Trade Liberalization and Labor Reform in LAC in 1990s: A Brief Note

Transcription

1 Trade Liberalization and Labor Reform in LAC in 1990s: A Brief Note 1 For a more extensive coverage on the contents of this section see From Natural Resources to the Knowledge Economy: Trade and Job Quality (Chapter 5). Latin American labor markets performed below expectations during the 1990s in terms of both formal employment creation and poverty reduction. As a result concern has been growing among different groups that trade liberalization and increasing integration in the world economy have caused workers in Latin America to lose ground. In this note we attempt to shed some light on the debate by revisiting the existing empirical evidence. For this purpose, we first examine the relationship between trade liberalization and labor market outcomes during the 1990s, concluding that globalization does not seem to be the main suspect behind the trends described above. We then turn to labor market rigidities and the extent to which reform efforts facilitated formal employment creation. Here we find substantial variation across countries in the policies chosen and thus different degrees of success. Finally, based on this discussion we end the note by taking stock of what we learned from the 1990s, and propose a new agenda for labor market reform that we hope is more in accordance with the new environment in which Latin American countries now operate. 1. Is Globalization to Blame? Trade Reform and Labor Market Performance in the 1990s 1. Although several countries in the region experienced low employment creation and growing unemployment during the 1990s, over the long run there seems to be no relation between these developments and trade openness. Specifically long-run trends in unemployment rates were not substantially higher in the 1990s than in the 1980s and 1970s, which cannot be characterized as periods of substantial trade liberalization. Instead the empirical evidence seems to suggest that high levels of unemployment in the last decade reflect the long transition periods associated with macroeconomic reforms, rather than their final outcomes. Table 1. Average unemployment rates in LAC (%) Country 1970s 1980s 1990s Entire Period Argentina Bolivia Brazil Chile Colombia Costa Rica Honduras n.a México Paraguay Peru Uruguay Venezuela, RB LAC - Mean LAC - Median Source: SIMA Database for the 1970s and 1980s, and UN and ECLAC for the 1990s.

2 Paired with poor formal employment creation most countries experienced substantial increases in informality over the 1990s 2. However, as in the case of unemployment, there is no evidence that this was due to subcontracting in global production chains and little evidence that it was more generally due to trade liberalization. In fact in many countries increases in informality at the beginning of the 1990s were driven partly by the expansion of opportunities in non-tradeable sectors, such as construction and services, which tend to be informal. Moreover, in those countries where labor market segmentation was a problem, with employment in the informal sector being a fall back option, this segmentation could not be attributed to trade reforms in most cases so much as to macro-adjustments and onerous labor legislation. In addition the jobs created in the emerging sectors appeared to be better than similar jobs in other sectors. In particular, jobs in EPZs, new agricultural sectors, tourism and tele-services compared favorably to work in more traditional sectors in terms of pay and benefit, while female labor force participation and relative wages were also high in these sectors, thus giving women more leverage in household 3. Finally, although wage inequality did worsen in several countries, numerous studies have failed to find direct links to trade liberalization. Rather this trend seems to result from skill biased technological change as well as from insufficient growth in the supply of medium/high skilled workers relative to their demand (technological progress). In fact average wages in the region rose at historically high levels during the 1990s and more open sectors appeared to pay higher than average wages (adjusted for human capital Figure 1). Figure 1. Wages Adjusted for Human Capital by Sector Wages adjusted for human capital 200.0% radables 150.0% % increase over non t 100.0% 50.0% 0.0% anl petr o mat f or tr op cer l ab cap mach chem Ar genti na* Br azil Costa Rica Mexico Rep.Dom Venezuel a -50.0% * ur ban wor ker s onl y Leamer categor ies Source: From Natural Resources to the Knowledge Economy: Trade and Job Quality (The World Bank, 2001) In sum, the evidence presented here shows that even though the region performed below expectations in terms of employment creation and poverty reduction during the 1990s, these 2 CEPAL estimates that 6 of every 10 new jobs were created in informal sector. 3 One should not forget, however, that there is some concern that working conditions other than pay fell below desirable standards due to tight competition with other developing countries and small comparative advantage margins that leave little room for improving working conditions.

3 developments were generally due to factors other than increasing globalization. Our discussion, however, should not be regarded as dismissive of the important effects that trade liberalization has had on the region, nor should it be read as saying that all these effects were positive. 2. Working for Us or Against Us? Labor Market Rigidities and Reform Efforts in the 1990s 4. The developments described above forced Latin American governments to turn their attention more directly to the functioning of labor markets in the region, in an attempt to understand why economic growth induced by macroeconomic stability and trade liberalization had not translated into higher employment creation in the formal sector. In doing so excessively rigid legislation was identified as a potential explanatory factor for this poor performance, hence making labor market flexibility the main goal of any labor reform. This, however, was easier said than done, and proposals to increase flexibility often faced substantial opposition, making them difficult to implement. 1.- Job Security Legislation: Cost and Cause of Dismissal Latin American labor legislation traditionally imposed high firing costs on employers to achieve job security and employment stability. Theoretically these costs included the stipulated severance pay and possibly other forms of compensation depending on the cause of dismissal. However, in practice employers also needed to take into account transaction costs associated with workers appeals and lengthy court procedures. While such a system may have had a certain logic in closed economies where firms were unlikely to go bankrupt and the slow process of innovation meant little need for job destruction, in today s increasingly open economy it imposes tremendous costs on workers and employers. In fact high firing costs have long been considered a barrier to job creation in Latin America and a fair amount of empirical work has then been devoted to exploring the effect that job security and severance pay have on employment and unemployment. Although the evidence on job security is somewhat mixed both at the regional and the country levels, it seems to point towards a negative effect of job security on employment creation. An impression that is then corroborated by more conclusive evidence regarding tenure-based severance pay systems (Heckman and Pagés, 2000; Pagés and Montenegro, 1999). However, recommendations for reform born from these studies have not translated into uniform policies across the region. Rather we find that while some countries decreased severance pay in the 1990s (Argentina, Colombia, Guatemala, and others), others did the opposite (Brazil, Chile, and the Dominican Republic). These differences respond mostly to the fact that in most countries severance pay serves also as unemployment insurance, so that reforms to the system are not only concerned with the potential negative effects of severance pay on employment creation but also with protecting workers at the time of separation. The outcomes of severance pay reforms, the only ones for which empirical evidence is available, were as varied as the reforms themselves. Some patterns arise, however, with reductions in severance pay generally favoring more disadvantaged workers (youth, female workers), and increases in severance pay favoring tenured workers and increasing the cost of formal employment relative to informal employment. 4 For a comprehensive review of labor market reform in Latin America and the Caribbean see Gill et alia, 2001, and Weller (2001).

4 2.- Unemployment Insurance and Compensation Funds Since severance pay systems are in most cases an inadequate way of providing protection for unemployed workers, some countries in the region started to experiment with alternative systems such as unemployment insurance and compensation funds. However, adoption of alternative compensation systems did not come hand in hand with a reduction in severance pay (see previous point). Rather these new systems were generally introduced on top of existing labor legislation, thus putting additional stress on labor taxes and introducing substantial inefficiencies. Moreover these alternative compensation systems were generally characterized by strict eligibility criteria (access to funds is often conditional on the cause of dismissal or requires a certain number of months of contributions to the unemployment fund) and low and poorly targeted coverage of the overall unemployed population. 3.- Employment flexibility and temporary contracts Prior to the 1990s labor legislation in most Latin American countries prohibited the use of temporary contracts for permanent activities, restricting their use to work that was temporary in nature, such as harvesting or construction. Moreover, it was common for the use of these contracts to involve high administrative costs (e.g. labor union consent on a case-by-case basis). These restrictions undoubtedly had negative effects on formal employment creation and encourage low compliance and informality, especially among small businesses for whom flexibility is key. Aiming to minimize these negative effects some countries changed the rules concerning temporary employment during the 1990s, introducing fixed-term and other types of contracts when they did not exist, or making their use more flexible and attractive when they were already available. These changes had significant effects on the labor market and although each country s experience was different, a few common patterns can be identified. First the introduction of temporary contracts generally translated into their extensive use. However, when the administrative requirements associated with the use of these contracts were too strict, temporary contract use did not expand as expected. Second the introduction of temporary contracts translated into higher employment creation that would have been otherwise achieved in some countries, especially formal employment creation. Third these reforms were often accompanied by higher worker turnover, although this did not seem to translate into substantially lower unemployment rates. Hence summing up the discussion in this section, while legislative changes implemented to increase flexibility were fairly similar across countries (e.g. new contractual forms, probation periods), there was substantial variation in the way these same countries tackled the worker protection issue. In particular some of them continued to rely on severance pay systems to protect workers in case of unemployment, whereas others experimented with alternative compensation systems. Finally it is also important to remark that despite the economic benefits associated with some of the reforms, they were generally unpopular; sometimes so much so that certain governments were forced to backtrack on some of the changes. This goes to show how important the political economy of reform is. 3. What Remains to Be Done? Guidelines for Labor Market Reform in the New Decade.

5 This final section takes stock of what can be learned from the 1990s and proposes a new agenda for labor market reform. This new agenda is meant to be more in accordance with the new environment in which Latin American countries now operate an environment of increasing integration in the global economy and rapid technological changes, the so-called new economy. 1. A change of focus from reducing stipulated firing costs to reducing transaction costs associated with firing. As we pointed out above, most of the reform efforts during the 1990s were devoted to changing the rules of game in the labor market, in order to make them more flexible. In the context of hiring and firing costs this translated into reductions in stipulated severance payments and other forms of worker compensation in case of job loss in some countries. Unfortunately playing under this new rules did not turn out to be any simpler than playing under the old rules, with most involuntary separations still leading to lengthy court cases with uncertain outcomes that were costly for both the worker and the employer. That is, even though stipulated firing costs diminished in most cases, transaction costs remained high. This reality then calls for our energies to shift from further reductions in stipulated firing costs to clearer, faster implementation of the existing rules. Shorter, less uncertain procedures would reduce transaction costs for both the worker and the employer without diminishing the actual compensation received by the former. Faster procedures would also allow workers to start looking for another job sooner if they so desire. This would also require clearer, more flexible rules concerning dismissals for economic reasons in order to allow firms to adjust to less favorable market conditions while continuing in business, at the same time that effective monitoring systems and effective safety nets are put in place to ensure that a more lenient policy on this regard is not misused or abused by employers nor does it makes workers systematically worse off. 2. A change of focus from labor-market based compensation mechanisms to portable benefits and effective safety nets. Worker compensation mechanisms in the region have traditionally been linked to the labor market. For instance, job loss severance payments are supposed to provided workers with (transitory) income while they find new employment. Unfortunately, as we discussed above, such a system has not performed as well as expected. It is costly due to lengthy litigation regarding separations, and it provides uneven coverage and uncertain outcomes due to the fact that severance payment liabilities are in most cases unfunded. In other words, relying on severance payments to provide coverage for workers in the event of job loss does not satisfy the principles of universal insurance and risk pooling both desirable features for a safety net that is expected to function in an environment of increasing globalization and more fluid and dynamic labor markets. In fact the new open economy calls for safety nets that offer workers portable benefits (i.e. compensation funds based on personal savings accounts), combined with a minimum level of universal coverage in case of job loss (i.e. unemployment insurance). This kind of system has important advantages over the current severance-payment-based one.

6 Portable benefits are more valuable to the worker since she can access them immediately after separation and administer them at will (note that this may also have effects on informal/formal employment decisions). For instance in Colombia, where workers are given the option of quitting the compensation fund and receiving slightly higher wages, Kugler (2000) finds that in 1995 only 1.5 percent of workers in manufacturing and 0.6 percent of workers in commerce preferred the higher salary to fund. In addition, employers view contributions as variable costs linked to employment rather than fixed costs linked to separation, and the cost of minimum insurance after separation is shared by all participants in the system rather than falling exclusively on the employer. 3. A change of focus from labor market flexibility to flexibility of labor as a production input. Today, as we take stock on the past reform efforts and think about new directions for change and improvement, we need to realize that Latin American labor markets are de facto rather flexible in the traditional sense of the word (e.g. flexible wages and non-binding minimum wages in most cases, large informal sector). This leads us to think about flexibility in a more innovative way. It leads us to think about flexibility in the context of the production process rather than in the context of labor market institutions. Increases the flexibility of labor as a production input requires investing in workers capacity to learn, adapt and contribute to the ever faster changing work environment. For this purpose we need to invest extensively in human capital emphasizing (i) education and training as part of a life-long learning process, and (ii) problem-solving skills rather than memorization of certain tasks/principles. It also requires that workers operate in a more fluid workplace where merit is rewarded over seniority. This means more mobility and flexibility of labor within the workplace (e.g. less restrictive occupational mobility regulations, development of internal labor markets to replace seniority-based promotion systems) 4. A change of focus from informality as a less-advantage sector of the labor market to informality as a largely voluntary and heterogeneous sector 5. A long tradition views informal workers as comprising the less-advantaged sector of a dualistic or segmented labor market. Above market-clearing wages force workers to queue for preferred jobs while subsisting in the informal sector, that is characterized by an absence of benefits, irregular work conditions, high turnover and, overall, lower rates of remuneration. A recent variant on the dualism view, albeit with different emphasis, sees informalization as an effort by firms facing international competition to reduce these legislated or union induced rigidities and high labor costs, particularly through subcontracting production out to unprotected workers. However, based on the (scarce) empirical evidence available, not only does the informal sector appear to be very heterogeneous but most informal workers seems to be so voluntarily. Workers may prefer to be informal because (i) they value benefits less than the corresponding wage deduction, (ii) there are weak linkages between taxes and benefits, (iii) the formal sector offers few openings for promotion (rigid, tenured-based system) or mobility, especially for the low skilled. 5 For more on this issue see Maloney (2001).

7 The informal sector is not simply or even primarily absorbing the unemployed from the formal sector, and neither does it seem to be a way for international firms to avoid protection. Rather informality seems to serve a double function as a port of entry for young, poorly educated workers into paid employment, and as a more flexible regime for business operation for more experienced and entrepreneurial workers. This dichotomy between voluntary and involuntary informality has important policy implications, since the causes for one or the other require very different treatment. However this new and more positive take on the informal sector, as a fairly heterogeneous and dynamic sector, should not make us forget involuntary informality. A large numbers of informal workers would prefer, if given the option, the benefits and security offered by formal employment to the flexibility that characterizes the informal sector. This in turn emphasizes the need to reduce the cost of formal employment relative to informal employment, and to extent social protection to the informal sector.. Summing up with optimism for the new decade In the early 1990s it would not have been an exaggeration to characterize labor reform as the forgotten reform. As mentioned above, most countries had by then put in place successful macroeconomic stabilization programs, and many had already launched financial sector reforms. By the middle of the decade, however, the labor policy agenda could be upgraded from forgotten to unfinished, and governments in the region seem committed to continue working on labor reform during this new decade. Operating under this belief, this note has tried to suggest new directions for labor reform efforts in the next few years, emphasizing the need to create dynamic labor markets and effective safety nets to fight vulnerability and foster productivity, rather than to perpetuate a culture of job stability and protection of the employed. Finally it is important to remember that, although a substantial change of direction has been suggested here, this second round of reforms would not be possible without the changes that took place in the 1990s, so that those years should be regarded as the time when the seeds of durable labor reform were sown in LAC.

8 References Gill, Indermit S., Claudio E. Montenegro and Dörte Dömeland, editors (2001) Crafting Labor Policy: Techniques and Lessons Learned from Latin America. The World Bank, Washington, DC. Heckman, James and Carmen Pages (2000) The Cost of Job Security Regulation: Evidence from Latin American Labor Markets. NBER Working Paper No. 7773, National Bureau of Economic Research, Cambridge, MA. Maloney, William (2001) Informality Revisited. Mimeo, The World Bank, Washington, DC. Pagés, Carmen and Claudio Montenegro (1999) Job Security and the Age-Composition of Employment. Evidence from Chile. Office of the Chief Economist. Working Paper No. 398, Inter-American Development Bank, Washington, DC. Weller, Jürgen (2001) Economic Reforms, Growth and Employment: Labour Markets in Latin America and The Caribbean. CEPAL, Santiago (Chile). World Bank, The (2001) From Natural Resources to the Knowledge Economy: Trade and Job Quality. PREM/LAC, The World Bank, Washington, DC.