Dismissal and hiring regulations – or employment protection legislation in short – are an important determinant of worker security and firm adaptability. This chapter provides an up-to-date review of employment protection legislation in OECD countries, building on earlier work by the OECD in the area. Taking into account legislation and actual practices, it describes the regulation of individual and collective dismissals of workers on regular contracts and the regulation for hiring workers on temporary contracts. It also discusses recent reforms in employment protection legislation. The comparison of employment protection across countries in this chapter brings evidence to the policy debate on the relative importance that different systems attach to the twin aspirations of protecting workers and promoting adaptable labour markets.
OECD Employment Outlook 2020
3. Recent trends in employment protection legislation
Abstract
In Brief
Key findings
Employment protection legislation is at the heart of the contractual working relationship between a firm and its employees. For the firm, it influences the ability to attract new staff and dismiss workers to react to changes in economic conditions and technology. For the employees, it is an important element of the stability of their job. Hiring and dismissal regulations therefore affect the rights and well-being of every employee in the labour market. As this chapter explains, one motivation for employment protection legislation is to induce firms to at least partially internalise the social costs of their hiring and dismissal decisions, in terms of unemployment benefit costs, psychosocial distress and income shocks.
The OECD has a long history of being at the forefront of international comparisons of employment protection legislation, also by providing comparable indicators via the OECD Employment Protection Legislation Database. Over the past three decades, these indicators have become one of the most widely used sources for benchmarking labour market regulation. The chapter extends the database, last updated in 2013, to 2019 and describes recent trends in employment protection in OECD countries.
Besides extending past series, the new indicators allow for a more systematic and accurate assessment of different types of job protection provisions, due to methodological improvements and the inclusion of important aspects that were not considered previously. For example, collective dismissals (i.e. dismissals of several workers) are now evaluated using the same methodology as the one for individual dismissals, and enforcement issues enter the quantitative comparisons in a substantial way for the first time. The new design of the indicators ensures better comparisons between countries and better assessments of reforms.
The main findings are as follows:
The OECD indicators measure job dismissal regulations along four dimensions: i) procedural requirements before notice is given; ii) notice period and severance pay; iii) the regulatory framework for unfair dismissals; and iv) enforcement of unfair dismissal regulation. The new dimension on enforcement takes account of policies that, by explicitly limiting the scope for unfair dismissal complaints, ease de facto dismissal regulations. This can be the case with advance validations of dismissals and pre-termination resolution mechanisms (that allow a worker to leave an employer by mutual agreement or resignation without losing the right to unemployment benefits).
Job dismissal regulations exhibit large differences across OECD countries: English-speaking countries are among those with fewer restrictions on dismissals, so that the layoff risk for workers is higher. Many European Union (EU) countries as well as a few non-EU countries have more restrictions on dismissals and high job security for workers on regular contracts.
Countries with strict job protection provisions for regular workers usually have strict hiring laws for workers on fixed-term, or temporary work agency, contracts. Strict job protection provisions for regular workers require strict hiring laws for temporary workers to limit labour market duality and segmentation (the degree to which firms substitute less flexible regular contracts with more flexible temporary contracts).
The granular comparisons of employment protection in this chapter highlight which elements of job dismissal regulations play a particularly important role in the different OECD countries. The chapter shows for example where notice periods and severance pay are the highest (in Turkey, Lithuania and Israel for workers who have been in their job for four years), where there are relatively fewer rights relating to unfair dismissals (in the United States, the United Kingdom and Canada) and where advance validations of dismissals and pre-termination resolution mechanisms are strongest (in Austria and the Netherlands).
Thirty-two of the 37 OECD countries impose more restrictions on collective dismissals than on individual dismissals, mostly because of stricter consultation requirements before notice can be given. These higher restrictions reflect the greater challenge for the economy of dealing with a collective dismissal. Nevertheless, pooling several individual layoffs in one collective dismissal can, in some cases, reduce the administrative burden of the firm.
In the aftermath of the global financial crisis, a number of countries – including Greece, Portugal and several other EU countries – had eased strict dismissal regulations for regular workers to lower dualism in the labour market. In the following years from 2013 to 2019, the period under study in this chapter, 12 OECD countries reformed job dismissal regulations for regular workers, while 17 OECD countries reformed hiring regulations for temporary workers. Some reforms had as objective to reduce the stringency of employment protection against dismissals (as in France, Italy, Lithuania and Slovenia). A second category of reforms focused specifically on hiring regulations for temporary workers. Belgium and the Netherlands aligned dismissal regulation for different types of workers and dismissals. Among the countries that undertook a reform during 2013‑19, more countries relaxed dismissal regulations for regular workers than strengthened them. Countries reforming hiring regulations for temporary workers were evenly split between those that reduced restrictions on temporary contracts and those that imposed additional restrictions on them. In the current COVID‑19 health and economic crisis, several EU countries have taken temporary action to considerably strengthen protection against dismissals.
Introduction
Regulations on the hiring and dismissal of employees – or employment protection legislation (EPL) in short – are one of the most discussed areas of labour market policy. They are an important policy intervention in the labour market as they usually affect every single employee and every firm, contrary to sector- or region-specific policies, for example. They are also important because they influence job security for employees (the risk of being dismissed and the chances of moving into a job) and firm adaptability (the scope for firms to respond swiftly to changing demand and new technologies). Losing a job and finding a job are pivotal moments in many people’s lives, making job protection provisions a key determinant of well‑being.
Employment protection is the central policy pillar that supports worker security, alongside publicly funded policies such as unemployment benefits, short-time work schemes and active labour market programmes (see Chapters 1 and 2). Well-designed dismissal regulations are motivated by the desire to protect workers against arbitrary dismissals and to have the company dismissing a worker bear some of the social costs (in particular the fiscal, psychological and health costs) of the dismissal (Pissarides, 2010[1]; Scarpetta, 2014[2]). At the same time, dismissal provisions tend to preserve existing jobs, rather than help workers move into better jobs; they therefore go against the often expressed principle that policy should “protect workers, not jobs”. Another consideration is the interaction between regulations for regular contracts and regulations for temporary contracts, which influences who has access to stable jobs. These different functions and effects make the design of employment protection an important, complex matter.
This chapter provides a comprehensive overview of employment protection legislation in OECD countries and recent reforms in employment protection legislation. The chapter builds on and refines earlier streams of work in the area by the OECD (Grubb and Wells, 1993[3]; OECD, 1999[4]; OECD, 2004[5]; OECD, 2013[6]; Venn, 2009[7]). These lines of work have fed into the OECD Employment Protection Legislation Database and associated research projects. Governments, organisations and researchers have used the OECD indicators extensively for cross-country benchmarking in reports, books and academic papers. The last vintage of the indicators dates from 2013, so the present analysis extends the database by six years to 2019.
The chapter is organised as follows. Section 3.1 provides an overview of the ways in which job protection matters for labour market outcomes. Section 3.2 describes the design of the OECD Employment Protection Legislation indicators and how they aim to capture the main elements of job protection. It also presents the key aspects of the revision of the indicators, made necessary to reflect ongoing changes in the labour market since the last revision in 2008. Section 3.3 describes employment protection legislation in all 37 OECD countries. Section 3.4 looks at employment protection reforms in the OECD during 2013‑19 and how they are reflected in the indicators. The new data are available for free download at http://oe.cd/epl.
3.1. How job protection matters for labour market and economic outcomes
The rationale for job dismissal regulations is mainly twofold: to protect workers against arbitrary dismissals and to have the firm dismissing a worker carry some of the social costs of the dismissal (Cahuc, Carcillo and Zylberberg, 2014[8]). The need to protect workers against arbitrary dismissals is especially relevant for firms facing high labour supply relative to their labour demand. In such situations, some firms may undercut labour standards and threaten their workers that, if they are unhappy, they will be replaced. Dismissal regulations also reduce another possibly excessive motive for firms to dismiss a worker that can arise when the firm does not take into account the consequences of the job separation for fiscal revenues (due to lower labour income) and fiscal costs (higher expenditure on unemployment benefits). Regulation can also induce employers to internalise the health consequences of dismissals (Bassanini and Caroli, 2015[9]) and the destruction of human capital following job loss and joblessness (Neal, 1995[10]). A further rationale for dismissal regulation, in particular an advance notice period, is that it helps workers, potentially with the early support of the public employment services, to smooth the transition to the next job (OECD, 2018[11]).
The intended effect of the two main motivations for job dismissal regulation – protecting workers against arbitrary dismissals and having the firm bear some of the dismissal costs – is that layoffs are less frequent than they would be in the absence of regulation. Economic theory predicts that job dismissal regulation also reduces hiring, since firms anticipate the higher layoff costs already at the time of hiring and because workers’ opportunity cost of moving to another job is higher. Different models have obtained the result that job dismissal regulations reduce both hiring and layoffs and hence job and worker flows (Bentolila and Bertola, 1990[12]; Garibaldi, 1998[13]; Mortensen and Pissarides, 1994[14]; Nickell, 1978[15]).1 Some reduction in hiring and layoffs as a consequence of job dismissal regulation is desirable to avoid excessive worker turnover; overly strict regulation can, however, reduce hiring and layoffs below their optimal level.
A large number of empirical studies, both cross-country and single-country analyses, confirm that dismissal regulation lowers job and worker flows (Autor, Donohue and Schwab, 2006[16]; Boeri and Jimeno, 2005[17]; Gielen and Tatsiramos, 2012[18]; Haltiwanger, Scarpetta and Schweiger, 2014[19]; Marinescu, 2009[20]; Micco and Pagés, 2006[21]; Millán et al., 2013[22]; OECD, 2010[23]; Salvanes, 1997[24]). Dismissal regulation is a multifaceted concept, which the design of the OECD Employment Protection Legislation indicators in this chapter will take account of. Among the various elements of job protection, the factor that has been found to reduce labour market fluidity the most is the regulatory framework for unfair dismissals (Bassanini and Garnero, 2013[25]). Of the features defining unfair dismissal regulation, long trial periods and strict reinstatement rules seem to play the most important role.
Job dismissal regulation therefore reduces job creation and job destruction. It also seems to lengthen the duration of unemployment, as most studies, including those relying on well-identified natural experiments, find that dismissal protection for regular-contract workers has no or a small negative effect on employment and that unemployment is little affected as well. These are the conclusions of the OECD Jobs Strategy (OECD, 2018[26]) and several literature surveys (Boeri, 2011[27]; Martin and Scarpetta, 2012[28]; OECD, 2013[6]). One case when employment may be lower is when job protection increases step-wise with job tenure (i.e. with large hikes at specific seniority levels), as this may encourage firms to anticipate layoffs before their cost becomes too high (Cahuc, Malherbet and Prat, 2019[29]; García Pérez and Osuna, 2014[30]).2 Moreover, in declining sectors strict employment protection may salvage jobs with no or little effect on hiring (Messina and Vallanti, 2007[31]), and in a macroeconomic downturn reforms relaxing dismissal regulation may depress employment, at least temporarily (Bassanini and Cingano, 2019[32]; OECD, 2016[33]).
The main downside of overly strict dismissal regulation is that, by lowering job and worker flows, it tends to make labour markets less adaptable to economic change, with too little worker movement from declining towards fast-growing businesses and reduced entry and exit of firms. Several empirical papers confirm that strict dismissal regulation dampens the scope for productivity-enhancing worker reallocation from low- to high-productivity firms (Andrews and Cingano, 2014[34]; Bottasso, Conti and Sulis, 2017[35]; Bravo-Biosca, Criscuolo and Menon, 2016[36]). It can thus weaken labour productivity growth and slow economic development.
Economy-wide labour productivity growth is determined not only by worker reallocation, but also by within-firm productivity growth, which depends on investment and innovation at a given firm. In principle, dismissal regulation can increase or decrease investment and innovation. Investment and innovation might be lower because actual and anticipated adjustment costs for firms are greater. However, investment could also be higher if firms substitute more flexible capital for less flexible labour. Innovation, too, could be higher to the extent that dismissal regulation reduces the risk of being laid off, making it more likely for employees with innovative ideas that they can reap the rewards for their initiative. The empirical evidence on the effects on investment is mixed (Autor, Kerr and Kugler, 2007[37]; Bai, Fairhurst and Serfling, 2020[38]; Cingano et al., 2010[39]; Cingano et al., 2016[40]), while most studies find that overly strict dismissal regulation is linked with fewer innovative activities and weaker multifactor productivity growth (Bartelsman, Gautier and De Wind, 2016[41]; Bassanini, Nunziata and Venn, 2009[42]; Bjuggren, 2018[43]; Griffith and Macartney, 2014[44]; Murphy, Siedschlag and McQuinn, 2017[45]).3
Summarising the effects of job dismissal regulations discussed so far: strict dismissal regulation tends to reduce layoffs, which is a direct result of its intended effect to raise the costs of dismissals. It also tends to reduce hiring, as firms factor in the higher costs for a potential dismissal already at the time of hiring. Dismissal regulation therefore reduces both flows out of jobs and flows into jobs. Aggregate employment and unemployment do not appear to be affected much. Fewer job flows means lower risk of job loss, which to a certain extent is a good outcome as it countervails an otherwise inefficiently high dismissal rate. When job protection is too high, however, efficient job allocation and innovation are likely to suffer. Hence, overly strict dismissal regulation tends to reduce productivity growth and increase the duration of unemployment spells.
One consequence of the negative effect of strict dismissal regulations on productivity growth is that strict dismissal regulation also limits the scope for pay increases, given that, at least to some degree, wage developments are tied to productivity developments. Higher dismissal costs may also dampen wage levels as they add to the total expected labour costs for firms which know that they will dismiss some workers. These negative effects of strict dismissal regulations on pay and pay increases may be countervailed by dismissal regulations increasing the bargaining power of workers and therefore the labour share – see the evidence from cross-country analysis and laboratory experiments (Ciminelli, Duval and Furceri, 2018[46]; Falk, Huffman and Macleod, 2015[47]). Moreover, the literature points to a distinction between newly hired workers and already employed workers: a higher stringency of dismissal regulations has been found to lower wages of new hires (Leonardi and Pica, 2013[48]), while raising wages of incumbent workers (Martins, 2009[49]; van der Wiel, 2010[50]). One question for future research is whether stricter job dismissal regulation influences the pace of automation, which in turn would affect productivity and wage growth.
An important difference needs to be drawn between workers on regular contracts and workers on temporary contracts. Workers on regular contracts usually benefit from greater employment protection than workers on temporary contracts. The evidence indicates that, the larger is the gap in employment protection between these two contractual forms of work, the more firms use temporary contracts (Centeno and Novo, 2012[51]; Hijzen, Mondauto and Scarpetta, 2017[52]; Kahn, 2010[53]). Youth, women and the low skilled tend to be the population groups for which temporary work relationships are particularly common. Large use of temporary jobs can amplify the increase in unemployment during a business cycle downturn (OECD, 2017[54]). Simultaneously strict dismissal regulations for regular workers tend to be positive for the resilience of the labour market initially, but can hinder job creation in the subsequent recovery.
Larger duality – in the sense of a segmented labour market between highly protected workers on regular contracts and little protected workers on temporary contracts – has been shown to be associated with weaker productivity levels and growth rates (Bassanini, Nunziata and Venn, 2009[42]; Cahuc, Charlot and Malherbet, 2016[55]; Damiani, Pompei and Ricci, 2016[56]; Dolado, Ortigueira and Stucchi, 2016[57]; Hijzen, Mondauto and Scarpetta, 2017[52]). One reason is that the limited scope for career advancement in the firm for people on temporary jobs tends to reduce their commitment to the job and hence their incentives to invest in firm-specific knowledge and skills.4 Another potential reason is that duality induces an inefficiently high share of temporary workers whose employment spell is too short to exploit all production opportunities of the firm.
A deeper divide between workers on regular contracts and others on temporary contracts has also been found to be associated with worse working environments, weaker job stability and greater wage inequality (García-Pérez, Marinescu and Vall Castello, 2018[58]; OECD, 2011[59]; OECD, 2014[60]). In addition, it can have negative effects from one generation to the next: children with fathers on a temporary contract have been shown to be more likely to drop out of the education system and be unemployed than children with fathers on a regular contract (Ruiz-Valenzuela, 2020[61]).
3.2. The design of the 2019 OECD Employment Protection Legislation indicators
Many of the papers referenced in the previous section rely on earlier vintages of the OECD Employment Protection Legislation Database. The indicators have been used to investigate the effects of employment protection on worker flows, employment, productivity growth, wages, investment, resilience to a downturn, the extent of use of temporary work, wage inequality, subjective job security and political economy aspects. They have also served as a control variable or descriptive tool in many other papers. Denk and Georgieff (forthcoming[62]) survey the academic papers that have used the OECD Employment Protection Legislation Database. They refer as well to some of the numerous policy reports that have drawn on the indicators – by the OECD (e.g. OECD Economic Outlook, OECD Economic Surveys), national governments and supranational bodies and institutions (e.g. European Commission and International Monetary Fund). The wide use of the database underlines its importance for informing and influencing the setting of job dismissal and hiring regulations.
Besides legislation, the OECD indicators quantify also actual practices, by considering court rulings and collective bargaining agreements. The main distinction in the database is between indicators that assess dismissal regulation for regular workers and indicators that assess hiring regulation for temporary workers. The first part of this section looks at dismissal regulation for regular workers and the second part at hiring regulation for temporary workers.5 Importantly, the OECD indicators quantify employers’ dismissal and hiring costs and not the degree of protection for workers.
So far, three annual time series (Versions 1‑3) existed, all ending in 2013. Version 1 begins in 1985, and the two subsequent versions sought to improve the ways in which the indicators capture job protection provisions. Version 2 begins in 1998 and has some coverage of collective (besides individual) dismissals of regular workers. Version 3 begins in 2008 and introduces certain aspects of enforcement for regular workers and additional items on hiring regulation for temporary workers. The present analysis extends Versions 1‑3 to preserve the time series dimension of the database. For regular workers, however, the chapter develops, and mostly relies on, the new Version 4, which is available from 2013 to 2019. The design of the indicators for temporary workers is unchanged.
The indicators consider information on employment protection legislation in a detailed, but also pragmatic way: they take account of national and sectoral, but not firm-level, collective bargaining agreements. They focus on the private, not public, sector and evaluate regulation applying to medium-sized and large, not small, firms and their employees.6 Where there are differences by firm size, the scored value is the average of the values for a firm with 35, 150 and 350 employees. Where there are differences between categories of workers (for example blue-collar and white-collar workers), the scored value is the average of the values corresponding to each category. These standardisations are necessary in light of the available information. A further consideration is the design of the scoring scale which, as from the beginning of the database, is constructed to enable quantitative comparisons of regulatory stringency. The nature of the indicators, which convert mostly qualitative information into numerical data, means that readers should nevertheless be cautious when interpreting small differences in scores across countries and over time.
3.2.1. The OECD Employment Protection Legislation indicators for dismissing regular workers
The new Version 4 of the OECD Employment Protection Legislation indicators for regular workers better reflects the differences in job protection regulation across countries and time. This is achieved in four ways: i) by improving the granularity of some elements of employment protection regulation that have already entered the indicators; ii) by adding important elements of employment protection regulation that have so far been absent from the indicators; iii) by expanding the assessment of employment protection regulation of collective dismissals to align it with the assessment of employment protection regulation of individual dismissals; and iv) by changing the way in which dismissal size thresholds for dismissals of several workers enter the indicator of employment protection regulation of collective dismissals.
All employment protection legislation indicators in this chapter refer to no-fault dismissals; hence, the stated reason for the dismissal is not related to illegitimate behaviour of the worker (such as theft, misconduct or unauthorised absence from work). The indicators, in the case of both individual and collective dismissals, take account of four aspects of dismissal regulations (Table 3.1): procedural requirements, notice period and severance pay, the regulatory framework for unfair dismissals and enforcement of unfair dismissal regulation. The first two of these categories are defined by two lower-level elements, the last two by four elements. The four broad categories determine with equal weight the aggregate score, and the lower-level elements determine with equal weight the scores for the four broad categories.7 Where there are differences between dismissals for personal and for economic reasons, the scored value is the average of the two. Annex 3.A provides further details on the methodology and the full scoring scale.
Table 3.1. The OECD Employment Protection Legislation indicators for dismissing regular workers
Note: The changes indicated in blue are with respect to the 2008 version (i.e. Version 3) of the indicators. The four broad categories of dismissal regulation determine with equal weight (25%) the aggregate score; the lower-level elements determine with equal – or almost equal – weight the scores of the four broad categories. Length of trial period is not included in the indicator for collective dismissals. Annex 3.A provides the full scoring scale.
Procedural requirements captures the actions that the firm must take before or when issuing the dismissal to the worker. It consists of two components: notification procedures and time delay before notice can be given. Notification procedures assess whether the dismissal notification needs to be provided with reasons, its legal value and whether it needs to be preceded by a warning procedure, a discussion with the worker and the consultation or authorisation of a third party, such as the relevant administrative body. This item on notification procedures was already included in previous versions of the indicators, but the new version modifies it to better reflect the differences in procedure linked with the various types of notification requirements. Also, in the new version, the scoring scale for the time delay before notice has been aligned with the one for the length of the notice period at four years of job tenure, so that days, weeks and months count the same in both these lower-level elements.
The second category considers the length of the notice period and the amount of severance pay, the two elements of dismissal regulation that often come first to mind. As in earlier versions, both are evaluated as the average of the values at three points of job tenure: 9 months, 4 years and 20 years.
The third category, the regulatory framework for unfair dismissals, is concerned with the breadth of the definition of fair and unfair dismissals and the stringency of remedies imposed by courts when a dismissal is judged to be unfair. It takes into account four aspects: the definition of unfair dismissal, the length of the initial (or trial) period during which the employee is not protected against unfair dismissal, the monetary compensation to the worker following an unfair dismissal and the possibility of reinstatement following an unfair dismissal.8
The new version substantially expands the item on the definition of (un)fair dismissal. This now considers unfair dismissals for economic reasons separately to those for personal reasons.9 Within unfair dismissals for economic reasons, the item assesses the freedom that judges have in their decision, the restrictions on the firm as to which worker is to be selected for the dismissal and the requirements in terms of alternative employment and training opportunities that the firm needs to offer to the worker and without which the dismissal is considered unfair. The item on the length of the trial period has been slightly refined as well. It now takes higher values when the regulation for dismissing workers before the end of the trial period is more stringent in terms of notice and severance pay.
The new version of the employment protection indicators for regular workers places a stronger emphasis on enforcement of unfair dismissal regulation. Three items have been newly introduced: i) whether the worker alone has the burden of proof when filing a complaint for unfair dismissal; ii) whether an ex‑ante validation of the dismissal limits the scope of unfair dismissal complaints; and iii) whether a pre-termination resolution mechanism exists that, by granting eligibility for unemployment benefits, offers an attractive alternative to dismissal for the employee.
In most countries, the burden of proof in unfair dismissal cases does not lie only with the employee. While this increases the cost and the uncertainty for the firm that it can prove that the dismissal was fair (Boeri, Garibaldi and Moen, 2017[63]), it reflects that the firm is the party that knows the reason or motivation for the dismissal. Validation of the dismissal as a preventive check is closely related with the item on notification procedures in the first category, procedural requirements. It tends to make notifications more burdensome for the firm, but has the advantage for the firm of limiting the risk that the dismissal will be judged as unfair later on. Pre-termination resolution mechanisms (termination by mutual consent or resignation) give the right to unemployment benefits in many OECD countries (sometimes with sanctions). France, for example, introduced in 2008 a formalised scheme of termination by mutual agreement that provides entitlement to unemployment benefits, and the reform has been found to have increased worker flows (Batut and Maurin, 2019[64]).
The coverage of enforcement issues in the OECD indicators remains limited overall, as the indicators do not take account of certain aspects of the functioning of the judicial system, such as access to labour courts or the length of proceedings. Such aspects of the complexity of judicial procedures matter for the decision of the firm whether to dismiss a worker (Espinosa, Desrieux and Ferracci, 2018[65]; Gianfreda and Vallanti, 2017[66]), but can also influence the incentives of the employee to file a complaint (Campolieti and Riddell, 2020[67]; Espinosa, Desrieux and Wan, 2017[68]; Fraisse, Kramarz and Prost, 2015[69]). Their effects on employers’ costs remain therefore ambiguous, which makes it difficult to incorporate them in the indicators. Another consideration is that integrating statistics on judicial procedures in the indicators would be problematic given lack of data and poor cross-country comparability and raise issues concerning the endogeneity of judicial outcomes to regulation and labour market conditions (Ichino, Polo and Rettore, 2003[70]).
The overall employment protection legislation indicators for dismissing regular workers assign a weight of 5/7 to individual dismissals and 2/7 to collective dismissals, as in previous versions. In the design of the indicators, a dismissal is seen as collective when a firm lays off several workers at around the same time. More precisely, the indicator for collective dismissals in Version 4 evaluates for each item the average for dismissals of 10, 45 and 120 workers by a firm within one month, rather than including a separate item on dismissal thresholds as before.10 In all OECD countries with specific legislation for collective dismissals (and for the firm sizes considered by the indicator), this legislation always applies in the case of dismissals of 120 workers or more in one month, which will be referred to as mass dismissals in the remainder of this chapter.11 Moreover, in contrast to individual dismissals, collective dismissals can only occur for economic reasons. Therefore, while the indicators for individual dismissals give the same weight to dismissals for personal and economic reasons, in the aggregate indicators dismissals for economic reasons take a weight of almost two‑thirds.
3.2.2. The OECD Employment Protection Legislation indicators for hiring temporary workers
The OECD Employment Protection Legislation indicators for temporary workers distinguish between fixed-term contracts and temporary work agency contracts. They focus on hiring restrictions instead of dismissal regulations, in contrast to the indicators for regular workers. This is natural to a certain extent, given that terminations of temporary contracts tend to be rare during the duration of the contract and easy at the end, while legislation in many countries aims to avoid an excessive use of temporary contracts. Nevertheless, dismissals of temporary employees do happen, and in some cases severance pay or other tools used in dismissal regulation protect temporary workers when their contract expires. Future work is planned to go beyond hiring restrictions for workers on fixed-term contracts, by also considering regulations in the context of dismissals of fixed-term workers and expirations of fixed-term contracts.
The two categories of hiring regulations, for fixed-term and temporary work agency contracts (Table 3.2), contribute in equal shares to the total score. This is to be kept in mind in an environment where in all OECD countries fixed-term contracts are more common than temporary work agency contracts (OECD, 2014[60]). The first three lower-level elements serve the same purpose for the two types of temporary contracts: to capture constraints to the ease with which such contracts can be used in place of regular contracts. The fourth item on authorisation and reporting obligations is in practice mainly relevant for temporary work agency contracts. The last item on equal treatment of temporary work agency workers and regular workers at the user firm concerns both pay and working conditions. The full scoring scale is in Annex 3.A.
Table 3.2. The OECD Employment Protection Legislation indicators for hiring temporary workers
Note: There are no changes with respect to the 2008 version (Version 3) of the indicators. The two broad categories of hiring regulation determine with equal weight (50%) the aggregate score; the lower-level elements determine with equal weight the scores for the two broad categories, except for the first element in both cases (“Valid cases for use of fixed-term contracts” and “Types of work for which temporary work agency employment is legal”) which carry a weight of 50% and 33% respectively. Annex 3.A provides the full scoring scale.
3.2.3. Comparison with other employment protection legislation databases
The last two decades have seen an increasing effort by international organisations and research centres to create indicators of labour market institutions, including of employment protection legislation.12 Country coverage of the OECD Employment Protection Legislation indicators themselves was extended in 2013 to include, for example, several non-OECD Latin American and Caribbean countries, in a partnership with the Inter-American Development Bank. Besides the OECD indicators, among the most well-known are the EPLex by the International Labour Organization, the Labour Regulation Index by the Centre for Business Research (CBR-LRI) at the University of Cambridge and the (suspended) Employing Workers indicator in the Doing Business dataset by the World Bank.13 While these three databases share some similarities with the OECD indicators,14 important differences remain. Overall, the OECD indicators continue to be the most widely used for cross-country comparisons in policy reports and academic papers (Denk and Georgieff, forthcoming[62]).
The three other databases cover more countries than the OECD indicators (from 100 countries for the EPLex to 190 countries for the Doing Business). However, they provide a less comprehensive overview of employment protection regulation.15 In particular, the three other databases do not cover enforcement of unfair dismissal regulation, and the EPLex and the Doing Business do not cover temporary work agency employment. In addition, the CBR-LRI and the Doing Business have a less deep evaluation of collective dismissals and of the regulatory framework for unfair dismissals (for example they have no information on compensation following an unfair dismissal). There are also certain aspects of job protection that the OECD indicators do not cover, but some of the other databases do (for example prohibited grounds for dismissal such as discrimination in the EPLex). Two other differences between the EPLex and the OECD indicators are that the EPLex is more descriptive, rather than quantitative, and available for a shorter time span, since the series starts in 2009, against 1985 for the OECD indicators. Also, the three other databases do not, or little, take account of sectoral collective agreements and case law.
3.3. Employment protection legislation in OECD countries in 2019
This section compares employment protection legislation in OECD countries, based on information in the OECD Employment Protection Legislation Database. It begins with a detailed look at the components of job dismissal regulation. It then presents assessments of the overall regulation of individual and collective dismissals of regular workers and the regulation for hiring temporary workers. These indicators of dismissal and hiring regulations are based on the methodology outlined in the previous section.
3.3.1. Country details on individual elements of dismissal regulations for regular workers
This part discusses country details on individual elements of job dismissal regulations for regular workers in the order of the four broad categories entering the indicator: procedural requirements before notice can be given, notice period and severance pay, the regulatory framework for unfair dismissals and enforcement of unfair dismissal regulation.
Procedural requirements
Procedural requirements for individual dismissals of regular workers vary significantly between countries (Figure 3.1). This variation is mostly due to differences in notification procedures, rather than in the time delay before notice is given, the other regulatory element entering this category.
The employer usually needs to inform the employee of the reason for the dismissal, at least if the worker requests it. This notification generally takes the form of a written statement. Some countries require, at least in some cases, a prior warning (Australia, Austria, Québec in Canada, the Czech Republic, Denmark, Estonia, Greece, Ireland, Lithuania, New Zealand, Portugal and the Slovak Republic), an interview with the employee (Australia, Colombia, France, Iceland, Ireland, Luxembourg, Slovenia and Turkey) or a consultation with a third party (the Czech Republic, Finland, Israel, Norway, Poland, the Slovak Republic and Sweden). Only in Canada (except Québec) and in most states of the United States, the firm never needs to provide a reason for the dismissal to the worker before or at the time of dismissal. By contrast, procedures are particularly stringent in the Netherlands, where a dismissal cannot occur without prior authorisation by the Public Employment Service or the Sub-district Court. In Germany and Sweden, the dismissal can be paused until the final judgement by the court if the work council (Germany) or worker (Sweden) objects to the dismissal.
Notification procedures for collective dismissals are more homogeneous and stringent than for individual dismissals for economic reasons (Denk and Georgieff, forthcoming[62]). All countries, except some provinces in Canada, Chile and the United States, require a consultation or even an authorisation before the dismissal can take place from a certain threshold number of workers dismissed. In particular, above this number of dismissals, a dismissal can never occur without the authorisation of the administration in Colombia and France (for firms with more than 50 employees). In Belgium, if the work council and worker object to the dismissal, the dismissal can be paused until the employer has provided evidence of compliance with the notification and consultation procedures. In Mexico, dismissals for economic reasons are allowed only if they involve several workers and they require an authorisation by the Labour Court.
Figure 3.1. Procedural requirements for individual dismissals of regular workers
2019
Note: Range of indicator scores: 0‑6. Procedural requirements consists of two components: notification procedures and time delay before notice can be given.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
Notice and severance pay
At the point at which a firm decides to dismiss a worker, it often cannot do so without informing the worker in advance (i.e. respecting a notice period) and providing severance pay. Notice period and severance pay represent a cost to the firm. To the worker, they reduce the economic and possibly psychological burden of a layoff, by preventing an abrupt loss of labour income. Employees usually continue to work for the firm during the notice period. However, in some cases, notified employees are released from work, and even employees continuing work during the notice period are likely to be less motivated. A combined estimate of the costs to firms and the benefits to workers from notice period and severance pay is thus provided by the number of months a firm must give notice before dismissal and the amount of severance pay (in months of work).
The database collects information on notice period and severance pay at three points of job tenure: 9 months, 4 years and 20 years. At 4 years of job tenure, compensation in the form of notice period and severance pay in the case of an individual dismissal varies widely among OECD countries, from no compensation in the United States to six months of pay in Turkey (Figure 3.2). The compensation of half a year of pay in Turkey corresponds to one‑eighth of the four years of labour income earned until then. Costs to firms and benefits to dismissed workers are also high in Israel and Lithuania, where more than an extra 10% of the labour income earned until then is paid as compensation to a dismissed worker. Notice periods are more widely used overall than severance pay is: only two countries have no notice period, while 12 countries have no severance pay. Nonetheless, countries with the highest total compensation stand out with very high severance pay. The variation in severance pay across countries is much greater than the variation in notice periods.
Figure 3.2. Notice period and severance pay for individual dismissals of regular workers
Four years of job tenure, measured in months of pay after dismissal notice, 2019
Note: These values are for individual (not collective) dismissals. They take the average of dismissals for personal and economic reasons.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
Notice period and severance pay both tend to increase with the length of job tenure. The longer workers stay with their firm, the more secure their job tends to be. This greater job security at higher job tenure may also contribute to the observed lower job mobility of long-tenured, and often older, workers. In all OECD countries, notice period and severance pay are increasing, or at least not decreasing, with longer job tenure across the three points of tenure in the database: 9 months, 4 years and 20 years. On average in OECD countries, together they are seven times as high at 20 years of job tenure as at 9 months of job tenure (Figure 3.3).
Severance pay increases more steeply with job tenure than notice period. On average in OECD countries, severance pay is smaller than notice period at 9 months of job tenure, but greater than notice period at 20 years of job tenure. At 9 months of job tenure, two‑thirds of OECD countries require no severance pay. At 20 years of tenure, workers in nine OECD countries – Belgium, Chile, France, Israel, Luxembourg, Mexico, the Netherlands, Spain and Turkey – have the right to a severance pay that is worth at least half a year of work. It is relatively more common for the notice period to be the same at the three points of job tenure.
Notice periods are similar for dismissals for personal and for economic reasons. They are also similar for individual and collective dismissals, with the exception of a few countries where they are longer for collective dismissals. For example, in the United States dismissed workers are entitled to two months of notice in the case of mass layoffs and large plant closures. In Canada, the notice that must be given to an employee affected by a collective dismissal is often longer than for an individual termination of employment. In Luxembourg and the United Kingdom, in the case of a large dismissal, a long time period must occur between notification to the labour authority and the day the dismissal takes effect, de facto prolonging notice periods for employees with short job tenure.
Figure 3.3. The role of job tenure for notice period and severance pay
OECD average at three different points of job tenure, measured in months of pay after dismissal notice, 2019
Note: These values are for individual (not collective) dismissals. They take the average of dismissals for personal and economic reasons. OECD average is the unweighted average for the 37 OECD countries.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
For severance pay, more notable differences arise between dismissals for economic and personal reasons and between individual and collective dismissals. At 20 years of job tenure, severance pay is higher in the case of dismissals for economic reasons in Australia, the Czech Republic, Estonia, Ireland, Poland, the Slovak Republic and the United Kingdom. At this point of tenure, no country has higher severance pay for dismissals for personal reasons. This on average somewhat lower level of security in the case of dismissals for personal reasons is potentially motivated by workers themselves sharing more of the responsibility for the layoff, for example due to insufficient performance. With regard to the comparison of individual and collective dismissals, additional compensation can often be granted for dismissals that exceed a certain number of workers, usually as a result of the consultation with worker representatives.
Counting months of advance notice and severance pay on equal terms as in the analysis of this section is necessarily a simplification. For the firm and the worker, notice periods are often somewhat less costly and more protective than severance pay: the worker is required to continue work while on notice and sufficiently long notice periods allow the employment services to intervene before the dismissal takes place, thereby facilitating the transition to another job. These considerations motivate OECD advice that countries, where notice periods are short and severance pay is high, could consider extending notice periods and lowering severance pay, while activating early interventions, to smooth job transitions without increasing employers’ costs (OECD, 2018[11]).
Regulatory framework for unfair dismissals
In almost all OECD countries, a dismissal based on a reason that is beyond the scope of allowed (or “fair”) reasons can, if it is challenged in court, lead the employer to pay specific compensation to workers or even reinstate workers to the positions from which they were dismissed. The category “regulatory framework for unfair dismissals” captures the definition of unfair dismissal, the length of the trial period during which all dismissals are fair and the compensation and reinstatement rules following an unfair dismissal (Figure 3.4). Fair reasons for dismissal generally include operational reasons (e.g. economic difficulties or technological changes) or personal reasons related to workers themselves (e.g. insufficient performance or unsuitability). In Canada (except Québec) and the United States, an employee can be fairly dismissed without reason, provided that the dismissal was not based on prohibited grounds.16 Contrasting examples are Chile, which forbids dismissals for insufficient performance and unsuitability, and Mexico, which allows dismissals for economic reasons only if they involve several workers.
Figure 3.4. Regulatory framework for unfair individual dismissals of regular workers
2019
Note: Range of indicator scores: 0‑6. Compensation following unfair dismissal is missing for Canada, Greece, Iceland and the United States, and length of the trial period is missing for Canada and the United States. Values for these subcomponents are set equal to the average of non-missing subcomponents of regulatory framework for unfair dismissals for the same country.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
The scope of fair dismissals for economic reasons depends to a large extent on the freedom that judges have in their decision. In about half of the OECD countries (including Finland, Germany, Poland, Spain and the United Kingdom), dismissals for economic reasons can only be challenged if the reason for the dismissal was false or patently irrational. By contrast, in the other half of the countries (including Australia, Chile, Italy, Japan, the Netherlands and Norway), judges can question the operational need of the dismissal decision. In some countries, when redundancy could concern several workers occupying similar positions, the employer should select the employees who are to be dismissed based on objective criteria other than performance. Job tenure may be part of these criteria, as for example in France, Latvia, Portugal and Sweden.17 In Italy, if the dismissal concerns five employees or more within a period of 120 days, judges cannot question the operational need for the dismissal and the firm must follow social and economic criteria for selecting the workers to be dismissed.
About two‑thirds of OECD countries require substantive conditions for a dismissal for economic reasons. These conditions generally include attempting the transfer of the worker to another position, possibly with retraining. They may also require priority for rehiring (e.g. Finland and France) or provision of outplacement services (e.g. Belgium). Other countries have no conditions, at least for individual dismissals: Canada, the Czech Republic, Denmark, Greece, Hungary, Iceland, Israel, Slovenia, Spain, Switzerland, Turkey, the United Kingdom and the United States. However, Canada, Denmark, Greece, Slovenia, Spain, Switzerland and Turkey have additional requirements from a certain number of workers dismissed. These requirements typically include the establishment of a social plan, i.e. a set of measures of reemployment, retraining, outplacement and, in some cases, extra monetary compensation for the workers.
Dismissals for personal reasons exist in most countries. Employers can dismiss workers who have become unsuitable for the position (due to medical or qualification reasons) or whose performance has become insufficient. In some countries, however, insufficient performance, without unsuitability, is not a fair reason for dismissal (e.g. Chile, Finland, France, Mexico, Norway, Portugal, Spain and Sweden). Attempting substantial alternatives can be required also in the event of a dismissal for personal reasons. For example, a transfer to a suitable position should be attempted in the case of a dismissal for medical unsuitability in Belgium and France and for all types of unsuitability in Finland and Italy. In Japan and Spain, workers should be trained to avoid a dismissal for insufficient qualification.
In almost all OECD countries, unfair dismissal regulation does not apply during an initial (or trial) period at the beginning of the employment relationship. Belgium, Chile, Greece, Israel, Japan and Poland are the only exceptions, although temporary contracts might sometimes act as a substitute for the trial period.18 The median value of the trial period is three months. It is longest in the United Kingdom (24 months) and Ireland (12 months).
When judges deem the dismissal (at some point after the completion of the trial period) to be unfair, they can order the payment of a compensation or the reinstatement of the worker to the position. Compensation following an unfair dismissal is particularly high in Italy. Reinstatement is always made possible to the employee in Austria, the Czech Republic, Korea, Latvia and Turkey. By contrast, reinstatement, except in the case of dismissals on prohibited grounds, cannot be imposed on the employer in Belgium, Colombia, Estonia, Finland, France (for individual dismissals), Iceland, Lithuania, Luxembourg, Spain, Sweden, Switzerland and the United States. In France, in the event of a dismissal of more than ten workers in a firm with more than 50 employees, the absence (or insufficient elaboration) of a social plan can entail the nullity of the redundancy procedure; in these cases, the judge may order the reinstatement of the employees upon their request. Overall, reinstatements tend to be more common in countries with a more stringent regulatory framework for unfair dismissals (e.g. Greece, Korea, Latvia, Norway, Portugal and Turkey).
Enforcement of unfair dismissal regulation
The indicators on enforcement of unfair dismissal regulation consider the maximum time to make a claim, the burden of proof, ex-ante validation of the dismissal and pre-termination resolution mechanisms (Figure 3.5). As explained in Section 3.2.1, this means that the coverage of enforcement issues in the indicators remains limited overall since the indicators, for various reasons, do not take account of certain aspects of the functioning of the judicial system, such as access to labour courts or the length of proceedings.
The median duration for the time period during which an employee can file an unfair dismissal complaint is two months among OECD countries. In some countries (Austria, Denmark, Hungary, Lithuania, Slovenia, Switzerland and Turkey), the maximum time available is so short that in practice claims must be filed before the dismissal takes effect. By contrast, it is longer than two years in Colombia, Iceland, Israel, Japan and the United States19 (where it varies by state).
In most countries, following an unfair dismissal complaint, it falls, at least in part, on the employer to provide evidence that the dismissal was fair. Who bears the burden of proof matters for the incentives of the firm to dismiss a worker and of the employee to file a complaint. The only countries where the burden of proof lies solely with the employee in cases of unfair dismissal (not based on prohibited grounds) are Australia, Colombia, the Czech Republic, Denmark, Israel, Poland, the Slovak Republic, Switzerland and the United States.
Figure 3.5. Enforcement of unfair dismissal regulation for individual dismissals of regular workers
2019
Note: Range of indicator scores: 0‑6. The minimum score for burden of proof and ex-ante validation of the dismissal is 0.5.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
Validation of the dismissal as a preventive check tends to make notification procedures more stringent, but it has the advantage for the firm of limiting the risk that the dismissal will be judged as unfair later on. Only in Austria and the Netherlands, all dismissals need to involve an advance validation that limits the scope for unfair dismissal complaints; dismissals should be approved by the work council in Austria and the Public Employment Service or the Sub-district Court in the Netherlands. A validation secures the dismissal for the employer, but only from a given number of workers dismissed, in Belgium (in some cases), Colombia, France, Greece, Mexico and Spain.
Resignation and some form of termination by mutual consent provide the right to unemployment benefits in many countries (possibly with sanctions) and are thus a popular alternative to dismissals. In a number of countries, they grant unemployment benefits under the same conditions as in the event of a dismissal. This is the case in Austria, Chile, Colombia, France, Hungary, Japan, Korea, Lithuania, Mexico (where there are no unemployment benefits), the Netherlands and the Slovak Republic. In other countries, resignation and termination by mutual consent entitle workers to receive unemployment benefits with long waiting periods (Immervoll and Knotz, 2018[71]). By contrast, resignation and termination by mutual consent never give access to unemployment benefits (in contrast to dismissals) in Canada, Greece, Italy, Luxembourg, Slovenia, Spain, Turkey and the United States20 (in most states and in the case of individual termination). These countries are also among those for which the overall level of enforcement is particularly high.
3.3.2. Aggregate assessments of dismissal regulations for regular workers
This part assesses job dismissal regulation for workers on regular contracts by aggregating the individual elements that the previous section discussed. It does so first for individual dismissals, then for collective dismissals and finally for a composite of individual and collective dismissals.
Regulation of individual dismissals of regular workers
The OECD indicators show wide variation in the strictness of regulation of individual dismissals of regular workers across countries (Table 3.3). Five of the ten countries with the lowest measured regulation have a legal system with British common-law origin: the United States, Canada, Australia, the United Kingdom and Ireland (in order). Regulation is assessed to be low as well in Switzerland, Austria, Hungary, Denmark and Estonia. At the other end of countries with relatively strict regulation are the Czech Republic, Israel, Portugal, the Netherlands, Turkey, Belgium, Italy, Latvia, Greece and Luxembourg. In between are the remaining countries where policies seem to attach more equal importance to firm adaptability and job security.
The aggregate score is determined by the four categories of regulation: procedural requirements, notice and severance pay, the regulatory framework for unfair dismissals and enforcement of unfair dismissal regulation. One question is whether countries with high overall regulation have strict regulations along all four dimensions or whether certain categories of regulation are more typical of high-regulation countries. This is investigated statistically through the correlations among the four categories of dismissal regulation and their correlation with the aggregate score. This analysis indicates that the fourth category, enforcement of unfair dismissal regulation, plays a different role to the other three categories, underlining the importance of including it in the indicators to obtain a more complete picture of regulations. Procedural requirements, notice and severance pay and the regulatory framework for unfair dismissals are all positively correlated, suggesting that they tend to be complementary, rather than substitute policies.
The aggregate score is only weakly correlated with enforcement of unfair dismissal regulation.21 Some countries with low regulatory protection in aggregate have a high score on enforcement of unfair dismissal regulation, notably Canada and the United States. Canada is one of the countries that provide no access to unemployment benefits except after dismissals and so offer no alternative pre-termination route that would be made attractive by granting unemployment benefits. In the United States, the time to make a claim of unfair dismissal is long (in case restrictions to dismissals exist in the contract or “implied contract”, in which the employer gives certain assurances for continued employment to create a contract of sorts). The Netherlands, by contrast, a country with overall high regulatory protection, has a low score: the Public Employment Service or the Sub-district Court provides an ex-ante validation of the dismissal, and termination via mutual consent gives right to unemployment benefits without sanctions.
In pairwise comparisons, enforcement of unfair dismissal regulation is negatively correlated with the three other categories.22 One reason may be that in countries, where unfair dismissals carry few rights (i.e. the regulatory framework for unfair dismissals is less strict), a high degree of enforcement of unfair dismissal regulation is less relevant. Another reason is design: in general, validation of a dismissal by a third party before the dismissal occurs enters the indicator negatively under enforcement as such validation reduces the chances for the dismissal to be qualified as unfair; and it enters the indicator positively under procedural requirements as it makes the notification procedure more stringent.23
The new Version 4 of the indicators changes the assessment of the stringency of job dismissal regulations in several countries compared with the previous Version 3. Job protection against individual dismissals is, for example, assessed to be less strict in Austria, France, Germany and the Netherlands and stricter in Canada, Ireland and the United States. Box 3.1 mentions some of the reasons behind the new evaluation, drawing on the analysis in Denk and Georgieff (forthcoming[62]). Conceptually, the differences in scores are due to three reasons: i) revisions to the categories procedural requirements and the regulatory framework for unfair dismissals; ii) the addition of the fourth category enforcement of unfair dismissal regulation; and iii) the reduction in the weight of the three categories that have already entered the indicator. As the examples in the box illustrate, the modifications to the indicator design ensure better overall comparisons between countries, because they improve the way in which the indicators map the costs for firms stemming from different regulatory aspects.
Table 3.3. The OECD indicators: Strictness of regulation of individual dismissals of regular workers
White / light blue / dark blue: countries with low / middle / high regulatory protection, 2019
Note: Range of indicator scores: 0‑6. The ten countries with the lowest and highest score are classified as countries with low and high regulatory protection. Scores are rounded to one decimal, while classification is done with the actual scores. The OECD Employment Protection Legislation indicator is the average of the scores for the four broad categories.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
Box 3.1. New design, new assessment: What is different with the new OECD indicators?
The main changes to the design of the OECD indicators for job dismissal regulations of regular workers are due to a substantive revision of the first category procedural requirements, the addition of the fourth category enforcement of unfair dismissal regulation and the expansion of the evaluation of employment protection regulation of collective dismissals. Each of the three aspects is important for improving the comparisons of countries in the new Version 4 relative to the previous Version 3. This box summarises the role of procedural requirements and enforcement of unfair dismissal regulation (Figure 3.6). Further details, also on the role of collective dismissals, can be found in Denk and Georgieff (forthcoming[62]).
Procedural requirements is the category with the largest influence on the differences between the two versions. Newly taking into account warning procedures and the obligation to consult the worker before the dismissal contributes to the higher scores for Québec (Canada) and Ireland. The need to notify a third party is now viewed as a constraint only when it involves a consultation or authorisation procedure, which is not always, or never, the case in Austria, France, Germany and Latvia. The alignment of the scoring scales for the time delay before notice and the length of the notice period is another explanation.
Two of the countries for which the addition of the enforcement category significantly lowers the scores are Austria and the Netherlands. In these countries, an advance validation of the dismissal acts as a preventive check, reducing the chances for the dismissal to be qualified as unfair, and terminations via mutual consent grant access to unemployment benefits under the same conditions as in the case of a dismissal. These aspects therefore enter the indicator with a minimum value for the two countries.
Figure 3.6. Changes in the assessed strictness of regulation of individual dismissals
Difference in the indicator score between the new Version 4 and the previous Version 3, 2019
Note: Range of indicator scores: 0‑6. Countries are ordered based on the difference in the overall indicator score. A negative (positive) number means that regulation is evaluated to be less (more) strict with the new Version 4 than the previous Version 3. For further details, see Denk and Georgieff (forthcoming[62]), “The 2019 OECD Employment Protection Legislation indicators: New insights on job dismissal regulation in OECD countries”, OECD Social, Employment and Migration Working Papers, OECD Publishing, Paris.
Source: OECD calculations based on OECD Employment Protection Legislation Database, http://oe.cd/epl.
Regulation of collective dismissals of regular workers
Periods of economic difficulty, due to for example a persistent decline in demand or required technological change, might lead firms to restructure their workforce, involving the dismissal of a large number of workers in relatively short time. It is common that specific regulations apply in these situations. Until the analysis in this chapter, the database considered such specific restrictions on collective dismissals only in selected dimensions of dismissal regulation and as a top-up to individual restrictions. The new OECD data present, for the first time, employment protection legislation indicators for collective dismissals that are calculated analogously to the employment protection legislation indicators for individual dismissals. The section presents dedicated indicators for individual dismissals for economic reasons, as collective dismissals normally occur for economic reasons and the OECD indicators for individual dismissals described above reflect both dismissals for personal and economic reasons.
While important in its own right, a comprehensive assessment of regulation of collective dismissals has gained further relevance given developments over the past decade. In the wake of the global financial and economic crisis, the number of “zombie firms”, i.e. firms with difficulty to meet their financial obligations, has increased (Andrews, Adalet McGowan and Millot, 2017[72]), potentially related with the regulation of collective dismissals. Moreover, the scope for firms in difficulty to adjust wages rather than the workforce has shrunk in the context of downward nominal wage rigidities, low inflation and weak nominal wage growth. Digitalisation and globalisation trends are also likely to make more firms restructure their workforce.
As mentioned in Section 3.2.1, the indicators define a dismissal as collective if several workers are laid off within one month, hence irrespective of whether specific regulations apply. More precisely, the indicators score the average of the values for 10, 45 and 120 dismissals in one month. In some countries specific regulations apply as of 10 dismissed workers or lower; in others they start only when the number of involved workers is greater. The dismissal threshold in all countries with specific regulations for collective dismissals is smaller than 120 workers. In this way, the indicators for collective dismissals reflect both the stringency of regulation, when several dismissals are subject to specific regulations, and the scope of the regulation, as captured by the dismissal threshold.
All OECD countries – with the five exceptions of Chile, Israel, Korea, Mexico and New Zealand – impose more stringent restrictions on collective dismissals relative to individual dismissals (Figure 3.7).24 Chile, Israel, Korea and New Zealand regulate collective dismissals the same as individual dismissals; Mexico is a peculiar case, as legislation does not allow individual dismissals for economic reasons, only collective dismissals for economic reasons. Protection against collective dismissals is 10‑15% higher than against individual dismissals in the OECD on average, mostly because of stricter procedural requirements before notice can be given – see Section 3.3.1 and Denk and Georgieff (forthcoming[62]). Countries’ widespread use of specific restrictions on collective dismissals is likely to reflect the greater challenge for the economy of dealing with a collective dismissal. Nevertheless, pooling several individual layoffs in one collective dismissal can, in some cases, reduce the administrative burden of the firm.25
Figure 3.7. The OECD indicators: Strictness of regulation of collective dismissals (defined as dismissals of several regular workers in one month)
2019
Note: Range of indicator scores: 0‑6. Individual dismissals refer to economic reasons, because collective dismissals are always for economic reasons. The indicators score the average of the values for 10, 45 and 120 dismissals within one month.
Source: OECD calculations based on OECD Employment Protection Legislation Database, http://oe.cd/epl.
Where regulation of individual dismissals is strict, regulation of collective dismissals tends to be strict, as in practice regulation of individual dismissals often serves as a minimum for that of collective dismissals. Six of the ten countries with the lowest as well as highest regulation of individual dismissals are also among the ten countries with the lowest, respectively highest, regulation of collective dismissals. Across countries, additional restrictions on collective dismissals (the difference in regulation between collective and individual dismissals) are not significantly related with the strictness of regulation of individual dismissals. Therefore, the extent to which collective dismissals are subject to specific regulations appears to be more the choice of governments and countries, rather than a natural consequence of the regulation of individual dismissals.
An extreme form of collective dismissals are mass dismissals which the chapter defines as layoffs of at least 120 workers in one month. This definition ensures that the threshold for specific regulations applying to a series of individual dismissals is passed in all OECD countries with dedicated legislation for a series of individual dismissals. The measured degree of regulation is generally the same for mass dismissals as for smaller-scale layoffs with a number of dismissals that is above the threshold for specific regulations. Therefore, the higher is the threshold for specific regulations, and the more extensive are the additional restrictions compared with individual dismissals, the greater is the difference between the indicator of regulatory strictness for mass dismissals relative to that for collective dismissals.
In one‑third of OECD countries, the dismissal threshold for collective dismissals to be subject to specific regulations is (or is equivalent to) 10 or less workers in one month. In these countries, the regulation of mass dismissals is identical to the regulation of collective dismissals (Figure 3.8). The dismissal threshold is higher in some other countries, in particular in Australia, Colombia, Japan and the United States. This explains the comparatively high additional restrictions on mass dismissals in these countries.
Figure 3.8. Strictness of regulation of mass dismissals (defined as dismissals of at least 120 regular workers in one month)
2019
Note: Range of indicator scores: 0‑6. The regulation of mass dismissals corresponds to the regulation of collective dismissals above the country‑specific threshold for a series of individual dismissals to be subject to different legislation. The definition of at least 120 regular workers ensures that this threshold is passed in all countries with dedicated legislation for a series of individual dismissals.
Source: OECD calculations based on OECD Employment Protection Legislation Database, http://oe.cd/epl.
Overall regulation of dismissing regular workers
The stringency of overall regulation of individual and collective dismissals of regular workers, as assessed with the aggregate indicators, is similar to the regulation of individual dismissals and the regulation of collective dismissals given their high correlation (Figure 3.9). Taking the indicators at face value suggests that dismissal regulation in the most regulated countries is close to twice as strict as in the least regulated countries.
The data bring out international differences in labour market and social models quite strongly. Three of the four countries with the least strict regulation are Anglo-Saxon: the United States, Canada and Australia. Geographically, OECD countries in North America and Australasia have regulation that is evaluated to be below the OECD average.26 By contrast, regulation in the majority of the OECD countries that are also members of the European Union is above the OECD average. Four of the five countries with the highest regulation are EU countries: the Czech Republic, the Netherlands, Portugal and Italy. English-speaking countries combine low costs for firms with relatively few protective measures for workers, while dismissal costs and job security for regular workers are comparatively high in many EU countries.
Figure 3.9. The OECD indicators: Strictness of regulation of dismissing regular workers
Contributions of each component, 2019
Note: Range of indicator scores: 0‑6. These aggregate indicators assign a weight of 5/7 to individual dismissals and 2/7 to collective dismissals.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
The role of collective bargaining and case law
Employment protection regulation is not based solely on the legislation, but it also depends on collective agreements and case law, and the indicators of employment protection legislation take this into account.
Collective agreements may influence dismissal protection for regular workers, for example by changing notice periods (e.g. in Australia, Austria, Denmark, France, Iceland, Italy, Sweden), the trial period (e.g. in Denmark, France, Hungary, Iceland, Italy, Sweden, Turkey), severance pay (e.g. in Australia, Denmark, France) or the criteria for selecting which workers to dismiss (e.g. in Finland, Norway, Sweden).27 Box 3.2 examines in more detail the role of national and sectoral collective bargaining in France, Italy and Sweden, three countries where the share of workers covered by collective agreements is high. In some countries, firm-level collective agreements may derogate from the law or higher-level agreements, thereby reducing the binding effect of the regulation. This is, however, beyond the scope of the indicators.
Case law also matters for assessing aspects that are little or not addressed in the legislation, for example the freedom that judges have in their decisions or the legal value of the written statement on the reason for the dismissal. Moreover, it can affect the interpretation of the legislation, as is the case for the transfer requirements and reinstatement options in Italy or the maximum delay to file a complaint for unfair dismissal in France.
Box 3.2. Collective agreements and employment protection in France, Italy and Sweden
Collective agreements can affect job protection provisions, especially in countries with high collective bargaining coverage rates like France (99%), Italy (80%) or Sweden (90%). Collective agreements typically strengthen the standards of protection that are in the law, but sometimes they may also be allowed to derogate, i.e. to set lower standards of protection.
The OECD indicators account for the rules in national and sectoral collective agreements. In France, sectoral agreements for managers and professionals (cadres) usually set longer notice periods and higher severance pay at long tenure. In Italy, sectoral agreements generally provide for shorter trial periods and longer notice periods. In Sweden, most sectoral agreements include exemptions to the “Last-In-First-Out” (LIFO) principle, according to which workers who were hired last are those to be dismissed first. In addition, collective agreements for white-collar workers frequently include a 55/10 provision (Söderqvist and Lindberg, 2019[73]): the notice period for workers aged 55 or older and with 10 or more years of tenure is 12 instead of 6 months. Taking account of collective agreements increases the job dismissal indicator for regular workers in France and Italy (Figure 3.10). In Sweden, the effect is small as the LIFO exemptions and the 55/10 provision influence protection in opposite directions.
Figure 3.10. How taking into account collective agreements affects job protection
Strictness of regulation for individual and collective dismissals of regular workers, 2019
Note: Range of indicator scores: 0‑6. “No CA” refers to the hypothetical score when collective agreements are not taken into account.
Source: OECD calculations based on OECD Employment Protection Legislation Database, http://oe.cd/epl.
The indicators do not capture a number of aspects of collective agreements due to limited coverage or the small size of the regulatory deviation involved. In France, many sectoral collective agreements for managers set longer trial periods, and derogations from several restrictions to the use of temporary employment are allowed (since 2017). In Italy, most collective agreements extend the length of the notification procedure. In Sweden, sectoral agreements for white-collar workers generally allow for an extended maximum duration for successive fixed-term contracts.
Small firm exemptions
All OECD indicators in this chapter focus on regulation as it applies to medium-sized and large firms, which generally employ most employees in OECD economies. For example, in 2013, firms with 20 employees or more employed 77% of all employees in Canada, 69% in Israel, 47% in Korea, 70% in Mexico, 73% in Turkey and 83% in the United States.28 In some countries, however, small firms are subject to less strict regulation, which is beyond the scope of the indicators. In Australia, for example, firms with less than 15 employees do not have to pay redundancy pay. In Austria, firms with less than 5 employees are not required to have, and therefore to inform, a work council. In Germany, firms with 10 employees or less are exempt from dismissal regulation (except in cases of discriminatory and arbitrary dismissal). In Korea, firms with 4 employees or less are exempt from dismissal regulation, except regarding advance notice or equivalent compensation. In Portugal, in the event of an unfair dismissal, firms with less than 10 employees may submit a request to the court to oppose reinstatement. In Spain, in firms with 25 employees or less, the maximum duration of the trial period is three instead of two months (except for workers with a higher education degree) and the Wage Guarantee Fund pays part of the redundancy pay (except if the dismissal is deemed to be unfair). In Turkey, in the event of an unfair dismissal, firms with less than 30 employees do not have to reinstate workers or pay compensation and back pay. A final example is the exemption from requirements for collective dismissals in firms with less than 20 employees in Belgium, the Czech Republic, Denmark, Germany, Hungary, Iceland, Switzerland and Turkey.
3.3.3. Regulation of hiring temporary workers
An important aspect of job protection regulations relates to the difference between regular and temporary workers. It is more difficult for firms to lay off regular workers than to not renew temporary contracts (OECD, 2014[60]). To counteract potential overuse of temporary contracts by firms, governments usually impose restrictions on their use. As Section 3.2.2 described, the OECD Employment Protection Legislation Database therefore also gathers indicators on the regulation of hiring temporary workers. It distinguishes between regulation of fixed-term and temporary work agency contracts, attaching equal weight to the two.
OECD countries exhibit an even wider variation in the regulatory restrictions to hiring temporary workers, as measured by the indicators (Figure 3.11), than in the restrictions to dismissing regular workers. On average, higher regulation of the two types of temporary contracts tend to go hand in hand, although there are many idiosyncratic situations. The geographic pattern is similar to the one for the regulation of regular workers. All common-law OECD countries (Canada, the United States, the United Kingdom, New Zealand, Australia, Ireland and Israel) are at or near the bottom of the distribution of this type of regulation. Four of the five countries with the highest regulation are EU countries: Luxembourg, Italy, France and Spain.
Figure 3.11. The OECD indicators: Strictness of regulation of hiring temporary workers
2019
Note: Range of indicator scores: 0‑6. These aggregate indicators assign the same weight to fixed-term contracts and temporary work agency contracts.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
The correlation between the two main indicators in the database – for dismissing regular workers and for hiring temporary workers – is highly positive (Figure 3.12). Countries with higher regulation on one tend to have higher regulation on the other. While this is generally the case, a small number of countries (including in particular the Czech Republic, Israel and the Netherlands) appear to step somewhat out of line. These are assessed to have relatively low regulation of temporary contracts given their high regulation of regular contracts.
The overall positive relationship between the regulation of regular and temporary contracts is likely to be the result of the differences in regulation of regular contracts together with policy makers’ desire to restrain the use of temporary contracts. Where regular contracts are not much regulated, firms have few incentives to replace regular with temporary contracts; the need to restrict the use of temporary contracts is therefore not there. In countries with high regulation of dismissals of regular workers, strict regulation of temporary contracts can help avoid that these are overused. As seen in the Netherlands, Portugal and Sweden for example (OECD, 2014[60]), relatively low regulation of temporary contracts in situations of high regulation of regular contracts can lead to strong, unintended labour market segmentation between highly protected regular workers and weakly protected temporary workers. Section 3.4 will shed further light on the extent to which recent reforms in job protection have gone in the direction of increasing or reducing the regulatory divide between regular and temporary workers.
Figure 3.12. Dismissal regulation for regular workers and hiring regulation for temporary workers are positively correlated
2019
Note: Range of indicator scores: 0‑6. The indicator for dismissals of regular workers is for individual dismissals only, as the hiring indicator for temporary workers is also based on hiring one worker.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
3.4. Recent reforms in employment protection legislation
The focus of the analysis of employment protection reforms in this section is on the period between 2013, the year of the previous release of the OECD Employment Protection Legislation indicators, and 2019, the latest year in the database. This period follows the more immediate aftermath of the global financial crisis, during which a number of countries – including Greece, Portugal and several other EU countries – had eased strict dismissal regulations for regular workers to lower dualism in the labour market.
In the 2013‑19 period, 21 OECD countries undertook at least one reform, as reflected in a change in the OECD indicator score for job dismissal regulation for regular workers or hiring regulation for temporary employment. The section focuses exclusively on those reforms that affected the indicators. It describes their main features and impact on the assessed level of job protection, as measured by the indicators. Country-specific OECD reports provide further details on some of the reforms (Carcillo et al., 2019[74]; OECD, 2018[75]).
These recent reforms can be classified into three categories. A first class of reforms went into the direction of reducing restrictions to dismissing regular workers. Second, several countries changed restrictions on the use of temporary employment, in some instances to reduce labour market duality. A third group of countries (Belgium and the Netherlands) aimed to make regulation fairer and simpler by standardising the protection of regular workers against different types of dismissal. Among the reforming countries, those that relaxed dismissal regulation were more numerous than those which strengthened it, while a similar number of countries increased and decreased the stringency of hiring regulation for temporary workers.
Three figures provide the basis for the discussion in this section. Figure 3.13 depicts the changes in the regulatory indicator for individual and collective dismissals of regular workers and in the regulatory indicator for hiring temporary workers from 2013 to 2019. The following two figures, Figure 3.14 and Figure 3.15, decompose these changes in terms of the main dimensions of the regulation.
Figure 3.13. Quantifying recent reforms in employment protection legislation
Note: Range of indicator scores: 0‑6. Data for Colombia and Lithuania refer to 2014 instead of 2013.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
Figure 3.14. Regulatory changes in recent reforms of dismissal regulation for regular workers
Indicator score
Note: Range of indicator scores: 0‑6. All countries with changes in the indicator of protection for individual dismissals are displayed. Data for Lithuania refer to 2014 instead of 2013.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
Figure 3.15. Regulatory changes in recent reforms of hiring regulation for temporary workers
Note: Range of indicator scores: 0‑6. All countries with changes in the indicator of regulation for hiring temporary workers are displayed. Data for Lithuania refer to 2014 instead of 2013.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
3.4.1. Reforms that reduced the stringency of employment protection against dismissals
A number of countries implemented reforms aimed at easing dismissal regulations, sometimes at the same time as they eased restrictions on the use of temporary employment. Major reforms (or sets of reforms), involving several aspects of regulation, took place in France, Italy, Lithuania and Slovenia. Four other OECD countries enacted more specific relaxations of regulations concerning dismissals of regular workers, with notable effects on the job protection indicators.
Successive reforms in France
Between June 2013 and December 2017, France enacted a number of reforms that reduced the stringency of job dismissal regulation. Among the main measures adopted, the August 2016 Labour law clarified the definition of fair economic reasons for dismissals. A substantial reduction (larger than a specified threshold) in at least one of several economic indicators listed in the law, such as orders or turnover, must now be considered as fair by the court, while economic difficulties and technological changes can still be invoked even if the reduction does not reach the threshold (as before the reform). Subsequently, in the second half of 2017, the Ordonnances established a schedule for compensation following an unfair dismissal and introduced a formalised scheme of collective termination by mutual agreement (rupture conventionnelle collective). They also clarified the definition of procedural irregularity, which is far less penalised than unfair dismissals. In particular, an irregular notification of the reason for the dismissal in the dismissal letter is no longer sufficient to make the dismissal unfair. In addition, the maximum period to challenge a dismissal in court successively decreased from five years in 2013 to twelve months in 2018. With regard to collective dismissals specifically, the June 2013 Loi relative à la sécurisation de l’emploi limited the risk of a dismissal being classed as unfair by requiring a validation of the social plan by the administration before the dismissal (in firms with more than 50 employees), although this came at the cost of making the notification procedure more complex.
Overall, the reforms moderately affected employment protection for regular workers, as measured by the aggregate indicator (see Figure 3.13, Panel A). The reduction is mostly due to the lesser legal value of the reason stated in the dismissal letter, which considerably simplified the notification procedure in the event of an individual dismissal. In addition, the shortening of the period during which an employee can file a complaint to enforce unfair dismissal regulation plays a role as well (see Figure 3.14, Panels A and D).29 The previous Version 3 of the OECD indicators did not capture the legal value of the dismissal letter for regular workers. Hence, France is a good example for how the new Version 4 better captures reforms of dismissal regulation. Box 3.3 provides further information on these improvements.
The Jobs Act and the elimination of the mobility allowance in Italy
The March 2015 Jobs Act in Italy considerably reduced employment protection for regular workers against individual dismissals. One of the major measures of the Jobs Act was to eliminate the possibility of reinstatement (in firms with more than 15 employees) in case of individual dismissals for economic reasons and most collective dismissals, as well as in some cases of dismissals for personal reasons.30 The Act also replaced the mandatory conciliation phase that was to take place before all individual dismissals for economic reasons, and before a dismissal for personal reasons if the employee requested it, with an ex-post conciliation procedure. These two changes significantly reduced the job protection indicator for regular workers (see Figure 3.13, Panel A), by limiting the consequences of both unfair individual and collective dismissals, and by simplifying the individual dismissal procedure (see Figure 3.14, Panels A, C and F).
Box 3.3. How the new version of the indicators helps capture reforms of dismissal regulations
A comparison of the changes from 2013 to 2019 in the aggregate dismissal indicator for regular workers between the previous Version 3 and the new Version 4 shows how the new version allows to capture regulatory changes that were not (or very little) captured by the previous version (Figure 3.16).
This is especially the case for France, Greece and Hungary.11 The more detailed assessment reflects the revision of existing, and the addition of new, items in a way that takes into account the legal value of the reason stated in the dismissal letter (for France), the authorisation requirements for collective dismissals (Greece) and the burden of proof in an unfair dismissal case (Hungary). Also, the changes in the indicator resulting from the reforms in Belgium, Lithuania and the Netherlands are considerably amplified in Version 4 with respect to those due to other reforms. For Belgium, Version 4 captures the shift of the burden of proof, which Version 3 does not. For Lithuania, the new labour code affected collective dismissal regulation in about the same way as individual dismissal regulation and hence is reflected in the new indicator for collective dismissal, while Version 3 only reflects collective dismissals to the extent that they represent additional restrictions relative to individual dismissals. For the same reason, Version 3 does not capture the extension of individual severance pay to collective dismissals in the Netherlands. By contrast, the Italian and Slovenian reforms are relatively less marked in Version 4 than Version 3. This is because the involvement of a third party in the notification procedure (for Italy) and transfer requirements (for Slovenia) matter less, as these are now part of a broader range of factors accounting for procedural requirements and unfair dismissal regulation.
Figure 3.16. The new version of the indicators tends to better capture reforms of dismissal regulations
Strictness of regulation for individual and collective dismissals of regular workers, change from 2013 to 2019
Note: Range of indicator scores: 0‑6. All countries with changes in the indicator of protection for individual and collective dismissals are displayed. Data for Lithuania refer to 2014 instead of 2013.
Source: OECD Employment Protection Legislation Database, http://oe.cd/epl.
1. Additional exercises conducted for important job protection reforms before 2013, such as the Spanish one, also show that the new Version 4 provides a more comprehensive assessment of reforms.
The decline in the indicator for collective dismissals (see Figure 3.14, Panel F) is not only due to the Jobs Act, but also, and more significantly, to an earlier reform of severance pay that came into force only in January 2017. Before then, in the event of a collective dismissal, an employee with at least 12 months of job tenure was entitled to a “mobility allowance”, replacing unemployment benefits, and the employer had to pay a contribution to this allowance that amounted to two to seven months of salary.31 The suppression of this allowance implies that the employer now needs to pay merely the same contribution as for individual dismissals, i.e. usually less than one month of pay.
The new labour code in Lithuania
In Lithuania, the new labour code, which entered into force in July 2017, provides more flexibility to employers, regarding both the dismissal of regular workers and the use of temporary forms of employment.
The reform reduced notice period and severance pay and made reinstatement subject to approval by the employer, although a specific compensation (capped at six months of wages) needs to be awarded if there is no reinstatement. In addition, the new labour code includes a special procedure for dismissals at the will of the employer, in which an employee can be dismissed for any reason at very short notice (three days) and high severance pay (six months of wages). This new procedure, together with the elimination of reinstatement obligations, significantly reduced the employment protection indicator for regular workers (see Figure 3.13, Panel A), by restricting both the definition and the consequences of an unfair dismissal (see Figure 3.14, Panel C).32
The new labour code also made the use of temporary employment easier. All restrictions on valid cases for the use of fixed-term contracts were lifted, provided that they do not account for more than 20% of all employment contracts. There are no limits on the number of successive fixed-term contracts, although they can only be used for a maximum of two years for a given employee in the same function and five years in different functions. As a result of these relaxations, Lithuania is the country with the largest decline in the indicator for temporary employment from 2013 to 2019 (see Figure 3.13, Panel B).
The new Employment Relation Act in Slovenia
In Slovenia, the new Employment Relation Act came into force in April 2013, with significant reductions in the protection of regular workers against dismissal and more flexible rules on the use of temporary work agency employment.
Following this reform, opposition by a trade union can no longer inhibit the dismissal procedure, notice period and severance pay are now lower, and a dismissal can be qualified as fair even when the employer did not attempt to retrain or transfer the worker to another position. These three aspects of the reform affected the indicator for regular employment (see Figure 3.14, Panels A, B and C), and the overall impact on the protection of regular workers is considerable (see Figure 3.13, Panel A).
Regarding the regulation of temporary contracts, the same Act waived the 12 months duration limit of successive temporary work agency assignments and the requirement for temporary work agencies to provide annual reports. At the same time, the Act established that temporary work agency employment can no longer exceed 25% of employment at the user firm. These changes reduce the indicator for temporary employment (see Figure 3.13, Panel B).
Reforms in other countries
Based on the indicators, four other countries softened employment protection for regular workers, without changing restrictions on the use of temporary employment. Finland extended the trial period from four to six months in January 2017. In Greece, collective dismissals can take place without the approval of the administration since 2017. In Portugal, two acts in August 2013 reduced severance pay from 20 days to 12 days per year of tenure and created a “compensation fund” to help finance it.33 In 2013, the United Kingdom halved the minimum period between notification to the administration and a collective dismissal from 90 to 45 days.
The regulatory changes in Finland, Greece and the United Kingdom, which were either small or limited to collective dismissals, do not change the assessed overall level of employment protection much. The impact of the Portuguese reforms to severance pay and its financing is of similar magnitude to that of the major reforms in France, Italy, Lithuania and Slovenia (see Figure 3.13, Panel A).
The large majority of countries among those reforming dismissal regulations for regular workers reduced their stringency. Hungary and Spain are the only two countries that made dismissal regulations more stringent (see Figure 3.13, Panel A), besides Belgium and the Netherlands where the increase in regulatory stringency came more as a by-product of standardising protection against different types of dismissals (see Section 3.4.3). In Hungary, the new code of civil procedure that entered into force in January 2018 shifted the burden of proof on the employer in labour law cases.34 In Spain, employers can since January 2019 no longer hire under a “Permanent Employment Contract to Support Entrepreneurs”. This contract was restricted to firms with less than 50 workers and included a trial period of one year instead of four months.35 The changes to dismissal regulations in Greece and Hungary are reflected in the new and finer Version 4 indicators, whereas the previous Version 3 would have shown no change (see Box 3.3).
Exceptional measures in the COVID‑19 crisis
The evaluation of employment protection legislation in this chapter is particularly pertinent given the current COVID‑19 health and economic crisis which has severely increased the dismissal risk for many employees in the private sector. Job dismissal protection, when coupled with effective short-time work schemes, has likely preserved jobs in countries badly affected by the crisis.36 As Box 3.4 details, a few EU countries have further strengthened job dismissal protection in the crisis. The policy priorities in the area of employment protection over the coming months will depend on the evolution of the pandemic, restrictions to economic activity and developments in the labour market. They will likely require a shift from the immediate need to help preserve existing jobs and incomes to increasing support for firm-to-firm worker mobility, also in light of the structural changes in the labour market that the crisis is bringing (for example increased demand of workers for health care and online and delivery services).
3.4.2. Reforms of hiring regulation for temporary workers
In Lithuania and Slovenia, as discussed, one component of a broader labour market reform was to facilitate temporary employment. Five other countries undertook specific reforms to reduce restrictions on the use of temporary employment, with corresponding effects on the indicator. By contrast, reforms in several countries imposed additional restrictions on temporary contracts. Some of them introduced a maximum cumulated duration for successive contracts or assignments, while Italy and, to a lesser extent, Denmark implemented restrictions on the valid cases for use of these forms of employment.
Reforms that focused on facilitating temporary employment
In France, contracting on temporary employment was made easier in August 2015 with an increase in the maximum number of successive fixed-term contracts from two to three, which is reflected in the corresponding decline in the indicator for temporary contracts (see Figure 3.13, Panel B). The possibility for collective agreements to derogate from restrictions to the use of temporary employment, introduced by the 2017 Ordonnances, may further facilitate the use of temporary contracts in the future, as new agreements are signed and extended.
Box 3.4. Exceptional measures in job dismissal regulations during the COVID‑19 crisis
The COVID‑19 health and economic crisis has drastically reduced economic activity and put many firms in financial difficulty. It has hence severely increased the risk for many private-sector employees to be dismissed based on economic grounds. The crisis has heightened the risk to be dismissed for personal reasons as well, especially for workers who have difficulties maintaining high work performance in the face of increased care responsibilities (for example because they have families with a sick household member or young children whose schools are closed). To a certain degree, differences in labour market developments between countries reflect the stronger protection of employees in the European Union where, up to this point, employment has been resilient and the increase in unemployment mild (see Chapter 1).
Several countries in the European Union have further strengthened job dismissal protection during the COVID‑19 crisis. Four EU countries (France, Greece, Italy and Spain) have taken significant, time‑limited action to discourage economic redundancies, favouring the continuation of existing employment relationships. Two EU countries (again Italy and the Slovak Republic) have strengthened protection of employees against personal dismissals. The remainder of the box gives an overview of these measures.
France announced increased scrutiny of collective dismissals for economic reasons by the authorities as part of the notification procedure in firms with more than 50 employees. Firms are allowed to dismiss employees, if they can show that they were already in economic difficulty before the COVID‑19 crisis and if they predict to be unable to restart part of their activity in the next six months. Economic dismissals related to the COVID‑19 crisis are, in principle, only allowed when a firm ceases its activity or based on other, rather tightly defined economic considerations. Another objective has been to relax restrictions relating to the renewal or prolongation of fixed-term contracts.
Greece prohibited dismissals of employees in firms that have suspended their operations because of the lockdown measures. However, employers who are significantly affected by the COVID‑19 crisis may suspend the contracts of their employees for up to one month. Upon expiry of the suspension of the contracts, companies must maintain the same number of employees for a period equal to that of the suspension.
Italy blocked individual and collective dismissals for economic reasons for the first two months of the crisis. The ban applies to layoffs on grounds connected to the reduction or transformation of activities, reorganisation of work or business closure. In addition, Italy limited the scope for dismissals for personal reasons. The absence from the workplace of parents of a child with a disability and of parents with a child aged between 12 and 16 cannot constitute a just cause for contract termination, provided that the employees communicate these reasons of absence to their employer.
Spain adopted the requirement that dismissals for reasons connected to the COVID‑19 pandemic need to be reviewed by a judge and will be qualified as either null or inadmissible. If the dismissal is judged as null, the employee will be reinstated to the position. If the dismissal is seen as inadmissible, the employee receives a compensation of 33 days of pay per year of tenure (in addition to the statutory severance pay), as in any other case of unfair dismissal. In addition, Spain prolonged the duration of fixed-term contracts that expire during the health emergency.
The Slovak Republic strengthened the protection of workers against personal dismissals. The measure considers employees with a personal obstacle to working, such as caring for a sick family member or a young child due to school closure, as temporarily unfit for work, thereby protecting them from dismissal for the duration of their inability to work. The same provisions apply when the employee is subject to quarantine or isolation.
Four other countries lowered the restrictions for firms to hire temporary workers. In Norway, since July 2015, firms can use fixed-term contracts without restrictions on the reasons for their use, for at most 12 months and within the limit of 15% of their workforce. Turkey, where temporary work agency employment was prohibited, introduced this form of work in 2016 in specific industries and for specific reasons. Belgium and Spain in 2013 extended the use of temporary work agency employment. The reforms in Norway and Turkey significantly reduced the restrictions to temporary employment, as reflected by the indicator, while Belgium and Spain experienced only a small decline in the assessed degree of restrictions to temporary employment (see Figure 3.13, Panel B).
Reforms that increased restrictions to temporary employment
A number of countries introduced a legal limit for the cumulated duration of fixed-term contracts or temporary work agency assignments: Poland with 33 months for fixed-term contracts (February 2016), Germany with 18 months for temporary work agency assignments (April 2017) and the Slovak Republic with 24 months for temporary work agency assignments (March 2015). In Japan, the 2013 revision of the Labour Contract Act made it possible for workers who have had a fixed-term contract for at least five years to have their contract converted into a permanent one. These changes resulted in small, but nevertheless noticeable increases in the indicator (see Figure 3.13, Panel B).37
In Italy, reforms first reduced, but later increased restrictions to temporary employment. The March 2014 Poletti decree abolished the obligation to provide a rationale when using fixed-term contracts and allowed for five successive renewals (so long as these contracts do not exceed 20% of the number of regular contracts in firms with more than five workers). The decree also allowed the use of temporary work agency contracts with no justification. However, the reform of July 2018 restored, and even reinforced, the obligation to provide a rationale when using a fixed-term contract for more than 12 months. Possible extensions (up to three) for a maximum duration of 24 months are allowed for temporary and objective needs or to replace some workers. The use of temporary work agency assignments was restricted to the same reasons and maximum duration. These policy changes explain why, of all countries, Italy exhibits the largest increase in restrictions to temporary employment during 2013‑19 (see Figure 3.13, Panel B). Restrictions on the use of both fixed-term and temporary work agency contracts explain this increase (see Figure 3.15, Panels A and B).
Denmark implemented similar restrictions in July 2013 by conditioning temporary employment on objective reasons. However, this reform focused only on renewals of temporary work agency assignments and therefore resulted in a small increase in the indicator (see Figure 3.13, Panel B).
3.4.3. Reforms that standardised protection against different types of dismissal
Belgium and the Netherlands undertook significant employment protection reforms to standardise regulation, either across workers with different employment status (blue-collar and white-collar workers in Belgium) or across different types of dismissal (via the Public Employment Service or the Labour Court in the Netherlands).
The single status in Belgium
The January 2014 reform in Belgium introduced a single status to abolish regulatory differences between blue-collar and white-collar workers, which the Constitutional Court had considered discriminatory and therefore unconstitutional.
The reform harmonised the time length of the period before the dismissal takes effect (delay before notice and notice period) and severance pay, resulting overall in an increase in protection for blue-collar workers and a decrease for white-collar workers. The reform also standardised the definition of an individual unfair dismissal and the course and consequences of the associated unfair dismissal procedure. In particular, the burden of the proof is always shared between the parties (it was previously with the employee for cases involving white-collar workers), and the compensation granted in the event of an unfair dismissal is now aligned with the lower compensation previously granted to white-collar workers. In addition, the reason for the dismissal now needs to be provided upon the request of the employee. The reform also abolished the trial period and expanded the use of outplacement regimes (i.e. services provided by the employer to help dismissed workers find a new job) following an individual dismissal, previously restricted to older workers. As of December 2016, the newly created “Reintegration programme” has continued the trend towards more protection, by ensuring that workers with long-term medical incapacity are transferred to suitable jobs.
These regulatory changes considerably increased the strictness of dismissal rules, as measured by the indicators (see Figure 3.13, Panel A). This mainly reflects the new obligation to provide a reason for dismissal in the notification procedure, as well as the elimination of the trial period within which workers were not protected by unfair dismissal law. The partial shift of the burden of the proof towards the employer in unfair dismissal cases concerning white-collar workers played a role as well (see Figure 3.14, Panels A, C and D).
The Work and Security Act in the Netherlands
The Work and Security Act in the Netherlands comprised several labour market reforms to simplify the dismissal law. Since July 2015, employers can no longer choose the procedure for dismissal (via the Public Employment Service or the Sub-district Court); it is now determined by the reason for the termination (the Public Employment Service deals with dismissals for economic reasons or long-term disability). In addition, these two procedures were made more comparable: the notice period was extended to termination via the Sub-district Court and severance pay (the “transition allowance”) to termination via the Public Employment Service, including for collective dismissals.
The extension of the notice period and severance pay to all dismissal procedures, and of severance pay to collective dismissals, explains most of the notable increase in the indicator for regular workers (see Figure 3.13, Panel A and Figure 3.14, Panels A, B, E and F).38 The Work and Security Act also increased restrictions to temporary employment by lowering the maximum duration of successive fixed-term contracts from three to two years, slightly increasing the indicator for temporary workers (see Figure 3.13, Panel B).
Employment protection reforms and labour market duality
Employment protection reforms can reduce labour market duality between secure and precarious jobs by lowering the opportunities and incentives for firms to replace regular with temporary contracts. This is the case for example for reforms that restrict the valid cases for use of temporary employment as in Italy 2018. Policy action to reduce labour market duality goes beyond restrictions on the use of temporary employment. In particular, it frequently involves aligning social contributions and working conditions between temporary and regular contracts. Such changes are largely beyond the scope of the indicators.39 For instance, Slovenia’s new Employment Relation Act introduced severance pay and additional social security contributions for fixed-term contracts. Another example is from the Netherlands, where the Balanced Labour Market Act, which entered into force in January 2020, increased the employer’s unemployment insurance contribution rate for fixed-term contracts.40
3.5. Concluding remarks
Employment protection legislation is a widely debated policy and, as this chapter has shown, governments in OECD countries continuously adapt regulations in this field. During the 2013‑19 period, 21 of the 37 OECD countries undertook one or several reforms in employment protection that are reflected in changes in the OECD Employment Protection Legislation indicators. Dismissal and hiring policies involve an inherent trade-off between job security for workers who have a job and firm adaptability to changes in demand conditions or technology. By comparing employment protection legislation in OECD countries, the chapter sheds light on the relative importance that different systems attach to these twin aspirations. The descriptive evidence in the chapter and the new indicators in the OECD Employment Protection Legislation Database can be used as tools to further analyse what design of employment protection may deliver better outcomes than others.41
The chapter gives a detailed description of dismissal regulations for regular workers and hiring regulations for temporary workers in OECD countries, issues that are of particular relevance in the current environment of high dismissal risk and low hiring chances. Subsequent work will look in greater depth at the regulations applying to dismissals of workers on fixed-term contracts and expirations of these contracts, examine the fate of temporary workers during the COVID‑19 crisis and extend the update of the OECD Employment Protection Legislation Database to several non-OECD countries. With greater country coverage and better comparisons of employment protection of regular workers, employment protection of temporary workers and the differences between the two, the aspiration is to make this work the most useful for policy makers and citizens as they decide how they would like to shape job protection provisions in the future.
References
[81] Acharya, V., R. Baghai and K. Subramanian (2014), “Wrongful discharge laws and innovation”, Review of Financial Studies, Vol. 27/1, pp. 301-346.
[72] Andrews, D., M. Adalet McGowan and V. Millot (2017), “Confronting the zombies: Policies for productivity revival”, OECD Economic Policy Papers, No. 21, OECD Publishing, Paris, https://doi.org/10.1787/f14fd801-en.
[34] Andrews, D. and F. Cingano (2014), “Public policy and resource allocation: Evidence from firms in OECD countries”, Economic Policy, Vol. 29/78, pp. 253-296.
[16] Autor, D., J. Donohue and S. Schwab (2006), “The costs of wrongful-discharge laws”, Review of Economics and Statistics, Vol. 88/2, pp. 211-231.
[37] Autor, D., W. Kerr and A. Kugler (2007), “Does employment protection reduce productivity? Evidence from US states”, Economic Journal, Vol. 117/521, pp. F189-F217.
[38] Bai, J., D. Fairhurst and M. Serfling (2020), “Employment protection, investment, and firm growth”, Review of Financial Studies, Vol. 33/2, pp. 644-688.
[41] Bartelsman, E., P. Gautier and J. De Wind (2016), “Employment protection, technology choice, and worker allocation”, International Economic Review, Vol. 57/3, pp. 787-826.
[9] Bassanini, A. and E. Caroli (2015), “Is work bad for health? The role of constraint versus choice”, Annals of Economics and Statistics 119/120, pp. 13-37.
[32] Bassanini, A. and F. Cingano (2019), “Before it gets better: The short-term employment costs of regulatory reforms”, Industrial and Labor Relations Review, Vol. 72/1, pp. 127-157.
[25] Bassanini, A. and A. Garnero (2013), “Dismissal protection and worker flows in OECD countries: Evidence from cross-country/cross-industry data”, Labour Economics, Vol. 21/C, pp. 25-41.
[42] Bassanini, A., L. Nunziata and D. Venn (2009), “Job protection legislation and productivity growth in OECD countries”, Economic Policy, Vol. 24/58, pp. 349-402.
[64] Batut, C. and E. Maurin (2019), “From ultima ratio to mutual consent: The effects of changing employment protection doctrine”, IZA Discussion Papers, No. 12440.
[12] Bentolila, S. and G. Bertola (1990), “Firing costs and labour demand: How bad is Eurosclerosis?”, Review of Economic Studies, Vol. 57/3, pp. 381-402.
[43] Bjuggren, C. (2018), “Employment protection and labor productivity”, Journal of Public Economics, Vol. 157(C), pp. 138-157.
[27] Boeri, T. (2011), “Institutional reforms and dualism in European labor markets”, Handbook of Labor Economics, Card, D. and O. Ashenfelter (eds.), North Holland, Amsterdam.
[63] Boeri, T., P. Garibaldi and E. Moen (2017), “Inside severance pay”, Journal of Public Economics, Vol. 145/C, pp. 211-225.
[17] Boeri, T. and J. Jimeno (2005), “The effects of employment protection: Learning from variable enforcement”, European Economic Review, Vol. 49/8, pp. 2057-2077.
[76] Boeri, T. and J. van Ours (2013), The Economics of Imperfect Labor Markets, Princeton University Press, Princeton.
[35] Bottasso, A., M. Conti and G. Sulis (2017), “Firm dynamics and employment protection: Evidence from sectoral data”, Labour Economics, Vol. 48(C), pp. 35-53.
[36] Bravo-Biosca, A., C. Criscuolo and C. Menon (2016), “What drives the dynamics of business growth?”, Economic Policy, Vol. 31/88, pp. 703-742.
[8] Cahuc, P., S. Carcillo and A. Zylberberg (2014), Labor Economics, MIT Press, Cambridge, MA.
[55] Cahuc, P., O. Charlot and F. Malherbet (2016), “Explaining the spread of temporary jobs and its impact on labor turnover”, International Economic Review, Vol. 57/2, pp. 533-572.
[29] Cahuc, P., F. Malherbet and J. Prat (2019), “The detrimental effect of job protection on employment: Evidence from France”, IZA Discussion Papers, No. 12384.
[67] Campolieti, M. and C. Riddell (2020), “Does mediation-arbitration reduce arbitration rates? Evidence from a natural experiment”, Industrial and Labor Relations Review, Vol. 73/1, pp. 211-235.
[82] Campos, N. and J. Nugent (2018), “The dynamics of the regulation of labour in developing and developed countries since 1960”, The Political Economy of Structural Reforms in Europe, Campos, N., P. De Grauwe and Y. Ji (eds.), Oxford University Press, Oxford.
[74] Carcillo, S. et al. (2019), “Assessing recent reforms and policy directions in France: Implementing the OECD Jobs Strategy”, OECD Social, Employment and Migration Working Papers, No. 227, OECD Publishing, Paris, https://doi.org/10.1787/657a0b54-en.
[51] Centeno, M. and Á. Novo (2012), “Excess worker turnover and fixed-term contracts: Causal evidence in a two-tier system”, Labour Economics, Vol. 19/3, pp. 320-328.
[46] Ciminelli, G., R. Duval and D. Furceri (2018), “Employment protection deregulation and labor shares in advanced economies”, IMF Working Papers, No. 18/186.
[80] Ciminelli, G. and D. Furceri (2020), “A new database of employment protection legislation”, Unpublished manuscript.
[40] Cingano, F. et al. (2016), “Employment protection legislation, capital investment and access to credit: Evidence from Italy”, Economic Journal, Vol. 126/595, pp. 1798-1822.
[39] Cingano, F. et al. (2010), “The effects of employment protection legislation and financial market imperfections on investment: Evidence from a firm-level panel of EU countries”, Economic Policy, Vol. 25/61, pp. 117-163.
[56] Damiani, M., F. Pompei and A. Ricci (2016), “Temporary employment protection and productivity growth in EU economies”, International Labour Review, Vol. 155/4, pp. 587-622.
[62] Denk, O. and A. Georgieff (forthcoming), “The 2019 OECD Employment Protection Legislation indicators: New insights on job dismissal regulation in OECD countries”, OECD Social, Employment and Migration Working Papers, forthcoming, OECD Publishing, Paris.
[57] Dolado, J., S. Ortigueira and R. Stucchi (2016), “Does dual employment protection affect TFP? Evidence from Spanish manufacturing firms”, SERIEs: Journal of the Spanish Economic Association, Vol. 7/4, pp. 421-459.
[78] Duval, R. et al. (2018), “A narrative database of major labor and product market reforms in advanced economies”, IMF Working Papers, No. 18/19.
[65] Espinosa, R., C. Desrieux and M. Ferracci (2018), “Labor market and access to justice”, International Review of Law and Economics, Vol. 54, pp. 1-16.
[68] Espinosa, R., C. Desrieux and H. Wan (2017), “Fewer courts, less justice? Evidence from the 2008 French reform of labor courts”, European Journal of Law and Economics, Vol. 43/2, pp. 195-237.
[47] Falk, A., D. Huffman and W. Macleod (2015), “Institutions and contract enforcement”, Journal of Labor Economics, Vol. 33/3, pp. 571-590.
[69] Fraisse, H., F. Kramarz and C. Prost (2015), “Labor disputes and job flows”, Industrial and Labor Relations Review, Vol. 68/5, pp. 1043-1077.
[30] García Pérez, J. and V. Osuna (2014), “Dual labour markets and the tenure distribution: Reducing severance pay or introducing a single contract”, Labour Economics, Vol. 29, pp. 1-13.
[58] García-Pérez, J., I. Marinescu and J. Vall Castello (2018), “Can fixed-term contracts put low skilled youth on a better career path? Evidence from Spain”, Economic Journal, Vol. 129/620, pp. 1693-1730.
[13] Garibaldi, P. (1998), “Job flow dynamics and firing restrictions”, European Economic Review, Vol. 42/2, pp. 245-275.
[66] Gianfreda, G. and G. Vallanti (2017), “Institutions’ and firms’ adjustments: Measuring the impact of courts’ delays on job flows and productivity”, Journal of Law and Economics, Vol. 60/1, pp. 135-172.
[18] Gielen, A. and K. Tatsiramos (2012), “Quit behavior and the role of job protection”, Labour Economics, Vol. 19/4, pp. 624-632.
[44] Griffith, R. and G. Macartney (2014), “Employment protection legislation, multinational firms, and innovation”, Review of Economics and Statistics, Vol. 96/1, pp. 135-150.
[3] Grubb, D. and W. Wells (1993), “Employment regulation and patterns of work in EC countries”, OECD Economic Studies, Vol. 21, pp. 7-58, http://www.oecd.org/economy/growth/33942847.pdf.
[19] Haltiwanger, J., S. Scarpetta and H. Schweiger (2014), “Cross country differences in job reallocation: The role of industry, firm size and regulations”, Labour Economics, Vol. 26/C, pp. 11-25.
[52] Hijzen, A., L. Mondauto and S. Scarpetta (2017), “The impact of employment protection on temporary employment: Evidence from a regression discontinuity design”, Labour Economics, Vol. 46, pp. 64-76.
[70] Ichino, A., M. Polo and E. Rettore (2003), “Are judges biased by labor market conditions?”, European Economic Review, Vol. 47/5, pp. 913-944.
[83] Ichino, A. and R. Riphahn (2005), “The Effect of Employment Protection on Worker Effort: Absenteeism during and after Probation”, Journal of the European Economic Association, Vol. 3/1, pp. 120-143, http://dx.doi.org/10.1162/1542476053295296.
[79] ILO (2015), Employment Protection Legislation: Summary Indicators in the Area of Terminating Regular Contracts (Individual Dismissals), International Labour Office, Geneva.
[71] Immervoll, H. and C. Knotz (2018), “How demanding are activation requirements for jobseekers? New evidence on activity-related eligibility criteria for unemployment and social assistance benefits”, OECD Social, Employment and Migration Working Papers, No. 215, OECD Publishing, Paris, https://dx.doi.org/10.1787/2bdfecca-en.
[53] Kahn, L. (2010), “Employment protection reforms, employment and the incidence of temporary jobs in Europe: 1996-2001”, Labour Economics, Vol. 17/1, pp. 1-15.
[48] Leonardi, M. and G. Pica (2013), “Who pays for it? The heterogeneous wage effects of employment protection legislation”, Economic Journal, Vol. 123/573, pp. 1236-1278.
[20] Marinescu, I. (2009), “Job security legislation and job duration: Evidence from the United Kingdom”, Journal of Labor Economics, Vol. 27/3, pp. 465-486.
[28] Martin, J. and S. Scarpetta (2012), “Setting it right: Employment protection, labour reallocation and productivity”, De Economist, Vol. 160/2, pp. 89-116.
[49] Martins, P. (2009), “Dismissals for cause: The difference that just eight paragraphs can make”, Journal of Labor Economics, Vol. 27/2, pp. 257-279.
[31] Messina, J. and G. Vallanti (2007), “Job flow dynamics and firing restrictions: Evidence from Europe”, Economic Journal, Vol. 117/521, pp. F279-F301.
[21] Micco, A. and C. Pagés (2006), “The economic effects of employment protection: Evidence from international industry-level data”, IZA Discussion Papers, No. 2433.
[22] Millán, A. et al. (2013), “How does employment protection legislation influence hiring and firing decisions by the smallest firms?”, Economics Letters, Vol. 121/3, pp. 444-448.
[14] Mortensen, D. and C. Pissarides (1994), “Job creation and job destruction in the theory of unemployment”, Review of Economic Studies, Vol. 61/3, pp. 397-415.
[45] Murphy, G., I. Siedschlag and J. McQuinn (2017), “Employment protection and industry innovation”, Industrial and Corporate Change, Vol. 26/3, pp. 379-398.
[10] Neal, D. (1995), “Industry-specific human capital: Evidence from displaced workers”, Journal of Labor Economics, Vol. 13/4, pp. 653-677.
[15] Nickell, S. (1978), “Fixed costs, employment and labour demand over the cycle”, Economica, Vol. 45/180, pp. 329-345.
[26] OECD (2018), Good Jobs for All in a Changing World of Work: The OECD Jobs Strategy, OECD Publishing, Paris, https://doi.org/10.1787/9789264308817-en.
[11] OECD (2018), OECD Employment Outlook 2018, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2018-en.
[75] OECD (2018), OECD Reviews of Labour Market and Social Policies: Lithuania, OECD Publishing, Paris, https://doi.org/10.1787/9789264189935-en.
[54] OECD (2017), OECD Employment Outlook 2017, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2017-en.
[33] OECD (2016), OECD Employment Outlook 2016, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2016-en.
[60] OECD (2014), OECD Employment Outlook 2014, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2014-en.
[6] OECD (2013), OECD Employment Outlook 2013, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2013-en.
[59] OECD (2011), Divided We Stand: Why Inequality Keeps Rising, OECD Publishing, Paris, https://doi.org/10.1787/9789264119536-en.
[23] OECD (2010), OECD Employment Outlook 2010, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2010-en.
[5] OECD (2004), OECD Employment Outlook 2004, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2004-en.
[4] OECD (1999), OECD Employment Outlook 1999, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-1999-en.
[1] Pissarides, C. (2010), “Why do firms offer ‘employment protection’?”, Economica, Vol. 77/308, pp. 616-636.
[61] Ruiz-Valenzuela, J. (2020), “Intergenerational effects of employment protection reforms”, Labour Economics, Vol. 62.
[77] Saint-Paul, G. (2014), Frictions and Institutions, Bookboon, London.
[24] Salvanes, K. (1997), “Market rigidities and labour market flexibility: An international comparison”, Scandinavian Journal of Economics, Vol. 99/2, pp. 315-333.
[2] Scarpetta, S. (2014), “Employment protection”, IZA World of Labor, No. 12.
[73] Söderqvist, F. and M. Lindberg (2019), “A technical note on improving the granularity and scope of the OECD EPL Index”, Paper presented at the OECD Expert Workshop on employment protection legislation, Paris, 27 June 2019.
[84] Turrini, A. et al. (2015), “A decade of labour market reforms in the EU: Insights from the LABREF database”, IZA Journal of Labor Policy, Vol. 4/12.
[50] van der Wiel, K. (2010), “Better protected, better paid: Evidence on how employment protection affects wages”, Labour Economics, Vol. 17/1, pp. 16-26.
[7] Venn, D. (2009), “Legislation, collective bargaining and enforcement: Updating the OECD employment protection indicators”, OECD Social, Employment and Migration Working Papers, No. 89, https://doi.org/10.1787/223334316804.
Annex 3.A. Methodology
Annex Table 3.A.1. Structure of Version 4 of the OECD EPL indicators for dismissing regular workers
Note: Where there are differences by types of dismissal, the scored value is the average of the values for a dismissal for personal and for economic reasons. Where there are differences by firm size, the scored value is the average of the values for a firm with 35, 150 and 350 employees. Approximately the same coding is used for individual and collective dismissals; the only differences are that, as collective dismissals can only occur for economic reasons, Items 5d and 6 are removed, and warning procedures are not taken into account in Item 1. The indicator for collective dismissals evaluates the average for dismissals of 10, 45 and 120 workers by a firm within one month (based on the average of the three firm sizes for 10 dismissals, the two larger firm sizes for 45 dismissals and the largest firm size for 120 dismissals).
Annex Table 3.A.2. Weighting in the OECD EPL indicators (Version 4) for dismissing regular workers
Annex Table 3.A.3. Structure of Version 3 of the OECD EPL indicator for hiring temporary workers
Note: Where there are differences by firm size, the scored value is the average of the values for a firm with 35, 150 and 350 employees.
Annex Table 3.A.4. Weighting in the OECD EPL indicators (Version 3) for hiring temporary workers
Notes
← 1. Overviews of the main theoretical frameworks can be found in academic books such as Boeri and van Ours (2013[76]), Cahuc, Carcillo and Zylberberg (2014[8]) and Saint-Paul (2014[77]).
← 2. However, a positive job tenure-protection profile tends to be efficient in the case of jobs requiring continuous firm-specific investments by workers (Boeri, Garibaldi and Moen, 2017[63]).
← 3. A notable exception is Acharya, Baghai and Subramanian (2014[81]) who provide evidence for the United States that the passage of wrongful discharge laws spurred innovation.
← 4. However, fixed-term contracts have been shown to induce higher effort by workers if they expect a high probability of conversion of their contract into an open-ended one (Ichino and Riphahn, 2005[83]).
← 5. The collection of the information and data for the OECD Employment Protection Legislation Database is the result of a large collaborative effort by the labour ministries of OECD countries and the OECD Secretariat, which remains however solely responsible for the indicators. The project benefited as well from the insightful comments and suggestions by a group of academic and policy experts.
← 6. One interesting avenue for future research would be to evaluate employment protection for public-sector employees, also in comparison to employment protection for private-sector employees.
← 7. The only exception is the category notice and severance pay, in which severance pay carries a slightly higher weight (4/7) than notice period (3/7). The rationale is that workers can still contribute to the firm’s output while on notice. Consequently, the net cost to the firm is higher for a month of severance pay than for a month of advance notice. Yet, advance notice may protect workers better than severance pay, in particular if early support by the public employment services is in place (OECD, 2018[11]).
← 8. The length of the trial period enters only the employment protection indicator for individual dismissals and not the indicator for collective dismissals, as trial periods are always associated with an individual worker.
← 9. This is not the case for the employment protection indicator for collective dismissals, because collective dismissals can only occur for economic reasons.
← 10. As mentioned, where there are differences by firm size, the scored value is the average of the values for a firm with 35, 150 and 350 employees. The indicator for collective dismissals is based on the average of the three firm sizes for 10 dismissals, the two larger firm sizes for 45 dismissals and the largest firm size for 120 dismissals.
← 11. As 120 workers exceeds the dismissal threshold in all countries with specific legislation for collective dismissals (but not 10 and 45 workers), the regulatory differences between individual and mass dismissals enter the indicator for collective dismissals with a minimum weight of around one‑third.
← 12. An alternative approach is to focus on reforms in employment protection. Duval et al. (2018[78]) compile a dataset of major reforms in this area with a 0‑1 reform variable covering 26 advanced economies over the period 1970‑2015. The Labour Market Reforms (LABREF) Database by the European Commission monitors qualitatively job protection reforms in the 27 EU countries since 2000 (Turrini et al., 2015[84]).
← 13. Other researchers have developed their own databases, among them Campos and Nugent (2018[82]) and Ciminelli and Furceri (2020[80]).
← 14. ILO (2015[79]) shows that the OECD indicator for regular workers against individual dismissals is strongly correlated with the EPLex (the correlation coefficient equals 0.81). Calculations by the OECD suggest a significant, albeit weaker, correlation with a constructed CBR-LRI indicator focusing on dismissal regulation (0.66).
← 15. The CBR-LRI and the Doing Business indicators go well beyond employment protection, which is only one focus of these databases in terms of institutional coverage.
← 16. The United States has a rising number of cases in which employees pursue wrongful termination claims by alleging that the dismissal was based on an “implied contract” for continued employment (despite no formal contract), because certain assurances for continued employment were given. The probability of succeeding in claiming the existence of an “implied contract” increases with seniority and is likely to be extremely low or zero for only a few years of job tenure.
← 17. In Sweden, a strict last-in-first-out rule applies except when collective agreements establish otherwise (see Box 3.2).
← 18. For example, Poland has a special type of fixed-term contracts which in practice is used as a trial period. Belgium allows the use of temporary work agency employment for “insertion”, i.e. to be hired under a regular contract for the same job after the assignment.
← 19. In the United States, the prescription period in the situation of an “implied contract” depends on the state jurisdiction, but it is typically very long (for example four years in California and six years in New York).
← 20. In the United States, when an employer offers voluntary separations but intends to implement involuntary dismissals in case of an insufficient number of volunteers, employees who take the voluntary separation package may be eligible for unemployment benefits.
← 21. In cross-country regressions of the aggregate indicator on the four broad categories (one at a time), enforcement of unfair dismissal regulation is positively correlated with the aggregate indicator, but not in a statistically significant fashion. The other three categories are positively correlated with the aggregate indicator, statistically significant at the 1% level. The estimate is also much lower for enforcement of unfair dismissal regulation than for the other three categories.
← 22. The pairwise negative correlations are not statistically significant. In line with the reasons provided, enforcement is more negatively correlated with procedural requirements and the regulatory framework for unfair dismissals.
← 23. While for individual dismissals this is only the case in the Netherlands and, to some extent, in Germany and Sweden, for collective dismissals other countries as well require an authorisation pre-dismissal that can serve as validation post-dismissal.
← 24. While additional notifications are required in Israel (to the Employment Service Bureau) and Korea (to the Ministry of Labour), they do not impose significant additional constraints on firms and hence are not reflected in the indicators.
← 25. The way the indicators evaluate regulation of collective dismissals is by the stringency that would apply to one worker if this worker was dismissed by a collective dismissal. While this definition does not allow for fixed (in particular procedural) costs being spread across several workers, it is not obvious how the design of the indicators could, in a pragmatic way, take account of this possible fixed cost nature of some of the aspects relating to dismissal regulation.
← 26. These are Canada and the United States (North America) and Australia and New Zealand (Australasia).
← 27. They can also affect restrictions on the use of temporary contracts (e.g. in France, Italy, the Netherlands, Sweden).
← 28. These are calculations for the whole business economy using the OECD Structural and Demographic Business Statistics Database.
← 29. By contrast, the clearer definition of fair economic reasons for dismissals on economic grounds had little impact; it limits the freedom that judges have in their decision to classify a dismissal as unfair, but this accounts for only one‑sixth of the score for the lower-level element “Definition of unfair dismissal”. Moreover, the new schedule for compensation following an unfair dismissal does not sufficiently reduce the compensation to be reflected in the indicators. Finally, the introduction of the rupture conventionnelle collective also did not change the indicators, as it only formalised the existing voluntary separation plans which already granted eligibility for unemployment benefits.
← 30. Initially, the Jobs Act set a schedule of two months of wages per year of tenure for compensation following an unfair dismissal for these types of dismissal. However, the Constitutional Court rejected this in a November 2018 decision. The indicators do not reflect this temporary change, as it is difficult to assess the extent to which the schedule was enforced and for how long. However, they reflect the new range of possible values for compensation, which now lie between 6 and 36 months of wages, compared with 12 and 24 before the Jobs Act. The resulting increase in the indicators mitigates the downward effect of the Jobs Act.
← 31. The contribution of the employer was higher in the case of use of Cassa Integrazione Guadagni (the Italian short-time work scheme) before the collective dismissal.
← 32. The changes in notice period and severance pay induced by the different reform measures, including the introduction of dismissals at will, exactly cancel each other out in the indicators, so that no change is visible on the component notice and severance pay (see Figure 3.14, Panel B).
← 33. In May 2014, the obligations to attempt a transfer before dismissal and to follow criteria provided by the Labour Code when selecting the worker(s) to be dismissed were reintroduced, to comply with the September 2013 ruling of the Constitutional Court. These obligations had been lifted in August 2012. The indicators do not reflect these temporary changes, as it is difficult to assess the extent to which these were enforced and for how long.
← 34. In particular, it is the employer’s responsibility to show the contents of any collective agreement or internal document required for adjudicating the dispute.
← 35. Moreover, a severance pay subsidy in the case of fair dismissals in firms with less than 25 employees was suppressed in December 2013 (OECD, 2016[33]).
← 36. In the absence of effective short-time work schemes, excessively strict job protection may, however, lead to firm bankruptcies, thereby failing to preserve jobs.
← 37. The indicator for temporary contracts in Portugal increases due to the end of an extraordinary regime of renewals applying to fixed-term contracts until December 2016. An additional increase from 2.46 to 2.58 is due to the further reduction in the maximum cumulated duration of fixed-term contracts implemented in September 2019.
← 38. The notice period can only start at the end of the month. Therefore, the extension of the notice period to all dismissal procedures also increased the delay before notice can be given.
← 39. Still, the indicators include one item on equal treatment of regular and temporary work agency workers, which makes it possible to reflect part of these changes. For example, in Québec (Canada), since June 2018, employers can no longer set different wages for workers solely based on their employment status, and temporary work agencies are prevented from paying wages that are lower than those of regular employees performing the same tasks. Similarly, in Iceland, since 2013, temporary work agency workers should be paid at least the same wages and benefit from the same facilities as regular employees. In the Netherlands, the possibility to deviate from the basic wage during the first 26 weeks of a temporary work agency assignment has been removed from the Collective Labour Agreement for Temporary Agency Workers. These changes explain the increase in the indicators for temporary work agency employment for these countries (see Figure 3.15, Panel B).
← 40. The Balanced Labour Market Act also made dismissals easier (employers can now combine personal reasons for dismissal that are by themselves insufficient to justify a dismissal), aligned severance pay for permanent and temporary workers, modified their calculation rules and extended the maximum cumulated duration for fixed-term contracts. The reform did not change the indicator of protection for regular workers, because the indicator does not capture the possible combination of insufficient grounds and the opposite changes in severance pay at different tenures cancel each other out. However, it reduced the indicator for temporary workers from 1.48 to 1.23.
← 41. The chapter mostly focuses on developments since 2013. Researchers and other users interested in the complete time series of the data and information available since 1985 can find these on the dedicated website (http://oe.cd/epl). Using these data, Denk and Georgieff (forthcoming[62]) depict long time series for employment protection.