Improving the Labor Market: Challenges and Options1
Labor market outcomes in Serbia are worrisome: the unemployment rate is one of the highest in Europe and the labor force participation rate is one of the lowest. These outcomes can be attributed to the unfinished transition to a market-based economy: the late start of reforms and their uneven implementation constrained FDI inflows, slowed down structural transformation and therefore hampered employment creation. In addition, a sharp increase in unemployment without adjustment in the labor cost is indicative of lack of flexibility in Serbia’s labor market. Cross-country comparison suggests that the average total labor cost in Serbia has risen faster than productivity: this could be partly attributed to (i) the strong role of the public sector in setting wages, (ii) excessive protection of insiders, and (iii) a centralized wage bargaining process. A comprehensive structural reform is needed to strengthen private sector and boost the demand for labor in Serbia. Reforms should aim to reduce the total labor cost and to increase labor market flexibility.
1. Serbia’s labor market is characterized by low labor force participation and high unemployment. The participation rate—defined as a percentage of working age population, either employed or actively seeking job—declined from 66 percent in 2004 to about 60 percent in 2012, which is lower than in most of Serbia’s peers and other European economies (Figure 1). The rate of unemployment, as measured by the Labor Force Survey (LFS), has increased to more than 23 percent and is now one of the highest in Europe. Low labor force participation and high unemployment result in exceptionally low rate of employment, which fell to 46 percent in 2012—much below most European economies and New Member States of the EU (NMS) in particular. High unemployment reduces labor force participation as many workers who lost their jobs became “discouraged” and leave the labor force.
Figure 1.Serbia: Labor Market Indicators, 2004-2012
Sources: SORS, FREN, and Eurostat.
2. High youth unemployment and a significant rate of long-term unemployment are particularly worrisome. The rate of unemployment for the group aged 15–24 exceeded 51 percent in 2012 (Figure 1). This entails negative implications for potential GDP growth given poor chances for young people entering the labor force to acquire needed on-the-job training and human capital. High rate of long-term unemployment—reaching 78 percent in 2012—could also be a detriment to accumulation of human capital.
3. Other notable features of the Serbia’s labor market are significant shares of informal and public sector employment. Informal employment—as estimated by a difference between overall employment and registered employment—amounts to almost 500 thousand persons, or 19 percent of overall employment.2 The public sector employs 43 percent of registered employment in legal entities, or 26 percent of overall employment.3
4. A significant share of unemployment in Serbia—as well as in other Balkan economies—is structural. Labor markets in the Balkan economies are characterized by a low value of Okun’s coefficient, which shows the responsiveness of unemployment to the business cycle. This suggests that unemployment in Serbia cannot be explained by the downturns over the business cycle and the roots of the problem are structural.
GDP Growth and Changes in Unemployment, 1993-2011
Sources: WEO; and IMF staff calculations.
1/ Albania, Bosnia and Herzegovina, Kosovo, Macedonia, Montenegro, and Serbia.
2/ Greece, Italy, Portugal and Spain.
5. The absence of adjustment in the real wages despite increasing unemployment since 2008 is puzzling. An excess supply of labor should lead to a downward pressure on wages, but this was not the pattern observed in the data (Figure 2). There could be a statistical explanation: wage pressures should be judged by looking at the wage for a given skill level. However, the wage data aggregates different skills level into an average, leading to a possible upward bias of wages as the composition of labor is rebalanced—it is likely that average wage was affected by attrition of low-skilled workers.4
Figure 2.Serbia: Wage Indicators, 2009-13
Sources: Eurostat, SORS and IMF staff calculations.
1/ 2011 data used in place of unavailable 2012 data.
6. There could also be significant labor market rigidities explaining the lack of wage adjustment. Possible rigidities are (i) the significant role of the public sector in setting the wage level, (ii) excessive insider protection imposed by the Labor Law and other institutional arrangements, (iii) restrictive wage bargaining system, and (iv) binding minimum wage. The remainder of this paper examines structural causes of Serbia’s high unemployment, assesses wage competitiveness using several indicators in the cross-country setting and sheds light on possible rigidities.
B. What Are the Possible Causes of Meager Labor Market Outcomes?
The Role of the Unfinished Transition to a Market Economy
7. Successful transitions of the 1990s in emerging Europe and particularly NMS were generally associated with comprehensive structural reforms. These reforms attracted significant inflows of foreign direct investment (FDI) and promoted job creation. Despite the fact that unemployment may have increased in the short run, the reforms allowed development of the private sector that eventually provided conditions for reducing unemployment rates. The infusion of capital from abroad—especially via greenfield FDI—played a key role as it helped to develop new businesses or even entire new sectors in the economy and provided a chance for workers dismissed from the declining sectors to be reabsorbed by new economic activity. While other types of capital movements, such as equity flows, have also shown positive relationships with growth under specific circumstances, only FDI proved to be a robust and significant driver of output growth (Aizenman et al., 2011). The reasoning behind that particularly beneficial relationship is that FDI offers a more stable and long-term foundation for technology, know-how, managerial skills, and international marketing networks to be transferred than other more volatile forms of capital flows. FDI inflows, however, critically depended on favorable business environment, macroeconomic stability and strong track record in structural reforms.
8. Unlike NMS, Serbia as well as other Balkan economies were delayed in their structural transformations. Significant reforms largely began nearly a full decade behind NMS. Serbia made significant progress in transition in the first half of 2000s, but many needed reforms still remained incomplete when the global financial crisis erupted. As a result, the unfinished structural reform agenda remained substantial, with more progress needed in such essential areas as privatization, enterprise restructuring, and competition policy (Figure 3).
Figure 3.EBRD Transition Indicators, 2012 1/
Source: 2012 EBRD Transition Report and IMF staff calculations.
1/ The measurement scale for the indicators ranges from 1 to 4+, where 1 represents little or no change from a rigid centrally planned economy and 4+ represents the standards of an industrialized market economy.
2/ Average of all six EBRD transition indicators.
3/ Average of large scale and small scale privatization indicators.
9. The delay in reforms led to a relatively weak business environment, reduced attractiveness to foreign investors and therefore constrained job creation. Serbia—as well as several other Balkan economies—are lagging behind in terms of ease of doing business, and the FDI stock per capita is significantly lower than in peer countries. These factors put Serbia at a disadvantage in terms of diversifying away from traditional sectors and therefore created a drag on employment creation. A scatter plot of per capita stock of FDI and unemployment rates in the Balkan countries and NMS suggests that countries that attracted more FDI tend to have lower unemployment rates.
Ease of Doing Business, 2012
Unemployment rates vs. FDI, 2011
Source: LFS unemployment rates (except ALB which publishes only registered rates); IIP database; Doing Buisness (2012); and IMF staff calculations.
10. Structural employment losses since 2008 were partly attributed to layoffs related to the pre-crisis privatization and restructuring. Former socially-owned enterprises had contractual obligations to refrain from layoffs for three years following privatization. As privatization peaked in 2005-06, the layoffs coincided with the onset of the 2008–09 financial crisis. Layoffs were significant as post-privatization restructuring appears to have been focused on employment cuts rather than wage cuts: wage data suggest that companies shed labor rather than reduced wages (Figure 2). Another important factor was that Serbian tradable sector faced growing competition—the first wave of trade liberalization took place in early 2000s and the second wave began in 2009 with the launch of the Stabilization and Association Agreement with the EU. The employment losses were undoubtedly compounded by the onset of the global financial crisis in 2008.
Wage Competitiveness: Are Labor Costs Too High?
Distinguishing between net wages, gross wages and overall labor costs
11. Analysis of labor market competitiveness calls for examining net wages, gross wages and total labor costs (TLC) separately. Net wage—defined as gross wage accrued to an employee less taxes and social security contributions paid by the employee—should be the most important determinant of labor supply. The demand for labor, however, should depend not on the net wage but rather on the TLC, which is defined as gross wage plus social security contributions and any labor-related costs (e.g. severance payments) paid by the employer. The wedges could be illustrated using data for the manufacturing sector in Serbia:5
Total labor cost and wages in manufacturing, 2006-12
Source: Structural Business Statistics, SORS and IMF staff estimations.
• Gross wage vs. net wage: This wedge consists of personal income tax (12 percent on a taxable base) plus employee’s contribution to the social security system (17.9 percent of the gross wage). In 2012, net average wage in manufacturing was about 27 percent lower than the average gross wage.
• TLC vs. gross wage: In 2011, TLC in manufacturing (expressed in Euros per worker per month), as reported in SORS’ Structural Business Indicators, was about 23 percent higher than the gross wage. The difference is explained mainly by the employer’s contribution to the social securing system (17.9 percent). The remaining difference arises from severance payments, allowance for business trips abroad and allowance for transport to work, jubilee awards, directors’ fees and scholarships but also wages for occasional employment other than fixed-term employments (external salesmen, security workers and others). Cross-country comparison suggests that this wedge is in line with the EU average, but it does not necessarily mean that this is optimal for Serbia.
Wedge between total labor cost and gross wage in manufacturing, 2010
Source: Eurostat, SORS and IMF staff calculations
1/ Total labor cost is calculated as a ratio of total personnel cost in manufacturing to the number of persons employed as reported by the Structural Business Statistics.
• The overall wedge. In 2011, the overall tax wedge in manufacturing, defined as a difference between TLC and the net wage, amounted to 41 percent of TLC or 70 percent of the net wage. For comparison, Fiscal Council estimated the fiscal burden to equal 64 percent of the net wage.6 Their estimate is lower because the fiscal burden does not include severance payments and other labor-related expenditures captured in structural business statistics.
Analyzing trends in wages and productivity
12. Average gross real wage in the economy grew faster than productivity in 2001-08, but this trend was reversed during the period from 2009 to 2011. Average real wage increased by 174 percent between 2001 and 2008 whereas productivity increased by only 47 percent, suggesting erosion of competitiveness (in manufacturing sector, this pattern appeared to be somewhat less pronounced). It should also be noted, however, that wage growth in the earlier part of the decade could have been overestimated as the low-wage earners were gradually dropping out of the sample, creating an upward bias in the average wage (World Bank Labor Market Review, 2006). During 2009-2011, the growth of real wages slowed down but the gap between wages and productivity persisted.
Real wages and productivity, 2001-12
Real wages and productivity in manufacturing, 2001-12
Sources: SORS and IMF staff calculations.
1/ Productivity index is based on the formal employment.
Analyzing trends in the unit labor cost
13. Unit labor cost (ULC) in manufacturing based on the average gross wage increased faster than in peers until 2008 but declined during 2009–10.7 ULC growth before the crisis led to concerns about competitiveness: IMF staff noted that while Serbia’s euro wage levels relative to labor productivity were in line with most transition peers, they significantly exceeded those of key regional competitors, including Bulgaria and Slovakia.8 The erosion of export competitiveness was reversed in 2009 and 2010 when ULC declined by 7 percent on account of significant dinar depreciation. However, ULC indexes do not provide information regarding the relative “level” of competitiveness (i.e. how “competitive” a country is relative to its peers).
Unit Labor Cost (Euro) in Manufacturing, 2005-2012
Sources: National authorities and IMF staff calculations.
Cross-country comparison of wages and productivity
• Cross-country comparisons reveal strikingly low productivity in Serbia. In 2012, Serbia’s productivity was four times lower than in the Euro Area, and was lower than in most peer countries: it was only about a half of productivity in Croatia and Romania (Figure 4). Productivity in manufacturing was also low.
Figure 4.Serbia: Wage and Productivity Indicators, 2012
Sources: Haver; EMED; WEO; and IMF staff calculations.
1/ Labor productivity = [(* nominal GDP in euros / * employment) / (EA Nominal GDP / employment)]
(*) = Each country
• Gross and net wages appeared to be broadly in line with productivity. In 2012, wages appeared to be broadly in line with these in peer countries taking into account low level of productivity and adjustment that took place since the onset of the crisis.
• On the contrary, TLC was somewhat higher than warranted by Serbia’s relative productivity. A regression line estimated on a sample of EU-27 countries in 2010 (excluding Ireland that was an outlier) suggested a “fitted” level of labor cost in manufacturing per employee for a given level of productivity. Serbia’s monthly total labor cost of 534 euros per employee was 11 percent higher than the TLC suggested by the regression.
|Relative productivity (percent of the EU productivity)||20|
|Actual TLC, Euros per worker per month||534|
|“Fitted” TLC, Euros per worker per month||475|
|Deviation from the “fitted”TLC, percent||11.4|
Is Employment Protection too Strong?
14. Employment protection legislations (EPLs) refer to laws, regulations, or customary procedures that govern employers’ ability to dismiss workers. These come in various forms such as restrictions on layoffs, required notice periods, penalties for unfair dismissals, mandatory severance payments, etc. (OECD, 1994). EPLs protect workers’ rights and are often guided by the International Labor Organization’s (ILO’s) Termination of Employment Convention. At the same time, excessive EPLs constrain job creation as employers adopt greater caution in hiring. This means that EPLs tend to protect those that are already employed but reduce opportunities for the unemployed. Over time, this creates a bottleneck for young workers joining the labor force and prolongs the jobless spells for the unemployed. Very rigid labor legislations, therefore, impede labor re-allocation and may protect the employed at the expense of the unemployed, thereby exacerbating unemployment. Indeed there is empirical evidence that protective labor regulations tend to increase unemployment, particularly for the young (Nickell et al., 2005, Feldmann, 2009, Botero et al., 2004), and decrease labor force participation (Feldmann, 2009).
15. The 2005 Serbia’s Labor Law Amendment increased the level of regulation, aligning it with more advanced European countries, primarily those in the Southern Europe. It made it more restrictive relative to Serbia’s regional peers. Enacted at the peak of privatization and restructuring, the amendment sought to strengthen job security by introducing disincentives for firing and shortening length of temporary employment.
16. Rules governing severance payments constitute a unique feature of the Serbia’s EPLs embodied in the Labor Law. The severance payment is calculated on the basis of lifetime employment rather than on the period with the last employer. It is calculated as the sum of ⅓ of most recent average monthly salary for each year of employment for the first 10 years and ¼ for each subsequent year of employment. In the case of an employee with 30 years of employment this adds up to 8⅓ monthly salaries. The rule creates disincentives for laying off older employees. Cross-country comparisons indicate that Serbia’s severance payment rule is unique: other European countries have implemented either fixed severance payments or linked it to the length of service with the last employer.
17. Aside from severance payments, dismissing permanent employees is subject to administrative and legal hurdles. The Labor Law distinguishes between collective or individual dismissals, and the latter comprises dismissals for (i) economic reasons, and (ii) poor performance/inadequate skills.
• Collective dismissals have to be preceded by attempts of alternative employment and preparation of a program intended to mitigate the consequences by providing training, shorter work hours and similar along with providing causes for dismissal and dismissal criteria. The program must be discussed with trade unions and the National Employment Service, both of which are required to provide non-binding suggestions.
• Individual dismissals due to inadequate skills, low productivity or other performance-related reasons generally require “proofs” of inadequate performance or skills. Following trade union’s opinion, the employee can be dismissed after a notice period of 1 to 3 months (without notice period in the case of misbehavior and criminal acts) depending on the length of career (no severance payment is required in this case). A dismissed employee can appeal to the court and be reinstated if the employer fails to justify the dismissals (in which case the employee has to be reimbursed for the loss of wage income).
• Individual dismissals for economic reasons require a letter from the employer justifying the reasons for dismissal and a severance payment. Also, the employer must rehire the same person if the same position opens up again within the following six months.
18. Resorting to fixed-term contracts—which is a common solution in other countries for bypassing excessive insider protection—is limited by the Labor Law. The Labor Law limits fixed-term contracts to one year and allows such contracts to be concluded only in the case of seasonal jobs, projects, temporarily increased workload, replacement of absent employees and other similar tasks. The likely motivation was to eliminate potential loopholes that would help avoid the above severance requirements. Thus, fixed-term contracts are effectively prohibited for hiring employees to perform permanent tasks. Additionally, employment lasting 5 days more than the prescribed maximum of 12 months, is treated as a contract of indefinite duration.
Do Unemployment Benefits Create a Disincentive for Work?
19. Unemployment benefits in Serbia do not appear to be overly generous and therefore are unlikely to create disincentives for work. They are on average below the minimum wage while their duration is limited to 12 months in the case of dismissed employees with more than 20 years of service. As the majority of unemployed persons have been unemployed for over 12 months, they are not eligible to receive it.
Unemployment Benefits, 2010
Maximum Duration of Unemployment Benefits, 2010
Sources: OECD; country authorities; and IMF staff calculations.
Role of the Public Sector
20. The average wage in the public sector is higher than in the private sector. The difference of euro-denominated wages in public and private sectors was modest in the beginning of the last decade but grew progressively larger, peaking before the 2008 global crisis. This might have created an impediment for wage adjustment: higher public sector wages created an incentive to seek employment in the public sector, effectively setting a “reservation wage” for a given level of skill. The differential was reduced during the 2009-10 period as public sector wages were frozen, but it still remains sizeable.
21. In addition, trade unions are dominated by member unions from the public sector. This tilts the collective bargaining and trade unions’ pressure for wage increases to the benefit of public sector employees.
Does the Collective Bargaining Process Create Rigidities in Serbia’s Labor Market?
22. Collective agreements play an important role in Serbia’s labor market. Collective agreements are concluded at the national, regional and company level, although the bargaining process is being moved to the industry level after the general collective agreement (GCA) expired in May 2011.The following features of the bargaining process pose challenges for the employers:
• Collective agreements specify so-called “base wages” for different levels of skills via a set of special multipliers. Base wages effectively set wage floors for different skill levels. Anecdotal evidence suggests that actual wages are generally higher than “base wages”; however, this mechanism is subject to risk of posing undue wage rigidities if the minimum wage pushes base wages above the market-clearing levels, or if there is a need of downward wage adjustment. Minimum wage negotiations take place under the Social Economic Council and when agreement is reached, the minimum wage becomes mandatory for all employers including those not represented by the participating employers’ association.
Average monthly gross wage, 2003-12
1/ Public administration (all levels), education, culture and social services.
2/ Excluding entrepreneurs and enterprizes with social and mixed ownership.
• The Ministry of Labor, Employment and Social Policy extends application of recently concluded industry-level agreements to companies not represented in the bargaining process (Foreign Investor Council’s White Book, 2011) and these companies were forced to abide by various obligations specified in the agreements. This poses a challenge for companies that do not have sufficient resources to comply with requirements specified in the agreements.
Labor Market Efficiency: Evidence from Global Competitiveness Report
23. Survey-based evidence confirms that Serbia is lagging behind in terms of overall flexibility of the labor market. Serbia ranked 122 out of 142 countries on the overall labor market efficiency in 2011 Global Competitiveness Report.9 The ranking was done using 9 sub-indicators: cooperation in labor market relations (136 place among 142 ranked countries); flexibility of wage determination (64); rigidity of employment (90); hiring and firing practices (89); redundancy costs (50); pay and productivity (103); reliance on professional management (133); brain drain (139), and female participation rate (69). The rankings suggest significant room for improvement along nearly all dimensions
C. Policy Implications
24. Poor labor market outcomes are symptoms of Serbia’s unfinished transition and therefore require a comprehensive strategy that goes well beyond the labor market itself. In order to alleviate unemployment, Serbia needs a dynamic private sector capable of generating demand for labor. This implies that there is no “silver bullet” solution for generating employment—Serbia faces the unfinished reform agenda that needs to be addressed with a comprehensive package of structural reforms (Selected Issues Paper “In Search of an Effective Growth Model”).
25. The structural reforms should include a set of measures to reduce labor costs and removing disincentives for hiring. The wedge between total labor cost and wage is too high and should be lowered, if fiscal space allows, by reducing employers’ contribution to social security as well as other labor-related expenses. The following measures could be considered:
• Reducing the social security contribution rates paid by employers to reduce the TLC and increase demand for labor when there is fiscal space. One way to achieve this is to strengthen capacity of the Tax Administration and Labor Inspection in order to increase collection rates—broadening the tax base could create scope for reducing the tax wedge.
• Revising the formula for calculating the severance payment could provide more opportunities to job-seekers, especially older workers. The severance pay should be calculated on the basis of the length of employment with the last employer (consistently with the practice in other European countries).
• Simplifying dismissal procedures. The Labor Law should clarify conditions for fair dismissals in order to reduce room for misinterpretation.
• Making collective agreements less restrictive and bringing bargaining process to the company level. As the first best, giving priority to firm-level agreements over higher level collective bargaining and allowing wages setting at the company level (firms that do not have resources to administer their own agreements could have a choice to “opt-in” for a higher level agreement). As a second best, discontinue the practice of extending these agreements to all companies in a given industry.
26. More active use of active labor market policies (ALMP) would alleviate the skills mismatch and help unemployed adjust to the changing environment. Serbia’s spending on ALMP is significantly lower than that of Serbia’s peers (World Bank Country Economic Memorandum) and boosting these programs could alleviate skills mismatch.
27. An adequate social safety net would be needed for those who would be affected by increasing labor market flexibility. A reform could lead to reallocation of employment between economic sectors and also to layoffs. In order to avoid social tension, there should be an adequate social safety net protecting the poor.
Figure 5.Serbia: Labor Market Indicators from The Global Competitiveness Report 2011
Source: World Economic Forum: Global Competitiveness Report 2011.
- Search Google Scholar
- Export Citation
AizenmanJoshuaYothinJinjarak and DonghyunPark2011“Capital Flows and Economic Growth in the Era of Financial Integration and Crisis, 1999–2010,”NBER Working Paper 17502.
ArandarenkoM.2011“Tržište rada u Srbiji – Trendovi, institucije, politike,” Beograd, Srbija. Centar za izdavačku delatnost Ekonomskog fakulteta u Beograd.
Commission of the European Communities2006Report by the Commission services on the implementation of COUNCIL Directive 1999/70/EC OF 28 JUNE 1999 concerning the Framework Agreement on Fixed-term Work concluded by ETUC UNICE and CEEP (EU-15) Commission Staff Working Document.
Commission of the European Communities2008“National legislation transposing directive 1999/70/EC on fixed term work in the EU 10,”Commission Staff Working Document.
Council of the European Union1999“Council Directive 1999/70/EC of June 1999 concerning the framework on fixed-term work concluded by ETUC, UNICE and CEEP.”Official Journal L 175 10/07/1999 P. 0043–48.
International Monetary Fund2011a“Spain—Staff Report for the 2011 Article IV Consultation,”Washington, D.C.
International Monetary Fund2011b“The Spanish Labor Market in a Cross-Country Perspective.”Working Paper 11/11. Washington DC.
International Labor Organization1982“Termination of Employment Convention.”
International Labor Organization2009“Rules of the Game: a brief introduction to International Labor Standards (Revised edition 2009).”
Fiscal Council2012“Suggested Fiscal Consolidation Measures for 2012–2016.”
FREN - Fond za razvoj ekonomska nauke2011“Spotlight on: Serbia’s competitiveness: Measuring competitiveness and country rankings in the World Economic Forum Global Competitiveness Report.”
Human European Consultancy2007“Implementation report Directive 1999/70/EC concerning the Framework Agreement on fixed-term work concluded by UNICECEEP and ETUC – EU Member states: Czech Republic, Estonia, Cyprus, Latvia, Lithuania, Hungary, Malta, Poland, Slovenia and Slovakia.
OECD - Organization for Economic Cooperation and Development2006“OECD Employment Outlook.”
OECD - Organization for Economic Cooperation and Development2008“OECD Reviews of Labor Market and Social Policies – Serbia, a Labor Market in Transition.”
OECD - Organization for Economic Cooperation and Development2011“Employment Outlook 2011,”Paris, France.
Milieu2009“Implementation Report for Bulgaria - Directive 1999/70/EC concerning the Framework Agreement on fixed-term work concluded by ETUC, UNICE and CEEP.”
Milieu2009“Implementation Report for Romania - Directive 1999/70/EC concerning the Framework Agreement on fixed-term work concluded by ETUC, UNICE, and CEEP.”
Foreign Investor Council2011“White Book 2011”.
World Bank2006“Serbia - Labor Market Assessment.”Report No. 36576 – YU. Washington, D.C.
World Bank2008“Doing Business – An Independent Evaluation,”Washington, DC.
World Bank (2009a). “Guidance Note for World Bank Group Staff on the Use of the Doing Business Employing Workers Indicators for Policy Advice,”Washington, DC.
World Bank (2009b). “Doing Business 2010,”Washington, DC.
World Bank (2010). “BEEPS At-A-Glance 2008 – Serbia,”Washington, DC.
World Bank (2011). “Doing Business 2012,”Washington, DC.
World Economic Forum (2010) “The Global Competitiveness Report 2010–2011,”Geneva, Switzerland.