2009 LaborRegulationsUnionsandSocial

• (Freeman, 2009) ⇒ Richard B. Freeman. (2009). “Labor Regulations, Unions, and Social Protection in Developing Countries: Market Distortions Or Efficient Institutions?." National Bureau of Economic Research.

Subject Headings: Labor Market Institution; Labor Regulation; Labor Union; Economic Growth

Cited By

• http://scholar.google.com/scholar?q=%22Labor+regulations%2C+unions%2C+and+social+protection+in+developing+countries%3A+market+distortions+or+efficient+institutions%3F%22+2009

Quotes

Abstract

This essay reviews what economists have learned about the impact of labor market institutions, defined broadly as government regulations and union activity on labor outcomes in developing countries. It finds that: 1) Labor institutions vary greatly among developing countries butthey vary among advanced countries. Unions and collective bargaining are less important in developing than in advanced countries while government regulations are nominally as important. 2) Many developing countries compliance with minimum wage regulations produce spikes in wage distributions around the minimum in covered sectors. Most studies find modest adverse effects of the minimum on employment so that the minimum raises the total income of low paid labor. 3) In many countries minimum wages "spill-over" to the unregulated sector, producing spikes in the wage distributions there as well. 4) Employment protection regulations and related laws shift output and employment to informal sectors and reduce gross labor mobility. 5) Mandated benefits increase labor costs and reduce employment modestly while the costs of others are shifted largely to labor, with some variation among countries. 6) Contrary to the Harris-Todaro two sector model in which rural-urban migration adjust to produce a positive relation between unemployment and wages across regions and sectors, wages and unemployment are inversely related by the "wage curve". 7) Unions affect non-wage outcomes as well as wage outcomes. 8) Cross-country regressions yield inconclusive results on the impact of labor regulations on growth while studies of country adjustments to economic shocks, such as balance of payments problems, find no difference in the responses of countries by the strength of labor institutions. 9) Labor institution can be critical when countries experience great change, as in China's growth spurt and Argentina's preservation of social stability and democracy after its 2001-2002 economic collapse. Cooperative labor relations tend to produce better economic outcomes. 10) The informal sector increased its share of the work force in the developing world in the past two decades. The persistence of large informal sectors throughout the developing world, including countries with high rates of growth, puts a premium on increasing our knowledge of how informal sector labor markets work and finding institutions and policies to deliver social benefits to workers in that sector.

Introduction

Once about a time – not so long ago – the international financial institutions and many in the economics and policy establishment believed that they knew how to create sustainable growth in developing economies. They had a tool kit of policy prescriptions that they could take from country to country to cure economic ills. For the labor market, the package called for reduced regulations and lower social protection, cuts in public sector pay and employment, weaker unions, and greater reliance on market wage setting compared to collective bargaining or administrative rules. The enemy of growth was “urban bias” (Lipton, 1977) -- government or union setting of pay and work conditions that benefit modern sector workers but that reduce the flow of workers from low productivity informal and rural sectors to the modern sector. The World Bank’s 1990 Development Report presented the prevailing wisdom: “Labor market policies – minimum wages, job security regulations, and social security – are usually intended to raise welfare or reduce exploitation. But they actually work to raise the cost of labor in the formal sector and reduce labor demand ... increase the supply of labor to the rural and urban informal sectors, and thus depress labor incomes where most of the poor are found.” (World Bank 1990, p. 63).

Underlying this perspective was the Harris-Todaro (1970) two-sector model that attributed joblessness in developing countries to institutionally imposed high urban wages. The model posited that the high wages induced rural workers to migrate to urban areas, where they became unemployed while waiting for good jobs. Migration continued until the rate of unemployment equated the expected urban sector earnings (the wage times the probability of employment) to rural earnings. In this situation, an increase in modern sector employment at the institutionally determined wage does not raise GDP. This is because the addition of a high productivity job induces enough rural workers to migrate into urban unemployment to reduce rural output by the increased output due to the new urban job.

World Bank and International Monetary Fund economists also worried that labor institutions would undermine structural adjustment programs designed to cure balance of payments deficits or other economic ills. Viewing the archetypical problem as one in which the developing country ran into a balance of payments deficit, they stressed the need to shift resources from labor-intensive non-traded goods and services to capital-intensive traded goods sectors. The least costly way to do this was to devalue the currency, which would raise the price of tradeable goods and services relative to non-tradeable goods and services and thus attract resources into the traded sectors. As long as tradeable goods were capital intensive, this would also shift the income distribution toward capital. The fear was that unions or other institutions that raised wages to preserve labor incomes would stop relative prices from moving in the desired direction. Absent a price-induced shift in resources, the country would have to undergo a recession to reduce imports and raise exports, which would be far more costly than a real devaluation.

At the 1992 World Bank Annual Conference on Development Economics, I reviewed extant evidence that labor institutions harmed economic development and stymied adjustments to macro-economic problems per this analysis and found it sparse and unconvincing (Freeman, 1993a). The strongest evidence was Fallon and Lucas's (1989) comparison of the response of employment to output and wages in 35 industries in India and 29 industries in Zimbabwe before and after these countries strengthened labor laws. Their analysis showed that industries adjusted employment to changes in output as rapidly after the laws as before the laws but that employment was lower at the same output after the laws (ie that productivity improved, which could be interpreted as a positive outcome). Absent evidence that firms complied with the laws and that other factors did not affect outcomes over the same period (they noted that Zimbabwe became independent co-terminus with the change in labor regulations) I viewed the results as inconclusive at best. I was more impressed by the large declines in real minimum wages and average earnings in many African and Latin American countries during the 1980s that suggested that labor regulations were more “sawdust” than “hardwood”.

The quantity and quality of research on labor institutions in developing countries has increased greatly since the early 1990s. Some countries changed labor regulations in ways that provide good pseudo-experiments of whether institutions help or hinder the working of labor markets. Many countries now regularly provide researchers with micro data files on individuals and establishments that permit deeper probing of hypotheses than is possible with aggregate data. Research institutions and individual researchers have developed new data sets with country labor codes and institutional practices that illuminate cross-country differences and provide input into cross-country growth and other regressions. In light of all this, what have we learned about how labor institutions affect outcomes in developing countries?

The recent research has not uncovered a general law for the effects of institutions on outcomes – economic circumstances and institutions probably vary too much among countries to support any single generalization – but it has yielded new and in some cases surprising findings on how institutions affect outcomes. This has led to a more measured view of what institutions do than in the World Bank’s 1990 proclamation. Here are the main findings:

1) Labor institutions vary greatly among developing countries butthey vary among advanced countries. Collective bargaining is weaker in developing countries than in advanced countries while labor regulations are nominally similar.

2) Contrary to my initial skepticism, compliance with regulations in the formal sector of many developing countries is sufficient that minimum wages appear to be binding. They produce spikes in the distribution of wages around minimum. Most studies find that minimum wages reduce employment sufficiently modestly so that minimums generally help the low paid.

3) Contrary to Harris-Todaro type models, minimum wages induce spikes in the distribution of earnings in the informal sector in several countries, suggesting that minimum determine reservation wages of workers in those sectors.

4) Wages and unemployment are negatively related across geographic areas, consistent with the wage curve and contrary to the Harris-Todaro model.

5) Mandated benefits increase labor costs and reduce employment modestly while the costs of others are shifted largely to labor, with some variation among countries.

6) Some mandated benefits increase labor costs and reduce employment modestly while the costs of others are shifted largely to workers and thus presumably do not impact employment.

7) Unions are associated with higher wages and non-wage shares of compensation and with lower turnover and less dispersion of pay. Estimates of the union effects on profits and productivity differ across countries.

8) Cross-country regressions yield inconclusive results on the impact of labor regulations on growth while studies of country adjustments to economic shocks, such as balance of payments problems, find no difference in the responses of countries by the strength of labor institutions. The research has led analysts at the World Bank and other international institutions to moderate their initially negative assessments of labor institutions. Readers familiar with the retreat of the Bank and the IMF from obiter dicta on free trade, unrestricted capital flows, and laissez faire policies will note that this fits with the new modesty of these institutions about what economists can scientifically assert about growth-inducing policies.

9) Labor institution can be critical when countries experience great change, as in China’s growth spurt and Argentina’s preservation of social stability and democracy after its 2001-2002 economic collapse. Cooperative labor relations tend to produce better economic outcomes.

10) In the 1990s-2000s, the informal sector’s share of employment increased or held steady in virtually all developing countries, including those with healthy growth and limited regulations (section 8). Even without deregulating the formal sector, an increasing proportion of workers in developing countries are working in largely unregulated markets.

1. The Debate over Labor Institutions

Developing countries, like advanced countries, evince substantial differences in labor institutions that could impact economic outcomes and growth. To quantify this variation, I summarize in exhibit 1 the mean and standard deviation of five measures of the institutional orientation of formal sector labor markets. The five measures are: the labor component from the Fraser Institute’s (2006) index of economic freedom; the Botero, et al (2004) indices of the strength of employment laws and laws regarding collective rights; the power of firms to set wages and hire and fire reported in the (World Economic Forum's Globa Competitiveness Report (2006) and rates of unionization from the ILO (1996). I have scaled the indices so that high values mean that a country relies more on market forces than on institutions in determining outcomes. I differentiate developing countries by level of income and distinguish the traditional advanced countries (the West and Japan) from recently the developed Asian Tiger economies. Appendix A gives the measures for each country.

References

,

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2009 LaborRegulationsUnionsandSocial	Richard B. Freeman			Labor Regulations, Unions, and Social Protection in Developing Countries: Market Distortions Or Efficient Institutions?