Determinants of the labour market institutions in post-socialist economies

Communist and Post-Communist Studies xxx (2015) 1e16

Contents lists available at ScienceDirect

Communist and Post-Communist Studies journal homepage: www.elsevier.com/locate/postcomstud

Determinants of the labour market institutions in post-socialist economies Michał Pilc
University of Economics, Poland Poznan

a r t i c l e i n f o

a b s t r a c t

Article history: Available online xxx

The aim of this study was to identify common determinants of the shape of labour market institutions in the group of 25 post-socialist countries during the transition period. The exclusion of countries' fixed effects (with the use of least squares dummy variable e LDSV model) showed that conducted market reforms (especially the privatisation of stateowned companies and enterprise restructuring) have determined the shape of labour market institutions in post-socialist countries. However, the levels of political liberty, protection of property rights, international trade freedom and government expenditures did not occur to have influenced the labour market institutions in these economies. © 2015 The Regents of the University of California. Published by Elsevier Ltd. All rights reserved.

Keywords: Labour market institutions Transition economies

1. Introduction For the last three decades the economic literature has seen a lively and continuously growing discussion concerning the role of institutions in economic performance. Such a lively debate was also present in the labour economics literature where heterogeneity in labour market institutions among countries has been widely described and analysed, especially as a possible determinant of heterogeneity in labour market performance (Boeri and Van Ours, 2008; Layard et al., 2005, p. XIII-XXXIX; Blanchard, 2006; Eichhorst et al., 2008) and as a determinant of the labour market flexibility (Kwiatkowski et al., 2001; Cazes, 2002; OECD, 2004, p. 61e125). However, despite growing amount of studies in that field, the available evidences in the literature still remain inconclusive and often contradictory (Lehmann and Muravyev, 2012, p. 237). As it is stressed by Blanchard (2006, p. 43e46), there are rather no doubts that labour market institutions matter, but the question which ones and how is still current. Moreover, the question ‘why labour market institutions are as they are, and to what extent the current configuration of labour market institutions might be desirable despite sometimes their unfavourable impact on labour market performance’ (Arpaia and Mourre, 2005, p. 17e18) is also still open. The analyses of labour market institutions presented in the literature focus predominantly on the OECD countries. In that context it worth taking notice of the post-socialist economies where labour market institutions were thoroughly rebuilt at the beginning of the transition period and significantly revised in later years. Also labour market outcomes were changing considerably and heterogeneously in these countries. Moreover, the economic transition was a period of deep reforms conducted in order to improve the macroeconomic performance and to transform economic and political institutions. Nowadays, after more than twenty years of transition, post-socialist countries can be seen as much more diversified in terms of the economic performance and the institutional framework than they were at the beginning of the transition process. In consequence, due to that significant diversification in time patterns of the conducted reforms, the transition countries can be perceived as a potentially fruitful ‘laboratory’ that allows to identify relationships among changes in different institutions, not only in the economic and political, but also in the labour market institutions. http://dx.doi.org/10.1016/j.postcomstud.2015.06.008 0967-067X/© 2015 The Regents of the University of California. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

2

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

Therefore, the goal of this study is to identify determinants of the changes in labour market institutions in the group of 25 post-socialist economies1 during the transition period. The following hypothesis is stated: although one of the main characteristics of the changes in labour market institutions in post-socialist countries is the cross-country heterogeneity, the common determinants of the changes in these institutions include the conducted market reforms and the achieved degree of political freedom. Thus, it is conjectured that the conducted labour market reforms were a consequence of both economic and political aspect of the transformation process. Although the analysis of the determinants of changes in the labour market institutions is present in the literature, it is conducted almost only for highly-developed countries where legislative processes are democratic and have not been so significantly changed in the last twenty years. Thus, the analysis of post-socialist countries is not only original, but can also shed some new light on the question concerning the determinants of labour market institutions. Moreover, thanks to the recent work of Lehmann and Muravyev (2012), who published the seminal dataset of labour market institutional indicators for transition countries, we will be able to identify these determinants empirically, which is the main novelty of the presented research. The paper is organised as follows. Section 2 provides a synthetic description of the changes in the labour market institutions in post-socialist economies based on the critical review of the literature and analysis of the collected data. It presents also results of the cluster analysis that shows that one of the main characteristics of the changes in the labour market institutions in post-socialist economies is the cross-country heterogeneity. Section 3 reviews the literature concerning determinants of the labour market institutions in highly-developed countries and identifies theories that may be applied also for post-socialist countries. Section 4 presents a fixed effects model, estimated in order to identify common determinants of the changes in the labour market institutions in post-socialist countries. Section 5 gives a brief summary of the conducted analyses and suggests areas of further research.

2. Changes in the labour market institutions during the transition 2.1. Remarks from the literature In the transition countries the processes of restructuring, privatisation and, what is characteristic for all post-socialist countries, the transformational recession (Havrylyshyn, 2001) caused significant changes on the labour markets during the transition period. It can be seen especially in the significant decreases in employment and economic activity levels, and also in the occurrence (almost all centrally planned economies were formally fully-employed) and rapid growth of unemployment and informal employment (Cazes and Nesporova, 2003, p. 7e19). What is more, the transition economies were characterised by the strong excess supply of low-skilled workers (demanded in generally heavy-industry oriented socialist economies) and by a high demand for better skilled workers, with qualifications more suitable for service and consumer oriented market economy (Winiecki, 2008). Therefore, in order to adapt to these changes, governments in post-socialist countries had to reform the labour market institutional framework substantially. However, due to the shortage of data, especially for the majority of CIS (Commonwealth of Independent States) countries, the labour market institutions analyses focused mainly on the transition countries that became EU members ~rk, 2007; Fialova and Schneider, 2009; Tonin, (Cazes and Nesporova, 2003; Boeri and Garibaldi, 2005; Anspal and Vo 2009), on the selected groups of countries (Eamets and Masso, 2005; Gligorov et al., 2008) or on particular countries (Kajzer, 2007; Micevska, 2008). In that context it is worth mentioning the recent work of Muravyev (2010) where the values of Employment Protection Legislation (EPL) index2 for former USSR countries for the years 1990e2009 are calculated. These results were further used in the work of Lehmann and Muravyev (2012) where labour market institutions were analysed as the determinants of labour market performance with the use of the new, hand-collected database of the labour market institutions indicators created for 27 post-socialist countries. Although their database allowed to include into the analysis also the CIS countries, the data in that study were collected only for four years: 1995, 1999, 2003 and 2007. Based on the mentioned literature, three general conclusions concerning the labour market institutions in the postsocialist countries can be stated. Firstly, labour market institutions matter as a framework for economic activity on the labour market and as an important determinant of the flexibility of that market. Secondly, in the majority of the transition countries labour market institutions do not provide less flexibility than the institutions in the EU-15, and in many cases even more. And thirdly, despite some significant similarities among labour market institutions in many countries (for instance the general tendency to decrease the union density rate), the general characteristic of the labour market institutions in the transition economies is their heterogeneity, both in time and especially in space.

1 The data were collected for the following countries: Albania, Armenia, Azerbaijan, Belarus, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Macedonia, Moldova, Poland, Romania, Russia, Slovakia, Slovenia, Tajikistan, Turkmenistan, Ukraine and Uzbekistan. 2 The EPL index has been proposed by OECD (2014) in order to measure the strictness of employment protection legislation. It takes values from 0 to 6, the higher the value the more protective the labour law is.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

3

Table 1 Variables and their sources. The type of variable

Variable

Source

Labour market institutions

EPL EPL EPL EPL

OECD, Lehmann and Muravyev (2012), Muravyev (2010), Gligorov et al. (2008), Micevska (2008), Kajzer (2007) Eurostat, Lehmann and Muravyev (2012) Eurostat, Doing Business. Employing Workers, World Bank Lehmann and Muravyev (2012)

(2nd version), for regular contracts, for temporary contracts, for collective dismissals

Tax wedge Minimum wage Average unemployment benefit, Maximum benefit duration Expenditure on ALMP Expenditure on PLMP Union density

Market reforms

Other

Union coverage, Wage bargaining coordination Large scale privatisation, Small scale privatisation, Governance and enterprise restructuring, Competition policy Political Rights Trade Freedom, Property Rights Government expenditure

Eurostat, Lehmann and Muravyev (2012) Eurostat ICTWSS Database, Amsterdam Institute for Advanced Labour Studies (AIAS), Lehmann and Muravyev (2012) ICTWSS Database, Amsterdam Institute for Advanced Labour Studies (AIAS) Transition Reports, EBRD

Freedom in the World, Freedom House Organization Index of Economic Freedom, Heritage Foundation EBRD

Source: own elaboration.

2.2. Descriptive analysis of the collected data The analysis of the changes in the labour market institutions in this study was based on the database of 14 indicators describing labour market institutions and created for 25 post-socialist economies (for the sources of particular variables see Table 1, for descriptive statistics see Table 2). It was decided to use the data on labour market institutions for four years: 1995, 1999, 2003 and 2007 respectively. Although it was possible to create longer time series of the labour market institutions for countries from the Central and Eastern Europe3 (CEE) and for some countries from the South Eastern Europe (SEE), in case of some labour market institutions for many CIS and SEE countries the only observations that could be collected were these published by Lehmann and Muravyev (2012) who created their database only for these four years. Although it is possible to estimate a panel data model based on a dataset with different amount of observations for particular countries, such specification in this case would produce East-European biased results. Therefore, it was decided to limit the data only to these four years with the smallest amount of missing values,4 in order to obtain results more equally representative for all the groups of post-socialist countries (CEE, CIS and SEE). The decision to choose these particular labour market institutions indicators was based on the literature (Blanchard and Wolfers, 2000; Layard et al., 2005, p. XXVII eXXXIX; brief comparison of various researches and used indicators is presented in Arpaia and Mourre (2005, p. 33e35)) and on the data availability. Analysis of the collected data indicates that various labour market institutions changed significantly and heterogeneously in the analysed countries (see Fig. 1aen). The level of employment protection legislation (EPL) generally did not change significantly in CEE and CIS countries but decreased in SEE (see Fig. 1a). Although some weakening of the protection for regular employment can be observed (see Fig. 1b), it is compensated by strengthening of the protection of temporary employment (see Fig. 1c) which is connected with the growing proportion of temporary contracts in employment contracts in the transition countries (Cazes and Nesporova, 2003, p. 44). The collective dismissals occurred to be generally far less regulated in CIS than in CEE and SEE (see Fig. 1d). It should be, however, added that the detailed analysis of the dataset for

3 In the research the following categorisation of the countries is used: Central and Eastern Europe (CEE): Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland and Slovakia; Commonwealth of Independent States (CIS): Armenia, Azerbaijan, Belarus, Georgia, Kazakhstan, Kyrgyzstan, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine and Uzbekistan; South Eastern Europe (SEE): Albania, Bulgaria, Croatia, Macedonia, Romania and Slovenia. 4 In case of the analysis of the institutional variables the popular approach is to use few years' averages. However, if this approach had been used in this study, it would have meant that for many CIS and SEE countries such few years' averages would have been calculated only with the use of one year's observation, which has no reasonable justification.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

4

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

Table 2 Descriptive statistics and short descriptions of variables. Variable

Short description

Obs.

Mean

EPL (2nd version) EPL for regular contracts EPL for temporary contracts EPL for collective dismissals Tax wedge Minimum wage (Eurostat) Minimum wage (Doing business) Average unemployment benefit Maximum benefit duration Expenditure on PLMP Expenditure on ALMP Union density Union coverage

Values form 0 to 6; 6 denotes the most restrictive employment protection legislation

100 84 84 83 81 24

2.29 2.92 1.48 2.29 38.6 37.04

0.55 0.7 0.95 1.53 5.39 5.77

24

20

9.1

0

36.1

81

24.38

10.25

0

45

80 16 72 76 21

10.75 0.28 0.19 41.98 50.04

5.22 0.13 0.18 21.94 27.08

40

1.95

1.02

1

4

100 100 100

2.8 3.59 2.23

0.91 0.84 0.73

1 1 1

4 4.33 3.67

100 100

2.16 3.39

0.63 2.1

1 1

3.67 7

93 93 100

68.58 43.33 36.45

12.14 15.33 10.52

17.6 10 12.3

Wage bargaining coordination

Large scale privatisation Small scale privatisation Governance and enterprise restructuring Competition policy Political Rights Trade Freedom Property Rights Government expenditure

% of the total labour costs of the earner % of average monthly earnings in industry and services Relation of the minimum wage for a 19 year-old employee to the economy's average value added per worker (in %) % of an average wage In months % of GDP % of GDP Union members as a % of wage and salary earners in employment Employees covered by wage bargaining agreements as a % of all wage and salary earners in employment with the right to bargaining Values from 1 to 5; 5 e economy-wide bargaining, 4 e mixed industry and economy-wide bargaining, 3 e industry bargaining, 2 e industry-and firm level bargaining, 1 e firm level bargaining Values from 1 to 4.3; 1 e lack of reforms, 4.3 e institutional framework characteristic for developed economies

Values form 1 to 7; 1- maximum level of political rights Values form 0 to 100; 100 e maximum level of trade freedom or property rights protection % of GDP

Std. Dev.

Min 0.41 0.73 0.25 0 23 21.3

0 0.1 0.01 7.31 12.5

Max 4.1 4.39 4.3 5.3 48.27 45.8

24 0.52 0.83 100 100

85 70 60.4

Source: own elaboration. Variables descriptions based on the sources listed in Table 1.

particular countries reveals that as far as the EPL is concerned, rather the cross-country not the cross-regional heterogeneity prevails. The tendency to weaken the employment protection coincided with the slight tendency to lower the tax wedge, especially in CIS countries (see Fig. 1e). In case of the minimum wage, values of two indicators were collected (from Eurostat and Doing Business), because they measure it in a different way.5 What is more, Eurostat does not collect values for CIS countries, while in the Doing Business the data have been presented since 2006. Thus, we are only able to say that the level of minimum wage (see Fig. 1f) decreased in CEE during the transition period, but rose in SEE. While at the end of the analysed period (see Fig. 1g) the average level of minimum wage was the lowest in CIS, greater in CEE and more than two times greater in SEE (comparing to CIS). As far as the labour market policy is concerned, once more the cross-country heterogeneity prevails in the transition countries, but some general slight tendency to lower the unemployment benefits (see Fig. 1h) and the maximum duration of paying the benefits (see Fig. 1i) can be observed. It coincided with the tendency to lower the rate of expenditures on ALMP as the percentage of GDP in CIS and SEE and to increase it in CEE countries in later years of the transition period (see Fig. 1k). In the transition countries a general tendency to significantly lower the union density rate (see Fig. 1l) can be observed, which corresponds with the changes in the union coverage rate (see Fig. 1m). At the beginning of the transition period the post-socialist countries had generally decentralised the process of wage bargaining, and in consequence the coordination of wage bargaining at industry and company levels predominated during the transition period (see Fig. 1n). In addition to the descriptive analysis for particular labour market institutions, also a cluster analysis for the set of six equally weighted variables (EPL, tax wedge, average unemployment benefit, maximum benefit duration, expenditure on ALMP and union density), which were chosen due to the best data availability, was performed. It was carried on in order to identify groups of similar countries over time in terms of the labour market institutions. However, the results of the

5 Eurostat (2014) collects data concerning the ratio of the national minimum wage to the average monthly earnings in industry and services, while the World Bank's (2014) Doing Business report presents the ratio of the minimum wage to the average value added per worker.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

5

Fig. 1. Changes in the labour market institutions in the transition period. Source: own elaboration, based on the data presented in Tables 1 and 2. Some calculated regions' averages for CIS or SEE countries can be misleading, as due to the missing values they were not calculated for all the countries from the mentioned regions.

conducted analysis (allocation of the particular countries to the clusters is presented in Table 3 and dendrograms in Figs. 2e5) do not allow to identify any durable similarities among the analysed countries, at least as far as the group of six mentioned labour institutions is concerned. Results show rather very significant cross-country heterogeneity among the time patterns of the changes in labour market institutions, which confirms and strengthens the third mentioned conclusion derived from the literature. 3. Possible determinants of the changes in the labour market institutions in the transition countries e remarks from the literature Although, to our knowledge, this is the first research that is trying to explain the shape of the labour market institutions in the post-socialist countries as a result of the process of economic and political transformation, the labour economics literature Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

6

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

Fig. 1. (continued).

provides some studies concerning the determinants of the labour market institutions in highly-developed countries, which may appear to be applicable also for the transition countries.6 Arpaia and Mourre (2005) give a brief summary of the knowledge in that field and identify three general types of views that are proposed in the literature. The first view suggests that current labour market regulations have their origins in the history of the country's law. This approach, however, is rather not adequate to explain the labour market institutions heterogeneity in the post-socialist countries due to their common legacy of the centrally directed and fully-employed economy. The second view underlines that particular labour market institutions are introduced after weighing their social costs against their social benefits. With the use of this approach explanations were given, why some labour market regulations may be beneficial for the society even if they can hamper the functioning of the labour market. For instance, Bertola and Koeniger (2004) show that strict employment protection and high unemployment benefits may be introduced, in order to reduce labour income fluctuations in countries with under-developed financial systems, where consumer credit is relatively scarce. Agell (2002) points that strict employment protection and generous unemployment benefits can be also a substitute for other instruments of social protection policy. These conceptions can be potentially important for the transition countries, because the financial systems in these countries were considerably under-developed. Moreover, the scale of the social protection significantly decreased at the beginning of the transition process since the state-owned companies had to reduce the scale of

6 Although, an attempt to apply the conclusions derived for highly-derived countries to post-socialist economies is not free from controversies, it was undertaken, because one of the crucial elements of the transition process was to introduce such rules of the market economy that were present in the highly-developed countries (Kornai, 2006; Pia˛ tek et al., 2013). Moreover, many empirical analyses (Grzymala-Busse and Innes, 2003; Schweickert et al., 2011; Staehr, 2011) indicate that one of the main determinants of institutional changes in CEE and many SEE countries was the desire to adapt the institutional framework to the EU standards in order to fulfil the accession criteria.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

7

Table 3 Clusters of countries in terms of the labour market institutions.

the social transfers to their employees, such as: subsidies for meals, childcare, health care services and others (Milanovic, 1995; Fakin and de Crombrugghe, 1997). The third view identified by Arpaia and Mourre uses the term of social conflict to show that the shape of institutions is rather a result of the rivalry between different social groups with diversified political power who are trying to maximise their own rents.7 This approach is used by Boeri (2005) to show that increase in product market competition (which can be a result of market reforms), which pressures the companies to increase their productivity, could lead to decrease in employment, and in consequence can cause stronger pressure from workers to make labour law more protective or unemployment benefits more generous.8 This mechanism can be important in the case of the transition countries, because they entered into the transition period with large amount of state-owned companies, characterised by the low labour productivity and high employment of low-skilled workers. These companies had to start to compete internationally and therefore had to conduct the process of restructuring which often included large employment reduction (Cazes and Nesporova, 2003, p. 8e10). Some authors have been also trying to explain the changes in the labour market institutions by the use of other labour market institutions. For instance Algan and Cahuc (2009) and Boeri et al. (2003) argue that since the employment protection

7 This way of determining the shape of formal economic institutions is also strongly supported by Acemoglu et al. (2005, p. 389e396) in their theory of a hierarchy of institutions which indicates that changes in de jure or de facto political power precedes the changes in economic institutions. 8 Saint-Paul (1996) argues that the same mechanism could also work in the times of economic slowdown. What is more, the model proposed by SaintPaul (2002) also shows that the scale of pressure to strengthen employment protection depends on workers' human capital. The lower the workers' general human capital and the higher their job-specific, the higher is the workers' fear of being redundant e and in consequence the higher the pressure to strengthen employment protection.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

8

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

Fig. 2. Dendrogram of the countries in terms of labour market institutions for 1995. Source: own elaboration. The Ward method for hierarchical clustering was used. The variables were standardised and missing data were filled with the average values of particular variables in the analysed year. The countries are listed on the vertical axis, and the horizontal axis represents the relative Euclidean distance (as a percent of the maximum observed distance) between the centroids of the clusters. A large Euclidean distance between the centroids of the clusters is interpreted as a large difference between the clusters.

legislation (EPL) and unemployment benefits have the same goal e which is to ensure some protection against unemployment risk e they can be treated as substitutes. Thus, they indicate that there is a trade-off between the EPL and the unemployment benefits. In addition, Boeri et al. (2003) have shown that countries with the majority of low-skill insiders or larger share of elderly workers tend to choose more restrictive EPL rather than generous unemployment benefits. Checchi and Lucifora (2002) also indicate that there is a negative correlation between the union density and the EPL, since e as they argue e weak employment protection can be an incentive for workers to join the union. However, in the discussion section of their article, Boeri shows (Checchi and Lucifora, 2002, p. 401e403) that there exists a rather positive correlation between the EPL and the union coverage, because one of the union goals should be to strengthen the employment protection. Also Saint-Paul (2002) argues that the greater the workers' bargaining power, the greater is the pressure on strengthening the employment protection. Thus, the relation between trade unions and employment protection must be considered as ambiguous. The role of informal institutions in determining the shape of formal institutions is indicated by North (1997), who argues that the pace and scale of the changes in formal institutions is influenced by the characteristic of informal institutions that shape the way individuals interpret received information from the economic, social and political environment. North (1997, p. 16) indicates also that different characteristics of the transformation processes in CEE and CIS result from differences in

Fig. 3. Dendrogram of the countries in terms of labour market institutions for 1999. Source: the same as for Fig. 2.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

9

Fig. 4. Dendrogram of the countries in terms of labour market institutions for 2003. Source: the same as for Fig. 2.

informal institutions between these two regions that were inherited from their history. Thus, a conjecture can be made that the lack of heritage of market economy and democracy in the CIS countries' societies has resulted not only in the lower level of democratisation in these countries, but also in the lower political pressure from the society to strengthen the protective role of the labour market institutions. This could be one of the explanations why in CIS countries have lower levels of the EPL and unemployment benefits than in CEE countries are observed. Generally, it can be stated that although the studies presented above are predominantly related to the highly-developed countries, they allow to conjecture that market reforms conducted in the transition countries, with the aim to strengthen the competitiveness of national economies, could have determined the shape of labour market institutions in these countries. These reforms, on the one hand, might gain public support, since liberalisation of the labour market institutions could have improved the companies' productivity and in consequence increase their chances to develop and raise wages in the future, but on the other hand, the companies' restructuring was unfavourable for many workers with poor qualifications who were in danger of being dismissed and carrying the costs of these reforms. The political power of these workers (and thus the ability to influence the shape of labour market institutions) could depend on the level of political liberty that was ensured to the citizens in particular transition countries. Thus, it can be concluded that the observed heterogeneity in the labour market institutions in the transition countries is not an exogenously given characteristic, but can be at least partially explained as a result of the heterogeneity in the pace and scale of the market reforms, and in the level of political liberty in the transition countries. The next section presents empirical verification of this conjecture.

Fig. 5. Dendrogram of the countries in terms of labour market institutions for 2007. Source: the same as for Fig. 2.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

10

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

4. Model 4.1. Data The sources of particular variables analysed as possible determinants of the labour market institutions in the post-socialist countries are presented in Table 1, while their descriptive statistics and short descriptions are shown in Table 2. In the previous section a conjecture has been stated that the shape of labour market institutions can be determined by the level of political freedom ensured to the citizens in particular transition countries. In order to measure it, the Freedom House's Political Rights index was used that evaluates the extent to which people are allowed to participate freely in the political process, including the right to vote freely, compete for public office, join political parties and organisations, and elect their representatives (Freedom House, 2012). As measures of conducted market reforms four indicators from the EBRD Transition Indicators were included into the estimations. These were: Large scale privatisation, Small scale privatisation, Governance and enterprise restructuring, and Competition policy (EBRD, 2012a). Large and small scale privatisation measure the extent to which private entrepreneurs have taken the control over the previously state-owned companies. Governance and enterprise restructuring captures the effectiveness of corporate governance, while the Competition policy measures the actions taken to reduce the abuse of large companies' market power and to promote a competitive environment (EBRD, 2012b). To measure the economy's openness for the international competition, the Trade Freedom index was used that was published by the Heritage Foundation and is a composite measure of the absence of tariff and non-tariff barriers that affect imports and exports of goods and services (Heritage Foundation, 2012). Since the level of some labour market institutions (for instance the tax wedge, the expenditure on labour market policy) can be determined by the size of government budget, also the total expenditure of the government as a percentage of GDP from the EBRD database (EBRD, 2012a) was included into the estimations. Moreover, the government expenditure may also be treated as a proxy for the level of social transfers9 which, as it was mentioned, can be substituted for other instruments of protection against unemployment risk (like the EPL or unemployment benefit). Additionally, in estimations for the EPL, average unemployment benefit and union density, the Heritage Foundation's Property Rights index was used, since it was assumed that the choice between the EPL and unemployment benefits can be influenced by the extent to which individuals are able to enforce the labour law e one of the aims of the Property Rights is to measure the independence and efficiency of the court system (Heritage Foundation, 2012). This index was also included in the estimations for the union density, because the literature (Checchi and Lucifora, 2002; Saint-Paul, 2002) shows that the union density can be correlated with the EPL and the average unemployment benefit. Due to the data availability it was decided to conduct the estimations for the following set of five labour market institutions with the smallest amount of missing values: EPL, average unemployment benefit, union density, tax wedge and expenditure on ALMP. 4.2. Details of the estimation procedure In the estimations the fixed effects10 (LSDV) model was used with the general form that can be written in the following way:

LMIit ¼ const þ b1 FH PRit1 þ b2 HF TFit1 þ b3 HF PRit1 þ b4 EBRD GOVit1 þ b5 EBRD MRit1 þ b6 O LMIit þ ci þ tt þ εit (1) where i represents country and t denotes time (years: 1995, 1999, 2003 or 2007 respectively), ci is a country fixed effect, tt is a time dummy and εit is a white noise disturbance. LMI represents particular labour market institution. FH_PR denotes the Freedom House's Political Rights index, HF_TF stands for the Heritage Foundation's Trade Freedom index, and HF_PR is the Heritage Foundation's Property Rights index which, as mentioned, was included in the estimations with the following independent variables: EPL, average unemployment benefit or union density. EBRD_GOV stands for the total government

9 Although this assumption has clear limitations, it may be applied to post-socialist countries during the transition process (especially in its first years), due to the shortage of proper data concerning the social transfers for the whole region and due to the fact that all these countries inherited the legacy of the centrally planned economy where the government was fully responsible for organising and providing the social protection (Rys, 2001). Moreover, as the data collected by Eurostat indicate (Freysson and Wahrig, 2013), the social protection still constituted the highest category of the government expenses in all new EU member states in 2011. 10 It was decided to use the fixed effects specification, because it allows to obtain consistent estimates even if the missing observations are not randomly distributed among the analysed objects (Wooldridge, 2001, p. 579e580), which is mostly probably the case in the analysed data, because there is a clear correlation between the progress of the transition reforms and the data availability for particular countries. In such a situation the random effects estimator would provide inconsistent results (Wooldridge, 2001, p. 580). Therefore, it was decided not to use this method. Moreover, the decision to use the fixed effects regression was also motivated by the fact that it allows to directly control the unobserved factors such as informal institutions (Wooldridge, 2001, p. 266).

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

11

Table 4a Results of the fixed effects regression for the EPL and the average unemployment benefit. Dependent variable

EPL

Independent variables

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

Political Rights

0.015 (0.088) 0.004 (0.004) 0.007 (0.01) 0.013 (0.01) 0.075 (0.084)

0.002 (0.09) 0.003 (0.005) 0.007 (0.01) 0.012 (0.01)

0.005 (0.09) 0.004 (0.005) 0.007 (0.011) 0.012 (0.01)

0.03 (0.09) 0.005 (0.004) 0.007 (0.01) 0.014 (0.01)

0.04 (0.11) 0.003 (0.01) 0.005 (0.012) 0.017 (0.012)

0.06 (0.115) 0.01 (0.01) 0.004 (0.012) 0.013 (0.012)

0.102 (1.06) 0.085 (0.068) 0.093 (0.094) 0.11 (0.176) 2.01 (3.24)

0.415 (0.714) 0.106 (0.08) 0.114 (0.087) 0.03 (0.14)

0.375 (0.864) 0.067 (0.093) 0.102 (0.1) 0.1 (0.191)

0.4 (0.765) 0.074 (0.09) 0.09 (0.095) 0.13 (0.19)

0.43 (0.77) 0.1 (0.077) 0.1 (0.08) 0.013 0.136

0.47 (0.77) 0.09 (0.077) 0.094 (0.061) 0.035 (0.15)

5.85** (2.65)

6.82*** (2.34)

Trade Freedom Property Rights Government expenditure Large scale privatisation Small scale privatisation Gov. and enterp. restructuring Competition policy

Average unemployment benefit

0.09 (0.134)

6.23*** (2.324) 0.121 (0.126)

3.449 (5.99) 0.297*** (0.105)

0.35** (0.149)

0.4*** (0.124)

1.035 (2.492)

EPL

1.696 (1.76)

Av. unemployment benefit Union density Const Observations Cross-sections included Fixed periods effects Adjusted R2 Within R2 DurbineWatson statistic P-value for the F statistic P-value for the F test with H0: all the cis are equal P-value for the Breusch ePagan test

0.009 (0.008)

0.81** (0.4) 79 24

1.77** (0.673) 79 24

1.17 (0.157) 79 24

0.7 (0.566) 79 24

1.75** (0.72) 79 24

0.11** (0.046) 2.87*** (1.055) 70 21

0.005 (0.01) 93 25

0.007 (0.012) 93 25

0.01 (0.013) 93 25

0.01 (0.01) 93 25

0.005 (0.06) 79 24

0.004 (0.004) 0.011 (0.084) 74 23

No 58.9% 7.65% 1.12

No 58.9% 7.84% 1.14

No 58.9% 7.64% 1.15

No 61.07% 12.5% 1.14

No 58.19% 13.9% 1.19

No 54.08% 12.9% 0.98

No 68.2% 4.77% 1.36

No 70.55% 11.7% 1.46

No 68.24% 4.78% 1.39

No 67.9% 3.77% 1.37

No 70.34% 12.85% 1.49

No 74.4% 20.2% 1.72

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Source: own estimates. The robust standard errors (obtained with the use of the Arellano's (2003, p. 18) estimator) are reported in parentheses. Asterisks denote significance levels: *** e significant at 1%, ** e significant at 5% and * e significant at 10%. In order to conduct the BreuschePagan test the models was reestimated as pooled models.

expenditure, and EBRD_MR are indicators of the market reforms. Since high correlation rates were observed among them, for each dependent variable four separate estimations were made, one for each market reform indicator respectively. Independent variables were lagged by one year (thus, they refer to years: 1994, 1998, 2002 and 2006 respectively) in order to avoid direct manifestation of the endogeneity problem. It was already stated that one may expect a correlation among different labour market institutions indicators. Therefore, in case of all dependent variables also some additional estimates were conducted, where other labour market institutions indicators were included (variable O_LMI in the equation). Since there were some missing values for the analysed labour market institutions indicators, such estimates were performed separately for each additional variable. It was done not only to measure the relations among different labour market institutions indicators, but also to give some robustness check of the previous estimates (that is without the inclusion of other labour market institutions) with different number of observations and cross-sections included into the model.11

11 It was also decided to perform additional robustness checks that are presented in Table 5. To perform these tests a simple specification is estimated at the first step: LMIit ¼ const þ ci þ εit . Then, with the use of the Wald test (based on the F-distribution) it is verified whether it is statistically justified to augment this equation with one particular independent variable that was either the Freedom House's indicator of political rights or one of the ERBD indexes of market reforms. We use only these independent variables, because there have no missing values. Therefore, in this robustness check we are sure that all observations of the labour market institutions indicators are included in the estimations (which is not necessarily the case in the estimates based on the Equation (1)). Moreover, these two types of indicators refer directly to the hypothesis stated in the research. Additionally, we estimate also the second specification which has the following form: LMIit ¼ const þ g FH PRit1 þ ci þ εit . Then, once more with the use of the Wald test it is verified whether it is justified to augment this equation with one of the ERBD indexes of market reforms.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

12

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

Table 4b Results of the fixed effects regression for the union density. Dependent variable

Union density

Independent variables

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

Political Rights

2.78 (2.74) 0.07 (0.21) 0.208 (0.136) 0.088 (0.218) 5.095 (4.367)

1.533 (2.789) 0.062 (0.191) 0.193 (0.142) 0.019 (0.256)

1.372 (2.57) 0.118 (0.226) 0.005 (0.159) 0.013 (0.248)

1.744 (2.917) 0.003 (0.235) 0.09 (0.15) 0.17 (0.273)

1.569 (2.759) 0.048 (0.196) 0.18 (0.14) 0.003 (0.245)

0.915 (2.89) 0.121 (0.248) 0.119 (0.165) 0.031 (0.383)

1.433 (2.42) 0.106 (0.222) 0.019 (0.165) 0.031 (0.241)

0.928 (2.775) 0.107 (0.26) 0.042 (0.129) 0.102 (0.296)

6.953** (3.153)

0.84 (7.74) 15.57** (5.83)

14.87*** (5.39)

Trade Freedom Property Rights Government expenditure Large scale privatisation

6.736** (3.16)

Small scale privatisation

15.1** (5.624)

Gov. and enterp. restructuring

7.928 (6.656)

Competition policy EPL

2.23 (3.38)

4.18 (2.722) 74 23 Yes 73.26% 17.45% 1.37 0.00 0.00

5.435* (2.89) 74 23 Yes 73.95% 19.6% 1.38 0.00 0.00

2.11 (2.35) 74 23 No 71.9% 7.2% 1.28 0.00 0.00

0.655 (1.818) 74 23 No 71.3% 5.3% 1.25 0.00 0.00

5.39* (2.767) 74 23 Yes 73.5% 20.1% 1.38 0.00 0.00

0.59** (0.222) 1.716 (3.313) 70 21 No 71.7% 8.76% 1.23 0.00 0.00

0.00

0.00

0.00

0.00

0.00

0.00

Av. unemployment benefit Const Observations Cross-sections included Fixed periods effects Adjusted R2 Within R2 DurbineWatson statistic P-value for the F statistic P-value for the F test with H0: all the cis are equal P-value for the Breusch ePagan test

2.305 (3.053)

2.12 (2.298) 74 23 No 71.44% 7.72% 1.29 0.00 0.00

0.612*** (0.193) 2.476 (3.229) 70 21 No 73.31% 13.9% 1.34 0.00 0.00

0.00

0.00

Source: own estimates. The robust standard errors (obtained with the use of the Arellano's (2003, p. 18) estimator) are reported in parentheses. Asterisks denote significance levels: *** e significant at 1%, ** e significant at 5% and * e significant at 10%. In order to conduct the BreuschePagan test the models was reestimated as pooled models.

Since the parameters' estimates occurred to be unstable while estimating the model on the levels of variables, it was decided to transform the variables into the differences from the cross-country average values for particular years.12 Such transformation lessened the correlation between particular independent variables and gave stable parameters' estimates. Variables were not logarithmically transformed, since the fit of the estimated models occurred to be worse after the logarithmic transformation. The decision to include the time dummies was based on the Wald test for their joint significance. It was also decided to use the estimator of robust standard errors proposed by Arellano (2003, p. 18), which is designed to handle both heteroscedasticity and autocorrelation in the data sets consisting of many units observed in relatively few periods. The omitted factors were controlled by the use of fixed-effects specification of the regression model. 4.3. Results The final results of the estimations are presented in Table 4aec. They show that the EPL index (Table 4a, columns 1e6) did not occur to be determined by any of the independent variables with the exception of the Competition policy index. The obtained relationship suggests that the transition countries which decided to strengthen their domestic competition decided also to strengthen the employment protection of their workers. However, this relationship was not confirmed by the additional robustness checks (see Table 5). Moreover, the level of openness to international competition (measured by the Trade Freedom) did not occur to be connected with the values of EPL index. Therefore, it can be noticed that these results give only a little support to the model proposed by Boeri (2005, p. 13), which indicates that there may be a connection between the scale of the product market competition and the EPL. Furthermore, the EPL index also occurred not to be correlated with the

12

0

These transformed values were calculated in the following way: xit ¼ xit 

P

xit =nx;t , where nx,t is the number of non-missing observations for variable

i

x in the year t. Therefore, the estimations were in fact based on the detrended data. Although, the panel data models are less vulnerable to the problem of spurious regression than pure time series models, because the independent cross-section data in the panels are a source of additional information (Baltagi, 2005, p. 251), this problem can still occur. Therefore, the applied transformation allowed not only to improve the fit of the model, but also to limit the potential problem with nonstationarity.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

13

Table 4c Results of the fixed effects regression for the tax wedge and the expenditure on the ALMP. Dependent variable

Tax wedge

Independent variables

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

Political Rights

0.36 (1.02) 0.07 (0.066) 0.001 (0.08) 0.12 (1.7)

0.316 (0.8) 0.067 (0.06) 0.001 (0.083)

0.437 (0.769) 0.07 (0.063) 0.007 (0.068)

0.019 (0.801) 0.12* (0.066) 0.048 (0.096)

0.74 (0.73) 0.026 (0.052) 0.05 (0.089)

0.131 (0.811) 0.042 (0.06) 0.001 (0.063)

0.436 (0.751) 0.071 (0.053) 0.015 (0.112)

0.111 (0.789) 0.1 (0.062) 0.04 (0.074)

0.016 (0.02) 0.003 (0.002) 0.002 (0.002) 0.022 (0.079)

0.02 (0.017) 0.001 (0.002) 0.001 (0.001)

0.019 (0.02) 0.003 (0.002) 0.002 (0.104)

0.019 (0.026) 0.003 (0.003) 0.002 (0.001)

0.02 (0.017) 0.001 (0.002) 0.003 (0.002)

0.018 (0.018) 0.002 (0.003) (0.002) (0.001)

0.188** (0.08)

0.194** (0.096)

Trade Freedom Government expenditure Large scale privatisation Small scale privatisation Gov. and enterp. restructuring Competition policy

Expenditure on ALMP

0.62 (1.45)

0.182** (0.083) 3.37** (1.467)

2.477 (1.81)

2.804 (1.709)

3.58* (1.834)

0.01 (0.104) 2.937 (1.9)

3.474** (1.628)

0.004 (0.07) 0.005 (0.004)

Tax wedge 1.463 (2.727)

Exp. on ALMP Av. unemployment benefit Const 0.25 (0.32) Observations 79 Cross-sections 24 included Fixed periods No effects 2 Adjusted R 54.4% Within R2 2.7% DurbineWatson 1.64 statistic P-value for the F 0.00 statistic 0.00 P-value for the F test with H0: all the cis are equal 0.02 P-value for the BreuschePagan test

1.574 (2.93) 0.198** (0.075) 0.44 0.097 (0.552) (0.467) 69 73 21 23

0.001 (0.004) 0.01 0.124** 0.009 0.006 0.14*** 0.119* (0.026) (0.053) (0.03) (0.017) (0.05) (0.059) 69 69 69 69 69 68 21 21 21 21 21 21

0.124 (0.39) 79 24

0.225 (0.435) 79 24

0.514* (0.3) 79 24

0.615 (0.956) 69 21

0.207** (0.085) 0.94 (0.8) 73 23

No

No

No

No

No

No

No

No

Yes

No

No

Yes

Yes

53.9% 2.94% 1.66

55.4% 5.95% 1.61

57.6% 10.7% 1.87

49.7% 6.3% 1.5

53.3% 14.05% 1.66

52.1% 10.8% 1.73

56.7% 20.3% 1.9

48.7% 4.91% 1.74

50.8% 14.9% 1.82

48.6% 4.72% 1.74

48.6% 4.7% 1.73

50.3% 16.2% 1.82

49.3% 15.5% 1.84

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.06

0.04

0.1

0.08

0.05

0.1

0.01

0.01

0.02

0.00

0.00

0.00

0.00

0.17

0.06

0.13

0.15

0.1

0.13

Source: own estimates. The robust standard errors (obtained with the use of the Arellano's (2003, p. 18) estimator) are reported in parentheses. Asterisks denote significance levels: *** e significant at 1%, ** e significant at 5% and * e significant at 10%. In order to conduct the BreuschePagan test the models was reestimated as pooled models.

average unemployment benefits and the union density, which shows that the correlation between these institutions identified for the highly-developed countries did not apply to the post-socialist economies. Average unemployment benefit (Table 4a, columns 7e12) occurred to be determined by the level of the Small scale privatisation. The greater the scale of that privatisation was, the higher unemployment benefits were offered. Quite surprisingly, the level of unemployment benefits was not correlated with the Large scale privatisation and with the Government Expenditure. However, the unemployment benefits happened to be connected with the union density. The direction of this relation is consistent with the relation between the union density and other form of the unemployment risk protection e the EPL e that was observed by Checchi and Lucifora (2002), and indicates that the higher the unemployment benefits were, the lower the union density was. However, the results for the union density show (Table 4b, columns 13e20) that in case of the post-socialist countries the relationship between the union density and the unemployment benefits was probably driven by the Small scale privatisation, since also the union density occurred to be determined by this market reform indicator. The greater the level of Small scale privatisation was, the lower the union density was. The union density occurred also to be connected with the Governance and enterprise restructuring. The greater the scale of this reform was, the lower the union density rate was. As far as the tax wedge is concerned (Table 4c, columns 21e28), quite surprisingly, it was not connected with the Government Expenditure, but it occurred to be negatively correlated with the Governance and enterprise restructuring and positively correlated with the Competition Policy. However, in the additional robustness checks (see Table 5) only the relationship Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

14

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

Table 5 P-values of the Wald tests for the significance of indicators of market reforms and political rights. Tested specification

LMIit ¼ const þ g1 Xit1 þ ci þ εit

LMIit ¼ const þ g1 Xit1 þ g2 FH PRit1 þ ci þ εit

Tested hypothesis

g1 ¼ 0

g1 ¼ 0

X:

LMI:

Political Rights Large scale privatisation Small scale privatisation Gov. and enterprise restructuring Competition policy

EPL

Average u. benefit

Union density

Tax wedge

Exp. on ALMP

EPL

Average u. benefit

Union density

Tax wedge

Exp. on ALMP

0.661 0.755 0.508 0.882

0.643 0.43 0.076 0.417

0.415 0.541 0.042 0.1

0.502 0.982 0.261 0.243

0.415 0.361 0.12 0.754

n/a 0.616 0.496 0.915

n/a 0.511 0.074 0.435

n/a 0.357 0.06 0.1

n/a 0.736 0.301 0.214

n/a 0.513 0.1 0.752

0.301

0.826

0.496

0.043

0.441

0.323

0.914

0.383

0.057

0.554

Source: own estimates. P-values (based on the F-distribution) not higher than 10% are in bold. Ci is a country fixed effect while FH_PR is the Freedom House's Political Rights index.

with Competition Policy could be confirmed. Moreover, both results did not occur to be robust to changes in the number of observations and cross-sections after the inclusion of other labour market institutions indicators into the model (Table 4c, columns 25e28). These additional equations were estimated in order to verify whether the tax wedge was correlated with the expenditure on ALMP and the level of unemployment benefits. In case of the ALMP no significant relationship was observed, but for the unemployment benefits a significant positive correlation was noticed. Expenditure on ALMP occurred to be determined only by the Small scale privatisation (Table 4c, columns 29e34). The larger the scale of that privatisation was, the higher the expenditure on ALMP was noticed. Although the level of the average unemployment benefits was also determined by the Small scale privatisation, it occurred not to be significantly correlated with the expenditure on ALMP. Quite surprisingly, such indicators as Political Rights, Trade Freedom, Property Rights or Government expenditure did not occur to significantly influence the shape of labour market institutions in post-socialist economies. Such a result suggests that changes in the labour market institutional framework during the transition period could be predominantly determined by the market reforms and only coincided with other elements of the transition process. Is should be also mentioned that in almost all the estimates the conducted F-test indicated to reject the null hypothesis that all group have a common intercept. This means that independent variables included into the estimates did not allow to explain the cross-country heterogeneity in labour market institutions sufficiently enough to remove the countries' fixed effects dummies from the estimates. In fact, the low values of the within R2 statistic indicate that a significant part of that cross-country diversity has remained unexplained. Thus, although we were able to identify some statistically significant relationships between indicators of the market reforms and the labour market institutions, in case of labour market institutions in the post-socialist economies the countries' specificity still seems to prevail. 5. Conclusions The aim of this study was to identify common determinants of the shape of labour market institutions in the group of 25 post-socialist countries during the transition period. It was found that the process of privatisation was connected with the increase in the level of unemployment benefits and expenditure on ALMP and also with the decrease in the union density rate. Some of the results indicate that the post-socialist countries which strengthened their domestic competition also decided to increase the EPL and the tax wedge. However, the results do not allow to confirm the expected relationship between the shape of political and labour market institutions e political rights index occurred to be insignificant in all the estimates. Generally, it can be stated that although heterogeneity in the scale of conducted market reforms can help to explain the heterogeneity in labour market institutions in post-socialist countries, the heterogeneity in political and trade freedom, in the level of the property rights enforcement and in the level of the government expenditure cannot. Therefore, it can be concluded that the conducted analyses give only a partial support to the stated research hypothesis, because the labour market reforms in post-socialist countries occurred to be connected with changes in the economic environment only and were independent from the level of freedom granted to labour market participants to express their political preferences and to enforce their privileges in courts. Thus, the labour market reforms in post-socialist countries can be perceived as a part of the economic transition process, but not as a part of the changes in political institutions. The conducted research also allows to formulate recommendations for further analyses. In this study the labour market outcomes were not analysed as possible determinants of the labour market institutions, since the small amount of observations would not allow to deal with the endogeneity problem. However, it can be assumed that they could help to explain the labour market institutions heterogeneity, since labour market institutions could have been changed in response to the actual situation on the labour market. Thus, it would be recommended to identify determinants of the shape of labour market institutions in post-socialist countries also among the labour market outcomes, if enough data had been gathered. Gathering longer time series would also allow to verify the obtained results with the use of dynamic panel models and the Granger causality tests, which is the second recommendation for further research. It would be also valuable to verify whether results Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

15

obtained in this research are characteristic only for the post-socialist countries or whether they are also correct for other developing regions. Acknowledgements This paper was written with the support of the National Science Centre in Poland, grant no. 2011/01/B/HS4/00802, titled: ‘State in relation to economic freedom. Theory and practice of transition’. An earlier version of this paper was presented at the 2012 EAEPE Symposium held in Rome and at the 24th Annual EAEPE Conference held in Cracow in 2012. Special gratitude is owed to anonymous reviewers who helped to improve the article substantially. The author would also like to give thanks to Katarzyna Szarzec and Larysa Tamilina for all the inspiring comments and suggestions; all remaining errors are the author's responsibility. References Acemoglu, D., Johnson, S., Robinson, J.A., 2005. Institutions as a fundamental cause of long-run growth. In: Aghion, P., Durlauf, S.N. (Eds.), Handbook of Economic Growth, vol. 1A. Elsevier, pp. 385e472. Agell, J., 2002. On the determinants of labour market institutions: rent seeking vs. social insurance. Ger. Econ. Rev. 3 (2), 107e135. Algan, Y., Cahuc, P., 2009. Civic virtue and labor market institutions. Am. Econ. J. Macroecon. 1 (1), 111e145. ~ rk, A., 2007. Labour market institutions and productivity in the New EU Member States. Prax. Work. Pap. 27, 1e30. Anspal, S., Vo Arellano, M., 2003. Panel Data Econometrics. Oxford University Press, Oxford. Arpaia, A., Mourre, G., 2005. Labour market institutions and labour market performance: a survey of the literature. Eur. Comm. Econ. Pap. 238, 1e39. Baltagi, B.H., 2005. Econometric Analysis of Panel Data, third ed. John Wiley & Sons, Chichester. Bertola, G., Koeniger, W., 2004. Consumption smoothing and the structure of labor and credit markets. IZA Discuss. Pap. 1052, 1e28. Blanchard, O., 2006. European unemployment: the evolution of facts and ideas. Econ. Policy 21 (45), 5e59. Blanchard, O., Wolfers, J., 2000. The role of shocks and institutions in the rise of European unemployment: the aggregate evidence. Econ. J. 110 (462), 1e33. Boeri, T., 2005. Reforming Labour and Product Markets: Some Lessons from Two Decades of Experiments in Europe. IMF Working Paper 05/97, pp. 1e34. Boeri, T., Conde-Ruiz, J.I., Galasso, V., 2003. IZA Discussion Paper. Protecting against Labour Market Risk: Employment Protection or Unemployment Benefits?, vol. 834, pp. 1e39. Boeri, T., Garibaldi, P., 2005. Are labour markets in the New Member States sufficiently flexible for the EMU? J. Bank. Financ. 30 (5), 1393e1407. Boeri, T., Van Ours, J., 2008. The Economics of Imperfect Labor Markets. Princeton University Press, Princeton. Cazes, S., 2002. Do labour market institutions matter in transition economies? An analysis of labour market flexibility in the late nineties. Int. Inst. Labour Stud. Discuss. Pap. 140/2002, 1e29. Cazes, S., Nesporova, A., 2003. Labour Markets in Transition: Balancing Flexibility and Security in Central and Eastern Europe. ILO, Geneva. Checchi, D., Lucifora, C., 2002. Unions and labour market institutions in Europe. Econ. Policy 17 (35), 361e408. Eamets, R., Masso, J., 2005. The paradox of the Baltic States: labour market flexibility but protected workers? Eur. J. Ind. Relat. 11 (1), 71e90. EBRD, 2012a. Forecasts, Macro Data, Transition Indicators. http://www.ebrd.com/pages/research/economics/data/macro.shtml#macro (accessed 05.11.12). EBRD, 2012b. Transition Indicators Methodology. http://www.ebrd.com/pages/research/economics/data/macro/ti_methodology.shtml (accessed 07.11.12). Eichhorst, W., Feil, M., Braun, C., 2008. What have we learned? Assessing labor market institutions and indicators. IZA Discuss. Pap. 3470, 1e38. Eurostat, 2014. Eurostat Metadata. Minimum Wages. http://ec.europa.eu/eurostat/cache/metadata/en/earn_minw_esms.htm (accessed 09.02.14). Fakin, B., de Crombrugghe, A., 1997. Fiscal adjustments in transition economies: social transfers and the efficiency of public spending. A comparison with OECD countries. In: World Bank Policy Research Working Paper, vol. 1803, pp. 1e32. Fialova, K., Schneider, O., 2009. Labor market institutions and their effect on labor market performance in the new EU member countries. East. Eur. Econ. 47 (3), 57e83. Freedom House, 2012. Freedom in the World 2012. Methodology. http://www.freedomhouse.org/report/freedom-world-2012/methodology (accessed 12. 07.12). Freysson, L., Wahrig, L., 2013. General government expenditure in 2011 e focus on the functions ‘social protection’ and ‘health’. Eurostat Stat. focus 9, 1e8. Gligorov, V., Iara, A., Landesmann, M., Stehrer, R., Vidovic, H., 2008. Western Balkan countries: adjustment capacity to external shocks, with a focus on labour markets. Vienna Inst. Int. Econ. Stud. Res. Rep. 352, 1e131. Grzymala-Busse, A., Innes, A., 2003. Great expectations: the EU and domestic political competition in East Central Europe. East Eur. Polit. Soc. 17 (1), 64e73. Havrylyshyn, O., 2001. Recovery and growth in transition: a decade of evidence. IMF Staff Pap. 48, 53e87. Heritage Foundation, 2012. Index of Economic Freedom. Methodology. http://www.heritage.org/index/book/methodology (accessed 16.06.12). Kajzer, A., 2007. Development of the Slovenian labour market in 1996e2006 and the main challenges of labour market policy. Post-Communist Econ. 19 (4), 471e482. Kornai, J., 2006. The great transformation of central eastern Europe. Econ. Transit. 14 (2), 207e244. Kwiatkowski, E., Socha, M.W., Sztanderska, U., 2001. Labour market flexibility and employment security: Poland. ILO Employ. Pap. 2001/28 1e77. Layard, R., Nickell, S., Jackman, R., 2005. Unemployment. Macroeconomic Performance and the Labour Market. Oxford University Press, New York. Lehmann, H., Muravyev, A., 2012. Labor market institutions and labor market performance: what can we learn from transition countries? Econ. Transit. 20 (2), 235e269. Micevska, M., 2008. The labour market in Macedonia: a labour demand analysis. Labour 22 (2), 2345e2368. Milanovic, B., 1995. Poverty, inequality, and social policy in transition economies. World Bank Policy Res. Work. Pap. 1530, 1e56. Muravyev, A., 2010. Evolution of employment protection legislation in the USSR, CIS and Baltic states, 1985e2009. IZA Discuss. Pap. 5365, 1e57. North, D.C., 1997. The contribution of the new institutional economics to an understanding of the transition problem. UNU WIDER Annu. Lect. 1, 1e18. OECD, 2004. OECD Employment Outlook 2004. OECD, Paris. OECD, 2014. OECD Indicators of Employment Protection. http://www.oecd.org/employment/emp/oecdindicatorsofemploymentprotection.htm (accessed 09.02.14). Pia˛ tek, D., Szarzec, K., Pilc, M., 2013. Economic freedom, democracy and economic growth: a causal investigation in transition countries. Post-Communist Econ. 25 (3), 267e288. Rys, V., 2001. Social protection in central and Eastern Europe ten years after. Int. Soc. Secur. Rev. 54 (2e3), 3e6. Saint-Paul, G., 1996. Exploring the political economy of labour market institutions. Econ. Policy 11 (23), 263e315. Saint-Paul, G., 2002. The political economy of employment protection. J. Political Econ. 110 (3), 672e704. Schweickert, R., Melnykovska, I., Belke, A., Bordon, I., 2011. Prospective NATO or EU membership and institutional change in transition countries. Econ. Transit. 19 (4), 667e692. Staehr, K., 2011. Democratic and market-economic reforms in the postcommunist countries: the impact of enlargement of the European Union. East. Eur. Econ. 49 (5), 5e28. Tonin, M., 2009. Employment protection legislation in central and east European countries. South-East Eur. Rev. 4, 477e491.

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008

16

M. Pilc / Communist and Post-Communist Studies xxx (2015) 1e16

Winiecki, J., 2008. Employment and unemployment in transition: the legacy of the communist past. Post-Communist Econ. 20, 377e390. Wooldridge, J.M., 2001. Econometric Analysis of Cross Section and Panel Data. The MIT Press, Cambridge MA and London. World Bank, 2014. Doing Business. Labor Market Regulations. http://www.doingbusiness.org/data/exploretopics/labor-market-regulation (accessed 09.02. 14).

Please cite this article in press as: Pilc, M., Determinants of the labour market institutions in post-socialist economies, Communist and Post-Communist Studies (2015), http://dx.doi.org/10.1016/j.postcomstud.2015.06.008