Applying conditionality to privatization
In opening remarks, IMF Managing Director Horst Köhler noted the timeliness and relevance of the seminar, given ongoing IMF reforms and the emphasis on streamlining conditionality. On privatization, Köhler was convinced that private sector activity is indispensable for growth and for improving welfare over the long run. Nonetheless, privatization should not be seen as a rigid ideology, because it is up to individual governments to make decisions about social organization. Thus, while privatization is often the preferred route, the best way to convince governments is through consultation and persuasion. Collaboration between the IMF and the World Bank is particularly important, he said, because the privatization process is a microeconomic issue and therefore falls more within the World Bank’s area of responsibility and expertise than the IMF’s.
Privatization and program design
In a presentation on the fiscal and macroeconomic implications of privatization, Jeffrey Davis of the IMF’s Fiscal Affairs Department noted that privatization proceeds should be transparently channeled through the budget and not into extrabudgetary funds, and they should be considered as financing in the fiscal accounts. Privatization proceeds, he added, should preferably be used for debt reduction, and any spending should be restricted to high-return projects and should also be consistent with macroeconomic objectives. He noted that privatization was correlated with better macrofiscal outcomes, including higher growth and lower unemployment (at least in the aggregate). Programs should estimate privatization proceeds cautiously, he said, and adjusters should be used to ensure that any excess privatization proceeds are saved.
Even though the evidence showed privatization improved enterprise and macroeconomic performance and enhanced government credibility, according to John Nellis, formerly with the World Bank, it remains a difficult, contentious, and unpopular action. The answer to this conundrum, he suggested, is that privatization involves trade-offs, including between multiple government objectives and between different segments of the population. In his view, efficiency and market structure are more important than revenue generation. Also, both speed and preparation are important, and technical assistance should be used to attain high-quality, yet fast, results. A regulatory framework needs to be in place before, or—at a minimum—in parallel with, privatization, particularly in the case of utilities, he emphasized. It is also important to ensure free entry to the market—perhaps as important as privatization itself.
Providing a summary of findings from the academic literature, William L. Megginson of the University of Oklahoma said the evidence shows that privatization improved performance. This improvement was evident even when the government retained a majority stake, provided the company was allowed to operate commercially. Also, sales prices tended to be higher when a regulatory regime was in place.
Timothy Lane of the Policy Development and Review Department reviewed the state of play in the IMF’s ongoing review of conditionality. He noted that, in deciding when and how to apply conditionality to privatization, one must bear in mind that the macroeconomic consequences of privatization depend considerably on how it is carried out. In particular, the transparency of the process and the corporate governance structure, degree of competition, and labor market restrictions newly privatized enterprises will face are very important. For this reason, notwithstanding the belief that privatization is generally beneficial, it is not always clear whether these benefits will materialize under the actual circumstances in which privatization takes place—which often include poor governance and weak administrative capacity.
A related issue, Lane said, is the question of ownership: if the authorities are not committed to the goals of privatization, they may either drag their feet on implementation or implement in a way that frustrates these goals. These considerations should be taken into account in deciding whether to attach conditionality to privatization in a particular instance. These considerations also pose the question of whether conditionality should specify how privatization takes place or just the outcome: the importance of the process for the macroeconomic benefits would argue for the former, but if the IMF tries to specify the process in too much detail, this may be seen as intrusive and weaken ownership.
Oleh Havrylyshyn of the European II Department noted that moving toward privatization conditionality that is based on process, rather than on time or other targets, helps. This shift started many years ago, for example in Bulgaria and Romania, though the results varied. How much success is a result of better design of conditionality, and how much of stronger ownership is difficult to assess. Strong ownership, he said, ensures success in privatization-related institutional reforms—for example, in the judicial system (as in Lithuania)—while devoting large resources may not achieve results if there is lack of ownership. A good regulatory framework helps ensure the sustainability of reforms, as was true for Moldova, while allowing privatized enterprises to “capture” the regulators leads to failure, as happened in the energy and railroad sectors of several countries.
Havrylyshyn suggested that two important issues in designing privatization conditionality are that it needs to be clear that privatization is critical for the program, and that more flexibility is needed in this area than in the design and monitoring of conditionality in other areas. In general, he felt it was undesirable to set targets for the number of enterprises to be privatized, total privatization revenue, or the timetable, although he said that indefinite delays should not be permitted. In establishing a competitive market environment for privatized enterprises, liberal entry to and exit from the market are crucial, he stressed. When dealing with natural monopolies, regulation must be part of the privatization program.
IMF–World Bank collaboration
There was a frank discussion on the cooperation between the World Bank and the IMF on privatization issues. Speaking from the perspective of the World Bank, Nemat Talaat Shafik said there was a need to better understand the political economy aspects of creating a robust market structure and to address distribution issues up front. Frequently, she noted, differences between the Bank and the IMF come down to speed, with the IMF often aiming for quicker results.
Looking at Bank-IMF collaboration from the IMF’s side, Anupam Basu of the African Department used the experience of trying to privatize the public groundnut company in Senegal as an example of cases where the process needed to be strengthened. The decision to privatize this company had been made explicit in the policy framework paper as far back as 1995, he said, but the opposition of certain groups, a lack of satisfactory bids, and the need to take broader liberalization measures had delayed its implementation. Indeed, by 2000, the objective had become a “future milestone.”
On the basis of this example, Basu saw a need to review the Bank’s usual practice of waiting until countries were ready to implement policies without tracking developments closely in the meantime. He noted, for example, the lack of financial analysis and infrequent auditing of companies put forward for privatization and suggested conducting periodic profit and loss assessments for these companies. There were also, he indicated, differences in the way the IMF and the Bank set structural conditions and in the extent to which nonobservance of these conditions affected aid inflows. In general, IMF conditions were policy-based whereas Bank conditions were results-based. In addition, when IMF conditions were not met, failure to complete a program review could lead to a drop in financial support from various sources. In the case of the World Bank, a loss of financial support as a result of poor policy implementation in a particular sector would be much more limited, as financing would continue to be provided to the country through loans in other sectors. Where there has been a reversal of measures (the delay in privatizing the public groundnut company in Senegal was a prime example), he noted it would be helpful for the World Bank to provide an analysis of the underlying causes.
More generally, Basu emphasized that it was essential to improve Bank-IMF collaboration in this area because World Bank advice had a macroeconomic impact.
There was a need for a shared analysis of objectives and greater clarification from the Bank side as to what kind of sectoral reforms were needed in countries.
Responding to Basu, Michael Klein of the World Bank said there had been a number of instances of successful collaboration between the two institutions on privatization. However, the lack of a mandate in the World Bank to conduct systematic surveillance had hampered its analysis. Klein took issue with the contrast drawn by Basu between the enforcement of World Bank and IMF conditionality, saying it was overdone in the case of privatization. Since IMF program reviews and the associated disbursements were often held up as a result of delays in policy implementation, de facto IMF conditionality had become floating conditionality. Finally, he acknowledged the need for better coordination between the two organizations in this area.
Although participants saw privatization as clearly beneficial, the seminar revealed a range of views on how prospective benefits could best be achieved, with some speakers emphasizing that the consequences of privatization depend on how, as well as whether, it is implemented. Thus, the process by which privatization is accomplished—including the transparency of the process—and the environment in which the privatized enterprise will operate are important. Several participants stressed the need for hard budget constraints on the privatized enterprise and for freedom of exit, as well as entry. Others, however, warned that worrying about such issues could become an excuse for delay and stressed that privatization is usually beneficial even if it does not happen under ideal circumstances.
One key theme was the need to differentiate the way conditionality is applied to privatization, depending on the circumstances of the country. When there is a hemorrhage from the budget caused by financial imbalances at state-owned enterprises, privatization conditionality is absolutely essential and should take the form of performance criteria and even prior actions when country finances are threatened. When the primary objectives are longer term—for example, in the case of transition economies—more flexibility is justified, and the implementation of conditionality could be monitored in the context of program reviews focusing, in particular, on the process and eschewing explicit targets for timetables, numbers of enterprises sold, or revenues. In many middle-income countries, capital markets are more likely to play a key disciplining role, and IMF efforts should thus be tilted toward advice rather than conditionality.
The discussion of Bank-IMF collaboration highlighted the need to strengthen collaboration to address the different time frames under which both institutions often work. This was seen as particularly important in cases in which quick action on privatization was critical for fiscal sustainability. Closer coordination and a clear understanding of the objectives and constraints of each institution were seen as essential given these differences. It was agreed that a process needs to be set up to guide Bank-IMF coordination and help set priorities and clarify further the responsibilities of each institution.