101596 Balancing regulations to promote jobs From employment contracts to unemployment benefits Arvo Kuddo | David Robalino | Michael Weber Balancing regulations to promote jobs From employment contracts to unemployment benefits Arvo Kuddo, David Robalino, Michael Weber Disclaimer: The findings, interpretations, and conclusions expressed herein are those of the authors and do not necessar- ily reflect the views of the World Bank Group, or its Executive Directors, or the governments they represent. The World Bank Group does not guarantee the accuracy of the data included in this work. The authors attest that the paper represents original work. It fully references and describes all relevant prior work on the same subject. Foreword Balancing labor regulations is a crucial policy challenge for no ‘one size fits all’ blueprint for reform, there are some gen- countries that want to protect their workers while promoting eral principles that can help improve the design of labor laws more, better, and inclusive jobs. These regulations help miti- and their implementation. The report also underscores the gate labor market imperfections, support social cohesion, importance of dialogue between representatives of employ- and enhance economic efficiency. ers and workers as well as other major stakeholders. Studies from both developed and developing countries show Significantly, this report reflects a shared vision between that labor market regulations can affect important economic the ILO and the World Bank Group to promote policies that outcomes such as productivity, labor force participation, encourage job creation and protect workers. This has been earnings, or informal employment. The evidence points to possible thanks to the commitment of both institutions to the need to ensure that policies to promote employment focus on the lessons derived from rigorous research and opportunities also protect workers and provide incentives for international experiences. We hope this report will inform work.   countries’ paths to achieve the Global Goal to promote inclu- sive economic growth, employment, and decent work for all. This report offers guidelines to design, implement, and reform labor market regulations in four areas: employment con- Arup Banerji tracts, minimum wages, dismissal procedures, and income Senior Director, Social Protection, Labor & Jobs protection for the unemployed. It shows that, while there is World Bank Group BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS iii Acknowledgements This report was prepared by Arvo Kuddo, David Robalino, and Jeff Vogt, and Carolin Vollmann. Brent Wilton provided com- Michael Weber under the general guidance of Arup Banerji, ments for the IOE.1 Senior Director of the World Bank Global Practice Social Pro- tection & Labor and the Jobs Group. The report was written in Comments within the World Bank were received from Diego close consultation with the International Labor Organization Angel-Urdinola, Omar Arias, Roberta Gatti, Julian Messina, (ILO), the International Trade Union Confederation (ITUC), Gonzalo Reyes, Jan Rutkowski, Setareh Razmara, Dena Rin- and the International Organization of Employers (IOE). The gold, and Joana Silva. Gordon Betcherman, Professor in the team from the ILO under the leadership of Sandra Polaski, School of International Development and Global Studies, ILO’s Deputy Director-General for Policy, included Mariya University of Ottawa kindly reviewed the report. Aleksynska, Christina Behrendt, Patrick Belser, Janine Berg, The authors would like to acknowledge the valuable com- Karen Curtis, Nancy Donaldson, Anne Drouin, Natan Elkin, ments and suggestions made by all reviewers during the Evelyn Elsaesser, Michael Henriques, Martine Humblet, Juan consultation process. de la Iglesia, Mélanie Jeanroy, Sangheon Lee, Philippe Marca- dent, Angelika Muller, Céline Peyron Bista, Bill Salter, Emman- uelle St-Pierre Guilbault, Kristen Sobeck, Corinne Vargha, and 1.  The participation of these reviewers does not signify that they or their Erick Zeballos. The team representing the ITUC/Global Unions organizations necessarily endorse the content or views expressed in the was composed of Peter Bakvis, Kwabena Otoo, John Schmitt, report. iv Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi Chapter 3: Dismissal Procedures . . . . . . . 23 Chapter 1: Employment Contracts . . . . . 1 3.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3.2. Objectives and Impacts . . . . . . . . . . . . . . . 24 1.2. Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3.3. Design and Implementation . . . . . . . . . . . 24 1.3. Labor Market and Social Impact . . . . . . . . 2 3.4. Resources . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.4. Design and Implementation . . . . . . . . . . . 4 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.5. Resources . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Annex—Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . 29 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Chapter 4: Income Protection for the Annex—Chapter 1 . . . . . . . . . . . . . . . . . . . . . . . 7 Unemployed: Severance Pay and Chapter 2: Minimum Wages . . . . . . . . . . 10 Unemployment Benefit Schemes . . . . 32 4.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 4.2. Objectives and Rationale . . . . . . . . . . . . . . 33 2.2. Objectives and Rationale . . . . . . . . . . . . . . 10 4.3. Labor Market Impacts . . . . . . . . . . . . . . . . 34 2.3. Labor Market and Social Impact of Minimum Wages . . . . . . . . . . . . . . . . . . 11 4.4. Design & Implementation of Income Protection for the Unemployed . . . . . . . . 35 2.4. Design and Implementation of Minimum Wage Policies . . . . . . . . . . . . . . . . . . . . . . . 13 Entitlement and Eligibility . . . . . . . . . . . . . . . 37 Setting the Level . . . . . . . . . . . . . . . . . . . . . . 13 Reforming Severance Pay . . . . . . . . . . . . . . . 38 Differentiation . . . . . . . . . . . . . . . . . . . . . . . . 14 4.5. Resources . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Adjusting Minimum Wages over Time . . . . . . 15 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Institutional Arrangements . . . . . . . . . . . . . . 15 Annex—Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . 42 Noncompliance and Enforcement . . . . . . . . . 16 2.5. Resources . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Annex—Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . 21 BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS v Introduction Labor regulations can correct imperfections in labor markets This report provides general principles for the design and resulting from inadequate information, uneven bargaining implementation of labor regulations in four areas: (a) employ- power, limited ability to enforce long-term commitments, ment contracts, (b) minimum wages, (c) dismissal proce- or insufficient insurance mechanisms against employment dures, and (d) severance pay and unemployment benefits. related risks.2 Thus, labor regulations can, if well designed, This selection of topics resulted from internal consultations, avoid inefficient and inequitable labor market outcomes and ongoing dialogue with client countries, as well as the impor- have an important role to play in any country. tance of these topics for labor markets. The report summa- rizes the main findings from the literature and discusses In addition, there are fundamental rights at work that should country experiences and policy implications. It targets task be part of any employment legislation. These include (a) the team leaders involved in policy dialog on labor regulations, right to engage in work conditional on meeting the minimum as well as policymakers. age requirement; (b) the right to pursue a freely chosen or accepted occupation without coercion; (c) prohibition of dis- The report suggests that there are general principles that can crimination on grounds such as race, color, sex, religion, polit- guide the design of labor regulations. For instance, in terms ical opinion, national extraction or social origin, that would of employment contracts, it is important to ensure conver- impair equality of opportunity or treatment in employment gence in the types of benefits and protections that workers or occupations; (d) the right to equal remuneration for both receive, regardless of the length of time they spend with a male and female workers for work of equal value; and (e) the given employer. For minimum wages, it is necessary to keep right to establish and join workers’ and/or employers’ organi- regulations simple and transparent, and to reduce discretion zations and to bargain collectively.3 by having an independent body that periodically assesses the level of the minimum wage and its economic and social The challenge is to establish the right balance between work- impacts. Regarding dismissal procedures, the report rec- ers’ protection and flexibility in the management of human ommends giving flexibility to firms in the management of resources at the firm level, that is, avoiding both over- or human resources, as long as there is appropriate advance under-regulation. Between these two extremes, there is a notice, an adequate system of income protection, and effi- ‘plateau’ where appropriately designed regulations can alle- cient mechanisms to detect and sanction discrimination. viate (labor) market failures, offer adequate protection to Finally, for severance pay, the recommendation is to rely more workers, and contribute to shared prosperity without impos- on unemployment benefit systems that reduce employees’ ing unreasonable costs on firms. risk of not receiving payments if their employers face liquid- ity constraints or go out of businesses. 2.  World Bank (2012). Beyond some of these general principles, however, there is no 3.  The ILO Declaration on Fundamental Principles and Rights at Work overall blueprint to design or adapt labor regulations. Rather, and its Follow-up (1998) declares that all Members States have an obligation to respect, promote and realize the principles concerning the there are different reform paths that depend on country char- fundamental rights which are the subject of the core ILO Conventions, acteristics and are shaped by social, political, economic, and namely: (a) freedom of association and the effective recognition of the historical circumstances combined with different legal tradi- right to collective bargaining; (b) the elimination of all forms of forced or compulsory labor; (c) the effective abolition of child labor; and (d) the tions.4 A recommendation is to reform labor regulations in a elimination of discrimination in respect of employment and occupation. See also the EU’s Charter of Fundamental Rights proclaimed by the Nice European Council on December 7, 2000. Retrieved from: http://www .europarl.europa.eu/charter/pdf/text_en.pdf. 4.  See, for example, World Bank (2012). vi systematic and comprehensive manner. In the past, several workers.6 Besides, in many countries, employment laws are countries narrowly focused on selected labor regulations often ineffective because of evasion and weak enforcement, without considering the complexity of effects on the labor even for workers with formal employment. market. Although the focus of the report is the main labor law, it is A key message throughout this report is the essential role important to note that it typically provides only for the mini- of social dialogue and tripartism in the development and mum legislative requirements that employers and workers implementation of labor regulations. As stated by the ILO, must comply with on commencing or terminating employ- the main goal of social dialogue is to promote consensus ment and during the period of employment. Minimum stan- building and ensure involvement of the main stakeholders dards may be extended by other laws and acts, collective in decisions and actions related to labor relations. This would agreements, local (internal) regulatory acts, or individual improve opportunities for workers to obtain decent and pro- employment contracts which define the rights and respon- ductive work in conditions of freedom, equality, security, and sibilities of the parties in greater detail. These instruments, human dignity. Social dialogue includes all types of negotia- however, cannot undermine the provisions for employees tions, consultations, and exchange of information between, established by labor law. or among, representatives of governments, employers, and workers or between employers and workers’ representatives The report is organized into four chapters. Chapter 1 dis- on issues of common interest.5 cusses essential elements of the employment contract as well as emerging forms of contractual relationships, with It is necessary to recognize, however, that there are limits to the focus on temporary employment and part-time employ- what labor laws and policies can accomplish. In particular, ment contracts. Chapter 2 examines the establishment of the labor regulations primarily benefit employees in the formal minimum wage, including setting the minimum wage, the sector. According to recent estimations, over 60 percent of level and differentiation of minimum wages, coverage, and jobs in middle- and low-income countries are in the informal adjustments in the level. Chapter 3 presents the main rules sector; these include farmers, the majority of own-account and regulations regarding the termination of employment workers and informal wage employees, or close to 1.5 billion contracts. Chapter 4 presents a discussion on the design and implementation of income protection schemes in case of unemployment, focusing on severance pay and unemploy- ment benefit schemes. 5.  See ILO, 2013c “National Tripartite Social Dialogue: ILO Guide on Improved Governance” available at http://www.ilo.org/wcmsp5/groups/ public/@ed_dialogue/@dialogue/documents/publication/wcms_231193 .pdf. 6.  See World Bank (2014); ILO (2015). BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS vii 1. Employment Contracts 1.1. Overview have had no choice but to agree to a temporary contract var- ies considerably, from 25 percent in the United Kingdom to 51 percent in Poland and 65 percent in Italy.8 As economies and labor markets evolve, a wide variety of employment contracts have been developed. These con- Fixed-term (temporary or limited duration) contracts are more tracts differ significantly in the degree of employment secu- common among young people, new labor market entrants, rity, associated working and living conditions, and the types and those with the lowest education or skill levels. Women of benefits that must be provided to workers. Although full- are more likely to be in temporary work than men. Among time employment contracts of indefinite duration are the youth aged 15 to 24 years, the share in OECD countries was most common form of an employment relationship in devel- 24.5 percent in 2012. The highest share of young workers on oped countries, variations, including temporary employ- temporary contracts was reported in Slovenia, 72.0 percent, ment (fixed-term) contracts, part-time contracts, on-call followed by Spain, 62.4 percent, and Poland, 56.2 percent. contracts, zero-hour contracts, contracts for workers hired through temporary employment agencies, or freelance con- Especially for young workers, fixed-term work can provide tracts, have become established features of modern labor an entry point into the workforce and an opportunity to gain markets. Looking at developing countries, where casual experience and skills, but such contracts can also be exploited work is widespread, many workers do not have written con- so that young people are trapped in perpetually renewed tracts at all. Moreover, many of the alternative work arrange- fixed-term contracts. The use of fixed-term contracts also ments offer a limited set of benefits for workers, for example appears to be growing in some developing countries. In this a lack of social insurance coverage or no protection against context, the policy objective is to ensure that workers have dismissal. access to proper working conditions, equal pay, and access to social and fringe benefits. To some extent, reforms in labor legislation around the world have been associated with easing the recourse to temporary In many countries of Latin America (Argentina, Uruguay, Para- or alternative forms of employment while leaving existing guay, Peru, Bolivia, Guatemala, Honduras, and others), part- provisions for regular or permanent contracts unaltered. As time work is widespread and particularly prevalent among a result, the share of workers currently on temporary con- women. For example, in the period 2002–2004, 43 percent of tracts is quite significant. According to data available from employed women worked part-time relative to 27 percent of Organization for Economic Co-operation and Development employed men.9 (OECD) countries, in 2012, 11.8 percent of the workers had Many workers are willing to take on part-time work as it temporary contracts. In some countries, the ratio of workers enables them to combine work with family responsibilities or on temporary contracts is much higher, for example, 26.9 per- education. While in OECD countries a majority of part-time cent in Poland, 23.6 percent in Spain, 20.7 percent in Portugal, employees work on a voluntary basis, with few moving into and 13.7 percent in Japan (Annex—Chapter 1 Figure  1.1).7 full-time work and many staying in part-time jobs for long The share of workers on temporary contracts who claim to 7.  OECD (2010). Labour Market Statistics: Employment by Permanency 8. http://epp.eurostat.ec.europa.eu/portal/page/portal/statistics/ of the Job Incidence. OECD Employment and Labour Market Statistics search_database. (database). doi: 10.1787/data-00297-en (accessed on 26 February 2014). 9.  López Bóo et al. (2009). BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 1 periods, there has been a substantial increase in involuntary supplementary guarantees to workers. These arrangements, part-timers in many advanced economies since the recent however, may not provide conditions for the employees financial crisis.10 which would be less favorable than those established by the main labor law as well as other laws and acts. The key policy issue when discussing regulations regarding employment contracts is how to balance the objectives of An employment contract typically itemizes essential aspects providing sufficient flexibility to employers and workers in of the employment relationship such as salary, benefits, and defining work arrangements while ensuring that workers are duration of employment if the contract is not open-ended fully able to exercise their rights and enjoy adequate legal (Box 1). benefits and protections. This chapter will discuss some gen- eral principles to be considered for achieving this. 1.3. Labor Market 1.2. Objectives and Social Impact The objective of an employment contract is to set up the Labor regulations affect the nature and types of employment rights and obligations of the parties to the employment rela- contracts, which in turn affect employment and salary levels, tionship. An employment contract is an important indicator access to training, protection from dismissal, ability to exer- for the existence of an employment relationship.11 cise freedom of association, collective bargaining, and trade union rights, and social protection at work, among others. The employer is obliged to secure the employee the work agreed upon by the parties in the employment contract. The Although international evidence is limited, some studies employer is required to provide the employee all the neces- have found that policy reforms that facilitated the creation sary means and materials for work, a safe and healthy work of fixed-term jobs in Europe in the late 1990s raised the prob- environment, and appropriate remuneration that may not be ability, on average, that a worker would be on a fixed-term lower than the existing minimum wage. The employer should contract. However, there is no evidence that such reforms take the necessary measures to ensure that each employee increased overall employment; instead, they appear to have has been given sufficient training. The worker’s duty is to encouraged substitution of temporary for permanent work.12 carry out the job that he agreed to in the employment con- tract, during the working hours, and at the location specified Several studies focus on major Spanish reforms in the early for carrying out the work. When entering into employment, 1980s that liberalized fixed-term contracts without chang- the worker should have access to basic social protection ing dismissal costs for regular contracts and find, in general, arrangements that cover old age, disability, employment that this led to a large increase of fixed-term contracts and injury, maternity, unemployment, sickness, and medical care. a reduction in permanent contracts.13 Not surprisingly, Spain has one of the highest concentrations of workers with tem- Labor law typically regulates the employment relationship, porary contracts. Evidence from Spain also suggests that thereby setting the rights and obligations for employers and when the regulatory gap between permanent and tempo- employees to comply with when commencing or terminat- rary employment is large, transition rates across these two ing employment and during the period of employment. states are low.14 Other legislative acts, internal regulations, collective agree- ments, and an individual employment contract provide An important finding is that although temporary jobs may increase employment opportunities for some, they do not lead to net job creation and can also result in undesirable 10.  In OECD countries, between 2000 and 2012, the share of labor market outcomes.15 From the workers’ perspective, involuntary part-timers as a percentage of part-time employment has fixed-term jobs are less secure and pay lower than average increased from 12.0 percent to 17.1 percent and the share of involuntary wages. For example, in the European Union (EU), tempo- part-timers in total employment from 2.4 percent to 4.5 percent. See http://stats.oecd.org/BrandedView.aspx?oecd_bv_id=lfs-data-en&doi= rary workers earn on average 14 percent less than workers data-00299-en. with open-ended contracts after controlling for a number of 11.  While the definitions of the basic terms ‘employee’ or ‘worker’, ‘employer’, and ‘employment contract’ are commonly left to national legislation, methods for determining the existence of an employment relationship are internationally recognized in the ILO Employment Relationship Recommendation, 2006 (No. 198). It is also important to 12.  Kahn (2010). make a clear distinction between employment and self-employment. 13.  See Bentolila et al. (2008); Aguirregabiria and Alonso-Borrego See for example, the updated “Code of Practice for Determining (2009). Employment or Self-Employment Status of Individuals” prepared by the Hidden Economy Monitoring Group (http://www.revenue.ie/en/tax/rct/ 14.  For example, Güell and Petrongolo (2007); Garcia-Serrano and Malo determining-the-correct-employment-status-of-a-worker.html). For more (2013). comparative examples, see also ILO (2007, 2013b). 15.  Kahn (2010). 2 Box 1: Suggested Essential Aspects of the Contract or Employment Relationship (a) The identities of the parties The place of work; where there is no fixed or main place of work, the principle that the employee is employed at vari- (b)  ous places and the registered place of business or, where appropriate, the domicile of the employer (c) The title, grade, nature, or category of the work for which the employee is employed or a brief specification or descrip- tion of the work The date of commencement of the contract or employment relationship (d)  In case of a temporary contract or employment relationship, the expected duration thereof (e)  The amount of paid leave to which the employee is entitled or, where this cannot be indicated when the information (f )  is given, the procedures for allocating and determining such leave The length of the periods of notice to be observed by the employer and the employee should their contract or (g)  employment relationship be terminated or, where this cannot be indicated when the information is given, the method for determining such periods of notice (h) The initial basic amount, the other component elements, and the frequency of payment of the remuneration to which the employee is entitled The length of the employee’s normal working day or week (i)  Where applicable, the collective agreements governing the employee’s conditions of work, or in the case of collective (j)  agreements concluded outside the business by special joint bodies or institutions, the name of the competent body or joint institution within which the agreements were concluded Probationary period, if any (k)  personal characteristics.16 Temporary contracts can enable effect may be to produce high turnover within these alter- firms to reduce labor costs, substituting lower-paid tempo- native types of contracts, with many workers going through rary labor for permanent workers. Temporary workers tend several unemployment spells before obtaining a regular to have reduced access to training provided or subsidized by job.18 There is a risk that part of the workforce will become firms as the limited duration of their employment relation- trapped in a succession of short-term, low-quality jobs with ship discourages investment in (firm-specific) human capi- inadequate social protection, leaving them in a vulnerable tal. Fixed-term workers are subject to higher turnover. It also position. In the case of economic difficulties, temporary appears that the prevalence of fixed-term work is highest for workers tend to be the first workers whom employers will those with the lowest levels of education and that it is more make redundant. This effect reduces the response of wages extensive in the primary, service and construction sectors to increases in unemployment, leading to lower permanent than in manufacturing. Taken together, these outcomes can employment rates than otherwise.19 be suboptimal for the labor market and the economy as a whole if they encompass lower training, effort, productivity, Part-time work can also have mixed outcomes. On the one and stability of household incomes. hand, employers have more possibilities to adjust working hours to the economic cycle. They can promote higher labor In addition, the expansion of alternative types of employment force participation and more family-friendly work sched- contracts can reinforce labor market duality.17 In fact, its main ules or arrangements. On the other hand, part-time workers have lower hourly wages, on average, than full-time workers 16.  EC (2010). 17.  A dual labor market reflects a situation where the workforce is reduced or non-existent, giving rise to a situation of uncertainty about divided between the permanently employed and relatively well-protected future prospects (see EC 2003). “insiders” and “outsiders.” Outsiders are precariously and informally employed as well as the unemployed. This group occupies a grey area 18.  Blanchard and Landier (2002); Cahuc and Postel-Vinay (2002). where basic employment or social protection rights may be significantly 19.  Blanchard and Landier (2002); Bentolila and Dolado (1994). BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 3 in almost all OECD countries. Part-timers face a higher risk of work, and casual work are the relationships most commonly poverty and are less likely to have access to unemployment associated with precariousness.24 In principle, employment benefits or reemployment assistance if they become unem- relationships should be legitimated in a written employment ployed. Part-time workers may also be excluded from coverage contract. When this is not the case, it is particularly impor- of social protection systems such as health care or pensions, tant to identify mechanisms to prevent abuse of employment receive fewer fringe benefits, and face more limited career pos- relationships. sibilities than full-time jobs. Moreover, increases in the level of involuntary part-time employment often coincide with peri- Temporary (fixed-term) contracts can be appropriate to ods of economic recession since fewer standard full-time jobs allow firms to cope with unexpected fluctuations of demand; are available and workers are forced to take jobs with reduced replace permanent staff on holiday, maternity leave, or sick working hours as ‘better-than-unemployment’ alternatives. leave; hire workers with specialized skills to carry out specific time-limited projects; or launch start-up ventures implying risky and uncertain returns. However, repeatedly renewing 1.4. Design and short-term contracts for ongoing work instead of hiring on a standard open-ended basis should be discouraged. Implementation In an employment relationship, fixed-term workers should not be treated in a less favorable manner than comparable This section discusses key issues in the design and implemen- permanent workers. The equal treatment should apply in tation of employment contracts, including temporary con- almost all fields: health and safety, remuneration, duration tracts and part-time contracts, as well as important aspects of and organization of working time, paid leave and public the probationary period.20 The fundamental principle encom- holidays, social benefits, and information on available vacant passing this approach involves ensuring adequate protec- posts and their terms (Annex—Chapter 1 Figure 1.2). Excep- tion for all workers, independent of the type of employment tions to the general principle of equal treatment are possible contract.21 with respect to terms which are directly related to the limited Due to the diversity of economic needs, employers should be duration of employment, for example, notice or severance able to offer different types of contracts provided that ade- pay.25 National regulations, such as in Australia for casual quate protection for the workers involved can be secured. workers, may provide that, if the equal treatment does not Therefore, legislation should treat workers in temporary or apply to all fields (such as paid or sick leave), workers should part-time employment to protection equivalent to that of be compensated with a substantially higher pay. In these comparable permanent full-time workers, determined in cases, effective monitoring or complaint mechanisms would proportion to hours of work or other objective indicators.22 be needed to prevent abuse. Furthermore, employers should facilitate atypical workers’ Equalizing benefits and dismissal procedures between access to appropriate training opportunities to enhance their fixed-term and open-ended contracts might prevent mis- skills, career development, and occupational mobility. use of successive fixed-term contracts. However, this should Employment legislation should include measures to combat not lead to lower protection, as it would otherwise exacer- disguised employment relationships that hide their true sta- bate the negative externalities for the labor market and the tus because these can have adverse effects on the affected economy as a whole in terms of suboptimal investment in workers and on tax revenue, thus externalizing the costs of training, productivity, and stability of household incomes. the firm to others.23 False or bogus self-employment, informal Clearly, firms have less incentives to convert workers from temporary to permanent contracts in cases where it is easy to hire temporary workers but costly to dismiss regular ones. Addressing problems of dismissal procedures (see Chap- 20.  Other forms of atypical work are beyond the scope of this report. 21.  See for example, the ILO Employment Relationship ter  3) is therefore essential to avoid temporary workers— Recommendation, 2006 (No. 198). often young, unskilled, or old workers—being forced into a 22.  ILO Recommendation No. 166 (Article 3(2)) provides examples of constant rotation across temporary jobs. This is important legal rules to prevent abusive recourse to fixed-term contracts. This may since there is evidence that their low probability of being be done by (a) limiting recourse to contracts for a specified period to cases in which, owing either to the nature of the work to be effected or converted into a permanent job, even if they are very skilled, to the circumstances under which it is to be effected or to the interests of the worker, the employment relationship cannot be of indeterminate duration; (b) deeming contracts for a specified period, to be contracts of employment of indeterminate duration; (c) deeming contracts for a specified period of time, when renewed on one or more occasions, to be their client is such that they are effectively employees of the client. They contracts of employment of indeterminate duration. The ILO Termination are treated as self-employed for tax evasion or other purposes. of Employment Recommendation, 1982 (No. 166). 24.  McKay et al. (2012). 23.  The term ‘disguised employees’ refers to workers who receive 25.  See for example, the ILO Termination of Employment Convention, payments from a client via an intermediary and whose relationship with 1982 (No. 158). 4 reduces incentives to perform well and to learn on the job, the work agreed upon. During this period, labor contracts are with obvious costs in productivity growth and well-being of not fully covered by employment protection provisions. If the worker.26 employers are not satisfied, they can terminate the employ- ment contracts of workers under probation with more flex- In all cases, there should be provisions that define ible conditions than for regular workers. Two main cases can 1. objective reasons justifying the length and renewal of be distinguished: (a) protection of unfair dismissal does not such contracts or relationships; apply (though dismissals cannot be made on prohibited or discriminatory grounds); (b) normal notice period and sever- 2. maximum total duration of successive fixed-term ance pay rules do not apply. employment contracts or relationships; and The duration of the probationary period should be reason- 3. number of renewals of such contracts or relationships. able. It usually ranges between 3 and 6 months but a shorter or longer trial period may be stipulated in the collective agree- Employers should also inform fixed-term workers about ment or agreed upon by the parties in individual employment vacancies that become available in the undertaking or the contract(s) (see also Annex—Chapter 1 Table 1.1).30 In some establishment to ensure that they have the same opportunity jobs, particularly high-level positions, employers need more to secure permanent positions as other workers. time to determine if a worker or employee is a good match. It is also important to ensure that part-time workers receive On the other hand, for many unskilled and semiskilled occu- conditions equivalent to those of comparable full-time work- pations, it is not necessary to have a lengthy probationary ers in the fields of (a) salary; (b) maternity protection; (c) termi- period to verify the abilities of the worker. However, exces- nation of employment; (d) paid annual leave and paid public sively short trial periods may not permit sufficient monitor- holidays; and (e) sick leave. Statutory social security schemes ing of workers and raise the risk of disciplinary or economic that are based on occupational activity should be adapted so dismissals for workers’ unsuitability at a later stage. that part-time workers enjoy conditions equivalent to those To prevent abuse of the probationary period, it might be of comparable full-time workers or to encourage the conver- appropriate to set a maximum number of trial workers for a sion of full-time posts to part-time contracts in order to avoid single position. Especially in micro enterprises and in the ser- providing certain benefits.27 (Annex—Chapter 1 Figure 1.3). vice sector, this could avoid the practice of some employers However, exceptions can be allowed to control transaction misusing probationary periods by hiring the workers only on costs associated with social entitlements of certain types of probation and replacing them at the end of the period with part-time work. Part-time workers whose hours of work or new employees. To mitigate the abuse of the probationary earnings are below specified thresholds may be excluded period, some countries also stipulate exemptions to specific from the scope of some measures taken in the above- groups. For instance, the probationary period may not be mentioned fields, provided they are not excluded from used for individuals with disabilities, pregnant women, or employment injury benefits or maternity protection mea- workers with children of up to three years of age and some sures.28 The thresholds should be sufficiently low so as not to other categories of employees. exclude an unduly large percentage of part-time workers.29 It is important to allow the inclusion of a probationary period in the employment contract to confirm that the employee 1.5. Resources has the necessary professional and social skills for performing ILO Equal Remuneration Convention, 1951 (No. 100) ILO Discrimination (Employment and Occupation) Conven- tion, 1958 (No. 111) 26.  Aguirregabiria and Alonso-Borrego (1999). 27.  According to the ILO Part-Time Work Convention, 1994 (No. 175) ILO Workers with Family Responsibilities Convention, 1981 and Part-Time Work Recommendation, 1994 (No. 182). (No. 156) 28.  According to the ILO Convention 175 (Article 8), part-time workers whose hours of work or earnings are below specified thresholds may be ILO Workers with Family Responsibilities Recommendation, excluded from the scope of some of the statutory social security schemes and other measures, except in regard to employment injury benefits and 1981 (No. 165) maternity protection measures. 29.  For example, with regard to dismissal protection, in Denmark, to ILO Termination of Employment Convention, 1982 (No. 158) qualify for relevant benefits, the part-time worker must work at least 15 hours per week; in Germany, 18 hours per week; in Spain, 12 hours Termination of Employment Recommendation, 1982 ILO  per week; in France and the Netherlands, there is no threshold; in Ireland, (No. 166) the threshold is 12 hours per week; in Austria, 12 hours per week; in Sweden, 17 hours a week; and in the United Kingdom, the threshold is earnings of at least 57 pounds per week. See http://ec.europa.eu/ employment_social/labour_law/docs/06_parttime_implreport_en.pdf. 30.  World Bank (2013). BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 5 ILO Part-Time Work Convention, 1994 (No. 175) Bignami, R., G. Casale, and M. Fasani 2013. “Labour Inspection and the Employment.” Geneva: ILO. ILO Part-Time Work Recommendation, 1994 (No. 182) Blanchard, O., and A. Landier. 2002. “The Perverse Effect of Employment ILO  Relationship Recommendation, 2006 Partial Labor Market Reform: Fixed-Term Contracts in (No. 198) France.” Economic Journal 112 (480). Council Directive 91/533/EEC of October 14, 1991 on an employ- Boeri, T., B. Helppie, and M. Macis. 2008. “Labor Regulations er’s obligation to inform employees of the conditions appli- in Developing Countries: A Review of the Evidence and cable to the contract or employment relationship Directions for Future Research.” Social Protection Discus- Council Directive 97/81/EC of December 15, 1997 concerning sion Paper 0833, World Bank, Washington, DC. the Framework Agreement on part-time work concluded Botero, J. C., S. Djankov, R. La Porta, F. Lopez-de-Silanes, and by UNICE, CEEP and the ETUC A.  Shleifer. 2004. “The Regulation of Labor.” Quarterly WB Doing Business database on labor legislation (http:// Journal of Economics 119 (4). http://www.doingbusiness www.doingbusiness.org/) .org/documents/labor_June04.pdf. ILO Working Conditions Laws Database (http://www.ilo.org/ Cahuc, P., and F. Postel-Vinay. 2002. “Temporary Jobs, Employ- dyn/travail/travmain.home) ment Protection, and Labor Market Performance.” Labor Economics 9 (1). Employment protection legislation database—EPLex ILO  (http://www.ilo.org/dyn/eplex/termmain.home) Cazes, S., and M. Tonin. 2009. “Employment Protection Leg- islation and Job Stability: An European Cross Country Analysis.” Discussion Paper Series, Economics and Econo- Bibliography metrics 0902, Economics Division, School of Social Sci- ences, University of Southampton. Addison, J. T., and P. Teixeira. 2003. “The Economics of Employ- ment Protection.” Journal of Labor Research 24 (1). Cazes, S., and S. Verick. 2010. “What Role for Labour Market Policies and Institutions in Development? Enhancing Aguirregabiria, V., and C. Alonso-Borrego. 1999. “Labor Con- Security in Developing Countries and Emerging Econo- tracts and Flexibility: Evidence from a Labor Market mies.” Employment Working Paper 67, ILO, Geneva. Reform in Spain.” Working Paper 99-27, Universidad Car- los III de Madrid. Deakin, S. 2013. “Addressing Labour Market Segmentation: The Role of Labour Law.” Dialogue Working Paper 52, ILO, ———. 2009. “Labor Contracts and Flexibility: Evidence from Geneva. a Labor Market Reform in Spain.” Economic Series Work- ing Paper 09-18, Universidad Carlos III de Madrid. EC (European Commission). 2003. “Jobs, Jobs, Jobs: Creating More Employment in Europe.” Report of the Employment Aleksynska, M. 2014. “Deregulating Labour Markets: How taskforce chaired by Wim Kok. Brussels. Robust Is the Analysis of Recent IMF Working Papers?” Conditions of Work and Employment Working Paper ———. 2006a. “Employment in Europe 2006.” Brussels. Series 47, ILO, Geneva. ———. 2006b. “Modernising Labour Law to meet the Angel-Urdinola, D., and A. Kuddo. 2010. “Key Characteristics Challenges of the 21st Century.” Commission Green of Employment Regulation in the Middle East and North Paper of 22 November 2006. http://ec.europa.eu/ Africa.” Social Protection Discussion Paper 1006, World employment_social/labour_law/green_paper_en.htm. Bank, Washington, DC. ———. 2007. “Towards Common Principles of Flexicurity: Auer, P. 2007. “In Search of Optimal Labour Market Institu- More and Better Jobs Through Flexibility and Security.” tions.” In Flexicurity and Beyond. Finding a New Agenda for http://ec.europa.eu/social/main.jsp?catId=102&langId= the European Social Model, edited by H. Jørgensen and en&furtherPubs=yes. P. K. Madsen. Copenhagen: DJOF. ———. 2010. “Employment in Europe 2010.” Brussels. Bentolila, S., and J. Dolado. 1994. “Spanish Labour Markets.” García-Serrano C., and M. Malo. 2013. “Beyond the Contract Economic Policy April. Type Segmentation in Spain: Country Case Study on Bentolila S., J. Dolado, and J. Jimeno. 2008. “Two-Tier Employ- Labour Market Segmentation.” Employment Working ment Protection Reforms: The Spanish Experience.” CESifo Paper 143, ILO, Geneva. DICE Report, Vol. 6. 6 Güell, M., and B. Petrongolo. 2007. “How Binding Are Legal Kuddo, A., and D. Robalino. 2013. “Survey of Worker Protec- Limits? Transitions from Temporary to Permanent Work tion Measures 2012.” Manuscript. World Bank, Washing- in Spain.” Labour Economics 14. ton, DC. Heckman, J., and C. Pages. 2004. “Law and Employment: Les- López Bóo, F., L. Madrigal, and C. Pagés. 2009. “Part-Time Work, sons from Latin America and the Caribbean.” Chicago: The Gender and Job Satisfaction: Evidence from a Developing University of Chicago Press. Country.” Inter-American Development Bank. ILO (International Labour Organization). 2007. “The Employ- McKay, S., S. Jefferys, A. Paraksevopoulou, and J. Keles. 2012. ment Relationship: An Annotated Guide to ILO Recommen- “Study on Precarious Work and Social Rights.” Carried out dation No. 198.” Geneva. for the European Commission. London. ———. 2013a. “Application of International Labour Standards OECD (Organisation for Economic Co-operation and Devel- 2012 (I).” Report of the Committee of Experts on the Appli- opment). 2013. “OECD Employment Outlook 2013.” Paris. cation of Conventions and Recommendations, Geneva. ———. 2014. “OECD Employment Outlook 2014.” Paris. ———. 2013b. “Regulating the Employment Relationship in Europe: A Guide to Recommendation No. 198.” Geneva. Rutkowski, J., and S. Scarpetta. 2005. “Enhancing Job Opportu- nities: Eastern Europe and the Former Soviet Union.” Wash- ———. 2013c. “National Tripartite Social Dialogue: ILO Guide ington, DC: World Bank. on Improved Governance”. Geneva. Venn, D. 2009. “Legislation, Collective Bargaining and Enforce- ———. 2015. World Employment and Social Outlook: The ment: Updating the OECD Employment Protection Indi- changing nature of jobs.” Geneva. cators.” www.oecd.org/els/workingpapers. Kahn, L. M. 2010. “Employment Protection Reforms, Employ- World Bank. 2012. “World Development Report 2013: Jobs.” ment and the Incidence of Temporary Jobs in Europe: Washington, DC. 1996–2001.” Labour Economics 17 (1). ———. 2013. “Doing Business 2014.” Washington, DC. Kuddo, A. 2009. “Labor Laws in Eastern European and Central Asian Countries: Minimum Norms and Practices.” Social ———. 2014. “Doing Business 2015.” Washington, DC. Protection Discussion Paper 0920, World Bank, Washing- Young, D. 2003. “Employment Protection Legislation: Its ton, DC. Economic Impact and the Case for Reform.” Economic Papers 86, European Commission, Brussels. Annex—Chapter 1 Table 1.1: Probationary Period (months) by Income Group of Countries in 2013 Including: Total Number of 6 and More Income Group Countries* 2 Months and Less 3–5 Months Months High income  51  7 24 20 Upper middle income  49  9 25 15 Lower middle income  45 19 14 12 Low income  32  7 11 14 Total 177 42 74 61 *Data not available for 12 countries Source: World Bank 2013 BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 7 Figure 1.1: Incidence of Temporary Employment in Some OECD Countries and in Russia, 2014, % Chile Poland Spain Netherlands Korea Portugal Sweden Slovenia France Finland Italy Canada Switzerland Turkey Germany Greece OECD Hungary Ireland Austria Russia Slovakia Belgium Denmark Norway Japan UK Australia* Latvia Estonia 0 10 20 30 40 50 60 70 80 15–24 Total Source: http://www.oecd-ilibrary.org/employment/data/labour-market-statistics/employment-by-permanency-of-the-job-incidence_data-00297-en. 8 Figure 1.2: Economies in Which the Law Requires Figure 1.3: Economies in Which the Law Requires That Temporary Workers Enjoy Conditions Equivalent That Part-time Workers Enjoy Conditions Equivalent to Those of Comparable Full-time Workers, % to Those of Comparable Full-time Workers, % 73 71 88 Maternity leave 84 Maternity leave 83 83 93 91 76 77 84 88 Sick leave 85 Sick leave 93 84 88 76 77 Paid annual leave 84 75 83 Paid annual leave 91 82 84 70 Termination of employment 81 59 83 77 95 Termination of employment 74 68 0 20 40 60 80 100 0 20 40 60 80 100 Low income Upper middle income Lower middle income High income Low income Upper middle income Lower middle income High income Source: Kuddo and Robalino 2013 Source: Kuddo and Robalino 2013 BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 9 2. Minimum Wages 2.1. Overview Important dimensions of minimum wage regulations that will be discussed in this chapter are (a) objectives and ratio- nale; (b) labor market and social impact of minimum wages; Statutory minimum wage regulations were first introduced in and (c) design and implementation arrangements (setting New Zealand in 1894 and the Australian state of Victoria in the level, differentiation, adjusting minimum wages over 1896. Their main purpose was to protect workers who were time, institutional arrangements, and noncompliance and not covered by collective agreements and who were particu- enforcement). larly vulnerable to low pay conditions. Minimum wages exist in most of the countries in the world. By mid-2015, 160 econ- omies out of 189 had minimum wages for a regular worker.31 2.2. Objectives Minimum wage levels vary significantly by country. Accord- ing to Doing Business data, in absolute terms, in 2015 the low- and Rationale est minimum wages were reported in Uganda and Burundi The primary purpose of minimum wages is to provide protec- (US$2 per month)32 and the highest in Norway (US$3,722 per tion for wage earners against ‘unduly low wages’ (see Box 2 month), Luxembourg (US$3,062 per month), and Australia for wage concepts and definitions).35 In this context, mini- (US$2,397 per month).33 mum wages can contribute to reduce wage inequality and, In practices, some workers are excluded from the scope of the under certain conditions, reduce or prevent poverty. general minimum wage in some countries.34 Minimum wage Clearly, a minimum wage is only part of the answer to the rules typically do not cover the self-employed or (unpaid) problem of low wages. Wages may be low because labor family workers. In practice, they often are not enforced in the productivity is low. This may occur as a result of low capital informal economy although recent research indicates some intensity per worker, inefficient production technologies, upward impact on wages of informal workers when minimum poor management, and/or an unskilled labor force.36 Very wages are increased. Small firms are sometime excluded from low wages are often the result of labor market imperfections coverage of laws. such as asymmetrical bargaining power and monopsonistic labor markets. In addition, many of the poor and low-income workers in developing countries are self-employed or sub- sistence farmers. Thus, minimum wages should be looked as one instrument in a menu of policy options available to effect 31.  World Bank (2015). For assumptions about the regular worker, see World Bank (2013a, b). income distribution and poverty. In most cases several tools 32.  World Bank (2013a). Minimum wages in these countries have not will be necessary for optimal effects. been adjusted since 1984 and 1988, respectively. 33.  Minimum wages can be set on either an hourly, daily, weekly, or monthly basis, which complicates cross-country comparisons of their levels. ‘Doing Business’ recalculates them on a monthly basis. 35.  According to ILO Convention No. 131 (1970). 34.  ILO (2014). 36.  Del Carpio et al. (2010). 10 Box 2: Wage Concepts and Definitions According to the ILO Protection of Wages Convention No. 95 (1949), the term ‘wages’ means remuneration or earnings, however designated or calculated, capable of being expressed in terms of money, and fixed by mutual agreement or by national laws or regulations, which are payable in virtue of a written or unwritten contract of employment by an employer to an employed person for work done or to be done or for services rendered or to be rendered. The minimum wage is defined as “the minimum amount of remuneration that an employer is required to pay wage earn- ers for the work performed during a given period, which cannot be reduced by collective agreement or an individual contract.”38 Minimum wages fixed by collective agreements and made binding by public authorities are included in this definition. The Council of Europe’s Social Charter provides for the right to ‘a fair remuneration’, referring to ‘a remuneration such as will give the workers and their families a decent standard of living’. The Social Charter defines the decency threshold as 60 percent of net national average wage (that is, after deduction of taxes and social security contributions).39 The provision of a living wage should allow the workers and their families to lead a simple but decent life considered acceptable by society, in light of its level of development, among other things, to live above the poverty line and partici- pate in social and cultural life.40 The concept of fair wages encompasses 12 dimensions. In particular, wages must be paid regularly and in full to the workers; they must comply with the minimum wage regulations; they should not require excessive working hours; and they must lead to a balanced wage structure, free of discrimination.41 2.3. Labor Market United States, the most precise estimates were heavily clus- tered at or near zero employment effects.42 A meta-analysis and Social Impact of 16 U.K. studies found no significant adverse employment effect.43 In its reexamination of the job strategy, the OECD came to a similar conclusion, basing its policy advice on the of Minimum Wages considerable number of studies that have found that the adverse impact of minimum wages on employment is modest The labor market and social impacts of minimum wages or nonexistent.44 Recent studies using natural experiments or depend particularly on the level at which the minimum detailed payroll data also point toward no significant effects wages are set, how many workers are directly affected by this on employment and some positive benefits such as reduced measure (that is, earning at the minimum wage or slightly turnover and increased productivity.45 above), and how well they are enforced. Minimum wages Most studies address the effects of the minimum wage in that are set too high can, in principle, reduce employment advanced economies.46 Empirical evidence on developing levels. At the same time, minimum wages that are set too low countries and emerging market economies is more limited would fail to protect workers. 37 38 39 40 and the results are mixed. The effects of minimum wages on (1) Impact on employment: Although the range of estimates employment appear to be negative in some countries and from the literature varies considerably, the emerging trend is some years and positive in others. For example, the impact that the effects of minimum wages on employment are usu- ally small or insignificant (and in some cases positive).41 For example, based on a meta-study of 64 minimum wage stud- 42.  Doucouliagos and Stanley (2009). ies published between 1972 and 2007 and measuring the 43.  Leonard et al. (2013). impact of minimum wages on teenage employment in the 44.  OECD (2006). 45.  Card and Krueger (1995, 2000); Dube et al. (2010); Hirsch et al. (2011). 37.  ILO (2014). 46.  The most comprehensive, analytical, and unbiased assessment 38.  Council of Europe (2008). of the effects of minimum wage increases, based on a rigorous meta- analysis of more than 200 scholarly publications published since 1991, 39.  Anker (2011); Belser and Sobeck (2012). has been compiled by Belman and Wolfson (2014). Their findings indicate 40.  Vaughan-Whitehead (2010). that increases in the minimum wage result in very small reductions in 41.  See Betcherman (2014); Belman and Wolfson (2014). employment. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 11 appeared to be negative in Colombia up until the end of the depends not only on how high the minimum wage is set rela- 1990s and small or negligible in Costa Rica and Mexico in the tive to the rest of the distribution but also on compliance. In same period. In Brazil, empirical evidence shows a positive addition, minimum wages have tended to narrow the earn- effect of an increase of the minimum wage on employment, ings differentials across demographic groups. For example, in mainly resulting from changes in the composition between Brazil, wage compression in both the public and private sec- hours worked and the number of jobs.47 tors was found to be a result of minimum wage increases.57 It has been documented that wage compression in Costa Rica Evidence for new EU member states suggests that adverse was due to positive wage impacts being strongest for low- effects introduced by minimum wages, although present, are wage workers.58 limited.48 Similarly, empirical studies differ on the impacts of minimum wages on various groups of workers. Some studies Despite the fact that minimum wages are often not enforced found that in a few countries (such as Brazil, Chile, or Mex- in the informal sector, it seems to influence informal sector ico), employment decreased for vulnerable groups such as wage distribution. From the labor supply side, the minimum women, youth and low-skilled.49 Other studies could not vali- wage may be a benchmark for ‘fair’ wages. On the demand date this effect, for example, in Brazil from 1982 to 2000 or in side, employers may pay wages higher than the minimum Thailand from 2001 to 2011.50 and comparable to the formal sector market wage for a par- ticular occupation so that their employees will not leave for (2) Impact on worker flows and informal employment: a similar job in the formal sector, a kind of efficiency wage There is mixed evidence regarding the effects of minimum (so-called ‘lighthouse effect’).59 Alternatively, they may not wages on worker flows and informal employment. Some be willing to provide legislated labor benefits but instead studies find a negative impact of minimum wages on job pay the minimum wage. In some countries, increases in the retention for individuals at, or close to, the minimum wage.51 minimum wage often raise, rather than depress, wages in the However, some other studies have found no significant informal sector. This finding has been most notable in studies impact.52 There are also no firm conclusions that higher mini- of Latin American countries.60 mum wages can shift employment from the formal to the informal sector. The studies that have looked at this question (4) Impact on poverty: Earlier OECD analysis found a rela- do not yield consistent results. Some do find a decrease in for- tively weak link between low pay and poverty in advanced mal employment and an increase in informal employment.53 economies.61 This was explained by the observation that No effect of minimum wages on informality was identified in poverty is predominantly associated with nonemployment Costa Rica and Brazil; a stronger negative effect of increased (including unemployment) rather than with low wages. minimum wages on wage workers was found in the informal Many poor households have no household member working sector than in the formal sector.54 and some minimum wage workers live in households with above-average incomes. Still, most of the additional income (3) Impact on wage distribution: Minimum wages also from minimum wages accrues to families with relatively low affect the wage distribution and therefore average wages, income, and a sizeable proportion of minimum wage earners the extent of which depends on the level of the minimum are poor. wage as well as on possible spillover effects that lead to salary increases higher up the wage distribution (‘ripple A review of literature in 2012 found that while most studies effect’).55 For example, according to evidence for Latin Amer- identified a poverty-alleviating effect of minimum wages, ica, a 10 percent increase in minimum wages would entail an some studies discovered no effect.62 In the United States, increase in average wages of between 1 and 6 percent.56 Also 20  percent of low-wage workers live in families whose almost all studies find that minimum wages do lead to a com- income is below the poverty line and another 16 percent live pression of the earnings distribution and reduction in earn- in families whose income does not exceed 1.5 times the pov- ings inequality. How much earnings dispersion is reduced erty line.63 In European countries, household surveys show that minimum wage earners live in relatively larger house- holds with substantially lower incomes and are much more at 47.  Bell (1997); Maloney and Mendez (2003). 48.  Koettl et al. (2012). 49.  Montenegro and Pages (2003). 50.  Lemos (2007); DelCaprio et al. (2014). 57.  Lemos (2007). 51.  Abowd et al. (2005); Portugal and Cardoso (2006); Draca et al. 58.  Gindling and Terrell (1995); Fajnzylber (2001). (2008). 59.  Cunningham (2007). 52.  Zavodny (2000). 60.  Gindling and Terrell (2004); Lemos (2004); Fajnzylber (2001); 53.  Maloney and Mendez (2003); Jones (1997). Maloney and Mendes (2003). 54.  Gindling and Terrell (2007); Fajnzylber (2001). 61.  OECD (1998). 55.  Harris and Kearney (2014). 62.  Betcherman (2012). 56.  Cunningham (2007). 63.  CBO (2014). 12 risk of poverty.64 In India, about 30 percent of salaried work- in response to the higher labor costs from minimum wage ers and 40 percent of casual wage workers earning less than increases. An additional factor is that higher wages tend to state-level minimum wages live in poor households.65 reduce turnover among low-wage workers and thus lead to a workforce with more experience. Gindling noted that “whether raising minimum wages reduces poverty depends not only on whether formal sector workers lose jobs as a result but also on whether low-wage workers live in poor households, how widely minimum wages 2.4. Design and are enforced, how minimum wages affect informal workers, and whether social safety nets are in place.”66 For example, a Implementation 1 percent increase in minimum wages lowered the incidence of poverty by 0.12 percentage points in Nicaragua and by of Minimum Wage 0.22 percentage points in Honduras.67 Consequently, the link between minimum wages and poverty can be stronger or weaker depending on the country context. Policies Additional policies to mitigate poverty will typically be Setting the Level needed to complement minimum wages. In-work benefits, for example, might be a complementary policy tool intended In setting minimum wages, two important considerations to provide financial incentives for workers to take low-paid need to be made. First, economic factors, including the jobs by offering additional earnings, thereby reducing pov- requirements of economic development, levels of productiv- erty. They may take the form of tax credits, wage-related ity, and the desirability of attaining and maintaining a high transfers, or other lump-sum payments.68 Employment guar- level of employment; and second, the needs of workers and antee schemes may also have a large and positive impact on their families, taking into account the general level of wages poverty and can also influence whether statutory minimum in the country, the cost of living, and the relative living stan- wages reach the poorest households. dards of other social groups.71 In practice, the level of the min- imum wage depends largely on the prevailing social norms (5) Impact on productivity: Some studies suggest that a with regard to inequality and fairness, as well as on relative minimum wage and minimum wage increases could increase bargaining strengths of workers and employers. productivity because of higher investments in human capital. A study on productivity effects of wage costs, using cross- The level of minimum wages should be carefully considered. country aggregate data for 18 OECD member countries, esti- Too low a level reduces the relevance of minimum wages and mated that a 10 percentage point increase in the minimum the potential for this policy tool to address inequality and the wage-to-median wage ratio was associated with an increase living standards of the working poor; too high a level runs of between 1.7 to 2.0 percentage points in long-run labor pro- the risk of firms evading minimum wage legislation. Too high ductivity and multifactor productivity levels.69 Recent studies a minimum wage may have an adverse impact on employ- using natural experiments or disaggregated data sets find a ment or may push vulnerable workers such as low-skilled positive impact on productivity from increases in minimum workers, young people, and women out of employment or wages in China, the United States and the United Kingdom.70 into informal sector jobs, although as discussed above, the One possible reason for this effect is the substitution of higher- limited empirical evidence available for developing countries skilled labor for less-skilled labor due to decreased demand is mixed and generally weak on this point. for unskilled labor as minimum wages rise. To the extent that employers make this substitution, productivity levels will rise There is no fixed formula to set the minimum wage level but without any change in employment levels. Another possible it is important to consider its social and economic impacts. reason is that employers could make productivity-enhanc- Some indicators that can be used as a reference include ing adjustments—that is, increased investments in training (a)  minimum wages relative to median or average wages or new technologies or better management techniques— (see Annex—Chapter 2 Figure 2.1)72 and (b) share of work- ers affected by minimum wages (see Annex—Chapter 2 64.  Rycx and Kampelmann (2012). 65.  Belser and Rani (2011). 66.  Gindling (2014). 71.  The ILO Convention No. 131 and Recommendation No. 135 provide the principles for setting up minimum wages. 67.  Alaniz et al. (2011); Gindling and Terrell (2010). 72.  Median rather than mean earnings provide a better basis for 68.  ILO (2010). international comparisons as they account for differences in earnings 69.  Bassanini and Venn (2007). dispersion across countries. The median wage is the middle observation in 70.  Mayneris, F., S. Poncet and T. Zhang (2014); Schmitt, J. 2013; the distribution of wages whereas the mean wage is the sum of all wages Croucher and Rizov (2012). divided by the number of wage earners. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 13 Table 2.1). These indicators should be complemented by an With multiple minimum wages, however, there is a risk that sim- assessment of the needs of workers and their families, such as ilar workers (for example, with the same occupation and skills through the calculation of an adequate living wage. levels) would be subject to varying minimum wage rates. This would affect the principle of equal treatment. Therefore, the An appropriate level of the minimum wage also depends on general principle of equal remuneration for work of equal value a host of country-specific factors, such as labor market con- has to be observed while fixing minimum wage levels. Fur- ditions; other labor market policies; and variations in worker thermore, too many minimum wage levels make it difficult for productivity across regions, industries, and occupations. In workers and employers to determine which rates are applicable developed countries, minimum wages typically range from to them and might also affect the mobility of labor and capital. about 35 percent to 60 percent of median full-time wages, with some clustering around 45–50 percent.73 In develop- Multiple minimum wages also add administrative complex- ing countries, this ratio is often higher, not least because the ity and are only successful if an adequately robust system median worker is frequently a low-paid worker.74 Given that of oversight accompanies them. The ILO 2014a General Sur- median wages are often not readily available in developing vey on minimum wage systems points out that “the more countries, a second-best option has been to compare mini- complex a minimum wage system is, and the more sectoral, mum wages with mean wages rather than median wages. occupational and geographical rates it involves, the more dif- Mean wages typically exceed median wages, by an extent ficult it is to monitor, particularly in countries where the labor which depends on the level of inequality in a particular administration services have very limited resources.”79 country. The general trend around the world is to reduce the number Based on data for 75 countries, the minimum wage level is of differentiated minimum wage rates.80 Nonetheless, national most frequently set at around 40 percent of mean wages.75 minimum wages are sometimes modified for first-time job While these different ratios represent a useful reference seekers or for economically depressed regions with high point, it is important to disaggregate the analysis also con- unemployment. In countries with exceptions to the standard sidering the distribution of wages separately for women minimum wage to promote the employment of vulnerable and men and for different industries; and to observe where workers, it is important to establish precautionary measures a minimum wage lies in these respective distributions.76 It is to avoid abuse and to ensure respect for the principle of equal also useful to compare the level of the minimum wage to the remuneration for work of equal value. Substantial evidence wages of those workers who are legally covered, as opposed exists that employers may use such exceptions to lower labor to all wage workers, which may include various groups with costs without providing any net employment creation, for low wages who are not actually covered by the legislation.77 example, by hiring ever younger workers to replace cohorts For all these reasons, ex ante and ex post impact evaluations that age and move up the wage scale, without providing net of the level set should be conducted to assess their relevance. employment creation even for the targeted group.81 According to the Doing Business 2014 database, 42 countries Differentiation have youth minimum wages to encourage employers to hire this age group, given their lack of experience. A large number While minimum wages differentiated by segments of a coun- of countries have renounced maintaining reduced minimum try’s labor market are common, these can increase admin- wage rates for young workers, notably for reasons related to istrative complexity, create enforcement challenges, and nondiscrimination.82 lead to wage discrimination. Depending on the country, minimum wages are sometimes differentiated according to Instead of using age as a criterion, it might be preferable region, sector, age, qualifications, education, or type of job. to take account of the qualifications and experience of the International reviews suggest that minimum wages are gen- workers concerned.83 Persons on apprenticeship or trainee- erally applied uniformly at the national level (45 percent) or ship contracts should only be paid at a differentiated rate by sector or industry (43 percent). The minimum wages fixed where they receive actual training during working hours at for specific occupational categories or sectors may include the workplace. In general, the quantity and quality of the regional variations.78 work performed should be the decisive factors in determin- ing the wage paid. 73.  ILO (2014). 79.  ILO (2014). 74.  ILO (2013c); Rani et al. (2013). 80.  World Bank (2013b). 75.  Belser and Sobeck (2012); ILO (2009a). 81.  Salverda (2009); Allegretto et al. (2011); Blázquez Cuesta et al. 76.  ILO (2014). (2011); Dickens et al. (2010). 77.  ILO (2013c). 82.  ILO (2014). 78.  ILO (2013a, 2014). 83.  ILO (2014). 14 Adjusting Minimum Wages over Time that stop such an indexation when there are concerns about economic, productivity, and employment growth. Minimum wage rates should be adjusted from time to time to take account of changes in the cost of living and other eco- As discussed above, an important principle would be to nomic and social conditions. The main consideration should be accompany the discussions of the adjustment of the mini- to reduce uncertainty in frequency and levels of adjustment. mum wage with ex ante and ex post assessments of the The following criteria are used in different countries for the potential social and economic impacts. These assessments adjustment of minimum wage rates: the cost of living, which could be more thorough than what a simple formula can infer. guarantees workers a certain purchasing power; consumer price index; the types of workers’ needs which the minimum wage must satisfy but also those of their families; the general Institutional Arrangements level of wages in the country (the mean wage); the economic Independent of the authority chosen to ultimately decide situation of the country; requirements of economic develop- on the minimum wage, full consultations and participation ment; productivity; the level of employment; economic com- of social partners need to be ensured. The process should be petitiveness and the financial capacity of enterprises.84 A first transparent and all social partners should be on equal footing principle is to establish predetermined dates for minimum and able to make proposals on the minimum wage. wage reviews and revisions, for example, once a year.85 The revi- sion does not imply that the minimum wage will be adjusted The authority for the final decision and institutional arrange- but that a possible adjustment would be studied. The results of ments for fixing the minimum wage can vary. The minimum the assessment, the decision to adjust, and the time when the wage can be set up by statute, decision of the competent new minimum wage is effective (in case of a change) should authority, decision of wage boards or councils, industrial or also be made public at a predetermined date. Such measures labor courts, and tribunals.87 Alternatively, minimum wages increase the predictability of the process for firms and workers. may be set by giving the force of law to provisions of collec- tive agreements (Box 3). In case the minimum wage level is One possible mechanism to reduce uncertainty and help set fixed by the public authority, it is essential that the decision expectations regarding the level of the minimum wage is to responds to the recommendations of an independent techni- set up a simple formula to compute a reference value. Some of cal assessment and consultations with social partners. 88 the factors that could enter the formula include labor produc- tivity, cost of living, poverty lines, and unemployment rates. The formula is not used to define what the new level of the minimum wage should be. That would depend on consulta- Box 3: Minimum Wage Fixing Machinery tions or negotiations with the social partners and additional technical analysis. The purpose of the formula is to set an The minimum wage fixing machinery may take a vari- anchor to facilitate discussions among different stakeholders. ety of forms. Automatic adjustments to the minimum wage could be • 46 percent of countries have legislation that provides considered, in principle, but there is a risk that the resulting that the government agency (typically the labor level is not consistent with changing macroeconomic condi- ministry) sets the minimum wage after consultation tions and reduces the role of social partners in the adjust- with or upon recommendation of a specialized body ment process. In some countries, in fact, adjustment takes (bipartite or tripartite) or of the social partners. place following a decision by the relevant authority as mini- • In 10 percent of countries, the legislation permits mum wages are linked to factors such as the cost of living the government alone to set the minimum wage. (for example, Belgium, Haiti, Luxembourg, and Paraguay) or the mean wage (for example, Belarus, FYR Macedonia, Israel, • In 11 percent of countries, specialized bodies and Montenegro).86 These automatic adjustments allow (wages boards or councils) legally determine mini- safeguarding the purchasing power of minimum wages, mum wages. avoid lengthy political negotiations, and reduce uncertainty among workers and employers. There should be provisions • In 9 percent of countries, minimum wages are determined through collective bargaining. Some countries consult social partners separately or 84.  See ILO (2014) for details. have mixed wage-fixing mechanisms.89 85.  However, ILO standards do not require the establishment of predetermined dates for reviewing minimum wages. Recommendation No. 135 provides that “Minimum wage rates should be adjusted from time to time to take account of changes in the cost of living and other economic conditions.” 87.  ILO Minimum Wage Fixing Recommendation, 1970 (No. 135). 86.  ILO (2014). 88.  ILO (2013b). BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 15 To this end, some countries have set up independent expert over collectively bargained minimum wages is often mainly mechanisms (for example, United Kingdom and Chile) delegated to trade unions or the social partners in general that make recommendations on the level of the minimum and industrial/labor tribunals or courts. There is also a role for wage based on an analysis of economic and social impacts NGOs in monitoring the application of regulations and help- and consultations with social partners. Other countries (for ing workers understand the rights associated with a mini- example, Tunisia) have set up social councils with represen- mum wage system, for example, submitting complaints if tatives from government, workers, and employers that are their rights were compromised. responsible for recommending adjustments to the minimum wages. Such councils can commission independent technical Third, sanctions for noncompliance with minimum wage assessments. rules have to be imposed to correct abuses and serve as a deterrent. Measures should be taken to (a) ensure that the labor legislation provides for appropriate sanctions against Noncompliance and Enforcement the infringement of minimum wages and (b) reinforce the labor inspection services or other agencies so they are able Besides those workers who are not legally covered, some to effectively oversee compliance with the legislation. Advice, earn wages below the minimum because their employer fails public awareness campaigns, and stimulation or motiva- to comply with the legislation.89 This issue is particularly pro- tion of employers to implement the legal requirements may nounced in countries with high informality. While necessary, sometimes be an effective approach to improve compliance. enforcement itself may not be sufficient to address low com- pliance. Rather, a combination of measures should be taken. Regardless of where the minimum wage is set, there will be firms where the average labor productivity is below the mini- First, countries should avoid overly complex minimum wage mum cost of labor. For example, recent estimates by coun- systems, be transparent, and have a broad communication try authorities put these numbers at 30 percent of nonfarm regarding the minimum wage regulations and levels. Legal activities in Morocco and 35 percent in Tunisia. It is important provisions and procedures should be simplified, thereby that countries devise explicit mechanisms to deal with this enabling workers to effectively claim their rights under mini- problem. One option could be to allow these firms to go out mum wage provisions. Likewise, arrangements need to be of business in the interest of discouraging low-productivity, put in place to give publicity to minimum wage provisions in low-efficiency firms. However, depending on the concentra- a way that raises awareness among employers and workers. tion of such firms, the economic and social consequences Wherever it is possible and capacity is available, employers’ could be unacceptable. The second option could be to implic- and workers’ organizations should be encouraged to protect itly ignore the problem. This would reduce the credibility against abuses and ensure the effective implementation of and effectiveness of the minimum wage policy and respect the minimum wage provisions. for rule of law. The third option could be to provide tempo- rary exemptions, accompanied by measures that increase Second, inspections should be carried out frequently with the productivity of these firms or facilitate the transition of appropriate resources allocated to ensure that sufficient the owner and workers to other economic activities.90 Such and adequately trained inspectors are employed. These exemptions should remain exceptional and temporary and inspectors should have the powers and facilities required be subject to monitoring by the authorities as well as consul- for enforcing minimum wage provisions. In most countries, tations with the relevant trade unions. enforcement of laws on minimum wages is the responsibility of labor inspectorates or alternatively of specialized supervi- sory bodies of the relevant ministry, financial organs, bodies of the State Tax Inspectorate, the Procurator Office, industrial 2.5. Resources or employment tribunals, ombudsman, or labor offices. How- ILO Conventions, Recommendations and Data Sources ever, inspections are unlikely to be sufficient when there is a large informal sector. Minimum Wage Fixing Convention No. 131 (1970) Trade unions could be involved in monitoring the compliance Minimum Wage Fixing Recommendation No. 135, 1970 with minimum wage legislation albeit in many developing Protection of Wages Convention, 1949 (No. 95) countries, their capacity is limited. In high-income countries without a national statutory minimum wage, supervision 90.  The ILO Committee of Experts on the Application of Conventions and Recommendations, a body of independent legal experts, has 89.  Compliance can be measured in a number of ways: first, based on expressed its view that “allowing for deviations, however temporary, the complaints made by workers in a judicial system; second, based on from the established minimum wage runs counter to the very concept workplace inspection by labor inspectors; and third, by calculating the of a minimum wage which is meant to be the minimum sum payable, share of workers’ earning less than the legal minimum wage. guaranteed by law and not subject to abatement.” 16 Protection of Wages Recommendation, 1949 (No. 85) Advanced Countries, edited by David Blanchflower and Richard Freeman, 427–472. Chicago: University of Chi- Equal Remuneration Convention No. 100, 1951 cago Press. Equal Remuneration Recommendation No. 90, 1951 Abowd, J., F. Kramarz, D. Margolis, and T. Philippon. 2005. Discrimination (Employment and Occupation) Convention, “Minimum Wages and Employment in France and the 1958 (No. 111) United States.” Mimeo, CREST-INSEE, Paris. Discrimination (Employment and Occupation) Recommen- Adam, A., and T. Moutos. 2006. “Minimum Wages, Inequality dation, 1958 (No. 111) and Unemployment.” Economics Letters 92: 170–176. ILO minimum wage legislation database (around 150 coun- Alaniz, E., T. H. Gindling, and K. Terrell. 2011. “The Impact of tries); http://www.ilo.org/dyn/travail/travmain.home Minimum Wages on Wages, Work and Poverty in Nicara- gua.” Labour Economics 18 (S1). WB Doing Business database on minimum wages (189 coun- tries); http://www.doingbusiness.org/~/media/GIAWB/ Alatas, V., and L. Cameron. 2003. “The Impact of Mini- Doing%20Business/Documents/Annual-Reports/ mum Wages on Employment in a Low Income Coun- English/DB14-Full-Report.pdf try.” Policy Research Working Paper 2985, World Bank, Washington, DC. LivingWageIndicator.org (around 100 countries). Allegretto, S., A. Dube, and M. Reich. 2011. “Do Minimum http://www.wageindicator.org/main/salary/living-wage/ Wages Really Reduce Teen Employment? Accounting for living-wage-map Heterogeneity and Selectivity in State Panel Data.” Indus- trial Relations 50 (2): 205–240. ILO (2014). Minimum wage systems. International Labour Conference, 103rd session. Geneva. Anker, R. 2011. “Estimating a Living Wage: A Methodological Review.” Geneva: ILO. Bibliography Autor, D. H., A. Manning, and C. L. Smith. 2010. “The Contribu- tion of the Minimum Wage to U.S. Wage Inequality over Three Decades: A Reassessment.” NBER Working Paper Aaronson, D. 2001. “Price Pass-Through and the Minimum 16533. http://www.nber.org/papers/w16533. Wage.” Review of Economics and Statistics 83 (1). Bassanini, A., and D. Venn. 2007. “Assessing the Impact of Aaronson, D., and E. French. 2007. “Product Market Evidence Labour Market Policies on Productivity: A Difference-in- on the Employment Effects of the Minimum Wage.” Jour- Differences Approach.”Social, Employment, and Migra- nal of Labor Economics 25 (1): 167–200. tion Working Paper 54, OECD, Paris. Aaronson, D., E. French, and I. Sorkin. 2013. “Firm Dynamics Bell, L. 1997. “The Impact of Minimum Wages in Mexico and and the Minimum Wage: A Putty-Clay Approach.” FRB Colombia.” Journal of Labor Economics 15 (3). of Chicago Working Paper 2013–26. http://papers.ssrn .com/sol3/papers.cfm?abstract_id=2369447. Belman, D., and P. J. Wolfson. 2014. “What Does the Minimum Wage Do?” Kalamazoo, Michigan, U.S.A.: W.E. Upjohn Aaronson, D., E. French, and J. MacDonald. 2008. “The Mini- Institute Employment Research. mum Wage, Restaurant Prices, and Labor Market Struc- ture.” The Journal of Human Resources XLIII (3). Belser, P., and U. Rani. 2011. “Extending the Coverage of Mini- mum Wages in India: Simulations from Household Data.” Abbott, L. F. 2000. “Statutory Minimum Wage Controls: A Criti- Economic and Political Weekly XLVI (22), May 28. cal Review of their Effects on Labour Markets, Employment, and Incomes.” Manchester, UK: ISR Publications. Second Belser, P., and K. Sobeck. 2012. “At What Level Should Coun- edition. tries Set Their Minimum Wages?” International Journal of Labour Research 4 (1). Abowd, J., F. Kramarz, and D. Margolis. 1999. “Minimum Wages and Employment in France and the United States.” NBER Betcherman, G. 2012. “Labor Market Institutions: A Review of Working Paper 6996. the Literature.” Washington, DC: World Bank. Abowd, J., F. Kramarz, T. Lemieux, and D. Margolis. 2000. “Mini- ———. 2014. “Labor Market Regulations: What Do We Know mum Wages and Youth Employment in France and the about Their Impacts in Developing Countries.” The World United States.” In Youth Employment and Joblessness in Bank Research Observer. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 17 Blázquez Cuesta, M., R. Llorente Heras, and J. Moral Carcedo. Literature Review for the Low Pay Commission.” http:// 2011. “Minimum Wage and Youth Employment Rates, www.lowpay.gov.uk/lowpay/research/pdf/croucher 2000–2008.” Revista de Economon Aplicada 56. whitefinal.pdf. Boadway, R., and K. Cuff. 2001. “A Minimum Wage Can Be Wel- Croucher, R., and M. Rizov (2012). “The Impact of the National fare Improving and Employment Enhancing.” European Minimum Wage on Labour Productivity in Britain.” Economic Review 45: 553–576. E-Journal of International and Comparative Labour Stud- ies. Vol. 1, No. 3–4. In: http://adapt.it/EJCLS/index.php/ Boeri, T. 2009. “Setting the Minimum Wage.” IZA Working ejcls_adapt/article/view/43. Paper 4335. Cunningham, W. 2007. Minimum Wages and Social Policy: Les- Boeri, T., P. Garibaldi, and M. Ribeiro. 2010. “Behind the Light- sons from Developing Countries. Washington, DC: World house Effect.” IZA Discussion Paper 4890. Bank. Boockmann, B. 2010. “The Combined Employment Effects of Del Carpio, X., A. Kuddo, Y. Cho, D. Margolis, and D. Robalino. Minimum Wages and Labor Market Regulation: A Meta- 2010. Issues to Consider Before Implementing (Revising) the Analysis.” IZA Working Paper 4983. Minimum Wage Policy, Malaysia as a Country Case. World Bosch, M., and M. Manacorda. 2010. “Minimum Wages and Bank. Earnings Inequality in Urban Mexico.” CEPR Discussion Del Carpio, X., J. Messsina, and A. Sanz-de-Galdeano. 2014. Paper 7882. “Minimum Wage: Does It Improve Welfare in Thailand?” Brown, C. 1999. “Minimum Wages, Employment, and the IZA Discussion Paper 7911, Bonn. Distribution of Income.” In Handbook of Labor Econom- Dickens, R., and A., Manning. 2004a. “Spikes and Spill-overs: ics. Vol. 3B, edited by Orley Ashenfelter and David Card, The Impact of the National Minimum Wage on the Wage 2101–2163. Amsterdam: Elsevier. Distribution in a Low-wage Sector.” Economic Journal 114. Card, D., and A. Krueger. 1994. “Minimum Wages and Employ- ———. 2004b. “Has the National Minimum Wage Reduced ment: A Case Study of the Fast Food Industry in New Jer- Wage Inequality?” Journal of the Royal Statistical Society sey and Pennsylvania.” American Economic Review. Series C. ———. 1995. “Myth and Measurement: The New Economics Dickens, R., R. Riley, and D. Wilkinson. 2010. “The Impact on of the Minimum Wage.” Princeton: Princeton University Employment of the Age Related Increases in the National Press. Minimum Wage.” Report prepared for the Low Pay ———. 2000. “Minimum Wages and Employment: A Case Commission. Study of the Fast-Food Industry in New Jersey and Penn- Dolado, J. 2000. “The Role of the Minimum Wage in the Wel- sylvania: Reply.” American Economic Review 90 (5). fare State: An Appraisal.” IZA Discussion Paper Series 152. Cardoso, A. 2009. “Long-Term Impact of Youth Minimum Dolado, J., F. Kramarz, S. Machin, A. Manning, D. Margolis, and Wages: Evidence from Two Decades of Individual Longi- C. Teulings. 1996. “The Economic Impact of Minimum tudinal Data.” IZA Discussion Paper 4236. Wages in Europe.” Economic Policy October. Castillo-Freeman, A., and R. Freeman. 1992. “Minimum Wages Doucouliagos, H., and T. D. Stanley. 2009. “Publication in Puerto Rico: Textbook Case of a Wage Floor?” NBER Selection Bias in Minimum-Wage Research? A Meta- Working Paper 3759. Regression Analysis.” British Journal of Industrial Relations. CBO (Congressional Budget Office). 2014. “The Effects of a Draca, M., S. Machin, and J. Van Reenen. 2008. “Minimum Minimum-wage Increase on Employment and Family Wages and Firm Profitability.” NBER Working Paper 13996, Income.” Congressional Budget Office, Congress of the Cambridge, MA. United States. Draca, Mirko, Stephen Machin, and John Van Reenen. 2011. Council of Europe. 2008. “Digest of the Case Law of the “Minimum Wages and Firm Profitability.” American Eco- European Committee of Social Rights.” http://www nomic Journal: Applied Economics 3 (1): 129–151. .coe.int/t/dghl/monitoring/socialcharter/digest/ DigestSept2008_en.pdf. Dube, Lester, and Reich. 2010. “Minimum Wage Effects Across State Borders: Estimates Using Contiguous Counties.” The Croucher, R., and G. White. 2011. “The Impact of Minimum Review of Economics and Statistics 92 (4) November. Wages on the Youth Labour Market: An International 18 EC (European Commission). 2012. “Towards a Job-rich Recov- ———. 2012. “The Impact of Minimum Wages on Wages, ery.” Communication from the Commission to the Euro- Work and Poverty in Nicaragua.” http://ideas.repec.org/p/ pean Parliament, the Council, the European Economic umb/econwp/10126.html. and Social Committee and the Committee of the Regions. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri= Gottvald, J., J. Hantvald, and M. Pytlikova. 2002. “Minimum COM:2012:0173:FIN:EN:PDF. Wage and Its Impact on Wage Distribution, Unemploy- ment and Hours Worked.” In Determinants of Individual EIRO. 2005. “European Industrial Relations Review.” Pay and Firms Pay Structures in the Czech and Slovak Republics, edited by J. Gottvald et al. Ostrava, Vstrava. EIRO. 2013. “Developments in Collectively Agreed Pay 2012.” http://w w w.eurofound.europa.eu/eiro/studies/ Hamermesh, D. M. 1986. “Inflation and Labor Market Adjust- tn1303028s/tn1303028s_4.htm. ment.” NBER Working Papers 1153, National Bureau of Economic Research, Inc. Eyraud, F., and C. Saget. 2005. “The Fundamentals of Minimum Wage Fixing.” Geneva: International Labour Office. Harris, B. H., and M. S. Kearney. 2014. “The ‘Ripple Effect’ of a Minimum Wage Increase on American Workers.” Brook- Fajnzylber, P. 2001. “Minimum Wage Effects throughout the ings Brief. http://www.brookings.edu/blogs/up-front/ Wage Distribution: Evidence from Brazil’s Formal and posts/2014/01/10-ripple-effect-of-increasing-the- Informal Sectors.” Working Paper Series 151, Centro de minimum-wage-kearney-harris. Desenvolvimento e Planejamento Regional, Belo Hori- zonte, Brazil. Hirsch, B., B. Kaufman, and T. Zelenska. 2011. “Minimum Wage Channels of Adjustment.” IZA Discussion Papers 6132, Freeman, R. 1994. “Minimum Wages—Again!” International Institute for the Study of Labor (IZA). Journal of Manpower 15 (2). ILO. 2009a. “Global Wage Report 2008/09—Minimum Wages ———. 1996. “The Minimum Wage as a Redistributive Tool.” and Collective Bargaining: Towards Policy Coherence.” The Economic Journal: Policy Forum: Economic Aspects of Geneva. Minimum Wages (The Royal Economic Society) 106 (436). ———. 2009b. “Update of the First Global Wage Report with a Freeman, R. 1996. “The minimum wage as a redistributive Focus on Minimum Wage Policy Development”. Govern- tool.” Economic Journal. ing Body, 306th Session, November 2009. Geneva. Gindling, T. H. 2014. “Does Increasing the Minimum Wage ———. 2010. “Global Wage Report: Wage Policies in Times Reduce Poverty in Developing Countries?” IZA World of of Crisis.” http://www.ilo.org/wcmsp5/groups/public/ Labor. @dgreports/@dcomm/@publ/documents/publication/ Gindling, T. H., and D. Newhouse. 2012. “Self-Employment in wcms_145265.pdf. the Developing World.” Washington, DC: World Bank. ———. 2013a. “Global Wage Report 2012/13: Wages and Equi- Gindling, T.G. and K. Terrell. 1995. “The Nature of Minimum table Growth.” Geneva. Wages and Their Effectiveness as a Wage Floor in Costa ———. 2013b. “Working Conditions Laws Report 2012. A Global Rica, 1976–91.” World Development 23(8): 1439–1458. Review.” Geneva. ———. 2002. “The Effects of Minimum Wages on the For- ———. 2013c. “World of Work Report 2013: Repairing the Eco- mal and Informal Sector: Evidence from Costa Rica.” nomic and Social Fabric.” Geneva. processed. ———. 2014a. “Minimum Wage Systems.” International ———. 2004. “Legal Minimum Wages and the Wages of For- Labour Conference, 103rd session. Geneva. mal and Informal Sector Workers in Costa Rica.” William Davidson Institute Working Papers Series 2004-647, Wil- ———. 2014b. “Global Wage Report 2012/13: Wages and Equi- liam Davidson Institute at the University of Michigan. table Growth: Wages and Income Inequality.” Geneva. ———. 2007. “The Effects of Multiple Minimum Wages Jones, P. 1997. “The Impact of Minimum Wage Legislation in throughout the Labour Market: The Case of Costa Rica.” Developing Countries Where Coverage Is Incomplete.” Labour Economics 14 (3). Working Paper No. WPS/98-2. Centre for the Study of African Economies. Institute of Economics and Statistics. ———. 2010. “The Impact of Minimum Wages on Wages, University of Oxford. Work and Poverty in Nicaragua.”  UMBC Economics Department Working Papers 10-126, UMBC Department Kertesi, G., and J. Köllő. 2002. “Labor Demand with Hetero- of Economics. geneous Labor Inputs after the Transition in Hungary BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 19 1992–1999—and the Potential Consequences of the Mayneris, F., S. Poncet and T Zhang. 2014. “The cleansing Increase of Minimum Wage in 2001 and 2002.” Budapest effect of minimum wage: Minimum wage rules, firm Working Papers on the Labor Market 2002/5. dynamics and aggregate productivity in China.” CEPII Working No. Paper 2014-16. ———. 2003. “The Employment Effects of Nearly Doubling the Minimum Wage—The Case of Hungary.” BWP 2003/6, Montenegro, C. E., and C. Pagés. 2003. “Who Benefits from Institute of Economics, HAS, Budapest. Labor Market Regulations? Chile 1960–1998.” NBER Work- ing Paper 9850, National Bureau of Economic Research, Knabe, A., and Schoeb, R. 2008. “Minimum Wage Incidence: Cambridge, MA. The Case for Germany.” CESifo Working Paper 2432. Neumark, D. 2014. “Employment Effects of Minimum Wages: Koettl, J., T. Packard, and C. Montenegro. 2012. “From the When Minimum Wages Are Introduced or Raised, Are Shadow: Integrating Europe’s Informal Labor.” Washington, There Fewer Jobs? Global Evidence Says Yes.” IZA. http:// DC: World Bank. www.iza.org/wol/Neumark.pdf. Kohen, A. 1983. “The Effect of the Minimum Wage on Employ- Neumark, David, M. Schweitzer, and W. Wascher. 2000. “The ment and Unemployment.” Journal of Human Resources Effects of Minimum Wages throughout the Distribution.” 18 (1): 3–31. NBER Working Paper 7519. Kristensen, N., and W. Cunningham. 2006. “Do Minimum Neumark, D., and W. Wascher. 1999. “A Cross-National Analy- Wages in Latin America and the Caribbean Matter? Evi- sis of the Effects of Minimum Wages on Youth Employ- dence from 19 Countries.” Policy Research Working Paper ment.” NBER Working Paper 7299, Cambridge, MA. 3870, World Bank, Washington, DC. Neumark, D. and W. Wascher. 2007. “Minimum Wages and Lee, D. 1999. “Wage Inequality in the United States During Employment.” IZA Discussion Paper 2570. Bonn. the 1980s: Rising Dispersion or Falling Minimum Wage?” Quarterly Journal of Economics 114 (3): 977–1023. OECD (Organisation for Economic Co-operation and Devel- opment). 1998. “OECD Employment Outlook 1998.” Paris. Lee, D., and E. Saez. 2008. “Optimal Minimum Wage Policy in Competitive Labor Markets?” NBER 14320. ———. 2006. “OECD Employment Outlook 2006.” Paris. Lee, S. 2012. “Varieties of Minimum Wage Systems: Through ———. 2010. “OECD Employment Outlook: Moving Beyond the the Lens of Indicator-based Country Rankings.” Interna- Job Crisis.” Paris. tional Labour Review 51 (3). Portugal, P., and A. R. Cardoso. 2006. “Disentangling the Mini- Lemos, S. 2004. “The Effect of the Minimum Wage on Prices.” mum Wage Puzzle: An Analysis of Worker Accessions and IZA Discussion Paper 1072. Separations.” Journal of the European Economic Associa- tion 4 (5). ———. 2006. “Minimum Wage Effects in a Developing Country.” University College London, unpublished. Rama, M. 2001. “The Consequences of Doubling the Mini- mum Wage: The Case of Indonesia.” Industrial and Labor ———. 2007. “The Effects of the Minimum Wage in the Pri- Relations Review 54 (4): 864–881. vate and Public Sectors in Brazil.” Journal of Development Studies. Rani, U., P. Belser, M. Oelz, and S. Ranjbar. 2013. “Minimum Wage Coverage and Compliance in Developing Coun- Leonard, M., T. D. Stanley, and H. Doucouliagos. 2013. “Does tries.” International Labor Review 152 (3–4). the UK Minimum Wage Reduce Employment? A Meta- Regression Analysis.” British Journal of Industrial Relations. Rutkowski, J. 2003. “The Minimum Wage: Curse or Cure?” Wash- http://www.hendrix.edu/uploadedFiles/Departments_ ington, DC: World Bank. and_Programs/Business_and_Economics/AMAES/ UkMinWage%20final%20edits.pdf. Rycx, F., and S. Kampelmann. 2012. “Who Earns Minimum Wages in Europe: New Evidence Based on Household Sur- Maloney, W. F., and J. N. Mendez. 2003. “Measuring the Impact veys.” ETUI Report 124. of Minimum Wages: Evidence from Latin America.” NBER Working Paper 9800, National Bureau of Economic Saget, C. 2011. “Is the Minimum Wage an Effective Tool to Research, Cambridge, MA. Promote Decent Work and Reduce Poverty? The Experi- ence of Selected Developing Countries”. Employment Marx, I., S. Marchal, and B. Nolan. 2012. “Wage Workers in the Paper 2001/13, ILO. http://www.ilo.org/employment/ EU and the US.” Discussion Paper 6510, IZA. 20 Whatwedo/Publications/WCMS_142310/lang--en/index Kingdom/Northampton, Massachusetts, United States: .htm. Edward Elgar. Salverda, W. 2009. “The Netherlands: Minimum Wage Fall World Bank. 2012. “World Development Report: Jobs.” Wash- Shifts Focus to Part-time Jobs.” Working Paper 09/71, ington, DC: World Bank. Amsterdam Institute for Advanced Labour Studies, Uni- versity of Amsterdam. ———. 2013a. “Doing Business 2014.” Washington, DC: World Bank. Schmitt, J. 2013. “Why Does the Minimum Wage Have No Discernible Effect on Employment? Center for Economic ———. 2013b. “The Minimum Wage Policy: Lessons with a and Policy Research.” http://www.cepr.net/documents/ Focus on the ASEAN Region.” Washington, DC: World publications/min-wage-2013-02.pdf. Bank. Schulten, T. 2014. “Minimum Wage Regimes in Europe … and ———. 2015. “Doing Business 2016.” Washington, DC: World What Germany Can Learn from Them. Friedrich Ebert Bank. Shiftung.” http://library.fes.de/pdf-files/id-moe/10558 Zavodny, M. 2000. “The Effect of the Minimum Wage on .pdf. Employment and Hours.” Labour Economics 7 (6). Vaughan-Whitehead, D. 2010. “Fair Wages: Strengthening Corporate Social Responsibility.” Cheltenham, United Annex—Chapter 2 Table 2.1: Full-time Employees Earning the Minimum Wage in Some Countries, %; Latest Available Data Belgium 3.7 Luxembourg 11.2 Czech Republic 2.5 Latvia 18.0 Estonia 4.6 Netherlands 1.6 Spain 2.6 Poland 2.0 France 10.6 Portugal 8.7 Greece 20.4 Slovenia 2.8 Hungary 2.7–2.8 U.K. 4.3 Lithuania 7.0 U.S. 4.9 Source: Marx et al. (2012). BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 21 Figure 2.1: Ratio of Minimum Wages Relative to Median Wages in OECD Countries in 2000 and 2014 Slovenia Mexico Israel Ireland Chile Australia France Luxembourg Belgium Greece Netherlands Turkey New Zealand Portugal Lithuania Spain Slovakia Latvia Canada UK Poland Hungary USA Estonia Japan Czech R. Korea Romania 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 2014 2000 Source: http://www.oecd-ilibrary.org/employment/data/earnings/minimum-wages-relative-to-median-wages_data-00313-en. 22 3. Dismissal Procedures 3.1. Overview countries have legislation regulating the valid grounds for dismissals and the procedure to respect before terminating a worker’s employment contract. Labor markets are characterized by a continuous movement of workers between employment, unemployment, and inac- In case of collective dismissals for economic reasons, most tivity as well as between jobs. Job displacement represents a governments set requirements for advance notification, non-negligible proportion of job flows. In countries for which negotiation, or advance approval of mass terminations. Pre- data is available, between 2 and 7 percent of workers face ventive measures can avert or minimize the impact of dismiss- dismissals in a typical year.91 Compared with prime-age work- als. Examples include spreading the workforce reduction over ers, older and younger workers are at greater risk of dismissal. a certain period to permit natural reduction of the workforce, Others at higher risk are workers in small firms and those internal transfers, (re)training, voluntary early retirement employed on fixed-term and temporary contracts whose with appropriate income protection, restriction of overtime, contracts might not be renewed. reduction of normal hours of work, and work-sharing.94 Both preventive measures and dismissal procedures need to bal- National legislation typically establishes the scope of valid ance the interests of workers and employers and should pro- reasons for dismissals. International treaties recognize the vide for a process of social dialogue to ensure that workers employer’s right to dismiss a worker for a valid reason and are being protected while firms can effectively manage their also aim to guarantee a worker’s right not to be deprived human resources. of work unfairly.92 Countries generally legislate some form of dismissal protection but there is a range of regulatory This chapter will discuss main areas of interventions associ- approaches and the right to recourse to an impartial body in ated with dismissals for economic and other reasons of work- case of unfair dismissal. On one end of this range stands the ers with regular contracts, including (a) legislative provisions legal concept known as ‘at-will’ employment, which prevails setting conditions under which a dismissal is ‘justified’ or in the United States and a few other countries. It provides ‘fair’; (b) procedural requirements that the employer may face that an employer is free to terminate an employee at any when starting the dismissal process; (c) advance notice; and time, without notice and without cause, as long as it is not (d) special provisions governing collective redundancies.95 for an unlawful reason as specified in laws on discrimination While the focus of this chapter is on statutory provisions, or military service, for example. At the other end, some coun- tries limit the valid reasons to disciplinary grounds only, as in Bolivia and Venezuela.93 Apart from these examples, most 91.  OECD (2013). one of these just grounds, dismissal is not legally possible on the basis of 92.  See the ILO Termination of Employment Convention, 1982 (No. redundancy. In Venezuela, under the provisions of the new Organic Labor 158), and the ILO Termination of Employment Recommendation, 1982 Law (2012), work stability has been extended to protect nonexecutive (No. 166). See also the Final Report of the Tripartite Meeting of Experts to employees with more than one month of service against dismissals Examine the Termination of Employment Convention (No. 158) and the without cause. Termination of Employment Recommendation, 1982 (No. 166), Geneva, 94.  See the ILO Termination of Employment Recommendation, 1982 18–21 April 2011. (No. 166). The terms ‘termination of employment’ and ‘terminated’ 93.  In Bolivia, Decreto Supremo 28699 requires the worker’s consent describe the termination of employment at the initiative of the employer. for any contract termination without ‘just grounds’. As redundancy is not 95.  Termination payments will be discussed in Chapter 4. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 23 dismissal procedures can also be regulated through collec- legislation (EPL) reduces both the creation and destruction of tive agreements, arbitration awards, or court decisions.96 jobs in declining sectors relative to expanding sectors, imply- ing that faster growth attenuates the impact of firing costs on 3.2. Objectives firms’ hiring and firing decisions. In contrast to these findings, some other studies indicate that there are no statistically sig- and Impacts nificant effects of dismissal protection legislation on worker turnover.101 In their study, the authors investigated the effects of variable enforcement of German dismissal protection In unregulated labor markets, employers would be able to legislation on the employment dynamics in small establish- hire and dismiss workers ‘at-will’ in the absence of any kind ments. A recent review of the literature found that overall, of dismissal protection, which can create considerable uncer- “the findings on employment impacts (of EPL) are mixed, but tainty and economic insecurity for households. Also, depend- the results can also be characterized as fragile.”102 ing on their bargaining power, workers could be subject to abuse or discrimination. Dismissal procedures aim to (a) miti- Empirical results from OECD countries also suggest that more gate the adverse effects of any termination on the workers rigid dismissal regulations have a negative impact on pro- concerned and on their communities when dismissals affect ductivity growth in industries where layoff restrictions are large groups of workers at once and (b) protect workers from more likely to be binding.103 On the other hand, countries abuse and discrimination. In general, dismissal rules intend with more protective dismissal regulations tend to have more to provide protection against sudden loss of income and durable or stable jobs, contributing to investments in human unfair or discriminatory practices. The fundamental princi- capital, including training.104 ple behind their design is that the employment of a worker Overall, a meta-analysis of studies on the effects of EPL on should not be terminated without a valid reason97 and that employment finds that the impacts are relatively small and there should be scope for redress. can be either positive or negative.105 The World Development The evidence of the impact of dismissal procedures on labor Report 2013: Jobs concludes that there is a broad ‘plateau’ cov- market outcomes is mixed. For example, employers may be ering a range of policy choices available to countries based reluctant to hire if constraints inhibit future dismissals for rea- on their preferences. sons related to business. As a result, dismissal procedures may lengthen job tenure and reduce turnover of those who are already employed. This may, however, have a negative effect 3.3. Design and on new employment opportunities. In particular, a study on the Italian labor market found that the increase in dismissal Implementation costs decreased accessions and separations for workers in small firms relative to large firms.98 This effect is particularly This section discusses key aspects of the design and imple- strong in sectors with higher employment volatility but has a mentation of dismissal procedures, including the justifica- negligible impact on net employment. Another study found tion of dismissals, requirements for individual and collective that stringent employment protection had a sizeable nega- dismissals, and notice periods. The fundamental principle tive effect on labor market flows and, through this channel, underlying the following recommendations is that employers hindered productivity growth.99 At the same time, their evi- should be allowed to manage human resources to respond to dence also shows that while greater labor market reallocation the operational and business needs of the firm but that there benefits many workers through higher real wages and better is a need for safeguards to ensure that (a) there is a valid rea- careers, some displaced workers will have longer unemploy- son to terminate the employment contract; (b) employers do ment durations or lower real wages in post-displacement not discriminate against workers; (c) there is advance notice; jobs. The study on the impact of firing restrictions on job-flow and (d) income protection and activation measures are in dynamics across 14 European countries suggests that more place. Special provisions, however, need to be considered in stringent firing laws dampen the response of job destruction case of mass redundancies. to the cycle.100 Furthermore, stricter employment protection Valid reason for dismissal. The EPL should define ‘valid rea- sons’, ‘just causes’ or ‘serious reasons’ for the termination of an employment relationship and the sanctions applicable to the 96.  See Article 1 of the ILO Termination of Employment Convention, 1982 (No. 158). 97.  See Article 4 of the ILO Termination of Employment Convention, 101.  Bauer et al. (2007). 1982 (No. 158). 102.  Betcherman (2012); see also Howell et al. (2007); Glyn (2003). 98.  Kugler and Pica (2008). 103.  Bassanini et al. (2009). 99.  Martin and Scarpetta (2012). 104.  Young (2003). 100.  Messina and Vallanti (2007). 105.  Betcherman (2012, 2014). 24 employer in case of neglecting this principle.106 The employer just cause for termination related to conduct or performance may be required to substantiate the reasons justifying the may include reasons such as unpermitted absence from dismissal. The cases for justified dismissal can be based on work, theft or intentional damage to property, violence, or (a) worker-related grounds, such as worker conduct or worker imprisonment. A collective agreement may further limit capacity, but not for discriminatory reasons (for example the causes for dismissal or require a disciplinary process be based on age, gender, ethnicity, religion, trade union activ- exhausted before termination.110 In many countries, a trade ity, or maternity and educational leave) or (b) economic or union founder or leader may not be dismissed without the technological grounds, for example, in case demand for a prior permission of a court given the risk such a dismissal given product or service plummets or a new technology that would have on the exercise of freedom of association. increases productivity is adopted.107 Disguising the real reasons for dismissals or pushing work- An employer should be able to terminate an employment ers into ‘voluntary’ separation should be prohibited and contract for an open-ended or fixed-term contract, among sanctioned as it occurs. For example, separations into retire- others, for reasons such as (a) decline in economic activity; ment and disciplinary dismissals can be misused. Further- (b) the liquidation of the enterprise, agency, or other orga- more, adequate safeguards should be put in place to prevent nization; (c) the declaration of bankruptcy of the employer; recourse to short-term contracts of employment to circum- (d)  unsatisfactory results of a probationary period; and vent the protection. 111 112 (e) breach of the employee’s duties. The following, among others, should not constitute valid rea- sons for termination:108, 109 Box 4: Examples of Procedural (a) Union membership or participation in union activi- Requirements in Case of Contract ties outside working hours or, with the consent of the Termination in Line with the Labor employer, within working hours Law (for 187 Countries)112 (b) Seeking office as, or acting or having acted in the capac- • In 93 countries, the employer is obliged to notify or ity of, a workers’ representative consult with a third party before an employer can (c) Filing of a complaint or the participation in proceedings make a worker redundant. against an employer involving alleged violation of laws • In 34 countries, the employer additionally needs the or regulations or recourse to competent administrative approval of a third party before an employer can authorities make a worker redundant. (d) Race, color, sex, marital status, family responsibilities, • In 115 countries, the employer has to notify or con- pregnancy, religion, political opinion, nationality, or sult a third party in order to dismiss a group of 9 social origin redundant workers. (e) Absence from work during maternity leave • In 41 countries, a third party approval to dismiss a (f ) Temporary absence from work because of illness or injury group of 9 redundant workers is required. (g) Age, subject to national law and practice regarding • In 48 countries, retraining should be provided, retirement or there is a reassignment obligation before an employer can make a worker redundant. (h) Absence from work due to compulsory military service or other civic obligations, in accordance with national law • In 76 countries, there are priority rules applying to and practice redundancies. National law and practice as well as collective agreements • In 67 countries, the labor law stipulates restrictions may provide further protection against dismissals. In general, or obligations on rehiring.113 106.  The principles governing dismissal procedures are laid down in the ILO Termination of Employment Convention, 1982 (No. 158) and in the 110.  See ILO Convention No. 158 which provides, for example, that ILO Termination of Employment Recommendation, 1982 (No. 166). “the employment of a worker shall not be terminated for reasons 107.  For a list of substantive dismissal requirements in 95 countries of related to the worker’s conduct or performance before he is provided an the world see Annex—Chapter 3 Table 3.1. opportunity to defend himself against the allegations made.” 108.  ILO Termination of Employment Convention No. 158 (1982). 111.  These figures only concern redundancy dismissals. 109.  ILO Termination of Employment Recommendation No. 166 (1982). 112.  See Annex—Chapter 3 Figures 3.1 and 3.2. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 25 Employers should explore options to provide retraining or Many countries distinguish notice periods according to ten- reassign the worker before he or she can be made redundant. ure; some also have multiple advance notice periods differen- Retraining might open up new employment opportunities tiated according to professional and social criteria.117 for the worker. Collective dismissals. In view of the economic and social To avoid lengthy and potentially expensive administra- implications arising from the dismissal of many employees tive processes, countries may wish to provide for adequate within a short period and a specific geographical area, addi- advance notice requirements rather than obliging employ- tional requirements are often stipulated (see Box 5 on criteria ers to seek third-party approval before proceeding with dis- pertaining to collective redundancies). In case an employer missals (see below).113 However, a worker who claims that his is contemplating collective redundancies, consultations with employment has been unjustifiably terminated should be the workers’ representatives should begin in a reasonable entitled to appeal against that termination to an impartial period with focus on measures to mitigate the adverse effects body, such as a court, labor tribunal, arbitration committee, of any termination on the workers concerned.118 To enable or arbitrator. If the employment is terminated without a valid workers’ representatives to make constructive proposals, reason, the worker should have the right for adequate com- the employers should submit detailed information about pensation or other appropriate reliefs, including reinstate- the upcoming redundancies (Annex—Chapter 3 Tables  3.3 ment where appropriate. Ex post audits of dismissals could and 3.4). be enabled by law or collective agreement to ensure that the workers’ rights have not been compromised. In case of collective redundancies, employers or the govern- ment should prepare a program of measures (a social plan) Advance notice. All workers should be entitled to a reason- aimed at mitigating the impact of mass layoffs or provide able period of advance notice before termination.114 The reemployment guarantees to employees made redundant labor law should therefore require employers to give advance because of mass layoffs. When appropriate, early retirement notice before terminating workers, indicating the reason or provisions could be considered. reasons for termination. In the course of the consultations with workers’ representa- Advance notice is a means to give workers ample warning of tives, the employer should be required to inform the public future dismissals and thus facilitate job search. It is possible, authority in charge of employment intermediation. This noti- however, that notice varies with tenure and that a minimum fication should contain all relevant information in connection job tenure is required to be eligible for the notice. The period with the planned mass redundancies and the consultations of notice may be a cost factor for employers, as it could with employees’ representatives. In particular, it should con- involve a period of unproductive employment.115 To facilitate tain the reasons for the layoff, the number of employees job search during the notice period, workers should also be being made redundant, and the period within which the entitled to a reasonable amount of time off without loss of layoff should occur. Governments, in turn, should consider pay, for example, through paid leave of absence. Given the special actions to reinstate workers, such as intermediation, impact of mass layoffs on a large number of workers and on job-search assistance, or training. local communities, the notice period should be extended in case of mass redundancies. The law should define the criteria for collective redundancies, since these instigate special procedures (see Box 5). In about There are no general rules for the length of the advance notice 80 percent of the 125 countries with relevant legislation, the period which therefore should be country specific and dis- law provides a quantitative definition of collective dismissals cussed among social partners. Out of 187 countries for which for economic reasons and prescribes specific procedures of the data are available on the notice period for redundancy consultation with workers’ representatives, or notification of dismissal, 62 countries have a notice period of less than one public authorities.119 month (of which 26 countries do not have legally mandated notice period); 66 countries have a notice period of one to two months, and 59 countries have a notice period longer than two months (see also Annex—Chapter 3, Table  3.2).116 117.  For example, in the Lao People’s Democratic Republic, the notice is 30 days for manual workers and 45 days for skilled workers. In Angola, the notice period is 60 days for executives and medium- to high-skilled technical workers, and 30 days for other workers. In Austria, it is a fixed term of 2 weeks for blue-collar workers but for white-collar workers, it 113.  Approval is, as a rule, ensured when the reasons justifying the depends on the tenure. Also, in Madagascar, the notice period depends dismissal are substantiated (for example, it is not an absolute ban). on the length of service and professional group (it is different for laborers, 114.  ILO Convention No. 158. skilled trades, foreman, middle managers, or senior managers). 115.  EC (2012). 118.  See, for example, the Council Directive 98/59/EC of 20 July 1998 116.  Notice period for workers with 10 years of tenure. In some on the approximation of the laws of the Member States relating to countries, the notice period varies according to the type of dismissal collective redundancies; ILO Convention 158 and Recommendation 166. (economic versus noneconomic). 119.  Muller (2011). 26 ILO Employment Relationship Recommendation, 2006 Box 5: Collective Redundancies (No. 198) ILO Employment protection legislation database—EPLex According to EU legislation, ‘collective redundancies’ (http://www.ilo.org/dyn/eplex/termmain.home). means dismissals effected by an employer for one or more reasons not related to the individual workers WB Doing Business database on labor legislation (http:// concerned where, according to the choice of the mem- www.doingbusiness.org/). ber states, the number of redundancies is either, over a period of 30 days: • at least 10 workers in establishments normally Bibliography employing more than 20 and less than 100 workers; Aguirregabiria, V., and C. Alonso-Borrego. 1999. “Labor Con- tracts and Flexibility: Evidence from a Labor Market • at least 10 percent of the number of workers in Reform in Spain.” Working Paper 99-27, Universidad Car- establishments normally employing at least 100 but los III de Madrid. less than 300 workers; Autor, D. 2003. “Outsourcing at Will: The Contribution of • at least 30 workers in establishments normally Unjust Dismissal Doctrine to the Growth of Employment employing 300 workers or more; or Outsourcing.” Journal of Labor Economics 21 (1). • over a period of 90 days, at least 20 workers, what- Autor, D., J. Donohue, and J. Schwab. 2002. “The Costs of ever the number of workers normally employed in Wrongful Discharge Laws.” NBER Working Paper 9425. the establishments in question.121 Bassanini, A., and R. Duval. 2006. “Employment Patterns in OECD 120 Countries: Reassessing the Role of Policies and Institutions.” Economics Department Working Paper 486, OECD. Lastly, workers undergoing dismissals, individual or collec- tive, should have access to income protection programs for Bassanini, A., L. Nunziata, and D. Venn. 2009. “Job Protection the unemployed (to be discussed in Chapter 4). It is desir- and Productivity.” Economic Policy 24 (58). able that these workers also have access to job-training and job-search assistance, or what is commonly known as active Bauer, T., S. Bender, and H. Bonin. 2007. “Dismissal Protection labor market programs based on activation principles.121 and Worker Flows in Small Establishments.” Economica 74 Active labor market programs, such as training, employment (296). services, wage subsidies, and public works can help facilitate Betcherman, G. 2012. “Labor Market Institutions. A Review job matching, mitigate the negative impacts of economic of the Literature.” World Bank Policy Research Working downturns, and fill the gap when employers or workers Paper 6276. underinvest in training as long as they are well designed and implemented.122 ———. 2014. “Labor Market Regulations What Do We Know about Their Impacts in Developing Countries?” Policy 3.4. Resources Research Working Paper 6819, World Bank. Boeri, T., B. Helppie, and M. Macis. 2008. “Labor Regulations ILO Termination of Employment Convention, 1982 (No. 158) in Developing Countries: A Review of the Evidence and Directions for Future Research.” Social Protection Discus- Termination of Employment Recommendation, 1982 ILO  sion Paper 0833, World Bank, Washington, DC. (No.˛166) Boeri, T., and J. van Ours. 2008. The Economics of Imperfect Labor Markets. Princeton University Press. http://onlinelibrary 120.  The EU Council Directive 98/59/EC of July 20, 1998 on the .wiley.com/doi/10.1111/j.1468-0335.2006.00562.x/full. approximation of the laws of the member states relating to collective redundancies. Cazes, S., S. Khatiwada, and M. Malo. 2012. “Employment Pro- 121.  The essence of activation is in many countries increasingly based tection and Collective Bargaining: Beyond the Deregula- on the principle of ‘mutual obligations’. The principle states that, in tion Agenda.” Employment Working Paper 133, ILO. return for receiving income support (unemployment benefits and other related entitlements or social safety nets) and being offered a range of (re)employment services, individuals must commit and comply with a set Djankov, S., and R. Ramalho. 2009. “Employment Laws in of eligibility requirements (for instance, active job search behavior) and Developing Countries.” Journal of Comparative Economics participate in training or other (re)employment programs. 37 (1). 122.  World Bank (2012). BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 27 Eichhorst, W., and R. Konle-Seidl. 2006. “The Interaction of Martins, P. S. 2009. “Dismissals for Cause: The Difference Labor Market Regulation and Labor Market Policies in That Just Eight Paragraphs Can Make.” Journal of Welfare State Reform.” Comparative Labor Law and Policy Labor Economics 27 (2) April. http://www.jstor.org/ Journal 28 (1). stable/10.1086/599978. EC (European Commission). 2012. “Labour Market Develop- Messina, J., and G. Vallanti. 2007. “Job Flow Dynamics and ments in Europe, 2012.” Brussels. Firing Restrictions: Evidence from Europe.” The Eco- nomic Journal 117 (521). http://onlinelibrary.wiley.com/ Glyn, A. 2003. “Labor Market Institutions and Unemployment: doi/10.1111/j.1468-0297.2007.02057.x/full. A Critical Assessment of the Cross-Country Evidence.” Economics Series Working Papers 168, University of Micco, A., and C. Pages. 2004. “Employment Protection and Oxford, Department of Economics. Gross Job Flows: A Differences-in-Differences Approach.” IDB Working Paper 421. http://papers.ssrn.com/sol3/ Heckman, J., and C. Pages. 2004. “Law and Employment: Les- papers.cfm?abstract_id=1818710. sons from Latin America and the Caribbean.” Chicago: The University of Chicago Press. Moser, C., D. Urban, and B. Weder di Mauro. 2010. “International Competitiveness, Job Creation and Job Howell, D. R., D. Baker, A. Glyn, and J. Smitt. 2007. “Are Protec- Destruction—An Establishment-level Study of German tive Labor Market Institutions at the Root of Unemploy- Job Flows.” Journal of International Economics 80 (2) ment? A Critical Review of the Evidence.” Capitalism and March. http://www.jstor.org/stable/10.1086/383107. Society 2 (1). Muller, A. 2011. “Employment Protection Legislation Tested Kahn, L. M. 2010. “Employment Protection Reforms, Employ- by the Economic Crisis: A Global Review of the Regula- ment and the Incidence of Temporary Jobs in Europe: tion of Collective Dismissals for Economic Reasons.” Dia- 1996–2001.” Labour Economics 17 (1). logue in Brief 3, ILO, Geneva. Kaplan, D. S. 2009. “Job Creation and Labor Reform in Latin OECD (Organisation for Economic Co-operation and Devel- America.” Journal of Comparative Economics 37 (1) March. opment). 2013. OECD Employment Outlook 2013. Paris: http://www.sciencedirect.com/science/article/pii/ OECD. S0147596708000760. Parsons, D. 2011. “Mandated Severance Pay and Firing Cost Kuddo, A. 2009. “Labor Laws in Eastern European and Central Distortions: A Critical Review of the Evidence.” George Asian Countries: Minimum Norms and Practices.” Social Washington University and IZA Discussion Paper 5776. Protection Discussion Paper 0920, World Bank, Washing- ton, DC. Pissarides, C. 2001. “Employment Protection.” Labour Econom- ics 8. Kugler, Adriana, and Gilles Saint-Paul. 2004. “How Do Firing Costs Affect Worker Flows in a World with Adverse Selec- Skedinger, P., ed. 2010. “Employment Protection Legislation: tion?” Journal of Labor Economics 22 (3): 553–584. Evolution, Effects, Winners and Losers.” Northampton, MA: Edward Elgar Publishing. Kugler, A., and G. Pica. 2008. “Effects of Employment Protec- tion on Worker and Job Flows: Evidence from the 1990 Wolfers, J. 2002. “Measuring the Effects of Employment Pro- Italian Reform.” Labour Economics 15. tection on Job Flows: Evidence from Seasonal Cycles.” Labour Economics 9 (1). http://www.sciencedirect.com/ Kugler, Adriana, Juan F. Jimeno, and Virginia Hernanz. 2003. science/article/pii/S0927537101000513. “Employment Consequences of Restrictive Perma- nent Contracts: Evidence from Spanish Labour Market World Bank. 2012. “World Development Report 2013: Jobs.” Reforms.” CEPR Working Paper 3724. Washington, DC: World Bank. Martin, J. P., and S. Scarpetta. 2012. “Setting It Right: Employ- World Bank. 2014. “Doing Business 2015.” Washington, DC: ment Protection, Labour Reallocation and Productivity.” World Bank. De Economist 160. http://download.springer.com/static/ pdf/578/art%253A10.1007%252Fs10645-011-9177-2 Young, D. 2003. “Employment Protection Legislation: Its Eco- .pdf?auth66=1399492544_57c77b58d1984a96c5ad9d6 nomic Impact and the Case for Reform.” Economic Papers 81c6be173&ext=.pdf. 186, European Commission, Brussels. 28 Annex—Chapter 3 Table 3.1: Substantive Dismissal Requirements Obligation to Provide Reasons to the Employee Valid Grounds (Justified Dismissal) Total Economic Reasons; Worker’s Any Fair Countries No Yes Capacity; Worker’s Conduct Reasons None Africa  5 21 19  4  3 26 Americas  4  9  3  7  3 13 Arab States  4  1  2  1  2 5 Asia  7 10  6  6  5 17 Europe  5 29 12 16  6 34 Total 25 70 42 34 19 95 Source: ILO EPLex at http://www.ilo.org/dyn/eplex/termmain.home. Table 3.2: Notice Period for Redundancy Dismissal* Average Notice Period Total Number Less than 8.7 Weeks for Workers with 1, 5 and Income Group of Countries 4.3 Weeks 4.3–8.6 Weeks and More 10 Years of Tenure, in Weeks High Income  56 18 23 15 5.7 Upper Middle Income  50 19 26  5 4.1 Lower Middle Income  47 11 30  6 4.8 Low Income  34  5 19 10 6.4 Total 187 53 98 36 5.2 *Note: Average for Workers with 1, 5, and 10 Years of Tenure, in Salary Weeks. Source: Doing Business 2014. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 29 Table 3.3: Procedural Requirements in Case of Collective Dismissals (No. of Countries) Prior Consultations with Trade Notification to the Public Notification to Trade Union Unions (Workers’ Representatives) Administration (Workers’ Representatives) No Yes No Yes No Yes Africa  4 22  5 21  1 25 Americas  8  5  4  9  7  6 Arab States  3  2  2  3  3  2 Asia  8  9  4 13  7 10 Europe  4 30  1 33  1 33 Total 27 68 16 79 19 76 Source: ILO EPLex at: http://www.ilo.org/dyn/eplex/termmain.home. Table 3.4: Procedural Requirements in Case of Collective Dismissals (No. of Countries) Approval by Public Approval by Trade Union Employer’s Obligation to Consider Administration or Judicial Bodies (Workers’ Representatives) Alternatives to Dismissal No Yes No Yes No Yes Africa 21  5 26 0 11 15 Americas 10  3 13 0 10  3 Arab States  3  2  5 0  4  1 Asia 14  3 15 2  9  8 Europe 31  3 34 0  2 32 Total 79 16 93 2 36 59 Source: ILO EPLex at http://www.ilo.org/dyn/eplex/termmain.home. 30 Figure 3.1: Share of Countries in Which a Third- Figure 3.2: Share of Countries with Priority Rules Party Notification and Approval Is Required If One for Redundancies, Retraining Obligation and Priority or Nine Workers Are Dismissed, 2013 (%) Rules for Reemployment in 2013 (%) 22 36 62 Total 26 Total 18 41 50 32 7 High income 36 High income 45 39 5 34 32 Upper middle income 20 20 28 Upper middle income 52 20 44 30 Lower middle income 23 38 36 Lower middle income 72 32 56 60 Low income 15 65 29 88 0 20 40 60 80 Low income 18 71 Priority rules for reemployment Retraining obligation 0 20 40 60 80 100 Priority rules for redundancies Third-party approval if 9 workers are dismissed Source: Doing Business 2014. Third-party notification if 9 workers are dismissed Third-party approval if 1 worker is dismissed Third-party notification if 1 worker is dismissed Source: Doing Business 2014. Note: Doing Business data are based on the survey of law firms and/or labor lawyers in the country and to the best of our knowledge should represent relevant provisions in labor legislation. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 31 4. Income Protection for the Unemployed: Severance Pay and Unemployment Benefit Schemes 4.1. Overview worker with 10 years of tenure, in salary weeks) was in Sierra Leone, 132 weeks’ salary; followed by Zimbabwe, 130; Sri Lanka, 97.5 and Indonesia, 95.3 weeks’ salary.124 Such a con- Most countries offer some form of income protection to tingent liability might be especially damaging to small and workers in case of dismissals or when individuals leave their medium size enterprises (SME). At the other end, 42 countries jobs. These arrangements are designed to provide a certain do not have a statutory severance pay. Some of these coun- degree of financial security to workers and their households tries may still have severance pay fixed in collective agree- in case of unemployment. Unemployment benefit schemes ments or in an individual employment contract.125 guarantee income protection to unemployed workers, usu- ally combined with measures that support jobseekers in There is a consistent relationship between national income returning to work. Severance pay is the most prevalent form and mandated severance arrangements; poor countries of protection in case of unemployment in most low-income impose firing costs 50 percent higher relative to national economies and in middle-income economies that have not per capita income than rich countries.126 Also, the absence implemented unemployment benefit schemes yet. of unemployment benefits in low- and middle-income coun- tries leads to protection through mandated severance pay It is important to note that severance pay and unemployment and dismissal rules. Globally, countries without an unem- benefits are very different in their nature as discussed in more ployment benefit scheme have, on average, 40 percent detail in the section on objectives. Severance payments typi- higher severance pay than the countries with such scheme cally provide lump sums to workers who involuntarily or volun- (Annex—Chapter 4 Table 4.1). tarily separate from their employers.123 The lump sum is usually based on the worker’s earnings and the length of service pre- Unemployment benefits are less common than severance ceding the dismissal. Unemployment insurance systems also pay mechanisms. Out of a total of 189 economies, only 86 present the advantage of assisting job seekers in finding have unemployment benefit schemes.127 The development employment through the, often mandatory, condition of peri- of government-organized protection against unemployment odically reporting to the employment services for continuing came after the introduction of other government-organized to qualify for unemployment benefits and through facilitating social protection benefits, for example, employment injury or the participation of unemployed workers in skills development sickness benefits in many countries. The first unemployment and active labor market policy (ALMP) measures. Country arrangements for severance pay are diverse. In 2015, the highest severance pay for redundancy dismissal (for a 124.  World Bank (2015). 125.  This chapter will only discuss statutory severance pay schemes in case of fair economic individual dismissals. As noted by Boeri et al. (2013), “although the definition of fair economic dismissal differs quite 123.  Severance pay includes several types of termination (separation) considerably from country to country, it generally implies that some payments whose typology depends on the reasons for dismissal and can ‘genuine and serious’ exogenous shocks in firm’s performance require be classified into (a) severance payments, which arise from terminating ‘operational changes’ in the scale, and possibly, nature of the work an employment relationship on worker-related grounds, such as worker organization, making the worker involved redundant. Often evidence of conduct or capacity and (b) redundancy payments that arise when ‘economic difficulties’ or ‘technological change’ is explicitly required.” terminating an employment relationship on economic grounds. Termination payments arise mainly when separation is initiated by the employer. 126.  Holzmann et al. (2011). Some countries do not specify the difference between redundancy and 127.  World Bank (2013). According to the ILO World Social Protection severance payments and bundle them into a single severance pay provision Report 2014–15, there are 85 countries with unemployment benefit but usually provide specific reasons for the payments, including possibly schemes providing periodic benefits and an additional 4 with lump-sum different schedules depending on the type of separation. payments (UISA or provident fund; ILO 2014). 32 benefit schemes were adopted at the beginning of the 20th The primary objective of severance pay is to provide income century and are now more than a century old. Nonetheless, protection in case of loss of employment depending, in most coverage is limited: due to the structure of the labor market cases, on job tenure. In certain cases, severance pay may and the predominance of the informal economy in many be part of the compensation package upon termination of countries, even where unemployment benefit schemes are employment at retirement, which may be considered as a in place, many of the unemployed are not eligible for unem- substitute to occupational pensions. Other objectives may ployment benefits. This may be the case for long-term unem- include the stabilization of employment and therefore pre- ployed who have exhausted their entitlement or for first-time vention of unemployment by discouraging redundancies jobseekers or workers who do not have a sufficient insurance as employers may consider it as additional cost. It can also period to claim contributory benefits and yet might be eligible promote longer-term relationships by making redundancies for noncontributory benefits. Around the world, 12 percent of more costly and thereby retaining (valuable) workers and the unemployed are covered by unemployment benefits. It is reducing transaction costs resulting from labor turnover. A noteworthy that 64 percent of unemployed persons receive justification to mandate payments from employers in case unemployment benefits in Western Europe compared to only of dismissals is to internalize the social costs of unemploy- one percent of the unemployed in Africa and two percent in ment.131 In this case, however, payments do not have to be the Arab states (Annex—Chapter 4 Table 4.2).128 made to employees directly but could take the form of a tax for the employer that, for instance, funds collectively- Several developing countries are considering reforms to their managed severance payments as well as passive and active income protection regulations in case of contract termination labor programs. to balance the needs of firms and workers. One of the prob- lems of severance pay is that, in general, it does not offer ade- The primary objective of unemployment benefit schemes, quate protection to workers. Indeed, employers seldom make as an integral part of the social protection system, is to pro- provisions for benefits and those who are facing economic vide adequate income protection to the unemployed while difficulties and need to downsize might not be able to finance also serving employment policy goals. Such schemes aim to the payments. In addition, the process to effect payments can (a) guarantee an appropriate level of income during periods often involve courts, and the judicial process in those courts of unemployment; (b) provide better protection of workers at tends to be lengthy and costly. Discussions about reform a relatively low cost, involving risk pooling and redistribution; options then revolve around how to improve the system and and (c) integrate unemployment payments with active labor whether to introduce an unemployment benefit scheme.129 market policies that facilitate effective job search and match- ing, additional training, or other assistance. Beneficiaries are This chapter discusses the rationale for having in place sev- typically required to look actively for work and/or participate erance pay and unemployment benefits, their potential in a training program to promote their job prospects. impacts on labor markets and workers’ protection, as well as issues related to design and implementation. To better protect workers from the risk of unemployment and internalize the social costs of unemployment, countries can 4.2. Objectives consider combining both unemployment benefits and sever- ance pay. Unemployment benefits are a better structure to and Rationale protect workers’ income because financial risks are pooled across participants (instead of being a liability of employers). At the same time, employers could be required to pay sev- Severance pay and unemployment benefits provide some erance or a dismissal tax that could replace severance pay, degree of income protection to workers who have lost their provided that adequate protection is available through an jobs and facilitate transitions between jobs.130 There are, unemployment benefit scheme. The revenues from this dis- however, important conceptual differences between the two missal tax could contribute to finance redistribution within schemes. Severance pay is regulated by the labor law and the UB system.132 based on the concept of employer liability whereas unem- ployment benefits are financed collectively and payments are secured regardless of the situation of employers. 131.  Blanchard (2000). 132.  For example, Baumann and Stähler (2006) note: “Financing 128.  ILO (2014). unemployment benefits through employment taxes leads to externalities 129.  The ILO Convention No. 158 provides for either of these two as firms that are firing workers create additional costs to the options—severance pay or unemployment benefit scheme—or a unemployment insurance that they do not have to bear. To internalize combination thereof as a vehicle for income protection for workers these externalities, firms can be made liable for their behavior by whose employment has been terminated. changing the financing system towards a scheme with dismissal taxes.” 130.  For severance payments, this is largely limited to financial support, See also Blanchard and Tirole (2010) and Cahuc and Zylberberg (2005). yet there usually is a lack of access to ALMPs and public employment If a dismissal tax is introduced, countries may choose not to levy it in case services. of dismissals for economic reasons. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 33 4.3. Labor Market thereby increasing the productivity of workers as well as their future employability.139 Impacts As far as impacts of unemployment benefit schemes are con- cerned, some studies suggest that higher benefit levels and This section reviews the evidence of the impact of severance longer benefit duration tend to be associated with longer pay, unemployment insurance, and unemployment individ- spells of unemployment and can, among other things, lead to ual savings accounts on labor market outcomes.133 Although a decline in the intensity of job searches.140 On the other hand, the evidence comes from developed economies, an effort there is also evidence that beneficiaries with longer benefit has been made to highlight the experience from developing duration are able to find jobs at higher wages and longer countries. tenure, thus allowing workers to find better-matching jobs.141 A recent study of the European Commission concludes that The evidence on the impact of severance pay systems on “the generosity of income support does not prevent returns labor market outcomes and its effectiveness as an unem- to employment” and that “benefit systems integrated with ployment support system are mixed. In the upper-middle to inclusive labor markets and enabling services facilitate the high-income countries, which usually also feature unemploy- returns to employment.”142 ment insurance schemes as part of their social protection systems, most of the studies found that the impact is seem- Some studies reveal a welfare-increasing income effect ingly insignificant. For example, a review of the empirical lit- of unemployment benefits, which is particularly impor- erature concludes that “severance pay, unaccompanied by tant in the presence of inefficient private insurance mar- other labor regulations, has little impact on worker separa- kets and high-risk aversion. The income effect is significant tion (and accession) or average employment level.”134 Differ- for liquidity-constrained households since unemployment ences in labor market impacts exist between severance pay benefits maintain a certain level of income, which increases due to dismissals only (also called severance insurance pay the opportunity for consumption while unemployed. This, or indemnity for dismissal) and severance pay due to any in turn, reduces the need for job searches, which leads to a separation (also called severance savings or seniority pay).135 substitution effect. If workers are unconstrained, the income The roles of the two schemes are different, as are their effects effect does not exist. The substitution effect and the income on the labor market; for example, there is a positive effect effect have contrasting welfare implications. 143 of seniority pay on employment.136 The latter in particular The evidence for developing countries is more limited. Some behaves like deferred earnings, and is often based on accu- early studies for Brazil showed little effect of unemploy- mulated funds, which also alleviates some of the issues linked ment benefits on unemployment rates. If anything, benefits to nonpayment. increased the share of self-employment over wage employ- In contrast to these results, some researchers argue that sev- ment. Cross-country studies including developing countries erance pay increases firing costs and thus reduces the prob- have found negative and not very robust effects of unemploy- ability of exit from employment to unemployment, but at the ment benefits on informality.144 Some authors argue on the same time, it imposes additional costs on employers and thus basis of a theoretical model calibrated to the Mexican case hinders job creation.137 Some studies show that high firing that well-designed unemployment benefits can increase for- costs slow the pace of structural change by reducing employ- mal employment by making it more attractive than informal ers’ incentives to introduce new technologies.138 According to work.145 However, recent analysis for Mexico suggests that some authors, these costs likely reduce productivity growth the main effect of unemployment insurance was to reduce and overall economic growth. Regarding the gains, severance the share of formal employment. In countries like Mexico, the pay may promote longer-lasting employment relationships trade-off between informal work and unemployment seems and thus improve employers’ incentives to provide training, to be driven by the size of the tax wedge. The higher the tax wedge, the higher the effect on unemployment.146 133.  A system of Unemployment Individual Saving Accounts (UISAs) has been proposed as an alternative to the traditional unemployment insurance system in which individuals are required to save a percentage of wages in special accounts and to draw unemployment compensation 139.  OECD (1999). from these accounts instead of taking state unemployment insurance benefits. If the accounts are exhausted, mechanisms may be in place to 140.  EC (2006); Arias et al. (2014). provide some level of protection, for example, through a solidarity fund. 141.  Caliento et al. (2013); Tatsiramos (2009). 134.  Parsons (2011a, b). 142.  EC (2013). 135.  Parsons (2011a, b). 143.  Chetty (2004, 2008); Chetty and Looney (2006). 136.  Heckman and Pagès (2004). 144.  Lehmann and Muravyev (2011). 137.  Blanchard (2000); Kaplan (2009). 145.  Bosch and Esteban-Pretel (2013). 138.  Calmfors and Holmlund (2000). 146.  Ribe et al. (2012). 34 Moreover, in developing countries, where the informal economy still predominates, the presence of unemployment 4.4. Design & Implementation benefits prevents formal laid-off employees from growing the ranks of the informal economy in times of economic downturns.147 Less is known about the effects of UISAs on labor markets. Their main perceived advantage over unem- ployment insurance is that workers have stronger incentives of Income Protection to seek and take jobs because benefits are financed out of individual savings. This seems to be the case, for instance, for the Unemployed in Chile, which has a mixed system involving unemploy- Severance pay faces several challenges to provide adequate ment individual savings accounts as well as a solidarity fund income protection to workers. There is a serious issue regard- for risk pooling.148 Because balances in individual accounts ing nonpayment of severance pay by employers. Particularly can increase the value of pensions at retirement and, under in developing countries, severance pay is frequently not some conditions, finance investments in education or home enforced and many workers fail to obtain benefits. Nonper- purchases, workers are less likely to exhaust their savings if formance is often due to the limited risk-pooling potential of there are work alternatives. However, as in the case of Chile, the program, coupled with its mostly unfunded nature and since these ‘savings’ belong to workers, it is very difficult to the fact that the liabilities often arise when a firm is in need of impose restrictive conditions on withdrawing the savings, so restructuring but may be cash strapped and have no funds for that in many cases the funds are used for other purposes and severance payments. Many workers who were made redun- are not available to provide sufficient protection in case of dant do not qualify for the severance pay since their length unemployment. In those cases, individual savings accounts of service might be too short or the firms are too small and will not help cope with vulnerability of those losing their job. therefore not obliged to pay severance under the labor law. In addition, pure savings schemes usually lack mechanisms Furthermore, litigation costs that arise from disputes over the to support unemployed workers in finding a job (for exam- cause of separation might also be significant.151 Therefore, ple, compulsory visits to employment service centers), which from the point of view of providing income security in case may increase poverty among unemployed workers and their of unemployment, severance pay mechanisms have obvi- families. Moreover, pure unemployment individual accounts ous disadvantages compared to unemployment insurance do not fulfill the condition of collective financing which is schemes. at the core of social insurance. As they often are exclusively financed by the workers and lack risk-pooling mechanisms, To provide comprehensive income protection for workers, the contribution rate of individual savings accounts that the recommendation would be to introduce an unemploy- would allow for a meaningful protection is much higher than ment benefit system financed by way of contributions from that of unemployment insurance and the benefit level could employers, or employers and workers, and complemented, be very low for those workers with short and interrupted if necessary, by public funds. If an unemployment benefit careers and thus cannot play a major role of income protec- scheme is introduced, existing severance pay provisions tion for unemployed workers in many cases. Therefore, these could be reduced to their seniority-related component (as accounts can be regressive, with risks of, for example, low- deferred compensation) while income protection for unem- skilled workers quickly exhausting their accounts, leading to ployment would be ensured by the unemployment benefit a situation where those most in need of protection are not system.152 covered.149 Moreover, individual savings accounts have the potential to increase labor turnover beyond efficient levels. However, introducing a comprehensive unemployment ben- High mandates to save can create incentives for workers to efit system may not be feasible in some low-income countries. leave their jobs to withdraw some of their savings. Below- For example, there could be low institutional capacity to effec- market interest rates on savings can have the same effect, tively operate this type of reform. In this case, the alternative particularly when access to credit is constrained and indi- recommendation would be to improve existing severance viduals cannot ‘dis-save’. For example, labor turnover seems payment provisions and make the scheme more reliable for to have increased considerably because of the system of indi- vidual savings accounts in Brazil and Chile.150 151.  De Ferranti et al. (2000); Ahsan and Pagés (2009). Other limitations of severance pay as an income protection system are discussed in Holzmann and Vodopivec (2012) 152.  ILO Social Security (Minimum Standards) Convention, 1952 147.  Peyron-Bista et al. (2014). (No. 102) and ILO Employment Promotion and Protection against Unemployment Convention, 1988 (No.168) as well as its attached 148.  Reyes et al. (2010). Recommendation, 1988 (No.176) provide comprehensive and sound 149.  OECD (2010). guidelines for the design and implementation of unemployment 150.  Ribe et al. (2012); Nagler (2013). protection systems. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 35 workers and, in some cases, less costly to employers, includ- which, however, puts at risk their income security in old age. ing through prefunding arrangements. This could require leg- In addition, since benefits are financed only out of savings, islative or institutional strengthening of appeal and dispute contribution rates are usually high. This is in contrast to the settlements mechanisms in case of noncompliance of the principle that contribution payments should not generate severance payment provisions by employers. Existing systems hardship. Higher contributions also reduce incentives to join do not work well in a number of countries, entailing extensive the system and can increase informality while leaving more delays, lack of transparency, or other flaws. households without income protection. Given the sensitivity of income-protection mechanisms, it is There is a set of three important elements for the design of an important to have a broad social dialogue involving all stake- unemployment benefit system: (a) the definition of the con- holders before deciding upon changing an existing income tingencies; (b) the features of the scheme and its parameters, protection scheme. Policymakers, in consultation with work- such as eligibility and entitlement criteria; and (c) the insti- ers’ representatives, employers, and other stakeholders, have tutional arrangements. The unemployment benefit system to take into account the national circumstances when design- should at least incorporate the following dimensions: the ing the unemployment benefit system. protected categories of workers (coverage), the benefit level or calculation of replacement rate, the eligibility criteria, the Designing unemployment benefit systems. Unemployment floor or ceiling on covered earnings (if any), the duration of benefit systems usually rely on risk-pooling arrangements. In benefits, the minimum/maximum benefit, financing method, these arrangements, benefit levels may be linked to workers’ and the level of contributions. Furthermore, adequate insti- previous earnings or set as fixed amounts. While the financing tutional arrangements should be in place.155 Overall, inter- of unemployment benefit schemes usually comes from contri- national labor standards provide guidance on setting these butions, general tax revenues can play a complementary role. parameters. Choices often reflect social preferences. There In the case of unemployment individual savings accounts, are general principles one can follow to ensure that the man- workers receive their savings, which accumulate through date is adequate and affordable.156 their own contributions and those of their employers, if Benefits. The contingencies of an unemployment benefit required. In general, there is no guaranteed minimum system should include suspension or loss of earnings due to replacement rate or guaranteed benefit duration, as is the the unavailability of suitable employment by a person who is case in unemployment benefit schemes and therefore are capable of and available for work. Contribution rates and the not in line with international social security standards. Poli- levels of benefits should be fixed in a way that allows for the cymakers, in consultation with social partners, have to make adequate protection of unemployed workers, including job a choice between these two arrangements when designing search costs when transitioning between jobs, while ensur- the unemployment benefit system.153 Unemployment indi- ing the financial, economic, and social sustainability of the vidual savings accounts have both proponents and oppo- unemployment benefit scheme. nents. The main advantage over unemployment insurance, as emphasized in the literature, could be that since benefits Systems with contributory unemployment benefits should are financed out of the individuals’ savings, workers have bet- provide a replacement rate at no less than 50 percent of ter incentives to seek and take jobs.154 The main shortcoming previous earnings in case of full unemployment.157 In case is that they do not provide adequate protection to workers, of noncontributory benefits, the amount should be fixed at particularly low-skilled and low-income workers who need not less than 50 percent of the statutory minimum wage or the protection most. These workers usually have shorter peri- at a level that covers essential living expenses, whichever is ods of contributions and more frequent periods of unem- higher. Countries should guarantee, as part of their national ployment. The fact that benefits are often paid as a lump sum social protection floors, at least a basic level of income secu- can also be problematic as there is a risk that they run out rity that allows for life with dignity.158 before the individual finds a job. In some countries, unem- ployed workers who have depleted their unemployment individual savings accounts are allowed to deplete their bal- 155.  ILO’s Social Security (Minimum Standards) Convention, 1952 ance of accumulated contributions in the pension insurance, (No. 102) as well as the Promotion of Employment and Protection against Unemployment Convention, 1988 (No. 168) define both covered contingencies and core parameters while leaving it to national legislation to go beyond these minimums given the national circumstances and social preferences. ILO standards also include provisions for the effective 153.  Robalino and Weber (2013). participation of representatives of the persons protected and social dialogue. 154.  Robalino et al. (2009). Chile combines risk pooling through a 156.  Robalino and Weber (2013). solidarity fund with mandated savings in the form of UISAs so as to mitigate the moral hazard problem of traditional unemployment insurance 157.  Employment Promotion and Protection against Unemployment programs (Hartley, van Ours and Vodopivec 2010). However, according Convention No. 168 (1988). Convention No. 102 sets minimum to administrative data from the system, only 4.2 percent of beneficiaries standards at a slightly lower level. benefited from the solidarity fund (Sehnbruch and Carranza 2014). 158.  Social Protection Floors Recommendation, 2012 (No. 202). 36 While there are no universal standards for the duration of separation from the company, except in the case of partial benefit reception and the level of the replacement rates, the unemployment benefits. The right to benefit may be sus- ILO suggests at least 13 weeks of benefit reception within a pended in defined circumstances,161 which include volun- period of 12 months as well as replacement rates of at least tarily leaving employment without just cause. For example, 45 percent of the previous earnings of the beneficiary.159 the right to receive unemployment insurance benefits should be restricted if the person’s last employment or service rela- Unemployment insurance benefits may be supplemented tionship was terminated (a) at the initiative of the employee with an unemployment assistance scheme, which can be a or public servant without just cause; (b) due to a breach of flat-rate benefit or means tested. This policy could save scarce duties of employment or service, loss of confidence, an inde- resources, which could instead be directed to employment cent act, or an act of corruption; or (c) by agreement of the services and ALMPs. parties.162 Maximum and minimum benefits should be established in To receive benefits, the applicant should meet both entitle- the national legislation.160 For example, the minimum unem- ment and eligibility conditions. ‘Entitlement’ conditions ployment benefit can be equal to the minimum wage (Brazil), restrict benefits to people who, in the case of fixed-duration between the minimum wage and the minimum living allow- unemployment insurance benefits, have a sufficient record ance (China), or a percentage of a legally defined reference of contributions from work or an assimilated status and have salary that increases with the number of dependents and been unemployed for a limited duration. Alternatively, in the decreases after a period of time (Spain). case of tax-financed assistance-type unemployment benefits, Financing. Contribution rates depend, among other factors, the recipients should have low total income. ‘Eligibility’ con- on the ratio of unemployed beneficiaries to contributors to ditions, on the other hand, should restrict unemployment the UB scheme, the replacement rate of the benefit, and the benefits to people who (a) are unemployed, namely, not duration of the benefit. The financing of unemployment ben- only out of work but also able to enter work at short notice efit system can have several options, often combining more and are undertaking active steps to find work and (b) meet than one: (a) contributions or other payroll tax paid by the administrative requirements. An example for the latter would employer; (b) contributions or tax on wages earned paid by be applying for benefits with the necessary documentation the worker; (c) general revenues mobilized by other taxes; and attending interviews with an employment counselor or and (d) an implicit tax paid by those UB contributors whose applying for vacancies as directed by a public employment unemployment risk is systematically low. service.163 For example, in Italy and Poland, only employers contribute In the event of refusal, withdrawal, suspension or reduction to the scheme, while in Hungary, only employees contrib- of the benefit, claimants should have the right to present a ute. In some other countries, unemployment insurance con- complaint to the body administering the benefit scheme and tributions are shared between the employer and employee to appeal thereafter to an independent body. (Republic of Korea, China, Germany, Ukraine), or relevant Institutional arrangements. The unemployment ben- costs are covered from the overall social (Ireland and Kazakh- efit system should be administered by an institution that stan) or health insurance (Belarus) contributions. In Russia, already efficiently handles the required business processes the costs are covered by federal and local government bud- for a similar social insurance program. For example, these gets. In Australia, benefits are financed out of general taxes. processes could be provided through the national social security administration which manages all social insurance programs. Also, public employment services can be in charge Entitlement and Eligibility of registration of unemployment claims and payments of Qualifying conditions. Entitlement to unemployment ben- unemployment benefits. Some countries have found it use- efits may be subject to the completion of a reasonable quali- ful to separate unemployment insurance from other social fication period, which must however not exceed the duration insurance payment schemes because the benefits are linked necessary to preclude abuse. Entitlements to unemployment to participation in training and job-search services. There are benefits are generally conditional on the circumstances of also examples of unemployment insurance administrations run by trade unions, for instance, in Sweden. In many cases, administrative reforms will be required to enable an efficient unemployment benefit delivery system, for example, by an 159.  ILO Convention No. 102. The ILO Convention No. 168 recommends that in case of full unemployment, the initial duration of payment of the benefit may be limited to 26 weeks in each spell of unemployment or to 39 weeks over any period of 24 months. 161.  ILO social security conventions enumerate such circumstances. 160.  ILO social security standards contain provisions with respect to the level of contribution ceilings, as well as minimum and maximum benefit 162.  See Article 20 of ILO Convention No. 168. levels. 163.  Grubb (2000). BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 37 introduction of performance incentives. Innovations in infor- Reforming Severance Pay mation and communication technologies can be used to improve identification, registration, eligibility checks, benefit There are three important policy choices when it comes to payments, or collection of contributions.164 severance pay: (a) the level of the benefits; (b) the vesting period for eligibility; and (c) financing and provision. Unemployment benefit systems are an important compo- nent of activation policies if benefit recipients are expected Benefits. The simplest benefit for severance payments is a to look actively for work or participate in a program to pro- flat amount or fixed percentage of wages for all beneficia- mote their job prospects. The systems should follow the ries.168 As the severance pay system evolves, a more complex principle of ‘mutual obligations’ meaning that, in return for severance pay formula may be required in which compensa- receiving unemployment benefits and other related entitle- tion is adjusted according to years of service or age of ben- ments and being offered a range of reemployment services, eficiaries. Under such structures, older workers or those with individuals must commit and comply with a set of eligibility long service records can be entitled to more generous sever- requirements. These requirements can for instance include ance pay. The severance payment may be limited by a ceiling. active job search, participation in training, or other reemploy- The generosity of severance benefits may also differ by the ment programs. type of separation, including dismissal, redundancy, collec- tive redundancy, or end of service. To receive the unemployment benefit, the job seeker should be ready to accept suitable job offers. ‘Suitable work165 is Vesting period. It is also recommended to establish a vest- described as work that registered jobseekers/unemployed ing period for eligibility, thereby making the benefit payment cannot refuse without the suspension of unemployment conditional on a minimum number of months of employment. benefits or their reduction. Typical reasons that may lead The labor legislation in several countries provides between to sanctions for the beneficiary include the failure to report 6 months and 3 years of job tenure with the employer before to the PES without justification; refusal to accept a suitable the worker is eligible for severance.169 Certain types of con- job offer; failure to meet the conditions of actively seeking tracts are typically excluded from severance payments (such employment and availability for work; or working informally as fixed-term contracts) and adequate safeguards should be while registered as unemployed. The reason for termina- provided against the use of such contracts with the aim of tion of the status of the unemployed beneficiary might also avoiding severance payments.170 be the refusal, without good reason, to participate in active labor market programs, such as training or public works. Financing and provisions. Usually, severance pay is directly financed by employers. As employers bear the risk of con- The two core qualifying conditions that apply to most unem- tract termination, governments need to consider regulations ployment benefit systems are (a) loss of earnings due to lack to manage and mitigate this risk. In the case of large-scale of suitable employment166 and (b) to be actively searching redundancies as a result of an economic slowdown, govern- for a job. In the case of middle- and low-income countries, ments may provide financial assistance. It is desirable, how- enforcing either one has proven difficult. The main chal- ever, to have an explicit criteria and procedures to engage lenge is the high prevalence of informal employment that this type of financial support. Alternatives for consideration can render traditional enforcement mechanisms ineffective. could be to mandate employers to create external reserves An alternative approach would be to focus on ensuring that for payments. Another option would be to require employ- individuals receiving unemployment benefits—particularly ers to make contributions (typically quite small) into a public through redistributive arrangements—engage in job search contingency fund which makes severance payments only in and activities to improve their employability.167 cases where the firm goes into insolvency. In some unemployment benefit systems, the unemployment Connecting to ALMPs. Severance pay schemes should also benefit may be suspended for the period in which severance include provisions aimed at stimulating early registration of payments are directly received from their employer or from workers made redundant and seeking a job using employ- any other source under national laws or regulations or collec- ment services. In particular, the duration of severance pay tive agreements. can be conditional on registration with the PES, typically within 10 days after contract termination, as long as a well- functioning PES exists. Rapid registration allows the 164.  Robalino and Weber (2013). 165.  ‘Suitable work’ and criteria for the suspension of benefits are defined by ILO Convention No. 168 (Articles 20 and 21) and in ILO 168.  At least 12 countries use flat rate severance payments, mostly Recommendation No. 176 (para. 14). in the Commonwealth of Independent States (CIS), but also in Algeria, 166.  ILO Convention No.168. It also considers a broader range of Kenya, and Mongolia. contingencies, including partial and temporary unemployment (Art. 10). 169.  World Bank (2013). 167.  Robalino and Weber (2013). 170.  See ILO Convention, No. 158 (Art. 2). 38 jobseeker/redundant worker to be immediately offered avail- WB Doing Business database on labor legislation (http:// able vacancies, provide employment services such as job www.doingbusiness.org/). counseling and job search assistance, or send him/her to par- ticipate in ALMPs, thus shortening the transition from one job to another.171 Bibliography Abidoye, B., P. F. Orazem, and M. Vodopivec. 2009. “Firing 4.5. Resources Cost and Firm Size: A Study of Sri Lanka’s Severance Pay System.” Social Protection Discussion Paper 0916, World ILO Social Security (Minimum Standards) Convention, Bank, Washington, DC. 1952 (No. 102). Available at: http://www.ilo.org/dyn/ normlex/en/f?p=NORMLEXPUB:12100:0::NO:12100 Addison, J., and P. Teixeira. 2005. “What Have We Learned :P12100_INSTRUMENT_ID:312247:NO About the Employment Effects of Severance Pay? Further Iterations of Lazear et al.” Empirica 32. ILO Termination of Employment Convention, 1982 (No. 158). Available at: http://www.ilo.org/dyn/normlex/ Ahsan, A., and C. Pagés. 2009. “Are All Labor Regulations en/f?p=NORMLEXPUB:12100:0::NO:12100:P12100_ Equal? Evidence from Indian Manufacturing.” Journal of INSTRUMENT_ID:312303:NO Comparative Economics 37 (1) March. ILO Termination of Employment Recommendation, 1982 Almeida, R., J. Arbelaez, M. Honorati, A. Kuddo, T. Lohmann, (No. 166). Available at: http://www.ilo.org/dyn/normlex/ M. Ovadiya, L. Pop, M. L. Sanchez Puerta, and M. Weber. en/f?p=NORMLEXPUB:12100:0::NO:12100:P12100_ 2012. “Improving Access to Jobs and Earnings Opportu- INSTRUMENT_ID:312504:NO nities: The Role of Activation and Graduation Policies in Developing Countries.” Background Paper for the World ILO Employment Promotion and Protection against Unem- Bank 2012–2022 Social Protection and Labor Strategy. ployment Convention, 1988 (No. 168). Available at: Social Protection and Labor Discussion Paper 1204, http://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB World Bank, Washington, DC. :12100:0::NO::P12100_ILO_CODE:C168 Arias, O., C. Sanchez-Paramo, M. Davalos, I. Santos, E. Tiong- ILO Employment Promotion and Protection against Unem- son, C. Gruen, N. de Andrade, G. Saiovici, and C. Cancho. ployment Recommendation, 1988 (No. 176). Available at: 2014. “Back to Work: Growing with Jobs in Europe and Cen- http://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB: tral Asia.” Washington, DC: World Bank. 12100:0::NO:12100:P12100_INSTRUMENT_ID:312514:NO Bassanini, A., L. Nunziata, and D. Venn. 2009. “Job Protection ILO Employment Relationship Recommendation, 2006 (No. Legislation and Productivity Growth in OECD Countries.” 198). Available at: http://www.ilo.org/dyn/normlex/en/ Economic Policy 24 (4): 349–402. f?p=NORMLEXPUB:12100:0::NO:12100:P12100_ INSTRUMENT_ID:312535:NO Baumann, F., and N. Stähler. 2006. “Financing Unem- ployment Benefits: Dismissal versus Employment ILO Social Protection Floors Recommendation, 2012 (No. Taxes.” Labour 20 (3). http://onlinelibrary.wiley.com/ 202). Available at: http://www.ilo.org/dyn/normlex/en/ doi/10.1111/j.1467-9914.2006.00349.x/full. f?p=NORMLEXPUB:12100:0::NO:12100:P12100_ INSTRUMENT_ID:3065524:NO Blanchard, O. 2000. “The Economics of Unemployment. Shocks, Institutions, and Interactions.” Lionel Robbins US Social Security Administration in collaboration with the Lectures, London School of Economics. http://econ- International Social Security Association: Social Security www.mit.edu/faculty/download_pdf.php?id=800. Programs Throughout the World. Published in: http:// www.socialsecurity.gov/policy/docs/progdesc/ssptw/ Blanchard, O. J., and J. Tirole. 2010. “The Joint Design of index.html Unemployment Insurance and Employment Protection: A First Pass.” Journal of the European Economic Associa- ILO Employment protection legislation database—EPLex tion 6 (1). http://onlinelibrary.wiley.com/doi/10.1162/ (http://www.ilo.org/dyn/eplex/termmain.home). JEEA.2008.6.1.45/abstract. Boeri, T., P. Garibaldi, and E. Moen. 2013. “The Economics of Severance Pay.” IZA Discussion Paper 7455. http://ftp.iza. org/dp7455.pdf. 171.  In most CIS countries, the duration of severance pay is conditional to finding a new job and/or registration at the employment service within Bosch. M., and J. Esteban-Pretel. 2013. “Labor Market a certain period (usually within two weeks after dismissal). Kuddo (2012). Effects of Introducing Unemployment Benefits in an BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 39 Economy with High Informality.” IDB Publications (Work- Heckman, J., and C. Pages. 2004. “Law and Employment: Les- ing Papers) 80558, Inter-American Development Bank. sons from Latin America and the Caribbean.” Chicago: The University of Chicago Press. Brown, A. J. G., J. M. Orszag, and D. J. Snower. 2008. “Unem- ployment Accounts and Employment Incentives.” Euro- Holzmann, R. 2005. “Reforming Severance Pay: Toward an pean Journal of Political Economy 24 (3). Understanding of Program Rationale, Economic Impact and Reform Options.” Empirica 32. Cahuc, P., and A. Zylberberg. 2005. “Optimum Income Taxa- tion and Layoff Taxes.” IZA Discussion Paper 1678. Holzmann, R., Y. Pouget, M. Vodopivec, and M. Weber. 2011. “Severance Pay Programs Around the World: History, Caliento, M., K. Tatsiramos, and A. Uhlendorff. 2013. “Benefit Rationale, Status, and Reforms.” Social Protection Discus- Duration, Unemployment Duration, and Job Matching sion Paper 1111, World Bank, Washington, DC. Quality: A Regression Discontinuity Approach.” Journal of Applied Econometrics 28 (4). Holzmann, R., and M. Vodopivec, eds. 2012. “Reforming Sever- ance Pay: An International Perspective.” Washington, DC: Calmfors, L., and B. Holmlund. 2000. “Unemployment and World Bank. Economic Growth: A Partial Survey.” Swedish Economic Policy Review 7 (1): 109–53. ILO (International Labour Organization). 2011a. “General Sur- vey Concerning Social Security Instruments in Light of the Cardinale, M., and M. Orszag. 2005. “Severance Pay and Cor- 2008 Declaration on Social Justice for a Fair Globalization: porate Finance: Empirical Evidence from a Panel of Aus- Social Security and the Rule of Law.” Geneva: International trian and Italian Firms.” Empirica 32. Labour Organization. Chetty, R. 2004. “Optimal Unemployment Insurance When ———. 2011b. “Termination of Employment Legislation Income Effects Are Large.” NBER Working Paper 10500. Digest.” http://www.ilo.org/public/english/dialogue/ Chetty, R. 2008. “Moral Hazard versus Liquidity and Optimal ifpdial/info/termination. Unemployment Insurance.” Journal of Political Economy ———. 2010. “World Social Security Report: Providing Coverage 116 (2). in Times of Crisis and Beyond 2010–2011.” Geneva: Interna- Chetty, R., and A. Looney. 2006. “Consumption Smoothing tional Labour Organization. and the Welfare Consequences of Social Insurance in ———. 2014. “World Social Protection Report 2014/15: Building Developing Economies.” Journal of Public Economics 90 Economic Recovery, Inclusive Development and Social Jus- (12). tice.” Geneva: International Labour Organization. De Ferranti, D., G. E. Perry, I. S. Gill, and L. Serven. 2000. “Secur- Jaramillo, M., and J. Saavedra. 2005. “Severance Payments in ing Our Future in a Global Economy.” In World Bank, Latin Latin America.” Empirica 32. and Caribbean Studies. Washington, DC: World Bank. Kaplan, D. S. 2009. “Job Creation and Labor Reform in Latin EC (European Commission). 2006. “Employment in Europe.” America.” Journal of Comparative Economics 37 (1) March. Brussels: European Union. http://www.sciencedirect.com/science/article/pii/ EC (European Commission). 2013. “Employment and Social S0147596708000760. Developments in Europe 2013.” Brussels: European Union. Koettl, J., and M. Weber. 2012. “Does Formal Work Pay? The Fredriksson, P., and B. Holmlund. 2006. “Improving Incen- Role of Labor Taxation and Social Benefit Design in the tives in Unemployment Insurance: A Review of Recent New EU Member States.” Research in Labor Economics 34. Research.” Journal of Economic Surveys 20 (3). Kuddo, A. 2009. “Labor Laws in Eastern European and Central Friesen, J. 2005. “Statutory Firing Costs and Lay-offs in Can- Asian Countries: Minimum Norms and Practices.” Social ada.” Labour Economics 12. Protection Discussion Paper 0920, World Bank, Washing- ton, DC. Goerke, L. 2006. “Earnings-Related Severance Pay.” Labour 20 (4) December. Kuddo, A. 2012. “Public Employment Services and Activation Policies.” Social Protection and Labor Discussion Papers Grubb, D. 2000. “Eligibility Criteria for Unemployment Ben- 1215, World Bank, Washington, DC. efits.” In OECD Economic Studies 31, 2000/II. Kugler, A. 2002. “From Severance Pay to Self-Insurance: Effects Grund, C. 2003. “Severance Payments for Dismissed Employees of Severance Payments Savings Accounts in Colombia.” in Germany.” Bonn: Institute for the Study of Labor. 40 Center for Economic Policy Research. http://www.cepr Pissarides, C. A. 2001. “Employment Protection.” Labour Eco- .org. nomics 8. Kugler, A. D., and G. Saint-Paul. 2004. “How Do Firing Costs Reyes Hartley, G. R., J. C. van Ours, and M. Vodopivec. 2010. Affect Worker Flows in a World with Adverse Selection?” “Incentive Effects of Unemployment Insurance Savings Journal of Labor Economics 22 (3). Accounts: Evidence from Chile.” IZA DP No. 4681 January 2010. Lalive, R. 2008. “How Do Extended Benefits Affect Unemploy- ment Duration: A Regression Discontinuity Approach.” Ribe, H., D. Robalino, and I. Walker. 2012. “From Right to Real- Journal of Econometrics 142 (2). ity: Incentives, Labor Markets, and the Challenge of Univer- sal Social Protection in Latin America and the Caribbean.” Lalive, R., J. C. van Ours, and J. Zweimüller. 2002. “The Effect Washington, DC: World Bank. of Benefit Sanctions on the Duration of Unemployment.” IEW Working Paper 110, University of Zurich, Zurich. Robalino, D. A., and A. Mataoanu. 2005. “Severance Pay in the Middle East and North Africa Region.” Middle East and ———. 2005. “The Effect of Benefit Sanctions on the Duration North Africa Region Working Papers, World Bank, Wash- of Unemployment.” Journal of the European Economic ington, DC. Association 3 (6). Robalino, D., M. Vodopivec, and A. Bodor. 2009. “Savings for Lehmann, H., and A. Muravyev. 2011. “Labor Market Institu- Unemployment in Good or Bad Times: Options for Devel- tions and Labor Market Performance: What Can We Learn oping Countries.” Discussion Paper 4516, Institute for the from Transition Countries?” Discussion Paper 5905, Insti- Study of Labor, Bonn. tute for the Study of Labor, Bonn. Robalino, D., and M. Weber. 2013. “Designing and Implement- Messina, J., and G. Vallanti. 2007. “Job Flow Dynamics and Fir- ing Unemployment Benefit Systems in Middle and Low ing Restrictions: Evidence from Europe.” Economic Jour- Income Countries: Beyond Risk-pooling vs Savings.” IZA nal 117 (521). Journal of Labor Policy 2:12. Micco, A., and C. Pagés. 2006. “The Economic Effects of Schwab, S. 2003. “Mandated-Severance-Pay Laws in Transition Employment Protection: Evidence from International Economies.” Ithaca, NY: Cornell University. Industry-Level Data.” Discussion Paper 2433, Institute for the Study of Labor, Bonn. Sehnbruch, K., and R. Carranza. 2014. Forthcoming. “The Chilean System of Unemployment Insurance Savings Nagler, P. 2013. “How unemployment insurance savings Accounts.”  Working Paper. accounts affect employment duration: evidence from Chile.” IZA Journal of Labor & Development 2013, 2:9. Svarer, M. 2007. “The Effect of Sanctions on the Job Finding Rate: Evidence from Denmark.” University of Aarhus, OECD (Organisation for Economic Co-operation and Devel- CAM, and IZA Discussion Paper 3015. opment). 1999. “OECD Employment Outlook 1999.” Paris: OECD. Tatsiramos, K. 2009. “Unemployment Insurance in Europe: Unemployment Duration and Subsequent Employment ———. 2010. “Latin American Economic Outlook 2011: How Stability.” Journal of the European Economic Association 7 Middle Class Is Latin America?” Paris: OECD. (6). Parsons, D. 2011a. “Mandated Severance Pay and Firing Cost Venn, D. 2012. “Eligibility Criteria for Unemployment Ben- Distortions: A Critical Review of the Evidence.” In Reform- efits: Quantitative Indicators for OECD and EU Coun- ing Severance Pay: An International Perspective, edited by tries.” OECD Social, Employment, and Migration R. Holzmann and M. Vodopivec. Washington, DC: World Working Papers 131, OECD Publishing. http://dx.doi Bank. .org/10.1787/5k9h43kgkvr4-en. Parsons, D. 2011b. “The Firing Cost Implications of Alterna- Vodopivec, M. 2004. “Income Support for the Unemployed: tive Severance Pay Designs.” In Reforming Severance Pay: Issues and Options. Regional and Sectoral Studies.” Wash- An International Perspective, edited by R. Holzmann and ington, DC: World Bank. M. Vodopivec. Washington, DC: World Bank. Vodopivec, M., and M. H. Tong. 2008. “China: Improving Peyron-Bista, C., J. Carter, and M. Bedard. 2014. “Comparative Unemployment Insurance.” Social Protection Discussion Review of Unemployment and Employment Insurance Sys- Paper 44779, World Bank, Washington, DC. tems in Asia and Worldwide.” Bangkok: ILO. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 41 Vodopivec, M., L. Madzar, and P. Dolenc. 2009. “Non- World Bank. 2015. “Doing Business 2016.” Washington, DC: performance of the Severance Pay Program in Slovenia.” World Bank. Social Protection Discussion Paper 0901, World Bank, Washington, DC. Annex—Chapter 4 Table 4.1: Average Severance Pay in Countries with and without Income Protection Scheme for the Unemployed in 2013 Countries with No Scheme Countries with Scheme Total Number Total Number Income Group of Countries Average Severance Pay of Countries Average Severance Pay High Income  10 10.3 46  6.7 Upper-Middle Income  26  8.2 24 13.3 Lower-Middle Income  35 17.0 12 12.5 Low Income  31 15.7  3  8.4 Total 102 13.7 85  9.7 Source: Doing Business database 2014. Table 4.2: Percentage of Unemployed Receiving or Not Receiving Unemployment Benefits Unemployed Receiving Unemployment Benefits (%) Unemployed Contributory and Not Receiving Contributory Noncontributory Noncontributory Unemployment Schemes Schemes Schemes Benefits (%) World 10.2 1.5 11.7 88.3 Africa 0.9 0.0 1.0 99.0 Middle East 2.2 0.0 2.2 97.8 Asia and the Pacific 6.8 0.4 7.2 92.8 Central and Eastern Europe 21.1 0.5 21.6 78.4 Latin America and the 4.6 0.0 4.6 95.4 Caribbean North America 28.0 0.0 28.0 72.0 Western Europe 44.6 19.2 63.8 36.2 Source: ILO (2014). World Social Protection Report 2014/15, Geneva. 42 Figure 4.1: Severance Pay for Redundancy Dismissal 80.0 70.0 60.0 Severance pay, in salary weeks 50.0 40.0 30.0 20.0 10.0 0.0 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 100000 -10.0 GNP per capita Source: Doing Business 2013. Note: Average for workers with 1, 5 and 10 years of tenure, in salary weeks. Depending on the country level of GNI per capita (in US$) in 2012. BALANCING REGULATIONS TO PROMOTE JOBS: FROM EMPLOYMENT CONTRACTS TO UNEMPLOYMENT BENEFITS 43 This report reviews international experiences and presents options for the design of four types of labor regulations: employment contracts, minimum wages, dismissal procedures, as well as severance pay and unemployment benefits. In each case, it discusses the market failures that these regulations aim to address and suggests options to better protect workers while preserving incentives to create good jobs. The main message from the report is that labor regulations are important to protect workers from abuse and exploitation and allow them to better manage labor market risks. The report also shows that it is possible to establish a balance between workers’ protection and flexibility in the management of human resources at the firm level, avoiding both over- or under-regulation. Between these two extremes, there is a ‘plateau’ where labor regulations can o er adequate protection to workers, and contribute to shared prosperity without imposing unreasonable costs on firms. This report reflects a shared vision between the ILO and the World Bank on policies that traditionally have been controversial. This has been possible thanks to the commitment of both institutions to focus on the lessons derived from rigorous research and international experiences.