Labour market flexibilisation in Lithuania: Outcomes and impacts on gender differences in work arrangements

Abstract

This paper evaluates the impacts of 2017’s labour law liberalisation on labour market flexibility in Lithuania. While employment did grow rapidly in 2017–2019, there was little change in labour market flexibility. Against expectations, part-time employment declined as labour relations continued to be administered under path-dependent institutional inertia inherited from previous decades. The prevalence of full-time, dual-earner employment was shaped by the country’s socialist legacy and was reflected in high employment rates and permanent open-ended contracts for both men and women. Analysis also showed that the revised labour law lowered the probability for women with family care responsibilities to be hired. Once hired, they were offered permanent employment albeit with the reduced protections that such contracts now provide. Impacts of the new labour law on shaping what is considered to be ‘men’s work’ and ‘women’s work’ in Lithuania are discussed.

Introduction

In July 2017 a new labour code reformed under the European Union (EU) initiative and guided by ideologically dominant neoliberal labour flexibilisation prescriptions came into force in Lithuania (European Commission, 2006; Heyes, 2013; Petrylaite, 2015). After 3 years of its operation, this article assesses the effects of labour market liberalisation in Lithuania, focusing on its gendered impacts on work arrangements, and in shaping what is considered to be ‘men’s work’ and ‘women’s work’ in Lithuania.

The 2017 labour law reforms in Lithuania were part of a broader post-2008–2009 recession wave of labour code reforms across the EU. Lithuania, a small East European country (population 2.8 million), after regaining its independence from the Soviet Union in 1991, zealously embraced neoliberal doctrines in the creation of a new market economy, with low controls on capital, open markets and reduced provisions for social protection (Bohle and Greskovits, 2007; Kolodko, 2002; von Hirschhausen, 1998). Characteristically, Lithuania responded to the severe 2008–2009 economic crisis by doubling down on neoliberal prescriptions favouring wage suppression and massive cuts to public sector expenditures. After 2010, economic growth rebounded although at the high cost of depressed and declining wages, a significant increase in poverty and unemployment, and a massive wave of emigration to core EU countries (Bieler and Salyga, 2020; Sommers et al., 2014). By the mid-2010s, the effectiveness of austerity prescriptions in stimulating economy decreased, while labour costs began to rapidly outpace increases in productivity (Juska and Lazutka, 2019: 540–541). It is within this context of EU prodding combined with waning effectiveness of ‘internal devaluation’ policies that the Lithuanian government made decision to pursue highly controversial neoliberal labour law reforms that led to passage of the new labour code.

By 2020 three annual national monitoring reports on post-reform employment contracts became available (SADM [Socialinės apsaugos ir darbo ministerija], 2018, 2019; VDI [Valstybine Darbo Inspekcija, State Labor Inspectorate], 2018) as well as European Union Labor Force Survey data covering the period until the fourth quarter of 2019 (EU LFS, 2020). This new data potentially allows assessment of the effects of the new labour code in the first two post-reform years.

For analytical purposes, studies on outcomes of labour law reforms can be subdivided into two overlapping areas (Cook, 2007: 1–24). The first, focusing on the political dynamics of labour reforms, investigates the capacity and actions of policymakers to implement reforms, and the capacities and roles of labour, business groups, transnational and other social actors in obstructing or facilitating reform goals (Adascalitei and Morano, 2016; Eichhorst et al., 2017).

The second focuses on legal frameworks and political legacies explaining their outcomes (Caraway et al., 2015; Cook, 2002). This path dependency approach sees social change as bound by preceding institutional conditions that constrain politics during a critical juncture in which actors make choices initiating a specific institutional trajectory that is difficult to reverse (Mahoney, 2000; Pierson and Skocpol, 2002). As noted by Cook (2007: 17), wholesale revisions of labour laws are very rare since they are reflective of historical evolution and embedded in formal and informal institutions and power relationships in a society. Instead, most labour law reforms occur in historically specific trajectories punctuated by critical junctures, each resulting in changes conditioned by their previous legacies.

This article, while drawing on strategic interaction research, relies on a path dependency approach in analysing the impacts of the 2017 labour code reform in Lithuania. We argue that Lithuanian labour laws and labour market, especially as relating to gender roles, continue to be shaped by a half a century-long socialist legacy which included and justified social policy covering full employment for both men and women, free education and health care and social security (Cook, 2002; Guogis and Bogdanova, 2012). In the post-independence period, pro-worker labour law inherited from the socialist era was only gradually replaced by an increasingly flexible, pro-employer or pro-capital regime. This path-dependent trend in labour law reforms was typical of other former socialist countries as well (Eichhorst et al., 2017; Koukiadaki et al., 2016; Pula, 2020).

More specifically, both left-centre and right-centre governments in Lithuania saw labour regulations inherited from the previous socialist system as hindrances to be dismantled in order to allow consolidation of market forces in a post-communist transition. However, attempts to liberalise the labour code were conflict ridden, often implying steep political costs for parties and politicians supporting them.

Therefore, in post-socialist Lithuania labour code reforms proceeded slowly and had a character of mostly ad hoc responses to crises and short-term political considerations rather than attempts to comprehensively restructure labour regulations inherited from the socialist era. For example, only in 2003 the old Soviet Labor Code of 1973, last revised in 1989 (while Lithuania was still a part of the Soviet Union) was finally changed, due mostly to pre-accession to EU membership requirements of harmonisation of national and EU labour regulations (Woolfson and Beck, 2003). Next, in 2010, in response to severe economic crisis, the Labor Code was amended to allow more ‘flexible’ industrial relations, and subsequent ‘tweaking’ of the law in 2011 through 2013 (Clauwaert et al., 2016). Finally, by 2017, the Labor Code was again re-written, in response to a slow post-recession recovery and under EU mandate (Council of the European Union, 2014; Juska and Lazutka, 2019).

While the politics of labour reforms in each of the three critical junctures was complicated and messy, they all were discursively framed by the Soviet labour law legacy. Thus, proponents of the 2017 reform continued to criticise the previous 2003 Labor Code as a Soviet-era leftover, representing ‘clutches from the past’ that created dependencies of workers ‘clinging’ to their lifetime jobs, making them afraid or unable to choose where to work in the modern economy (Juska and Woolfson, 2017: 140–141). Soviet labour law legacy in shaping gender work in post-socialist Lithuania proved to be equally significant.

More specifically, the socialist labour code codified gendered work arrangements that were reflective of prevailing ideas of what constituted ‘men’s work’ and ‘women’s work’ in Soviet Lithuania. Soviet authorities, driven by their ideological commitment to communist egalitarianism, significantly advanced the emancipation of women. However, such feminist ‘from above’ policies were designed and implemented within the framework of the dominant culture and practices of Soviet masculinity. This contradictory mixture of equality and patriarchal subordination of women under socialism was legitimised by an essentialist interpretation of gender as based on the assumption that ‘men and women were equal, but different, and that the state had a strong role to play in ensuring that women’s reproductive biology did not disadvantage them’ (Ghodsee and Mead, 2018: 103). Thus, the Soviet Labour Code of 1973 not only guaranteed women rights to full-time employment and full and equal participation in the labour force, but also provided the necessary support of parity of education and employment training for women, public childcare and maternity leave.

Despite advances in gender equality under socialism, gendered work remained entrenched. Women were concentrated in sectors of the economy that paid less and were of lower prestige such as agriculture and light industry, and in service professions such as law, medicine, accounting and teaching. In comparison, men were employed predominantly in manufacturing, engineering, construction and other blue-collar and technical jobs with higher wages, and were dominant in managerial positions. While formal gender equality was practised in economic and public life, private and family life continued to be dominated by ingrained patriarchal arrangements: women were disproportionately burdened with both full-time jobs and unpaid household responsibilities (Ghodsee and Mead, 2018; Gray, 1990).

Since the early 1990s the intersection between emerging capitalism and the prevalent practices and culture of masculinity in Lithuania began to redefine understandings of gender and gender equality inherited from the socialist era. This process was driven by neoliberal reforms that attempted to challenge essentialised interpretations of women as equal to but fundamentally different from men therefore in need of extensive welfare provisions and support. Moreover, like definitions of gender developed under socialism, attempts at gender redefinition in support of the project of restructuring society under market logic market were also framed for women, within the culture and practices of the dominant masculinity (Cornwall et al., 2008; Garlick, 2020; Vasavi and Kingfisher, 2003).

Thus, neoliberal reforms constructed both employers and employees as self-interested rational actors following the logic of the market. However, these interpretations of economic actors were modelled upon men whose reproduction outside markets depended upon women’s unacknowledged domestic work. In the case of Lithuania, this close affinity between masculinity and neoliberalism was criticised by women’s rights advocates, who pointed out that the new Labor Code assumed that ‘the ideal employee is a young, educated man, of good health and single’, while treating social responsibilities, especially ones associated with women’s roles, as of secondary importance (Saukienė, 2015).

Thus, gender in the post-communist Lithuanian labour market can be conceptualised as a path-dependent process of coercion, persuasion and negotiation between (a) essentialist interpretations of gender supported by extensive welfare protections, inherited from the socialist era and (b) neoliberal interpretations of men and women as primarily rational economic actors responding to availability of increasingly flexible employment arrangements via more active participation in the labour force and in business and entrepreneurship activities in general. Their enhanced economic activity is assumed to stimulate economic growth and increase gender equality ‘from below’ via the market (Ahl et al., 2016). In other words, neoliberal contestation of the Soviet legacy mobilised masculinity under the ideological veneer of gender-neutral economic rationality, even if it created different labour market incentives and constraints for employed men and women.

The masculine neoliberal market subject brackets out the domestic labour of household reproduction. Yet women’s market-driven behaviour is constrained by household and family responsibilities. For women, increased labour market flexibility, while providing increasing incentives and rewards, was occurring in conjunction with the dismantling of welfare support and increasing commodification of previously socialised domestic work. Therefore, for women choices in the liberalised labour market were much more circumscribed.

Analytically these choices can be differentiated as falling along the following scale. At one end of the scale is a ‘retraditionalisation’ or ‘refamilisation’ of women’s roles, when women choose or are forced to prioritise family responsibilities and retreat from the labour force, leading to the decline of dual earner and re-emergence of male breadwinner and female housewife model (Gal and Kligman, 2000; True, 2003). Retraditionalisation also includes increased feminisation of poverty as a result of the re-emergence of gender discrimination previously contained by the Soviet policies leading to suppression of women’s wages and increasing precariousness of their employment (Motiejūnaitė, 2010: 240; Kanopienė, 2000).

On the opposite end of the scale is a choice of new venues for professional advancement and business entrepreneurship opened by labour market liberalisation (Palalić et al., 2020; Ramadani et al., 2015; Tillmar et al., 2021). For example, gender differences in employment in Lithuania are also affected by the so-called revaluation of resources strategy that is available mostly to women but only to some degree to men (Fodor 1997; Ghodsee, 2005). This strategy was derived from path-dependent legacies of women’s employment in sectors of a socialist economy that were not a priority to central planners such as services, but became the drivers of economic growth and expanded dramatically in the post-independence era. Occupational segregation of women in white-collar jobs, their higher levels of education and professional and administrative experience during the Soviet era, in turn, facilitated their relatively greater success than men’s, in adapting to radical economic changes.

The labour codes are crucial in facilitating or constraining gendered or gender-neutral labour market changes because they define and codify the balance of incentives and constraints that women face when balancing home and work. However, what specific gendered impacts the new labour code is having on the Lithuanian labour market is an empirical question to which we will turn following the brief review of changes in the 2017 reformed labour law reform and a discussion of methodology used evaluate labour market flexibilisation in the post-reform period. We will conclude with the discussion of impacts that the new labour law is having on shaping what is considered to be ‘men’s work’ and ‘women’s work’ in Lithuania.

The 2017 labor law reform

The new version of the Labor Code (LC) (2016.09.14, Nr. XII-2603) significantly changed Lithuania’s labour law (LR Seimas (Lietuvos Respublikos Seimas), 2016). Our focus here is the new provisions designed to increase employment flexibility and provide additional opportunities for employees to balance work and family responsibilities.

Among the most controversial and contested provisions of the new labour law were those introducing new types of contracts and working-time arrangements. In addition to already existing open-ended, fixed-term and seasonal employment contracts, the law established temporary agency, apprenticeship, project-based, job sharing and multiple-employer employment contracts (LC, Article 66). A proposal to introduce zero-hour contracts was turned down by the Seimas owing to overwhelmingly negative public reaction. New types of working-time provisions however included fixed duration, annualised hours, flexible work schedule, split shift working time and individualised working-time arrangements (LC, Article 113).

Article 28 of the Labor Code provides guidelines on balancing employees’ work and family obligations. The law directs employers to implement a number of measures enabling employees to better attend to their family obligations. In response to an employee’s requests relating to the meeting family obligations, employers must provide a written response outlining practical measures to accommodate them. For example, the employer must arrange at least one-fifth of standard working hours remotely when it is requested by a pregnant or breastfeeding woman, or a single parent raising children under 3 years of age (LC, Article 52). The employer must accommodate requests for a temporary part-time work schedule to enable an employee to provide care for a child under 3 or for nursing other family members (LC, Article 40). The new law also retains benefits for single parents, parents of two and more children and those raising children with disabilities, entitling them to one extra day’s paid leave each month, and those with three or more children – to 2 days paid leave per month (LC, Article 138).

The new Labor Code, however, also significantly reduced severance compensation and enabled more flexible lay-off and dismissal procedures. The list of causes for dismissal at the employer’s initiative and without worker fault was extended, while dismissal notice periods shortened (LC, Article 57). Overall, while providing for more flexibility in employment, the new LC shifted socio-economic risks from capital to labour and enhanced employers’ abilities to control labour costs.

Measuring labour market flexibilisation

In order to describe changes in the labour market produced by the labour code reform, just measuring increases or decreases among various employment types and employment by gender categories is not enough. Assessing the nature and size of changes also requires criteria to assess and compare impacts. Here, the labour law reforms will be evaluated according to the degree to which claims of pro- and against-liberalisation of labour law were validated in the first two post-reform years, with special focus on gender differences in demand for employment flexibility.

For 2.5 years preceding the passage of the new labour laws, Lithuanian politics was dominated by contentious debates on the pros and cons of labour market liberalisation (Juska and Lazutka, 2019; Juska and Woolfson, 2017). Proponents of the reforms claimed that the new labour code would allow the creation of up to 90,000 new jobs, reduce unemployment to 7% and lead to growth in wages by 3%–5% (BNS [Baltic News Service], 2015; Lietuvos, 2015) because ‘when it is easier to hire and fire, businesses are more willing to risk [expansion]’ (Kupetytė, 2015). It was also claimed that more flexible employment would be especially beneficial in extending possibilities to combine work and family care responsibilities. Thus, gender wage inequalities in Lithuania would decrease because flexible, part-time employment contracts for primary caregivers would help shorten breaks in labour market participation for childbearing and family care needs (Davulis, 2017).

In comparison, critics of the liberalisation reforms asserted that the new jobs forecasts were greatly inflated. They claimed that the introduction of new forms of ‘flexible’ employment would lead to increasing labour market precariousness, that is, to a potential rise of the ‘gig economy’, which could also increase gender wage inequality in the country. Furthermore, critics pointed out that in neighbouring Poland recent labour law liberalisation had led to a significant increase in employment precariousness and gender wage gaps, with disproportionate numbers of women relegated to the secondary, low-skill and poorly paid sector (Dagys, 2015; Gatti, 2014).

For 2.5 years prior to the passage of the new labour law, the Lithuanian business and financial lobby put enormous pressure on politicians and state officials, campaigning relentlessly and promising numerous benefits from liberalisation (Juska and Woolfson, 2017). Since capital had succeeded in re-writing the labour law on its terms, it would be reasonable to expect that businesses would make use of these highly desirable employment flexibility opportunities once they became available.

However, caution is needed when measuring and interpreting changes in post-reform outcomes. Both advocates and opponents of labour liberalisation in Lithuania tended to claim that changes in the labour law would directly affect the structure of the labour market. However, such argumentation is reductionist and simplistic, because labour law liberalisation is only one among many factors that affect labour markets in general and gender differences in employment in particular.

For example, in Czech Republic and Estonia, unlike in Poland, the introduction of provisions for flexible employment did not result in an increase in part-time employment. It is suggested that in comparison to EU core countries, underdevelopment of part-time employment in Central and Eastern Europe is due to low wages that make income from part-time work insufficient to provide for personal and household needs, while requests for more flexible working arrangements may be limited owing to fears of losing job promotion opportunities and fears of dismissal (Hamplova, 2006; Helemae and Saar, 2006). The latter consideration is especially relevant if dismissal costs to employers are relatively low while social protection in case of unemployment is weak, as is the case in Lithuania (Masso and Krillo, 2010). For employers low wages could, paradoxically, act as disincentive for flexibility, because part-time and other flexible forms of employment do not necessarily ‘bring sufficient cost reduction to counterbalance the negative effect of the unavailability of part-time employees’ (Cazes and Nesporova, 2004: 35). Additionally, the reluctance of women in Central and Eastern Europe to partake in more flexible employment arrangements may also be shaped by the socialist-era legacy of full-time/dual-earner types of employment (Hamplova, 2006; Helemae and Saar, 2006).

Furthermore, new types of employment contracts represent only one way of increasing employment flexibility. In Lithuania compliance with the labour law is spotty, and its enforcement ineffective. Employers quite often achieve ‘employment flexibility’ by simply disregarding labour law provisions. The discrepancy between de facto vis-à-vis de jure aspects of labour laws became an issue in parliamentary debates when, in the aftermath of the 2008−2009 financial crisis, 93% of those laid-off were classified as quitting ‘voluntarily’, thus becoming ineligible for severance pay. This was probably the most egregious mass noncompliance with the labour laws in recent years (Juska and Woolfson, 2017: 142–143).

The labour taxation regime, among other factors, can also limit the use of part-time and other flexible forms of contracts in Lithuania. A World Bank study had shown that the labour market in Lithuania is characterised by ‘employers . . . hav[ing] a substantial degree of flexibility with employment adjustment [and] limited flexibility to wage adjustment due to a high statutory wage’ (Rutkowski, 2003: ii). Since social insurance and other taxes levied on wages in Lithuania are relatively high, this could potentially limit employer incentives to hire employees on part time or other flexible contracts, as savings would be rather limited (Lazutka at al., 2018; Okunevičiūtė-Neverauskienė et al., 2017).

Finally, reluctance to more widely adopt flexible forms of employment can be influenced not only by the dual-career and dual-earner legacy of the socialist era, but also by costs of the so-called ‘flexibility stigma’ attached to women who have a higher demand for flexibility in order to balance work and family care responsibilities. Flexible employment from the point of view of the employer might be interpreted as ‘inflexibility’, especially when long working hours and working particular hours are of importance (Goldin, 2014; Walby and Olsen, 2002).

To date, unfortunately, no data is available to test how the factors described above facilitated or constrained adoption of new flexible employment arrangements in Lithuania. Nevertheless, analysis of the most recent reports on the post-reform labour market in Lithuania do enable us to (a) describe labour market changes in 2017−2019 and (b) analyse gender differences in demand for flexible working arrangements and employment-related outcomes in pre- and post-labour law reform periods. As it is well established that women tend to have a higher demand for workplace flexibility to balance work and family care responsibilities (Chung and van der Lippe, 2020; Clawson and Gerstel, 2014; Singley and Hynes, 2005), we expect the legislation to have a greater effect, if any, for women.

We are aware that the time frame may be too short to register meaningful labour market changes, since post-reform data have been available for only 2 years since the new labour law took effect. However, it is also necessary to consider the duration of the political process preceding the passage of the new labour code. Debates over the reforms lasted for almost 2 years, beginning in December 2014, when the government made public the first draft of the new law. Until September 2016 labour law reform was extensively covered by news media. The core provisions, especially introduction of the new types of contracts and new types of working-time arrangements, remained virtually unchanged when enacted into law. It could thus be expected that at least some of the new employment contracting opportunities would have been used. Furthermore, labour turnover in Lithuania is very high; in 2018 close to 625,000 new contracts were signed and a similar number terminated in a labour market of 1.4 million, thus providing ample opportunity for employers and employees to use the new provisions (VDI, 2018).

Methodology

Empirical analysis of the impact of the new labour law on labour market flexibility in Lithuania will be conducted in two blocks. First, the aggregate data from national monitoring reports will be used to assess pre- and post-reform changes in shares of fix-term versus other forms of contracts and changes in prevalence of part-time work. This will provide a broader overview of changes in Lithuanian labour market flexibility since 2017.

General employment dynamics capture both the effects of the new legislation, as well as other effects, such as those of the economic cycle. Therefore, in the second block we will attempt to disentangle and measure effects of the new labour code (as opposed to effects of other exogenous variables) by analysing employment trends by gender before and after the reform, including gender employment gaps among those with and without family care responsibilities. For this purpose, the difference-in-differences (DiD) approach will be used. DiD is a statistical technique that attempts to mimic an experimental research design using observational study data. It calculates the effect of a treatment (i.e., change in the labour code) on an outcome (i.e., change in employment arrangements) by comparing the average change over time in the outcome variable for the treatment group, compared to the average change over time for the control group (Josselin and Le Maux, 2017).

To measure the impact of the new labour law on employment flexibility, we compared the outcomes of the labour code reform on women with family care responsibilities with the outcomes for two control groups: (a) women with no such responsibilities and (b) men with family care responsibilities. These two control groups are used to test an assumption that interaction of two variables – first, female gender and second, having family care responsibilities – would produce the highest demand for flexibility at the workplace. To put it differently, we assume that the direct labour law effect (as opposed to effects of other exogenous variables) will be expressed in the degree to which women with care responsibilities were more likely than employees in two control groups to be employed and use the flexible working arrangements provided by the new labour law.

Following Josselin and Le Maux (2017: 492–498), we ran six regression models to analyse the effects on the following indicators: employment status (1), type of employment (2–4) and working time flexibility (5–6). The specifications of the models are as follows:

(1) $\begin{array}{c}{y}^{PFT}=\alpha +{\beta }_{1}P+{\beta }_{2}F+{\beta }_{3}T+{\gamma }_1\left(P\times F\right)+{\gamma }_2\left(F\times T\right)+{\gamma }_3\left(P\times T\right)\\ +{\gamma }_4\left(P\times F\times T\right)+\delta X+\epsilon \end{array}$

where P is a period effect, F is an effect for women and T is the effect of belonging to a group of people with care responsibilities. Interactions between these effects are also included (i.e., the period effect for women ( $P\times F$ ), effect for women with care responsibilities ( $F\times T$ ) and the period effect for people with responsibilities ( $P\times T$ ). Also included are controls X for such socio-demographic characteristics as age, marital status, education level and status, level of urbanisation, employment status of the partner and, where appropriate, for profession, industry, size of the firm and form of employment. The reform effect is identified after controlling for all other effects and controlled individual characteristics through the combined period effect for women with care responsibilities ( $P\times F\times T$ ), building on an assumption that higher demands for flexibility and therefore higher effect of the new legislation, if any, can be expected for women with care responsibilities.

We used European Union Labour Force Survey (EU LFS), including its ad hoc module on balancing career and family care responsibilities (EU LFS, 2020). The sample is about 10,000 respondents. We used quarterly LFS data for 2014−2019 to analyse the labour market situation before and after the labour code reform; 2014 was chosen as the start of the data series since it was the year when the economic expansion cycle began following the post-2008−2009 crisis. We applied parametric t-tests on the quarterly pooled LFS data for testing statistical significance of the results. For DiD analysis we used an ad hoc module which included additional variables on family care responsibilities (2018q2 [quarter 2]) together with the data before the reform (2017q2). The two data points were selected to neutralise seasonal differences in the labour market. Working-age population was defined as being 15−64 years old.

Labour law liberalisation and changes in labour market flexibility

The data provided by the Ministry of Social Security, the State Labor Inspectorate (VDI, 2018) and Eurostat indicate that since the mid-2010s, the Lithuanian labour market was characterised by increasing employment and rising wages. In 2017q2−2019q2 total employment rate increased from 70.6% to 73%, exceeding the EU28 average of 69.3% in 2019 (Eurostat, 2020). However, these increases were due to more cyclical recovery and expansion following the 2008−2009 recession and other changes (including a 46% increase in the minimum wage between 2017 and 2019), rather than changes in labour law. Overall, employment in Lithuania gradually increased and unemployment declined from 2010 until March 2020, when the economy entered a COVID-19-induced recession (Statistics Lithuania, 2020).

While the number of people employed in Lithuania increased post-reform, employment flexibility – as measured by heterogeneity of types of employment contracts used – actually declined. The number of fixed-term contracts post-reform declined by about 10%, while the number of permanent contracts continued to grow. Among 625,000 new employment contracts completed in 2018, only 3616 (or less than 1%) were of the new types established by the revised labour code. Project contracts, accounted for 77% of new-style contracts, while other types remained rarely used (SADM 2018: 3).

Why were businesses and the financial lobby so heavily invested in promoting the new labour code while seemingly disregarding the new flexible employment opportunities once they became available? The zeal of the pro-liberalisation drive to reform the labour law without its actual application in 2017−2019 may have been a case of ‘competitive policy signalling’, a strategy embraced by East and Central European countries ‘to assert their capitalist credentials and differentiate themselves in a region . . . in competition for foreign investment in conditions where information about local investment conditions is limited’ (Appel and Orenstein, 2018: 5–6). Labour law liberalisation in Lithuania may have been undertaken primarily because it was reflective of ‘widespread belief among investors and policymakers [that such reforms] were correlated with profitable foreign investment’ (Appel and Orenstein, 2018: 116; see also Bandelja et al., 2015; Campos and Kinoshita, 2008). It seems that with signalling accomplished, business interests gave little consideration to using new employment flexibility provisions. Instead, labour relations continued to be administered under path-dependent institutional inertia inherited from previous decades.

In sum, aggregate data for 2017−2019 suggests rapid growth in employment (in absolute terms as well as in comparison with EU28 countries), but little if any change in labour market flexibility. In order to estimate how much of these post-reform labour market changes were due to impacts of the new labour law as opposed to other exogenous variables, we now turn to analysis of gender differences in demand and outcomes of flexible work arrangements in the post-reform labour market.

Flexible work arrangements in post-reform labour market: Gender differences in demand and outcomes

As discussed above, the Lithuanian labour market is still shaped by its socialist-era legacy of dual-earner employment. An average employment rate for the period of 2014−2019 was 69.1% for women, 70.3% for men, with an increasing trend for both sexes (Figure 1). However, women were more often inactive (at an average 25.8% for women, 22.6% for men), while the unemployment rate was higher for men (an average of 7% for women, 9.2% for men). There was no significant increase in women’s employment rate in the first post-reform year, while an increase was observed for men (from 70.5% to 73.6% between 2017q2 and 2018q2).

Figure 1. Employment, unemployment and inactivity rates, %.

Source: Own calculations based on LFS data, see data in the Appendix (Table A1).

However, there were significant gender differences among those 8.3% of employed who worked part-time in 2014−2019. A larger proportion of women – on average 10.3% – were working part-time compared to 6.2% for men.

Since women tend to have higher labour market flexibility demands, we will now focus on differentiating and describing the impact of the labour code reform on flexible work arrangements for women with family care responsibilities by using the DiD approach. We ran six regression models to analyze the effects on the following indicators: employment status (1), type of employment (2–4) and working time flexibility (5–6). Table 1 shows the identified effect for women (F), time effect (P) and effect for people with family care responsibilities (T), as well as their interactions.

Table 1. Effects of the labour code reform on working arrangements of working-age women with family care responsibilities.

Source: Own calculations based on merged the European Union Labour Force Survey model and the European Union Labour Force data for 2017q2 (EU LFS, 2020).

Non-significant effects at 0.001 significance level are not shown. All models control for the following characteristics: age, marital status, education level and status, level of urbanisation, employment status of the partner and, where appropriate, for profession, industry, size of the firm and form of employment.

The results presented in Table 1 suggest that women compared to men were less likely to be employed, to be working full-time, to be working standard hours and to indicate that their working time allowed for balancing their care responsibilities (i.e., they did not wish to work more or work less). On the other hand, employed women were more likely to be on permanent employment contracts (rather than self-employed; see Figures 1 and 2).

Figure 2. Share of employed on fixed contracts and working part-time by sex.

Source: Own calculations based on the European Union Labour Force data (EU LFS, 2020), see data in the Appendix (Table A2).

Those with family care responsibilities (T) were less likely to be employed, to work full time or standard hours and to consider their working time arrangements satisfactory, compared to those without such responsibilities. Those with family care responsibilities had a higher probability of being on permanent employment contracts. Most likely, the majority of employees in this group were employed in the sizable public sector, about 22% of total employment in Lithuania (Eurostat, 2018).

The period effect (P) reflects the general trends in the expansion phase of the economic cycle which was characterised by increasing employment, increase in full-time and permanent contracts, as well as rise in employees’ satisfaction with working time arrangements. The latter could be attributed to the fact that economic growth and structural labour shortages in the late 2010s led to a modest increase in employees’ bargaining power (Lietuvos, 2019: 20).

Table 1 measures effects of three more interactions: the effect on women with family care responsibilities (F * T), period effect on women (P * F) and period effect on employees with family care responsibilities (P * T). F * T interaction effects indicate that women with family care responsibilities had a significantly lower probability of being employed compared to women without such responsibilities. If employed, they were also less likely to have a permanent contract and more likely to be self-employed. However, those with a contract had a higher probability of working a standard, 8 am to 5 pm Monday through Friday schedule – one compatible with day-care and school operating hours. Two more period effects were observed. First, there was an 11% increase in the share of women who were satisfied with their work time arrangements. Second, employees with family care responsibilities (P * T) saw a 9% increase in their employment.

It is important to note that all the effects presented in Table 1 are cumulative. In Table 2 all period effects (P + P * F + P * T) for 2017q2 through 2018q2 are summed, indicating that although employment increased for both sexes, an increase was almost twice as high for people with family care responsibilities, despite a modest decline of the proportion of women with family care responsibilities employed full-time. Finally, work schedules became more flexible for men as well as women with family care responsibilities. Positive period effects are most likely due to more employers willing to hire and accommodate employees with family care responsibilities in a tight labour market.

Table 2. Cumulative change due to period effects by gender and care responsibilities.

Source: Own calculations based on merged the European Union Labour Force Survey (2018q2) model and the European Union Labour Force data for 2017q2 (EU LFS, 2020). Non-significant effects at 0.001 significance level are not shown. All models control for the following characteristics: age, marital status, education level and status, level of urbanisation, employment status of the partner and, where appropriate, for profession, industry, size of the firm and form of employment.

As suggested in the methodology section, we conceptualised the direct effect of the new labour code on increasing flexibility in balancing work and care responsibilities as an additional, ‘premium’ effect. That is, we examined the degree to which women with care responsibilities were more likely to use the flexible working arrangements provided by the new labour law than employees from the two control groups, while controlling for period effects.

The direct labour law reform effects (P * F * T) identified in Table 1 after controlling for other effects and individual characteristics are significant in three areas: negative effects on employment level (−9.8%) and contracted work (−17.3%), and a strong positive effect on permanent contracts (+76.9%).

Probably the most important DiD analysis finding was close to a 10% reduction in the probability of employment for women with family care responsibilities due to the new labour law. This suggests that while employment of women with family care responsibilities did grow in 2017−2019, their probability to be employed was about 10% less than that of the control groups. At the same time, as the new labour law was suppressing employment of women with family care responsibilities, it also had a very strong effect on increasing probability to be hired on a permanent basis (+76.9%), albeit with the reduced protection it now provides, and decreasing probability of contract work (−17.3%). All three identified direct labour law effects were contrary to the reform’s design as it intended to increase employment and expand the variety of contracts available especially for women with family care needs.

We posit two reasons for the lower probability for women with family care responsibilities to be hired, but if hired, then under permanent employment contracts. First it reflects institutional inertia or strong path-dependency effects that seem to operate in the Lithuanian labour market. Secondly it reflects changes introduced by the revised labour law. Employment suppression for women with family care responsibilities is reflective of reduced employee dismissal costs (especially reduction in severance pay) and the simplified dismissal procedures that were introduced by the new labour code. This new flexibility modified costs-benefits calculations to the comparative disadvantage of women with family care needs. They became less likely to be hired and more likely to be fired than employees in the control groups. At the same time, when the decision was made to hire, it was done following the historical continuity of preference for and familiarity with permanent contracts going back to socialist era.

Conclusions

We have attempted to measure the impacts of the labour law liberalisation in Lithuania on labour market flexibility in two respects. We defined formal impacts as being indicated by increasing variation in new types and scope of employment contracts resulting from legal and statutory changes in the labour code, with a special emphasis on gender differences in work arrangements. We defined informal effects in terms of the impact of the labour law reforms on ideas and practices concerning what constitutes gender and gendered work in post-communist Lithuania.

Impacts of legal and statutory changes in the new labour code were evaluated with regard to the claims made in the 2.5 years of campaigning for and against liberalisation before the 2017 passage of the new legislation. Reform proponents claimed that newly introduced flexible contracts would enable an increase in employment and expand options to combine work and family care responsibilities by shortening breaks in labour market participation due to childbearing and family care needs. The opponents of the reform argued that liberalisation would result in increase in precariousness of employment, weakening of social protection, and lead to a decline in wages of low-paid workers and increased gender wage gaps.

Contrary to predictions by reform critics, proportion of permanent contracts in post-reform the labour market continued to grow, while contract work declined, and use of the new flexible forms of employment remained marginal. Pro-liberalisation claims, similarly, failed to materialise. Despite availability of the new flexible employment provisions in the post-reform period, employers actually increased their reliance on permanent contracts.

This indicates that despite changes in laws, labour relations in Lithuania continued to be administered under path-dependent institutional inertia inherited from previous decades. Following Appel and Orenstein (2018) we suggested that lobbying for flexibility provisions that were (so far) ignored by employers represented, in part, a competitive policy signalling campaign designed to convey a message about the favorable business climate in the country.

The increased employment rate in 2017−2019 continued a trend for close to a decade preceding the abrupt economic downturn caused by the COVID-19 pandemic. This indicates that the increase in hiring and decrease in unemployment in 2017−2019 were strongly influenced by period effects, that is, a phase in an economic cycle.

We used the DiD analysis to differentiate between and measure period effects and effects of the new labour code on employment. We hypothesised that a greater effect of the new legislation, if any, can be expected for women with family care responsibilities. Our analysis showed that despite availability of flexible employment options, the new labour code depressed employment growth among women with family care responsibilities, which is especially worrisome.

Most recent data on the changes in the Lithuanian labour market due to the COVID-19 pandemic seem to confirm our findings. In the second quarter of 2020 unemployment in Lithuania grew by 2%, a rate that was among the highest in the EU despite the relative stability of other macroeconomic indicators. Lithuania was the only euro area state that did not see real GDP decline in the first quarter of 2020; even as GDP declined in the second quarter, net exports mitigated the decline as exports fell less than imports. Furthermore, industrial production picked up considerably in June 2020 and this positive trend continued into the third quarter (European Commission, 2020: 104; Eurostat, 2020). This indicates that labour is increasingly and disproportionately bearing the brunt of the recession, as the labour code facilitates shifting the costs of the economic downturn from capital to labour. In arguing that this would be a result of the new labour code, opponents of labour law liberalisation seem to be proven correct.

We also argued that the labour law reform attempted to redefine informal institutional practices and ideas of what constitutes gender and gendered work in Lithuania. Neoliberal reforms challenged essentialist understanding of gender typical to the socialist era, that is, of women being equal to but fundamentally different from men due to their reproductive biology, therefore in need of extensive welfare support. Instead, the new neoliberal subject that was constructed by the reformed labour code was supposed to be genderless, that is, driven primarily by the imperatives of market rationality. However, the very act of bracketing labour markets out of their social and cultural embeddedness and conceiving them as a realm of neoliberal, market-based freedoms was accomplished by mobilising masculinity. The new neoliberal subject was modelled under the man who, in the words of the main architect of the reforms Professor Tomas Davulis, from Vilnius University, was intent on breaking the ‘shackles’ of a labour market burdened with outdated restrictions (i.e., social protections inherited from the Soviet era) and ready ‘to work any time of the day or night, and anywhere opportunities arise’ (Davulis, 2013). It seems that the affinities between masculinity and neoliberal labour market reforms in Lithuania were confirmed by our study. The effect of the new legislation in suppressing employment growth for women with family responsibilities was indicative of the trend towards ‘retraditionalisation’ of definitions of what constitutes the work of men and work of women in Lithuania.

However, we must be cautious when considering the impact that retraditionalisation of work might have on increasing gender inequality in the Lithuanian labour market. First, our analysis is still provisional because of the limited data available as well as the limited time to register the impact of the new laws on the labour market. Second, as we have pointed above, liberalization of labour codes can also contribute to increasing bifurcation of gender-based employment benefiting some of its segments (such as women with higher levels of education in service sectors via revaluation of resources strategy) while depressing wages and increasing precariousness of low-skill and low-wage employment for women. Missing in our analysis and requiring further investigation is the impact of increasing labour market flexibility on entrepreneurship strategies available to women, and on the entrepreneurship of women in general. Higher levels of education of women and more opportunities for professional advancement in the service sector (as compared to men) and their increasing entrepreneurship could potentially counterbalance the impact of retraditionalisation of women’s work and, instead of increasing gender inequality, indicate a trend towards growing social and economic stratification and fragmentation of gender-based groups in the Lithuanian labour market.