Privatization and Labor Issues in the Context of Economic Reform

5 Privatization and Labor Issues in the Context of Economic Reform

Rolph van der Hoeven, Massoud Karshenas, and Gyorgy Sziraczki

ince the early 1980s, in the so-called post-oil boom era, the S economies of the Middle East and North Africa region (MENA) have faced major adjustment problems that have substantially slowed the growth of the regional economy as a whole.1 The required adjustments and the constraints on economic growth since the 1980s can best be analyzed in the context of the experience of growth during the oil boom years. The MENA countries achieved high rates of GDP growth and rapid structural change during the 1960s and the 1970s—some of the fastest rates of growth in the world economy. This applied to output growth rates in all the main sectors of the economy in almost all the individual countries in the region as well as the average growth rates for the region as a whole (Karshenas, 1996). Rapid rates of growth of oil and other primary commodity exports constituted the engine of growth during these years. Up to the 1980s, the strategy in the MENA region, except in the low-population oil surplus economies, was import-substitution industrialization in predominantly primary exporting economies. Historically, direct

Note: The authors thank Nadar Fergany, Director of the Almishkat Centre for Research and Training, and John Page of the World Bank for useful comments and suggestions on an earlier draft. This paper draws partly on van der Hoeven and Sziraczki (1997) and Karshenas (1996). 1See p. 139 of this volume for a list of the countries that comprise the MENA region.

193

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI public sector involvement in industrial production has also been an important feature of the import-substitution industrialization process in the region. The availability of ample foreign exchange revenues in this way allowed fast rates of growth of industrial investment and output. The problems associated with this phase of rapid industrialization were not due to low investment in the traded goods sectors. Major investments were undertaken in industry and agriculture in all the economies in the region in this period. The overall growth of investment and the growth of traded goods, such as industry and agriculture, in this period were among the highest in the world economy.

Economic Reform and Consequences for Labor in the Middle East and North Africa Region

The major problems that haunted the economies of the region in the post-oil boom era were associated rather with the production inefficiencies that resulted from the inward orientation of the industrialization strategy. With the end of the oil boom era from the early 1980s, the productive assets that were created during the earlier period became a liability in that they were a considerable net foreign exchange drain with high recurrent import requirements, but were not competitive enough to export. The problems of economic reform in general, and privatization in particular, should be examined in the context of the production inefficiencies inherited from the earlier phase of development. The developments in the labor markets in various countries in the MENA region have also been closely intertwined with the processes of growth and structural change during two distinct phases: (1) before 1980—a period characterized by rapid inward-oriented industrialization fostered by fast growth of primary commodity export volumes and prices; and (2) the period since 1980, which has wit- nessed a sharp decline in primary commodity export volumes and prices as well as economic adjustment and restructuring. The fast rates of increase in primary commodity export revenues were only one of the conditions that allowed the rapid rates of industrialization and growth to be achieved in the MENA region during the 1960s and the 1970s. The other important factor was the abundant supply of labor. During the early 1960s in the region as a whole

194

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES more than 60 percent of the labor force was engaged in the agricultural sector. The share of agricultural labor in the total labor force was at or above 50 percent in all the countries except Jordan, and in Turkey it was 75 percent. Although by 1980 substantial changes in the structure of employment had taken place, the share of the agricultural labor force in the region as a whole was still as high as 45 percent. The relatively large shift of labor out of the agricultural sector, together with high natural rates of population growth, ensured an abundant supply of labor during the high-growth period of the 1960s and the 1970s. During the oil boom years, the high rates of investment and rapid rates of growth in the MENA region kept the rate of growth of employment high enough to prevent mounting unemployment and underemployment of labor. The remarkable degree of labor mobility across the different countries in the MENA region both ensured the availability of labor for the small oil surplus economies and alleviated the mismatches between the supply of and demand for different types of labor within individual countries. The overall structure of the labor force in the region both in 1965 and in 1980 was remarkably close to the structure of employment in the middle-income countries. The slight difference between the two country groupings is that the share of the labor force in the agricultural sector for MENA as a whole in both years is higher than in the middle-income countries, and the share of services was propor- tionately smaller. To what extent do the observed patterns of growth and structural change in the MENA region during the 1960s and the 1970s shed light on the problems of structural change and growth during the 1980s and after? The availability of external resources during the 1960s and particularly the 1970s allowed a massive investment boom in the traded-goods sectors, namely, manufacturing and agriculture, on a scale that would not have been possible otherwise. Agricultural and manufacturing growth rates during this period compared favorably with the rest of the developing world and with their own performance in other periods, both for the region as a whole and for individual countries. It was indeed in the post-oil boom era that overall investment growth came to a standstill and the growth of output in traded-goods sectors, particularly in manufacturing, was substantially reduced in all the countries, especially in the major oil- exporting ones. During the investment boom years of the 1970s, the growth of nontraded-goods sectors, such as construction and ser-

195

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI vices, in most countries in the region surpassed growth in other sectors, although the growth of one sector did not seem to compromise growth in other sectors. Also, most countries in the region wit- nessed rapid rates of real wage increase as a result of the fast rates of growth of sectors other than those for traded goods during the 1970s boom. This, however, seems to have been counterbalanced by the rapid rates of labor productivity growth in the manufacturing sector, where the availability of external rents made it possible to maintain profitability and growth by relying more heavily on capital-intensive imported technology. To regard the structural problems in the MENA region economies at the end of the oil boom as merely or even largely arising from high real wages and the overdevelopment of the nontraded-goods sectors would put too much of the burden of adjustment on real wage reduction. Surely with the collapse of the 1970s commodity boom, the prevailing real wages and overvalued real exchange rates in the region were no longer sustainable and had to be reduced. If the structural problems of these economies were due to the over- concentration of labor and other resources in the nontraded-goods sector, wage price adjustments would be sufficient to reinstate structural balance, with a time lag that would be longer or shorter depending on the labor market institutions and the mobility of labor. Two of the distinct empirical features about the labor markets in the MENA region on the supply side, however, signify that the problems of adjustment in the labor markets go far beyond a mere substitution of labor between the traded- and the nontraded-goods sectors. One of these features relates to the quantitative aspects and the other to the qualitative aspects of the labor force. First, this region has the highest rate of growth of the labor force among the developing countries. This is due partly to demographic factors, namely, the high rates of growth of the population, and particularly the working-age population, and partly to the low but growing labor force participation of women. The other, and perhaps larger part of the labor market adjustment problem, which remains unrecorded in the developing countries but is implicit in the differential rates of growth of output and the labor force, is the growing underemployment of labor in the regional economy. With the end of significant emigration from the region during the 1990s, the problems of unemployment and underemployment of labor are likely to be exacerbated. Rather than the mobility of labor between the non- 196

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES traded- and traded-goods sectors, the main problem of the post-oil boom economy of the MENA region thus seems to be to achieve a high enough rate of job creation and to provide adequate training for the new entrants into the labor market to fill these jobs. This leads us to the second general feature of the labor market in the MENA region, that is, the quality of the labor force. An important structural feature of the MENA economies, inherited from the experience of rapid growth and structural change during the oil boom era, is that the quality of the labor force in terms of skills, training, and education is well below that implied by the level of income and structure of production in these economies. In terms of per capita income levels and the structure of output and employment, most of the countries by the early 1980s fell into the category of the World Bank's middle-income-country grouping, and many belonged to the higher-middle-income bracket. However, in terms of the educational attainment of the adult population, the region seems to lag far behind the middle-income countries. The rate of adult illiteracy is more than twice as high as the average for the middle-income countries, and indeed it is even well below the average for the low-income countries. Adult illiteracy among the female population is particularly high in the MENA region by any standards. The fact that such low rates of adult literacy were recorded during the 1990s after massive investment in education during the 1970s and the 1980s is indicative of the low priority historically given to formal education in the region and the considerable amount of resources required to upgrade the education of the population. Formal education of course plays a largely complementary role in learning and skill formation in the economy—although some may consider basic literacy and numeracy of the labor force a minimum requirement for modern economic growth. The larger part of the necessary skills in the economy is generated within the production process through learning by doing or through on-the-job training by firms. The growth of skills and know-how in the economy would therefore depend on the growth of investment in physical capital and output as well as on the existence of an appropriate incentive structure to encourage such learning, on the part of both the employers and workers, including, for example, stable employment relations, job security, and "fair" remuneration for labor. Given that the majority of workers in the MENA region either are engaged in traditional agriculture or are first-generation migrants from the rural

197

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI

areas, and considering their young age structure, the existing stock of industrial skills in these economies is likely to be lower even than that suggested by the data on the rates of illiteracy among the adult population discussed above. Of course, the level of skills and industrial know-how of the population of the MENA economies should be considered relative both to the level of income that such skills are required to sustain and to relative skill and income levels in other developing countries. This is particularly important in the post-oil boom era, when the MENA economies have had to develop alternative sources of foreign exchange earnings by developing non-oil exports in competition with other newly industrializing economies. A comparison with Asian economies in this respect would be instructive (Table 1). Until the late 1970s, the MENA economies managed to sustain relatively high per capita income levels, well above the average for the newly industrializing economies in Asia. Their rapid rate of GDP growth in this period suggests that, despite their high rates of population growth, they managed to reduce somewhat their per capita income gap in relation to the industrial countries and widen their distance from Asia in general and even from the fast-growing economies in East Asia. With the end of the oil boom, however, these rapid rates of growth were no longer sustainable. In the ensuing period, the MENA economies developed substantial external and internal imbalances, and per capita incomes have since followed a long declining trend. As can be seen from Table 1, by 1990, many of the East Asian economies had surpassed the MENA region in per capita income terms, and Asia as a whole had considerably narrowed its income gap. Of course, the different MENA economies have varied in terms of the timing and intensity of economic retrogression since the end of the oil boom. The majority of the countries, namely, Al- geria, Egypt, the Islamic Republic of Iran, Iraq, Jordan, and the Syr- ian Arab Republic, have experienced declining per capita incomes since the 1980s. Morocco, Tunisia, and Turkey have posted moder- ate growth rates, but, even in these countries, the trend growth rates are well below those achieved in earlier periods. This economic retrogression has been combined with the collapse of the investment process in these economies. By now, almost all MENA region countries have taken far-reaching measures to restructure their economies, make them more competitive, and create the conditions for the resumption of investment

198

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES

Table 1. Per Capita GDP by Region 1960 1979 1990

MENA1 100 194 176 Algeria 102 167 165 Egypt 48 91 114 Iran, Islamic Rep. of 177 271 202 2 Iraq 204 510 190 Jordan 69 191 73 Morocco 49 112 128 Syrian Arab Rep. 94 247 232 Tunisia 65 141 173 Turkey 96 176 222 Yemen, Republic of2 45 74 118

East Asia3 50 115 199 Hong Kong 134 472 979 Korea 53 197 396 Malaysia 84 206 304 Singapore 97 395 695 Taiwan Province of China 75 252 479 Thailand 56 128 212 South Asia and China 41 63 97 Bangladesh 56 65 83 China 34 52 79 India 46 50 75 Indonesia 38 71 117

4 Industrial bloc 387 708 885 Source: World Penn Tables, Mark 5. 11960 = 100, at 1985 international prices in U.S. dollars. 2Dates for Iraq refer to 1960, 1979, and 1987, and for the Republic of Yemen to 1969, 1970, and 1990. 3Asia-Pacific region, excluding China. 4Member countries of the Organization for Economic Cooperation and Development, excluding Mexico and Turkey. and growth in an outward-oriented environment. What is important to note in this respect is that international competitiveness and growth of non-oil exports are central to any successful structural adjustment in the economies of the region in the post-oil boom era. As previously noted, the relatively low levels of skill and industrial efficiency in the region may imply that a prerequisite is some degree of real wage reduction. Declining per capita income trends since the early 1980s suggest that real wage compression has already begun to take place. It is also known that, even in countries such as Morocco, Turkey, and Tunisia, where per capita incomes have been rising moderately, substantial real wage compression has taken place (see Table 2). Because real wage reductions beyond a certain limit can

199

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI exacerbate production inefficiencies and discourage private investment by creating social tension and uncertainty, the success of the adjustment program crucially depends on the countries' ability to improve competitiveness without excessive pressure on the standard of living of the workers. Such an adjustment process is therefore much more complex than a mere shift of labor between the nontraded- and the traded-goods sectors, which a real devaluation of the exchange rate or sufficient reduction in real wages could bring about (as the Dutch disease theory would have it). A successful adjustment policy in the context of the MENA region economies needs to address, among other things, two basic issues. The first one relates to the development of skills through education and training and the improvement of productive efficiency within the economic sectors, particularly industry and agriculture. Im- proved competitiveness resulting from efficiency gains within the traded-goods sectors, especially in nontraditional manufacturing, is necessary for growth to resume without exacerbating the external imbalances that have emerged in the economies in the region since the end of the oil boom. The second essential requirement is the resumption of investment growth at a rate that would be adequate to address the problems of growing unemployment and underem-

Table 2. Exchange Rate, Inflation, Real Wages, and Employment in the Manufacturing Sector Average Annual Rate of Change of Exchange rate1 Prices2 Real wages3 Output4 Employment4 Productivity4 MENA Egypt (1985-92) 22.2 17.3 -9.5 0.1 1.9 -2.0 Iran, Islamic Rep. of (1977-92) -0.5 16.3 -2.2 1.4 3.9 -2.5 Morocco (1978-92) 5.1 7.1 -5.7 2.2 5.5 -3.3 Tunisia (1980-92) 6.5 6.5 -1.4 2.5 1.4 1.1 Turkey (1979-88) 42.5 39.3 -3.5 8.3 2.0 6.3 Median 6.5 16.3 -3.5 2.2 2.0 -2.0 Sources: World Bank, World Development Report (various years); and United Nations Indus- trial Development Program (various issues). 1Local currency per U.S. dollar; official rate, period average. 2Consumer price index. 3Real product wages in the manufacturing sector. 4Manufacturing sector.

200

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES

ployment of labor and the inadequacies of human capital formation. The resources made available through efficiency gains, as well as through lower real wage and consumption levels, may help finance the required rates of investment growth. However, depending on the level of industrial development and the existing capital stock in the economy, external finance may be needed to a lesser or greater degree to supplement domestic resources in order to achieve an adequate rate of investment in the different economies. Although labor market flexibility can play an important role in structural adjustment, to rely "excessively" on real wage reductions and on casual employment can hinder this process by reducing work effort and learning, alleviating the pressure on firms to improve efficiency, and reducing investment incentives.

Economic Reform and Privatization in the MENA Region

Most of the restructuring in the MENA countries during the past decade seems to have taken place against the backdrop of falling wages and falling productivity in the manufacturing sector, which essentially arises from a lack of new investment as well as from an in- ability to remove the productive inefficiencies inherited from the old investments during the oil boom period. Most commentators (World Bank, 1995a, b; El-Erian and others, 1995) favor expanding private sector activities in order to increase efficiency and attract foreign capital (or repatriated domestic capital). Privatization forms part of such strategies. However, privatization has been more limited in the Arab countries than in other regions. The existing evidence on privatization activities since 1988 indicates that the MENA region has had the lowest proceeds and number of transactions in the world (Anderson and Martinez, 1995; Sader, 1993). Thus far, only Tunisia and Morocco and, to a lesser extent, Egypt have made progress. But even in these countries the number of enterprises transferred to private owners constitutes less than 10 percent of total public enterprises (Anderson and Martinez, 1995). Also noteworthy is the fact that this region has had barely any foreign participation in privatization (Anderson and Martinez, 1995; Sader, 1993, p. 19), emphasizing the absence of foreign investment in the region in general. Of course, the difficulties in constructing a market-based and private-sector-dominated development strategy in general, and privati-

201

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIACZKI zation of existing assets in particular, vary across the different MENA economies depending on the preexisting role of the private sector during earlier phases of development. The problems in some of the countries, for example, Algeria, Egypt, Iraq, Libya, and the Syrian Arab Republic, where the domination of the state during the import- substitution phase was almost total, are closer to those of the former centrally planned economies. In other countries, for example, Jordan, Tunisia, and Turkey, where the import-substitution phase took place within a mixed-economy framework, the problems are closer to those facing other developing economies in Latin America and Asia. Of course, as discussed above, the MENA region countries have unique characteristics that must be taken into account. To discuss labor issues of privatization, it is therefore useful to look at experiences in different types of economies in other regions where substantial privatization programs have taken place, to draw pertinent lessons for the countries of the MENA region. The next section of the paper discusses some fundamental issues about privatization as well as some of the labor market consequences of recent privatization schemes in a number of developing and transition economies (Bulgaria, Czech Republic, former German Democ- ratic Republic, Hungary, India, Republic of Korea, Mexico, and Pak- istan). Among the issues considered is the effect of privatization on productivity and on the level and structure of employment, including changes in employment and job security, employment practices, and forms of employment contracting. The evolving patterns of industrial relations in privatized firms and the subsequent changes in wages, remuneration systems, and nonwage benefits are also examined.

Objectives and Scope of Privatization

Privatization has different meanings for different people. Three distinct kinds of activities that can be privatized may be distinguished. First are those performed by public enterprises that operate in a competitive market (for example, plants producing consumer goods or capital goods). Second are those activities performed by public enterprises that operate in a monopoly situation and that can be subdivided into "natural" monopolies (electricity, gas, and water) and artificial monopolies created by limiting the ability of other enterprises, either foreign or domestic, to enter the market (car facto

202

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES ries are often a case in point). A third kind of activity involves the contracting out of public services, such as garbage collection and the prison service. In contrast to other types of privatization, no individual products or services are sold to individuals; rather, the state en- ters into lump-sum or piecemeal contracts to pay for services. Al- though this form of privatization is gaining in importance and is often controversial, especially in the United States and certain parts of Europe, case studies in this paper concentrate mainly on the privatization of enterprises that sell products or services to the public. To assess privatization properly, it might be relevant to recollect briefly the reasons why public enterprises are created. One reason for establishing public enterprises has been national security: public defense industries and public transport enterprises have often been created for this reason. Another reason has been to raise revenue: product procurement and marketing boards have often served this end, especially where tax collection would be difficult or impossible. Economic control and self-reliance have been another important motive behind the creation of public enterprises. Another has been provided by the infant industry argument, with protection extending into ownership. The lack of private investment for undertaking large-scale activities has been a powerful incentive for the creation of public enterprises, especially in poorer developing countries. Equity considerations have frequently played a role in the establishment of public enterprises, or in the takeover by the public sector of private enterprises that no longer function profitably. Fi- nally, the fear of a private monopoly situation has often resulted in the establishment of public enterprises, which, it was assumed, would use the monopoly revenues in the public interest. Privatization of public enterprises is often advocated in cases where the original reasons for establishing such enterprises are not seen to be relevant anymore (for example, where well-functioning capital markets have emerged that were absent earlier) or where the management of public enterprises, because of multiple demands, has been inefficient (corrupt in certain cases), thus diminishing the profit to the state or increasing its losses. In most cases, however, privatization is a political act, undertaken as part of a broader structural adjustment or economic transformation program. In many cases, the very act of privatization signals domestic and international markets that a new economic climate is being established. As we will discuss later, some of the more far-reaching privatization pro-

203

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI grams in Latin America have been in effect the outcome of "debt for equity" swaps as part of a debt-reduction strategy having the implicit expectation not only of reducing public sector debts but also of gaining access to international capital markets to obtain fresh funds to replace technologically obsolete capital. They have also provided signals to foreign investors to bring in new money: in recent years one-third to over half of all foreign domestic investment in Latin America and Eastern Europe has been for privatization (Sader, 1993). The fact that privatization is often undertaken in tandem with other economic reforms makes it difficult to evaluate its effects sep- arately. The success or failure of privatization is often closely related to the success or failure of the whole economic reform process. Fre- quently, layoffs and retrenchments in government itself dominate the political discussion around privatization of public sector enterprises. It is therefore difficult, often impossible, to develop a coun- terfactual relating to what would have happened in the absence of privatization, other things being equal. For this reason, we have followed in this paper the more modest approach of discussing the results of privatization in a number of countries. A further complication in assessing the effects of privatization is that economic theory is generally agnostic about ownership. According to Commander and Killick (1988, p. 102), "while main- stream economic theory does point in the direction of the allocative superiority of competition, it is actually silent on the ownership issue." Adam, Cavendish, and Mistry (1992) therefore advance the notion of principal-agent behavior. The principal is the owner of the enterprise and the agent is the manager. These two do not necessarily have the same objectives. The objectives of the public sector as owner may be different from those of a private person as owner, while the manager, too, will have his or her own preference function. The complex question then becomes: to what extent does ownership influence management? Added to the interaction between owner and manager is another consideration, namely, that the free- dom of the manager (or agent) is determined not only by the owner of the enterprise but also by the whole gamut of regulations that govern product and factor markets. If, for example, competition is seen as the best vehicle for increasing efficiency, a change in ownership in the absence of changes in regulations to allow for competition will result not in increased efficiency, but in rents being trans-

204

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES ferred from public sector entrepreneurs (or bureaucrats in the public sector cashing in on these rents) to private sector entrepreneurs, who would be more than willing to pay a high price for a protected enterprise that allows them to cash in on rents. Yet, privatization is often recommended on the implicit premise that the process in itself will create less protection and more competition. One case that is often made for privatization is that it generates revenue for the government. However, an IMF publication argues that, in theory, the fiscal advantages are the same, over the long run, under private or public ownership (Hemming and Mansoor, 1988). The reasons for this are easy to understand. Where there are perfect markets and perfect discounting, the (asset) price of the enterprise is determined by the expected discounted future flow of revenues, and in a competitive market the enterprise will be sold for that price; this is the assumption made in many privatization programs. Hence, all that is involved is a portfolio shift between public and private sectors, in which the public sector exchanges an asset with a future stream of revenue for instant payment. The proceeds of privatization could be invested in bonds to ensure a future stream of income or used to reduce public debt, thus lowering future debt payments. Although many developing and industrial countries alike have advocated privatization in order to obtain current financing, an evaluation of privatization practices in Latin America shows that privatization has done little to reduce fiscal deficits (Pinheiro and Schneider, 1995).

Privatization Activities in Developing and Transition Economies: An Overview

The considerations advanced in the previous section led to the establishment of many state-owned enterprises in developing countries following their independence. By the early 1980s, state-owned enterprises accounted for 17 percent of GDP in Africa and 12 percent in Latin America, but only 3 percent in Asia (excluding China, India, and Myanmar) (Kikeri, Nellis, and Shirley, 1992). On average, 45 percent of domestic capital formation in developing countries was effected by state-owned enterprises (Pfeffermann and Madarassy, 1992). In Eastern Europe, state-owned enterprises were responsible for up to 90 percent of total domestic production.

205

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI

Privatization has increased rapidly since the second half of the 1980s for two distinct reasons. For many developing countries, in the general climate of stabilization and adjustment policies and programs to decrease the debt burden, privatization was seen as an omnibus policy that could reduce the budget deficit by forgoing subsidies to public enterprises, reduce debt by writing down the proceeds re- ceived from the sale of public enterprises, stimulate private sector investment and thus lower public sector investment, signal a new political climate, and increase the efficiency of former state-owned enterprises. While these privatization programs started to take place in developing countries, the systemic changes in Eastern Europe also induced several waves of privatization. The motive in those countries was to change from a command economy to a market economy rather than to achieve specific budgetary objectives. A third factor that has certainly contributed to privatization in developing countries is imitation of privatization efforts in the industrial countries, which came into vogue in the early 1980s. Since privatization in industrial countries often involves large entities, the number of sales in industrial countries is far lower than in developing and transition economies, but the proceeds of sales were supe- rior until very recently. In 1992, however, the total volume of sales in developing countries was for the first time larger than that in industrial countries ($23 billion versus $17 billion). In 1993, the picture was reversed again (Sader, 1993). Privatization in developing countries has often taken place in sectors that had a monopoly or were labor intensive, attributes that made them candidates for public sector ownership at an early stage. Privatization was often popularized by pointing to the poor quality of products and services delivered by many state-owned enterprises. During 1988-92, privatization sales amounted to $62 billion, of which $34 billion derived from sales in infrastructure sectors, including electricity, transportation, and telecommunications; 26 percent from sales of financial institutions; and 20 percent from sales of manufacturing enterprises, especially chemicals. However, these global distribution figures represent a differing regional pattern. Given that Latin American privatization revenues represent two-thirds of all privatization in developing countries, its overall sectoral distribution has a great influence on the total picture. The Latin American picture is in turn heavily dominated by privatization in Argentina, Mexico, and Venezuela. Privatization in East

206

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES

Table 3. Sectoral Distribution of Privatization by Region, 1988-92 (In percent, unless otherwise indicated) Total Amount Primary Financial (in billions of Region Sector Industry Infrastructure Services Other U.S. dollars) Latin America 18 15 34 30 3 40.9 Europe and Central Asia 8 54 4 25 9 10.6 East Asia and Pacific 4 9 80 3 4 7.8 South Asia 25 50 3 14 8 10.6 Sub-Saharan Africa 20 44 1 9 26 0.3 North Africa and Middle East 2 32 — — 66 0.1 Source: Sader (1993).

Asia is dominated by infrastructural projects, while in South Asia sales of manufacturing enterprises amounted to 50 percent of total sales (see Table 3). The picture in Eastern Europe is rather different. Privatization of infrastructure has been marginal. In the early 1990s, reforms started with "small" privatization, through which governments divested them- selves of such small assets in the service sector as restaurants, shops, retail units, and service workshops. Marked progress was also made in agriculture, the urban housing stock, and the restitution of property confiscated by previous regimes. All this was followed, within a short time, by the second wave of privatization, which involved the commercialization of medium- and large-scale industrial enterprises and their transfer to private hands. Some countries, especially Ger- many, the Czech Republic, and, to some extent, Hungary, have almost completed this task, while others are progressing more slowly. The various case studies confirm this general picture. In developing regions, the key issues have become the ways in which efficiency can be increased and public deficits reduced through privatization. In Mexico, growing foreign debt and fiscal crises induced the government to launch a privatization program in 1982. Since then, almost 900 state-owned enterprises have been privatized, including Telmex (Telefonos de Mexico), the third largest company in the country. Revenues from privatization have been used to reduce internal debts and to finance an increase in social expenditures (Botelho and Addis, 1997).

207

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI

In the Republic of Korea, privatization has been one component of the government's economic liberalization policy. The current privatization scheme started in 1987, when the Privatization Proceed- ing Committee decided on partial or total divestment of 11 public companies. By selling shares, the government expected to ease its financial burden, generate revenues for social development, and boost the stock market (Park, 1997). India has taken a cautious approach to privatization, owing to the country's employment situation, the government's fear of large-scale job losses, and fierce opposition to privatization, mainly from trade unions. Privatization has been, perhaps as a deliberate strategy, de- centralized and diluted to take the route of deregulation and partial divestment. The emphasis is on reducing support to loss-making firms, increasing the autonomy of public enterprises, easing entry barriers for private firms, expanding private sector involvement in the management of partially privatized enterprises, and encouraging competition in the economy (Venkata Ratnam, 1997). Pakistan followed a similar strategy in the 1980s, but in 1991 it changed its policy and launched a major privatization project. At the recommendation of the Divestment and Deregulation Committee, the government decided to pull out of the industrial sector and to privatize some companies in the services sector. So far, 71 industrial enterprises and 2 banks have been divested, and the management of 53 firms and 2 banks has been transferred to the private sector (Kemal, 1994). In Central and Eastern Europe, where the national economies were managed by the state until the end of the 1980s, privatization has been a fundamental element in the creation of a competitive market environment. The speed of reforms, however, varies greatly among the different countries. Germany has pursued a policy of rapid reorganization and privatization of all state-owned enterprises of the former German Democratic Republic. It has been implemented by Treuhandanstalt (THA), a public institution set up in July 1990. As owner of the bulk of state-owned companies, THA initially controlled thousands of firms with a total workforce of 4.1 million, about half the total workforce in east Germany in 1989. Between mid-1990 and mid-1993, 12,360 firms were partially or wholly privatized. The remaining 1,900 firms owned by THA were transferred to private owners or local authorities by the end of 1994, thus completing the privatization process in east Germany (Kuhl, 1997).

208

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES

Some of the Central and Eastern European countries have managed, within a short time, to establish the basic conditions for private sector development, including the adoption of laws and regulations related to private ownership and the creation of product and labor markets. In Hungary, new small-scale private firms are mush- rooming, and the privatization of small service and retail units is almost complete; the demonopolization and commercialization of state firms, through the creation of a string of limited-liability companies and joint-stock companies from the old conglomerates in the state sector, are progressing. The government privatization plan en- visages the transfer of 50 percent of state-owned assets to new owners by the mid-1990s (Neumann, 1997). Bulgaria embarked on market-oriented reforms somewhat later than Hungary. However, since the Bulgarian Parliament adopted the Transformation and Privatiza- tion Law in the spring of 1992, denationalization has accelerated (Rock, 1997). The government of the Czech Republic has pursued a policy of rapid privatization. The restitution of property, "small" privatizations, and the encouragement of entrepreneurship had created thousands of small businesses employing more than 600,000 workers by mid-1994, while most large enterprises have been transferred to new owners through the voucher scheme. The first round of voucher privatization took place in 1992 and 1993 and involved 1,500 enterprises. Through the scheme, about 8.5 million citizens gained ownership rights in the former state-owned firms. The second round started in the autumn of 1993 and ended in early 1995. The government has thus implemented a speedy and radical privatization program, leaving the uneasy task of restructuring enterprises and shedding labor to the new owners (Paukert, 1997).

Will Privatization Increase Efficiency and Employment?

Privatization is widely expected to promote productivity and efficiency. However, this largely depends on whether an environment exists in which privatized firms can operate efficiently and, even if it does exist, on how efficiently individual firms operate. The case studies show a rather varied picture. The performance of 15 firms in the Republic of Korea, privatized in the early 1980s, suggests that not all firms benefit equally from the change in ownership. While

209

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI six firms showed increases in efficiency, two actually experienced drops in efficiency after privatization. For the other seven firms, the effects were not significant. Five companies privatized since the late 1980s also display mixed performances. In two firms, productivity has improved, while in the others it has remained unchanged (Park, 1997). Whether the improvements are a direct result of privatization or the result of an economywide reorganization of production is, for reasons explained earlier, difficult to say. Incomplete market structures, continued government regulation and interference in enterprise-level decision making, and neglected human resource management have been suggested as causes of the failure of some enterprises to improve their operation. The experience of India is also instructive. Industrial malaise is pervasive not only among public enterprises but also in the private sector, owing to bureaucratic controls and firms' soft budget constraints, and privatization by itself has not increased efficiency. Deregulation and the creation of a market environment conducive to technological, product, and process innovation, which would provide a firm basis for the growth of productivity and the long- term viability of enterprises, stand out as issues still to be tackled. In Central and Eastern Europe, cuts in subsidies, the collapse of the Council for Mutual Economic Assistance, and trade liberalization have left most firms in a hopelessly uncompetitive situation because of their obsolete technology and the poor quality of their products. The success of privatization depends largely on whether it is accompanied by a restructuring and modernization of enterprises. In east Germany, a survey of industrial firms reveals that productivity has increased much faster in privatized firms than in those waiting to be privatized (Kuhl, 1997). Workforce reductions apart, the most important sources of productivity gains are innovations both in products and production processes and in firm-specific training programs. Modernization of capital equipment, adequate supplies of materials, smaller production units, and a more efficient organization of production have also contributed to increased productivity. Whereas many old firms produce a considerable range of goods and services, the privatized firms tend to concentrate their production on a limited range. This picture from east Germany is confirmed by evidence from Bulgaria and Hungary showing that those firms that cannot gain access to resources and modern technology through privatization are rapidly disappearing.

210

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES

What are the employment effects of privatization? This question, which is often raised, is difficult to answer in the abstract. The first thing to be clarified is whether we are talking about the effects on workers in the enterprises to be privatized or the effects on workers in the rest of the economy. It is also important to clarify whether employment is being viewed from a static, short-term point of view or from a dynamic, long-term one. Usually, privatization will involve immediate job losses because the introduction of more capital-intensive techniques in a period of slow growth— a characteristic of the general setting in which many privatization exercises take place—will lower the demand for labor. In these cases, the direct employment effects depend on the bargaining power of the workers in the privatized enterprise (Edgren, 1990). The employment effects on workers in the rest of the economy are limited. A positive gain in the employment situation of privatized enterprises can be expected over the longer run, stemming from a supposed increase in their profitability that might result in more investment and greater labor demand. Another positive effect can be expected from greater forward and backward linkages of the privatized industry. The magnitude of the employment effect of privatization is determined by the relative share of public enterprise employment in total employment, the number of layoffs expected just before and after privatization, and the potential of the economy to generate employment, both immediately and in the longer run, for those who have been laid off. Many of these factors do not depend on the privatization process itself, but on general economic and social policies, some of which are linked to privatization but many of which are not. The share of public enterprise employment in total employment varies widely among developing and industrial countries alike. Heller and Tait (1983) give an overview of the situation in the early 1980s for the various regions of the world. If public enterprise employment is represented as a share in nonagricultural employment, then the employment share in developing countries is more than three times as large as for the industrial countries, with the lowest share in Latin America and the highest in Africa and Asia (the latter because of the high share of India and China). However, if public enterprise employment is represented as a share in total population (as a rough proxy for total labor force), then the situation for industrial and developing countries becomes fairly similar, with

211

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI

Africa and Latin America having a lower share and Asia a higher share than the industrial market economies. Marinakis (1992) has investigated the changes in public enterprise employment over time. Despite the increased attention given to privatization, amazingly little attention has been paid to employment issues. Marinakis looked at various data sets and compared general government employment growth and total public sector growth (the latter being the sum of general government employment and public enterprise employment) and found that the growth of total public sector employment was much slower or even negative, especially during the second half of the 1980s (Table 4). The possible excep- tion was Latin America, where general government employment was seriously compressed because of the debt crisis and where in the 1980s some general government activities were transferred to public enterprises. What emerges from the figures is that the percentage share in employment of public enterprises in most developing countries is relatively small and that employment growth in public enterprises was often modest or negative even before privatization took place. The effect of privatization on total employment in developing countries can therefore be deemed to be small. This does not of course mean that the issue of employment can be neglected, as it too often is, but it does point to a situation where measures to deal with the immediate effects of privatization should

Table 4. Public Sector Employment Growth (Percentage growth rate) General Government Total Public Sector Region Employment Employment Africa 1975-80 6.4 3.8 1980-85 4.9 3.9 1985-latest 3.7 -0.8 Asia 1975-80 4.1 6.8 1980-85 4.3 2.6 1985-latest 4.1 3.6 Latin America 1975-80 2.1 3.9 1980-85 4.1 2.9 1985-latest 0.4 1.2 Source: Marinakis (1992), Table 5.

212

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES be taken in the realm of labor market policies—for example, nego- tiated severance pay, retraining programs, and support for the establishment of small and medium-sized enterprises—as would be applicable in normal restructuring processes. Several of the case studies support the general picture sketched above. In the Republic of Korea, almost no employees have been laid off because of privatization. This can be attributed to the performance of the economy and the enterprises concerned and to the guarantee of both employment and job protection in the privatized firms. In the few cases where the level of employment was reduced after privatization, reductions were achieved by freezing recruitment and offering generous retirement schemes. The picture just described for most of the developing countries is not, however, accurate for countries in Eastern Europe. There, public enterprises were the backbone of the economy and responsible for a large share of total output; rather than being single production units, they provided social infrastructure for a large part of the population. In addition, enterprises under soft budget constraints tended to hoard much more labor than they actually needed to meet production targets. This was "rational" behavior in a centrally planned economy because the larger a firm was in terms of employment, the more bargaining power its management had in se- curing subsidies, preferences, and special treatment from the plan- ners (Kornai, 1980). Because of the sheer size of public enterprises and the high level of labor surpluses, it is intuitively clear that the employment effects of privatization cannot be dealt with in the realm of normal or additional labor market policies. Its direct and indirect effects have to be part of a larger social, economic, and political calculus that takes into account not only the immediate effects, but also the second- and higher-order effects of privatization. In most Eastern European countries, privatization accompanied by redundancies and the introduction of hard budget constraints on enterprises has resulted in large-scale job losses. This has created a pool of unemployed workers that often cannot be absorbed by other privatized firms or new firms (Jackman, 1994). Furthermore, as has been demonstrated clearly in Russia, the Baltic countries, and the other countries of the former Soviet Union, many workers never appear in the unemployment statistics because, either before or after privatization, they are sent on leave without pay or with much reduced pay (Standing, 1994, 1995; Windell, Anker, and Sziraczki,

213

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI

1995). It is therefore not surprising that in many Eastern European countries the speed of privatization has been slowed because of the fear of its social and political consequences. In telecommunications, an industry that involves rapid technological changes, slight gradual reductions in employment have been observed. However, the reduction in employment took place in both public and privatized enterprises and has often been the consequence of technological development rather than of privatization itself. The restructuring of the various Hungarian telecommunications companies provides a good example of this. In Mexico, the April 1989 labor contract, which paved the way for the privatization of Telmex, both provided employment protection and introduced a flexible labor relations framework to accommodate new technolo- gies and rationalizations. Thus, while employment was secured, the management obtained greater flexibility to redeploy workers within the company. Ultimately, the combination of employment security and job flexibility (agreed in management-union negotiations) ben- efited both parties. Some argue, however, that the win-win situation for workers and management was partly due to a favorable set of regulations and generous concessions for price increases (Galal and others, 1994). It is easy to provide safeguards in profitable and expanding companies, but hardly possible in contracting and unviable firms. In Pakistan, where many public enterprises are grossly overstaffed, the privatization of manufacturing firms has sometimes led to substantial job losses. But most of the job losses in Pakistan's public sector are the consequence of restrictions in the government sector itself. The number of job losses in privatized industries is relatively small and often voluntary. In Central and Eastern Europe, redundancies have occurred on a massive scale in the wake of economic reform, privatization, and the resultant restructuring. For example, a survey carried out by the International Labor Office (ILO) in 1992 has shown that between December 1989 and December 1991 industrial employment in Bulgaria fell by 31.3 percent, with no significant difference between state-owned and commercial companies (Standing, Sziraczki, and Windell, 1993). In Germany, privatized firms employed about 1 million workers in mid-1993, which was about one- fourth of the number employed in those firms four years before. In total, about 3 million jobs have been destroyed owing to downsiz- ing and privatization.

214

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES

The examples discussed here demonstrate that attempts to assess the impact of privatization on employment are confounded by the same difficulties that hamper assessments of the effect of privatization on efficiency. One must separate the impact of privatization measures from other policy changes that occurred at the same time, which is particularly difficult when privatization is part of a more general economic reform, when changes in product markets occur, and when enterprises are restructured. In addition, it would be desirable to consider the effects that emerge over the longer term. Ide- ally, such an analysis requires a comprehensive database that en- ables one to assess the significance of the various factors associated with privatization measures. Unfortunately, one is not available. Only tentative conclusions may therefore be drawn from country re- ports and case studies, leaving many unanswered questions about the impact of privatization on efficiency and employment. It seems that, in the short term, the employment effect either will be negative or will preserve the status quo. (In many cases, privatization itself does not seem to be the main cause of employment reduction.) But what is likely to happen in the longer run is less clear from the available studies. Some argue that privatization will increase the capacity of the economy to create employment by generating more resources for investment and growth (Vuylsteke, 1988). Others disagree: while private industry has demonstrated a higher capacity for generating productive employment through a more efficient use of both capital and labor, privatizing existing public enterprises will not necessarily generate more employment in the long run (Edgren, 1990). Without an improved database for sys- tematic analysis, the question of whether or not the employment- generating potential of the economy will be increased by a transfer of ownership remains open.

Privatization and Labor Retrenchment: Employment Policy Responses

In many cases, privatization results in labor retrenchment. Privat- ization can be very difficult unless special measures are implemented to cushion its negative impact on employment. Three major groups of possible policy measures can be distinguished. The first group consists of measures to delay employment reduction or

215

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI spread it over a longer period after privatization. In many cases, privatization includes arrangements to protect employment in a privatized enterprise for a specific period after the change in ownership. Subsidized short-time employment, widely used in east Germany, is another measure used. Such arrangements have helped smooth the implementation of privatization, allowing a more favorable macroeconomic environment to emerge before the employment protection arrangements expire. The main disadvantages of such measures are that they tend to slow employment adjustment and may involve a financial burden on the state and the employers concerned. On the other hand, increased unemployment benefits are avoided. The debate should thus be carried on at a broader level than that of the enterprise. Freezing of recruitment and early retirements have also been used to minimize the number of redundancies and ease the transition. A second group of policy measures relates to severance payment regulations and a system of early warning for mass layoffs. A number of countries have implemented such measures. Payments to laid-off employees have varied from the minimum legally required severance pay to more generous ad hoc compensation (Kemal, 1994). But some Eastern European case studies note that these measures have sometimes been circumvented through unfair manage- rial practices (Sziraczki, 1988). Employers have, for example, laid off workers in small numbers, thus evading the necessity of observing the stricter regulations relating to mass redundancies. Furthermore, where a large company has been split into small firms, collective agreements and employment contracts have sometimes been mod- ified in such a way that the workers have lost their seniority rights and severance pay entitlement (Neumann, 1997). Another form of compensatory measure involves bonuses for employees who resign voluntarily instead of paying out severance pay to them as retrenched workers. To secure political support for privatization and the cooperation of the unions, the governments of Pakistan and the Republic of Korea have offered generous termination payments to employees who voluntarily leave overstaffed companies after privatization. In Pakistan, for example, of the almost 17,000 workers employed in the newly privatized industrial enterprises in the early 1990s, 43 percent left with a "golden handshake." As a consequence, a massive employment reduction was achieved without any involuntary retrenchment. But the scheme entailed con-

216

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES

siderable costs to the government, and it also backfired on the enterprises. The golden handshake provision encouraged the most productive workers to quit, leaving relatively less productive workers with the firm (Kemal, 1994). The third group of policy measures aims to facilitate the reinte- gration of laid-off workers into other forms of employment. Such measures have included job search and mobility assistance, retraining or vocational training, and job-creation schemes. The development of business advisory services and credit facilities linked to enterprise restructuring or privatization projects has enhanced the mobility of some groups of laid-off employees, while public works programs have provided temporary employment opportunities for those threatened by long-term unemployment. Counseling and support mechanisms for the general promotion of entrepreneurship, the enhancement of productivity growth, and the development of micro, small, and medium-sized enterprises have also played an important role in generating new jobs. Evidence about the relative magnitude and effectiveness of these programs is still scattered, however, and it remains unclear how many workers have found new jobs through these special measures and how many additional jobs have been created. Oliveira (1993), for example, cautions against too high expectations of special vocational training programs for workers made redundant as a consequence of privatization. Drawing on various case studies in Hungary, the Czech Re- public, and Slovakia, he observes a marked decrease in training activities undertaken by enterprises and public training institutions. However, even if training programs were to be increased, Oliveira argues that training plays an important but limited role in helping people find jobs. Newly established enterprises prefer to train their employees, and selection is often based more on attitudes and general skills than on specifically acquired skills in training programs for unemployed workers. Some countries (east Germany, Hungary, Czech Republic, and, to a lesser degree, Bulgaria) have introduced a comprehensive system of compensatory and proactive measures to deal with labor disloca- tion. In countries where social safety net and labor market policies are underdeveloped or largely lacking (India and Pakistan), there is a greater emphasis on privatization-induced social protection measures, focusing on the regions, the companies, and the workers affected by the transfer of ownership. However, these measures have often been

217

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI taken after privatization has occurred. A number of the measures to deal with retrenchment follow logically from ILO Convention 158 (1982) concerning termination of employment at the initiative of the employer. The main reason for such termination is the economic, technological, and structural position of the enterprise. Convention 158 (which is ratified by 22 countries) calls for various measures, including two types of social protection—severance payments and benefits from unemployment insurance and other forms of social security—and obliges employers to consult workers.

Privatization and Labor Relations

The employees of state-owned enterprises often oppose privatization because they fear a loss of job security and social protection as well as a deterioration in the industrial relations system. In many countries, though certainly not all, employees in state-owned enterprises have conditions of service equal to or resembling those of civil servants, in terms of job security and social protection as well as regulations relating to union membership and collective bargaining. Workers often belong to a public service union, which, in many countries, is the largest union. Fear of insecurity and of being marginalized in terms of collective bargaining often turns public enterprise or public sector unions against privatization schemes, especially when the unions have not been involved in initial discussions on privatization (thus augmenting the fear that one of the intentions of the privatization exercise is to marginalize the unions). Often because of close links between certain parts of the trade union movement and political opposition parties, something that started as a union-government dispute may develop into a political dispute, thus blocking privatization exercises or making them extremely difficult to implement. However, privatization does not always mean a threat to collective bargaining and worker involvement. As Schregle (1992) has pointed out, there may already be constraints on public sector enterprises in the sense that the management of these enterprises, in negotiations with the unions, is bound by a limited bargaining mar- gin or by bargaining guidelines dictated by budget limitations and imposed by government. Furthermore, in certain cases, labor legis- lation imposes constraints on state-owned enterprises in the form of

218

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES limitations on full collective bargaining and restrictions on the right to strike. A shift from civil service to private employment is therefore not always negatively perceived. For example, when the Korea Tobacco and Ginseng Corporation changed its status (from government authority to autonomous public enterprise) in the 1980s, its employees also changed status from public servants to civilian employees. Those who did not want to lose their public servant status were given the option of transferring to another government authority. In fact, not many workers took advantage of this possibility, because wages and fringe benefits were higher in autonomous public enterprises than in government authorities. Furthermore, a special provision was made to guarantee the same pension benefits they had en- joyed as public servants. With the privatization of the Korea Stock Exchange, the employees lost their lifelong employment security, which had been guaranteed under public ownership, but their wages and welfare benefits increased and their promotion prospects improved. These examples indicate that the impact of privatization on employment conditions is not necessarily straightforward. There may be some trade-off between the different effects, although it is hard to say whether the overall balance has been positive or negative, particularly in the long term. In Pakistan, the 1991 decision of the government to divest public enterprises aroused great dissatisfaction among workers. Prior to privatization, the government therefore entered into an agreement with the trade unions operating in the public sector. The agreement provided for employment protection for a year after privatization, generous termination payments for voluntary leavers, and retraining and unemployment benefits or credits to encourage laid-off workers to become self-employed (Kemal, 1994). In a similar move, the Indian government set up the National Renewal Fund in 1992 to cope with the social consequences of public sector reform and privatization. It provides assistance to firms to cover the costs of retraining and re- deployment of redundant workers, to provide compensation to employees affected by restructuring and closure, and to support job creation. Currently six regions where industrial malaise is rampant have been selected to implement the program on a pilot basis. However, the National Renewal Fund has been designed by the government; it is not the outcome of a collective bargaining process, and most unions do not regard it as their program (Venkata Ratnam, 1997).

219

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI

While retrenched workers are given severance pay, the remaining workers are often offered stock options as part of the privatization package. The aim is to help forge a new firm-based identity among workers and to give them an incentive to increase productivity. Our case studies suggest, however, that the level of financial participation by workers in most of these schemes is low, and workers are often excluded from decision making. In Mexico, the government guaranteed a low-interest loan to the telephone workers' union to acquire 4.4 percent of Telmex's public shares. Unionized workers bought 3 percent as individuals, and the union's retirement fund purchased 1.4 percent. Nonunionized employees were also autho- rized to purchase a certain number of shares through an additional fund established by the government. The union did not acquire a seat on the company's board, but it was agreed that workers would be allowed to continue purchasing shares and, when they reached 10 percent of all shares, they would be entitled to a seat. However, the union leadership's goal of participating on Telmex's board appears difficult to achieve. Control of 10 percent of the company has remained elusive because many workers cash in their shares and others choose to exert direct control over their shares rather than hand them over to the union's share fund management (Botelho and Addis, 1997). The lessons learned from the privatization of the Hungarian telecommunications sector are similar in many respects: the workers' share generally remains negligible, much less than would be needed to secure participation in decision making or to provide a real incentive to increase productivity (Neumann, 1997). In Bulgaria, the 1992 Privatization Law grants employees in com- mercialized state-owned enterprises the right to buy shares in their own enterprise at a reduced price, but with no voting rights for three years. A survey of 4,600 employees in manufacturing estab- lishments revealed that workers are rather cautious in purchasing shares in their own establishment. Only 10-15 percent are willing to risk more than a fourth of a year's income to buy shares. Employ- ees are uncertain about future ownership of their enterprise and generally feel that they have little influence on the process of privatization and the future of the enterprise (Rock, 1997). In Bulgaria, Hungary, and India, efforts have been made to reform the labor laws to ease labor adjustment and increase labor flexibility after privatization. At the same time, there is a tendency to move away from regular employment toward temporary, casual,

220

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES and contract labor. Enterprises have begun to use more flexible forms of labor relations, in part to enable them to adjust more quickly and without cost to fluctuating demand for their products, in part to reduce their fixed labor costs, and in part to respond to the increasing technological options. For the workers, this trend usually implies less employment security, although the level of wages is sometimes higher than with regular jobs. An apparent feature of the growth of external flexibility in Mex- ico is the increasing use of nonunionized contract labor. In Pakistan, contract workers may constitute as much as 32 percent of the workforce in the large-scale manufacturing sector. In Bulgaria, more and more employers are putting newly recruited workers on so-called civil contracts, which give no security of employment and from which workers can be released either without notice or with less than the normal notice and severance pay. In addition, employers are not obliged to pay social contributions for civil contract labor; consequently, workers are not entitled to unemployment compensation if they lose their jobs (Rock, 1997). The growing use of nonregular forms of employment raises the issue of what safeguards should be put in place to ensure that the benefits to enterprises of having flexible labor relations are not coupled with a general deterioration in wages, benefits, and working conditions. For both employers and workers, the benefits of flexibility have to be balanced by a careful consideration of equitable conditions of employment to ensure that those employed in relatively precarious work relations are not penalized by comparison with those in regular employment. This is important not only for countries where privatization is taking place but also for other countries responding to a process of globalization. As former U.S. Labor Secretary Robert Reich has stated, "Unless people have the security they need to adapt to the future, I believe they will seek security by trying to preserve the past" (Reich, 1993).

Conclusions: Privatization in a Reforming Arab World

Although the cases of privatization discussed in the previous sections have taken place in countries at various levels of economic development, with a wide range of social and economic systems and in which there were different reactions to balance of payments

221

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI problems and budget deficits, a number of lessons for countries that are undertaking or planning to undertake privatization can be in- ferred from the experiences so far. Perhaps the most important lesson is that privatization is not an end in itself. In cases where it was regarded as such it often failed, either because supporting structures had not been put in place or because it led to insurmountable opposition from various interest groups. For privatization to be successful—for the new owners of enterprises, for workers in the enterprises, and for consumers buying the products of the enterprises—privatization should be seen as part and parcel of a medium-term policy and a logical component of a well-defined long-term set of policies, which includes development of a healthy private sector with well-functioning goods, capital, and labor markets in a system of widely respected democratic control. In many cases, it has been found that it is the exposure to competitive forces rather than the nature of ownership that creates the greater pressure for improved efficiency. The absence of one or more of these factors may produce any of various side effects that often result from privatization programs, such as declining consumer surplus because of the continuation of monopolies, the development of crony capitalism through favoring ownership of political clients, and the concentration of restricted capital markets on a limited number of sectors. Political support for privatization is greatest if it is seen as part of a wider process of increasing human concerns and participation in society and as part of a democratic process of deciding whether the state should subsidize loss-making enterprises or invest more in health and education, which would benefit a greater part of the population. Such considerations can be part of a nationwide dialogue on new development directions. The experiences of the countries that have undergone privatization suggest that when privatization has been combined with thorough reforms, both within the enterprise and in its policy environment, it has produced substantial positive results. But it remains to be debated whether these results could have been achieved without privatization, with the same type of internal and external reforms. The evidence gathered so far on the privatization of public enterprises does not allow us to answer the question of whether a change in ownership will in itself increase efficiency. It is difficult to separate the impact of different factors on the performance of an enterprise. In addition, no sys-

222

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES tematic studies have been carried out to monitor the efficiency of the divested enterprises over a longer period and to compare this finding with projections based on the assumption that they had remained in the public sector. What has also become clear is that there must be recognition of the costs of privatization and of the uses to which the resulting revenues can be put. In many countries, privatization has entailed higher costs than anticipated (Kuhl, 1997), and it has rarely played (although it was supposed to do so) a significant role in stabilization programs. Governments seeking to achieve internal and external equilibrium and to privatize at the same time have found this to be difficult and often impossible because the accompanying structures and regulatory measures take time to put in place. Furthermore, the question of who should bear the social cost of privatization, including the accompanying retrenchment programs and other measures affecting social security, is often a highly political one that is difficult to answer in an economy undergoing stabilization. It is much easier to settle such questions at times when the economy is growing and the government is not caught in a straitjacket by its need to stabilize. Fear of privatization is often great among workers and is in many cases justified. Especially in Eastern European countries, but also in some developing countries with far-reaching privatization programs, workers face a great risk of being marginalized, either through worsening employment conditions or through being made redundant and having to accept more vulnerable employment conditions in other enterprises, particularly in the immediate future. The establishment of proper social security programs, income maintenance measures, and social safety nets, as well as training, placement, and other labor market programs, is becoming an essential part of the privatization process, but this places a heavy financial and organizational burden on the government. Two problems regarding social safety nets must be mentioned. First, social safety nets should be part of a larger process of structural change and not organized in an ad hoc way. Second, they are not an alternative to job creation and wider participation in decision making. Several experiences have shown that hasty programs, introduced under the pressure of high levels of unemployment, are not the right way to address the social consequences of privatization and broader structural change. Workers' fears of privatization can be addressed in part through proper government policies. Management, too, must make all pos-

223

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI sible efforts to speak to workers and alleviate unjustified fears. Con- sultation can address the merits and rationale of privatization and the unions' concerns. Where severe employment problems are expected to arise, a productive dialogue on possible measures to lessen the adverse consequences for employees is of crucial importance. It is essential to address employee concern as early as possible when initiating and implementing a privatization program. The weight placed on personnel factors and workers' cooperation will determine the degree of employee confidence in, and support for, privatization. The experience of the most successful programs has been that genuine consultation and involvement in the privatization process, from start to finish, not only smooth the process, but also, by mobilizing the cooperation of the workforce, enhance the prospects of a favorable outcome. These general lessons of privatization schemes seem to be especially relevant to the MENA region. As discussed earlier, the region is characterized by a higher share of workers in the primary and tertiary sectors, often with lower skill qualifications, than in countries with a comparable income level. Privatization should be part of an overall development strategy that, for most countries of the Middle East and North Africa region, would entail reducing dependence on primary and tertiary sector activities and emphasizing the development of the secondary sector and the services related to its development. Furthermore, most countries in the region have no choice but to foster their secondary sector through a gradual opening up of their economy, which will increase foreign competition. Therefore, state-owned enterprises in sectors that would be instrumental in opening up the economy, such as telecommunications and transport, would seem to be prime candidates for privatization. Privatization under such conditions will most likely have a negative effect on workers with limited skills. Furthermore, competition from countries with lower wages and skill patterns similar to those of the MENA region may result in the closure of a number of enterprises and in a real wage decline for workers with limited skills. In such a context, workers are justifiably concerned. One of the challenges leaders in the Arab region face is to indicate to workers, through dialogue and design of coherent policies, that they share their concerns and are working on solutions in cooperation with representatives of workers who take their constituents' concerns se-

224

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES riously. This process might lead to the conclusion that, viewed against past experience, privatization should be avoided under a stabilization program, not only because privatization in a shrinking economy will unnecessarily accentuate some of the adverse social effects, but also because it is expensive when it is implemented properly. Instead, privatization should be part of a longer-term restructuring plan, although this does not mean that action has to be postponed. During the stabilization phase, various measures could be taken to create a favorable climate for privatization, which would include more emphasis on basic education and skill development (through the enterprise structure), on the improvement of social security systems and social safety nets, and on increased involvement of workers' and employers' organizations and of the civil society as a whole in the process of restructuring. Furthermore, systems to establish correct property rights, guarantee an independent judiciary and a free press, and serve as a public control on privatization measures are only some examples of policies that can be introduced even if the economy is in a stabilization program and that will set the conditions for a successful privatization later on. Privatization experiences in the Czech Republic, the Republic of Korea, and Malaysia have shown that workers will respond favorably to privatization if the enterprises are regarded as viable and dynamic enterprises in a growing economy and if workers share in the proceeds of privatization.

References

Adam, Christopher, William Cavendish, and Percy Mistry, 1992, Adjusting Pri- vatization: Case Studies from Developing Countries (London: Currey). Anderson, R., and A. Martinez, 1995, Supporting Private Sector Development in MENA, paper presented at the workshop Strategic Visions for the Middle East and North Africa MENA Region (unpublished; Washington: World Bank). Botelho A., and C. Addis, 1997, "Privatization of Telecommunications in Mexico" in Lessons from Privatization: Labour Issues in Developing and Transition Economies, ed. by R. van der Hoeven and G. Sziraczki (Geneva: Interna- tional Labor Office). Commander, S., and T. Killick, 1988, "Privatisation in Developing Countries: A Survey of Issues," in Privatization in Less Developed Countries, ed. by P. Cook and C. Kirkpatrick (Brighton, Sussex, United Kingdom: Wheatsheaf; New York: St. Martin's Press).

225

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI

Edgren, G., 1990, Privatization, Efficiency, and Employment, International In- stitute for Labour Studies, Discussion Paper No. 23 (Geneva: IILS). El-Erian, Mohamed, and others, 1995, Macroeconomy of the Middle East and North Africa: Exploiting Potential for Growth and Financial Stability (Wash- ington: International Monetary Fund). Galal, Ahmed, and others, 1994, Welfare Consequences of Selling Public Enter- prises (Washington: World Bank). Heller, Peter S., and Alan A. Tait, 1983, Government Employment and Pay: Some International Comparisons, IMF Occasional Paper No. 64 (Washington: In- ternational Monetary Fund). Hemming, Richard, and Ali Mansoor, 1988, Privatization and Public Enterprises, IMF Occasional Paper No. 56 (Washington: International Monetary Fund). Jackman, Richard, 1994, "Economic Policy and Employment in the Transition Economies of Central and Eastern Europe. What Have We Learned?" Inter- national Labour Review, Vol. 133 (No. 3), pp. 327-45. Karshenas, Massoud, 1996, "Macroeconomic Policies, Structural Change and Em- ployment in the Middle East and North Africa," in Employment Expansion and Macroeconomic Stability Under Increasing Globalization, ed. by A. Khan and M. Muqtada (London: MacMillan). Kemal, A.R., 1994, Retrenchment Policies and Labour Shedding in Pakistan, In- terdepartmental Project on Structural Adjustment, Occasional Paper No. 17 (Geneva: International Labor Office). Kikeri, Sunita, John Nellis, and Mary Shirley, 1992, Privatization: The Lessons of Experience (Washington: World Bank). Kornai, Janos, 1980, Economics of Shortage (Amsterdam; New York: North- Holland). Kuhl, Jurgen, 1997, "Privatization and Labour Markets in Eastern Germany," in Lessons from Privatization: Labour Issues in Developing and Transition Economies, ed. by R. van der Hoeven and G. Sziraczki (Geneva: International Labor Office). Marinakis, Andres, 1992, Public Sector Employment in Developing Countries: An Overview of Past and Present Trends, Interdepartmental Project on Structural Adjustment, Occasional Paper No. 3 (Geneva: International Labor Office). Neumann, Laszlo, 1997, "Privatization of Telecommunications in Hungary" in Lessons from Privatization: Labour Issues in Developing and Transition Economies, ed. by R. van der Hoeven and G. Sziraczki (Geneva: Interna- tional Labor Office). Oliviera, Joao, 1993, Training and Privatization in Eastern European Coun- tries: Suggestions for an Operational Agenda, Interdepartmental Project on Structural Adjustment, Occasional Paper No. 13 (Geneva: International Labor Office).

226

©International Monetary Fund. Not for Redistribution CHAPTER 5 • PRIVATIZATION AND LABOR ISSUES

Park, Y.B., 1997, "Privatization and Employment in the Republic of Korea," in Lessons from Privatization: Labour Issues in Developing and Transition Economies, ed. by R. van der Hoeven and G. Sziraczki (Geneva: Interna- tional Labor Office). Paukert, L., 1997, "Privatization and Employment in the Czech Republic," in Lessons from Privatization: Labour Issues in Developing and Transition Economies, ed. by R. van der Hoeven and G. Sziraczki (Geneva: Interna- tional Labor Office). Pfeffermann, Guy, and Andrea Madarassy, 1992, Trends in Private Investment in Developing Countries (Washington: World Bank). Pinheiro, Armando, and Ben Schneider, 1995, "The Fscal Impact of Privatization in Latin America," Quarterly Review of Economies, Vol. 34 (Summer), pp. 9-42. Reich, Robert, 1993, Address to the Employment Week organized by the European Commission, Brussels (unpublished; Washington: Department of Labor). Rock, C, 1997, "Privatization and Employment in Bulgaria's Reform," in Lessons from Privatization: Labour Issues in Developing and Transition Economies, ed. by R. van der Hoeven and G. Sziraczki (Geneva: International Labor Office). Sader, Frank, 1993, Privatizations and Foreign Investment in the Developing World, 1988-92, Policy Research Working Papers, No. 1202 (Washington: World Bank). Schregle, J., 1992, Privatization and Industrial Relations: General Perspectives, paper presented at Top Management Forum on Labour Management Co- operation, Privatization and Industrial Relations, Tokyo, September 21-26 (unpublished; Geneva: International Labor Office, Industrial Relations and Labor Administration Department). Standing, Guy, 1994, "Labour Market Implications of 'Privatization' in Russian In- dustry in 1992," World Development, Vol. 22 (No. 2), pp. 261-70. , 1995, Enterprise Restructuring in Russian Industry and Mass Unemploy- ment, Labour Market Papers, No. 1 (Geneva: International Labor Office). , Gyorgy Sziraczki, and James Windell, 1993, Employment Dynamics in Bulgarian Industry, Labour Market Analysis and Employment Policies Working Papers, No. 63 (Geneva: International Labor Office). Sziraczki, Gyorgy, 1988, "Redundancy and Unemployment in a North Hungar- ian Steel Valley," Labour and Society, Vol. 13 (October), pp. 399-413. United Nations Industrial Development Program, Industry and Development (Vi- enna), various issues. van der Hoeven, R., and Gyorgy Sziraczki, 1997, "Privatization and Labor Issues: An Overview," in Lessons from Privatization: Labour Issues in Developing and Transition Economies, ed. by R. van der Hoeven and G. Sziraczki (Geneva: International Labor Office).

227

©International Monetary Fund. Not for Redistribution ROLPH VAN DER HOEVEN, MASSOUD KARSHENAS, AND GYORGY SZIRACZKI

Venkata Ratnam, C.S., 1997, "Adjustment and Privatization in India," in Lessons from Privatization: Labour Issues in Developing and Transition Economies, ed. by R. van der Hoeven and G. Sziraczki (Geneva: International Labor Office). Vuylsteke, Charles, 1988, Techniques of Privatization of State-Owned Enter- prises, Vol. 1, World Bank Technical Papers, No. 88 (Washington: World Bank). Windell, James, Richard Anker, and Gyorgy Sziraczki, 1995, Kyrgyzstan: En- terprise Restructuring and Labour Shedding in a Free-Fall Economy, 1991-1994, Labour Market Papers, No. 5 (Geneva: International Labor Office). World Bank, 1995a, Claiming the Future: Choosing Prosperity in the Middle East and North Africa (Washington). , 1995b, Will Arab Workers Prosper or Be Left Out in the Twenty-First Cen- tury? Regional Perspectives on the World Development Report, No. 1020-3648 (Washington). , World Development Report (Washington), various issues.

228

©International Monetary Fund. Not for Redistribution Comment

Nader Fergany

he title of the paper addresses two major issues of economic re- Tstructuring policies and programs—privatization and employment—both of which are quite controversial. Everyone recognizes the adverse impact of economic restructuring on employment. Pro- ponents of structural adjustment claim that these negative "side effects" are temporary, lasting only until economic reform takes hold. Actual experience, however, attests to the resilience of the rise in unemployment that accompanies structural adjustment programs. Privatization, especially divestiture, is advocated as the linchpin of structural adjustment. Fast and large-scale divestiture is promoted as an essential requirement for market mechanisms to function efficiently. Whether in the prevailing institutional context of Arab countries privatization can produce the intended results is doubtful. In addition, an overemphasis on divestiture overshadows the wider issue of private sector development as well as the critical questions of reforming the civil service and ensuring that governance regimes make it possible for market mechanisms to take hold. To be sure, privatization and labor issues deserve thorough cov- erage in a seminar devoted to the social impact of economic reform on Arab countries. However, the paper at hand does not do justice to the treatment of the two issues in Arab countries. It essentially consists of two disjointed pieces. The first is taken from a paper by Massoud Karshenas on macroeconomic policies, structural change, and employment in the Middle East and North Africa, published as a chapter in a book. The other, constituting the bulk of the paper,

229

©International Monetary Fund. Not for Redistribution NADER FERGANY is based on a book soon to be published by the International Labor Office, Lessons from Privatization. On the pretext that privatization has been limited in Arab countries, the treatment of privatization and labor market consequences is entirely based on the experience of a number of developing countries and transition economies in East Europe, Latin America, and Asia. The paper concludes with a section on privatization in a reforming Arab world that stops at general advice derived from the experiences of the non-Arab countries studied. While the absence of the Arab context is deplorable (a previous seminar sponsored by the Arab Monetary Fund and the Arab Fund for Economic and Social Development was devoted to privatization), the two main sections of the paper are instructive, and the present writer is generally in agreement with the labor market characterization of the first section and the general lessons of the privatization experience of the second. More to the point of this seminar, however, is a treatment of labor market issues in the context of economic restructuring in Arab countries and how these issues are related to privatization and private sector development at large. On this, the paper is quite silent. In what follows, an attempt will be made to make up for this deficiency.1 The starting point for an assessment of the social impact of economic restructuring is charting trends in the pertinent phenomena. Then the question of the link between economic restructuring and the observed trends can be addressed. Establishing a causal link, however, is fraught with difficulty, technical as well as ideological. The more important task is to consider the likely future development of social phenomena if the present regime of economic restructuring continues. Of ultimate relevance to the purpose of economics then is the following question: on the basis of trends and likely future developments, what policies should be advocated in order to maximize the social welfare of people in the region? Here, it is important to consider the role divestiture, or private sector development at large, can play. The first section of the paper under discussion aptly character- izes the main labor market problems in Arab countries under economic restructuring: a low—and, relative to other developing

1The following discussion is based, in part, on Fergany (1995b).

230

©International Monetary Fund. Not for Redistribution CHAPTER 5 • COMMENT countries, deteriorating—stock of human capital and growing unemployment and underemployment. The end result is sagging productivity and slow growth in total output. Indeed, in order to sustain adequate and rising levels of social welfare, Arab countries face two main employment challenges: to form high-quality human capital stock and generate enough job opportunities to minimize open unemployment in such a way as to enhance productivity growth. Major institutional reform is needed to allow such an environment to develop. Two approaches stand out—civil service reform and efficient labor market mechanisms, such as employment exchanges.

Human Capital Stock

Let me start by documenting the low level of human capital stock in the region and dispelling the myth that Arab countries are big spenders on education. The number of mean years of schooling (United Nations Development Program, 1994) in the Arab countries averaged 2.8 in 1990, less than in developing countries elsewhere (4.2). Even the countries of the Gulf Cooperation Council (GCC) (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates) at 4.0, did not reach the average of the rest of the third world. Using the ratio of expenditures on education to total product leads to the conclusion that Arab countries are big spenders on education, 6.4 percent of GDP compared with only 3.2 percent for other developing countries and even 6.0 percent for the member countries of the Organization for Economic Cooperation and De- velopment (OECD). A "paradox" has been created: Arab countries spend a lot on education and get little in return. Not so. The indicator used is rather misleading. It does not account for differences in the size of total output, population size, or age structure. Even expenditure on education per student is not a good indicator for international comparisons comprising industrial and developing countries, for it does not account for those not in school. A better indicator would be the average expenditure on education per person in the education age range, which is also a good proxy for quality. Using this indicator, it turns out that Arab countries are

231

©International Monetary Fund. Not for Redistribution NADER FERGANY very poor spenders on education. In 1990, Arab countries that are not in the GCC spent slightly less than developing countries outside the Arab world and less than one-thirtieth of the amount OECD countries spent. Egypt, for example, spends less than half the value of non-Arab developing countries. An international cross-sectional regression analysis of the rela- tionship between GNP per capita, as an indicator of economic productivity or welfare, and educational wealth, measured by average number of grades completed per person and current expenditures on education per person in the education age bracket 6-21 years reveals an extremely strong level of determination. The highest levels of economic development are associated with more than ten mean years of schooling and expenditure on education of more than $5,000 (1990 U.S. dollars). These values can be taken to de- fine a threshold of educational and economic maturity. In contrast, low levels of expenditure on education coupled with educational poverty (less than six years and less than $100 (1990 U.S. dollars)) are correlated with miserable economic performance. When expenditure on education is less than $500 (1990 U.S. dollars), an increase in mean years of schooling hardly results in higher per capita gross output. Higher levels of expenditure on education, however, show increasingly handsome economic returns, but only beyond the educational wealth threshold of six years. It is this lower threshold of educational and economic takeoff—six years of schooling and education expenditures of $500 for each individual 6-21 years old—that Arab countries should strive to attain, and surpass, as soon as possible. Compared with the present levels of the two variables considered, the required investment in education is huge and is probably beyond the means of most Arab countries. Alternative educational systems must be devised that produce much higher quality without the level of expenditure associated with best international practice. This is, to be sure, a major challenge. To stress the international context of the link between the level of human capital accumulation and economic performance, Arab countries can be compared with East Asia (excluding China), which has managed to achieve great progress over the past three decades. This is exactly the same time period during which a number of Arab countries were considered to have had a real opportunity to advance—one that appears to have been missed, at least for now.

232

Table 1. Selected Indicators of Development: A Comparison Indicator Year Arab Countries East Asia Real per capita GDP 1960 21 17 1990 29 62 Average education per person 1980 21 70 (in years) 1992 33 79

How do Arab countries compare with East Asia on these two gross indicators of development?2 Table 1 shows that, in terms of per capita GDP, Arab countries were better off than East Asia in 1960. Three decades later, however, average Arab economic welfare is less than half the level of East Asian countries. The relative position of Arab countries is even worse if the GCC members are excluded (because oil revenues significantly raise their per capita GDP). The second indicator is more important. It indicates a crucial dimension of human wealth—average educational attainment in society. The dramatic difference between Arabs and East Asians in this respect is immediately evident. It attests to one of the fundamental lessons of history: the decisive importance of investment in human capital as a prerequisite of genuine development. Clearly, Arab countries could not significantly narrow the gap separating them from East Asian countries in this respect during the 1980s. A closer examination of educational attainment, and trends therein, is essential for our purpose here, but is, unfortunately, ham- pered by a lack of detailed information. Only in Egypt does an information base sufficient for a preliminary assessment exist. By virtue of size, average economic structure, and a long educational history, the Egyptian case is significant. Official figures place illiteracy at about 50 percent of the population 10 years of age and older. But disadvantaged social groups, women and the poor, suffer relatively higher illiteracy rates, particularly in rural areas. Furthermore, basic education is less than uni- versal. All girls and the population in poor areas, especially in the countryside, suffer lower rates of access to education. The differen-

2The comparison is given in terms of a percentage of the level of industrial countries in the category "North" in United Nations Development Program (1994).

233

©International Monetary Fund. Not for Redistribution NADER FERGANY tials in access to basic education, across gender, social strata, and regions of the country follow the same patterns of differentials in illiteracy. Thus, differentials in access to basic education exacerbate differentials in literacy by gender and poverty. Breaking long-term trends, in the past 12 years access to primary education has been stagnant at the 90 percent level, and the gender differential in access seems to have widened. There is evidence that girls from poor urban households have suffered failing access to primary education in the same period, with the decline in access associated with increasing poverty. Thus, changes in the macroeconomic environment under economic restructuring (for example, the onset of inflation and the introduction of fees under cost recovery) that started in the mid-1970s seem to have been associated with adverse developments in educational attainment. The poor and girls seem to have paid the heaviest price. This author estimated that in Egypt about 600,000 girls 6-10 years old in 1993 were lacking a primary education. Assuming a desired average classroom size of 30, 20,000 classrooms are needed to only mop up the 1993 backlog of girls deprived of primary education. Given that (1) the estimates above exclude boys not in school, (2) population growth continues at about 2 percent a year, (3) there is a need to reinstate the previously canceled sixth grade in the primary education cycle, and (4) a majority of pupils in primary education attend school in double and triple shifts in crowded classrooms, the required provision of new schooling capacity in any future time frame increases considerably beyond the 20,000 classroom estimate. In contrast, a national campaign enjoying high official patronage and designed to overcome this problem before the end of this century by building girl-friendly, one-classroom schools was able to enroll fewer than 10,000 girls during the first year (1994). At this rate, it will take about 75 years to enroll the 1993 backlog of girls. Similar evidence exists in Morocco, where access to primary education, particularly in rural areas, has suffered in the years following the beginning of structural adjustment. Again, girls suffered a higher level of deprivation. The quality of education is a more serious problem. Less than uni- versal access to basic education is bad. Widespread access when the quality is poor, however, is outright dangerous, for it has dire consequences for productivity. Unfortunately, information on quality, in terms of real output, is even scantier than is the information on ac-

234

©International Monetary Fund. Not for Redistribution CHAPTER 5 • COMMENT cess. This is an area in which rigorous and comparable research is badly needed. In Egypt, there is strong evidence that competency in reading and writing (40 percent) and in mathematics (30 percent) is declining (especially since the educational system was "reformed" in 1988 by eliminating a grade from the primary education cycle in schools supervised by the Ministry of Education). Even worse, completion of secondary education (11 grades) raises competency, on the primary stage level, to only 80 percent in reading and writing and a scan- dalous 50 percent in mathematics. If competency in reading and writing only is to be taken as the definition of literacy, a very low standard for the end of the twentieth century, then completing secondary education should be the minimum educational attainment for a person to be counted as literate (the current census definition requires completing four grades of primary education only). If such a definition is adopted, the proportion of illiterates would probably rise to more than 80 percent. To my best knowledge, no comparable data are available for other Arab countries. Jordan, hailed for its high-quality educational system, was the only Arab country that joined an international educational assessment study and registered at the bottom of the scale among 16 countries in all categories. Could privatization of education help improve the formation of human capital stock? It is unlikely. For evidence, we turn again to the Egyptian case. Schools supervised by the Ministry of Education, government and private, account for more than 90 percent of primary education pupils (the rest go to schools run by Al-Azhar). Pri- vate schools enroll less than 7 percent of primary education pupils. Compared with government schools, private schools have lower internal efficiency and do not significantly improve acquisition of pupils' basic literacy skills.3 I believe that access to education and, more critically, educational quality would benefit in the long run from the presence of a strong not-for-profit educational system that can compete with the government, with strict control for quality through rigorous accreditation criteria. A revitalized movement by the nongovernmental organizations (NGOs) represents a window of opportunity in this respect. In the prevailing institutional climate, if

3The elite language schools accommodate a small fraction of private school pupils. Many private schools cater to the "rejects" of government schools.

235

©International Monetary Fund. Not for Redistribution NADER FERGANY private education grows to be a big for-profit business with influential investors, quality control by a weak and inefficient government administration that is also keen on attracting investment in education would be almost impossible.

Job Creation

Egypt has a relatively good database on unemployment, although it is riddled with problems. Results of recent rounds of the Labor Force Sample Survey (LFSS) show that unemployment has been on the rise. A detailed analysis of the final results of the LFSS produced a lower estimate of the rate of open unemployment, about 14 percent corresponding to approximately 3 million unemployed persons (Fer- gany, 1995a). Regardless of the sometimes contestable level of open unemployment, there is agreement that open unemployment has become "structural," or "organic," with grave social and economic consequences. Deficient demand for labor, stemming from stagnant economic growth, as well as the labor-light and capital-intensive structure of growth and technological choice in the last two decades are the primary causes of large-scale open unemployment. Unemployment is expected to rise even further in Egypt, because the first phase of the structural adjustment package was, by design, recessionary; divestiture of public enterprises is expected to result in layoffs; the economy has continued to stagnate even after completion of the first phase of economic "reform"; and prospects for labor migration are diminishing. More significant than the level of unemployment is its structure. Available data show that the vast majority of the unemployed are educated young persons entering the labor market for the first time (a result of discontinuing guaranteed employment of such individuals in the government and public enterprise sector). Unemployment rates are highest for graduates of in- termediate education and next highest among graduates of higher education (indicating a negative societal rate of return to education, taking employment as a criterion). The unemployment rate for fe- males is at least double that of males, and the poor suffer higher rates of unemployment. Unemployment data in other Arab countries are generally more deficient than in Egypt. Estimates of the level of unemployment in other Arab countries around 1990 include 6 percent in the Syrian

236

Arab Republic, 11 percent in Morocco and Tunisia, 16 percent in Sudan, 17 percent in Jordan, 21 percent in Algeria, 33 percent in Iraq, and 48 percent in the West Bank and Gaza Strip (Fergany, 1994). (Note the particularly high values for Algeria, Iraq, and the West Bank and Gaza Strip, the three hottest spots in the region.) For the Arab region as a whole, an overall open unemployment rate of 15 percent in 1990 is reasonable. This corresponds to almost 10 million unemployed persons, mostly among the poor and among educated youth. And the trend is on the rise. If we add to the present pool of unemployed future entrants to the labor market, comprising 2-3 percent of the labor force annually, the huge number of jobs that must be created represents an awesome challenge to these economies, especially against the backdrop of stagnant growth. Could privatization as practiced in most Arab countries at present, especially in the form of divestiture, help generate jobs on the vast scale required? It is an improbable scenario. On the contrary, divestiture is widely feared to result in layoffs designed to wipe out the overstaffing of public sector enterprises. Serious civil service reform is expected to have a similar outcome. Private sector development in the past two decades has been both capital intensive and labor sparse. This is indeed one of the main reasons for rising unemployment—the government stopped creating jobs in the public sector, and the private sector has not taken on the role of the main employer. Once again, a disabling institutional environment is easily identified as the main culprit. Bureaucratic red tape and corruption as well as macroeconomic and political instability drive capital out rather than bring it in. Although big business is decidedly favored in the prevailing institutional environment, a strong bias toward big capital under present economic restructuring regimes has left the small investor at the mercy of the worst of government inept- ness and the hostility of the legal and administrative framework toward private investment. Large-scale employment generation as part of private sector development is possible if it leads to labor-intensive growth as well as productivity enhancement. A tall order indeed.

Productivity Enhancement

Could divestiture lead to higher productivity in privatized enterprises? Here, the paper says that it depends. In the Republic of Korea,

237

"incomplete market structures, continued government regulation and interference in enterprise-level decision making, and neglected human resource management have been suggested as causes of the failure of some enterprises to improve their operation" (p. 210). In India, where "industrial malaise is pervasive not only among public enterprises but also in the private sector . . . privatization by itself has not increased efficiency" (p. 210). I concur. In many developing countries, including Arab countries, inefficiency is not limited to public ownership. It is a generalized trait of such economies. With low levels of human capital and a disabling institutional environment, private enterprise, especially if large and influential, can be as inefficient as public enterprise. Worse, in the absence of competitive market mechanisms and strict market regulation, large private enterprises can be harmful to the public interest (through private monopolies, restriction of access to information, and market entry to political clients). According to Messrs. van der Hoeven, Karshenas, and Sziraczki, "in many cases, it has been found that it is the exposure to competitive forces rather than the nature of ownership that creates the greater pressure for improved efficiency. . . . When privatization has been combined with thorough reforms, both within the enterprise and in its policy environment, it has produced substantial positive results. But it remains to be debated whether these results could have been achieved without privatization, with the same type of internal and external reforms" (p. 222).

Civil Service Reform

Essential economic functions remain the unique domain of government. Nevertheless, governments in developing countries are notoriously inefficient. The criteria for efficiency, however, differ from one perspective to another. The conventional wisdom in civil service reform concentrates on problems of pay and employment practices. This characterization, although accurate, can be considered "technical" on the basis of efficiency considerations. But the level of government efficiency has implications for economic performance at large and for that in the private sector in particular. It is also important to consider the social dimension of the role of government, that is, the effectiveness of government in attaining the social objectives of economic activity.

238

Assessment of the economywide, particularly social, implications of government performance and attempts at civil service reform are generally rather weak. Issues like the impact on unemployment, productivity, and poverty are rarely addressed, partly because it is more difficult to carry out such assessments and partly because there is little concern in structural adjustment programs for other than macroeconomic and international trade considerations. It is sometimes claimed that civil service reform has been "politi- cally easy." The claim is then followed by the recommendation that "bolder" reforms be attempted in the future. But it is also recognized that the small political price paid so far is probably due to the limited extent of civil service reform in the past, the repressive nature of most third-world regimes, and the related weakness of civil society in countries where civil service reform has been attempted, which renders the recommendation almost irresponsible. Civil service reform, however, remains an essential component of a labor market reform policy package. Reward for government service must be reformed through incorporation of a transparent structure, adequate wages, a decompression of scales, and equalization of discrepancies across the various branches of government service. Allocations for equipment, operation, and maintenance, necessary for efficient functioning, should be made available. Sound public administration practices, leading to higher productivity, need to be instituted, including the following: recruitment, ad- vancement, and termination of service based on merit, including competency exams for employees; rigorous use of information, planning, follow-up, and evaluation (including customer satisfaction as a major criterion for service departments and periodic functional reviews) in government administration; continuous, but serious, training for employees; and decentralization of government. To reduce the budget deficit, the tax structure should be reformed to ensure fairness as well as to raise the efficiency of tax collection, and government spending on other than civil service costs (for example, conspicuous consumption and spending on armaments and the mil- itary) should be rationalized. If public service employees are retrenched, remedial measures, such as severance packages, rede- ployment and retraining, credit programs, and public works programs, should be implemented. In conclusion, the sale of public enterprises is no panacea for labor market problems in Arab countries undergoing economic re-

239

©International Monetary Fund. Not for Redistribution NADER FERGANY structuring. Private sector development within the present institutional context does not hold much promise either. Nothing short of a complete paradigm shift is needed. Based on an integrated policy package that weaves intensive investment in human capital formation with extensive labor market reform, the paradigm shift also requires institutional change, spanning the range of civil service to governance reform while guaranteeing the social welfare of the people. Such changes could raise the growth of output and productivity, leading to adequate levels of social welfare. Only in this way can the foundation for economic efficiency as well as equity throughout society be laid down. Here again, I am in agreement with the general lessons given in the conclusion of the paper.

References

Fergany, Nader, 1994, "Employment in Arab Countries" (Cairo, Egypt: Almishkat Centre for Research and Training). , 1995a, "Recent Trends in Economic Activity and Open Unemployment" (Cairo, Egypt: Almishkat Centre for Research and Training). 1995b, "Strategic Issues of Education and Employment in Egypt" (Cairo, Egypt: Almishkat Centre for Research and Training). United Nations Development Program, 1994, Human Development Report (New York: Oxford University Press for the UNDP).

240

John Page

would like to begin by supporting Fergany's view of the regional I labor market. Real wages adjusted for skill levels in the Middle East are relatively high compared with those in other regions, although there are some exceptions. Thus, in the absence of rapid growth, sectoral shifts from the production of nontraded goods to traded goods will not be sufficient to absorb the growth in the labor force. Greater flexibility in the labor market across sectors is not in itself enough to achieve full employment at adequate real wage levels. This judgment reinforces the basic proposition of the paper that the political and economic challenge for leadership is to create more good jobs, and not simply more jobs. Good jobs are those that provide wages high enough to improve the standard of living of households throughout the region. Creating such good jobs requires, in addition to adjustments in relative prices, increased investment and improvements in the skills of the labor force. The paper maintains that privatization, by stimulating private investment, has the desirable effect of both increasing investment and improving learning and the acquisition of skills. To establish this, the authors discuss at length the experiences of several countries outside the region, but do not say much about the relevance of those experiences to countries of the Middle East and North Africa. That leaves a number of important questions unanswered. Does the pattern of privatization relevant to the Middle East and North Africa look like the mass privatization in Eastern European countries or more like the selective strategic privatization taking place in Latin

241

America? Is credibility the principal objective of privatization in the Middle East and North Africa; that is, should it be viewed primarily as a signal to promote increased investment by the private sector? Or is the principal objective the enhancement of efficiency in the use of the capital stock currently held in public sector enterprises? In short, the paper discusses international experience, but not its ap- plicability to the MENA region. I believe there are two issues of central importance to privatization policy in MENA. First, fiscal constraints limit the capacity of governments to undertake new public investments and to renew the capital stock of existing public enterprises. Thus, a simple, prag- matic motivation for governments to privatize is to enable formerly public enterprises to undertake needed investments, free from fiscal constraints. Second, the major political vulnerability of privatization policy stems from the fact that in most cases in MENA privatization involves labor redundancy. This conflict of investment and employment objectives is the public policy dilemma of privatization. With- out increased investment, the region's economies will fail to create enough good jobs; in the short run, however, privatization destroys good jobs. The answer to the conflict may lie in improving the skill and adaptability of the labor force. But the authors provide little strategic guidance on how this is to be accomplished. To enable MENA economies to reconcile their privatization and employment objectives, a privatization strategy is needed that can meet the requirements for critical investment in new physical capital while increasing labor force learning and skills acquisition. The experience of East Asian countries suggests that manufacturing of nontraditional exports is a powerful mechanism for the enhancement of learning among workers and managers in exporting firms. Furthermore, spillovers from the export sector to the rest of the economy create a dynamic learning process that tends to transform the comparative advantage of the country as a whole. Thus, a potential employment-oriented privatization strategy is to make privatization complementary to an export push. Such a strategy should consist of selective privitizations designed to relieve the constraints on trade-related infrastructure, such as communications, ports, power, and roads; break up trade-related monopolies in such activities as air and sea freight; and undertake other measures that lower the transactions cost of export activities. Another element of the strategy should be to privatize public industrial sector firms that

242

©International Monetary Fund. Not for Redistribution CHAPTER 5 • COMMENT are upstream producers of inputs to private investments in export- oriented activities. These actions are particularly relevant for Arab countries that have entered, or are about to enter, economic part- nership agreements with Europe, for example, Egypt, Jordan, Mo- rocco, and Tunisia. Some labor redundancies are inevitable during large-scale privatizations and would constitute the largest part of the political cost. This cost can be sustained, however, if rapid export expansion takes place and the net effect on employment and skills in the medium term is positive.

243