatized companies. In one or two extreme cases, special privileges granted in earlier privatizations are to be reviewed and amended as the privatized firms are exposed to greater market competition (though such a retroactive process may well affect future sales negatively). In one country a commission has been set up to ensure that agreements remain within the framework of a liberal investment code. Safeguards of this nature have costs as well as benefits. Attempting to ensure a competitive environment may scare away potential purchasers who insist on high returns to justify the high risks and costs of developing country investments. On the other hand, in the absence of safeguards, poor and fraudulent deals may be concluded. On balance, some minimal guidance must be established if efficiency gains are to be derived from privatization.

A third lesson is that greater efforts must be made to strengthen country capacity to manage privatization effectively. This sort of reform is complex: appropriate enterprises must be selected for privatization, restructuring measures may have to be undertaken, workable sale techniques must be chosen, private sector partners must be identified, complicated deals have to be struck, post-privatization evaluations must be conducted, and new regulatory agencies may have to be established. Institutional capacity must be built up to undertake each of these steps effectively. This involves developing local expertise in areas such as law, finance, accounting, and investment banking, training local counterparts in privatization techniques, strengthening government apparatus with donor-financed technical assistance, and designing optimal institutional arrangements to design and implement privatization programs.

Fourth, a key problem to date is the lack of any systematic evaluation of the post-sale economic (not financial) performance of privatized enterprises. This makes it very difficult to assess the impact on overall efficiency and the utility of privatization as an instrument in PE reform. In response, the World Bank’s operational units are launching follow-up studies on privatization. In Togo and Niger, for example, the Governments, with Bank assistance, are preparing assessments of privatization experiences and strategies. Other programs will be undertaken to assess the net benefits of privatization to these societies. The Bank’s central department for policy, planning and research is also preparing a comparative study of post-privatization performance of divested firms, to assess both financial and efficiency outcomes.


Analysts of the differences in performance of productively similar public and private firms generally conclude that privately owned firms will outperform PEs, when (and these are most important qualifiers) the two are responding to the same set of pricing signals and when the two are operating in competitive, nondistorted markets. Based on this reasoning, and on the Bank’s and borrower countries experiences with privatization to date, one can argue:

  • the performance of both public and private enterprises will be improved by immersion in a competitive environment;
  • where there is no competition there can be no supposition of the superiority of private ownership; where competition is present there can be such a supposition;
  • governments, especially in developing countries, have not proved to be able regulators of public monopolies;
  • privatization into noncompetitive environments is apt to create more problems than it will solve.

It seems reasonable to conclude, therefore, that privatization, or a change in ownership, will not by itself bring about efficiency improvements.30 But only the most ideological of proponents of privatization would argue that ownership was a sufficient condition for efficiency gains. There is, in sum, a general consensus that the enhancement of competitive forces is equally if not more important. In the absence of competitive market forces, an adequate or appropriate private sector “supply response” may not be forthcoming, and the viability and performance of privatized enterprises may be endangered. The problem, of course, is that creating competitive markets is a far more complex and lengthy issue than a relatively simple change of owners. It involves a host of policy, legal, and institutional changes, all properly phased and sequenced, none easy to achieve.

Once again, the unlikelihood and limitations of the “quick fix” are revealed. Countries and aid agencies must continue to struggle with the whole hard package—determining the appropriate reform set leading to macro policy change and overall liberalization, finding the optimal sequence and phasing for the implementation of reforms, putting the changes into effect in the teeth of resistance from those who benefited from the previous policies, and modifying them as opportunities and obstacles arise. This is the complex reality of development.


    Aylen, Jonathan,“Privatization in Developing Countries,”Lloyds Bank Review (London), No. 163 (January1987), pp. 15–30.

    Bovet, David,“Financial Impacts of Public Enterprises in Sub-Saharan Africa” (unpublished study submitted to the World Bank, August1985).

    • Search Google Scholar
    • Export Citation

    Choksi, Armeane M.,State Intervention in the Industrialization of Developing Countries: Selected Issues, World Bank Staff Working Paper No. 341 (Washington: World Bank, 1979).

    • Search Google Scholar
    • Export Citation

    Commander, Simon, and TonyKillick,“Privatisation in Developing Countries: A Survey of the Issues,”paper presented at Conference on Privatization, Manchester University, 1987.

    • Search Google Scholar
    • Export Citation

    de Chalendar, Jacques,Introduction générale sur les relations entre l’Etat et les entreprises publiques en France, Ministère de l’Economie, des Finances et du Budget (Paris, 1984).

    • Search Google Scholar
    • Export Citation

    Eckert, Gerhard, and WernerPuschra,“Public Enterprises and Development,”Vierteljahresberichte (Bonn), No. 98 (December1984), pp. 331–35.

    • Search Google Scholar
    • Export Citation

    Grosh, Barbara,“Agricultural Parastatals Since Independence: How Have They Performed?”Working Paper No. 435, Institute for Development Studies (University of Nairobi, January1986).

    • Search Google Scholar
    • Export Citation

    Hemming, Richard, and Ali M.Mansoor,Privatization and Public Enterprises, Occasional Paper No. 56 (Washington: International Monetary Fund, 1988).

    • Search Google Scholar
    • Export Citation

    Jones, Leroy P.,Public Enterprise in Less Developed Countries (Cambridge, England: Cambridge University Press, 1982).

    Jones, Leroy P., with PankajTandon and IngoVogelsang,The Economics of Divestiture: Ex Ante Valuation and Ex Post Evaluation, Provisional Papers in Public Economics, No. 87–4, Public Economics Department (Boston: Boston University, 1987).

    • Search Google Scholar
    • Export Citation

    Killick, Tony,Policy Economics (London: Heinemann, 1981).

    Lorch, Klaus,The Privatization Transaction and Its Longer Term Effects: A Case Study of the Textile Industry in Bangladesh, Working Paper of the Center for Business and Government, John F. Kennedy School of Government (Cambridge, Massachusetts: Harvard University, 1988).

    • Search Google Scholar
    • Export Citation

    Nair, Govindan, and AnastasiosFilippides,“State-Owned Enterprises and Public Sector Deficits in Developing Countries: A Comparative Statistical Assessment,”background paper to the 1988 World Development Report (Washington: World Bank, 1988).

    • Search Google Scholar
    • Export Citation

    Nellis, John R.,Public Enterprises in Sub-Saharan Africa, World Bank Discussion Paper No. 1 (Washington: World Bank, 1986).

    Nellis, John R.,Contract Plans and Public Enterprise Performance, Policy, Planning and Research, Working Paper No. 118 (Washington: World Bank, 1988).

    • Search Google Scholar
    • Export Citation

    Shirley, Mary M.,Managing State-Owned Enterprises, World Bank Staff Paper No. 577 (Washington: World Bank, 1983).

    Shirley, Mary M., (1988a), “Bank Lending for State-Owned Enterprise Sector Reform: A Review of Issues and Lessons of Experience,”Country Economics Department (Washington: World Bank, 1988).

    • Search Google Scholar
    • Export Citation

    Shirley, Mary M., (1988b), “The Experience with Privatization,”Finance & Development (Washington), Vol. 25 (September1988), pp. 34–35.

    • Search Google Scholar
    • Export Citation

    Shirley, Mary M., and ElliotBerg,Divestiture in Developing Countries, World Bank Discussion Paper No. 11 (Washington: World Bank, 1987).

    • Search Google Scholar
    • Export Citation

    Short, Peter,“The Role of Public Enterprises: An International Statistical Comparison,” in Public Enterprise in Mixed Economies: Some Macroeconomic Aspects, ed. by Robert H.Floyd, Clive S.Gray, and R. P.Short (Washington: International Monetary Fund, 1984).

    • Search Google Scholar
    • Export Citation

    Song, Dae-Hae,Economic Effects of New Korean Public Enterprise Management Policy and Its Replicability to Other Developing Countries (Seoul: Korean Development Institute, 1988).

    • Search Google Scholar
    • Export Citation

    Susungi, N. N.,The Caveats on Privatization as an Instrument of Structural Adjustment in Africa, African Development Bank Research Paper (Abidjan, 1988).

    • Search Google Scholar
    • Export Citation

    Svejnar, Jan, and MoncerHariga,Public vs. Private Ownership, Export Orientation and Enterprise Productivity in a Developing Economy: Evidence from Tunisia, Department of Economics, Working Paper No. 217 (Pittsburgh, Pennsylvania: University of Pittsburgh, 1987).

    • Search Google Scholar
    • Export Citation

    Van de Walle, Nicolas,“Privatization in Developing Countries: A Review of the Issues” (unpublished, Washington, 1988).

    Vernon, Raymond,Economic Aspects of Privatization Program, Economic Development Institute (Washington: World Bank, 1987).

    Vuylsteke, Charles, HelenNankani, and R.Candoy-Sekse,Techniques of Privatization of State-Owned Enterprises, World Bank Technical Papers Nos. 88, 89, and 90, 3 vols. (Washington: World Bank, 1988).

    • Search Google Scholar
    • Export Citation

The views expressed in this paper are those of the authors; they do not necessarily reflect the viewpoint or the policy of the World Bank.


Short (1983).


For a complete listing and discussion of the large number of reasons—economic and sociopolitical—why states have created public enterprises, see Choksi (1979); Jones (1982); Killick (1981); Nellis (1986); and Shirley (1983).


The usual justification given for aid agency support of PEs was that one could not expect weak existing line administrative agencies to exhibit the dynamism necessary to implement development projects; and that isolation from the surrounding lethargy and special incentive systems were needed if the goals of the project and the agency were to be achieved. Externally supported PEs were especially common in the rural and regional development field, though a number of industrial PEs were started with aid agency assistance. Many of these units achieved minimal or only momentary success; it would have been wiser to take a longer-term perspective. With the wisdom of hindsight, one should have expended efforts to reinforce the capacities of the line ministries and devoted more, and earlier, efforts to developing local private sectors.


Susungi (1988), pp. 14–16.


For Kenya, see Grosh (1986); for West Africa, see Bovet (1985).


One must also mention Korea, where none of the 25 “government invested enterprises” made a loss in 1986, and only 1 of the 25 ran losses in 1984 and 1985. See Song (1988).


One is not talking here of the natural monopoly public utilities; even though they are frequently the major fiscal offenders, they normally cannot be permitted to disappear. Both natural monopolies and market failures do exist; there are, in short, reasons which legitimate the existence and the subsidization of certain public enterprises.


Concerning performance agreements, see Nellis (1988).


These comments refer to what economists call “static” or productive or microeconomic efficiency, that is, the issue of the productivity of a given resource. Longer-term issues of “dynamic” or macroeconomic efficiency—measures of how economic units apply resources over the medium and long run to maintain and maximize their position—are not dealt with here.


For a review of problems, see Shirley (1983).


Commander and Killick (1987), p. 11, emphasis in the original. Hemming and Mansoor (1988) reach a similar conclusion. “Allocative efficiency” is a measure of the extent to which relative output prices operating in the economy as a whole reflect their scarcity values.


These themes are discussed at greater length by Hemming and Mansoor (1988).


On this point the developing country observers are in agreement with some Western neoclassical economists who reason that given the degree of government intervention in markets in many developing countries, and given the prevailing extent of macroeconomic distortions, then the same factors which cause public enterprise inefficiency will act on private owners. See Hemming and Mansoor (1988); and especially Van de Walle (1988).


Perhaps combined with some more politically determined notion of what is and what is not “strategic” for the society in question. The concept of market failure is neat; the concept of “strategic” is not, since it will vary according to criteria which shift from one society to another.


Mary Shirley, Bank Lending for State-Owned Enterprise Sector Reform: A Review of Issues and Lessons of Experience (World Bank Working Paper, forthcoming).


It is important to note that many countries such as Bangladesh, Chile, Mexico, and Thailand, for example, have conducted or are in the process of conducting large privatization programs without World Bank assistance. These experiences are only tangentially discussed in this paper.


For a detailed analysis of the various techniques being used, see Vuylsteke, Nankani, and Candoy-Sekse (1988), Vol. I.


In 1987, the IFC approved, as a small part of its total portfolio, seven equity investments with privatization components that had a total transaction value of about $175 million and an IFC participation of $22 million.


Except in those few cases where new and radically different regimes can seize the opportunity to take bold, abrupt privatization measures; and these usually are programs to return nationalized firms to their former owners, as in Bangladesh, Chile, and Jamaica.


This does not mean that governments can or should blithely continue their past policies of supporting PEs; it means rather that they must strenuously attack their macroeconomic policy deficiencies and consider the least palatable PE reform option: closure and liquidation of failed ventures. The choice in some countries and in some sectors is not between inefficient public and efficient private operation. It is between inefficient public and no operation whatsoever.

    Other Resources Citing This Publication