Labour market reforms in times of globalisation

Nicol Foulkes Savinetti & Aart-Jan Riekhoff (Eds), Shaping and re-shaping the boundaries of working life. Tampere: Tampere University Press, 141–158. http://urn.fi/URN:ISBN:978-952-359-020-5 Labour market reforms in times of globalisation Aart-Jan Riekhoff Introduction Globalisation1 is commonly considered to have major effects on labour market performance and outcomes in advanced industrial societies. The internationalisation of markets for goods and capital has brought about increased competition between firms on a global scale. The need for creating competitive advantage has pressured countries into creating business and investment-friendly environments. Deregulation and privatisation of labour, capital and goods markets are aimed at improving competitiveness and efficiency of the economy, but at the same time increase the vulnerability of 1 In this chapter, when referring to “globalisation” what is meant is economic globalisation, i.e. the opening up of markets for goods and capital. Obviously, there are other forms of globalisation, such as social, cultural and technological, but these are not studied here. 142 Aart-Jan Riekhoff Nicol Foulkes Savinetti & Aart-Jan Riekhoff (eds) labour markets to economic shocks and increase the exposure of workers to labour market risks (Buchholz et al. 2009; Genschel 2004). In the literature on globalisation and the welfare state, there are two main hypotheses on how governments react to these challenges of globalisation (Koster 2009; Meinhard & Potrafke 2012). The “globalist” or “efficiency” thesis, on the one hand, claims that under increasing international openness, the welfare state will contract because of pressures to reduce taxes in the world-wide competition to attract investors. With capital being more mobile, firms tend to reallocate away from regulated and unionised sectors (Potrafke 2013). The “compensation” hypothesis, on the other hand, states that globalisation and economic openness are generally expected to enhance feelings of economic insecurity among the population and will lead to increased demands for redistribution and protection (Agell 2002; Rodrik 1998). Empirical evidence has been provided for both hypotheses with no definite conclusion in favour of either (Koster 2009). In fact, Meinhard and Potrafke (2012) have interpreted this as a possibility that both effects co-exist: governments both compensate globalisation losers as well as try to improve efficiency. Given the data and methods used so far, most studies have only addressed the question of which effect offsets the other, rather than which hypothesis is true (Meinhard & Potrafke 2012, 273). Quantitative comparative studies on the nexus of globalisation and labour market protection have used either (changes in) spending on unemployment benefits (UB) or UB replacement rates in year t as dependent variables (Allan & Scruggs 2004; Gaston & Nelson 2004; Jensen, Knill, Schulze, & Tosun 2014; Potrafke 2010; Swank 2005). Analyses of absolute levels of these indicators are, however, problematic given the path-dependency of government programmes: the absolute level of UB in t is highly correlated to its level in t-1 (Kittel & Winner 2005, 280). Changes in expenditures and replacement rates are not necessarily policy reforms in themselves, but rather the Labour market reforms in times of globalisation 143Shaping and re-shaping the boundaries of working life results of reforms. There might also be a time lag effect: a reform and a change in outcomes do not always follow each other in the same year (Green-Pedersen 2007). The net outcome of a reform can be that expenditures and replacement rates remain unchanged within a certain year, whereas actually major shifts in entitlements, eligibility, or accessibility do take place, but perhaps over a longer period. An additional shortcoming in the empirical literature on globalisation and labour market policies is that most studies focus on changes in UB as the only indicator of social protection, ignoring that fact that other policies and institutions can perform as “functional equivalents” (Boeri, Conde-Ruiz, & Galasso 2003; Bonoli 2003). The effects of globalisation on employment protection legislation (EPL) have been especially under-researched (for exceptions, see: Fischer & Somogyi 2009; Potrafke 2010; 2013). EPL, however, performs a very different role in the labour market than UB. First, it is commonly viewed as more distorting and less efficient as a protection mechanism than UB (Blanchard, Jaumotte, & Loungani 2013; OECD 2013; Saint-Paul 2002). Second, in terms of labour market protection, EPL provides some degree of job security to workers, whereas UB provides a certain level of income security. Third, while the terms “institution” and “policy” are used interchangeably in this article, EPL resembles the concept of an “institution” more closely. Elmelund-Præstekær and Baggesen Klitgaard (2012) argued that policy retrenchment (i.e. cutting back on expenditure or benefit levels) follows a different logic than institutional retrenchment (i.e. changing programmatic rules and procedures). Taking these differences into account, there exists the additional possibility that one policy is used to offset or compensate the negative outcomes of the other in the face of globalisation (Jensen, Knill, Schulze, & Tosun 2014). In order to address some of these above-mentioned shortcomings in the literature and to re-test both hypotheses on the basis of new data, in this article I identify how globalisation affects the likelihood 144 Aart-Jan Riekhoff Nicol Foulkes Savinetti & Aart-Jan Riekhoff (eds) for different types of labour market reforms. To do so, I make use of a new dataset based on the Social Reforms Database that was collected by the Fondazione Rodolfo DeBenedetti and the Institute for the Study of Labor (fRDB-IZA 2010). On the basis of this data, it is possible to take into account expansion and retrenchment of unemployment benefits as well as the regulation and deregulation of employment protection for a set of 14 European countries for the period of 1980– 2007. This data allows focusing on the reforms as discrete policy events, analysing the effects of globalisation on each type of reform, as well as controlling for the economic, political and institutional settings at each point in time. Data Dependent variables This chapter makes several innovations in relation to the existing literature. First, it treats reforms as discrete policy events and not as changes in policy outcomes, such as expenditures and replacement rates. Second, it analyses reforms in both UB and EPL with the assumption that not all labour market policies are affected similarly by globalisation. Third, it treats expansionary and regulatory reforms as separate from retrenching and deregulatory reforms in order to identify any offsetting effects. For operationalising the dependent variables that suit the aims of this study, I made use of the fRDB-IZA Social Reforms Database (2010) to identify the variation in reforms in UB and EPL for 14 European countries over the period of 1980–2007. This database records and describes adjustments and reforms that have been implemented in UB systems and EPL for Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden and the United Kingdom. Labour market reforms in times of globalisation 145Shaping and re-shaping the boundaries of working life Using the descriptions and classifications in the dataset, I recoded each of the reforms in order to make quantitative analysis possible. For details on how the recoding was done, I refer to my previous work (Riekhoff 2015; 2017). The result is a series of indicators with different levels of comprehensiveness of the reforms. Distinguishing levels of comprehensiveness is important because small and corrective policy changes occur on a continuous basis, but in this study, the aim is to explain the implementation of labour market reforms that are intended to have a significant impact. Since this is a new dataset and there is no established or tested cut-off point for what makes a reform comprehensive, I create a set of dependent variables with various comprehensiveness levels for analysis. UB expansion and EPL regulation are considered as reforms in one direction (increased compensation, decreased economic efficiency), whereas UB retrenchment and EPL deregulation deviate in the opposite direction (decreased compensation, increased efficiency). The dependent variables are dichotomous categorical variables, expressing whether, in a specific year and in a specific country, a reform took place and whether this reform classifies as 1) any reform, 2) a reform having at least one comprehensive feature, 3) having at least two comprehensive features, 4) having at least three comprehensive features or 5) having all comprehensive features. If in the same year more than one reform takes place, the more comprehensive reform is taken into account for analysis. Independent variables For the selection of the independent variables for the model, I largely followed Allan and Scruggs’s regression specification from their 2004 study. Their article is commonly used as a benchmark of robust findings, which have been reproduced by others in related studies (Jensen et al. 2014, 537). Apart from two variables measuring the exposure to globalisation (trade and financial openness), economic 146 Aart-Jan Riekhoff Nicol Foulkes Savinetti & Aart-Jan Riekhoff (eds) control variables include GDP growth rate, the government balance and the unemployment rate. Political and institutional control variables include a corporatism indicator (level of collective bargaining, “veto points” (the extent to which a government might face institutional obstacles in decision-making) and the percentage of left-wing cabinet seats in a given year. Table 1 describes the independent and control variables (see also Riekhoff 2015; 2017). Table 1. Summary statistics independent variables Indicator Source Mean SD Min. Max. Trade openness OECD: Trade-to-GDP-ratio 74.05 32.51 31 184 Financial openness Quinn & Inclan (1997): Capital openness indicator 86.58 16.08 38 100 Unemployment rate OECD: Annual unemployment rates 8.42 4.03 1.62 24.17 Change in unemployment rate OECD: Change in unemployment rates (Ut-Ut-1) 0.03 1.13 -3.32 5.06 Government balance IMF, World Economic Outlook Database: General government balance as % of GDP -3.15 3.84 -15.70 6.90 GDP growth rate OECD: Annual GDP growth rate 2.58 2.03 -6.00 10.92 Corporatism Visser (2011): Level of wage coordination 3.42 1.11 1.00 5.00 Left cabinet portfolios Armingeon et al. (2014): Percentage of left-wing cabinet seats 41.07 38.64 0100 Veto points Armingeon et al. (2014): Veto points composite indicator 1.23 1.26 0 4 UB replacement rates Van Vliet & Caminada (2012): UB net replacement rates for average production worker 56.48 19.99 292 Methods The likelihood of each type of reform taking place in each particular year was analysed using times-series-cross-section analysis for binary data with discrete-time logit models (Allison 1982; Beck, Katz, & Tucker 1998). Because there are recurrent events, for example, each country can have more than one reform during the period of 1980– Labour market reforms in times of globalisation 147Shaping and re-shaping the boundaries of working life 2007, it is likely that the occurrence of one event is dependent on the event history of a particular country. The likelihood of a reform taking place is assumed to depend on the duration of the period of non-reform. In other words, the observations are temporally related. Therefore, to prevent artificially inflated t-values, I followed Beck, Katz and Tucker’s (1998) strategy of treating the data as grouped duration data and including a series of dummy variables for each of the number of years since 1980 or since the last reform occurred. In total, 15 dummies were included, taking into account the duration effects of 15 years since the start or last event. All economic independent variables (trade and financial openness, unemployment, government balance and economic growth) have been lagged with one year. Country dummies were added to all models to account for unobserved country heterogeneity. Logit-regression analysis was applied separately for each of the reform types at varying levels of comprehensiveness, starting with the least comprehensive reform (1) and continuing to the most comprehensive type (5). This was done in order to test for the sensitivity of the models to the coding of reforms and cut-off points. Because the number of events of reform comprehensiveness 4 and 5 were so rare, these were excluded from the reporting of the findings. The likelihood of each reform type at the different comprehensiveness levels was analysed in two steps. In the first models (a), only the globalisation variables (+ time and country dummies) were entered to analyse the direct effects of trade and financial openness. In the second models (b), the economic, political and institutional control variables were entered. Findings Tables 2–5 show the results of the logit-regression analyses. Overall, models performed rather well, with Pearson’s Chi-squared (not 148 Aart-Jan Riekhoff Nicol Foulkes Savinetti & Aart-Jan Riekhoff (eds) reported) being significant at the p<0.05 level for most models apart from models for regulation 2a and 3a and deregulation 1a and 2a (significant only at a p<0.1 level), whereas deregulation model 2b was not significant even at a p<0.1 level. Hence, the models explaining reforms in EPL generally performed worse than those for UB reforms. Trade openness was found to have a significant negative effect on the likelihood of reforms aimed at UB expansion, while no effects of financial openness were found (Table 2). This suggests that UB expansion was more likely to take place when the economy was more closed to trade. This effect is consistently significant throughout all models. These findings are not so much in support of the efficiency hypothesis, but rather a rejection of the compensation hypothesis: when more exposed to foreign trade, there is less room for expanding labour market protection in the form of UB (although not necessarily leading to retrenchment). Moreover, UB expansion was found to be more likely to occur in times of fiscal surpluses (expansion models 1b and 2b); also, the significance of this effect does not hold for the more comprehensive type of reform (3b). Hence, with more resources in their budgets, governments have more possibilities to increase spending on UB. Labour market reforms in times of globalisation 149Shaping and re-shaping the boundaries of working life Table 2. Results logit-regression for reforms aimed at UB expansion Dependent variables Independent variables UB expansion Expansion 1a Expansion 1b Expansion 2a Expansion 2b Expansion 3a Expansion 3b t0.21 (0.03) -0.01 (0.03) -0.05 (0.04) -0.08* (0.04) -0.08 (0.09) -0.19* (0.10) Trade opennesst-1 -0.03* (0.02) -0.04** (0.02) -0.04* (0.02) -0.06** (0.03) -0.16*** (0.06) -0.19** (0.08) Financial opennesst-1 0.00 (0.02) 0.00 (0.02) 0.02 (0.02) 0.03 (0.02) 0.00 (0.03) 0.04 (0.03) Unemployment ratet-1 0.08 (0.05) 0.07 (0.07) -0.11 (0.19) Government balancet-1 0.18** (0.07) 0.19** (0.08) 0.11 (0.17) GDP growtht-1 -0.03 (0.08) -0.05 (0.09) -0.03 (0.87) Corporatism -0.12 (0.22) 0.12 (0.26) -0.95 (0.15) Left cabinet share -0.00 (0.00) 0.00 (0.01) 0.00 (0.01) Veto points 0.36 (0.48) 0.77 (0.78) 5.83* (3.29) N Number of events Log likelihood 392 92 379.25 392 92 368.20 392 62 283.59 392 62 272.19 392 20 103.83 392 20 93.32 Note: Indicated are coefficients (standard errors). 15 temporal dummies and 14 country dummies not reported. * p < 0.1, ** p < 0.05, *** p < 0.01 Table 3 shows that financial openness has consistently significant positive effects on the likelihood of UB retrenchment. It means that when capital markets are more deregulated, governments are more likely to retrench. This is in support of the efficiency hypothesis. Trade openness was not found to have any significant effects on UB retrenchment. Higher unemployment rates did increase the likelihood of UB retrenchment. Hence, in times of growing unemployment, governments will more often cut down on UB. The fiscal situation does not directly play a role here, unlike in the case of UB expansion. Moreover, a significant effect was found for the share of left cabinet seats in the case of the least comprehensive UB retrenchment reform type 156 Aart-Jan Riekhoff Nicol Foulkes Savinetti & Aart-Jan Riekhoff (eds) References Agell, J. (2002). On the determinants of labour market institutions: rentseeking vs. social insurance. German Economic Review, 3(2), 107–135. Allan, J.P., & Scruggs, L. (2004). Political partisanship and welfare state reform in advanced industrial societies. American Journal of Political Science, 48(3), 496–512. Allison, P.D. (1982). Discrete-time methods for the analysis of event histories. Sociological Methodology, 13, 61–98. Armingeon, K., Knöpfel, L., Weisstanner, D., & Engler, S. (2014). Comparative Political Data Set I 1960–2012. Bern: Institute of Political Science, University of Bern. Beck, N., Katz, J., & Tucker, R. (1998). Taking time seriously: Time-seriescross-section analysis with a binary dependent variable. American Journal of Political Science, 42(4), 1260–1288. Blanchard, O., Jaumotte, F., & Loungani, P. (2013). Labor market policies and IMF advice in advanced economies during the Great Recession. IMF Staff Discussion Note, SDN/13/02. Boeri, T., Conde-Ruiz, J.I., & Galasso, V. (2003). Protecting against labour market risk: employment protection or unemployment benefits? IZA Discussion Paper, No. 834. Bonoli, G. (2003). Social policy through labor markets. Understanding national differences in the provision of economic security to wage earners. Comparative Political Studies, 36(9), 1007–1030. Buchholz, S. Hofäcker, D., Mills, M., Blossfeld, H.-P., Kurz, K., & Hofmeister, H. (2009). Life courses in the globalization process: The development of social inequalities in modern societies. European Sociological Review, 25(1), 53–71. Elmelund-Præstekær, C., & Baggesen Klitgaard, M. (2012). Policy or institution? The political choice of retrenchment strategy. Journal of European Public Policy, 19(7), 1089–1107. Emmenegger, P. (2009). Barriers to entry: Insider/outsider politics and the political determinants of job security regulations. Journal of European Social Policy, 19(2), 131–146. Fischer, J. A., & Somogyi, F. (2009). Globalization and protection of employment. KOF Swiss Economic Institute Working Paper, No. 238. fRDB-IZA (2010). Social Reforms Database. Retrieved from http://www. frdb.org/topic/data-sources/doc_pk/9027. Labour market reforms in times of globalisation 157Shaping and re-shaping the boundaries of working life Gaston, N., & Nelson, D. (2004). Structural change and the labor-market effects of globalization. Review of International Economics, 12(5), 769–792. Gaston, N., & Rajaguru, G. (2008). The rise (and fall) of labour market programmes: Domestic vs. global factors. Oxford Economic Papers, 60(4), 619–648. Genschel, P. (2004). Globalization and the welfare state: A retrospective. Journal of European Public Policy, 11(4), 613–636. Green-Pedersen, C. (2007). More than data questions and methodological issues: Theoretical conceptualization and the dependent variable ‘problem’ in the study of welfare reform. In J. Clasen, & N. Siegel (Eds), Investigating Welfare State Change: The ‘Dependent Variable Problem’ in Comparative Analysis (pp. 13–23). Cheltenham and Northampton: Edward Elgar. Hicks, A., & Zorn, C. (2005). Economic globalization, the macro economy, and reversals of welfare: Expansion in affluent democracies, 1978– 94. International Organization, 59(3), 631–662. Jensen, C., Knill, C., Schulze, K., & Tosun, J. (2014). Giving less by doing more? Dynamics of social policy expansion and dismantling in 18 OECD countries. Journal of European Public Policy, 21(4), 528–548. Kittel, B., & Winner, H. (2005). How reliable is pooled analysis in political economy? The globalization-welfare state nexus revisited. European Journal of Political Research, 44(2), 269–293. Koster, F. (2009). The welfare state and globalisation: Down and out or too tough to die? International Journal of Social Welfare, 18(2), 153–162. Meinhard, S., & Potrafke, N. (2012). The globalization-welfare state nexus reconsidered. Review of International Economics, 20(2), 271–287. OECD (2013). Protecting jobs, enhancing flexibility: A new look at employment protection legislation. In OECD Employment Outlook 2013. Paris: OECD Publishing. Potrafke, N. (2010). Labor market deregulation and globalization: Empirical evidence from OECD countries. Review of World Economics, 146(3), 545–571. Potrafke, N. (2013). Globalization and labor market institutions: International empirical evidence. Journal of Comparative Economics, 41(3), 829–842. Quinn, D., & Inclan, C. (1997). The origins of financial openness: A study of current and capital account liberalization. American Journal of Political Science, 41(3), 771–813. 158 Aart-Jan Riekhoff Nicol Foulkes Savinetti & Aart-Jan Riekhoff (eds) Riekhoff, A.J. (2015). Globalisaatio ja työmarkkinauudistukset Euroopassa: kompensaatioja tehokkuushypoteesit uudelleenarvioinnissa (Globalisation and labour market reforms in Europe: The compensation and efficiency hypotheses revisited). Työelämän Tutkimus, 13(3), 174–192. Riekhoff, A.J. (2017). The divergent logics of labour market reforms in times of permanent austerity. International Journal of Social Welfare, 26(1), 5–20. Rodrik, D. (1998). Why do more open economies have larger governments? Journal of Political Economy, 106(5), 997–1032. Saint-Paul, G. (1996). Exploring the political economy of labour market institutions. Economic Policy, 11(23), 263–315. Saint-Paul, G. (2002). The political economy of employment protection. Journal of Political Economy, 110(3), 672–704. Swank, D. (2002). Global capital, political institutions, and policy change in developed welfare states. Cambridge: Cambridge University Press. Swank, D. (2005). Globalisation, domestic politics, and welfare state retrenchment in capitalist democracies. Social Policy and Society, 4(2), 183–195. Van Vliet, O., & Caminada, K. (2012). Unemployment replacement rates dataset among 34 welfare states, 1971–2009: an update, extension and modification of the Scruggs’ welfare state entitlements dataset. Visser, J. (2011). Database on institutional characteristics of trade unions, wage setting, state intervention and social pacts, 1960–2010 (ICTWSS), version 3.0. Amsterdam Institute for Advanced Labour Studies AIAS. Retrieved from http://www.uva-aias.net/208. Wenzelburger, G., Zohlnhöfer, R., & Wolf, F. (2013). Implications of dataset choice in comparative welfare state research. Journal of European Public Policy, 20(9), 1229–1250.