Abstract
This article asks the question of why some social policies can be path dependent, while others are not, even if the country goes through what clearly seems to be a ‘critical juncture’ caused by exogenous shocks. We argue that in the Czech Republic labour market policies represent a clear break with the past, while healthcare and family policies have been path dependent to various degrees. There are several reasons. First, during the first years of the transition policymakers gave greater priority to labour market issues. Second, labour market policies were less constrained than the other policy areas, both because the government had to create new institutions in this area rather than rely on old ones and because these new institutions had not yet built up popular support as the old institutions had. Third, labour policy was not as influenced by policy legacies.
Keywords
Introduction
The collapse of communism represents one of the most dramatic events of the last century: governments fell; political prisoners were released from confinement; state enterprises were privatized and the command economy was dismantled; and multiparty competition and free elections replaced elaborate systems of police surveillance and control of citizens. The idea of ‘exogenous shocks’ has played a major role in theorizing about historical institutionalism and path dependency (Katznelson, 2003; Pierson, 2000) and surely the collapse of communism presents one of the greatest exogenous shocks that European countries have experienced in recent decades. According to mainstream historical–institutional theory, exogenous shocks give rise to ‘critical junctures’ (Collier and Collier, 1991) that allow policymakers to introduce far-reaching changes that set countries down different trajectories of development, after which they are likely to continue following the same path. Yet, while such insights have greatly enhanced our understanding of why different countries pursue different policies, this approach by itself has trouble explaining why the same exogenous shock (the collapse of communist-led regimes) has sometimes influenced policy differently, even within the same country.
As the Czech case shows, despite the radical exogenous shocks that came about with system collapse, in most social policy areas the changes that took place were much more limited than the path-dependency approach expects. This raises the question of why most social policies – with some modifications – continued down previous paths, while other policies in the same country, in the same period, became more sensitive to exogenous shocks and changed paths. This article compares three cases of how Czech social policy has developed. Peter Hall (1993) divides policy changes into three levels: first-order change, in which the instruments change (for example, changing benefit levels or benefits periods), but the policy structures and goals remain the same; second-order change, in which the structure and the instruments change (for example, changing the types of policy instruments used), but the goals remain the same; and third-order change, in which the goals themselves change.
As we explain below, to a large extent, Czech family policy basically represents a case of first-order change with much stronger path dependency, and it corresponds closely to the continental conservative model. Meanwhile, healthcare basically represents a case of second-order change and less path dependency than family policies. Finally, labour market policies represent a case of third-order change, in which a new path emerges that moves in a market liberal direction. Thus, these policies make for interesting comparisons, as they show how in the same period within the same country, three different policy areas can undergo three different types of change.
Although we use Hall’s typology, our article shows that, in contrast to Hall’s model, much more is needed than a change in the goals of policymakers to bring about third-order change. It is generally acknowledged that laissez-faire, market liberal ideology, as supported by international organizations and Washington, reached near-hegemonic status among Czech policymakers (Potůček, 2004; Weiner, 2007). Thus, the question arises: given this free market orientation on the part of policymakers, why did they only dare to carry out third-order, systemic changes in the area of labour market policy? Neoliberal ideas have been strong in all of the post-communist Central European countries, but the Czech Republic makes for an especially interesting case because it does not have any strong forces (such as the Catholic Church in Poland or large scale nationalist movements as in Hungary and Slovakia) to counterbalance these neoliberal ideas. Thus, at the ideational level, we would expect the Czech Republic to have the most market liberal policies in these areas – and yet, as we show, only the labour market reforms in that country went in a clear market liberal direction.
Explaining institutional changes
This article basically agrees with Cerami’s (2009: 46) claim that incremental transformative change took place after 1989, rather than path divergence. As he notes, after the fall of the communist regimes the new governments developed policies with interwar ‘Bismarckian characteristics already present during communism’. Besides the usual problems of path divergence, such as those caused by the entrenched interests of bureaucracies and increased costs associated with going against these bureaucracies, other problems existed, such as the lack of public support for radical welfare reforms and the lack of knowledge about how to reshape social policies (Inglot, 2008). Nevertheless, in the Czech Republic abrupt change did in fact take place in some selected areas, such as in labour market policy. Thus, we still need an explanation as to why some policy areas became relatively resistant to change, while others were able to shift paths or to go down a new path.
According to Hall (1993), the main reason policymakers make changes in policies and introduce third-order changes is because of social learning. Policymakers conclude that previous policy instruments no longer work and new methods are needed, and new goals as well. Similarly, in the view of discursive institutionalists, when politicians make decisions that change path trajectories, their decisions are influenced by the ideas that they hold, as well as the ideas of groups pushing for these changes (Schmidt, 2002, 2010). Discursive institutional approaches help explain why policymakers tried to implement market liberal policies when given the greatest amount of space for enacting reforms (as in the case of labour market policies), but they cannot explain the failure to implement market liberal reforms in the majority of social policy areas.
One reason why ‘insurrectionary’ policymakers (Mahoney and Thelen, 2010) were only able to bring about a clear path change in the area of labour market policy is that it was not possible for them to change everything overnight, so they had to give some policy areas priority over others. Offe (1994) observes that post-communist policymakers first went through an ‘early stage of emergency measures’, in which they had to quickly create unemployment insurance and some kind of social safety net, as many people would surely lose their jobs, while other weak groups (such as pensioners) would have trouble getting by on their incomes. Thus, in the first post-communist years, they decided to give priority to policies that established a functioning labour market, while introducing policies (such as unemployment benefits) that would pacify the population and prevent the unemployed or those afraid of becoming unemployed from revolting (Vanhuysse, 2006).
Even if it made logical sense for policymakers to give higher priority to labour market policies than to family policies or health policies, this by itself cannot explain why they did not push for stronger market liberal policies in these latter areas when they actually did introduce reforms. An important reason for favouring conservative over liberal policies comes from the country’s Bismarckian policy legacy. Policymakers often develop a tradition of looking at policies from a certain perspective or dealing with issues in a certain manner (Weir and Skocpol, 1985).
Finally, Pierson’s (2000) discussion about increasing returns leading to vested interests plays some role here. Since family policy institutions and health institutions already existed, so did groups of professionals with vested interests in these organizations. No such vested interests existed for labour market policies. In addition – as the literature on retrenchment has pointed out – universalist or generous programmes build up popular support. As a result, radical changes in such programmes face greater popular resistance (Bonoli and Taylor-Gooby, 2000; Pierson, 2000). Again, since labour market policies for job retraining and unemployment compensation did not exist under communist rule, constituencies supporting policies in these fields did not emerge either. 1
This article describes family policies, healthcare policies and labour market policies in the Czech Republic, in the process showing how ideas, policy legacies and the strength of institutional structures all combine to lead to different degrees of path dependency or to the emergence of a new path. We use ISSP data to analyse how institutional developments interact with public opinion, and to compare the Czech Republic with other countries. In addition, we complement our historical institutional account with references to interviews that we carried out with 10 policymakers in the field of health policy (including former ministers, former deputy ministers and advisors). We analyse why health policy did not go in a more market liberal direction and break with past patterns as labour market policy did, despite the fact that policymakers generally would have preferred market liberal policies. This example shows clearly why social learning among policymakers is not enough to induce third-order change: policymakers must also believe that they can gain some amount of popular support for their policy preferences before they dare to introduce radical changes.
First-order change: family policy
Some authors have written about the ‘refamilialization’ of post-communist family policies (Hantrais, 2004). This implies that communist era policies were not previously ‘familialized’, and that they first became so after the fall of communism. This suggests that, under communist rule, the state played the main care-giving role, but that the state reneged after 1989 and returned responsibility to the family (which in practice mostly means the mother). This is partly true, as the state has closed down most of the nursery schools for children under 3 years old in the Czech Republic. Yet the percentage of pre-school children aged 3–5 years attending kindergartens has actually increased slightly, from 78.9 percent in 1989 to 79.1 percent in 2007–2008. 2
The parental leave scheme has also remained substantially unchanged in the Czech Republic (Saxonberg, 2013). As under communist rule, mothers in the Czech Republic enjoy a 6-month insurance-based maternity leave, followed by a flat rate ‘extended leave’ (which in 2002 officially became a ‘parental leave’). Successive Czech governments have introduced some adjustments, but these represent typical first-order changes, in that they have only modified the instruments of policy, but did not change their structures or goals. For example, in 1991, the extended leave benefit was made available to men. In 1995 an extra year was added, so that parents can now stay at home for 4 years; however, parents are only guaranteed their job if they return to it within 3 years. Other adjustments have included the doubling of the parental leave benefit level in 2006 and the introduction of a more complicated three-track system in 2007. Parents can choose a fast track, in which they stay at home for 2 years and receive more money per month; a regular track, whereby they stay at home for 3 years and receive around the same amount as before; or a slow track, whereby they stay at home for 4 years and receive less money per month (Saxonberg, 2013).
In theory, this three-track reform might have made it easier for mothers to return to work after 2 years; thus, it might have signified a slight change in policy goals as it gives some incentive for mothers to return to work after 2 years, since their parental leave benefit per month would be higher than before (although the total amount of benefits that they would receive over the entire leave period would be the same as if they stayed at home for 3 years). However, in practice in did not imply a change in policy goals, because the government did nothing to promote increased access to daycare for children under 3 years old. Thus, the minister responsible for the reform admitted that he did not expect many mothers to take advantage of this reform and to shorten their leave to 2 years (Čápová, 2008). Given the fact that all of the reforms have used the same instruments and basically kept the main policy goals intact, they clearly represent what Hall would consider to be first-order changes.
Communist era and post-communist policies in the area of parental leave in today’s Czech Republic basically follow the conservative model. As Esping-Andersen (1990) notes, conservative corporatist welfare policies aim to protect social hierarchies and maintain traditional gender roles within the family. By contrast, parental leave policies based on the income replacement principle (as in Scandinavia and more recently in Germany) help break down traditional gender roles, because they give fathers an economic incentive to share in the leave time. This is because families no longer face a large drop in income if the father stays in home with the children (given that fathers usually have higher incomes than mothers). As Bussemaker and van Kersbergen (1999) observe, conservative countries tend to have maternity leave followed by parental leave with flat rate benefits. This provides fathers with such low levels of benefits that they have little incentive to take a share of the parental leave.
One reason the Czech Republic has continued its conservative family policies is that the long leave is popular among the population. Moves to introduce means-tested leave would therefore likely have led to widespread protests. As Haney (1997) shows, protests arose when the socialist-led Hungarian government tried to abolish its parental leave scheme 3 and to replace it with a means-tested one. The socialist government lost the elections and the new government immediately reinstated the popular parental leave scheme. No other governments have dared tamper with the basic system since then.
In the Czech Republic, surveys indicate that the parental leave system is popular, and that Czechs prefer receiving parental leave benefits to having the state provide daycare facilities for children under 3 years old. In a recent survey, 80 percent of Czech mothers with a child under the age of 6 years stated that, in the ideal case, the mother should stay at home on a full time basis until their youngest child is 2 to 3 years old (Dudová and Hašková, 2011). Support for paid leave remains extremely high throughout Central Europe. The latest ISSP survey on family issues from 2002 shows that well over 90 percent of the population in the Czech Republic, Hungary, Poland and Slovakia believes that working mothers should have the right to paid maternity leave. 4 With support that high, no post-communist government (except the Hungarian one in 1995) has dared suggest systemic changes in parental leave.
As already noted, the percentage of children attending kindergartens has remained rather stable since the communist era. The big change has taken place in the area of childcare for children under 3 years old, as most nurseries have been closed down. Kindergartens are popular because they have been part of the Czech nationalist project since the mid 1800s. The original kindergartens in the Austrian Empire taught in German, charged fees and were only open 4–5 hours per day; thus, they catered mainly to the middle class (Mišurcová, 1980). However, ‘Volkskindergärten’ emerged during that century. These taught in the local language and combined Fröbelian pedagogy with long opening hours, so that poor mothers could work (Fellner, 1884). Czech nationalists established an association, Matice česká, which promoted and financed Czech Volkskindergärten, so that their children would learn Czech (Mišurcová, 1980). Since the Volkskindergärten in the Czech lands taught in Czech, they became much more popular and widespread.
This connection of Volkskindergärten with nationalism also helps to explain the fact that, while post-communist governments with conservative views on gender roles let most nurseries close, they have kept most of the kindergartens open. Nurseries tend to be seen as a ‘communist’ idea, since the communists radically expanded them; by contrast, kindergartens still have a positive connotation, both among the population in general and among conservative groups in particular. One survey from the year 2000 found that 70 percent of parents agreed that Czech ‘kindergartens provide good care to all children of pre-school age’ (Ministry of Education, Youth and Sports of the Czech Republic, 2000: 105). Closing kindergartens, therefore, would have provoked much greater opposition than closing nurseries.
The question remains as to why nurseries were not as popular as kindergartens. First, while nurseries have existed in the Czech lands since the middle of the 19th century, they were not very common until the communists came to power. While 30 percent of children aged 3–5 years old attended kindergartens in 1946, only 1 percent of children aged 0–2 years old attended nurseries in 1948 (Saxonberg, 2014). Despite the small number of nurseries, the communist regime rapidly expanded them as part of its policy of ‘forced modernization’. 5 As a consequence, the number of nurseries in the Czech part of Czechoslovakia increased from 83 in 1937 to 1,330 by 1978 (Český statistický úřad, 1979: 1). Since the expansion of nurseries was so rapid under communist rule, the population came to associate them with communism.
A second reason why nurseries became less popular than kindergartens is that, when the communist regime closed down the Ministry of Social Caring in 1951 (Schiller, 1971), it moved responsibility for nurseries to the Ministry of Health, thereby turning early pre-school care into a health matter. 6 Consequently, nurses (with only secondary education) rather than university-trained teachers took care of the children. Rather than having pedagogical goals for developing the children, the main goal for policymakers was to make it easier for mothers to work. From this point of view, nurseries were more ‘efficient’ when the ratio of children per nursery was rather high. In Czechoslovakia under communist rule, nurseries had groups of at least 20 children, and there were about six children per nurse and 20 children per childminder. 7 In Sweden, by contrast, where citizens are the most satisfied in the EU with the quality of their daycare (according to the Eurobarometer, report no. 321), there are about four children per teacher, and the groups each have about 12 children (Andersson, 1989: 858). The Swedish system also relies mostly on university-educated teachers, rather than on nurses or childminders. In Denmark, which has the highest percentage in the world of children under 3 years old attending daycare, the ratio of children to teachers is 3:1 (Cooke and Henehan, 2012: 41).
Paradoxically, even though the main goal of the communist era nurseries was to prevent children from becoming sick so their mothers could work, because of the problem of overcrowding, it became easier for illnesses to spread among the children. Paediatricians complained, for example, that children under 18 months of age who attended nurseries became sick three times to seven times more often than those staying at home (Dunovský, 1971: 154).
Thus, nurseries gained a rather poor reputation among the public. For example, a survey taken in 1956 in Czechoslovakia showed that only one-third of Czech and Slovak women in gainful employment wanted to place their children in a nursery. The authors of the survey concluded that one of the reasons for the unpopularity of nurseries was that they were frequently closed due to the outbreak of contagious diseases (Srb and Kučera, 1959: 115–120).
Since mothers often had to stay at home with their sick children, it made economic sense from the standpoint of the communist regime to introduce an extended maternity leave rather than to increase support for nurseries. In order to reduce reliance on nurseries, then, the communist regime introduced the extended leave in 1964, and successively increased it to 3 years (Klíma, 1969). Then, since the 3-year maternity leave had already become the norm in Czech society, little opposition existed to the closing of nurseries when the communist regime fell. The basic institutions of Czech childcare policy – 3-year parental leave plus widespread access to kindergartens – remained largely intact. Rather than changing paths in a ‘refamilializing’ direction, the country simply continued down the more conservative path that had already been developing under communist rule.
Second-order change: healthcare
Healthcare policies basically moved in a conservative direction too. According to Hassenteufel and Palier (2008), the basic package of Bismarckian policies in this area comprises a ‘health insurance system’, which they contrast to the ‘national health system’. In a Bismarckian health insurance system, most hospitals are public, but most other providers (such as general practitioners, some health clinics, etc.) are private. Price competition, however, is rather limited. Competing health insurance funds pay the main healthcare expenses and are financed by social contributions.
After the Second World War, the coalition government introduced a universalist system of national health insurance in Czechoslovakia that provided the same level of coverage to all groups. However, this system only lasted for a brief time. When the Communist Party installed a one-party dictatorship, it reintroduced some Bismarckian elements. As with the universalist model, all healthcare was now publicly run and officially free (even if patients were known to pay bribes at times), and all sickness benefits were paid out of the state budget. On the other hand, the communist regime reintroduced the country’s Bismarckian tradition of providing different levels of benefits for different occupational groups (Matějček, 1973). It remains an open question whether the main motivation for this change derived from the pre-war policy legacy, or from pressure from the Soviet Union, or from a combination of the two.
Healthcare policies in the Czech Republic resemble family policies in having a relatively strong institutional legacy that makes it difficult to diverge radically from the established path. As in the area of family policy, physical buildings remain from the communist and pre-communist era (but they are hospitals rather than nurseries and kindergartens). In the area of labour market policy, by contrast, buildings – for housing unemployment offices, job retraining centres, unemployment insurance organizations, and the like – did not exist when the communist regime fell. In both policy areas, moreover, the communists could not simply start from scratch: they had to build on pre-existing institutions. The same inertia could then be seen at a later point – when the communists lost power and market liberals took over. Thus, while post-communist policymakers have had market liberal preferences, when the communist regime fell in 1989, they faced a situation in which all providers of healthcare were public.
Although it was relatively easy to privatize healthcare centres and allow doctors to start their own private practices, it was much more difficult to privatize hospitals. Doctors can simply set up a private practice by their own initiative, but starting a new private hospital is an expensive proposition. Few people in the Czech Republic have the funds needed to start their own hospital or to buy a state-run one; furthermore, healthcare is not a field in which it is easy to attract foreign investors. No government can afford to let inefficient hospitals – unlike industrial enterprises – simply go bankrupt and close down when private buyers are lacking; doing so would have the consequence of making healthcare unavailable for most of the population. Notwithstanding some ambitious plans to privatize hospitals, then, private hospitals only accounted for 31.8 percent of the total number of hospitals in 2002, and 10.6 percent of the total number of beds (Rokosová and Háva, 2005: 54). In contrast, already by the end of 1993, about 90 percent of general practitioners and some 70 percent of paediatricians and adolescent care physicians had private practices (Vepřek et al., 1995: 57). This mixture of private doctors and mostly public hospitals is similar to that in many other conservative continental healthcare systems – and to the system in interwar Czechoslovakia as well – in which a variety of public, private and semi-public hospitals exist side by side (cf. Mášova, 2005; Tůma, 1936). West Germany supplies a striking comparison here. In that country in 1989 (the year the communist regimes collapsed), 49.8 percent of hospitals were publicly run, while only 15.8 percent were private and the remainder were run by non-profit organizations (Hoffmeyer, 1994: 506).
Health insurance in the Czech Republic also approximates the Bismarckian model, although it is more universal in the sense that the same rules apply to all professional groups and everyone must be a part of the system. Hall places great emphasis on the changing attitudes of policymakers in his model of policy change. Yet, it is clear from our interviews with policymakers – including several former ministers of health – that they would have preferred third-order change in a more market liberal direction (although not as market oriented as the American model); however, they did not believe they could get popular support for such measures. For example, a former head of the General Insurance Fund (the state health insurance scheme) complained that Czechs were accustomed to receiving healthcare services for free. 8 Even on the issue of having patients pay the costs themselves up to a certain level, versus having the insurance system pay the entirety of the costs from the beginning, policymakers caved in to the argument that it was ‘not possible’ in the Czech context to expect people to pay anything. 9 As a former deputy minister of health acknowledges, they did not make their reforms in a vacuum: they were aware of the country’s Bismarckian tradition. 10
Despite their own preference for a more market-oriented system, then, policymakers found it wise to avoid risking the wrath of the population, which had been accustomed to free healthcare under communist rule. For example, the 1996 ISSP survey on the role of government shows that 96.8 percent of the population agreed or strongly agreed that it ‘should be the government’s responsibility to provide health care for the sick’. This was actually 0.2 percent lower than the EU average among the 15 countries that took part in the survey. 11 Another 82.7 percent thought the government should spend more money on healthcare (compared with 79.5 percent for the EU average). The vast majority of Europeans, quite clearly, are favourable to government intervention in the healthcare sector. In contrast, support for newly created programmes, such as unemployment insurance, was much lower. Only 19.7 percent of Czechs thought the government should spend ‘much more’ or ‘more’ on unemployment benefits, which was much lower than the EU average of 41.6 percent. The fact that the average for all of Europe is much lower when it comes to support for unemployment benefits than for healthcare or maternity leave reflects the general stigma attached to being unemployed: Europeans are more likely to blame people for becoming unemployed than for getting sick, while getting pregnant and having children is considered something positive and desirable. Nevertheless, the fact that support for unemployment benefits was much lower in the Czech Republic than in the rest of Europe – and even much lower than in the post-communist EU countries, which average 41.1 percent – shows that market liberal beliefs express themselves among the Czech population much more strongly in the area of labour market policy than in that of healthcare and family policy. The government would have probably faced mass protests if it had introduced an unregulated private system that left a large portion of the population without health insurance. In an interview in 2011, accordingly, a former minister for health, Luděk Rubáš, admitted that the American healthcare model ‘is not applicable’ to Czech conditions.
Even countries with generally liberal social policies and governments committed to welfare retrenchment have trouble making third-order changes in national healthcare programmes, because of the popular support that such programmes enjoy. As Oliver and Mossialos (2005) note, ‘even Margaret Thatcher – at the height of her political powers in the late 1980s – was reluctant to challenge the widespread public support for the fundamental financing tenet of the NHS’.
In addition to facing pressure from a population accustomed to affordable healthcare, the Czech government also faced pressure from doctors. Kamil Kalina, onetime Deputy Minister of Health, claims that Czech doctors had observed that countries with the Bismarckian health insurance model, such as Germany and Austria, were also the countries where doctors were best off and had the latest medical technologies. 12
The country went back, then, to its pre-war Bismarckian roots in the area of health insurance. 13 The system it introduced has some similarities to the German model as it existed in the early 1990s – that is, to the German model as it was at the time when the Czech Republic introduced its reforms. Health insurance in Germany is mandatory, and although private companies exist, the state-run insurance company dominates the market. However, the Czech system is more statist than the German one, which may be attributable in part to its much more statist starting point (everything having been state-run under communist rule.) In Germany, regional insurance funds (AOK) have the greatest market share, but this has declined from about 40 percent of the population in the 1990s to about 34 percent today. 14 In the Czech Republic, one centralized state fund completely dominates the market, with over 6.5 million members in a country of slightly more than 10 million citizens (Kulatý stůl, 2007: 13). In 2002, the state-run health insurance scheme still accounted for 68 percent of the market, meaning that less than a third of the population chose an alternative (Rokosová and Háva, 2005: 30). As in the German system, moreover, the competing health insurance funds are not allowed to make a profit (Rokosová and Háva, 2005: 31). Here again, however, the Czech system is more statist, as the German system still allows people with income over a certain level to opt out of the system and to chose a private, profit-making insurance company instead; the Czech system does not offer such an option. Nevertheless, about 92 percent of the population in Germany belonged to the non-profit, statutory health insurance system in 2004 (Altenstetter and Busse, 2005: 126). But while the Czech system is somewhat more statist than the German one, it is still much less statist than either the Polish or Hungarian one. Hungary only offers one state insurance fund (Gaál et al., 2011), and Poland has a single state-run national health fund with 16 regional branches (Sagan and Panteli, 2011).
One reason why the state insurance fund dominates in the Czech Republic is that non-state insurance schemes have had difficulty staying afloat. Many private health insurance funds have gone bankrupt. Another reason is that some non-state schemes have had trouble getting parliamentary approval, due to their unwillingness to insure high-risk people (Bošková, 1996: 1–2). As a result, the number of insurance funds has declined from 27 in the early 1990s to nine today. Finally, as in the German system, citizens can ‘top up’ the non-profit health insurance with a private, for-profit arrangement (Act 277/2009).
However, notwithstanding the Bismarckian structure of the health insurance system, neoliberal discourse eventually made an impact: recent reforms are causing the healthcare system to drift (or ‘decay’ – see Saxonberg and Sirovátka, 2009) in a more market liberal direction. This is especially true in regard to health insurance and benefits for care of the long term ill. Since 2008, patients have been forced to pay a modest fee for each visit to the doctor, for prescribed drugs and for hospital stays. Of course, many social democratic and conservative countries have introduced such reforms also. Nevertheless, the Czech reforms are ‘liberal’ in the sense that they are based on liberal arguments concerning the need for patients to pay part of the costs (besides many social democratic and conservative governments in Western Europe have been influenced by the neoliberal discourse too).
Another recent reform, which took effect in 2009, denies health insurance benefits during the first three days of illness. 15 Many European countries have waiting periods today: i.e. citizens do not receive sickness benefits until a certain number of days have passed since they first fell ill. However, the exact number of waiting days varies greatly between countries. Some conservative countries, such as Germany, do not have any waiting days at all. Conservative Belgium has a single waiting day, while conservative Austria has no waiting day for the 6 weeks during which the employer pays. Meanwhile, France has a 3-day waiting period (Heymann et al., 2009: 7). Given the wide variety of models to choose from, and given the fact that most ministry officials with whom we spoke showed awareness of their country’s pre-war past, it is not surprising that the government chose to make the number of waiting days exactly as the same as in interwar Czechoslovakia (Deyl, 1985: 84).
Care for the long-term ill (i.e. those who are permanently disabled) has also drifted in a more market-oriented direction recently. Previously, those who cared for the long-term ill received a flat rate benefit. As in the case of policies for the care of children, flat rate benefits give little incentive for fathers to become caregivers for the long-term ill, because in most families the father has the highest income. In 2006, a social democratic-led coalition government took away the flat rate benefit for those who do the caring and replaced it with one going directly to the disabled person. 16 The reform was based on the neoliberal idea that people have greater freedom of choice if they receive cash benefits rather than services. The argument is that the long-term ill can now decide how to spend the money: they can keep it themselves, they can pay family members, they can hire professional caregivers or they can pay for institutional care. In practice, we can expect them often to choose to keep the money themselves, and to expect family members (i.e. their daughter or wife) to take care of them for free. Thus, the liberal drift puts more pressure on women to return to the home. Interestingly, party politics do not seem to have mattered much here: it was a social democratic-led government, rather than a right-wing one, which introduced this reform (which the centre-right parties, unsurprisingly, did not oppose). Moreover, rather than meeting opposition from non-government organizations (NGOs), this reform was actually proposed by one: the National Council of Disabled (Hutař and Krása, 2006).
The widespread support among political parties and NGOs for liberal reforms shows how strong the neoliberal discourse has been. However, the fact that it took nearly two decades since the fall of communism to introduce such reforms indicates how difficult it has been for post-communist policymakers to break with the past.
Third-order change: labour market policies
As we have seen, then, family policies show a high degree of path dependency and healthcare policies a moderate degree of path dependency. In contrast, labour market policies mark a clear break with the communist era. Policymakers have embarked on a new path and have instituted third-order change. They have enjoyed freer rein in this area, because the main institutions for labour market policy did not exist in 1989. The country had no unemployment insurance, no employment agencies and no organizations for carrying out job retraining programmes (since unemployment officially did not exist under communist rule). Not only did policymakers have no institutions to build upon, they did not even have much of a policy legacy to look back to for guidance. In the interwar period as well, as Inglot (2008: 65) observes, Czechoslovakia failed to introduce any comprehensive, countrywide unemployment insurance. As Offe (1994) notes, furthermore, enacting policies such as unemployment insurance became a top priority for post-communist policymakers.
In the field of labour market policy, we find not only a more radical shift towards a liberal model, but also quite frequent changes in policy design. During the first stage of transformation (1990–1992), before Czechoslovakia split, policymakers considered the creation of a ‘social safety net’ (employment policy, unemployment protection and social assistance) to be the key issue due to fears about a possible explosion of unemployment. Unemployment insurance was introduced in 1990, and codified in February 1991 (Employment Act 1/1991 Coll.). Since the original anti-communist coalition government had a social liberal orientation toward social policy (Orenstein, 1995), it introduced a relatively generous benefit with a replacement rate of 60 percent of previous income and a special rate during the first 6 months (90 percent of previous income without a ceiling) paid to employees made redundant for ‘structural reasons’, i.e. as a result of mass layoffs (Decree 195/1989 of the Federal Ministry of Labour and Social Affairs).
Eventually, market liberals around the then finance minister, Václav Klaus, started gaining in influence. They could point to rising costs for unemployment benefits, as unemployment levels were rising. They succeeded in making the benefit more residualist by reducing the special rate to 65 percent in August 1991, and limiting the maximum level to 1.5 times the minimum wage. The following year they shortened the benefit period from 12 to 6 months. They also lowered the benefit level to 60 percent of previous net income for the first 3 months, and 50 percent for the next 3 months.
In 1997, after Czechoslovakia had split and a more ideologically free-market government had taken power under Klaus – who by then had become prime minister – the Czech government reduced the replacement rate still further: to 50 percent for the initial period and 40 percent for the next (Sirovátka and Hora, 2011). The unemployment benefit was now plainly structured in accordance with the more residualist, liberal model: replacement rates were low, the ceiling was low (some 55 percent of the average wage), and the benefit period was uniform and extremely short.
While health and family policies show considerable continuity across different governments, we see greater changes in the area of unemployment insurance. The social democrats came to power in 1998, and 1 year later they raised the ceiling for unemployment benefit from 1.5 to 2.5 times the subsistence level (and 1.8 to 2.8 times the subsistence level for participants in active labour market policies). When the Czech Republic joined the EU in 2004, moreover, a new social democratic-led coalition government increased the benefit rate during the second 3 months of unemployment – from 40 to 45 percent of net income – in order to meet the minimum demands of ILO treaty no. 168 (the new Employment Act 435/2004 Coll. in October 2004).
When the centre-right returned to power in 2007, it pushed the unemployment insurance system further in a liberal direction, by shortening the benefit period by 1 month (although it increased the replacement rate for the first 2 months – to 65 percent). 17 It also introduced workfare measures. Starting in 2008, persons unemployed for longer than 12 months could only receive a level of social assistance corresponding to the existence minimum, which is about 50 percent lower than the official minimum subsistence level, and they could only receive such benefits if they participated in public service activities for at least 20 hours a month. In 2011, the liberal government introduced further measures conforming to the liberal ‘work first’ strategy: after 2 months, unemployed persons may be required to take part in public service activities for at least 20 hours per week in order to earn the right to unemployment benefit. 18 Similar policies were introduced in the UK in the mid 1990s, with the Job Seeker’s Allowance. In general, Czech policies come relatively close to those applied in the market liberal UK: that is, replacement rates are low, the ceiling is very low, and benefits are of brief duration (see Table 1).
Unemployment benefits in 2008.
Source: EC 2012: MISSOC (Mutual Information System on Social Protection in the EU).
Job retraining programmes usually constitute the other main pillar of labour market policy, alongside unemployment benefits. Both left-leaning and right-leaning governments in the Czech Republic have consistently pursued market liberal policies in this area. Thus, in 1999 – 1 year after the social democrats had come to power – expenditures on active employment policy accounted for just 0.2 percent of GDP in the Czech Republic, while only 1.3 percent of the labour force was participating in active labour market schemes. As Table 2 shows, this is not only much lower than in social democratic Sweden and conservative Germany, it is also much lower than in the liberal UK and the other post-communist Central European countries.
Active labour market policy measures – expenditure as percentage of GDP and participants as percentage of labour force, in 1999 and 2008.
Source: Organisation for Economic Co-operation and Development, 2002, 2010. n.d., not documented.
Thus, in the area of labour market policy, Czech policymakers succeeded in introducing a market liberal system that clearly represents third-order change. In the areas of healthcare and family policy, conversely, they only made adjustments in their instruments or settings, thus achieving no more than first- or second-order change.
Conclusion
Scholars of historical institutionalism have recently made great strides in developing theories of path dependency further, but they have tended to ignore the question of why some policies are more path dependent than others. This is particularly pertinent in cases where we would expect a large exogenous shock – such as the collapse of a dictatorship or the break-up of a command economy – to lead to a critical juncture. Using the example of the Czech Republic, we have argued that, notwithstanding the dominant neoliberal discourse, family policies and healthcare policies have displayed various degrees of path dependency, and have mainly moved in a rather conservative direction. Labour market policies, in contrast, really do represent a break with the past.
It was much more difficult to make a break in family policy and healthcare policy, due to institutional and policy legacies in these areas dating back more than a century. Parents were accustomed to being able to take a 3-year parental leave and to send their children to kindergartens after their 3-year leave. Both of these institutions enjoy strong popular support. Similarly, citizens were accustomed to free and universal healthcare, making a privately based health insurance system difficult to implement. Thus, while policymakers would have preferred a more market-oriented approach, the political costs of implementing such policies would have been quite high. However, since unemployment benefit and active labour market institutions did not exist under communist rule, popular support for non-liberal policies in this area was lacking, giving policymakers a relatively free hand.
It was also more difficult in physical terms to make radical changes in family policy and healthcare policy. It would have been difficult to privatize kindergartens and hospitals, as the private sector has shown little interest in investing in these areas. Thus, the state was only able to privatize a few hospitals, although the Ministry of Health would have liked to privatize many more.
Finally, healthcare and family policies were not as vulnerable to the ‘window of opportunity’ that existed during the first years of the transition to democracy. Governments gave a higher priority to labour market issues, as they believed new measures in this area – such as a social safety net and unemployment insurance – to be necessary for a smooth transition to a market economy. And since family and healthcare policies were already in place, the first postcommunist governments did not give reform in these areas as high priority.
Footnotes
Funding
This research received funding from Czech Grant Agency (GAP404/10/1586) and Riksbankenens Jubileumsfond (The Swedish Foundation for Humanities and Social Sciences) (P09-0413:1-E).