Privatization and Social Exclusion in Latin America
After the widespread crises of the 1980s, most Latin American countries embraced institutional and economic reforms intended to reduce fiscal deficits and inflation, liberalize economies, and modernize the state apparatus (Lora, 2007). These reforms inevitably changed the relationship between citizens and the state. Before the reforms the state was seen as a great and even paternalistic social benefactor, serving as both a large-scale employer and a provider of a vast array of goods and services through active participation in markets. This view has changed substantially in recent decades.
While the economic and institutional effects of reform in Latin America have been extensively analyzed, only recently has research focused on how reform has affected the way in which citizens relate to the state. Effects on particular groups and their social and economic inclusion are only now beginning to be considered. This chapter addresses these effects by assessing how one of the most important reforms, the privatization of state-owned enterprises (SOEs), has measurably changed both perceptions and realities of social exclusion.
Latin Americans have generally disapproved of privatizations since the initial stages of structural reforms in the early 1990s. Nevertheless, opinion poll approval ratings for privatization have substantially increased in recent years as the benefits of specific privatizations have become more apparent. A growing majority of the Latin American population is able to enjoy the benefits of privatized enterprises, especially in improved access to and quality of basic services, which in turn improves access to a wide array of economic and social activities. Recent progress notwithstanding, approximately two out of every three Latin Americans take a negative view of privatizations. In the last round of polling by Latinobarometer (2006), covering seventeen countries, only 30 percent of Latin Americans said that they were “satisfied or very satisfied” with the results of the privatization of public services, “considering price and quality.”
What shapes Latin Americans’ opinions of privatization? Gaviria (2006) has shown that support for privatization is closely linked with wealth and the perception of social mobility. The richest quintile of the region’s households are on average 8 percent more likely to approve of privatizations than the poorest quintile. Also, regardless of income level, households whose members perceive that they have experienced or may yet experience social mobility are much more likely than others to approve of privatizations.
More subjective explanations for popular discontent—and perceptions that privatization has exacerbated social exclusion—suggest that this discontent stems from a widespread belief that privatizations have given private investors, who are seen as members of the economic elite, control over assets considered important for the country involved (Birdsall and Nellis, 2002). These feelings became especially apparent during the privatization of telecommunications in Mexico and Peru.
Popular approval of privatizations is further diminished by the absence of a political consensus on which activities should be under government control. According to a 1998 survey conducted by the Wall Street Journal Americas in fourteen Latin American countries, on average 31 percent of those interviewed thought that airlines should be under government control, with 26 percent supporting government management of television stations and 61 percent believing that water services should be provided by the government.
It should also be noted that public perceptions of who benefits from privatizations, at least in the short term, have some basis in fact. Those who approved of privatizations in the Wall Street Journal Americas poll were also those who identified themselves as belonging to the political right, being in the higher deciles of the income distribution, and actively participating in as well as being the first beneficiaries of economic improvements in their countries. Though some may benefit more than others in the short run, there is no definitive answer to questions such as whether privatizations ultimately exclude groups that are already worse off in favor of wealthier interests.
Most research on the effects of privatizations has evaluated the efficiency or productivity gains of private over public management. These gains, however, hold little immediate interest for the public at large, which is probably much more aware of direct welfare effects. This chapter therefore attempts to assess widespread claims that privatizations increase social exclusion and inequality. In particular, the chapter analyzes whether the effects of privatization per se can explain why it has found relatively little favor in the region.
Privatization and Exclusion in the Short Run
It is generally agreed that privatizations in Latin America have brought significant gains in productivity and efficiency.[1] This is notable because Latin America was a pioneer in promoting private participation in infrastructure projects; between 1990 and 2003, about half of the total US$768 billion private sector investment in developing countries was directed to the region. The role of various sectors in the privatization process can roughly be gauged by their respective shares of total privatization revenue. In Latin America, 75 percent of that revenue has been derived from the public service and infrastructure sectors, with 11 percent coming from the financial sector, and the rest from the fuel and manufacturing sectors.
Nonetheless, the extent of privatization in Latin America has varied across industries, and in no country of the region have all SOEs been privatized. Most Latin American countries have privatized their telecommunications, electricity, and, to a lesser extent, water and sanitation services. In contrast, privatization of railway companies, airlines, airports, and expressways has been less widespread. Privatization of the financial and industrial sectors has not been substantial, because private participation in these sectors was already extensive. Most countries have maintained the presence of at least one official bank and retained government control over companies related to natural resources such as oil, natural gas, and copper. Even Chile, one of the countries that has most aggressively embraced the privatization of state-owned enterprises, has maintained official control of companies in key sectors such as copper, oil, banking, the postal system, railways, and ports.
Within the overall trend toward privatization, its extent has differed greatly across countries. As shown in Figure 8.1, for example, some countries with large state sectors such as Costa Rica, Ecuador, and Uruguay privatized only a few companies in the last decade of the twentieth century, whereas other countries, including Argentina, Bolivia, Panama, and Peru, sold off state companies with total values of more than 10 percent of gross domestic product (GDP). Uruguay was the only country that did not privatize companies in its electricity, oil, and telecommunications sectors over this period, perhaps because proposed privatizations were explicitly subject to a popular referendum, a mechanism used in no other country in the region. At the other extreme, Argentina underwent a privatization process that affected practically all infrastructure sectors as well other sectors in which the state was involved; notable exceptions involve some provincial health companies as well as some national and provincial banks.
In Latin America, state companies have been transferred to the private sector primarily to achieve greater efficiency.[2] Before privatization, Latin American SOEs had largely displayed failings common to firms managed according to political criteria: decisions were made in regard to employment, investment, location, or innovation that proved detrimental to profitability and efficiency, thus producing fiscal deficits and undermining institutional frameworks. Evaluating the microeconomic effects of privatization in five Latin American countries, Chong and López-de-Silanes (2005a, 2005b) found that privatization considerably improved companies’ profitability and efficiency. Typically, after privatization, companies increased their net-income-to-sales ratio by fourteen percentage points, mainly through improved efficiency, as unit costs dropped by an average of 16 percent. Figure 8.2 summarizes the study’s main findings related to profitability.
Other indicators yield similar results. For example, the sales-to-assets ratio increased on average 26 percent in Latin American companies that were privatized, and the sales per employee indicator rose notably as well. In Chile and Mexico, the two most outstanding cases, sales per employee doubled in privatized companies, and in certain companies the increases were several times larger. These results might at first glance seem to stem from simply lowering costs and reducing the workforce. In fact, however, in the wake of privatization, company production substantially increased even as the average sales-to-assets ratio and sales per employee improved. Mexico and Colombia registered the greatest average production gains (of 68 percent and 59 percent, respectively). Brazil, which trailed the other countries in the study, increased production by a nonetheless impressive 17 percent.
Increasing productivity in these formerly state-owned enterprises, however, is generally perceived to have come at the cost of labor force reductions and social benefit cuts. In particular, it is commonly held that most workers dismissed from public enterprises in Latin America have been forced to enter the informal sector, thereby losing a stable source of income and access to social benefits. This view is not without a basis in fact: since public companies have often been used to create employment for political reasons, short-run job reduction has generally been necessary to make these companies viable as part of the privatization process.
The magnitude of job losses due to privatization in six Latin American countries is shown in Figure 8.3. Whereas industry-adjusted job losses in Chile averaged only about 5 percent, in Peru and Argentina the industry-adjusted average of job reduction in privatized SOEs was in excess of 37 percent. However, the effect of privatizations on unemployment in Colombia seems to have been relatively modest, at least in the electricity sector, where most privatizations took place (Chong and López-de-Silanes, 2005b).
These short-term findings, however, do not tell the whole story of privatization and employment. In the medium term, many firms ended up rehiring workers who had initially been fired during the privatization processes—once it became clear that the “wrong” workers had been dismissed. As shown in Figure 8.4, privatizations in Latin America have offered a prime example of this so-called adverse selection problem. Since some workers who were let go did in fact end up in the informal sector (Chong, López-de-Silanes, and Torero, 2007), it is clear that social exclusion as a result of privatization layoffs did occur, but this problem was in some cases mitigated by the rehiring of the laid-off workers.
Whether the efficiency gains from Latin American privatization have been driven by productivity-enhancing investment, or by reductions in jobs and social benefits, remains unclear. Although in some large countries large numbers of workers have been dismissed as a result of privatization, in other countries privatized firms have actually created a significant number of new jobs in the medium term. Still, net job loss or creation is not the only factor to consider. More relevant is how particular cases, with particular changes in the relationship between citizens and the larger institutional and economic framework, have led to greater integration or greater social exclusion.
Dynamics of Privatization, Employment, and Exclusion
Critics of privatization maintain that employment reductions are both the primary means of driving up productivity and the major cause of the exclusion of low-skilled and elderly workers from the formal labor market. Although the limited evidence available suggests that labor cost reductions do contribute to profitability gains after privatization, these savings do not explain the bulk of increased profitability (La Porta and López-de-Silanes, 1999). Moreover, job reductions are not the only means of increasing labor productivity, and even when they do occur, they may be accompanied by other cost-cutting measures such as lower wages and benefits.
Productivity, though, cannot be viewed in isolation. Taking into account other labor indicators, Chong and León (2007) find mixed evidence on the benefits of privatization. Managers in privatized firms earn significantly higher wages than their counterparts in either state-owned firms or firms that have always been private, whereas the wages of lower-skilled workers in privatized firms do not differ significantly from those of similar workers in private or state-owned firms. On the other hand, working conditions appear to have significantly deteriorated in the transition from public to private ownership. In addition to labor deregulation throughout the region, there has been a clear trend toward reduction in nonwage labor costs, especially social benefits. In other words, privatized and private firms seem to be favoring temporary workers over those with permanent contracts and employing more low-skilled workers. Under these circumstances, workers’ ability to organize has been diminished, and privatized firms display significantly lower unionization rates than state-owned enterprises.
In some instances, though, productivity gains from privatization have led to higher wages and increases in other forms of remuneration. In Mexico, wages in a broad sample of privatized companies increased an average of 76 percent from 1983 to 1994, well above those in the rest of the economy. Even more surprising, wages increased substantially more for blue-collar workers than for office staff (122 percent compared with 77 percent in the 1983–1994 period). Workers in many privatized companies have additionally benefited from ownership participation programs introduced to increase worker interest in privatization. In Colombia, average wages in privatized manufacturing firms increased by 25 percent after privatization. As in other countries, however, it appears that other labor conditions have deteriorated and the influence of labor unions has been eroded.
A further concern is what has happened to those workers who were laid off during the restructuring process, either before or after privatization. This segment of the population, usually drawn from the groups that are most vulnerable to economic shocks of any kind, arguably faces the greatest risk of social exclusion due to privatization. In fact, one of the leading concerns surrounding privatization has been that laid-off workers may be unable to obtain a similar job in the private sector because of age, low skills, or the accumulation of human capital that is not transferable to other industries. Although data limitations have made it nearly impossible to seriously address this issue on a large scale in most countries, Chong, López-de-Silanes, and Torero (2007) have analyzed the conditions of laid-off workers in Peru both before and 10 years after privatization. Even though these laid-off workers were given a compensation package, the study found that the average worker suffered a significant initial hit after being fired, which validates concerns regarding the impact of privatization on inequality and social exclusion. On the other hand, the study also found that workers’ wages and benefits eventually recovered to the same level as those of private sector workers in their industry.
Perhaps more surprising is that “stayers” in Peru command higher wages and benefits than comparable workers who had been laid off because of privatization or who had always been in the private sector. Workers in Peru’s former SOEs have apparently been able to extract more than other workers as a result of firm market power, union power, or favorable terms of a collective contract that remains in force. This result also helps to explain why the compensation of workers who have lost their jobs because of privatization reverts to the mean of the corresponding private sector industry.
As noted above, social exclusion does not operate only through specific markets such as the labor market. It can be assessed as well through outcomes related to the welfare of the excluded population. Such an approach reveals that the typical worker in Latin America who is laid off as the result of privatization consumes his or her compensation package within the first two years, usually through investment in a home or through the creation of a small business. Unfortunately, however, the average new business fails by the end of that period, and the worker moves on to activities related to his or her previous employment. These results contrast with the belief that workers lose in the long term after privatization and that workers who lose their jobs are condemned to unemployment or poverty.[3] In general, workers laid off as a result of privatization experience a significant setback at first, then quickly recover until their wages and benefits converge with those of similar workers in the private sector.
Distributive and Social Effects
The most important social and distributive effects of privatizations in Latin America may involve citizens in their roles as consumers rather than workers. In the postprivatization era, some firms in the region have been seen as exerting monopolistic power, affecting both supply and prices. Whether these perceptions have a factual basis is debatable, but they have undoubtedly influenced how the region’s population views privatization.
The privatization of public resources provides an important example, as its effects on poverty and inequality are a combination of two factors. First, prices usually increase significantly in the wake of privatization, which can prove particularly difficult for low-income groups. On the other hand, privatization of services often leads to expanded coverage, greatly benefiting groups that are on average even poorer. In addition, symbolic issues as well as material outcomes are relevant, as access to public services may generate a sense of inclusion and provision of basic rights to historically excluded populations.
Recent research on the effects of privatization on income and inequality has reached revealing conclusions. A comparative study by McKenzie and Mookherjee (2003) found that electricity and water privatizations had positive effects for all income groups in Argentina.[4] In Bolivia, however, telephone privatization was found mainly to have benefited the middle class, among whom the expansion of coverage was most evident. It should nonetheless be noted that the approach pursued by these authors offers only a partial overview of the benefits of privatization, because it does not consider effects other than income.
Several recent studies, including work sponsored by the IDB, offer very convincing evidence that privatization yields benefits in addition to those just described. Several studies have found important benefits from privatization in terms of health, time management, and beneficiaries’ employment opportunities in Argentina, where privatizations have met with widespread opposition. Galiani, González-Rozada, and Schargrodsky (2007), for instance, found that the expansion of waterworks to marginal shantytowns around Buenos Aires by a privatized firm led to a significant reduction in the number and severity of children’s episodes of diarrhea. Households additionally realized significant savings of both money and time because they no longer had to look for water. Even households that had illegal connections before privatization experienced improvements in health, as the water available through those connections was of very poor quality. Those who previously had no connections saved time, as they no longer had to bring water from faraway dwellings. These results parallel those from previous work (Galiani, Gertler, and Schargrodsky, 2005) showing that the privatization of water services in Argentina was followed by a 5–7 percent reduction in child mortality, with the greatest effects in the poorest areas.
In another category of services, González-Eiras and Rossi (2007) have addressed the effects of the expansion of privatized electricity networks on health outcomes in Latin America. Access to electricity and continuity of service have important indirect effects, as refrigeration problems can lead to food spoilage and public health problems such as food poisoning and malnutrition, particularly among children. The authors found that in Argentine provinces where electricity distribution was privatized, the frequency of low birth weights decreased relative to provinces that maintained public networks, with similar results for child mortality.
Somewhat more mixed results have emerged from research by Barrera-Osorio and Olivera (2007) on water service in Colombia. Privatization of Colombia’s water services has shown positive effects overall, according to these authors, especially in urban areas, where both coverage and water quality have increased. In addition, poorer areas have experienced an increase in the frequency of water service, at the expense of decreasing frequency in richer areas, and the weight-to-height index—a proxy for health status—has increased for children in privatized municipalities. At the same time, major increases in water service prices may have outweighed the benefits yielded by other improvements.
In Peru, however, as in Argentina, IDB-sponsored research has found privatization to be largely beneficial, particularly for campesinos and poor rural workers. Torero, Nakasone, and Alcázar (2006) found that the continuity and reliability of privatized electrical service in Peru allowed users to spend less time on agricultural work and more on nonagricultural or leisure activities, with positive effects on income and welfare. Similarly, Chong, Galdo, and Torero (2005) found that privatizing Peru’s telephone service delivered important benefits to the rural poor. In randomly selected villages where the government required the private company, Telefónica del Perú, that replaced the SOE to install public telephone booths, residents experienced improvements in income; particularly notable increases occurred in nonagricultural income, which is crucial for stabilizing the income of the rural poor. Although this benefit resulted from government requirements rather than the goodwill of the company, it is also true that the public company could not or did not want to provide public telephone booth service in the past, which suggests that government can target the benefits of privatization to favor less privileged social groups. In this case, privatization has clearly served to reduce social exclusion.
The expansion of service that has accompanied privatizations not only affords the less well-off the immediate opportunity to use the services provided, but also offers the possibly more important benefit of a sense of inclusion in society. Increased access to such services further allows Latin Americans to enjoy a higher quality of life and provides them with the opportunity to generate more stable sources of income.
Conclusion
Privatizations can be socially inclusive, but delivering their benefits to the poor requires government regulation of privatized companies. Although the benefits of privatization per se may be debatable, it is clear that a carefully designed transition, sound management, and adequate oversight of the process are necessary, and that social costs and benefits must be taken into account as well. It is especially important to consider the psychological and ideological factors that influence perceptions of whether privatization has inclusive benefits. Research in behavioral economics, particularly the field of prospect theory pioneered by Kahneman and Tversky (1979), has identified three such psychological factors that are particularly notable. The first is the tension between individual experience and aggregate statistics: exclusion as a result of negative personal outcomes such as job loss weighs much more heavily than small and widely dispersed gains.
A second and related psychological factor is that, in social exclusion as in other areas, individuals give much greater weight to losses than gains in respect to their initial situation. Finally, perception is biased toward short-term outcomes. Individuals are thus more likely to notice sudden changes than gradual changes such as increases in the coverage or quality of certain services.
Even if widespread perceptions of privatization’s effects do not necessarily match officially recorded and “objective” measures of progress, policymakers ignore those perceptions at their own peril. Politicians and officials cannot simply dismiss objections to privatization as mistaken or vaguely assure voters that their conditions will eventually improve. In Latin America, where public confidence in the competency and honesty of government is low, a different approach is needed to convince much of the electorate that privatization can and does reduce rather than increase social exclusion.
