40526 D I R E C T I O N S I N D E V E LO PM E N T Human Development Minimum Wages and Social Policy Lessons from Developing Countries Wendy Cunningham Minimum Wages and Social Policy Minimum Wages and Social Policy Lessons from Developing Countries Wendy V. Cunningham ©2007 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved 1 2 3 4 5 11 10 09 08 07 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The bound- aries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. ISBN-10: 0-8213-7011-1 ISBN-13: 978-0-8213-7011-7 eISBN-10: 0-8213-7012-X eISBN-13: 978-0-8213-7012-4 DOI: 10.1596/978-0-8213-7011-7 Cover art: Construction Workers, 1976 woodcut, by Vernal Reuben, Jamaica. World Bank Art Collection 466540. Reproduced by permission of Susan Reuben and with assistance from the World Bank Art Program. Library of Congress Cataloging-in-Publication Data Cunningham, Wendy V. Minimum wages and social policy : lessons from developing countries / Wendy V. Cunningham. p. cm. Includes bibliographical references. ISBN-13: 978-0-8213-7011-7 ISBN-10: 0-8213-7011-1 ISBN-10: 0-8213-7012-X (electronic) 1. Minimum wage--Latin America. 2. Social policy--Latin America. I. Title. HD4920.L38C86 2007 331.2'3098--dc22 2007003319 Contents Acknowledgments ix Introduction xi Chapter 1 Overview 1 Introduction 1 Report Findings 2 Issues for Policy Discussion 3 Note 5 Chapter 2 Context and Framework 7 The Historical and Social Justice Perspective--Birth and Evolution of the Minimum Wage 7 Theoretical Underpinnings--the Classical Economic View 10 Two-Sector Economies--Models with a "Covered" and "Uncovered" Sector 12 Empirical Evidence from the OECD Countries 13 The Latin American Research 15 Notes 16 v vi Contents Chapter 3 Minimum Wage Institutions in LAC: What Are They and Who Earns Them? 19 What Is a Minimum Wage? 19 There Is No Common "Minimum Wage System" for LAC 20 Who Earns the Minimum Wage? 26 How High Is the Minimum Wage in LAC? 31 Notes 34 Chapter 4 The Worker: How Do Minimum Wages Affect Other Wages and Employment? 35 Minimum Wages Are Somewhat Binding 35 Minimum Wage Policies Increase Wages throughout the Wage Distribution 38 The Level of the Minimum Wage Determines Whether It Increases or Decreases Wage Inequality 41 Wage Benefits Are Not Concentrated on Any Particular Group of Workers 42 Minimum Wages Increase Unemployment 44 Subnational Minimum Wages Are Standard Practice 47 Notes 48 Chapter 5 The Households: The Minimum Wage as an Antipoverty Tool 51 The Value of the Minimum Wage Is below the Household Subsistence Level 52 The Minimum Wage Decreases Poverty Rates but Does Not Help the Most Poor 53 High Minimum Wages Increase Household Income Inequality 56 Notes 57 Chapter 6 The State: The Minimum Wage Implications for Public Expenditures 59 The Minimum Wage May Have Large Impacts on the Public Sector Wage Bill 59 An Increase in the Minimum Wage Can Have Substantial Impacts on the Cost of Social Benefits 61 Notes 62 Contents vii Chapter 7 The International Community: Lessons from Their Experiences 65 Setting and Managing Minimum Wages 66 Enforcement 68 Note 70 Chapter 8 Report Conclusions and Policy Considerations 71 Report Conclusions 71 Considerations for Policy Debates 75 Note 77 References 79 Appendix I Research Methodologies 87 Notes 99 Appendix II Summary of Literature 101 Appendix III Source of Data for Cross-Country Comparisons 119 Appendix IV Kernel Density Plots 121 Index 127 Boxes 2.1 Employees against High Minimum Wages? 8 2.2 The Motivation for the Creation of Minimum Wages 9 3.1 Protecting the Wages of Soccer Players 25 5.1 Methodology for Testing the Minimum Wage Effects on Household Poverty and Inequality 55 6.1 Methodology for Simulations 60 Figures 3.1 Distribution of Minimum Wage Earners 26 3.2 Minimum Wage Relative to the Mean or Unskilled Wage 33 4.1 Cumulative Density Functions Showing "Spikes" at Multiples of the Minimum Wage 40 4.2 Average Wage Increase due to a 1% Increase in the Minimum Wage, by Position in the Wage Distribution before the Wage Change (Full-Time Workers) 41 viii Contents 4.3 Mexico, Average Wage Increase due to a 1% Increase in the Minimum Wage 43 4.4 Kernal Density and Cumulative Density Plots of Wages, by Region 47 5.1 Minimum Wage Relative to the Household per Capita Poverty Lines--US$2 per Day or the Consumption-Basket Poverty Line 52 5.2 Impact of the Minimum Wage on Household Income in Mexico, 1999 56 Tables 3.1 Main Institutional Characteristics Affecting Minimum Wages in Selected Latin American Countries 21 3.2 Ratio of Each Subgroup's Share of the Minimum Wage Population, Relative to Its Share of the Total Workforce (Full-Time Workers Only) 28 3.3 Ratio of Each Subgroup's Share of the Subminimum Wage Population, Relative to Its Share of the Total Workforce (Full-Time Workers Only) 29 3.4 Proportion that Earns at or below the Minimum Wage, by Demographic Characteristic 30 3.5 Minimum Wages in LAC 32 3.6 Ratio of the Minimum Wage to the Median and 10th Percentile Wages 33 4.1 Degree to Which the Minimum Wage Is Binding in Wage Employment in LAC and Year of Analysis, by Sector 37 4.2 Summary of Key Empirical Literature on the Effect of the Minimum Wage on Wages in Latin America 39 4.3 Change in the Wage-Gini Coefficient (by Demographic Group) due to a Decrease in the Minimum Wage, 1988­99 44 4.4 Summary of Key Literature on the Effect of the Minimum Wage on Unemployment and Employment in Latin America 45 5.1 Review of the Key Literature Examining the Impact of the Minimum Wage on Household Poverty and Inequality in Latin America 54 6.1 Increase in the Public Sector Wage Bill due to a 20 Percent Increase in the Minimum Wage, by Percentage 60 Acknowledgments The report was prepared with inputs from a team of professionals. Special recognition is deserved by the team who wrote the background papers: Carlos Arango, Rolando Guzman, Nicolai Kristensen, Magdalena Lizardo, Dayna Lora, Naercio Menezes Filho, David Neumark, Angela Pinchon, Laura Ripani, and Lucas Siga. Andres Lopez and Laura Saenz provided research assistance for the entire report. The process of report development greatly benefited from the technical inputs, debates, and direction provided by Ariel Fiszbein, William Maloney, Jorge Moreno, Carolina Sanchez-Paramo, Sergei Soares, Francisco Carneiro, Mauricio Santamaria, Luis Serven Diez, Guillermo Perry, Helena Ribe, and Jennie Litvack. Of course, all views presented and errors herein are the sole responsibility of the author. ix Introduction The minimum wage is an attractive policy tool for poverty reduction and social justice. It does not require significant direct government expenditures, is a simple and visible way for the government to show its commitment to social justice and support those at the bottom of the income distribution, is easily targeted to the poorest workers, and affects a market--the labor market--in which Latin American govern- ments are comfortable intervening. Other social programs with poverty reduction objectives have been tried, such as cash transfers or public works, but they tend to be difficult to target and monitor, impose high nonlabor costs, and create political economy disputes. Thus, the self- targeting, lower monitoring, low leakage, "right" worker incentive and labor market­focused characteristics of the minimum wage may make it an attractive social protection tool. The minimum wage was created in the late 19th century in New Zealand and Australia, and within 30 years it had a strong presence in Latin America and the Caribbean (LAC). The design of LAC's minimum wage institu- tions was based on two principal objectives. The fair wage objective1 centered on the idea that each occupation has a fair wage, which may be different from the level determined by the market. Ideally, collective bar- gaining would correct the imbalance, but if that was not possible, it was xi xii Introduction the government's responsibility to set and enforce the wage structure by creating occupation-specific wages so that each worker received his or her relative due wage. The poverty alleviation objective was based on the idea that market wages would not necessarily be at a level that was socially acceptable to society, so the minimum wage would ensure enough income for all citizens to maintain a minimum standard of living, regardless of occupation. Over time, the role of the minimum wage in LAC con- tinued to develop to meet the changing economic and social needs in the Region, resulting in a heterogeneous mix of institutions today. Despite the long history of the minimum wage, very little is known about its effectiveness in meeting its social justice or poverty alleviation roles. Although it may increase consumption by elevating wages above their market level, it may also lead to job loss (through layoffs) and decreased consumption. Thus, its usefulness for poverty and inequality reduction is uncertain. Similarly, in the social justice context, an exoge- nously imposed wage structure may be a means of overcoming unfair market allocations of wages, but the market will still drive labor demand, and higher-than-market wages may lead to unemployment. This begs the question of how the minimum wage can be designed to be an effective tool to ensure social justice or poverty alleviation. The literature in the United States and western Europe shows that the minimum wage has modest impacts on wages and employment (mostly with respect to youth), with few poverty or inequality impacts, but there are many reasons to think that the effects on wages, employment, and household income in LAC may differ from those in the Organisation for Economic Co-operation and Development (OECD). Relative to the OECD, LAC has very large informal sectors in which labor laws are not enforced, a greater proportion of adults who earn very low wages and thus may be affected by a minimum wage, incomplete social safety nets for counterbalancing the negative effects of the minimum wage, and a frequent practice of linking the minimum wage to government expen- ditures. Thus, minimum wages in LAC may affect different sectors of the population and have higher impacts on household welfare and public finance than in the United States and Europe. Although the minimum wage exists in every country in the LAC Region, there is very little research on the wage and employment effects of the minimum wage and even less on the poverty and household inequality effects. This is particularly troubling, since the minimum wage is a very active policy tool--renegotiated regularly--and labor markets are particularly sensitive to the negative impacts of excessive labor market Introduction xiii regulations (Heckman and Pages 2004). Currently, governments across the region are rethinking their social protection systems to make them more integrated and cohesive, so this is an ideal time to empirically sort out the impacts of the minimum wage in LAC and provide policy makers with new information on this policy tool. MinimumWage Debates in the Region Policy makers have been pondering the question of how the minimum wage can be used as a social protection tool. Several issues are being debated in the LAC Region today regarding minimum wages: · Is the minimum wage an effective poverty reduction tool? Throughout the Region, the minimum wage is believed to benefit the poor. In many countries, it served as a benchmark to protect the poor during hyper- inflationary periods, and it is still debated regularly when the (semi-) annual adjustments to the minimum wage are made. These debates are largely speculative, with little empirical evidence to back them up. · Do minimum wages exacerbate the unemployment situation? Throughout the Region, unemployment is an increasing problem. In these market economies, an above-market wage is expected to create unemployment, so the question is how high can the government set the wage without exacerbating unemployment? · Are minimum wage policies a means to increase low wages? As the economies of LAC open to the world, and as highly skilled labor becomes more valuable, the wage gap between skilled and unskilled workers is increasing. Returns to education--one of the principal factors in wage determination--among the most skilled are very high (nearly 20 percent for some countries) and increasing, whereas the returns are much lower, and stagnant, among the less skilled, thus leading to increasing wage inequality and the problems that accompany it.2 Clearly, an increase in education and investments in technology are the long-run solution, but in the short run, wage policies may be useful to control inequality. · Are minimum wage policies effective in economies with a large informal sector? Informal sector workers earn the lowest wages and are therefore the group most in need of wage protection. However, informal sector xiv Introduction workers are by definition outside the government's sphere of direct influence, so minimum wage policies may skip over exactly the group that they are intended to benefit. · Can the minimum wage offer labor protection to vulnerable groups? Youth unemployment is a grave concern in LAC. Minimum wages are thought to deter labor force entry of young individuals, since young people's marginal productivity may be lower than the minimum wage. Thus, discussions about the need to waive the minimum wage for young people through a subminimum apprenticeship wage is being tested in countries as diverse as St. Vincent and the Grenadines and Colombia and is being discussed in other countries in the region. However, others argue that the minimum wage should be above the marginal productivity of young workers to give them the incentive to remain in school rather than join the labor market. The increasing spread between the wages of low- and high-skilled workers is also becoming a concern, particularly with trade liberalization. · What is the cost of the minimum wage to the public deficit? Minimum wages are tied to social programs in many LAC countries, so changes in the minimum wage may exacerbate deficit issues. Furthermore, the large public sector in some countries, the increasingly trouble- some pension deficit, and increasing or stubborn poverty lead to much wider consequences from minimum wage increases. The Objective of the Report and Methodology This report contributes to the debate by attempting to better understand the distributional effects of the minimum wage and thus its usefulness as a policy tool for reducing poverty and inequality. It does not discuss whether or not minimum wage policies should exist. Instead, it assumes that minimum wages will continue to be a part of social policy and pres- ents new research useful to policy makers in designing the minimum wage and the accompanying larger social policy. The study begins with the existing literature on minimum wages in the Region, but expands the discussion in three ways. First, the household is placed at the center of the debate. Although it is important to understand the employment and wage effects of minimum wages, ultimately we are interested in the general equilibrium effects at the level of the household, which is the unit of observation for poverty measurement. This report Introduction xv presents new research that allows for risk pooling at the household level, where the net implications of job loss and income can be quantified. Second, new research is presented on how the minimum wage affects groups whose labor market participation and success is considered "vulnerable," that is, youth, women, the low-skilled, and informal sector workers. Third, the report provides new discussions on the implications of the minimum wage for state finances by moving beyond the pension issue and into questions of the public wage bill and indexing of social benefits to the minimum wage. Whereas much of the information presented in this report is drawn from the existing literature, additional information was created to round out the lessons. Specifically, six background papers that apply the latest methodologies from U.S. and European literature to data from the Region were commissioned for this study. The results of the papers are discussed in this study, and the methodologies are presented in appendix I. Particular attention is given to the unique characteristics of LAC, which may differentiate the research findings from those in the United States and Europe, where most minimum wage research has been done. This requires treating the formal and informal sectors separately and allowing for minimum wage effects beyond the low-wage population. Although this is a Regional study, a caveat is in order. The results are presented for LAC, although every country is not discussed. General information is presented for most countries in the Region, but a subset of countries is selected for deeper analysis that could provide wide- reaching lessons. Mexico, Brazil, and Colombia provide results for countries where the minimum wage is low, medium, and high, respectively, relative to unskilled wages.3 Report Organization The report has eight sections following this introduction. Chapter 2 presents a history of the minimum wage in LAC, the theory behind the functioning of the minimum wage, and empirical evidence from the OECD to lay a foundation for the Latin American experience. Chapter 3 presents an overview of the minimum wage in the Region, including a discussion of the definition of a minimum wage, institutional design, and who earns it. Chapter 4 focuses on the worker; it summarizes the exist- ing literature, presents new evidence on the wage and employment effects of a minimum wage, and gives special attention to "vulnerable" labor market groups. Chapter 5 turns its attention to the household and xvi Introduction presents the new (and only) evidence on the effects of the minimum wage on household poverty and inequality in LAC. Chapter 6 considers the state and discusses the cost of minimum wages to the government. Chapter 7 opens the discussion to the rest of the world and considers the lessons learned in other countries about setting, managing, and enforcing the minimum wage. Finally, chapter 8 concludes and presents policy considerations. Specific details about research methodologies and regres- sion results are given in appendix I. Notes 1. Since 1999, the International Labour Organization (ILO) has argued that the fair wage concept should be the basis for minimum wage regimes. 2. High wage inequality is associated with higher levels of violence (as the have-nots are increasingly faced with the wealth that they do not have), political discord, and the creation of two societies, side by side, with different levels of development (Fajnzylber et al. 1999). 3. In addition, these countries have long enough (noninflationary) time series and sufficient variance in the minimum wage over the period of the data to permit in-depth analysis. C H A P T E R 1 Overview Introduction The minimum wage originated as a social justice tool to provide socially acceptable wages to the most unskilled workers. By the time it was adopted in Latin America, it had become more generous, guaranteeing a wage that would allow for a good life for a worker and his or her family, including adequate food, shelter, clothing, hygiene, and leisure. However, economic theory points out that setting a wage above its market value will lead to unemployment, primarily among the most unskilled workers-- precisely those the minimum wage is intended to protect. Thus, the question for policy makers is how a minimum wage policy can ultimately help or hurt the poorest workers. Despite widespread debate over the appropriate level and design of the minimum wage throughout the region, there is surprisingly little empirical evidence on its actual impact. This report examines how minimum wages affect the income poverty of workers, their households, and the state. It does not question whether or not the minimum wage is a good policy: instead, it focuses on the trade- offs in setting the minimum wage level. It takes as a starting point the literature on the wage and employment effects of minimum wages in Latin America and expands the discussion in three ways. First, the household is 1 2 Minimum Wages and Social Policy placed at the center of the debate. Poverty and inequality are measured at the level of the household, rather than at the individual level, to allow for employment and wage trade-offs among individuals who pool their income. Employing recently introduced statistical techniques, this study is the first that uses estimated, rather than simulated, measures of the effect of the minimum wage on household poverty and inequality in the Region.1 Second, new research is presented on how the minimum wage affects groups whose labor market participation and success is considered "vul- nerable": that is, youth, women, the low-skilled, and informal sector workers. Third, the implications of the minimum wage on wage and social expenditures of the government are measured. In the end, the report argues that the minimum wage by itself is not a sufficient tool for protecting the income of the poorest households, and that other social protection tools are necessary to complement it. Report Findings Minimum wage policies matter in Latin America and the Caribbean (LAC), unlike the United States or western Europe. A large fraction of the labor force in Latin America and the Caribbean is not covered by minimum wage laws since these persons work in the informal sector. Among those covered, the minimum wage is often not enforced, partly because of weak institu- tions. Despite these facts, the minimum wage appears to have a larger effect than expected for two reasons, as follows: (a) Both formal and informal sector workers' salaries are affected by the minimum wage. In fact, the minimum wage is more binding in the informal than the formal sector. This may be due to supply- or demand- side factors. On the supply side, the minimum wage is considered a "fair wage" in the economy, regardless of sector. On the demand side, even informal sector employers will voluntarily offer the fair wage to attract labor and minimize labor turnover. (b) The effects are felt by a larger fraction of the labor force than just minimum wage earners. Only youth in the OECD, whose wages are clustered around the minimum wage, benefit from minimum wage increases. In Latin America and the Caribbean, wage gains are largest for those earning near the minimum wage, but the wage benefits (and job losses) are experienced throughout the entire wage distribution. Overview 3 The minimum wage can be a tool for poverty and inequality reduction. Although the minimum wage falls below the poverty line in households with a single worker, it can serve to increase the incomes of the poorest workers. In countries with relatively low minimum wage levels, it increases the wages of poorer workers, has minimal effects on unemployment, and leads to an aggregate net increase in household labor income for the poor households. It also reduces income inequality because the positive impacts shrink to zero in higher-earning households. The minimum wage can be set so that it leads to greater income inequality. However, in countries with relatively high minimum wages, the minimum wage can increase wages of higher earners, increase unemployment among the poor, cause greater poverty among the poor, and cause an increase in household labor income inequality. Thus, whereas a high minimum wage may serve a social justice objective, setting it too high can have the oppo- site effect. The most vulnerable workers are the first to be hurt by high minimum wages. Even where minimum wages are relatively low, they dispropor- tionately affect certain groups in the labor market. The wages of young, low-skilled, and female workers benefit from a higher minimum wage--as do those of prime-aged, skilled, and male workers--but the unemploy- ment effects are also concentrated on the former groups. In essence, their income is redistributed to less vulnerable segments of the labor force when the minimum wage increases. Social justice is costly to public finances. Social expenditures are closely tied to the minimum wage throughout Latin America and the Caribbean, so any increase in the minimum wage also has large implications for the public deficit. The largest expenditure category affected is the public sector wage bill. For example, a 10 percent increase in the minimum wage would increase total government expenditure by 1.4 percent in Panama. However, high costs come through tying the minimum wage to other social benefits. For example, in Brazil, a 10 percent increase in the minimum wage would increase pension payouts by 3.3 times more than the increase in receipts. Issues for Policy Discussion Careful thought should be given to the structure of minimum wage institutions to enable their use as effective policy tools. Simply having a minimum wage is not sufficient to cause or prevent poverty reduc- tion and income equality. Instead, it is a tool that must be carefully 4 Minimum Wages and Social Policy calibrated--in the context of other institutions, norms, values, and constraints of the economy and a society--to reach policy makers' objectives. The most important dimensions for development of an effective minimum wage are as follows: Clear objective. The minimum wage has played roles ranging from poverty reduction to macroeconomic stabilization. It cannot be a cure-all for society's problems; it is most effective if it has a focused role and if the institution is designed to fulfill this role. Simple structure. The least effective minimum wage institutions are those that are too complex to be effective. A clear objective will simplify the de- sign of the minimum wage institutions, but even within this, it is better to design a simple system that is well understood by all, rather than trying to fully address the heterogeneous needs of the labor force. A system should have only as many minimum wages as it can enforce, and all criteria should be well defined. It is also crucial to delink the minimum wage from other policies that affect public finance so that minimum wage decisions are independent from public finance issues or from other social decisions. Enforceable. Although there is a degree of self-enforcement of the mini- mum wage in Latin America, and clear objectives and a simple structure will facilitate this, it is still necessary to create incentives so that the policy will be followed.This requires setting a wage that is sensitive to the impact on employers; identifying, publicizing, and enforcing punishments for lack of compliance; clearly identifying who is responsible for enforcement of each minimum wage (for example, if the minimum wage differs by occu- pation, region, or demographic group, it is best if specific bodies, whether unions, regional offices, or advocacy groups, watch over the enforcement of their respective minimum wage); and adequately funding and supporting the administrative body responsible for enforcement. Set at a level that balances social goals and market responses to minimum wage. A minimum wage that is too high may actually increase poverty, so social justice guidelines are not sufficient to set a minimum wage. Simi- larly, the market may recommend wages that are too low to be acceptable by society. A careful balance must therefore be struck between the objec- tives. Clear guidelines have not been identified, primarily because the social expectations and the market reaction to a certain value of a minimum wage will differ across countries. Instead, trial and error with high-quality Overview 5 impact evaluation will define a set of parameters for accurate minimum wage setting in the context of each country. The minimum wage may cause unemployment, so there is a need to combine it with other social welfare tools. Whereas a well-targeted policy of tax and redistribution would most efficiently achieve the income inequality reduction objective, the minimum wage may be a more realistic means for achieving this goal. It avoids many of the pitfalls of other social policies: self-targeting, market (rather than government) income redistrib- ution, minimal public administration, no perverse labor supply incentives (receipt of benefit is contingent on being in the labor force), and it does not ring of paternalism (asistencialismo) or political favoritism (clientelismo), as the beneficiaries are selected by the market. However, minimum wages can cause unemployment, particularly among the most vulnerable workers. Thus, a minimum wage that is high enough to affect wages should be accompanied by a tax-and-redistribute scheme that reallocates income from all workers--not just the poor, as in the case of the minimum wage--to the poor who lose their jobs as a result of the policy. Note 1. The research on the household poverty effects of the minimum wage is scarce in the Organisation for Economic Co-operation and Development (OECD) literature as well. In the Latin America Region, there exists research that sim- ulates the poverty effects of the minimum wage and examines the "poverty" of the individual as a result of the minimum wage, but both are based on strong assumptions that may drive their results. C H A P T E R 2 Context and Framework The historical development of minimum wage institutions and the theo- retical underpinning provide a context for deeper analysis of the tool. The brief review in this chapter is not intended to be comprehensive. Rather, it demonstrates the high expectations countries have for the tool and that under fairly plausible conditions, the minimum wage may have a more limited effect (or even the reverse) than that hoped for. The OECD literature is also reviewed, exposing its limitations for under- standing the LAC situation. Theoretical ambiguity about the impact of the minimum wage makes the empirical work presented in this study critical to formulating minimum wage policy. The Historical and Social Justice Perspective--Birth and Evolution of the MinimumWage The minimum wage was created in the late 19th century in response to public demands for social justice among workers.1 The objective was to alleviate the situation of "sweating," defined as working at wages so low that they did not support "a socially acceptable level of wholesome fam- ily life" (Nordlund 1997). The low wages were attributed to the failure of a competitive market to offer an adequate wage level and the lack of 7 8 Minimum Wages and Social Policy bargaining power by the most vulnerable workers, namely women, chil- dren, and the unskilled. Or, as stated by the English economist, Sidney Webb, who is credited with coining the term "sweating" (Webb 1912): Experience has demonstrated, to the satisfaction of public opinion, as well as of economists, that to leave the determinants of wages, in a capitalist organization or industry, to the unfettered operation of "individual bargaining" and the "haggling of the market" between individual employers and individual wage-earners, is to produce, in the community, a large area of "sweating"-- defined by the House of Lords Committee of 1890 as "earnings barely suf- ficient to sustain existence." The original proponents of the minimum wage argued that low wages impose substantial negative externalities on society, and that government has a responsibility to correct the market and serve as a collective bar- gaining proxy for poor workers without representation to ensure them a standard of living acceptable to society. Early court cases show that despite the unconstitutionality of minimum wages in some countries, and legal suits by workers against the minimum wages (because of their adverse employment effects--see box 2.1), public pressure slowly won as minimum wages were instituted across the world. Latin America was one of the leaders in instituting minimum wages, but the policy objective differed from that of the originators.2 After World War II, many Latin American countries began to adopt detailed legislation to socially engineer the labor markets in response to the population's expec- tation that the government was responsible for their well-being and that markets were not able to produce the socially desirable results (Starr Box 2.1 Employees against High MinimumWages? The wage-employment trade-off of the minimum wage can lead employees to fight the government to lower minimum wages. For example, in 1923 in the United States, the Supreme Court ruled on the case Adkins v. Children's Hospital (chi. 8.88), in which a group of women sued the District of Columbia to halt the implementation of a minimum wage that had led to their losing their jobs. The court ruled in favor of the prosecution, stating that the minimum wage was a form of price fixing and an unreasonable infringement on individuals' freedom to choose the price at which they would sell their services. Context and Framework 9 1993). Constitutions and legislation specified that wages should be sufficient to provide for food, shelter, clothing, transport, leisure, and other basic necessities of life. Thus, the motivation for a minimum wage was partly social justice and partly to affect the wage distribution in order to offer "fair" wages (see box 2.2 for other motivations). Box 2.2 The Motivation for the Creation of MinimumWages The rationale for creation of a minimum wage varied across country, time, pressure group, and institutions. Some of the motivations included the following: Protectionofthemostvulnerable.Thiswastheearlyobjectiveoftheminimumwage, where those with the least bargaining power and the most inhumane living stan- dards were the target of the policy. Poverty reduction. Linked to the concept of protection of the most vulnerable, this objective identifies the"most vulnerable"as the poor and argues for a single minimum wage that allows a basic standard of living for all citizens. Payment for inputs. Early court cases in the United States argued that employers are obligated to pay for the production of the human energy required to supply the labor to their firms. Thus, wages should be sufficient to cover the cost of food, shelter, leisure, clothing, and other inputs that create labor. A minimum wage should be equal to this cost, and it would be unjust for employers to pay a lower wage. Fair labor standards. The International Labor Organization (ILO) 1919 charter argued that all occupations and industries have a "fair" wage. Ideally, collective bargaining would identify the fair wage for each industry, but since some indus- tries were unable to organize, a minimum wage set by the government was the second-best solution. This philosophy extends the concept of a bargained wage for all workers, not just the most vulnerable. Fair competition. Early employers in favor of minimum wages argued that com- petition for factor inputs was unfair, as it would give certain employers an unfair advantage in production costs. A minimum wage could promote fair competi- tion among entrepreneurs. Macroeconomic objectives. The minimum wage can also be used to affect the entire wage distribution, which may lead to economic growth, inflation control, or political gains, to name a few. Source: Starr (1993). 10 Minimum Wages and Social Policy By the 1960s, many Latin American countries assigned a larger role to the minimum wage: to be an instrument of macroeconomic policies (Starr 1981). This ambitious role for the minimum wage is particular to LAC. It became a tool for inflation stabilization, economic growth (through increasing purchasing power), poverty reduction, income inequality reduc- tion, and political gains. It took on the role as a numeraire for other social programs, and became so prevalent that workers and employers would specify the value of wages in terms of the "number of minimum wages." Many of these macroeconomic objectives failed, leading to a paring- down of the role of the minimum wage, including legislation to delink wages from the minimum wage and use of alternative macroeconomic tools. However, this has not eliminated use of the minimum wage as a social equity tool, and it still is expected to play a large part in poverty and inequality reduction in the Region. Minimum wage institution in the English-speaking Caribbean evolved differently. These countries adopted the minimum wage system of their British colonial occupiers, which was the classic implementation of a fair wage system: a series of wage boards that set wages by industry. With independence, several of the countries abandoned the complex wage board system in favor of simpler minimum wages with a poverty reduction objective (Starr 1993). Theoretical Underpinnings--the Classical EconomicView Economists use the labor market as the starting point for understanding the role of the minimum wage. The most basic view of the impact of the mini- mum wage begins with a downward-sloping aggregate labor demand curve, capturing the declining marginal return to labor with greater employment. A minimum wage set above market-clearing wage forces firms up the demand curve and reduces employment, either by reducing aggregate output or substitution away from labor to other factors of production. Such a wage floor is likely to have the impact of reducing the dispersion of wages and guaranteeing a fair or living wage for the poorest workers who retain their jobs. However, moving from the individual to the impact on household income distribution or poverty is less straightforward. Although some workers are now paid more, other workers earning near the minimum will lose their jobs, leaving the net impact on the target population--the poor--dependent on several factors: The magnitude of the accompanying job loss. A very low elasticity of demand implies that large wage gains can be imposed with little job loss, Context and Framework 11 implying a positive total transfer of income from the firms to the target worker group. In the case of one-company towns where the firm may have monopsony power, there may be no job loss and the minimum simply transfers rents from the firm to the workers. At the other extreme, where the elasticity of demand is high, the loss in income of dismissed workers exceeds that gained by those remaining. An elasticity of 1 implies that there are no net transfers--there is simply redistribution of earnings from the newly unemployed poor to the working poor. The structure of the social unit. If all families had only one worker, the impact of raising the minimum wage on poverty would not be ambiguous. Those retaining their jobs would gain, and if they were pushed above the poverty line, poverty would decrease. Those dismissed would now be without income and would fall below the poverty line. Often, however, there are multiple workers in a family who pool risks and hence one worker's job loss may be offset by another's wage gain. As the number of working family members increases, the impact on the family approaches that of the working sector as a whole: that is, the net transfer to poor households depends on the elasticity of demand.3,4 The overall social insurance context. In most advanced countries, the minimum wage is only one element of an overall social protection system that includes an unemployment benefit that replaces some fraction of dismissed workers' wages. This implies that even in the case of unitary elasticity, there are transfers from the society at large to the poorer groups, generally financed by a progressive tax. Latin America generally lacks systems to support the unemployed.5 Further, to the degree that the unemployed may include those who are looking for their first job, the usual severance pay system will provide no protection. The classical view of the minimum wage does not necessarily dovetail with the social justice objectives of the policy. The concept of a "fair" wage is not defined in the model. Instead, the market adjusts to the policy intervention without any consideration of the welfare effects or the social acceptability of the change. Thus, although a higher minimum wage may be socially desirable, the market may not support it. This places policy makers in the difficult role of determining a level of the minimum wage that maximizes social justice objectives while minimizing market distortions. The challenge is even greater in LAC than in the United States or western Europe, given the "two-sector economy" nature of their labor markets. 12 Minimum Wages and Social Policy Two-Sector Economies--Models with a"Covered" and"Uncovered"Sector The presence of a large informal sector uncovered by labor legislation in LAC alters some predictions of the classical view. In most LAC countries, 30­70 percent of workers are covered (Maloney 2003), so a change in the minimum wage law will not necessarily affect employment of all low- wage earners. Much of the development literature assumes that the two sectors con- stitute a dualistic labor market where those who cannot find a job in the formal sector find inferior jobs in the informal sector (Harris and Todaro 1970, Mazumdar 1983, Stiglitz 1984, World Bank 1995, IDB 2000). In this view, a nominal rigidity, such as the minimum wage, that forces the wage in the formal sector above market clearing causes dismissed workers to be absorbed into the informal sector, where wages fall to equate labor supply and demand. The wage adjustment in the informal sector makes the impact of a rising minimum wage on the distribution of wages ambiguous. Earnings are com- pressed in the formal sector, but the informal wage relative to the formal wage will fall and informal dispersion may increase as well. In the classical model, where there is no informal sector, these workers would be unem- ployed and hence a zero wage would enter in dispersion measures. In a dual market, this is no longer the case as they may enter the informal sector. In fact, the minimum wage may have effects emerging from other channels, for instance, through the increased demand for informal goods from richer formal sector workers. Analytically modeling such general equilibrium impacts becomes quite complex (see, for example, Fiszbein [1992], Agenor and Aizenman [1999], and estimating the final impacts of such nominal rigidities has been attempted by calibrating Computable General Equilibrium models (see May [1993], and Devararjan, Ghanem, and Thierfelder [1997], among others). Such exercises are useful for clari- fying the channels of influence and getting plausible orders of magnitude. Analysis is further complicated by recent literature that argues that the dualistic structure underlying much of this literature, and in particular the assumption of the informal sector as primarily the refuge of those unable to find jobs in the formal sector, describes only a minority of the sector. Rather, the informal sector can be conceived as very heterogeneous, with a large, unregulated entrepreneurial sector where many entrants are, in fact, voluntary and enjoy welfare and even wage gains on leaving formal sector employment (Cunningham and Maloney 2002). Recent work on Context and Framework 13 Mexico and Argentina (Arango and Malone 2000, Arias 1999) finds that, in fact, most unemployment is generated by the informal sector, and hence it is unlikely that the sector is primarily absorbing formal sector unemployment. Numerous authors (see, for example, Geldstein [2000] argues that entry into formality is not trivial and hence is not be the sec- tor of easy entry for the unemployed. Numerous Brazilian authors and Maloney and Nuñez (2004) have found a "lighthouse" effect where, in fact, informal salaried workers appear to be more affected by the minimum wage than the formal sector. This suggests that the sector is less characterized by being infe- rior, but rather is able to pick and choose the regulations it complies with. Even though, for a variety of reasons, an informal salaried worker may not receive pension or other benefits, a notion of "fairness" may dictate that his or her wage broadly follows the minimum in the formal sector (Maloney 2003; Maloney and Nuñez 2004; Foguel 1997). This may imply, however, that the effect of minimum wages in LAC, with its formal and informal sectors, is a bit closer to the classical world than originally thought. A rise in the minimum wage may lead to displacing workers in the informal sector to unemployment. In sum, the impact of a rise in the minimum wage on employment, earnings, the distribution of wages, the distribution of household con- sumption, and poverty depend on a host of factors that do not allow us to predict, on the basis of theory alone, whether the impact is positive or negative. For this reason, we turn to the empirical evidence. Empirical Evidence from the OECD Countries The evidence from the United States and Europe gives a useful starting point for testing these models and the discussion of the effects in Latin America.6 A wealth of research on the employment and, to a lesser extent, the wage effects of the minimum wage in the United States and western Europe give clear lessons. The studies analyzing the poverty and income inequality effects are much fewer. An increase in the minimum wage is associated with an increase in wages for those who earned below or near the new minimum wage, but these effects are neutralized in the long run. In the United States and the OECD countries, young people constitute the majority of minimum wage earn- ers; thus, teenagers' wages are primarily affected by the minimum wage, with little effect on adult wages. On average, teenagers who were earning 14 Minimum Wages and Social Policy above the old minimum wage at the time of the increase experience an immediate increase in wages (Grossman 1983, Neumark, Schweitzer, and Wascher 2000, Katz and Krueger 1992, Card and Krueger 1995, Card, Kramarz, and Lemieux 1999), but a wage decrease in the long run, as the supply of labor increases more than the demand for more skilled workers. An increase in the minimum wage has little or no impact on employ- ment. Most of the empirical evidence on minimum wage impacts on unemployment in developed countries focuses on teenage workers. For the United States (1970s and 1980s), Canada, Portugal, and Greece, a 10 percent increase in the minimum wage is estimated to decrease teenage employment by 0­3 percent, with higher disemployment elas- ticities for female than male teenagers (Brown 1999, Ghellab 1998). Conversely, in the United Kingdom (Machin and Manning 1996), the United States (Currie and Fallick 1996; Abowd, Kramarz, and Margoliz 1999; Neumark, Schweitzer, and Wascher 2000), and France (Card, Kramarz, and Lemieux 1999) in the 1990s,7 employment effects have not been generally identified. An increase in the minimum wage does not reduce household poverty. The paper written by Neumark and Wascher (1997) is the only one that uses microdata to directly estimate the poverty effects of a change in the minimum wage. They show that an increase in the minimum wage causes an increase in incomes of the poor and near-poor in the United States so that 4.1 percent of rise out of poverty, but 3.9 percent of those who are near-poor fall below the poverty line. The 0.2 percent net decrease in poverty is not statistically significant, indicating no real improvement in aggregate poverty. Instead, the minimum wage redistributes income and jobs among households clustered around the poverty line rather than redistributing income from wealthy households to poor households. Burkhauser, Couch, and Wittenburg (1996) use simulations with similar results, and explain that the absence of any effects is a result of the dis- tribution of minimum wage earners across the population, rather than being concentrated in low-income households. The minimum wage effects on family income inequality are mixed. In the United States, an increase in wage inequality from 1973 to 1992 is attributed to a fall in the real value of the minimum wage, especially among women (DiNardo, Fortin, and Lemieux 1996), but no effects on family income inequality were observed (Horrigan and Mincy 1993). Card and Krueger (1995) find that a 26.6 percent increase in the mini- mum wage leads to an increase in family income by an average of 6 per- cent for families in the lowest wealth decile and an increase in family Context and Framework 15 income of 2.5 percent for families in the fifth (median) decile, whereas no changes in family income are detected in the top wealth deciles. However, the results are not strongly statistically identified. Minimum wage effects differ little by gender and race. In the United States, the minimum wage does not have a larger effect on the wages or employment of women, as compared to men, or of black workers as com- pared to white workers (Brown 1999). An exception is DiNardo, Fortin, and Lemieux (1996), who find that a higher minimum wage increases female within-gender inequality, but this is not true of men. The Latin American Research If the preceding findings, mixed as they are, could be applied to Latin America, policy makers would have good information with which to develop minimum wage policies. However, there are various reasons to expect that the OECD findings will not tell the whole LAC story. First, LAC has a very large informal sector. As discussed earlier in this chapter, minimum wage effects in a dualistic economy are difficult to predict. Since 30­70 percent of LAC's workers are employed in the informal sector, this is a significant issue, requiring empirical work to sort out the effects. Unfortunately, the evidence of minimum wage impacts on informal sector workers in the United States and western Europe is very scarce, largely because their informal sectors are very small. Second, the low-wage population in LAC includes adults. Most of the minimum wage effects in the OECD were identified among youth, since they are the population group clustered around the minimum wage. In LAC, though, adults are also low-wage earners. Thus, the conclusion by Burkhauser, Couch, and Wittenburg (1996) that minimum wages have little effect on household poverty since they largely affect secondary house- hold workers, is less applicable to LAC, where minimum wage policies may affect the primary income earner in the household. Third, the minimum wage acts as a numerator in LAC. During peri- ods of hyperinflation, wages were indexed to inflation in several Latin American countries, so the minimum became a standard against which to measure wages. Although such indexing does not formally exist in LAC at the present, the informal practice of denominating wages in terms of the value of the minimum wage still does. Fourth, LAC has weak social protection systems. The Region has gone through an extensive overhaul of its labor market since its rapid entry into the world market in the 1980s. The social protection systems that 16 Minimum Wages and Social Policy were effective under closed economies are no longer suitable. Thus, countries are in the process of redesigning their social protection models to create integrated, cohesive, and comprehensive systems. These com- plementary programs are crucial for changing incentives and thus how minimum wages will affect wages, employment, household poverty, and income inequality. Although research on minimum wages in Latin America is relatively new, there is a body of work, mostly for Brazil, examining the employ- ment and wage impacts of minimum wages.8 These papers will be reviewed in detail throughout this report (and summarized in appendix II), but in general, they show that minimum wages have different effects in Latin America from those in the United States and Europe: The mini- mum wage has strong effects on employment and wages, the effects are felt throughout the wage distribution, and the minimum wage plays a role in the informal sector. These findings will be further tested in this report and the question of the minimum wage effects on poverty, inequality, special groups, and the public finances will also be addressed. Notes 1. The concept of the "maximum wage" first emerged in medieval Europe. The Black Death killed between one-third to one-half of the population in the 14th century, leading to a severe labor shortage and increasing the price of labor. In England, the landowners who depended on this labor were also the policy makers. They passed acts defining a maximum wage: workers and employers could bargain on the amount of time worked, but the payment was capped. Violations of the law were punishable by a fine equal to six months of the maximum wage. The severe shortage of labor led to secret informal arrangements among skilled laborers and their employers; these skilled laborers eventually banded together in trade unions and the famous "secret societies," including the Freemasons (Ridley 1999). 2. Mexico was the first country in the world to include in the constitution a guarantee for minimum wages. Argentina (1918), Costa Rica (1934), Brazil (1938), Ecuador (1896--transport of kerosene and 1900--agricultural sector), Uruguay (1923--agricultural sector), and Peru (1916--indigenous peoples, 1922--maritime workers, 1937--home workers) were also early leaders in implementing some limited occupation-specific version of a minimum wage (Starr 1993). 3. Various models of household dynamics suggest that income is not shared equally among household members. Unlike the Mincer (1962) and Becker Context and Framework 17 (1976) models, where the household head (the person with the most power in the household) is altruistic and ensures an equitable distribution of household resources, these alternative models assume a cooperative or even an uncooperative bargaining game where the outcome is not necessarily Pareto Optimal. In other words, the person who has the most power in the household and makes resource allocation decisions will not necessarily distribute household income in such a way that everyone is equally well off. See Haddad, Hoddinott, and Alderman (1997) for a review of household bargaining models. 4. The minimum wage may also contribute to noneconomic well-being. For exam- ple, if an increase in the minimum wage leads to higher earnings of women, they may have more say in how household resources are spent or on the formation and dissolution of the household. The intrahousehold dynamics of minimum wages are not covered in this paper. 5. Notable exceptions include Brazil, Argentina, Uruguay, and Chile. 6. For a full exposition of the OECD literature, see Brown (1999). 7. More recent studies in the United States and France, using longer time series and more careful analysis, find mixed results, but maintain that the employment effects are small to nil. 8. No minimum wage research using data from the Caribbean was identified during the preparation of this study. C H A P T E R 3 Minimum Wage Institutions in LAC: What Are They and Who Earns Them? What Is a MinimumWage? In the simplest terms, a minimum wage is a legally mandated lower bound for wages, but the term "legally mandated" is vague, leading to many dif- ferent kinds of minimum wage institutions. In the most straightforward cases, such as Brazil or Bolivia, the federal government identifies a wage level and all employers in the country must pay at that level or above it. In other countries, such as the United States, a federal minimum wage may be increased by a state-specific minimum wage. Yet other countries, such as Italy, have a collection of wages that are negotiated by trade unions, thus blurring the distinction between a minimum wage and a contract wage. In such cases, the question may be asked whether Italy has a minimum wage at all, or whether it simply has many negotiated wages that are backed by an effective monitoring network and an able judicial system (Trinder 1984). The coverage, enforcement, and degree to which the minimum wage affects the wage distribution differ across countries. The coverage of the minimum wage is that fraction of the population for whom the policy is legally guaranteed--that is, the formal sector. A minimum wage is enforced if everyone who is covered earns at least the minimum. Finally, it is bind- ing if it actually affects wage distribution, whether through enforcement or other factors. It is completely binding if it creates a wage floor, whereas it is 19 20 Minimum Wages and Social Policy somewhat binding if it creates a distortion of the wage distribution, which may not be a wage floor. This report finds that in LAC, minimum wages cover 30­70 percent of the population (that is, the formal sector), and although they are generally not enforced, they are somewhat binding. There Is No Common"MinimumWage System"for LAC All LAC countries have a legislated minimum wage system. The com- mon goal of the systems is to set a wage floor that provides a minimum standard of living for the worker and, in some cases, his or her family. Minimum wages are usually constitutionally mandated, in some cases dating back to the 1930s, when the concept of a minimum wage was new to the world. The individual institutions that have developed in LAC over time cover a wide range of minimum wage levels, categories, and wage-setting mechanisms (see table 3.1). Minimum wage systems vary widely across the Region. Whereas Argentina, the Bahamas, Bolivia, Brazil, Haiti, and Trinidad and Tobago are the only countries in LAC with a single minimum wage, and many others have a few well- defined minimum wages (two in Colombia and Jamaica, three in Chile, and four in Belize and Peru), some countries, such as the Dominican Republic, Ecuador, Guatemala, Mexico, Paraguay, and the República Bolivariana de Venezuela have hundreds of legislated minimum wages. Special minimum wages may be set for apprentices (Colombia, Chile, El Salvador, Guatemala, Honduras, Paraguay, Trinidad and Tobago, and the República Bolivariana de Venezuela); the public sector (Argentina, Brazil, Chile, Costa Rica, the Dominican Republic, Ecuador, Honduras, Jamaica, Panama, and Peru); youth (Argentina, Belize, Chile, Costa Rica, Paraguay, Trinidad and Tobago, and the República Bolivariana de Venezuela); part-time workers (Mexico); domestic workers (Haiti, Nicaragua, Paraguay, and Uruguay); or specific occupations (see box 3.1). All countries define a minimum wage by time spent working (hour, day, or month), but some have extra legislation by task (Uruguay, the Dominican Republic), or output quantity (the Dominican Republic, Guatemala, and Jamaica). The coverage of the minimum wage may be national (Brazil, Colombia, Chile), regional (Mexico), by occupation or industry or activity (Ecuador), task, firm size, or any mix of these (table 3.1). Tripartite wage setting is associated with complex minimum wage systems. Although the government alone sets the minimum wage in some countries (Brazil, Bolivia, Ecuador, Jamaica, Uruguay), a tripartite com- mittee composed of government, worker representatives, and employer Table 3.1. Main Institutional Characteristics Affecting MinimumWages in Selected Latin American Countries Number of Work period Sub- Mw applied Sanctions minimum covered by Scope of mw Body that Frequency Criteria for minimum to public for non- Country wages the mw setting sets the mw changed adjustment payments sector compliance Notes monthly (M), national (N), government inflation (I), yes, no, or weekly (W), regional (R), (G) or needs of own rate daily (D), sector (S), tripartite (T), worker & hourly (H), occupation (O), in order of family (N), by task (T), task (T), dominance cost of by output (O) firm size (W) living (L), economic development, (D), labor market conditions (M), firm capacity to pay (C), equity (E), other/unclear/ political (O) Argentina 1 M, H N (S, O) T discretion N, E workfare own $250 to d, h programs, $1,000 per disabled worker workers, trainees, youth 21 (continued) 22 Table 3.1. Main Institutional Characteristics Affecting MinimumWages in Selected Latin American Countries (continued) Number of Work period Sub- Mw applied Sanctions minimum covered by Scope of mw Body that Frequency Criteria for minimum to public for non- Country wages the mw setting sets the mw changed adjustment payments sector compliance Notes Bolivia 1 M N G annual I, N, D no yes n.a. c, i Brazil 1 M N G annual I no own $170 per worker Bahamas 1 W, H N G, T discretion O no yes n.a. Haiti 1 D N G, T discretion L, I domestic yes n.a. workers Trinidad 1 H, D, W, M N T, G discretion L, M, D, C trainees, yes n.a. and Tobago apprentices, workfare, students, volunteers Colombia 2 D N T, G annual L, M, C, I, G apprentice yes 1 to 100 x mw Jamaica 2 W, H N, O G 1­3 years I, N casual labor own none Chile 3 M N G, T annual I. L, M, O under 18, own fines by firm size f apprentices, over 65, mentally disabled workers Belize 4 H I T, G discretion O students yes n.a. Peru 4 M N T, G discretion N, D no no fines by firm size g Cuba 6 n.a. O, S G, T n.a. D, E no yes n.a. El Salvador 8 D S, T G, T 3 years N, L, D, O apprentices yes n.a. Honduras 12 D S, W T, G 6 months P, L, G, C, I apprentices, no $30 to $300 disabled workers Nicaragua 12 M, D S T 6 months­ domestic yes 25% mw b 1 year N, D workers Uruguay 21 M, D, T N, R, S, O G annual N, D, C, I, O domestic or no 1 to 150 mw rural workers per worker Costa Rica 25 H, D, M S, O T 6 months I, L, D youth own 1 to 23 mw e Panama 30 H R, S, W G, T 2 years L, D, C, O no no $25 to $150 Mexico 91 D R, O T yearly/ N, D, C, part-time yes none under discretion I, L, M workers federal law Dom. Rep ~271 M, D, T, S, O, T T discretion N, L, M, C no own 3 to 6 mw f (by sector) H, O Ecuador ~150 M S, O G 6 months I no own 2 to 5 mw (by sector) Guatemala hundreds D, O S, I T, G annual L, N, M, C apprentices own fines/prison (by sector) (continued) 23 24 Table 3.1. Main Institutional Characteristics Affecting MinimumWages in Selected Latin American Countries (continued) Number of Work period Sub- Mw applied Sanctions minimum covered by Scope of mw Body that Frequency Criteria for minimum to public for non- Country wages the mw setting sets the mw changed adjustment payments sector compliance Notes Paraguay hundreds M, D N, O, S T, G discretion N, L, M, apprentices, no 10 to 30 mw a (by sector) C, O youth, disabled workers, domestic workers Venezuela, hundreds M R, S, I, O T, G annual P, G, L apprentices, yes n.a a R.B. de (by sector) youth Source: Interviews with labor ministries; Gonzaga and Scandiuzzi (1998); Ruiz (2001); Starr (1993); www1.umn.edu/humanrts/esc/bolivia2001.html; www.salaryexpert.com/seco/careerjournal/hrcodes/COUNTRIES.htm; www.ilo.org/public/english/dialogue/govlab/legrel/papers/brfnotes/minwages; www.mintrab.cl; www.mtps.gob.pe; www.stps.gov.mx; www.set.gov.do/legislacion/salariomin/index.htm; www.ilo.org/travaildatabase/servlet/minimumwages; www.dol.gov/ILAB/media/reports/oiea/wagestudy. n.a. data are not available. mw = minimum wage. ~ = about. a. Automatic adjustment if the inflation rate rises by 10 percent or more. b.Fines are rarely imposed, only large firms are generally inspected. c. Incomplete information since the department of labor inspection is located outside ministry and without telephones. d.Enforcement generally at regional level--no information currently available. e. The ministry does not impose fines; it only warns and takes to court. f. Codigo del trabajo Art. 44 & Art. 477: firms with 1 to 49 workers: $40 to $800 monthly per worker affected. Firms with 50 to 199 workers: $80 to $1,600 monthly per worker affected. Firms with more than 200 workers: $120 to $2,400 monthly per worker affected. g. The fines are a function of the number of workers and the number of infractions. Fines range from the equivalent of US$200 for first infraction with 1­5 workers to US$1,800 for multiple infractions involving hundreds of workers. h.Legally, 1,500 minimum wages are on the books, as a result of the bargaining under the military dictatorship. In practice, though, only one minimum wage applies. i. According to Article 121 of the Supreme Decree 21615 (29 May, 1987), the labor judges may impose a fine of $1,000­10,000 bolivianos for infractions. Minimum Wage Institutions in LAC: What Are They and Who Earns Them? 25 representatives is the wage-setting body in others (Argentina, Bolivia, Colombia, Costa Rica, the Dominican Republic, Guatemala, Honduras, Mexico, Nicaragua, Peru, Paraguay, Trinidad and Tobago, and the República Bolivariana de Venezuela).1 In several countries (the Bahamas, Chile, Cuba, El Salvador, Haiti, and Panama), the minimum wage is recommend- ed by nongovernmental groups or tripartite committees, but the final level is set by the national government. Those countries with the most complex wage structure--hundreds of minimum wages--tend to be those where the wage is set by a tripartite council.2 The minimum wage is changed every six months or one year in about half the countries, with discretionary changes in others (Argentina, the Bahamas, Belize, the Dominican Republic, Haiti, Paraguay, Peru, and Trinidad and Tobago). Adjustments to the minimum wage may be tied to inflation (in half the countries), GDP fluctuations, the poverty line, or market wages (table 3.1). In Paraguay and Haiti, the legislation states that inflation rates equal to or higher than 10 percent should trigger renego- tiation of the minimum wage. Sanctions for noncompliance are regulated in many, but not all, coun- tries. The fines imposed on firms that do not abide by minimum wage laws range from one to 150 times the minimum wage (table 3.1). No Box 3.1 Protecting theWages of Soccer Players Occupation-specific minimum wages are widespread in Latin America, partic- ularly in those countries where occupational or industry groups bargain the minimum. Although wages are set for typical low-paying occupations, others are assigned to very specific, not necessarily low-paying, occupations or to those that are difficult to enforce. For example: · Peruvian soccer players, miners, and journalists have their own minimum wage. · A wage floor for the self-employed is legislated in Chile. · Street vendors--pica pollos or chimichurras--each have their own minimum wage in the Dominican Republic. · Workers on bee or rabbit farms are specially protected in Mexico. · Guatemalan bakers and pastry workers who are not paid daily are awarded a premium above the minimum wage. Source: Minimum wage legislation for each country. 26 Minimum Wages and Social Policy fines or punishments are legislated in Jamaica, Costa Rica, or Mexico. In some countries, enforcement is legislated but not carried out because of a lack of resources--ranging from no vehicles for the inspectors to conduct inspections, to no telephones to receive complaints from workers. The scarce data on enforcement show that fines are rarely imposed. In Chile, for example, 244 fines were issued in 2001, totaling $60,000, as compared to the $2.9 million collected in fines for all labor violations that year. Who Earns the MinimumWage? Up to 20 percent of the labor force in LAC countries earns the minimum wage. Figure 3.1 shows the proportion of the labor force that earns below, equal to, and above the minimum wage in each of 19 countries, using the most recent year for which data could be obtained (appendix III describes the data).3 More than 10 percent of the labor force earns the minimum wage in Brazil, Ecuador, Panama, and the República Bolivariana de Venezuela, whereas less than 5 percent are minimum wage earners in 12 other countries. The lowest proportion is in Uruguay, with 0.5 percent of the population; the highest proportion Figure 3.1. Distribution of MinimumWage Earners Paraguay Ecuador Guyana Nicaragua Colombia Peru Venezuela, R.B. de Guatemala Panama El Salvador Costa Rica Chile Bolivia Honduras Brazil Jamaica Mexico Argentina Uruguay 0% 20% 40% 60% 80% 100% wage < 0.95*mw wage +/­ 5% from mw wage > 1.05*mw Source: Kristensen and Cunningham (2006). Note: Although several of these countries have hundreds of minimum wages (table 3.1), the most frequently repeated level or (if this was not evident) the average minimum wage was used for this exercise. Minimum Wage Institutions in LAC: What Are They and Who Earns Them? 27 is in the República Bolivariana de Venezuela and Panama, with 20 and 13 percent, respectively.4 A low proportion of minimum wage earners does not indicate that the minimum wage is unimportant: as will be shown in this report, it may have effects throughout the wage distribution. Or it may be important for certain subgroups of the population, such as young workers, as is the case in the United States and Europe. The minimum wage is not strictly binding in LAC. Figure 3.1 shows that there are subminimum workers in all countries. The size of the subminimum population is larger than the minimum wage population in all countries, ranging from 0.7 to 45 percent of the workforce. More than 40 percent of the workforce of Nicaragua, Guyana, Ecuador, and Paraguay are subminimum workers, whereas less than 3 percent earn below the minimum in Mexico, Argentina, and Uruguay. Young, informal, and less educated workers are overrepresented among the minimum wage population. Similar to the OECD countries, young workers are overrepresented among minimum and subminimum wage earners.The first column of table 3.2 shows the ratios of 16- to 19-year-olds who are minimum wage earners (relative to the whole minimum wage population) to the ratio of the working population age 16­19 (relative to the whole working population). A value greater than 1.0 indicates that the young are a larger share of the minimum wage earners than they are of the labor force as a whole. The table shows that in all but four coun- tries in the sample, those who are age 16­19 are a larger share of the minimum wage population than the general population. Of the coun- tries in the sample, the young are the most overrepresented among min- imum wage earners in Uruguay (ratio = 2.6), whereas they are the least overrepresented in Paraguay (ratio = 0.3) and Ecuador (ratio = 0.4). Young workers are even more overrepresented among the subminimum population (first column of table 3.3). Employees in the informal sector are overrepresented among minimum and subminimum wage workers. In all countries except Honduras, Panama, and Paraguay, the ratio in table 3.2 is greater than 1, with par- ticularly high ratios in Mexico (3.7), Guatemala (2.0), Chile (1.9), and Brazil (1.8). These trends are reflected in the subminimum population. Although it may be argued that this is spurious correlation, as informal sector employees tend to be low-skilled and their marginal productivity therefore may happen to coincide with the minimum wage, later sections of this report show that this is not necessarily the case. The less skilled are more represented in the minimum wage sector in most of the Region. With the exception of Costa Rica, Ecuador, Panama, 28 Minimum Wages and Social Policy Table 3.2. Ratio of Each Subgroup's Share of the MinimumWage Population, Relative to Its Share of theTotalWorkforce (Full-TimeWorkers Only) 0.95 < w/mw 16­19 55­64 No Primary Informal Self- < 1.05 years years school school employee employed Female Argentina n.a. n.a. n.a. 2.0 n.a. n.a. 1.3 Bolivia 1.6 3.0 2.1 1.9 1.0 1.5 1.0 Brazil 2.1 0.9 1.4 1.2 1.8 0.5 1.4 Chile 1.7 0.7 1.7 1.5 1.9 n.a. 1.1 Colombia 1.0 0.7 1.2 1.2 1.4 0.7 1.0 Costa Rica 1.5 1.2 0.6 1.5 n.a. n.a. 1.0 Ecuador 0.4 0.8 0.7 1.1 0.9 0.9 0.9 El Salvador 1.4 0.3 1.4 1.6 1.0 n.a. 1.1 Guatemala 1.5 1.2 1.1 0.9 2.0 0.7 1.0 Guyana 1.1 1.1 1.7 0.8 n.a. 0.4 1.1 Honduras 0.3 n.a. 1.4 n.a. 0.4 n.a. 1.3 Jamaica 1.8 1.9 6.5 1.6 1.5 3.0 1.2 Mexico 1.9 1.6 3.4 1.3 3.7 1.7 1.3 Nicaragua 1.1 1.1 1.2 0.9 n.a. n.a. 1.1 Panama 1.3 0.7 0.9 1.3 0.9 0.9 1.0 Paraguay 0.3 0.8 0.3 0.9 0.9 n.a. 0.8 Peru 1.2 0.4 0.2 1.0 1.2 0.4 0.8 Uruguay 2.6 1.0 1.7 1.4 n.a. n.a. 1.3 Venezuela, R.B. de 1.2 0.8 0.8 1.1 1.2 0.7 1.1 Source: Kristensen and Cunningham (2006). Note: Omitted categories are, respectively, age 20­54, male, formal employee, and secondary school or above. n.a. data were not available to generate the statistic. A value greater than 1 indicates that the reference group is overrepresented among the minimum wage population. Paraguay, Peru, and the República Bolivariana de Venezuela, those without any formal education are a larger share of the minimum wage population than the general working population. This trend is even stronger for the subminimum population, where they are overrepresented in all countries. The trends for primary education are similar (tables 3.2 and 3.3). Women, older workers, and the self-employed are neither more nor less likely to be minimum wage workers. Table 3.2 shows that throughout the Region, women's share of the minimum wage population is similar to their share of the working population (near the value of 1). The exceptions are Argentina, Brazil, Honduras, Mexico, and Uruguay, where the ratio is a value of 1.3 or higher. Similarly, women are not necessarily overrepresent- ed among the subminimum population, although certain outliers emerge, namely Bolivia, Honduras, Mexico, and Uruguay. However, the countries where women are overrepresented among minimum wage workers are not Minimum Wage Institutions in LAC: What Are They and Who Earns Them? 29 Table 3.3. Ratio of Each Subgroup's Share of the Subminimum Wage Population, Relative to Its Share of theTotalWorkforce (Full-TimeWorkers Only) 16­19 55­64 No Primary Informal Self- 0.95 > w/mw years years school sector sector employed Female Argentina n.a. n.a. 8.1 1.7 n.a. n.a. 1.0 Bolivia 1.9 1.6 2.6 1.3 0.8 1.9 1.8 Brazil 2.2 1.6 3.3 1.2 2.3 1.8 1.0 Chile 2.9 0.7 1.8 1.5 2.4 n.a. 1.1 Colombia 1.8 0.9 1.6 1.1 1.3 1.4 1.0 Costa Rica 2.3 1.3 3.1 1.8 0.2 n.a. 1.3 Ecuador 1.9 1.0 1.8 1.5 1.0 1.1 1.1 El Salvador 2.5 1.8 3.0 1.4 1.0 n.a. 1.0 Guatemala 1.7 2.8 2.1 1.2 1.2 2.0 1.1 Guyana 1.8 0.8 1.4 1.1 n.a. 1.0 1.3 Honduras 1.9 1.3. 1.1 n.a. 1.5 n.a. 1.5 Jamaica 1.5 1.3 n.a. 1.4 1.0 3.8 1.1 Mexico 2.0 2.0 4.7 1.4 3.4 2.4 1.8 Nicaragua 1.6 0.9 1.4 1.3 n.a. n.a. 1.1 Panama 2.8 1.1 2.4 1.7 1.7 1.3 0.7 Paraguay 1.9 0.8 1.9 1.6 1.5 n.a. 1.1 Peru 2.4 0.9 2.6 1.4 2.0 1.1 1.2 Uruguay 5.0 0.8 3.4 1.5 n.a. n.a. 1.5 Venezuela, R.B. de 1.9 1.2 2.1 1.2 1.7 1.4 1.2 Source: Kristensen and Cunningham (2006). Note: Omitted categories are, respectively, age 20­54, male, formal employee, and secondary school and above. n.a. the data are not available to calculate the statistic. A value greater than 1 indicates that the reference group is overrepresented among the subminimum wage population. necessarily those where they are overrepresented among subminimum wage workers. A note of caution is necessary, however, since the table considers only full-time workers; expanding the sample to include part- time employees, among which women are overrepresented, may lead to a different conclusion. Across the Region, older workers (age 55­64 years old) are not uniformly over- or underrepresented among minimum wage earners. Although age-earning profiles tend to show that real wages for workers begin to decline around age 50, they do not necessarily revert to the minimum wage. Older workers are a lower share of the minimum wage population than the entire working population in countries as diverse as Brazil, Chile, Colombia, Ecuador, Panama, Paraguay, Peru, and the República Bolivariana de Venezuela. Conversely, in the equally diverse countries of Bolivia, Costa Rica, Guatemala, Jamaica, Mexico, and 30 Minimum Wages and Social Policy Nicaragua, workers age 55­64 are a higher share of the minimum wage population than the general population. The self-employed are largely underrepresented among minimum wage earners. In the small sample where the self-employed could be iden- tified, they were overrepresented among the minimum wage earners only in Bolivia, Jamaica, and Mexico, with ratios between 0.14 and 0.9 for the other eight countries in the sample. Alternatively, the self-employed are overrepresented among the subminimum population in all countries in the sample. This may suggest a greater relative number of workers with very low incomes in the self-employment sector, with a thinning of the population near the minimum wage. Or it may suggest a smooth distribution for the self-employed whose incomes we would not expect to be affected by the minimum wage, as opposed to a "sweeping up" of low-income employees whose incomes are near the mandated minimum. Groups that are overrepresented among minimum wage earners may still have low proportions who actually earn at or below the minimum wage. Although tables 3.2 and 3.3 show how over- or underrepresented various demographic groups are relative to their share of the labor force, they do not reveal what proportion of that group earns the minimum wage. Table 3.4 shows the proportion of workers within each demographic group who earn at or below the minimum wage, which suggests that even if individuals are overrepresented among the minimum wage population, they may not be overwhelmingly minimum wage workers. For example, young workers are overrepresented among minimum wage workers in Mexico, but only 25.6 percent of the young earn the minimum wage. A similar situation prevails for informal sector workers, where less than 30 percent of them earn less than the minimum wage. Table 3.4. Proportion that Earns at or below the MinimumWage, by Demographic Characteristic Argentina Brazil Colombia Mexico (1999) (1996­2000) (1984­2001) (1999) Young (age < 19) 70.0 40.0 63.0 25.6 Female 19.9 21.8 25.7 13.2 Informal sector 31.7 23.0 n.a. 16.9 Primary school or less 24.6 26.0 55.0 11.9 Household wealth quintile 1 24.5 n.a. 75.0 14.6 Household wealth quintile 5 5.6 n.a. 7.7 4.1 Source: Argentina and Mexico (Cunningham 2002), Brazil (Neumark, Cunningham, and Siga 2003), Colombia (Arango and Pachon 2003). n.a. the value was not reported. Minimum Wage Institutions in LAC: What Are They and Who Earns Them? 31 Wages of workers whose marginal productivity is near the minimum wage may converge to the minimum wage. Uruguay, Jamaica, Brazil, Argentina, Mexico, and Bolivia have the lowest ratio of minimum wage workers and also have more women, low-educated, and youth at the minimum than expected. Thus, even though the minimum wage popu- lation is small in these countries, it disproportionately affects certain population groups--the groups that are often the hardest to employ, whether because of time constraints (school, family care) that limit job choice, frequent entry into and exit from the labor market, or lower human capital because of less job experience or lower education levels. Conversely, in countries where a high share of the population consists of minimum or subminimum wage earners--namely Paraguay and Ecuador--women, unskilled, and the young are underrepresented among minimum wage earners. Thus, given the distribution of productivity and corresponding wages, the minimum wage may have the effect of sweeping up those whose productivity (and market wage) are near the minimum, but may not affect the wages of those whose market wage is far from the mandated minimum. How High Is the MinimumWage in LAC Minimum wage levels vary widely across Latin America. Table 3.5 shows the level of the minimum wage for 20 countries in the Region, in national currency and in PPP-adjusted U.S. dollars. The minimum wage is the most generous in Paraguay, at more than US$500 monthly (PPP-adjusted), and the least generous in Uruguay, at PPP-adjusted US$45 monthly. Although Paraguay has the highest proportion of individuals at or below the mini- mum and Uruguay the lowest, the absence of correlation between these two factors for the other 18 countries in the sample suggests that the pro- portion of minimum wage earners is largely independent of the PPP- adjusted level of the minimum wage. The minimum wage is far below the mean wage in all countries in the Region, but this does not imply that the minimum wage is not well set. Figure 3.2 shows that the ratio of minimum to mean wages ranges from 18 percent of the mean wage (Uruguay) to 72 percent of the mean wage (Paraguay). Half of the sample countries have ratios between 20 and 40 percent of the respective mean wages. Three of the four countries with the highest ratio also have the largest share of their population earning the minimum wage; Brazil is the outlier, with a large share earning the minimum but a low wage ratio. Similarly, the country ranking of the 32 Minimum Wages and Social Policy Table 3.5. MinimumWages in LAC Monthly mw, Mw in PPP- Country Year domestic currency adjusted US$ Paraguay 2000 680168 $546 El Salvador 1998 1,083 $446 Dominican Republic 1997 2,412 $429 Costa Rica 1999 54,938 $423 Chile 2001 100,000 $335 Colombia 1999 236,438 $293 Guatemala 2000 712 $289 Honduras 1999 1419 $221 Brazil 1999 136 $205 Argentina 2000 200 $200 Panama 1998 206 $185 Guyana 1999 19,000 $181 Peru 1999 345 $124 Bolivia 1999 330 $90 Jamaica 1998 800 $75 Ecuador 1998 762,967 $75 Venezuela, R.B. de 1998 100,000 $54 Mexico 1999 888.81 $50 Uruguay 1998 990 $45 Nicaragua 2001 1,000 n.a. Source: Kristensen and Cunningham (2006). n.a. no data were available. share of the population earning below the minimum wage (figure 3.1) is similar to the country ranking of the minimum wage to mean wage ratio (figure 3.2)--that is, countries with higher relative minimum wages are also those with more people earning less than the minimum. Perhaps a more appropriate relative measure is the median wage, as it omits very high earners. Maloney and Nuñez (2004) find that the ratio of the minimum wage to the median wage ranges from 27 percent (in Uruguay) to 69 percent (in Colombia) (table 3.6), compared to the ratio of the minimum wage to the mean wage, which is 18 percent and 52 percent, respectively. The minimum wage is similar to the average wage of unskilled work- ers in most countries. To control for the skill differentials in the compari- son of minimum wages to market wages, a better comparator group may be "unskilled workers," who have productivity levels more similar to workers at the bottom of the wage distribution. The ratio of minimum wages to mean wages of the unskilled ranges from 0.21 to 1.4, indicating Minimum Wage Institutions in LAC: What Are They and Who Earns Them? 33 Figure 3.2. MinimumWage Relative to the Mean or UnskilledWage Paraguay Ecuador Guyana Venezuela, R.B. de Colombia Nicaragua Panama El Salvador Costa Rica Chile Peru Guatemala Honduras Bolivia Mexico Argentina Brazil Jamaica Uruguay 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 mw/mean mw/mean low-skilled Source: Kristensen and Cunningham (2006). Table 3.6. Ratio of the MinimumWage to the Median and 10th PercentileWages, % Mw/median Mw/10th percentile Argentina 33 67 Brazil 43 100 Bolivia 34 80 Chile 55 109 Colombia 68 100 Mexico 48 87 Uruguay 27 64 Source: Maloney and Nuñez (2004). that the minimum wage is 40 percent higher than the average wage of unskilled workers in Paraguay and 89 percent below the similar compara- tor group in Uruguay (figure 3.2). Ten of the 18 sample countries show that minimum wages are greater than 70 percent of the average unskilled wage. A similar exercise by Maloney and Nuñez (2004) uses the 10th per- centile wage as the base (table 3.6); they show an even higher correlation between the minimum-wage and low-wage workers. A standard established by the Brazilian Institute for Applied Economics (Instituto de Pesquisa Economica Aplicada, IPEA) is that the international norm is a minimum wage that is at least 40 percent of the average manufacturing wage. This conclusion was reached by examining 34 Minimum Wages and Social Policy the minimum wage to manufacturing wage ratio in countries across the world and observing a clustering at 40 percent (IPEA 2002). The con- clusion is complicated by the fact that the productivity of manufactur- ing workers is likely to be much higher than that of unskilled workers in other sectors. Thus, the conclusion may be the result of productivity differences between the manufacturing and low-skilled workers and is therefore determined by the market. If, indeed, low-skilled workers are more than 60 percent less productive than the manufacturing sector, then this 40 percent may reflect a global sense of fairness. However, it would be useful to identify a comparator group in which the skill level is similar between the types of workers who earn the minimum wage and the comparator group, as shown above. Notes 1. In Colombia, for example, workers and employer representatives select the minimum wage. If they cannot agree on a wage, the government sets the wage. 2. If unions and employers define the minimum wage, it is preferred to have multiple minimum wages so that the agreed-on union wage is not imposed on the whole country in the form of a single minimum wage, which may be too high for nonunion employment. 3. A wage is considered "equal to" the minimum wage in this exercise if it is within 5 percent of the minimum wage. A range rather than a single value of the minimum wage is used to compensate for measurement errors in the data: wages are reported in the surveys as monthly wages, but minimum wages are often reported hourly, daily, or weekly. Thus, aggregating the "minimum wage" to a monthly comparison requires assumptions about the hours worked per day and days worked per week. 4. Since the minimum wage is a "round number," equal to 100,000 in Chile and the República Bolivariana de Venezuela, the apparent exact payment of the minimum may actually be due to a convention to pay round numbers or due to reporting bias that rounds the wage. C H A P T E R 4 The Worker: How Do Minimum Wages Affect Other Wages and Employment? The usefulness of the minimum wage as a social protection tool depends not only on its ability to redistribute income, but also on how well it does this compared to alternative tools. Ultimately, we are interested in how the minimum wage affects poverty and inequality, but this section focuses on the worker to give us the building blocks for the next chapter, which con- siders the household. It allows us to identify who is distributing income to whom, and whether this meets the government's and society's objectives. MinimumWages Are Somewhat Binding An examination of wage distributions can give us some insight into whether or not the minimum wage is enforced and binding. If workers are paid a wage equal to their marginal productivity, and if the latter is continuously distributed across the population, we would expect a smooth wage distribution. However, if there is an exogenous factor such as a minimum wage that affects wage setting, the distribution will not be smooth. Instead, we will see a "spike" in the distribution. To be sure that our "spikes" are not statistical artifacts, we also plot the number of people earning up to a specific wage (cumulative density plot) and examine 35 36 Minimum Wages and Social Policy whether a "cliff" is visible at the wage corresponding to the minimum wage. If the wage level at the spike and at the cliff for a particular country coincides with the value of the minimum, we can then assume that the minimum wage is to blame for the discontinuity in the wage distribution: it is binding.1 If everyone earns above the minimum and it is binding, this suggests that the minimum is enforced. The minimum wage is not enforced in Latin America. In all countries in the Region, there are individuals in both the formal and informal sectors who earn below the minimum wage.2 Appendix IV plots the wages of formal and informal sector workers and indicates where the minimum wage falls in each distribution (the vertical line). Very few individuals are subminimum earners in Argentina, Jamaica, Mexico, and Uruguay, which are also the countries with the lowest PPP-adjusted minimum wage. Although this may suggest that the minimum is well enforced in those countries, the absence of a "spike" at the vertical line (left graph in each row of appendix IV), which would graphically show that those who would be subminimum workers are pulled up to the minimum wage, suggests that it does not serve as a wage floor. Furthermore, it is unlikely to shift the whole distribution to the right, as the low ratio of minimum to unskilled wages shown in figure 3.2 suggests that the minimum wage is below market wages in these coun- tries, and not that it successfully creates a wage floor. This is supported by the information presented in table 3.1, in which low sanctions or inef- ficient monitoring systems suggest that enforcement is very low. The minimum wage is somewhat binding, particularly among those in the informal sector. If skills were continuously distributed across the labor force and institutions that affect the wage distribution were absent, the graphs in appendix IV would be smooth. While the graphs are smooth in some cases, in others there is a spike at the minimum wage, suggesting that the minimum wage affects the wage distribution in these countries. Table 4.1 summarizes which country-sectors show spikes at the minimum wage and which do not. The formal sector graphs have spikes in Brazil, Chile, Colombia, Ecuador, Nicaragua, Panama, Paraguay, Peru, and the República Bolivariana de Venezuela (left graph on each axis in appendix IV, top left cell in table 4.1), where there is a clustering of wages around the minimum wage (vertical line) suggesting that the minimum wage alters the formal sector wage distribution for some workers. It is not fully binding, as shown by the many observations to the left of the vertical line; those are the subminimum workers.3 Given the few enforcement mech- anisms in LAC, the effects of the minimum wage are likely due to other factors, including union pressures, efficiency wages, or a sense of fairness. The Worker: How Do Minimum Wages Affect Other Wages and Employment? 37 Table 4.1. Degree toWhich the MinimumWage Is Binding inWage Employment in LAC andYear of Analysis, by Sector Mw is somewhat binding Mw is not binding (spike at the minimum wage) (smooth wage distribution) Formal wage sector Brazil (1999) Argentina (2000) Chile (2001) Bolivia (1999)a Colombia (1999) Costa Rica (1999) Ecuador (1998) El Salvador (1998)a Guyana (1999)b Guatemala (2000)a Nicaragua (2001) Honduras (1999)a Panama (1998) Jamaica (1998)a Paraguay (2000) Mexico (1999) Peru (1999) Uruguay (1998) Venezuela, R.B. de (1998) Informal wage sector Brazil (1999) Argentina (2000) Chile (2001) Bolivia (1999) Colombia (1999) Costa Rica (1999) Ecuador (1998) Dominican Republic (1997)a El Salvador (1998) Guatemala (2000) Mexico (1999) Honduras (1999) Nicaragua (2001) Jamaica (1998) Panama (1998) Uruguay (1998)a Paraguay (2000) Peru (1999) Venezuela, R.B. de (1998) Source: Kristensen and Cunningham (2006). a. The kernel density wage distributions show a spike, but the cdf (right graph) does not show a cliff, suggesting that the minimum wage"impacts"we see in the kernel density may be a statistical artifact. b. Wage workers could not be identified as formal sector or informal sector. Contrary to the assumption that minimum wage legislation is relevant only to the formal sector, the left graph of each figure in appendix II shows that it affects the informal wage distributions in more countries than it affects the formal sector wage distributions. In Brazil, Chile, Colombia, Ecuador, El Salvador, Mexico, Nicaragua, Panama, Paraguay, Peru, and the República Bolivariana de Venezuela, spikes are evident at the minimum wage in the informal sector wage distribution (left graph on each axis, bottom left cell in table 4.1). The fact that minimum wages are somewhat binding in the informal sector may be due to various factors. From the labor supply side, the mini- mum may be a benchmark for "fair" wages. Foguel (1997) argues that workers value not only their absolute wage, but also the wage relative to others of similar skill level. Whereas the exact value of a comparator wage in the formal sector may not be measured easily, the value in terms of 38 Minimum Wages and Social Policy minimum wages may be estimated. On the demand side, employers may pay the number of minimum wages comparable to the formal sector mar- ket wage for a particular occupation so that their employees will not leave for a similar job in the formal sector, a kind of efficiency wage. Or the results may be a statistical artifact: as informal sector wages are lower than formal sector wages, the presence of more individuals in the lower part of the distribution--where the minimum wage tends to be--may lead to more of a "piling up" around the minimum wage in the informal sector than in the formal sector. Thus, minimum wages may be equally binding in both sectors, but the lower wages of informal sector workers may give an appearance of being more binding. The empirical evidence presented later in this chapter shows that this last explanation is not the case. An increase in the minimum wage has a positive effect on formal and informal sector wages. Turning from graphs to more rigorous analysis, table 4.2 summarizes the findings of key studies that estimate the effects of a minimum wage increase on average wages. Appendix II presents a more detailed version of the table, including information on the methodology used by the authors, the data, the time period, and the estimated elastici- ties. Regardless of the methodology used in the study, the time period studied, or the country of study, a 10 percent increase in the minimum wage increases the mean wage by 1­6 percent. The increase is particularly strong for workers who earn near the minimum, similar to the findings in the OECD. Notably, the positive effects are detected for formal salaried, informal salaried, and self-employed workers across the region.4 MinimumWage Policies IncreaseWages throughout theWage Distribution Numeraire effects are observed in several countries in the Region. Spikes are detected at multiples of the minimum wage in various countries, sug- gesting that the minimum wage may be used as a benchmark for other wages. This was noted by Neri, Gonzaga, and Camargo (2000) for Brazil, where numeraire effects were shown up to six times the minimum wage for Brazil. Testing in additional countries shows that the phenomenon is more widespread than just Brazil. For example, figure 4.1 shows that the wage distribution has cliffs--as highlighted by the vertical lines--at 1.0, 1.5, 2.0, 3.0, and 4.0 times the minimum wage in Jamaica, and the same for Mexico.5 Rigorous analysis confirms the numeraire observations in the graphs. Whereas the OECD literature tells us that minimum wages will affect only The Worker: How Do Minimum Wages Affect Other Wages and Employment? 39 Table 4.2. Summary of Key Empirical Literature on the Effect of the MinimumWage onWages in Latin America Country and MW effect on years analyzed average wagesa Note Source Brazil 1984­99 +b effects throughout the Fajnzylber (2002) wage distribution Brazil 1980­93 +, 0c first half of period, Carneiro and second half of period Faria (1997) Brazil 1983­99 +b bigger effect in the Foguel, Ramos, informal than formal Carneiro (2001) Brazil 1984­2000 + throughout the wage Lemos (2002) distribution Brazil 1995­2001 + only those clustered Neumark, near the mw Cunningham, and Siga (2006) Brazil 1996, 1999 + only for poor near mw Soares (2002) Colombia 1997­2000 +b throughout the Maloney and wage distribution Nuñez (2004) Colombia 1962­92 + manufacturing sector only Bell (1997) Colombia 1984­2001 0 lowest 40% of population, Arango and +c 40th­60th percentile Pachón (2003) of population Costa Rica 1980­96 + formal sector, Gindling and Terrell ­ small firm, (2005) ­ self-employed, 0c part-time Mexico 1988­98 +b throughout the wage Cunningham and distribution Siga (2006) Mexico 1985­2001 + formal sector only Castellanos, Garcia- Verdu, & Kaplan (2004) Mexico 1972­90 0 manufacturing sector only Bell (1997) Note: A description of the methodology and the elasticity estimates is given in appendix II. a. A positive sign indicates a positive effect of a minimum wage increase on other wages, a negative sign indicates a negative effect of a minimum wage increase on other wages, and a 0 indicates that an increase in the minimum wage does not have any identifiable effect on average wages. b. Effects for the formal and informal sectors, estimated separately. c. Each sign corresponds to the group listed in the"Note"column, respectively. the wages of those who earned at or below it (Brown 1999), in Latin America, they also affect wages above the minimum. Figure 4.2 gives the estimated percent increase in the hourly wages of Brazilian, Colombian, and Mexican workers in various parts of the wage distribution. An increase in the minimum wage has a positive effect on the wages of workers up to many multiples of the minimum wage. In Brazil, any increase in the 40 Minimum Wages and Social Policy Figure 4.1. Cumulative Density Functions Showing"Spikes"at Multiples of the MinimumWage Jamaica Mexico formal sector informal sector formal sector informal 1 1 sector 0.5 0.5 0 0 0 5 10 15 4 6 8 10 12 in(monthly wage) in(monthly wage) Jamaica, 1998, monthly Mexico, 1999, monthly Source: Kristensen and Cunningham (2006). Note: The vertical lines for Jamaica are set at 1, 1.5, 2, 3 and 4 times the minimum wage. The vertical lines for Mexico are set at 1, 1.25, 1.5, 2, 2.5, and 3 times the minimum wage. minimum wage is perfectly reflected in the wages of very poor Brazilian workers (earning one half to one minimum wage), with a smaller increase for those earning at the old minimum wage and yet smaller increases up to those earning six­nine minimum wages, until no effects are felt for those earning more than 9 minimum wages. The benefits of the minimum wage increase in Mexico run out at seven times the minimum wage (which is a much smaller monetary value than that in Brazil) whereas the only benefits in Colombia are experienced by workers at the 45th­60th wage centiles, who earn two­three minimum wages. Similar to the argument for the reasons that minimum wage effects are observed in the informal sector, wage effects above the minimum may be due to the numeraire effect. Because of the history of the minimum wage as a price index during hyperinflationary periods in many countries, it may still be used to guide wage values throughout the wage distribution. Or it may be due to employers trying to maintain some level of fairness so as to maintain productivity levels among workers who earn above the minimum wage. Card and Krueger (1995), who found a similar result for low-wage workers in the United States, argue that the relative wages in a firm are important. If an increase in the minimum wage "sweeps up" those who had earned below the new minimum wage to levels earned by those who were "above minimum wage earners" before the increase, employers will increase the wages of the former "above minimum wage earners" a bit to maintain a gap between these workers and minimum The Worker: How Do Minimum Wages Affect Other Wages and Employment? 41 Figure 4.2. AverageWage Increase due to a 1% Increase in the MinimumWage, by Position in theWage Distribution before theWage Change (Full-TimeWorkers), % (a) Brazil 1 0.8 0.6 0.4 0.2 0 ­0.9 ­1.1 ­2.0 .5 .0 ­3.5 ­4.0 12 4­5 5­6 6­9 ­40 0.5 0.9 1.1­1.5 1.5 2.0­2 2.5­3 9­ 3.0 3.5 12 wage, in terms of minimum wages before increase (b) Mexico 2.5 2 1.5 1 0.5 0 .95 .2 0 ­1.15 ­1.5 ­1.8 ­2.5 ­2.8 ­3.1 3.5 .5 ­5.5 ­7 7­10 0.2­0 1.5 1.8­2 3.5­4 4­4 5.5 2.2 2.5 2.8 3.1­ 4.5 10­4 0.95 1.15 wage, before mw increase, stated in terms of the minimum wage (c) Colombia 1.2 1 0.8 0.6 0.4 0.2 0 ­0.2 10 15 20 25 30 35 40 45 50 60 70 80 90 ­0.4 by wage centile Source: Cunningham and Siga (2006) for Mexico and Brazil; Arango (2004) for Colombia. Note: A colored bar indicates the coefficient is statistically significant at the 5% level, a white bar indicates that it is not. wage workers, thus encouraging them to continue exerting a level of effort above the minimum effort. The Level of the MinimumWage DeterminesWhether It Increases or DecreasesWage Inequality An increase in the minimum wage reduces wage inequality in countries with a low relative minimum wage. The downward-sloping curves of figures 4.2 (a) and (b) show that the lowest-paid workers have the greatest increase in wages when the minimum changes. Low-paid but above-the- minimum workers see their wages increase by a lesser amount, and work- ers at the top of the distribution do not see any change in their wages. 42 Minimum Wages and Social Policy This results in compression of the wage distribution. Direct tests of the change in wage inequality using Brazilian (Lemos 2002, Angel-Urdinola and Wodon 2003, Soares 2002, Rodrigues and Menezes 2003), Colombian (Bell 1997), Argentine and Mexican (Rodrigues and Menezes 2003), and U.S. data (Neumark, Schweitzer, and Wascher 2000) show that wage inequality decreases with an increase in the min- imum wage.6 However, this cannot be generalized worldwide as no effect is found in the case of the United Kingdom, which has low rela- tive minimum wages (Machin and Manning 1994). However, the minimum wage can exacerbate wage inequality in high- minimum-wage countries. In Colombia, where the relative minimum wage is high, the benefits of the minimum wage skip over low-wage workers and raise the wages of those in the middle of the wage distribu- tion. An increase in the minimum wage increases wages of those in the 45th to 60th wage deciles, with no effects on the lower or upper parts of the wage distribution. This implies that higher minimum wages increase wage inequality between the poor and middle-income workers (Arango and Pachón 2003). Wage Benefits Are Not Concentrated on Any Particular Group of Workers In Brazil and Mexico, youth and older adults, men and women, primary and secondary school, and formal and informal workers all benefit from wage increases. In Mexico, where the more vulnerable groups benefit from minimum wage increases, the "less vulnerable," such as men, formal sector workers, and the most educated, experience even larger wage gains than female, informal sector, or less educated workers. In Brazil, however, a clear pattern does not emerge. Some groups gain and others lose, but these gains continue up the wage distribution for all groups (Cunningham and Siga 2006). An increase in the minimum wage causes greater gains for men's wages than for women's wages. In Mexico and Brazil, both women and men benefit from a higher minimum wage, but the gains are greater for men throughout the wage distribution. Although the distribution of women's wages shows sharper spikes than men's wages (Soares 2001, and Rodrigues and Filho 2003), indicating that more women are clus- tered around the minimum wage, women's wages do not necessarily respond to a change in the minimum wage to a greater degree than men's wages. Figure 4.3 shows that except for the lowest earners, men's The Worker: How Do Minimum Wages Affect Other Wages and Employment? 43 Figure 4.3. Mexico, AverageWage Increase due to a 1% Increase in the Minimum Wage, % 4 3.5 in e eg 3 wa 2.5 hangc eg 2 1.5 cent veraa 1 per 0.5 0 15 5 3.5­4 4­4. 5.5­7 7­10 0.2­0.950.95­1. 1.15­1.51.5­1.8 1.8­2.2 2.2­2.5 2.5­2.8 2.8­3.1 3.1­3.5 4.5­5.5 10­40 wage, in terms of the minimum wage, before the 1% increase Source: Cunningham and Siga (2006). Note: The percent increase for men is given in the left bar and for women in the right bar. A solid bar indicates the coefficient is statistically significant at the 5% level; a white bar indicates that it is not. gains exceed women's at all wage levels in Mexico. Thus, while women are overrepresented among minimum-wage earners, their wages are less responsive to increases in the minimum. A 10 percent increase in the Mexican minimum wage led to a 10­36 percent increase in men's wages, but was responsible for only a 0­10 percent increase in women's wages (figure 4.3).7 The minimum wage premium to men also appears in Brazil, where men enjoy a 0­14 percent increase in wages from a 10 per- cent increase in the minimum, whereas women experience gains equi- valent to 0­6.5 percent (Cunningham and Siga 2006). As in the OECD countries, young workers with low wages gain from an increase in the level of the minimum. However, unlike the OECD, adults whose wages are clustered around the minimum make even more significant wage gains. A 10 percent increase in the minimum wage in Mexico results in a 10 percent increase in the wages of youth who earn the minimum wage, but a 23 percent increase in the wages of adults in the same initial earnings groups. The differences between the age groups are much smaller in Brazil. Also, unlike OECD countries, wage gains are experienced by youth (and adults) with initial earnings that are above the minimum wage. Workers with only a primary or secondary education benefit from higher minimum wages. In both Mexico and Brazil, a 10 percent increase in the minimum wage leads to an increase in wages as high as 7 percent in Brazil and 23 percent in Mexico for low- to medium-skilled workers. Brazilian workers with no formal schooling (not estimated for Mexico) do not benefit. 44 Minimum Wages and Social Policy Table 4.3. Change in theWage-Gini Coefficient (by Demographic Group) due to a Decrease in the MinimumWage, 1988­99 Brazil Mexico (mw = ­8.6%) (mw = ­37.7%) % of change in % of change in Gini due to Gini explained Gini due to Gini explained change in the by the change in change in the by the change minimum the minimum minimum in the minimum wage wage wage wage Men 0.02 10.7 0.02 29.3 Women 0.01 5.6 0.05 59 Unskilled 0.02 9.3 0.05 51.9 Skilled 0.01 7.9 0.03 40 Young 0.00 0 0.04 58 Adults 0.02 13.7 0.03 47.4 Formal ­0.01 4.8 0.01 7 Informal ­0.03 18.4 0.05 37.3 Source: Souza Rodrigues and Menezes Filho (2003). Note: Assume no unemployment effects and no numeraire effects. The decrease in the real minimum wage across the Region since the 1980s is partly responsible for the increased dispersion in the wage distribution of women and youth. The findings of this section are sup- ported by another exercise that simulates the change in the Gini coeffi- cient (a measure of inequality) owing to a decrease in the real value of the minimum wage in the 1990s.8 Table 4.3 shows that when the mini- mum wage decreases, the dispersion of wages increases 5­59 percent among men, women, unskilled, skilled, and adults in both Mexico and Brazil. The effects are particularly strong for the wage distributions of women, the unskilled, and the informal sector. Thus, increased minimum wages would decrease the inequality in these countries, as suggested by figure 4.2. Unfortunately, the "high relative wage" case of Colombia was not estimated. MinimumWages Increase Unemployment An increase in the minimum wages causes job loss. Throughout the Region, and regardless of year or methodology used for the analysis, an increase in the minimum wage results in higher unemployment and lower employment (table 4.4). The effects may be very small (Lemos (2002) for Brazil) or substantial (Bell (1997) for Colombia), with most countries experiencing a job loss of 2 percent for a 10 percent increase in the mini- mum wage. As with the wage impacts, job loss occurs most among those The Worker: How Do Minimum Wages Affect Other Wages and Employment? 45 Table 4.4. Summary of Key Literature on the Effect of the MinimumWage on Unemployment and Employment in Latin America Country Effecta Note Source Unemployment Brazil 1982­87 +b Foguel (1997) Brazil 1982­99 +b especially in the Fajnzylber (2002) informal sector Brazil 1996­2001 +b informal wage sector only Cunningham and Siga (2006) Chile 1957­96 + and increases the duration Montenegro (2003) of unemployment Colombia +b throughout the wage distribution; Maloney and Nuñez 1997­99 more in formal than s.e. sector (2004) Mexico 1983­2000 0b positive if time and Garza Cantu and state controls not included Bazaldua (2001) Mexico 1988­98 0 no strong effects in the formal or Cunningham and Siga informal (2006) Employment Brazil 1982­87 ­b decline in number employed and Foguel (1997) increase in labor market exit Brazil 1982­2000 +b small effects Lemos (2002) Brazil 1995, 1999 ­, +c formal, informal Carneiro and Corseuil (2001) Brazil 1982­2001 ­, +c formal, informal Foguel, Ramos, and Carneiro (2001); Carneiro (2000) Brazil 1996­2001 ­, +c household heads, household Neumark, Cunningham, dependents and Siga (2006) Colombia 1980­87 + manufacturing employment only Bell (1997) Colombia 1984­2001 ­, +c household heads, dependents Arango and Pachón (2003) Costa Rica 1980­96 +, ­c full-time workers, part-time Gindling and Terrell workers (2005) Mexico 1984­90 0 manufacturing employment only Bell (1997) Source: Cited in last column of table. Note: For full details, see appendix II. a. A positive sign indicates a positive effect of a minimum wage increase on employment and unemployment, a negative sign indicates a negative effect of a minimum wage increase on employment and unemployment, and a zero indicates that an increase in the minimum wage does not have any identifiable effect on employment and unemployment. b. Effects for the formal and informal sectors, estimated separately. c. Each sign corresponds to the group listed in the"Note"column, respectively. who earned near the old minimum wage, but is felt further up the wage distribution as well (Cunningham and Siga (2006) for Brazil; Maloney and Nuñez (2004), Arango and Pachón (2003) for Colombia). The only country analyzed for which no job loss was detected was Mexico, where 46 Minimum Wages and Social Policy the authors argue that the minimum wage is too low to have any impact on the labor market (Bell 1997). The minimum wage causes unemployment among formal sector workers, but the evidence is mixed for the informal sector.9 In Brazil and Colombia, an increase in the minimum wage corresponds to a lower likeli- hood of retaining one's job in the formal sector. The implications for informal sector employment are not as clear (table 4.4). Some Colombian and Brazilian literature find the expected disemployment effects, since the minimum wage appears to be binding in the informal sector, but other studies in Brazil show an employment effect in the infor- mal sector (Carneiro and Corseuil 2001; Corseuil and Morgado 2000; Foguel, Ramos, and Carneiro 2001; Carneiro 2000). These authors argue that higher minimum wages force formal sector workers into informal sector jobs. No employment effects in either sector were detected in Mexico, with its low minimum wage, even though the minimum wage is somewhat binding. An increase in the minimum wage leads to greater job loss among women, youth, and low-skilled workers. 10 Although wage gains are shared by most in the population, with greater gains among the less skilled, job loss is less equally shared. In Brazil (Cunningham and Siga 2006), an increase in the minimum wage leads to greater job loss for women, youth, and low-skilled workers whose wages are clustered around the minimum. No job loss is experienced by men, prime-aged or older workers, and the high-skilled. In Mexico, a pattern is not discernible, which is not surprising given the lack of evidence of job loss in the aggregate. Caution should be taken in applying these results to all countries in LAC, since the findings are for only two countries, one with a low rela- tive minimum wage but strong numeraire effects (Mexico) and the other for a country with a moderate relative minimum wage and some numeraire effects (Brazil). The extent to which the minimum wage sup- ports certain segments of the labor force depends on how high the mini- mum wage is relative to market wages of various demographic groups. For example, in countries with a very low minimum wage relative to the wage of low-skilled workers, minimum wages are very binding for those with primary school or less, as shown in chapter 3. The minimum wage may be too low to be meaningful for more highly skilled workers. However, in economies with a relatively high minimum wage, such as Paraguay and Ecuador, the wage distribution of the more highly skilled is most affected; the wage distribution of workers with a secondary edu- cation is affected more than that of workers with primary school or less. The Worker: How Do Minimum Wages Affect Other Wages and Employment? 47 Subnational MinimumWages Are Standard Practice A national minimum wage is generally not feasible because of regional price variation. Although national minimum wages may officially exist in most countries, heterogeneity in labor markets across a country creates multiple real regional minimum wages. For example, even though Brazil eliminated its regional nominal minimum wages in 1984, figure 4.4 shows that the real minimum wage variation by region still exists. The plots of the wage distribution in the different regions of Brazil in 1999 clearly show that the national minimum wage is too low to be relevant in the wealthy state of São Paulo, whereas it appears to play a role in the poorer state of Bahia. In real terms, the minimum wage, although identical in nominal terms, is, in fact, regionally specific. Figure 4.4. Kernal Density and Cumulative Density Plots ofWages, by Region region with high mean wage region with low mean wage Brazil 1 1 probability probability e 0.5 e 0.5 ulativ ulativ cum cum 0 0 2 4 6 8 10 2 4 6 8 10 in(monthly wage), in(monthly wage), Bahia São Paulo Mexico 1 1 0.5 0.5 0 0 4 6 8 10 12 5 10 15 in(monthly nominal wage), in(monthly nominal wage), region 1, 298, monthly region 3, 298, monthly Source: Author's calculations from PNAD (2001; Brazil) and ENEU (1998; second semester, Mexico). Note: Vertical line is minimum wage. Mexico: left-hand curve is informal sector; right-hand curve is formal sector. 48 Minimum Wages and Social Policy Mexico has multiple regional-specific minimum wages that allow for regional variation in labor markets. Although the minimum wage is low across Mexico, the minimumå wage in region 1, if applied to region 3, would have much bigger implications for employment and wages in region 3 than does the lower minimum wage applied to region 3. By setting different nominal minimum wages for regions, a national "real" minimum wage exists, so that the impacts are similar across the country (figure 4.4). One of the often voiced concerns about regional minimum wages is that they give incentives for migration from low to high nominal minimum wage regions. A counterexample is the United States, in which each state chooses to use the federal minimum wage or to set a higher level, resulting in a high variance in minimum wages across the country. The state-level minimum wages have not been shown to create an incentive for migration from low- into high-minimum wage states, because the real minimum wage varies less than the nominal wage (Brown 1999). Notes 1. The methodology presented by DiNardo, Fortin, and Lemieux (1996) and Maloney and Nuñez (2004) is used rather than the probability density function (pdf). Kernel density plots are estimated for each country-sector combination. They are basically a continuous version of discrete histograms; that is, they smooth a line between each observation xi along the x-axis or group of observations within a certain bandwidth of the x-axis to obtain an estimated density. We also present the cumulative density plots (cdf) as they do not require any assumptions about bandwidths, thus giving an alternative illus- tration of the wage distribution. For the cdf, a "cliff" would correspond to the "spike" in the kernel density plot. For full discussion of the methodology, see Kristensen and Cunningham (2006). 2. To control for measurement errors introduced by part-time workers in coun- tries where monthly minimum wages are legislated, the analysis is confined to "full-time employees," defined as individuals working 30­50 hours weekly who are not in the self-employment sector. 3. The spikes in the graphs may be a result of employers who reward or respon- dents who report a round number. In Chile and the República Bolivariana de Venezuela, with a minimum wage of 100,000 pesos monthly, this may be the case, but in the other countries, the minimum wage is not a round number, so the evidence of the role of the minimum wage in distorting the wage distribution is more compelling. 4. The differences in the estimated elasticities from different studies of the same country are due to different methodologies, segments of the labor market The Worker: How Do Minimum Wages Affect Other Wages and Employment? 49 studied, and the period studied. The latter is particularly important in Brazil, since the hyperinflationary period and wage indexing during the 1980s may have created a spurious correlation between wages and the minimum wage. 5. Although the minimum wage in Jamaica is J$800 dollars per week, the cliffs are not noted for every multiple of 100. The minimum wage in Mexico was Mexi $888 monthly. The surveys for which this phenomenon is noted do not use the minimum wage as a response in their surveys. Instead, an actual cur- rency value of wages or income is recorded. 6. Lemos (2002) finds that a 10 percent increase in the Brazilian minimum wage reduces the wage gap between the 50th and 90th centiles by 34 percent and the gap between the 10th and 90th centiles by 15 percent. Neumark, Schweitzer, and Wascher (2000) find that an increase in the minimum wage increases the wage of those earning less than two minimum wages in the short run, with no long-term effects. Angel-Urdinola and Wodon (2003) find a decrease in the Gini coefficient by 0.0033­0.0037 percentage points for a 6 percent increase in the minimum wage in Brazil. 7. The lowest wage category has a particularly large minimum wage effect. This phenomenon has been observed in the other studies that use this methodology (Neumark, Schweitzer, and Wascher [2004] for the United States, Maloney and Nuñez [2004] for Colombia), that attribute the particularly high impacts to mismeasurement. 8. The author constructs a counterfactual distribution of the wage distribution in 1999, if the real minimum wages, characteristics of the labor force, and unobserved characteristics in 1988 had not changed. The difference in the Gini can then be attributed to each of these characteristics. For a full discussion of the methodology, see appendix I. 9. An increase in unemployment is not synonymous with a decrease in employ- ment, since a higher minimum wage may induce labor force entry, thus increasing the unemployment rate. 10. "Young" are defined as workers age 16­24, "prime-aged" are age 25­50, and older workers are age 51­65. "Unskilled" are those with no more than a pri- mary education ("no school" are omitted), "semiskilled" are those with no more than a secondary education, and "skilled" are those with a university education. C H A P T E R 5 The Households: The Minimum Wage as an Antipoverty Tool Chapter 4 showed that minimum wages generally have a positive effect on the wage distribution and a negative effect on employment. Ultimately, though, we are concerned with the impact of the minimum wage on poverty. This requires us to shift our unit of analysis from the individual worker to the household for two reasons. First, households pool income so the net effect of the minimum wage on household earnings is more relevant than the effect on the labor status or earning of an individual within the household.1 Second, poverty is measured at the level of the household, not the individual. By examining the level of the minimum wage relative to household needs and the effect of the minimum wage on household income, we will have a better understanding of the useful- ness of the minimum wage as an antipoverty tool. This section is concerned with three questions. First, it determines whether the minimum wage is a viable household poverty reduction tool. Second, it aggregates the gains and losses across households and determines the net impact of an increase in the minimum wage on households at different parts of the income distribution. Finally, it identifies which households win and which households lose when the minimum wage increases. 51 52 Minimum Wages and Social Policy TheValue of the MinimumWage Is Below the Household Subsistence Level The minimum wage exceeds the US$2 daily poverty line for most countries. If the objective of the minimum wage is to guarantee that all workers, regardless of productivity level, receive a living wage, a good benchmark for the "fairness" of the minimum is its level relative to the poverty line--that is, the minimum income necessary to purchase basic good and services. In 17 of the 20 countries studied, the minimum wage exceeds the crude US$2 per day (PPP-adjusted) poverty line (figure 5.1). In Paraguay, the minimum wage is equivalent to more than nine times the PPP-adjusted US$2 per daily poverty line, whereas in the República Bolivariana de Venezuela, Mexico, and Uruguay, it is less than this poverty line. Figure 5.1. MinimumWage Relative to the Household per Capita Poverty Lines--US$2 per Day or the Consumption-Basket Poverty Line Paraguay El Salvador Dominican Republic Costa Rica Argentina Chile Colombia Guatemala Honduras Brazil Panama Guyana Nicaragua Peru Bolivia Ecuador Jamaica Venezuela, R.B. de Mexico Uruguay 0 1 2 3 4 5 6 7 8 9 10 mw/US$2 per day poverty line mw/poverty line Source: Kristensen and Cunningham (2006). The Households: The Minimum Wage as an Antipoverty Tool 53 The "basic goods" poverty measure gives similar conclusions.2 Although the value of the basic goods basket is generally lower than the PPP-adjusted $2 per day, the minimum wage is still above the poverty line in all countries except Jamaica, Uruguay, Mexico, and El Salvador (figure 5.1). Thus, for both poverty measures, the minimum wage seems sufficient to meet the minimum needs of the individual. The minimum wage is not sufficient to meet household consumption needs. Much minimum wage legislation in Latin America states that the minimum should be sufficient to cover the needs of the family (Starr 1981). If the worker's income also must provide for the needs of house- hold dependents, the minimum wage falls far short: in households with one dependent per worker, the minimum wage can provide the basic subsistence (basic goods basket) in only 7 of the 17 sampled countries (Peru, Nicaragua, Dominican Republic, Honduras, Guyana, Chile, and Costa Rica). And in households with one minimum-wage earner and three dependents, the level of the minimum wage is only sufficient to provide income above the poverty line in Guyana, and perhaps in Chile and Honduras.3 The Minimum Wage Decreases Poverty Rates but Does Not Help the Most Poor There is very little evidence in LAC about the impacts of the minimum wage on household poverty and inequality (table 5.1).4 Two studies in the mid-1990s used aggregate country data and found that higher minimum wages have some poverty-increasing effects and some poverty-decreasing effects (Lustig and McLeod 1996, Morley 1992). More careful simula- tions using Brazilian microdata found that an increase in the minimum wage has no effects on poverty, after taking into account the unemploy- ment effects of a minimum wage increase (IPEA 2000). Similarly, a direct test of how an increase in the minimum wage affects the poverty status of U.S. families also finds no effects (Neumark and Wascher 1997). However, the U.S. paper does find that an increase in the minimum wage increases the wages of the poorest, not by enough to bring them above the poverty line but enough to have a significant effect. In countries where the minimum wage causes unemployment, the poorest do not benefit from higher minimum wages; instead, the bene- fits are concentrated among those households near the poverty line. In Colombia, the poorest households--those in the bottom 25 percent of the household income distribution--neither gain nor lose from higher 54 Minimum Wages and Social Policy Table 5.1. Review of the Key Literature Examining the Impact of the MinimumWage on Household Poverty and Inequality in Latin America Poverty/inequality impact of an Country increase in the mw Note Source Poverty Household-level data Brazil 1997 0 household-level IPEA (2000) data Colombia 1997­2001 ­ household-level Arango and data Pachon (2003) Mexico 1988­99 ­ household-level Cunningham and data Siga (2006) United States 1986­95 0 household-level Neumark and data Wascher (1997) Aggregate estimates Asia, Africa, LAC 1950­80 ­ Lustig and McLeod (1996) Latin America 1981­89 +, ­a recovery, Morley (1992) recessions Inequality Brazil 1995­2001 + indirect test, only Neumark, Cunningham, losses for poor and Siga (forthcoming) Colombia 1984­2001 0, +a direct test, short Arango and run, long run Pachon (2003) Mexico 1988­98 ­ indirect test, Cunningham and faster gains Siga (2006) among poorest households a. Each sign corresponds to the group listed in the Note column, respectively. minimum wages. As only households with labor income are included in the sample, this suggests that either the wage and unemployment effects balance each other out in this sample--some households gain while others lose in equal proportions--or that the minimum wage laws do not have any effects on this segment of the wage-earning population. The latter explanation is more likely, since, as shown in chapter 4, the high minimum wages in Colombia pass over workers in the bottom half of the wage distribution. Only households in the middle of the house- hold income distribution--25th to 80th centiles--benefit from an increase in the minimum wage. The poverty line is in the 50th centile of the wage distribution, so poverty rates may change, but this is due to increases in household income among the better-off poor families The Households: The Minimum Wage as an Antipoverty Tool 55 (Arango 2004). Box 5.1 summarizes the methodologies used to test minimum wage effects on household poverty. Mexico presents the case where the minimum wage, although low, increases household earnings among the poor. The workers analysis for Mexico showed positive wage gains, especially among the poor, and no unemployment shocks (chapter 4). When taking the household as the unit of observation, the poorest households experience the highest wage gains following an increase in the minimum wage (figure 5.2). The gains are con- centrated among households that earn up to one minimum wage per capita. Notably, these are the very poorest, so there are still many poor Mexicans who do not benefit from the minimum wage increase (Cunningham and Siga 2006), and the effects on the poverty rate are likely to be negligible. The Brazil case demonstrates the dynamic poverty effects of the minimum wage, where short-run gains are offset by long-run losses. On average, the higher minimum wage has a positive short-run effect on very poor families, those in the bottom 30 percent of the household income distribution when wages increase. But in the long run, once labor contracts expire and workers are fired, the net effect is a loss of household income per capita among poor families. In other words, the income loss Box 5.1 Methodology forTesting the MinimumWage Effects on Household Poverty and Inequality The findings of this section are surprising: there is no previous literature from Latin America and little from the OECD to corroborate findings. Thus, two general methodologies were used to ensure that the findings were robust. The methodo- logies are defined in detail in appendix I, but in summary they are as follows: · Using short panel data (small t, large n), household income levels are tracked before and after minimum wage changes. The variation in the minimum wage (also, the mw/median wage was used) is correlated with changes in household income (used for Colombia and Mexico). · Cross-section time series methods were used to examine the correlation between wage movements at different income centiles and variations in the minimum wages over long periods (big t, small n). Aggregating across households, the effects of the minimum wage on different population groups are estimated (used for Brazil and Colombia). 56 Minimum Wages and Social Policy Figure 5.2. Impact of a 10% Increase in the MinimumWage on Household Income in Mexico, 1999 4 in 3 income 2 1 increase % 0 household ­1 0.01