Low-wage mobility in Central Europe

Abstract

The article provides analyses of the mobility and resilience to mobility among low-wage earners in four Central European (CE) countries. It examines transitions into higher-paid jobs, unemployment/inactivity, and the stability of low-wage status. In addition to standard transition matrices and summary mobility indices, it employs multinomial logit models with the aim of identifying individual determinants of low-wage earners’ prospects. The findings show that the CE countries do not represent a homogeneous group in terms of presence of low wages when the period of 2010–2016 is considered. In regard to future prospects, low-wage employees in the countries with higher incidence of low pay are more likely to reproduce their status, as compared with countries with lower incidence. Upward mobility is more likely among younger, high-educated employees and among those who work in “better” occupations.

Keywords

Central Europe low-wage employment low-wage persistence panel data relative mobility

Introduction

Low-wage employment is a double-face phenomenon. On one hand, it raises concerns about an ability to maintain decent living standards, the risk of becoming stuck in low-paid jobs for a long time, or even risk of a “low-pay no-pay” cycle (Clark and Kanellopoulos, 2013; Lucifora and Salverda, 2011; Uhlendorf, 2006). On the other hand, it is seen as a way of avoiding the unemployment trap and providing labor market re-integration for the (long-term) unemployed, as well as a transitory state that can serve as a stepping stone for transition to better-paid jobs (Cai et al., 2018; Knabe and Plum, 2013).

The article deals with low-wage employment in terms of the prospects of low-wage earners in the labor market, with the aim being to examine patterns of mobility and low-wage persistence in the four Central European countries (hereinafter, CE countries): Czech Republic, Hungary, Poland, and Slovakia. Based on longitudinal data from the European Union Survey on Income and Living Conditions (EU-SILC), which allows for tracking of individual-level changes over 4-year periods, the study analyses the chances of low-wage earners to experience the following three situations: transition into higher-paid jobs, transition into unemployment/non-employment, and stability of low-wage status. The study contributes to the large body of literature on low-wage jobs in several respects. It focuses on an EU region characterized by a wage level that is lower than in the western EU countries, and as such represents a source of international competitiveness. The four countries share similar specific development trajectories, including transition to market economy or the changing role of foreign direct investments, which shaped the general economic context as well as the nature of labor markets and structure of earnings, including the extent and nature of low-wage employment. These countries, while well covered by previous research on the labor market situation of various population groups and labor market dynamics (Flek and Mysíková, 2015; Flek et al., 2018), have only attracted limited attention when it comes to low-wage earners’ future prospects. The analyses, which include the four CE countries (European Commission, 2016, 2020), provided rather limited evidence, primarily related to the supranational level and/or which covered only very short periods of time. Furthermore, it analyses the period following the Great Recession (2008–2011),¹ studying low pay dynamics in the context of economic recovery and post-crisis growth. Finally, by identifying individual factors that affect transitions from low-wage employment or its continuity, it extends the evidence on determinants of chances of various population groups in labor markets in Central Europe. Thus, the study wants to fill the existing gaps concerning the CE countries, with the aim of exploring whether and to what extent low-wage employment provides chances for improving future prospects in the labor market or for their deterioration, as well as for how this varies across different categories of workers.

The article seeks to find answers to the following research questions. What are the most likely prospects for low-wage earners in the CE region? How do they differ among various categories of workers? To what extent and for whom do low-wage jobs operate as a vehicle for upward wage mobility? And what are the main dividing lines among the CE countries in terms of the low-wage earners’ prospects?

The article is structured as follows: the first part provides a review of the recent theoretical and empirical arguments. The second part describes the methodology of the empirical analysis. Next, the results of the empirical analysis are presented and discussed. The final part summarizes the findings and provides conclusions.

Review of theoretical arguments and empirical evidence

The ambiguous nature of low-wage employment is reflected by the theory of signaling and human capital accumulation theory. The theory of signaling is based on the assumption that the past labor market status represents an important signal of productivity for potential employers (McCormick, 1990). As one of the most quoted arguments in the literature reads—while unemployment is a bad signal, being in a low-quality job may well be a worse one (Layard et al., 2005). This applies, in particular, to high-qualified workers, as it indicates that there is a gap between their real productivity and the productivity suggested by their level of education/qualification.² Thus, according to the theory of signaling, high-skilled workers tend to avoid low-paid jobs, assuming that low-paid jobs do not necessarily work as a springboard to better-paid employment. For low-qualified workers, the signaling effects are ambiguous. On one hand, taking up a low-paid job may signal a willingness to work (Plum et al., 2021) and increase probability for upward wage mobility. On the other hand, it may serve as an indicator of low productivity for employers who scan the labor market history of potential employees, hampering access to better-paid jobs and making low-pay a persistent labor market status.

Opportunities for accumulating human capital also play a role in shaping the future prospects of low-wage earners. Low-paid jobs, which allow for an accumulation of human capital that corresponds to the demand in the labor market, may lead to higher-paid jobs in the future (Cai et al., 2018). If development of human capital is hampered or even tends to deteriorate in low-paid jobs, the future prospects look much worse. In this case, persistence of low pay is reinforced, and the probability of a low-pay, no-pay cycle increases.

Arguments about the role low-paid jobs play for the workers’ future prospects, formulated by both approaches, have been the subject of repeated empirical examination. Low-wage employment acting as a springboard has been identified by several studies. For Germany, Knabe and Plum (2013) showed that unemployed persons can improve their chances to get high-paid jobs by taking up a low-paid one. This is notably the case for persons without a college degree, persons who experienced several unemployment spells in the past, and persons who took a low-paid job with relatively high social status.³ In their most recent study, Plum et al. (2021) examined wage transitions in New Zealand using administrative data, confirming the strong stepping-stone effect. Workers continuously on low pay for 12 months have a substantially higher probability of moving to high-paid jobs in the next month compared with continuously non-employment persons. Cai et al. (2018) came to a similar conclusion for the United Kingdom, revealing that low-wage earners are more likely to experience upward wage mobility than those who are unemployed or out of the labor force. However, the study also provided evidence for persistence in low pay: low-wage earners are more likely to remain in low-wage employment in the future, as compared with persons in better-paid jobs as well as jobless persons. Persistence in low pay was also supported by other research, as Stewart and Swaffield (1999) found a considerable persistence in low pay in the United Kingdom and Clark and Kanellopoulos (2013) confirmed its presence in 12 EU old member states (after controlling for individual differences among workers).

In regard to the low-pay, no-pay cycle, there is rather mixed evidence. On one hand, Stewart (2007) and Cappellari and Jenkins (2008) for the United Kingdom, and Fok et al. (2015) for Australia, confirmed the thesis that low pay increases the probability of becoming unemployed in the future. On the other hand, Cai et al. (2018) did not find evidence for the low-pay, no-pay cycle among British workers. In addition, Mosthaf (2014) in his influential study showed that low-paid workers with a low level of qualification face a significantly lower risk of future non-employment than low-qualified persons who are out of the labor market. For middle- and high-qualified workers, low-wage employment produced the same risk of future non-employment as non-employment itself. According to the author, there are two plausible interpretations of the obtained results. First, they confirm the role of signaling effects, with larger negative effects for workers with higher qualifications. Second, they suggest that human capital accumulation plays an important role for high-qualified workers.

Signaling effects and human capital accumulation approaches put emphasis on workers’ job search behavior, their characteristics, capabilities, as well as employers’ strategies in matching supply and demand in the labor market. However, low-wage employment and its dynamics are also the results of structural characteristics in the labor market and more general socio-political forces. This is reflected in dual labor market theory and the theory of labor market segmentation. The origins of dual labor market theory date back to the early 1970s, when it emerged as a response to neoclassical theories regarding the labor market, with focus on low wages and unemployment in labor sub-markets. Rather than the characteristics of workers, it drew attention to characteristics of jobs and institutional processes that create differences between the categories of jobs (Peck, 1989), attributing a theoretically central role to the structural asymmetry in the labor market (Kreckel, 1980).

A well-known distinction was drawn between primary and secondary segments of the labor market. The primary segment consists of “good jobs” characterized by good remuneration, job security, and career progress. These jobs are created within the technologically advanced production sectors. The second segment contains precarious jobs, which feature low wages, poor working conditions, and few career advancements. Mobility between the primary and secondary labor market is assumed to be low, with a tendency to institutionalization of the existing differences (Emmenegger et al., 2012).

Dualization and segmentation of labor markets in developed countries has been confirmed in numerous studies (Dolado, 2017; Emmenegger et al., 2012; Rubery and Piasna, 2017), including those conducted in the CE countries (Trif et al., 2016). The research confirmed that non-standard work in the secondary labor market sector represents an increased probability of wage penalties and negative effects on career mobility (Rubery and Piasna, 2017), as well as that EU countries differ in terms of labor market segmentation (Emmenegger et al., 2012) and in terms of mobility between the primary and secondary segments (Eurofound, 2016).

In the CE countries, dual labor markets started to emerge during the economic transformation in the 1990s, which was marked by the collapse of the state socialist heavy industry and overall de-industrialisation, with serious social consequences, including the spread of unemployment, a dramatic decline in employment opportunities for unskilled workers, and a drop in the real wages (Bohle and Greskovits, 2007). This was accompanied by increasing returns to education and rising wage disparities (Scharle, 2012). Primary labor markets attracted a greater proportion of men, persons residing in capitals, persons in occupations that demanded more skills, as well as persons working for private companies (Pailhé, 2003). However, precarious forms of employment, defined as uncertain, unpredictable, and risky from the point of worker, did not come to the fore with the very beginning of the economic transformation. This “delay” was caused by the fact that standard forms of employment remained the norm for some time as a result of the socialist legacy (Mrozowicki et al., 2013).

Since then, the four CE countries followed similar paths in terms of labor market segmentation, but at uneven paces and with some different specific elements. Since the 2000s, various policy reforms facilitated use of external flexibility by employers and non-standard contracts. With the onset of the global economic crisis, all four CE countries also created conditions for an increased use of internal flexibility for employers on standard contracts, undermining individual and collective rights (Trif et al., 2016).

In Poland, a significant expansion of fixed-term contracts has taken place since 2000, which was caused by labor market flexibilization and low labor costs policies, placing Poland among the “leaders” of the EU in this respect (Maciejewska et al., 2016). Temporary employment tended to represent an involuntary destiny, not a free choice, and was accompanied by limited chances to move to the primary sector and a higher risk of poverty. According to Mrozowicki (2016), a process of “normalisation” of precarious employment took place in Poland, as it became an increasingly expected, taken-for-granted destiny by some categories of workers, as well as employers and policy-makers. In Hungary, increase in employment levels after the global economic crisis was accompanied by deterioration of employment standards: employer-friendly flexibilization and labor market deregulation, which took place particularly after 2010, led to expansion of precarious employment, including fixed-term contracts, part-time jobs, temporary agency work, as well as public employment programs targeted at the persons with weak attachment to labor market (Meszmann, 2016).

In Czechia, changes in the labor market regulations intensified the precariousness of employment in the period after the global economic crisis too, which was accompanied by increasing use of fixed-term contracts, temporary agency work and agreements on work performed outside a regular employment relationship. Slovakia introduced most non-standard forms of employment in the 2000s as part of liberal welfare state reforms and efforts to increase the attractiveness of foreign direct investments. Later, the insecure nature of some typical precarious forms of work was reduced, in particular that of temporary agency work and bogus self-employment (Fabo and Sedláková, 2017; Kahancová, 2016; Martišková, 2021).

Low-wage employment and its dynamics must be also seen in a more general economic context. The CE countries belong to the middle-income economies with wages below the level in western EU countries. The economic model characterized by low wages as a source of competitiveness became a characteristic feature of the CE region. This was also a result of the fact that re-industrialisation, which was driven by foreign direct investments (Barta et al., 2008) and relied on low labor costs as a comparative advantage, did not succeed in integration of the CE countries into high value-added segments of global value chains (Stojčič and Aralica, 2018). This model now represents a serious problem, as the CE countries have a subordinated role in the international division of labor, which is based on low value-added production activities. To describe this situation, Myant (2018) uses the term “dependent” economy, derived from the term “dependent development” that was applied to the Latin America economies. For the CE countries, the term dependent economy refers to the fact that their economic development depends to a large extent on inward investments by multinational companies, while domestic capital and other factors play a less important role.⁴

Strategy of empirical analysis, data, and methodology

Low wage is defined in line with common practice in comparative research and statistics as a wage below the two-thirds of median hourly wage in a country (European Commission, 2016; McKnight et al., 2016). It is a relative definition that takes into account an overall wage level in a given country and in a given year. The category of low-wage earners is constructed based on wage distribution in each year.

Focusing on transitions from low pay, three potential outcomes are taken into account: transition into higher-paid jobs, transition into unemployment/inactivity, and the stability of low-wage status. The strategy of the empirical analysis comprises four steps. First, we describe the situation in the CE countries in terms of low wages. Second, we examine individual characteristics that may affect any incidence of low pay, applying panel logistic regression. Third, we describe transitions from low pay, based on transition matrices. Finally, to analyze the future prospects of low-wage workers, we estimate four multinomial logit models—one model for each CE country. We opt for individual-country models instead of one pooled data model, as our aim is to find out whether and in which manner individual characteristics affect transitions from low-wage status in each country and to compare the findings across them.

Quantitative analyses of transitions from low-wage status typically rely on longitudinal micro-data (Cai et al., 2018; Plum et al., 2021; Stewart, 2007; Uhlendorf, 2006) because repeated measurements at an individual level allow for studying changes of the wage status and their persistence. We use data from the European Union Survey on Income and Living Conditions (EU SILC), which has a longitudinal component that allows for following respondents over a 4-year period.⁵ To examine movements over a longer time span and to increase the sample size, we pooled the data and created a data set consisting of EU SILC 2011–2016. As income reference period is the previous calendar year, this data set covers the period 2010–2015, which followed the economic and financial crisis that started in 2007/2008.

The countries’ samples are restricted to employed persons aged 20–64 years, for which information on key earnings variables are available for at least three consecutive years. The overall sample size ranges from 41,651 persons in Slovakia, 47,260 persons in Czech Republic, to 52,407 persons in Hungary and 73,168 persons in Poland.

To analyze three potential outcomes of transitions from low pay, we use data on wages and economic status that are provided on a yearly basis. Economic status is conceptualized as a self-declared current economic status that captures a respondent’s own perception of main activity, determined on the basis of the most time spent.⁶ When it comes to wages, gross hourly wage is used to ensure cross-country comparability as well as comparability of the wages of individuals with a varying number of hours worked. Gross hourly wage was calculated in two steps. First, a total annual wage was calculated as a sum of four variables: PY010G (employee cash or near cash income), +PY020G (non-cash employee income), +PY030G (employer’s social insurance contribution), and +PY050G (cash benefits or losses from self-employment). In the second step, an hourly wage was calculated by dividing the annual wage by the product of numbers of hours usually worked per week in a main job⁷ (variable PL060), the number of weeks per month (4), and number of worked months (PL211A—PL211L). Thus, our results are based on hourly wages. All observations with negative or missing values were eliminated from the analysis.

In logistic regression, which examines individual determinants of low pay, only working persons aged 20–64 years with non-zero wages are taken into account. In this case, the sample sizes are smaller, as compared with the samples used in other analytical steps. Outcome variable is represented the dummy variable “incidence of low wage,” which can take a value of 1 if a person earns a low wage or 0 if she or he has a higher wage. The list of explanatory variables include a set of socio-demographic and work-related characteristics that are considered as potentially important factors of becoming low-wage earners (European Commission, 2020; Fok et al., 2015; Uhlendorf, 2006).

Markov transition matrices were calculated for three potential outcomes: transitions from low wage to higher wage, transitions from low wage to unemployment, and reproduction of low-wage position. The Markov transition matrix contains the conditional probability that a person who starts in one position in time t, will be found there in a different position in time t + 1. The formal expression can be written as

P R [v_{t + 1} = v | v_{t}, v_{t - 1}, \dots, v_{0}] = P R [v_{t + 1} = v | v_{t}]

The positions were constructed for each year: t and t + 1 in the observed period. Two wage categories (low wages, higher wages) were constructed based on given wage distributions in each year. As we consider movements between three labor market states, the Markov transition matrix contains three rows and three columns where the diagonal elements represent unchanged labor market states.

In multinomial models, the list of explanatory variables includes individual socio-demographic (age, gender, educational level, increase in education level), job-related characteristics (working experience, occupation captured by the International Standard Classification of Occupations (ISCO) categories, type of employment, wage position in the initial year, change of job), household characteristics (presence of children, work intensity), as well as years-related dummy variables. More precise operationalisation will be provided in the section devoted to the discussion of the results. Following the study of Bachmann et al. (2016), we interpreted the relationship between explanatory variable(s) and the dependent variable in terms of predicted probabilities and marginal effects. Marginal effects, with their standard errors, are reported in the tables in the appendix.

Low wages in the Central Europe

Low wages are shaped by an overall wage level and earnings inequality. Prior to the economic and financial crisis, a dynamic growth of real wages accelerated a convergence between Central and Western Europe. The crisis brought about a slowdown of wage growth, followed by the fall in real wages⁸ and subsequent subdued increases up to 2016 (Galgóczi, 2017). However, as we already mentioned, the CE countries belong to the middle-income economies that remain caught in the trap of a competitiveness model based on low wages. In terms of earnings inequality, the four CE countries belong to the EU countries with above-average earnings differentials. At the onset of the economic and financial crisis in 2008, the ratio of the highest and lowest decile (D9/D1 ratio) reached 4.11 in Hungary and Poland and 3.50 in the Czech Republic and Slovakia (the average value for the EU was 3.25).⁹ During the crisis and immediate subsequent years, the inequality rose in all CE countries, with the exception of Poland. In the period 2010–2015, which is analyzed in this study, the growth of earnings differentials has deaccelerated (Czech Republic) or halted (Hungary, Poland, Slovakia). In 2015, the highest D9/D1 ratios could be still found in Hungary (3.72) and Poland (4.05).

As low pay threshold, which separates low-wage earners from those with higher earnings, is defined in relative terms in this study, its development is expected to reflect changes in the overall wage distribution. As Table 1 shows, the thresholds evolved differently in the CE countries during the observed period. However, in general, their development reflected the evolution of the average wages in the region after the economic and financial crisis, where acceleration in wage dynamics—related to the improvement in the labor market conditions—only started in 2014/2015 (Astrov et al., 2019). The highest values in each year of the period under consideration were reported in the Czech Republic. Poland, a CE country with the highest earnings inequality, had the third lowest value of the low-wage threshold. The lowest values could be found in Hungary.

Table 1.

Proportion of low-wage earners (%) and low-wage thresholds (in EUR per hour) in the pooled EU SILC data set.

	Czech Republic	Hungary	Poland	Slovakia
% of low-wage earners
2010	19	21	29	17
2011	19	18	29	17
2012	20	14	28	17
2013	21	13	28	18
2014	21	17	26	12
2015	20	17	25	16
Low-wage threshold
2010	4.13	2.56	2.65	3.21
2011	4.32	2.64	2.72	3.32
2012	4.26	2.56	2.76	3.43
2013	4.20	2.61	2.86	3.83
2014	4.17	2.51	2.83	3.95
2015	4.37	2.59	2.95	4.00

EU SILC: European Union Survey on Income and Living Conditions.

Source: Authors’ calculation based on pooled longitudinal data EU SILC 2011–2016.

The relative size of the category of low-wage earners varied across the CE countries (Table 1). The highest proportion of low-wage earners among employees was reported by Poland, followed by the Czech Republic. In Poland, low-wage earners represented, on average, 28 percent of the employees during the observed period. In the years 2010 and 2011, immediately affected by the economic and financial crisis, the proportion reached almost 30 percent. A high share of workers with temporary contracts and poor prospects of moving to permanent jobs (OECD, 2016) contributed to the specific position of Poland. In the Czech Republic, one fifth of the employees worked on average for low pay. In Hungary and Slovakia, their relative size fluctuated around quite lower percentage values (16% and 17% on average, respectively).¹⁰

How can low-paid workers be characterized? We ran a panel binary logit regression with random effects where the outcome variable had two categories: “being low-paid” and “being higher-paid.” The latter one served as a reference category. We then ran four regression models, one for each country. The list of explanatory variables include socio-demographic variables (gender, age, skills level), job-related characteristics (type of contract, occupation class, change of job), as well as other individual characteristics (health status, improvement in education) and the years’ dummy variables. All were inserted into the model as categorical variables, that is, as relevant sets of dummy variables. When it comes to estimation of the models, we report average marginal effects (Table 2), which express a change in probability of the outcome (being low-paid) as a function of the change in independent variable(s).

Table 2.

Average marginal effects from binary logit regression on being low-wage earner in the period 2010–2015.

	Czech Republic	Hungary	Poland	Slovakia
Male	−0.105***	−0.033***	−0.152***	−0.070***
25–34 years	−0.056***	−0.047***	−0.068***	−0.063***
35–44 years	−0.065***	−0.067***	−0.120***	−0.062***
45–54 years	−0.058***	−0.079***	−0.143***	−0.071***
55–64 years	−0.058***	−0,077***	−0.130***	−0.075***
Low-educated	0.117***	0.101***	0.134***	0.082***
High-educated	−0.118***	−0.092***	−0.085***	−0.069***
Job change	0.025***	0.077***	0.031***	0.009***
Improved education	0.019***	0.032***	0.035***	0.009***
Poor health	0.046***	0.042***	0.029***	0.039***
Part-time job	−0.007***	0.018***	0.003	0.031***
ISCO 4–5	0.047***	0.074***	0.198***	0.092***
ISCO 6–8	0.051***	0.089***	0.324***	0.106***
ISCO 9	0.112***	0.185***	0.268***	0.178***
Year 2012	0.013***	−0.036	0.006	0.017***
Year 2013	−0.018***	−0.043***	0.008	0.019***
Year 2014	−0.023***	−0.090***	−0.019***	0.020***
Year 2015	0.002	−0.016***	0.001	−0.043***
Number of observations	47,255	52,412	73,165	41,650

ISCO: International Standard Classification of Occupations.

Source: Authors’ calculation based on pooled longitudinal data EU SILC 2011–2016.

Note: Reference categories include females, persons aged 18–24 years, middle-skilled persons, persons who did not change the job, persons who did not participate in trainings, persons in good health, persons in full-time job, and persons in jobs in categories ISCO 1–3.

***p < .001.

In all four countries, females are more likely to become low-wage earners. Young persons under the age of 25 years represent the most risky age category (it serves as a reference category in the models). Older employees show markedly lower probability of becoming low-paid workers. In particular, it is Poland where being a member of the higher-age groups protects strongly against the low-wage status. A higher exposure of young people to low wages stems from the fact that they are more likely to enter labor market through non-standard forms of employment, including temporary contracts or low pay (Eurofound, 2016). Level of education plays an expected role. Employees with low skills are more likely to become low-wage earners. In a similar vein, occupations that require lower level skills are linked to higher probability of low wage. In general, the category of occupation represents one of the strongest determinants of low pay. All these findings are in line with the previous patterns identified in EU countries (Ozdemir and Ward, 2015).

An impact of another characteristic of job—whether it is part-time or full-time—does not show a clear pattern. While it only slightly increases the probability of low pay in Slovakia and Hungary, it has no significant effect in Poland. One of the explanations of a lacking clear pattern may lie in the fact that there is no link between low pay and the number of hours worked in Central Europe. Employees on low earnings do not work less as compared to higher-paid employees. Poland is a front runner in this respect: hours worked by low-wage employees represented, on average (across all observed years), 107 percent of the hours worked by other workers. In the Czech Republic and Slovakia, the ratio was 104 percent and 105 percent, respectively. The smallest differences were found in Hungary (101%).

There are two variables in the models that capture the change in individuals’ characteristics: job change and change in education level. Both operate in the same direction: the change is linked with higher probability of low pay. First, employees who have changed their job during the last year are more likely to find themselves in a low-paid job compared with employees who have kept their jobs. While supposing the negative consequences of job fluctuation, this relationship needs to be interpreted with caution. The reason is that job changes are more frequent among low-paid workers in all four countries. Second, persons who participated in education or training during the last year and increased their educational level tend to show higher propensity toward low wages. This group includes, in particular, graduates and young people aged 20–24 years who try to improve their initial position in the labor market.

Our models also suggest that increased probability of low pay is associated with poor health. Although there is a complex relationship between health and low-wage employment, which also includes the negative impact of low-wage jobs on health, there is evidence that poor health increases the risk of low pay by limiting the type or amount of work that an individual can do (McKnight et al., 2016).

We also performed a robustness test based on different specification of the number of low-wage episodes. While originally we focused on all low-wage earners who met the abovementioned conditions, in this exercise we ran binary logit models for the employees who experienced low wages for a period of 1 year, 2 years, 3 years, and 4 years. The obtained results correspond with the results presented in this study.

Low-wage mobility and the contributing factors

The mobility of low-wage earners is grasped by calculating Markov transition matrices, which capture conditional transition probabilities from the three original states (low wage, higher wage, unemployment/non-employment) at the time point t to the three destinations at the time point t + 1 and which are presented in the appendix (Table A1). Here we focus on transition rates for the movements from low wage.

As Figure 1 shows, the future prospects of low-wage earners varies across the CE countries. The countries differ in terms of the extent to which the low-wage status is reproduced, resulting in the two distinct clusters. In the Czech Republic and Poland, the probability of remaining in low-wage positions represent 71 percent. Slovakia and Hungary form a distinct cluster, characterized by a lower propensity of reproduction of the low-wage status. In Slovakia, the probability of being stuck in the low-wage trap represents 56 percent. In Hungary, it was even less than 50 percent.

Figure 1.

Transition rates for movements from low wage in the period 2010-2015 (%).

Lower propensity to reproduce a low-wage status in Hungary and Slovakia goes hand in hand with lower incidence of low pay. The question is whether this combination means that low-paid workers more easily move to better wage positions or whether they tend to exit the labor market, as compared with other countries. Transition probabilities support the first hypothesis. There are marked differences between the two clusters in terms of probability that low-paid workers improve their wage position. While it represents 42 percent in Hungary and 35 percent in Slovakia, it reaches values around 20 percent in the Czech Republic and Poland. However, when appraising the prospects of low-paid workers in Hungary, one has to also take into account the fact that their probability for exiting the labor market is twice as high as in other countries.

The finding that Slovakia and Hungary provide better prospects for low-paid earners is supported by research on transitions between wage deciles (Gerbery and Miklošovič, 2021) that focused on changes in the wage positions in wage distribution during the years 2008–2014 in Central Europe. It showed that Hungary and Slovakia reported the largest extent of upward wage mobility among employees in the first and second wage decile. Furthermore, this research also provided evidence that employees in Hungary and Slovakia showed lower probability for staying in the same wage position, supporting our finding that low wages in these two countries tend to be less persistent than in the Czech Republic and Poland.

The dynamics of temporary jobs were among other characteristics of Hungary and Slovakia that also shaped their specific position in terms of the future prospects of low-paid workers. Compared with the Czech Republic and Poland, the countries had higher proportions of temporary jobs with a shorter duration and higher proportions of employees in temporary jobs who moved to permanent jobs in the given period (Eurofound, 2016). Both characteristics make upward wage mobility more likely and suggest that, in Hungary and Slovakia, low-paid temporary jobs served as stepping stones more frequently.

Transition matrices represent a useful analytical tool when it comes to description of the mobility patterns. However, they do not allow us to answer the question of what determines whether low-paid employees remain caught in low-wage jobs or whether they are able to leave them, and, if yes, in which direction. To do so, we ran a panel multinomial logit regression with random effects for each country, with the categorical outcome variable having three levels: reproduction of low-wage position, upward wage mobility, and exit out of the labor market to unemployment or non-employment.

The focus is on the role of job-related characteristics (working experience, occupation, type of employment), individual socio-demographic (age, gender, educational level, increase in education level) and household characteristics (presence of dependent children, work intensity). To control for the initial conditions, wage position in the initial year (t0) is also included in the models. We estimated the average marginal effects of the determinants for each transition from low wage. Comparisons of the average marginal effects and their confidence intervals among the CE countries are reported in Figure 2 (stability of low wage), Figure 3 (upward wage mobility), and Figure 4 (transition out of the labor market). Marginal effects with their standard errors are reported in the tables in Appendix B.

Figure 2.

Marginal effects on stability of low wage.

Figure 3.

Marginal effects on upward wage mobility.

Figure 4.

Marginal effects on exit from labor market.

In two countries—the Czech Republic and Poland—low-paid men have better chances to improve their wage position as compared with women. They are less likely to remain caught in low-wage positions and more likely to experience upward wage mobility. As these gender differences are not present in Hungary and Slovakia, the role of low employment in mothers of children aged less than 3 years, which characterized all CE countries with the exception of Poland and which could affect career paths and lead to persistently lower wages for the mothers, is questionable. However, the Czech Republic and Poland have a specific position in terms of gender division of temporary jobs. The Czech Republic shows the largest gender difference in the share of temporary jobs in the region, with women facing a higher probability to take this type of job. In 2014, this gender gap represented three percentage points, as compared with 0.5 p.p. in Poland and Hungary and 0.9 p.p. in Slovakia.¹¹ Furthermore, the Czech Republic and Poland reported the largest gender differences in terms of proportion of involuntary temporary contracts. Here again, women are more frequently involved in this type of jobs, contrary to the situation in Hungary and Slovakia where involuntary temporary contracts are more frequent among men.

In terms of age, members of the oldest age category are, as expected, the most likely to exit labor market. Furthermore, employees aged 45 or above are less likely to experience upward mobility in the Czech Republic, Hungary, and Poland. The effects of education also follow expected patterns. High-level of education increases chances for upward wage mobility and decreases the probability of being trapped in low wage in all four countries. In addition to educational level, the models also contain information whether employees took part in training in each observed year. In two countries, it contributes to improvement of the employees’ wage situation: it makes upward wage mobility more likely in Hungary and reproduction of low-wage status less likely in the Czech Republic. What differentiates Hungary and the Czech Republic from Poland and Slovakia is the fact that Poland and Slovakia reported a significantly lower proportion of workers involved in job-related trainings in the examined period (OECD, 2013: 208). This relates, in particular, to workers with the lowest level of proficiency and literacy who typically perform low-skilled and low-paid jobs.

Apart from formal education and trainings, the role of gained experiences was investigated. However, the impact of length of work experience does not show a clear pattern in the CE countries. In the Czech Republic and Slovakia, low-wage earners who work longer (6 years or more) tend to have a lower probability of climbing up the wage ladder as compared with employees with a short work experience (5 years or less), and they are more prone to reproduce their low-wage status. The opposite is true in Hungary and Poland.¹² We cannot presume that a longer work history automatically translates into accumulation of human capital, which is one of the important drivers of upward wage mobility, and then look for the reasons why it does not work in Hungary and Poland. We think that the relationship between length of work experience of low-wage workers and their mobility is rather complex and requires further investigation, which would focus on nature and forms of human capital and its role in wage mobility.

Health represents an important personal asset when it comes to the future prospects in the labor market. Our analysis shows that it systematically affects the probability of leaving the labor market: employees with poorer health are more likely to go out of the labor market in all four countries. This is partly related to the fact that it is older employees who are more likely to exit labor force.

In terms of job characteristics, the occupational category, indicated by the ISCO-08 version, plays an important role. As expected, workers in elementary occupations (category 9 in the ISCO-08 classification) show a markedly higher probability of being caught in a low-wage position as compared with employees who worked as managers or professionals (categories 1, 2 and 3). At the same time, workers in elementary occupations show a markedly lower probability of moving up along the wage ladder. A similar—although weaker—effect can be observed among skilled workers in the occupational (categories 6–8) and among clerical support workers and services and sales workers (categories 4–5).

Part-timers are not systematically exposed to higher risk of wage stagnation, as a higher probability of being caught in the low-wage position was only identified among low-paid part-timers in Poland. And it is Poland, together with Hungary, where low-wage earners who work part-time are markedly less likely to experience upward wage mobility, as compared with full-time employees. Supposing that part-timers, who do not chose part-time work freely, face more precarious future prospects than their counterparts, this finding partly reflects the different proportions of involuntary part-time workers in the CE countries. In the given period, it was Hungary that reported the highest incidence of this phenomenon (43% in 2012, 41% in 2014). While both Poland and Slovakia showed the same levels of involuntary part-time work (32% in 2012, 31% in 2014), use of part-time jobs in Slovakia is very low, making the contribution in the estimated model weaker.

Furthermore, it is evident that low-wage part-time employees are more prone to leaving low-wage positions due to exiting from the labor market. Retirement represents the most important contributing factor, as low pay employees, who work part-time, are over-represented among the age category 55–64 years. However, an early stage of work career also plays a role: the category of persons aged less than 25 years shows the second highest share of low pay part-timers.

In addition to the abovementioned job’s characteristics, our analysis shows that wage position in the time period preceding the beginning of the “analysed period” affects the future prospects of low-wage earners. This means that what will happen to employees identified as low-wage earners in time t depends on wage position in time t0. Having a wage above the low-wage threshold in time t0 increases chances—compared with employees with low wages in this time period—for upward wage mobility from low-wage status in all four countries. Being without a wage (i.e. being without paid job) in the time t0 has a similar effect: unlike persons who earned low wages in t as well as period t0, people who were without a paid job in t0 have better chances for upward wage mobility after becoming low-wage earners.

Finally, there are two variables in the models that refer to household level: presence of children in the household and work intensity. Neither of them has the effects that would show a uniform pattern across the CE countries. Care for dependent children does not seem to be an important factor of wage mobility in Central Europe. When it comes to work intensity, no clear patterns emerge.

Discussion and conclusion

The economic development of the CE countries has been formed for a long time by the economic model of competitiveness based on low wages. Now, the countries are trying to escape it and change their role in the international division of labor. Our aim was to explore whether and to what extent low wages represent a long-standing or a transitory phenomenon and what differentiates the future prospects of low-wage earners.

The CE countries do not represent a homogeneous group in terms of presence of low wages when the period 2010–2016 is considered. While low-paid employees represented, on average, more than one-quarter of all employees in Poland, Slovakia and Hungary reported only 16% or 17%, with the Czech Republic being in the middle between the two extremes. As regards the future prospects of low-paid employees, two clusters in the Central Europe emerged that correspond to the division based on the incidence of low pay. On one hand, there is Poland and the Czech Republic—two countries with higher incidence of low pay—where low pay is a transitory phenomenon only for a limited part of low-wage employees. On the other hand, Slovakia and Hungary, where incidence of low wages is lower, also show significantly lower probability of being trapped in low-wage positions. Taking into account the patterns of trajectories among low-paid employees in Slovakia and Hungary, we suppose that it is particular in Slovakia, which allows them easily to improve their wage status.

An existence of the two regimes of low-wage persistency/mobility has been translated into the differences in terms of determinants of wage transitions. However, this has only been done to a certain extent, as there are important similarities as well as differences among the countries. On one hand, in all (or by majority) of the CE countries, upward mobility is more likely among younger (under the age of 45 years), high-educated employees and among those who work in “better” occupations. Furthermore, an initial wage position plays an important role: employees who had higher wages prior to their experience with low pay have higher chances increasing their wages. A similar position holds true for low-wage earners who were previously without a job. The differences consisted of nature of labor market dualism (incidence of precarious forms of jobs) in coverage by job-related trainings, or differences based on gender. They stem, in particular, from the differences in presence of involuntary non-standard forms of employment.

Looking at findings through the lens of theories that are used in relation to wage mobility, rather limited support for arguments of signaling and human capital accumulation theories which would be applicable to all four countries was found. However, echoes of the labor market segmentation arguments can be found in the Czech Republic and Poland, where low-paid workers tend to maintain their position and their transitions to primary labor markets are more limited than in Slovakia and Hungary. In the latter two countries, low-wage jobs are more likely to serve as stepping stones to better-paid jobs.

Although all four CE countries arrived at similar regimes of international competitiveness based on low wages and underwent a period of liberalization and flexibilization of labor laws since the 2000s, opening doors for an increased use of non-standard forms of employment, they do not represent a homogeneous world in terms of low-wage employment. However, the differences between the two clusters that emerged in relation to the future prospects of low-paid workers do not fully reflect the differences in the social and labor market policies. While Hungary and Slovakia, which provided better prospects for low-paid workers, shared some institutional features in the examined period (liberalization of the labor laws), they also differ in some aspects. In 2014, Slovakia addressed the precariousness of temporary work agencies and some other non-standard forms of work, increasing their security and narrowing the differences in relation to the standard contracts (Martišková, 2021). In Hungary, focus on internal and external flexibilization of standard and non-standard contracts still persisted. What both countries have in common, and what distinguished them from Poland and the Czech Republic, was the nature of temporary job dynamics that supports upward wage mobility. One of the reasons might lie in the fact that both countries reported a higher proportion of employees in elementary occupations where temporary contracts were used more frequently. Nonetheless, tracing the evolution of institutional divergence/convergence of social and labor market policies within the four CE countries after the great recession (2008–2011) remains a challenge that will also feed the analyses of wage mobility.

The presented findings shed some light on the existing differences across various categories as well as countries, contributing to the understanding of wage transitions. What remains to examine in future research, in addition to the role of institutional similarity/divergence, are the consequences of transitions from low-wage status, particularly in terms of quality jobs and their sustainability.

Supplemental Material

sj-docx-1-cos-10.1177_00207152231156436 – Supplemental material for Low-wage mobility in Central Europe

Supplemental material, sj-docx-1-cos-10.1177_00207152231156436 for Low-wage mobility in Central Europe by Daniel Gerbery and Tomáš Miklošovič in International Journal of Comparative Sociology

Supplemental Material

sj-docx-2-cos-10.1177_00207152231156436 – Supplemental material for Low-wage mobility in Central Europe

Supplemental material, sj-docx-2-cos-10.1177_00207152231156436 for Low-wage mobility in Central Europe by Daniel Gerbery and Tomáš Miklošovič in International Journal of Comparative Sociology

Supplemental Material

sj-xlsx-3-cos-10.1177_00207152231156436 – Supplemental material for Low-wage mobility in Central Europe

Supplemental material, sj-xlsx-3-cos-10.1177_00207152231156436 for Low-wage mobility in Central Europe by Daniel Gerbery and Tomáš Miklošovič in International Journal of Comparative Sociology

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Slovak Research and Development Agency under the contract no. APVV-20-0621 and no. APVV-20-0449.

ORCID iD

Daniel Gerbery

Supplemental material

Supplemental material for this article is available online.

Notes

References

Astrov

Holzner

Leitner

, et al. (2019) Wage developments in the Central and Eastern European EU member states. Research Report 443, December. Vienna: The Vienna Institute for International Economic Studies.

Bachmann

Bechara

Schaffner

(2016) Wage inequality and wage mobility in Europe. The Review of Income and Wealth 62(1): 181–197.

Barta

Czirfusz

Kukely

(2008) Re-industrialisation in the World and in Hungary. European Spatial Research and Policy 15(2): 5–26.

Bohle

Greskovits

(2007) Neoliberalism, embedded neoliberalism and neocorporatism: Towards transnational capitalism in Central-Eastern Europe. West European Politics 3: 433–466.

Cai

Mavromaras

Sloane

(2018) Low-pay employment in Britain: Estimating state-dependence and stepping stone effects. Oxford Bulletin of Economics and Statistics 80(2): 283–326.

Cappellari

Jenkins

(2008) Estimating low pay transition probabilities accounting for endogenous selection mechanisms. Journal of the Royal Statistical Society (Applied Statistics) 57(2): 165–186.

Clark

Kanellopoulos

(2013) Low pay persistence in Europe. Labour Economics 23(C): 122–134.

Dolado

(2017) European Union dual labour markets: Consequences and potential reforms. In: Matyas

Blundell

Cantillon

, et al (eds) Economics without Borders: Economic Research for European Policy Challenges. Cambridge: Cambridge University Press, pp. 73–112.

Emmenegger

Hausermann

Palier

, et al. (2012) How we grow unequal. In: Emmenegger

Hausermann

Palier

, et al. (eds) The Age of Dualization: The Changing Faces of Inequality in Deindustrialising Societies. Oxford: Oxford University Press, pp. 3–26.

10.

Eurofound (2016) Recent Developments in Temporary Employment: Employment Growth and Wages and Transitions. Luxembourg: Publication Office for the European Union.

11.

European Commission (2016) Employment and Social Developments in Europe 2016. Luxembourg: Publications Office of the European Union.

12.

European Commission (2020) Employment and Social Developments in Europe 2020. Luxembourg: Publications Office of the European Union.

13.

Fabo

Sedláková

(2017) Impacts of the liberalization and re-regulation of the labour market in Slovakia. In: Piasna

Myant

(eds) Myths of Employment De-Regulation. Brussels: European Trade Union Institute, pp. 123–141.

14.

Fialová

(2020) Nízke mzdy? Perspektivy mzdového vývoje. In: Komentář. Sociologický ústav Akademie Věd ČR. Available at: https://www.soc.cas.cz/aktualita/nizke-mzdy-perspektivy-mzdoveho-vyvoje-v-cesku (accessed 10 February 2021).

15.

Flek

Mysíková

(2015) Unemployment dynamics in Central Europe. A labour flow approach. Prague Economic Papers 24(1): 73–87.

16.

Flek

Hála

Mysíková

(2018) Nezaměstnanost a věková segmentace trhu práce. Politická Ekonomie 66(6): 709–732.

17.

Fok

Scutella

Wilkins

(2015) The low-pay no-pay cycle: Are there systematic differences across demographic groups? Oxford Bulletin of Economics and Statistics 77(6): 872–896.

18.

Galgóczi

(2017) Why Central and Eastern Europe needs a pay rise. Working Paper 2017.01. Brussels: European Trade Union Institute, January. https://www.etui.org/sites/default/files/WP%202017-01-pay%20rise-HD-web%20version%20finale.pdf

19.

Gerbery

Miklošovič

(2021) Wage mobility in Central Europe: Patterns and micro-level drivers of movements along the wage ladder. Eastern European Economics 60: 171–188.

20.

Kahancová

(2016) The rise of the dual labour market: Fighting precarious employment in the new member states through industrial relations. Country Report: Slovakia, CELSI Research Report no. 15, April. Bratislava: CELSI.

21.

Knabe

Plum

(2013) Low—wage jobs—Springboard to high-paid ones? Labour 27(3): 310–330.

22.

Kreckel

(1980) Unequal opportunity structure and labour market segmentation. Sociology 14(4): 525–550.

23.

Layard

Nickell

Jackman

(2005) Unemployment: Macroeconomic Performance and the Labour Market. Oxford: Oxford University Press.

24.

Lucifora

Salverda

(2011) Low pay. In: Salverda

Nolan

Smeeding

(eds) Oxford Handbook of Economic Inequality. Oxford: Oxford University Press, pp. 257–283.

25.

McCormick

(1990) A theory of signalling during job search, employment efficiency and “stigmatised” jobs. Review of Economic Studies 57(2): 299–313.

26.

Maciejewska

Mrozowicki

Piasna

(2016) The silent and crawling crisis: International competition, labour market reforms and precarious jobs in Poland. In: Myant

Theorodopoulou

Piasna

(eds) Unemployment, Internal Devaluation and Labour Market Deregulation in Europe. Brussels: European Trade Union Institute, pp. 229–254.

27.

McKnight

Stewart

Himmelweit

, et al. (2016) Low Pay and In-Work Poverty: Preventative Measures and Preventative Approaches. Luxembourg: Publications Office of the European Union.

28.

Martišková

(2021) The role of social partners in the regulation of non-standard forms of work and the impact on labour market dynamics in Slovakia. In: Mihes

(ed.) Non-Standard Forms of Employment in Selected Countries in Central and Eastern Europe. Budapest: ILO, pp. 127–143.

29.

Meszmann

(2016) The rise of the dual labour market: Fighting precarious employment in the new member states through industrial relations. Country Report: Hungary, CELSI Research Report no. 12, April. Bratislava: CELSI.

30.

Mosthaf

(2014) Do scarring effects of low-wage employment and non-employment differ between levels of qualification? Scottish Journal of Political Economy 61(2): 154–177.

31.

Mrozowicki

(2016) Normalisation of precariousness? Biographical experiences of young workers in the flexible forms of employment in Poland. Przegląd Socjologii Jakościowej 12(2): 94–112.

32.

Mrozowicki

Roosalu

Senčar

(2013) Precarious work in the retail sector Estonia, Poland and Slovenia: Trade union responses in time of economic crisis. Transfer: Review of Labour and Research 19(2): 267–278.

33.

Myant

(2018) Why are wages still lower in Eastern and Central Europe? Working Paper 2018.01. Brussels: European Trade Union Institute, January.

34.

Nölke

Vliegenhart

(2009) Enlarging the varieties of capitalism: The emergence of dependent market economies in East Central Europe. World Politics 61(4): 670–702.

35.

OECD (2013) OECD Skills Outlook 2013. Paris: OECD Publishing.

36.

OECD (2016) OECD Economic Surveys: Poland 2016. Paris: OECD Publishing.

37.

Ozdemir

Ward

(2015) The characteristics of workers on low wage. Research note no. 9/2015, December. Luxembourg: Publication Office for the European Union.

38.

Pailhé

(2003) Labour market segmentation in central Europe during the first years of transition. Labour 17(1): 127–152.

39.

Peck

(1989) Labour market segmentation theory. Labour & Industry 2(1): 119–144.

40.

Plum

Pacheco

Dasgupta

(2021) When there is no way up: Reconsidering low-paid jobs as stepping-stones. Economic Record 97(318): 387–409.

41.

Rubery

Piasna

(2017) Labour market segmentation and deregulation of employment protection in the EU. In: Piasna

Myant

(eds) Myths of Employment Deregulation: How It Neither Creates Jobs Nor Reduces Labour Market Segmentation. Brussels: European Trade Union Institute, pp. 43–60.

42.

Scharle

(2012) Job quality in post-socialist accession countries. In: Fernández-Marcias

Hurley

Storrie

(eds) Transformation of the Employment Structure in the EU and USA, 1995–2007. London: Palgrave Macmillan, pp. 180–200.

43.

Stewart

(2007) Interrelated dynamics of unemployment and low-wage employment. Journal of Applied Econometrics 22(3): 511–531.

44.

Stewart

Swaffield

(1999) Low-pay dynamics and transition probabilities. Economica 66(261): 23–42.

45.

Stojčič

Aralica

(2018) (De)industrialisation and lessons for industrial policy in Central and Eastern Europe. Post-Communist Economies 30(6): 713–734.

46.

Trif

Koukiadaki

Kahancová

(2016) The rise of dual labour market: Fighting precarious employment in the new member states through industrial relations. Comparative Report, Dublin City University, Dublin, 12 December.

47.

Uhlendorf

(2006) From no pay to low pay and back again? A multi-state model of low pay dynamics. IZA Discussion Paper no. 2482, December. Bonn: Institute for the Study of Labour.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

0.04 MB

0.02 MB