98274 BALANCING FLEXIBILITY AND WORKER PROTECTION UNDERSTANDING LABOR MARKET DUALITY IN POLAND October 2014 Social Protection and Labor Global Practice 1 Abstract This paper analyzes the legal and actual gaps in working conditions and returns between permanent and temporary contracts, with a focus on civil law contracts, which are at the center of the debate on widening labor market duality in Poland. We find that net employment creation in recent years occurred largely through involuntary temporary contracts, especially among the low-skilled and young people. Even accounting for workers heterogeneity, we find a persistent wage gap between permanent and temporary contracts, and limited mobility across contract typologies. Insights from qualitative evidence suggest that narrowing duality without hurting the most vulnerable workers will require a reduction in administrative burdens and implicit costs associated with permanent labor contracts, while eliminating firms’ incentives to strictly prefer temporary contracts. Acknowledgements: This note was prepared by Roberta Gatti, Karolina Goraus, and Matteo Morgandi, with Ewa Korczyc and Jan Rutkowski. Jan Rutkowski managed the qualitative analysis summarized in Uhl (2014). We thank DM Jacek Męcina, Bogumił Czerwiński, Małgorzata Sarzalska, Monika Siergiejuk (Ministry of Labor), and Gordon Betcherman and Marcel Jansen for detailed comments. We are grateful to Elsa Fornero, Anke Hassel, Marcel Jansen, Pietro Ichino, Piotr Lewandowski, and Jan Svejnar for productive discussions. Cite as: Gatti R., Goraus K., and Morgandi, M (2014). Balancing Flexibility and Worker Protection. Understanding Labor Market Duality in Poland. The World Bank, Washington DC 2 Table of Contents ACKNOWLEDGEMENTS 2 ABBREVIATIONS 4 1. INTRODUCTION AND KEY MESSAGES 5 2. LABOR MARKETS IN POLAND: THE BROADER CONTEXT 7 3. THE ROLE OF TEMPORARY EMPLOYMENT IN POLAND 11 4. WHY IS TEMPORARY EMPLOYMENT USED AND WHY IS IT ON THE RISE IN POLAND? 18 5. WHO WORKS WITH TEMPORARY CONTRACTS IN POLAND? 21 6. IS TEMPORARY EMPLOYMENT IN POLAND (IN)VOLUNTARY? 25 7. TEMPORARY EMPLOYMENT: A STEPPING STONE OR A NEW REALITY? 29 8. BALANCING FLEXIBILITY WITH FAIRNESS IN THE POLISH LABOR MARKET 31 REFERENCES 36 ANNEX 39 3 Abbreviations CC civil law contract EPL employment protection legislation EU European Union FTLC fixed-term labor contract GDP gross domestic product LFP labor force participation LFS Polish Labor Force Survey OECD Organisation for Economic Co-operation and Development 4 1. Introduction and Key Messages The perceived gap in labor market outcomes and conditions between those who are employed with labor contracts and those who are employed with atypical contracts is a subject of much debate in Poland and the central concern of this paper. Civil contracts (CCs), which are flexible forms of employment not regulated by the labor code, carry a narrow set of benefits, no protection against dismissal, and limited or no accrual of pension rights. Yet, together with fixed-term labor contracts (FTLCs), they have been a substantial part of net employment creation in the country in the past decade. A secondary concern of the paper regards the increasing use of FTLCs as a substitute for open-ended labor contracts. While FTLCs carry the same set of social security rights, the overutilization of FTLCs appears to be a reaction to the excessive rigidity of open-ended contracts. Based on existing available data and newly collected qualitative evidence, this note describes the main features of atypical labor contracts and the characteristics of the workers associated with them. It discusses why these contracts are increasingly used and the advantages and disadvantages they present to workers and society as a whole. The note concludes by presenting policy options to balance labor market flexibility with a reduction in the existing gap between labor and civil contracts in the context of the Polish labor market. Box 1. A note on terminology The central concern of this paper is the duality between civil and labor contracts. However, data allow us mostly to measure temporary employment, which encompasses both civil contracts (CCs) and Fixed Term Labor Contracts (FTLCs). As such, this paper refers to "temporary employment" whenever inferences are possible only regarding the combination of civil and fixed-term labor contracts, and to "civil contracts" when data or information refer directly to this subcategory of employment. Section 3.1 elaborates in more detail on the features of these contracts in contrast to labor contracts. The National Statistical Office GUS is fielding a module of the LFS which will allow the differentiation between civil and labor contracts. Once these data are available, the statistical analysis conducted in this paper could be updated to validate and refine results. This paper has four main findings: 1) While Poland did better than other comparator countries in employment, including during the 2008–2009 economic crisis, net employment creation in the past decade was largely in terms of temporary jobs. In Poland, the share of temporary employment has been steadily increasing from the early 2000s, up until the financial crisis, when it stabilized at around 25 to 27 percent of wage employment (about 3.2 million individuals), the highest share among the 5 EU-28. Just as importantly, the work of about 1.4 million individuals is regulated by civil law contracts (CCs), which offer a limited set of benefits for workers. Most temporary work is in the private service sector and in industry, and CCs are mostly concentrated in administrative and support services. 2) Temporary contracts are mostly used among youth (especially young women), older workers (55+), and low-skilled workers. In most cases these are involuntary employment arrangements. The distribution of temporary employment is highly driven by age: temporary contracts are often viewed as an entry to labor markets in lieu of probation for young workers. In all age categories, these contracts are especially used for workers with lower education—in 2012, 40 percent of the workers with primary education were in temporary contracts. 3) The vast majority of young people are in these jobs because they have no other option, while as many as 40 percent of older workers are in temporary contracts voluntarily. In addition, the reliance on temporary employment is not always a stepping stone to more stable employment; the data show that between 2008 and 2013, only one out of three workers employed with a temporary contract moved to permanent employment. When compared with countries such as Norway, the U.K., and even Spain, the probability of moving to a more stable job is low. This fact can be of particular concern for young people, especially given that temporary employment is associated with employers making lower investments in workers’ skills and training. Overall, temporary employment is increasingly becoming the norm in Polish labor markets, with the average tenure of temporary employment lengthening significantly. Interestingly, temporary employment is an effective way to bring older workers into the labor market in a way that is aligned with their preferences. This is an important finding in light of Poland’s low participation to the labor force, particularly so among older workers. Finally, temporary workers—even after controlling for differences in a number of characteristics—earn on average 15 percent less than permanent workers. However, this average actually masks two opposite situations: those workers who report to be involuntarily employed with temporary contracts actually receive 20 percent lower earnings, while those who are voluntarily employed with temporary contracts earn 4 percent more than permanent workers. 4) Low transitions, an important wage gap, and a high share of involuntary temporary employment point to a significant gap between permanent and temporary employment. Reducing this gap entails a multipronged strategy that increases the cost of using flexible contracts; reduces the cost and complexity associated with labor contracts (both permanent and temporary); and strengthens workers’ protection during transition. This requires first making civil contracts subject to the same social security contributions as labor contracts. It also requires limiting the use of flexible contracts to instances in which they are justified by the irregular or temporary nature of a job, which calls for strengthening the enforcement of existing regulations and giving greater power to the Labor Inspectorate. Second, the costs of labor contracts need to be lowered. This primarily requires simplifying labor regulation and increasing awareness around it, especially among small and medium firms. It could also entail lowering the procedural cost of dismissals (including the expected costs associated with judicial review), 6 which employers perceive as an important barrier. One way to do this is to institute a severance pay that increases with job tenure in lieu of the “just cause� requirement and the judicial review process. There is also room to rethink other aspects of labor code contracts design, such as the structure of the probation period, which is seen as too constraining in its ability to lead to an open-ended contract after three months. Finally, international experience shows that these measures need to be accompanied by strong social protection. 2. Labor Markets in Poland: The Broader Context Upon joining the European Union (EU) in 2004, policy reforms and economic integration with Europe fueled a remarkable expansion in Poland. The country recorded the strongest economic expansion among the new member states, and this growth brought about rapid convergence to EU income levels: GDP per capita increased from around 50 percent of the EU average in 2004 to 67 percent in 2012. When Poland joined the EU in 2004, its labor market faced significant challenges; an unfinished restructuring agenda translated into a high share of population working in low productivity sectors and a high structural unemployment rate. In addition, early retirement schemes and easy access to disability benefits created substantial disincentives to participate in the labor force. The EU accession process has been recognized as an important catalyzer of economic reforms. These reforms helped to foster continued reallocation of labor from agriculture to more productive sectors, while better education outcomes facilitated the transition from low to higher skilled occupations. Rapid economic growth resulted in improvements in some labor market outcomes. The unemployment rate dropped significantly from nearly 20 percent in 2004 to 9.6 percent in 2008, at the peak of the economic cycle (Figure 1). Government policies, in particular those that tightened the criteria for obtaining disability benefits in early 2000s and limited access to early pensions in 2008, helped boost certain groups’ labor force participation. At the same time, the opening of EU labor markets to Polish workers resulted in sizeable migration flows, especially to the U.K. and Ireland. The 2009 economic slowdown deteriorated labor market conditions. Unemployment started to pick up gradually, especially among young and lower skilled workers, which led to a large increase in the share of long-term unemployed (Figure 2). 7 Figure 1. Unemployment Rate in Poland, 2004–2012 (%) Figure 2. Increase in Unemployment Rates by Different Groups of Unemployed, 2009–2012 (%) male female unemployment rate Overall unemployment rate Youth Low-skilled Elderly 25 12 20 10 8 15 6 10 4 5 2 0 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 EU27 PL Source: Eurostat, World Bank staff calculations. Between 2001 and 2012, the size of the working age (15–64) population remained fairly stable, with a slight decrease from 25.9 to 25.7 million. However, Poland’s employment structure has changed substantially in the past decade. Positive economic performance resulted in a decreasing number of people remaining out of work. As a result, their share in the working age population decreased from 46 percent (in 2001) to 40 percent (in 2012). The number of self- employed people also decreased, but the change was much smaller—by 1 percentage point.1 The number of dependent employees increased from 10.2 million in 2001 to 11.9 million in 2012. This increase was overwhelmingly driven by temporary employment—the number of temporary dependent workers increased by 2 million people, while the number of permanent workers decreased by 0.3 million (Figure 3). The growth in temporary contracts took place largely in the early 2000s when Poland experienced high unemployment, though it did not contract again during the following years of economic growth. 1 Within the group of self-employed people there were important changes: the number of self-employed people in agriculture sharply declined, while the number of self-employed people outside of agriculture increased. It should also be mentioned that in both 2001 and 2012, 2 percent of the working age population is classified by Eurostat as working, but they are neither included in the separate estimates of dependent employment nor self-employment. For transparency, in Figure 3 we add those persons (who are possibly working as helping family members) to the group of self-employed. 8 Figure 3. Structure of Poland’s Working Age Population 2001 Not in employment Self-employment Permanent dependent employment Temporary dependent employment 14% 35% 40% 46% 5% 2012 Not in employment Self-employment Permanent dependent employment Temporary dependent employment 13% 34% 47% 40% 13% Source: Eurostat. 9 Notwithstanding the positive employment growth track record, Polish labor markets face a number of substantive challenges. The workforce is rapidly aging, and labor force participation among older workers—and especially women—continues to be substantially lower than in comparator countries (Figure 4). Long-term unemployment is on the rise, especially among young and low-skilled workers (Figure 5). Regional differences persist, which also reflects the limited internal mobility of workers. Finally, in-work poverty is relatively high—the fifth highest in the EU. Figure 4. Activity Rates Among People Aged 55–64 for Males and Females in Selected EU Countries, 2012 Males Females 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Romania Greece Latvia Germany Portugal Austria Croatia Slovenia Spain Slovakia Lithuania France Belgium Hungary Italy POLAND Estonia Bulgaria Czech Republic Hungary Bulgaria Germany Latvia Portugal Austria Croatia Slovenia Spain Slovakia Greece Lithuania Romania France Belgium Italy POLAND Estonia Czech Republic Source: Eurostat. Figure 5. Labor Force Participation (LFP) Rates, Unemployment Rates, and Long-Term Unemployment (as Percentage of Total Unemployment) for the Population Aged 15–64 (Poland and EU-15), 2000–2012 LFP Unemployment Long-term unemployment 74 25 60 EU 15 72 55 20 PL 70 50 68 15 45 66 40 64 10 35 62 30 5 60 25 58 0 20 2000 2002 2004 2006 2008 2010 2012 2000 2002 2004 2006 2008 2010 2012 2000 2002 2004 2006 2008 2010 2012 Source: Eurostat. 10 3. The Role of Temporary Employment in Poland Among these key challenges, Poland stands out internationally for the substantial share of temporary contracts among the employed. At about double the EU average, temporary employment in 2012 was the highest in Poland among all EU countries (Figures 6 and 7).2 Figure 6. Share of Temporary Employment (%) in EU Countries, 2012 30 EU 28 20 10 0 Czech… Netherl… United… Lithuania Poland Belgium Latvia Romania Sweden Bulgaria Greece Austria Portugal Hungary Slovakia Estonia Finland France Italy Spain Denmark Slovenia Germany Ireland Croatia Source: Eurostat. Figure 7. Share of Temporary Employment (%) in Poland and EU-15 Over Time 30 20 10 0 Poland EU 15 Source: Eurostat. Overall, the growth in temporary employment represented the totality of net employment growth in the past decade. This growth has been associated with a significant structural transformation from agriculture to services, as well as substantial churning of open-ended wage employment (Figure 8). The substitution of permanent workers with temporary ones, even in industry, is particularly striking. The international evidence shows that employment tends to grow after hiring regulations become more flexible, since firms adjust their labor force along the margin of temporary contracts. When a crisis intervenes, however, firms also adjust their total labor force downward, especially at the expense of temporary workers, exposing them to substantial vulnerability. In Poland, temporary employment grew significantly in period of 2 According to Eurostat, “Employees with temporary contracts are those who declare themselves as having a fixed term employment contract or a job which will terminate if certain objective criteria are met, such as completion of an assignment or return of the employee who was temporarily repla ced.� As per the Polish LFS, every employee that does not have an open-ended labor code contract is defined as a temporary worker. 11 increasing unemployment (2000–2004). Yet, when unemployment started to decrease, duality in the labor market remained entrenched, likely due to competition on labor costs, since the use of CCs allows a significant cost containment and might force other firms to follow suit or exit the market.3 Figure 8. Change in Dependent Employment, 2001–2012 Total Services Construction Temporary Industry Permanent Agriculture -1000 -500 0 500 1000 1500 2000 2500 thousands Source: Eurostat, World Bank staff calculations. 3.1 An Overview of Employment Relations in Poland Behind the broad term “temporary employment� used in the Polish Labor Force Survey (LFS) lies a heterogonous set of contracts (Figure 9). These could be broadly distinguished among (i) term contracts regulated by the labor code; (ii) CCs; and (iii) collaborations of firms with self- employed workers. Each type of employment relationship determines the mutual obligations of employers and employees regarding working time and conditions, termination, social security contributions, and other benefits. Table 1 summarizes employment relations in Poland. Throughout this paper the expression “temporary employment� refers to any dependent employment arrangement except for permanent contracts. As such, this taxonomy, which is driven by the classification of the LFS, lumps together FTLCs and CCs, which are very different in terms of benefits and accrued rights. Whenever evidence—qualitative or quantitative—allows us to single out features or dynamics of CCs, it will be explicitly indicated. Contracts of employment under the labor code are in writing and characterized by the following set of provisions: (i) the employee is subordinate to the employer—that is, the employee should follow instructions while he/she performs duties; (ii) the employee is obliged to perform his or her work in the place specified by the employer; and (iii) the employee is obliged to observe the 3 We are grateful to Marcel Jansen for this observation. 12 working time. If the employment relationship fulfills the provisions of the labor code contract, the employer is legally bound to offer such a contract to the employee. There are several types of labor code contracts of employment, and these differ depending on their duration. Contracts for a trial period, a fixed period, or for the time of completion of a specified task are time-bound, and their duration is set in the contract of employment, which needs to legally adhere to the labor code provisions. For example, the contract for a trial period cannot exceed three months and a second trial contract is not allowed after the first one has expired; in the case of fixed-term contracts, the labor code does not specify the maximum duration, but it limits the number of renewals to two, after which the open-ended contract has to be stipulated. The Polish Labor Code also distinguishes the replacement contract, which is a special kind of contract for a fixed period. In this case the termination date is defined as the moment when the regular employee returns to work. Most of the rules overlap with the standard contract for a fixed period, but there are a few differences that result in less protection of such workers: the notice period is always three days; in case of pregnancy, the contract is not automatically prolonged to the child’s delivery date; and the regulations regarding the number of fixed-term contracts’ renewals do not apply here. Figure 9. Contractual Employment Arrangements in Poland Self- Dependent employment employment Labor code contracts Civil law contracts Contract for Commission a specified Permanent contract Fixed term contracts task contracts (umowa (umowa o zlecenia) dzieło) Contract for Contract for the time of Contract for Temporary a fixed completion period of specified a trial period employment task Source: World Bank staff. 13 Labor code contracts offer numerous benefits to both employers and employees, but are also seen as the most costly and rigid form of employment. By signing a contract of employment, the employer gains a permanent employee and can be certain that the employee will arrive at the workplace at a specified time every day and will observe instructions. The employee, on the other hand, enjoys paid holidays, maternity and child care leave, severance pay when employment is terminated for reasons unrelated to the employee, and the right to effective legal protection against unfair dismissals.4 Labor code contracts are subject to all social security contributions (old age, disability, sickness, health, work accident insurance, and the Labor Fund and Guaranteed Employee Benefits Fund; see Table A1) and legal provision of the labor code in the areas of working time and conditions, redundancy, minimum wage, and other benefits. CCs and self-employment include commission contracts (umowa zlecenie), contracts of result (umowa o dzieło) or cooperation with self-employed persons (samozatrudnienie). Such contracts do not provide the protection defined in labor law provisions, and this is why the law limits their application even though they are very common. At the same time, they give substantial flexibility to employers and employees with regard to working time and conditions, and social security contributions. Just as importantly, a person working under these types of contract does not have the employee rights set out in the labor code. In such contracts there is no element of subordination (which is instead typical of employment relations regulated by the labor code); thus the person carrying out the work under such contract has the freedom to decide how the work should be performed. There is also no statutory minimum wage, no vacation, no severance pay, as well as no limitation on the number of subsequent CCs concluded. Table 1 summarizes some of the key differences between labor contracts and CCs. Table 1. Features of Various Employment Arrangements in Poland Benefits and rights Labor code Civil contracts of workers contracts (permanent or fixed Commission contract Contract of result term) (umowa zlecenie) (umowa o dzieło) Social security Yes Yes, but can be No benefits relatively low Health insurance Yes Yes No Holidays Yes No No Minimum wage Yes No No requirement Period of notice Yes Upon agreement Upon agreement Source: World Bank staff. 4 In companies employing less than twenty labor contract workers, there is no severance pay. 14 Figure 10 below shows the total cost of labor as percentage of net wage for minimum and average wage earners.5 For employers of low-skilled and low-wage workers, civil contracts (especially contracts of result) are a much more cost-efficient option than labor code contracts. In case of higher wages, self-employment is also a cheaper arrangement. Social security contributions, which are a significant part of labor cost, are a total of 32 percent of net wage. Tables A1 and A2 in the annex detail respectively the different components and who pays the cost and obligations in social security contributions and social security benefit obligations for different types of employment and contracts. Figure 10 offers a visual comparison of labor costs across different contracts for a hypothetical job with the same earnings/wages. Figure 10. Employment Arrangements in Poland 200% 180% 160% 140% Other 120% 100% 80% 60% Income tax 40% 20% 0% Health contribution Commission contract Commission contract Labor code contract Labor code contract Conract of result Self-employment Self-employment Conract of result Social security contributions Net wage Minimum wage earner Average wage earner Source: World Bank staff. In 2012, about 1.4 million workers (slightly less than half of temporary employment) worked with contracts that are regulated outside of the labor code, such as contracts of result (umowa o dzieło) or commission contract (umowa zlecenie).6 Compared with 2008 and according to the Polish Panel Survey the share of employees with a labor code open-ended contract was similar in 2008 and 2013,7 but the share of CCs has risen somehow since 2008, from 7 to 11 percent of all wage employment. As discussed above, these contracts provide minimal benefits to workers (no annual or sick leave, no maternity benefits), little or no social security protection, and, by their very design, no job protection against dismissal. For some (umowa o dzieło), a written contract is not even necessary. In much of the common press these are often referred to as “junk contracts.� With the exception of the obligation to pay income tax, the more flexible CCs 5 After recent proposed regulations introducing obligatory social security contributions for commission contract for earnings lower than or equal to the minimum wage. 6 We thank the National Statistical Office (GUS) for sharing these estimates. 7 The Polish Panel Survey is fielded every five years. In the last two waves (2008 and 2013), respondents were asked about the type of contract under which they were working. 15 bear significant similarities to informal work arrangements—a potentially contributing factor to the observed low level of estimated informality in Poland. The earlier discussion indicated that temporary (and among them civil) contracts have played an important role in Poland’s employment dynamic: between 2001 and 2012 the change in net employment in industry and services was 1.7 million workers (17 percent), and that increase was only due to an increased number of temporary workers (1.8 million), as the number of permanent workers in industry and services decreased by 0.1 million. The most rapid switch from permanent employment arrangements to temporary contracts occurred in the economic downturn period at the beginning of 2000s (see Figure 11). Figure 11. Change in Dependent Employment in 2001–2005, 2005–2009, and 2009–2012 2009-2012 Total 2005-2009 2001-2005 2009-2012 Services 2005-2009 2001-2005 Construction 2009-2012 2005-2009 Temporary 2001-2005 Permanent 2009-2012 Industry 2005-2009 2001-2005 Agriculture 2009-2012 2005-2009 2001-2005 -2,000 -1,500 -1,000 -500 0 500 1,000 1,500 2,000 Source: Eurostat, World Bank staff calculations. Temporary (including atypical) contracts have made up an important part of recent employment growth not only in Poland but also in other countries. For instance, the increase in mini-jobs, which have more than doubled since the mid-1990s, has been an important feature of Germany’s strong employment performance (see Figure 12). 16 Figure 12. Different Forms of Employment in Germany, 1996–2010 (1996 = 100) Source: Eichhorst, 2012. Despite the positive impact on employment growth, there are a number of concerns associated with the rising use of temporary contracts:  Temporary contracts that are indefinitely repeated can put workers in a permanent state of precariousness, rather than serving as a stepping stone to more permanent forms of work. This may especially impact vulnerable workers and young people.  The use of temporary contracts limits employers’ incentives and ability to invest in workforce skills and build human capital, with a potential loss of long-term productivity. In the long run, this could incentivize firms to specialize in activities that compete on costs and with lower value-added.  CCs deprive workers, who are de facto wage employees, from essential protections against sickness and unemployment.  The use of these contracts has long-term fiscal costs, because of the implicit debt associated with having a large share of the workforce not contribute to their retirement costs.  Temporary contract workers—especially if vulnerable—absorb in full the risks associated with the business cycle. As recently shown by the case of Spain, in countries with high labor market duality, the contraction on employment happens largely at the expense of temporary workers during crises. 17 The costs associated with temporary employment need to be understood in the broader context of potential counterfactual, which in addition to options of more secure employment could include increased informality and unemployment. 4. Why Is Temporary Employment Used and Why Is It on the Rise in Poland? The widespread utilization of temporary contracts has been a singled out as an important aspect of concern for the Polish labor market and has come to the center of national debate.8 More recently it was the object of a country–specific recommendation by the EU Commission European Semester mechanism for policy coordination, which recommended that Poland should “combat labour market segmentation by stepping up efforts to ensure a better transition from fixed-term to permanent employment and by reducing the excessive use of civil law contracts� (European Commission, 2014). Globalization, increased competition, and technological advances are rapidly changing the nature and composition of the new jobs that are created. In particular, structural transformation and the progressive shift away from industry towards the service sector has implied an increased creation of jobs that are commonly known as “white collar,� as well as new “atypical� professions in the growing service economy. Moreover, the widening variation in the productivity of workers has made it complicated to apply a standard treatment (and the same coverage) to all employees (Ichino, 2014). In the mid-1990s, many European countries responded to this technological change with often piecemeal labor markets reforms, which typically increased flexibility in hiring by introducing a range of contractual forms—including fixed-term contracts regulated by the labor code and flexible contracts regulated by the civil code. This translated into an increasing share of temporary contracts that differed from the mainstream wage employment. The scope of these reforms, which increased flexibility in hiring without easing dismissal provisions, was driven by important political economy dynamics in the dialogue among traditional social partners. In countries with otherwise strong employment protection, such as in Europe, Italy or Spain, these 8 This topic is the object of significant research in Poland. See for example Ministry of Labour and Social Policy, Department of Economic Analyses and Forecasts, 2008; Arak et al., 2014; Kiersztyn, 2014; Bednarski and Frieske, 2012; Baranowska et al., 2001. There is also an ample debate in the media, for example, “Pole wants permanent employment,� http://wyborcza.biz/biznes/1,100896,15881008,Polak_chce_etatu.html#MT; “Target—keep permanent employment,� http://gazetapraca.pl/gazetapraca/1,90443,14353440,Cel__utrzymac_etat.html; “Youths without permanent employment and opportunities. Lost generation,� http://wyborcza.biz/biznes/1,100896,15713351,Mlodzi_bez_etatu_i_bez_szans__Stracone_pokolenie_.ht ml#ixzz31Sk0uRpC; and “Junk contracts are a financial burden for credit and independent life,� http://wyborcza.biz/finanse/1,105684,15719462,Umowy_smieciowe_to_zyciowa_blokada_finansowa__T rudniej.html. 18 reforms resulted in an increase in labor market segmentation/duality between insiders— typically prime age males—and outsiders—women and youth.9 The challenge is to guarantee that the flexible arrangements are used where they are necessary and in the interest of employers and employees. Problems arise when these contracts offer important advantages in terms of flexibility or costs in regular activities that used to be covered by open-ended contracts. The creation of new technology and globalization-driven jobs suggests a taxonomy of employment, whereby job features—typical and atypical—could be matched to contract features (see Table 2). The top left and the bottom right boxes configure a priori good matches between jobs and contracts. On the other hand, the bottom left box delineates a mismatch, one where atypical contracts (in Poland these would be CCs for project or commission), are improperly used to define labor relationships that, by their characteristics (in terms of economic dependence or effective subordination to the employer) should be regulated by labor code contracts. These abuses are likely to result in costs borne by employees (in terms of lack of benefits or job security). Conversely, the top right box would comprise those cases in which labor code contracts regulate atypical forms of work, which are likely to introduce excessive rigidity for firms and employers. Table 2. Taxonomy of Employment Arrangements and Types of Jobs Jobs Atypical: Contract type Typical (unsupervised, deliverable oriented) Labor code contracts Mismatch (inefficiency, cost Match (open-ended and fixed-term) borne by the employer) Mismatch (abuse, cost borne CCs (and self-employment) Match by the employee) Source: World Bank staff. Many factors can explain the increasing use of CCs in Poland. Qualitative evidence from focus groups with employers indicates that firms rely on civil contracts primarily to cut costs: the lack of social security resulted so far in a cost advantage for certain contracts, as do—and significantly so—the lack of employers’ obligations for annual, maternity, or sick leave. These 9 Labor market duality is a particular two-segment case of labor market segmentation. As described by Fields (2007): “the idea that (i) the labor market consists of various segments with qualitatively distinct types of employment and (ii) access to the good job segments is limited in the sense that not all who want to work in those segments are able to be employed. � Dual labor markets in Europe are often characterized by the coexistence of two types of contracts within a country: (i) rigid permanent contracts with an array of contributions and benefits and high employment protection legislation, and (ii) temporary short-term contracts, often with lower contributions and lack of benefits, with easy dismissal rules. 19 contracts are also used abundantly for low-skilled jobs in areas with relatively depressed labor markets, where the nationally set minimum wage is binding (see Uhl, 2014). As they are currently designed, labor contracts also seem to poorly fit the characteristics of new jobs and of production, thus providing a further incentive to use (and potentially abuse) civil contracts. The labor code is perceived as exceedingly complex, and lack of awareness of labor code norms seems to be a particular barrier for small firms. Moreover, while in international indicators, Poland’s restrictiveness of employment protection legislation appears to be average (see Figure 13), the lengthy judicial review for just cause dismissal creates substantial uncertainty for employers, so much so that in focus group settings, some employers equate hiring a worker with a permanent labor contract to “adopting a child: it is forever,� as one manager of a small manufacturing company in Lublin put it (Uhl, 2014). This is consistent with international experience suggesting that strict employment protection on permanent contracts—especially on dismissal—promotes the use of temporary employment contracts (see for example Boeri, 2010). Figure 13. OECD EPL Index for Regular and Temporary Workers OECD EPL index for regular workers 4 3 2 1 0 OECD EPL index for temporary workers 4 3 2 1 0 Source: OECD, 2013a. Interestingly, Poland fares in the middle range in terms of employment protection legislation (EPL) among OECD countries. However, its share of temporary contracts is substantially higher than what the level of employment protection legislation would predict among OECD countries (see Figure 14). In fact, Poland and Spain are the two countries where strictness of regulations on permanent workers seems to least explain the share of temporary employment. While some 20 of the unexplained variation in temporary employment could be attributed to different production structures, it is also possible that the EPL index might not capture the constraints that are actually binding in Poland, as appears to be the case for Spain.10 The procedural costs linked to dismissal could be one reason—one that is repeatedly cited by employers (but not captured by the index and difficult to verify)—for resorting to temporary employment. In this respect it is important to highlight that legal uncertainty of dismissal can be more costly to employers than redundancy pay, since the latter can be factored into wages ex ante. Figure 14. OECD EPL Index for Regular Workers and the Share of Temporary Employment 31 Poland 26 Share of temporary employment Spain 21 Portugal Netherlands Slovenia 16 Finland Sweden France Italy Germany 11 Ireland Greece Austria Denmark Czech Republic Belgium 6 United Kingdom Slovakia Estonia 1 1.5 1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1 OECD EPL index for permanent workers Source: World Bank staff calculations based on OECD (2013a) and Eurostat data. 5. Who Works with Temporary Contracts in Poland? Temporary employment is most widespread among youth (especially young women), older workers, and low-skilled workers.11 Among the age group 15 to 24, and to a lesser extent among 10 Spain ranks as the second strictest in the EPL for temporary workers, which reflects a very restrictive regulation that in practice has not been enforced. The same could be true of CCs in Poland. We are grateful to Prof. Marcel Jansen for pointing this out. 11 Existing data on CCs is scant and in most cases (including our main data source, the LFS) the data lump together CCs with FTLCs in the category of temporary employment. For the purpose of this paper, temporary employment is therefore used to mean CCs + FTLCs. The term “civil contract� is used whenever there is available and specific evidence, qualitative or quantitative, about umowa o dzieło or umowa zlecenie. 21 those aged 25 to 34, the share of temporary workers is substantially higher than the national average. Temporary working arrangements are less common among prime age workers (only 18 percent of wage employees in that group have a temporary contract in the main job), which likely reflects different factors: vintage effects as well as experience and different needs of workers in this age range. At any given age group, the incidence of temporary employment is higher among wage employees with lower education (see Figure 15). In 2012, 40 percent of the workers with primary education were in temporary labor contracts. Gender-wise, temporary employment appears to be balanced between women and men, though when looking at age subcategories, young women are overrepresented in temporary employment compared to men. Figure 15. Who Is in Temporary Employment? Share of temporary Share of temporary employment employment by age by education 70% 45% 60% 40% 35% 50% 30% 40% 25% 30% 20% 15% 20% 10% 10% 5% 0% 0% 15-24 25-34 35-54 55-64 prim. vocat. sec. ter. Source: LFS 2012 data, World Bank staff calculations. Over time, the incidence of temporary employment increased across the board, with the highest increase observed among young workers (Figure 16). For example, among people aged 15 to 24, the share of temporary employees increased by 34 percentage points between 2001 and 2012 (from 28 percent to 62 percent). For the 25- to 34-year-olds, the share of temporary employment is lower, but the increase was also drastic (20 percentage points) from 12 percent in 2001 to 32 percent in 2012. 22 Figure 16. Share of Workers in Temporary Contracts in 2001 and 2012 by Age (left) and Education (right) 60% 60% 40% 40% 20% 20% 0% 0% 15-24 25-34 35-54 55-64 prim. vocat. sec. ter. 2001 2012 2001 2012 Source: LFS 2012 data, World Bank staff calculations. In addition, while in 2001 temporary contracts were similarly distributed across gender, by 2012 the gender gap was significant between young women and young men (Figure 17). It remains an open question whether this is related to women’s choice of more flexible jobs when having small children or to the employers’ potential concerns with providing women labor contracts that would entail maternity leave. Figure 17. Difference Between the Incidence of Temporary Employment Among Women and Men by Age Categories in 2001 and 2012 55-64 35-54 25-34 15-24 -3 -2 -1 0 1 2 3 4 5 6 2001 2012 Source: LFS 2012 data, World Bank staff calculations. Services are the largest sector of the Polish economy and 60 percent of all temporary wage employees work in this sector. However, in relative terms, construction is the sectors that make most use of temporary contracts, as a share of dependent employment (Figure 18). As shown previously in Figure 11, the increase in employment in industry, construction, and services altogether occurred mostly through temporary work. 23 Figure 18. Structure of Temporary Employment in 2012 Agricult 35% ure 30% Share of workers with tempoarary contracts 3% 25% 20% Industr 15% y 27% 10% 5% Services 58% 0% Constru ction 12% 2001 2012 Source: Eurostat, World Bank staff calculations. One of the few pieces of evidence on the incidence of CCs (rather than on temporary employment in general) in different sectors are the estimates produced by the Polish National Statistical Office for 2012. This data indicate that employment regulated by civil contracts makes up the overwhelming majority of jobs in administrative and support services, comprising occupations such as security guards, administrative support, call centers workers, and cleaning services (Figure 19). 24 Figure 19. Share of Civil Law Contracts out of Total Employment (2012) Administrative and support services Arts, entertainment and recreation Accommodation and catering Information and communication Professional and technical activities Water supply; waste management Financial, real estate activities Manufacturing, construction Transportation and other services Non-market services Agriculture, mining, elctricity 0% 10% 20% 30% 40% 50% 60% 70% 80% Share of employment in CCs (out of total employment) Source: World Bank staff calculations based on data provided by the Polish National Statistical Office. 6. Is Temporary Employment in Poland (In)voluntary? Overall, about 15 percent of workers in temporary employment in Poland strictly prefer it to open-ended employment — a level that is about average across EU-28 countries (see Figure 20). Figure 20. Share of Voluntary Temporary Employment among Temporary Workers in EU-28 Countries, 2012 45 40 EU 28 35 30 25 20 15 10 5 0 Source: Eurostat, World Bank staff calculations. Certain categories of Polish workers, in particular older workers and especially older women, prefer to work with temporary contracts to retain flexibility and reduce taxation (Figure 21). 25 Most likely, the limited accumulation of pension rights is not a relevant consideration for older workers. As a result, temporary contracts appear to be a suitable instrument to increase participation to active economic life, especially in a rapidly aging country like Poland, where older workers’ low labor force participation continues to be a concern. Professionals, who are over-represented among the higher educated, also prefer temporary contracts, in a way that also reflects a relatively favorable legislation on taxation and social security contributions. Finally, CCs also reduce the labor costs of workers whose productivity level is below minimum wage, allowing them to work formally, although in such cases the appropriate policy response should be a correction or differentiation of the minimum wage. Figure 21. Who Most Often Prefers Temporary Contracts in Poland? Share of voluntary temporary Share of voluntary temporary employment by age employment by education 40% 20% 35% 30% 15% 25% 20% 10% 15% 10% 5% 5% 0% 0% 15-24 25-34 35-54 55-64 prim. vocat. sec. ter. Source: LFS 2012 data, World Bank staff calculations. A related question is whether workers in temporary contracts earn salaries comparable to permanent workers. The ample literature on compensating differentials (Rosen, 1986; Kaufman and Hotchkiss, 2005; Averett et al., 2003) would indicate that workers might be willing to be paid less (more) for a particularly pleasant (unpleasant) job; for example, workers with a stronger preference for flexible arrangements might be willing to accept a lower wage for a job with those characteristics. A wage differential could also be the indication that segmentation between different types of contracts exists, in that the labor market does not work efficiently in allocating labor where its returns is the highest. Looking at unconditional means of full-time workers’ monthly wages12, we find that in Poland, temporary workers earn on average around 30 percent less than permanent workers (Figure 22). As we saw earlier, workers with different contracts differ along many dimensions, including age, education, and sex. Controlling for relevant observable differences (age, education, sex, geographical location, and employment sector) through a nonparametric matching method reduces this differential substantially—to 15 percent. In other words, it appears that workers 12 Looking at hourly wages leads to very similar results. 26 who are in permanent contracts differ significantly in endowments from those in temporary contracts. Figure 22. Wage Gap between Permanent and Temporary Workers 35% 30% 25% 20% 15% 10% 5% 0% Raw Adjusted Source: LFS 2011 data, World Bank staff calculations. Taken at face value, a positive wage gap of 15 percent in favor of permanent workers would be consistent with temporary workers preferring to work with temporary arrangements (thus accepting lower pay for the same work). However, there is an important caveat to this methodology, which is that there might be characteristics unobservable to the econometrician but observable to the parties in question—ability, preferences—that once accounted for might reverse the compensation differential. When accounting also for workers’ preferences – which are observable in the LFS - involuntary temporary workers earn almost 20 percent less than permanent workers (see red bar, Figure 23, left). On the other hand, workers who are voluntarily employed under temporary contracts earn only about 3 percent less than workers with permanent contracts, controlling for all other observable characteristics (Figure 23, right). This points to a certain amount of segmentation in the labor market, which prevents workers who are involuntarily in temporary employment from moving to permanent jobs. Figure 23. Wage Gap between Permanent and Voluntary/Involuntary Temporary Workers 40% 30% 20% 10% 0% Involuntary temporary workers Voluntary temporary workers Raw Adjusted Source: LFS 2011 data, World Bank staff calculations. 27 Finally, an important feature of CCs is that they allow for wages below the minimum wage. A priori, minimum wage could be a significant constraint in low productivity regions, hence it is important to look at wage distribution to understand the extent to which wages fall below minimum wage. Overall, in Poland, around 10 percent of full-time workers report wages below minimum wage. More than half of those workers are employed under temporary contracts, and thus they constitute as much as 20 percent of all temporary workers; among permanent workers, however, it is 6 percent (see Figure 24). While the split between labor and civil law contracts is not yet available in the LFS, it would be safe to assume that workers earning below minimum wage are mostly under CCs. Figure 25 plots net wage kernels for full-time workers under permanent (red line) and temporary (green line) contracts for a high productivity region (śląskie region, left quadrant) and a low productivity one (warmińsko-mazurskie region, right quadrant).13 Although there may clearly be some measurement error,14 the share of full-time temporary workers reporting wages below the minimum wage is higher in the low productivity region. While additional analysis—including on dispersion of wages within regions and through administrative data—would be needed to draw policy conclusions on potential options to rethink the minimum wage, reform proposals that call for the abolition of CCs should account for the potential increase in informality and/or unemployment for workers who are paid below minimum wage, which are quite a significant share in low productivity regions. Figure 24. Share of Full-Time Workers That Report Wages Below the Minimum Wage Among Temporary and Permanent Workers 25 20 15 Permanent 10 Temporary 5 0 Total Urban Rural Source: LFS 2011 data, World Bank staff calculations. 13 Low productivity in this context is defined as a region with the lowest average wages and highest share of temporary employment. 14 Technically, it should not be possible to work full-time with labor contracts at salaries below minimum wage. 28 Figure 25. Kernel Densities in Better-Performing (left) and Worse-Performing (right) Regions Source: LFS 2011 data, World Bank staff calculations. 7. Temporary Employment: A Stepping Stone or a New Reality? An important question is whether the reliance on temporary employment—and in particular, CCs—can be seen as a stepping stone to better and more stable employment, especially for youth, who have more difficulty entering the labor market. Understanding transitions from temporary employment over time can help shed light on this question. The Polish Panel Survey has provided information on the socioeconomic situation of respondents every five years since 1988. In the last two waves the question on type of contract was included, and for around 1,050 people it is possible to analyze how their situation changed between 2008 and 2013. As shown in Figure 26, only one-third of workers with temporary contracts in 2008 moved to open-ended employment by 2013. Interestingly, these probabilities are not different for youth and for older workers, suggesting that temporary contracts are more of a new reality of labor markets for all age categories than an entry point to jobs specific to youth. 29 Figure 26. Labor Market Status in 2013 of Temporary Workers in 2008 Source: Kiersztyn, 2014. Consistent with this is the evidence that temporary workers’ tenure with the same employer has increased quite significantly since 2001—from 2 years to 3.5 years (see Figure 27). Figure 27. Tenure with Current Employer (in Years) Among Temporary Workers in 2001 (left) and 2011 (right) Source: LFS data, World Bank staff calculations. Finally, because CCs do not provide access to most social protection instruments, their prevalence among low-skilled workers, women of childbearing age, and other vulnerable groups is of potential concern. As it is concentrated among low-skilled workers, temporary employment is likely to expose these more vulnerable segments of the population to high turnaround and the risk of unemployment. Moreover, while younger women prefer the flexibility of temporary employment, the discontinuity of treatment of maternity leave across contract types (covered by labor contracts, but not envisaged in civil ones) distorts incentives for employers and for young women, who have the incentive to have children once they are working under a permanent labor contract. This bunching can in turn reinforce the view that it is risky for companies to offer permanent contracts to young women and skew contract choices for women and men. 30 8. Balancing Flexibility with Fairness in the Polish Labor Market The evidence suggests that temporary employment has played a positive and important role in employment growth in Poland. Temporary employment accounts for all net employment in the period 2001–2012, and in turn job creation has been a key element of shared growth. Yet, in light of the significant degree of distance between the characteristics of permanent and temporary employment—regarding cost and flexibility—this is of concern for the potential risk associated with future crises. Depending on the extent to which sectors that are intensive in temporary employment are exposed to the variability of the business cycle—less so those that rely on public procurement, more so those in tradable goods—the employment adjustment following a shock could be more or less deep. At the same time, temporary employment seems to be an effective way to increase activity among older workers, one that is aligned with individual preferences and responds to the country’s need to increase activity among a group—older workers—which registers particularly low labor force participation. The evidence indicates that duality in Poland is deepening: the vast majority of those who are employed under temporary contracts would prefer to work under permanent contracts. For those who are employed under temporary contracts involuntarily, there is a significant wage disadvantage, estimated at about 20 percent. Transitions from temporary employment to permanent employment are relatively slow. Finally, the Labor Inspectorate estimates that CCs are misused in about 18 percent of inspected firms. Only five years earlier this share was equal to 4 percent.15 While the potential drawbacks of temporary employment need to be understood in the context of counterfactuals—which range from open-ended employment to informality, or unemployment—there is scope to reduce the gap between permanent and temporary employment, especially for what concerns the gap between labor and civil law contracts. Increasing the likelihood that workers can reach a more stable and secure employment will require a remodulation of both the costs and incentives that firms take into account when determining the contractual form for new hires. This necessarily requires intervening on multiple fronts, especially to avoid large disemployment effects originating from an increase in costs or rigidity of civil contracts. Some of these interventions appear more feasible in the short run, while others are more ambitious because of their political economy implications. 15 From this percentage a broader estimate of abuse of CCs cannot be inferred, since the Inspectorate’s checks are conducted in firms with a higher a priori chance that abuses will be identified. 31 Box 2. Emerging lessons from reforming non-standard contracts in Slovakia Besides standard employment contracts, Slovakia has three types of non-standard contracts: (1) agreements to work up to 10 hours per week (40% of all non-standard contracts); (2) agreements to perform specific work up to 350 hours in one year (34%); and agreements to work up to 20 hours per week for students up to 26 years of age (26%). All of these contracts are fixed-term up to a maximum period of one year. In 2014, more than three quarters of people working on these non-standard contracts earned less than 150 euro monthly (which is less than half of the minimum wage of 352 euro). These contracts were used increasingly after the 2009 crisis, also in an effort to cut costs. Compared to standard employment contracts, these agreements offered more flexibility mainly due to relatively easier hiring and firing procedures as well as due to limited regulations protecting agreement workers: minimum wage regulation nor paid leave apply to these contracts. Moreover, with a security social contribution rate of 1.05% - compared to 48.6% - the related labor cost was significantly low if compared to the standard employment contracts. While the personal income tax rate of 19% was the same for agreements and the standard employment contracts, this rate was applied only from the income above 304 euro monthly in 2013 (raised to 317 euro in 2014). The massive use of agreements highlighted several problems such as the limited protection of workers who were not entitled to social benefits such as social security, unemployment benefits, paid sickness leave, etc. The government of Slovakia adopted in 2013 a reform that increased social and health contributions for non- standard contracts to the same level as for employment contracts, e.g. from 1.05% to 48.6% of the gross income paid by both workers and employers, though few exceptions remained. The government also applied minimum wage regulation to all employment contracts. The reform had immediate impacts. The number of people working on agreements decreased by almost 60% in January 2013 compared to December 2012. Later, the decline slowed down and the year-on-year decrease was almost 40% in December 2013 corresponding to a reduction from 552 thousand in December 2012 to 344 thousand in December 2013. Number of non-standard working agreements and people working with them 700,000 600,000 500,000 400,000 300,000 200,000 100,000 0 4-14 1-10 4-10 7-10 1-11 4-11 7-11 1-12 4-12 7-12 1-13 4-13 7-13 1-14 7-14 10-10 10-11 10-12 10-13 Number of people working on agreements Number of agreements Source: INEKO based on data from the Social Insurance Agency 32 Box 2, continued What happened to people who ceased to work on agreements? There was no significant change in the number of employees (1.96 million both in 4Q 2012 and 4Q 2013), registered unemployed (0.39 million both in 4Q 2012 and 4Q 2013) and self-employed (0.36 million both in December 2012 and December 2013). About half of those who stopped canceled their non-standard agreements were already working with standard employment contract or as self-employed and continued to work under those arrangements. Their net income decreased on average by more than 60 euro monthly (Source: INESS, 2013).There are no data on the remaining 97 thousand, workers. Some of them might be pensioners and others might have continued working informally. The reform effectively transferred the full payroll-tax burden on workers regardless of their income. It is also likely that those with low productivity have been hit hard and had to move to working informally. Source: Peter Goliaš (2014), INEKO. Short-term changes would include: 1) Confine the use of CCs to within the scope of their intended use in the law. This change will necessarily require an improvement in the effectiveness of labor inspections and an increase in sanctions for abusing CCs—in line with measures adopted by other countries. Two of the legal tests commonly used to detect abuses are whether the worker under civil contract is operating in subordination to the employer, or alternatively whether he is in direct economic dependence on the employer (Ichino, 1996). Besides the monetary sanctions, a dissuasive measure may be to allow courts to retroactively declare freelance contracts as open-ended labor contracts, in the event freelance contracts disguised an employer-employee relationship.16 2) For atypical employment relationships for which CCs are warranted, social security contribution requirements should be aligned with those in labor contracts. This could ensure adequate contribution for old age pensions across categories of workers. Additional measures that could be considered are restricting CC renewals. Medium-term changes would help reduce the gap between permanent and temporary contracts: 3) The costs of temporary labor contracts should be reduced, especially for low-paid workers. This can be achieved at no significant economic cost by simplifying the labor code, which may lower the administrative burden on firms. A recent enterprise survey shows that medium-sized firms especially complain of the complexity of labor regulations. Interesting in this respect is the Italian initiative for a Simplified Labor Code, thanks to which a draft of a simplified labor code (in 16 This norm was introduced in Italy’s most recent labor market reform in 2012, together with an increase in contributions to progressively match those of term contracts. The reform resulted immediately in a net decline in the hiring of freelance workers (-25%), though less is known about the forms of contracts that substituted these jobs, due to the lack of longitudinal data. Another interesting fact is the modest increase in the use of term contracts (+6%) after the reform, though a new reality was the increase of “ultra-short� term contracts (1–-3 days). See: Ministero del Lavoro e Delle Politiche Sociali (2014) 33 70 simple articles) has just been completed. Such a measure would, however, require a significant amount of time, given the technical and potentially political complexity of changing the labor code. A quicker but more resource-intensive reform would be to raise the nontaxable allowance of labor income in order to make labor contracts more accessible to firms for current workers on civil contracts. This is particularly important because a share of CCs appears to be used to hire workers at wages below the minimum wage, and a tax exemption would maintain stable the net minimum wage without excessively increasing the gross labor costs for firms. In addition, costs could be reduced by limiting the abuses of contractual rights—such as sick leave—for prolonged periods, and without strong documentary evidence. For instance, Poland has one of the highest average periods of pre-birth sick leave among pregnant women. Part of this effort would include also (i) more effective access to information regarding the administration of labor contracts and sanctions for abuse of civil contracts; and (ii) making the regulation of hours in some professions more flexible—in some professions hours are still very rigid, which leads to hiring with civil contracts, under which workers might have overall lower protection. 4) Decrease the implicit costs of permanent contracts, so as to make them the natural progression of the employment relationship after a period of temporary work. This could be obtained by reducing the implicit costs of hiring permanent workers, including reducing procedural costs of dismissal and uncertainty of labor adjudication, which are perceived as an important barrier by employers. One way to do this is to put in place a severance pay that increases with job tenure in lieu of the “just cause� requirement and the judicial review process. There is also room to rethink other aspects of labor code contract design—one of these could be the structure of probation periods, which are seen as excessively constraining in their ability to lead to an open-ended contract after three months. Changes in this direction are likely to result in:  A smaller gap between CCs and FTLCs, with a likely improvement in benefits for workers. Because of the increased labor costs due to social security contributions, there might be a downward adjustment on wages. In the presence of reduced taxation this would not necessarily translate into a proportional decrease in take-home pay. Alternatively, it could result in an adjustment in working hours.  Relatively more flexible permanent contracts, akin to the changes that moving to a prevalent contract with a basic set of rights and increasing levels of protection (also known in the literature as a “single contract�) that is discussed in proposals for Italy or Spain. While there is currently much debate over the benefits of single labor contracts, their introduction in Poland would require an overhaul of the labor code into a system with significantly greater ease of dismissal, in order to allow employers to hire workers who are needed only for a limited period of time. This seems unlikely at present.  Many factors will determine the potential impact on unemployment and/or informality. Interventions on the labor demand side might be necessary to avoid, in some sectors, workers with low productivity for whom labor contracts are too costly being left 34 unemployed—for example, moving away from cost being the only criteria for public procurement selection. It is also important that enforcement provide suitable disincentives for firms to hire informally and create a level playing field across firms. The international evidence suggests that these types of adjustments might be associated with some increase in unemployment in the short term, but might not impact unemployment significantly in the long term. The government has recently proposed the increase of social security contribution for civil contracts at the minimum wage level. Recent literature (Onaran, 2008) does not provide evidence of the negative employment elasticities with respect to wages in Poland. However, in focus groups settings, managers of low-value added companies indicated that in response to these changes, firms might hire workers informally or hire younger workers. Firms that rely on a highly skilled work force reported that they would reduce employment to some extent. 35 REFERENCES Aleksynska, Mariya, and Martin Schindler. 2011. "Labor Market Regulations in Low-, Middle- and High-Income Countries: A New Panel Database." IMF Working Papers 11/154, International Monetary Fund. Arak, Piotr, Piotr Lewandowski, and Piotr Zakowiecki. 2014. “Dual Labor Market in Poland— Proposals for Overcoming the Deadlock.� IBS Policy Paper no. 01/2014, Instytut Badań Strukturalnych, Warsaw. http://ibs.org.pl/files/publikacje/pp/pp_2014/IBS_Policy_Paper_01-2014_eng.pdf. Averett, Susan, Howard Bodenhorn, and Justas Staisiunas. 2003. "Unemployment Risk and Compensating Differential in Late-Nineteenth Century New Jersey Manufacturing." National Bureau of Economic Research Working Paper 9977, Cambridge. Baranowska, A., and M. Gebel. 2008. “Temporary Employment in Central and Eastern Europe: Individual Risk Patterns and Institutional Context.� MZES Working Papers, 2008 (106), Universität Mannheim Mannheimer Zentrum
für Europäische Sozialforschung, Germany. Baranowska, A., M. Gebel, and I. E. Kotowska. 2001. “The Role of Fixed-Term Contracts at Labor Market Entry in Poland: Stepping Stones, Screening Devices, Taps or Search Subsidies?� Work, Employment and Society 25 (4): 777–93. Bednarski M., and K. Frieske. 2012. „Zatrudnienie na czas określony w polskiej gospodarce.� Społeczne i ekonomiczne konsekwencje zjawiska, Instytut Pracy i Spraw Socjalnych, Warszawa. Boeri, T. 2010. “Institutional Reforms in European Labour Markets.� Handbook of Labour Economics. Bruno, Giovanni S. F., Floro Ernesto Caroleo, and Orietta Dessy. 2012. “Stepping Stones Versus Dead End Jobs: Exits from Temporary Contracts in Italy After the 2003 Reform.� IZA Discussion paper 6746, Institute for the Study of Labor, Bonn, Germany. Cahuc, P., O. Charlot, and F. Malherbet. 2013. "Explaining the Spread of Temporary Jobs and its Impact on Labor Turnover." IZA Discussion paper 6365, Institute for the Study of Labor, Bonn, Germany. Eichhorst, W. 2012. “The Unexpected Appearance of a New German Model.� IZA Discussion paper 6625, Institute for the Study of Labor, Bonn, Germany. Esping-Andersen, G. 1990. The Three Worlds of Welfare Capitalism. Cambridge, UK: Policy Press. 36 Eurofound. 2013. "Young People and Temporary Employment in Europe." Eurofound, Dublin, Ireland. European Commission. 2014. “Recommendation for a COUNCIL RECOMMENDATION on Poland’s 2014 national reform programme and delivering a Council opinion on Poland’s 2014 convergence programme.� European Commission, Brussels. http://ec.europa.eu/europe2020/pdf/csr2014/csr2014_poland_en.pdf Eurostat. n.d. “LFS series—Detailed annual survey results (lfsa).� http://epp.eurostat.ec.europa.eu/cache/ITY_SDDS/en/lfsa_esms.htm. Fields, G. 2004. “Dualism in the Labor Market: A Perspective on the Lewis Model after Half a Century.� Manchester School 72 (6): 724–35. Fields, G. 2007. “Labor Market Policy in Developing Countries: A Selective Review of the Literature and Needs for the Future.� Policy Research Working Paper 4362, World Bank, Washington, DC. Fields, G. S. 1975. “Rural Urban Migration, Urban Unemployment and Job Search Activity in LDCs.� Journal of Development. Goliaš, P. 2014. “The reform of non-standard contracts in Slovakia,� mimeo, INEKO. Harris, John R., and Michael P. Todaro. 1970. “Migration, Unemployment and Development: A Two-Sector Analysis.� American Economic Review 60 (1): 126–42. Ichino, Pietro. 2014, May. Presentation delivered at the Conference on Labor Market Duality organized by the World Bank and the Polish Ministry of Labor, Warsaw. Ichino, Pietro. 1996. Il lavoro e il mercato. Milan: Mondadori. International Labour Office. 2012. Global Employment Trends for Youth: 2012. ILO, Geneva. International Monetary Fund. 2014. “Jobs and Growth: Supporting the European Recovery.� IMF, Washington DC. Kaufman, Bruce E., and Julie L. Hotchkiss. 2005. "Education, Training, and Earnings Differentials: The Theory of Human Capital." The Economics of Labor Markets (5th ed.). Harcourt College Publishers. Kiersztyn, A. 2014. „Niepewność zatrudnienia. Kto jej doświadcza? Czy to stan chwilowy?� Zespół Porównawczych Analiz Nierówności Społecznych, Instytut Filozofii i Socjologii Polskiej Akademii Nauk, Warszawa. Lewis, W. Arthur. 1954. “Economic Development with Unlimited Supplies of Labor.� Manchester School of Economic and Social Studies (22): 139–91. 37 Ministero del Lavoro e Delle Politiche Sociali. 2014. “Il primo anno di applicazione della legge 92/2012�. Quaderno 1. January 2014. http://www.camera.it/temiap/temi17/QUADERNO1_I%20anno%20applicazione%20L%2 092_2012.pdf Ministry of Labour and Social Policy, Department of Economic Analyses and Forecasts. 2008. Employment in Poland 2007: Security on a flexible labour market. Instytut Badań Strukturalnych, Warsaw. http://ibs.org.pl/files/Employment_in_Poland_2007_IBS.pdf OECD. 2013a. Indicators of Employment Protection Database. http://www.oecd.org/employment/emp/oecdindicatorsofemploymentprotection.htm OECD. 2013b. OECD Employment Outlook. OECD, Paris. OECD, 2014. OECD Economic Surveys: Poland. OECD, Paris. Onaran, Ozlem. 2008. "Jobless Growth in the Central and East European Countries: A Country- Specific Panel Data Analysis of the Manufacturing Industry." Eastern European Economics 46 (4): 90–115. POLPAN. Polish Panel Survey. 1988–2013. http://polpan.org/english/project-description Rosen, Sherwin. 1986. "The Theory of Equalizing Differences." In Ashenfelter, Orley; Layard, Richard. The Handbook of Labor Economics. New York: Elsevier. pp. 641–92. Uhl, Hanna. 2014. “Qualitative Research on Labor Market Duality in Poland: the perspective of employers and young worker.� Millward Brown, Warsaw. 38 ANNEX Table A1. Components of Social Security Premiums Paid by Employer and Employee Type of insurance % premium Breakdown of premium Employer Employee Retirement 19.52% of remuneration 9.76% 9.76% Disability 8% of remuneration 6.5% 1.5% Accident 0.67% to 3.33% of 0.67% to 3.33% - remuneration depending on occupational risk in a given industry Sickness 2.45% of remuneration - 2.45% Additional contributions: Labor Fund 2.45% of remuneration 2.45% - Guaranteed 0.10% of remuneration 0.10% - Employee Benefit Fund Source: World Bank staff. 39 Table A2. Social Insurance Obligations Under Different Employment Contracts Labor code Civil Labor costs contracts law contracts Self-employment Contract All types Commission contract of result Before changes After changes Yes, at least Yes, at least based on Yes, first contract based on Old age 60% of average wage Yes mandatory, other minimum No contribution in enterprise sector, voluntary wage, later later voluntary voluntary Yes, at least Yes, at least based on Yes, first contract based on Disability 60% of average wage Yes mandatory, other minimum No contribution in enterprise sector, voluntary wage, later later voluntary voluntary Voluntary for the Sickness Yes first contract, then Voluntary No No, only voluntary contribution* impossible Yes, at least Yes, at least based on Yes, first contract based on Accident 60% of average wage Yes mandatory, other minimum No contribution in enterprise sector, voluntary wage, later later voluntary voluntary Yes, at least based on Yes, obligatory Health Yes, obligatory for 75% of average wage Yes for every No contribution every contract in enterprise sector, contract later voluntary Yes, at least Yes, at least based on Yes, first contract based on 60% of average wage Labor Fund Yes mandatory, other minimum No in enterprise sector, voluntary wage, later later voluntary voluntary Yes, at least Guaranteed Yes, first contract based on Employee Yes mandatory, other minimum No No Benefits Fund voluntary wage, later voluntary Source: World Bank staff. 40 41