There is significant variance in the strategies of labor market flexibility under the same pressure of globalization. This article attempts to explain that variance by examining closely the Korean case, with particular attention to the response of labor, one of the most intractable actors in the

reform process. After theorizing the nature of social welfare as a quasi-collective good and hypothesizing labor's responses based on Olson's theory of collective action, the study seeks to explain Korea's low commitment to flexicurity and the resultant dualism in the labor market. The core argument here is that the collective action problem among atomized corporate unions has led to high employment protection for regular workers in big business at the expense of marginal workers without appropriate social protection.

KEYWORDS: South Korea, corporate unionism, big business, collective action problem, globalization, flexibility, labor market, flexicurity, employment protection

**********

The postwar social contract was predicated on securing risk pooling, social equity, and redistribution through such welfare programs as pensions, unemployment benefits, and health care. But the social contract was not confined to the provision of the welfare state. It was extended to the idea of labor market protection. Institutionalization of collective bargaining on wages and working conditions, as well as legislation protecting jobs and basic employees' rights, exemplify such trends. (1) Thus, it can be said that the welfare state and labor market protection have grown in tandem.

Since the 1980s, however, profound changes have taken place. Waves of globalization have begun to threaten the foundation of the postwar social contract, calling for a renegotiation on the conditions of the welfare state and on labor market protection. (2) Especially in Europe, labor market protection became one of the major issues of the renegotiation, since it is believed to be responsible for the high level of unemployment and for reduced competitiveness. Globalization has been exerting the same pressures for greater flexibility on the developing world. (3) Gradually, labor market flexibility became a norm, and many countries pushed forward with reforms aimed at greater flexibility in the use of labor to enhance competitiveness in the world market. The 1997 financial crisis and subsequent neoliberal labor market reform in South Korea offer a classic example in this regard. The Korean government deregulated the labor market to facilitate corporate restructuring and economic recovery.

Nevertheless, there is significant variance in the nature, perspectives, and strategies of labor market flexibility. Countries have employed very diverse reform strategies, from comprehensive reform packages--notably in Denmark, Finland, and the Netherlands--to reforms more narrowly targeted on specific fields. Even the same reform strategies have produced different outcomes. (4) Why have similar preoccupations by policymakers under the uniform pressure of globalization resulted in various responses and industrial market outcomes? This article attempts to explain these variations by closely looking at the Korean case with special emphasis on the response of labor to the pressure on labor market flexibility. As many studies point out, state strategies vary by internal political dynamics led by labor, one of the most intractable actors in the reform process. (5) Intensified competition and the accompanying threat to employment and income security stimulate workers' interest in and reliance on the traditional function of trade unionism as a protective mechanism. (6) Therefore, analyzing labor's response and behavior is one of the keys to understanding the diversity of national labor markets in the globalization era.

In this study, I draw on the logic of rational choice institutionalism, which focuses on the action and interaction of relevant actors based on each actor's preference and interest framed by institutional setting. Predicated on methodological individualism, my study attempts to provide microfoundations to better explain the distinct form and content of labor market policies and structure in Korea, as an outcome of collective human behaviors. After theorizing the nature of social welfare as a collective good and hypothesizing labor's responses based on Olson's theory of collective action, (7) I examine Korea's low commitment to social welfare and the resultant dualism in the labor market. My core argument is that the collective action problem among atomized corporate unions has led to high employment protection for regular workers in big business at the expense of marginal workers without appropriate social protection.

In the next two sections, I discuss the typology of labor market reforms in member countries of the Organization for Economic Cooperation and Development (OECD) and portray recent developments in South Korea, with particular attention to employment protection legislation and to social expenditure on labor market programs. The following explanatory section analyzes the Korean case, and the concluding section provides a summary and a discussion of theoretical implications.

Globalization and Labor Market Reforms: Three Ideal Types and Korea

Globalization and Labor Market Flexibility

The causes of labor market flexibility are diverse. Globalization is just one of them. Nonetheless, flexibility of the labor market emerges as one of the key neoliberal managerial responses to the challenges of globalization. (8) Liberalization of the world economy brings increases in competition among national economies and firms. Along with accelerated exogenous technical change, market volatility due to increased global competition has raised costs of employment security. The need of firms to adjust production and employment to the fluctuations in the world market has grown more rapidly in recent times than ever before.

Free of domestic politics, international financial institutions such as the International Monetary Fund (IMF) have been swift to address the challenge of globalization and to deliver the neoliberal reform agenda throughout the world. (9) In 1994, the OECD published a comprehensive set of country-specific recommendations--known as the OECD Jobs Strategy--to deal with labor market flexibility in each of the member countries. (10) Although there is a time gap, virtually all industrial countries have begun to realize the need for greater labor market flexibility, which consists of "less generous unemployment insurance provision in terms of benefit payments, duration of benefits, and qualifications of benefits; wider earnings dispersions; lower levels of unionization and less centralized wage bargaining; less government intervention in the wage bargaining process; fewer restrictions on hiring and firing of employees; and lower social insurance charges and other non-wage labor costs, such as the amount of paid vacation." (11)

Diversity of Flexibility: Three Ideal-Type Responses

Responses to neoliberal labor reform mandates, however, have varied by country, depending on the governing structure of labor (12) and more specifically on patterns of interactions between the labor market and the welfare state. (13) At the risk of oversimplification, the literature suggests that interactions between labor markets and welfare states can be categorized into three different ideal types: (1) the liberal regimes of the Anglo-Saxon countries, (2) the conservative regimes of the continental and southern European countries, and (3) the social-democratic regimes of the Nordic countries. (14)

Liberal regimes can be characterized by a limited role of the state, which intervenes in "bad risks" only--such as long-term unemployment and poverty. The liberal model prioritizes unregulated labor markets. Hiring and firing is relatively easy, and employment protection is governed by common law, which relates only to the violation of individual rights such as unfair dismissal. Decentralized and uncoordinated industrial relations at the firm level govern wage determination. Social spending on the labor market, such as unemployment benefits and active labor market policies (ALMPs), is relatively low.

The impacts of globalization are most visible in liberal regimes, which, facing the challenges of globalization, have consolidated a process of combining weak, decentralized industrial relations with weak employment protection, while emphasizing fiscal restraint at the expense of social welfare schemes. Liberal regimes have achieved relatively high private sector employment, averting the high unemployment problem plaguing many European countries. Nevertheless, deteriorating employment protection and social welfare systems have resulted in increasing inequality in the labor market and a relatively low-skill equilibrium.

Meanwhile, conservative regimes are predicated on the combination of high employment protection regulation and compulsory social insurance. The conservative model favors a passive approach to employment management (e.g., generous unemployment benefits and easy take-up of early-retirement allowances), whereas ALMPs are relatively sidelined. The management of unemployment tends to be a question of family support or of induced labor supply reduction, which discourages married women's careers and favors early retirement.

Employment protection and earning-related social insurance programs help maintain skilled workers and earnings equality, but the priority on fiscal stability makes it hard for the state to absorb increasing labor market participants through the expanding public provision of services, as in the social democratic regime. The responses of southern European countries are somewhat different from those of continental Europe. Their employment protection is exceptionally high for core workers, while external flexibility is obtained through the use of short-term and early-retirement schemes and of expanding employment in small firms and self-employment.

Finally, social democratic regimes are based on the notion of a citizen's right to comprehensive risk coverage. Therefore, employment protection is relatively strong. But the labor market is not as rigid as in conservative regimes. Rather flexible employment arrangements are buttressed or compensated by generous social wages and well-developed ALMPs. Based on Nordic "productivism," (15) ALMPs offer various training, retraining, and employment promotion programs for the unemployed to enhance (re)employability and to facilitate industrial restructuring. In principle, wage setting and regulation of working conditions are left to centralized collective bargaining between unions and employers' associations.

The Korean Labor Market

South Korea has also been under the same pressure from globalization to reform its labor market, especially since the 1997 economic crisis. What strategies have been taken to meet the challenge of globalization? Korea has eased regulations on employment protection and, at the same time, expanded labor market programs such as unemployment insurance and ALMPs, as seen in many OECD countries. But reform efforts and policy combinations have differed from those in some OECD member countries. Where can we locate Korea? Is Korea close to the liberal regime, to the social democratic regime, or to the conservative regime?

Legal protection and financial commitment. To figure out the overall characteristics of Korea's strategies for labor market flexibility, a simple scatter plot is displayed in Figure 1. It delivers, on the horizontal axis, an index of the strictness of employment protection legislation (EPL version 2) (16) compiled by the OECD on the basis of an assessment of national legislations as of 2003. The vertical axis indicates public expenditures on labor market programs (passive and active measures) in 2003 as a percentage of GDE Twenty out of thirty OECD countries for which data from 2000 onward are available are distributed along a trend line. (17)

[FIGURE 1 OMITTED]

Although there is no definite clustering, some patterns do exist. As most literature suggests, Anglo-Saxon countries show low levels of EPL and public expenditure on labor market programs, whereas most Nordic and continental European countries display high financial commitment to labor market measures and relatively high EPL levels. There are some variants. Portugal and Spain have higher levels of EPL than other European countries but relatively low spending on labor market programs. Denmark could be singled out as an outlier for its unique combination of a very high level of labor market expenditure and a low level of EPL.

Japan and Korea could be clustered with Anglo-Saxon liberal countries for its low level of labor market expenditure. But this cluster is distinctive in that its employment protection levels are discernibly higher than liberal regime countries, more or less equal to Finland and Austria. Put differently, legal protection for Korean workers is as high as for some European workers, but social safety nets for unemployment are much less generous.

As Table 1 reaffirms, the Korean labor market is more or less protective (ranked eleventh). But, its commitment to social safety nets is exceptionally low (ranked twentieth), not to mention proactive measures like ALMPs (ranked nineteenth). Why does Korea adopt a strategy to combine regulatory protections with weak supportive labor market programs?

Korea as a small open economy and its lack of flexicurity. Korea is quite interesting when compared to small open economies in Europe. An OECD report, assessing the past development and reforms of member countries, raised the question of why countries that implement far-reaching reforms are typically small open economies. A tentative answer was the very pressure exerted by globalization: high trade openness and the resulting high elasticity of labor demand in open economies due to more competition in product markets. (18) In order to survive in a more competitive world market, small open economies have no choice but to ease regulatory protection for employment. And at the same time, they tend to construct comprehensive social safety nets for workers ousted from the labor market in the form of compensations and ALMPs to enhance their skills and (re)employability. In short, they pursue a harmonious combination of flexibility and security, or "flexicurity." Indeed, this is not a new story. This is the way small open economies have survived so far. (19)

As seen in Table 1, Korea is an open economy (ranked fifth) with a continuing emphasis on manufacturing (ranked third), comparable to Sweden, Belgium, the Netherlands, Denmark, Ireland, Austria, and Finland. This begs a question: Why is flexicurity invisible in Korea? Despite seemingly vigorous expansions of unemployment insurance coverage and ALMPs since 1997, the Korean government's financial commitment is the lowest among the twenty countries compared. Korean labor market programs are in place but seem not to receive due attention from the state, business, and labor. Why is that?

Explaining the Korean Case

Union Structure and Hypothetical Responses to Pressure for Flexibility

Labor has been the main thrust for the modern social welfare state, since it is workers who suffer directly from the vagaries of the market. And, more importantly, it is workers who have organizational bases for overcoming the collective action problem. However, their responses to social questions and achievements have never been uniform.

So how can we understand unions' diverse responses to the need for labor market flexibility? One way is to trace how the nature of social welfare, as a "quasi-collective good," shapes the preferences and strategies of unions in the politics of labor markets. Social welfare is a collective good in the sense that the benefits of collective action for social welfare go to every individual in a given group, whether or not the individual contributed to the costs of the action. Universalistic social welfare provisions are the best example. Because of this nonexcludability of social welfare in a group, it follows that unless the group is small or meets certain other special conditions, the collective good will not be provided through the market or voluntary arrangements. Here, we observe the collective action problem arise. (20)

However, social welfare is, by nature, not a pure public good in the sense that it does not satisfy the principle of nonrivalry in consumption. The use of social benefits by one person reduces the allotment for others. This zero-sum nature becomes subject to exclusion. Thus, we often observe that people are excluded from certain social benefits and that different social welfare schemes have different redistributive effects. For these reasons, I view social welfare (including labor market programs) as a quasi-collective good and argue that the uneasy combination of nonexclusion and rival consumption inherent in social welfare is a key starting point for understanding actors' preferences and strategies (see Figure 2).

[FIGURE 2 OMITTED]

This provides a clue to a wide range of differential responses to social questions regarding labor market flexibilities, which are ubiquitous in all of today's industrialized societies. If union movements are based strongly on firm-level unions and lack institutional means for comprehensive negotiations of distributional issues at the industrial or the national level, collective action problems are most likely to be prominent and pervasive. Under such conditions, voices for universal social welfare and industrywide/nationwide unemployment protection are unlikely to prevail. Instead, major distributional struggles will take place at the firm or plant level, which tend to be particularistic and temporally shortsighted and thus are likely to favor wage maximization strategies for insiders.

Likewise, in response to neoliberal pressure for labor market flexibility, individual unions would prioritize job security over flexibility, since losing a job means losing everything. Therefore, they are likely to fiercely oppose easing regulation on employment and to try their best to protect jobs through collective bargaining. Not only universalistic social insurance programs but also employability-enhancing programs such as ALMPs are remote interests for individual firm unions. Therefore, a huge cleavage will occur between the internal and external labor markets, or between organized and unorganized workers. It is likely that unorganized workers situated in the ever-growing external labor markets will bear the burden of labor market flexibility. However, despite unions' efforts at preserving legal protection, the atomized corporate unions will not defend it in the long run because they lack political power as a united class. It will lead to a labor market close to that of a liberal regime.

However, if labor movements are divided along sectoral or industrial lines, demand for sectorally segregated social welfare schemes is likely to prevail. Sectoral/industrial union leaders may have an interest in achieving national unified social welfare schemes to enhance solidarity among workers as a basis for the long-term stability of the schemes and unionism itself. However, that is unlikely to occur, at least in the short run, since competing sets of leadership within and among industrially divided unions will narrow top leaders' vision to prioritize the interests of their own industry and member workers. Also, the rank and file, especially privileged workers within affluent sectors who are aware of rivalry in consumption, will be unwilling to sacrifice parts of their share to benefit others, unless they are ideologically determined to do so or are governed by strong organizational discipline.

In this situation, regulatory employment protections and sector-specific/income-related unemployment insurance programs would be preferred so as to preserve specific job profiles and skills (re)employable for specific industries. (21) Although industrially divided, unions are politically powerful enough to veto labor market reform. Such labor responses are often seen in conservative regimes.

On the contrary, one or a small number of powerful, all-encompassing, and centralized national-level trade unions composed of disciplined subunions will have incentives to install a nationwide/unified universalistic social welfare system to enhance solidarity among workers and thereby increase the political power of the working class. Also, relatively free from firm-specific or sector-specific interests, union leaders are likely to be farsighted and to pursue long-term economic interests for the working class as a whole. Due to centralization of bargaining power, union leaders are better situated to overcome the collective action problem and to control the segregating tendency within and across companies and sectors. Under this condition, it is more likely that the interests of nonregular (low-paid) workers (whose interests are not represented or are underrepresented by firm-level and fragmented sectoral unions) will be articulated, which in turn will increase political power of labor.

In this context, "coordinated flexibilization" or flexicurity would be in the interest of national union leaders. Accommodating the inevitable need for flexibility, they will instead seek to weave comprehensive social safety nets for outsiders (the young, the unemployed, the old, the disabled, the unskilled) and to provide them with more opportunities to upgrade their human capital and build a career. Universal income-supporting programs with low entry barriers and ALMPs would take policy priority over regulatory protections or segmented income-maintenance programs for insiders in the labor market. This kind of approach is likely to be seen in social democratic regimes.

Corporate Unionism in Korea

Labor unionism in the past. Korean trade unionism inherited two legacies from the authoritarian development state. One is tight control by the state, (22) and the other is corporate unionism. The former has gradually disappeared since the 1987 democratic opening, but the latter still thrives. (23)

In the three decades following the Korean War, labor legislation evolved with a view toward ensuring labor peace and minimizing the influence of trade unions. One of the features was the requirement for corporate unions to affiliate with industry federations through a single, state-sponsored national center, the Federation of Korean Trade Unions (FKTU), in 1961. This mandatory affiliation with a monopolistic peak labor organization was to optimize the control of workers. In 1973, when then president Park Chung Hee instituted the outright authoritarian Yushin Regime, the requirement was deleted. It is simply because corporatist control was no longer necessary: basic 'labor rights to collective bargaining and strike were outlawed with the imposition of martial law.

The Chun Doo-hwan government (1980-1987) also sought to further limit labor's institutional power. Through an amendment to the Trade Union Law, the government altered the union structure to place union activity at the company level, explicitly banning industry-level unionization. This measure further decentralized, fragmented, and atomized the labor movement. Moreover, the fear of outsiders supporting and "radicalizing" trade union activities led to the requirement that union membership be restricted to workers within a particular enterprise. Any political activities--for instance, fundraising for a particular political party--were also banned. Moreover, the new Labor Dispute Law prohibited the intervention of a "third party" in collective bargaining and labor disputes. This forbade any involvement of outside groups like prodemocratic independent labor organizers: "While the concept of the company and establishment level as the primary 'locus of bargaining' also exists in other countries, for example Japan and the United States, [these clauses] in Korean labor law carried the concept to an extreme." (24)

The 1987 democratic opening drastically changed the landscape of labor politics. Labor's response to the new political climate was striking. Between July and September 1987, about 3,500 labor conflicts occurred, more than the total number of labor disputes during the entire Park and Chun regimes. The explosion of local labor unions was also remarkable. Since the democratic opening in 1987, their number increased to 6,242 in 1988 and to 7,883 in 1989, from 2,534 in 1985.

The post-1987 labor movement qualitatively differed from the previous one in several ways. (25) First, the center of labor conflicts shifted from light manufacturing to heavy chemical industries and from small-and medium-sized enterprises to large chabols. Second, independent unions, which workers proudly called "democratic unions," grew rapidly and sought to establish an independent national-level union. For two years, between 1987 and 1989, they organized more than 3,700 new unions. The culmination of this organizational movement was the creation of the Korean Confederation of Trade Unions (KCTU), which comprised 619,204 workers in 744 democratic unions grouped into thirty-five industry-level associations as of 2004. Although the KCTU was not recognized by the government until 1998 and it had a smaller number of members than the mainstream KFTU, the KCTU has since led the labor movement in Korea, because its member unions were large factory unions in strategic exporting sectors such as metal, automobiles, and shipbuilding.

The current state of union structure and collective bargaining. Despite the surge of democratic labor unionism, however, the main characteristics of Korean labor politics have remained almost the same, due to the strong institutional tradition of corporate unionism.

One of the striking characteristics is that Korea is still one of the countries with the lowest levels of union density (see Table 2). It is because, as Figure 3 shows, union density is extremely low in small firms with less than 300 employees, which account for 87.5 percent of total employment. In contrast, the organization rate of the small number of large companies is very high, reaching more than 70 percent.

[FIGURE 3 OMITTED]

Accordingly, union membership is highly skewed toward big business. Table 3 illustrates that workers in a big company with more than 1,000 employees make up 62.4 percent of the total union membership, while big company unions account for only 3 percent of the total number of unions. In a similar vein, few nonregular workers, including daily, temporary, and part-time workers, are unionized. On average, the unionization rate among nonregular workers is 5.2 percent, while the rate for regular workers is 16.7 percent (2004). (26)

Second, collective bargaining covers only a small share of all employees (about 10 percent in 2000; see Table 2). It is a reflection of collective bargaining remaining at the firm level, where corporate unions hold exclusive bargaining power. Although there are no detailed surveys on the extent of collective bargaining coverage, a study shows that more than half of companies with over 300 workers, but less than 5 percent of firms with 10 to 100 workers, are governed by a collective contract. (27) The meager and asymmetric coverage of collective bargaining reflects the extremely low rate of organization in small companies. For example, a government survey reveals that only 0.5 percent of nonregular workers and 1.0 percent of regular workers in a small workplace (fewer than five employees) are organized. (28)

The low organizational rate would be offset by the mandatory coverage of collective bargaining. European countries like France and the Netherlands show organizational rates as low as Korea's. But their coverage rate is very high, since collective bargaining at the national/industrial level is applied to unorganized firms by law. In Korea, there is no such mechanism in place, with the exception of a minimum wage set by tripartite compromise at the national level.

All in all, despite strong labor activism, often dubbed "militant," the basic structure of the Korean labor market has not changed. It is still extremely atomized and fragmented, and a huge cleavage has occurred between large and small companies.

Unions' Responses, Labor Market Reforms, and Outcomes: Employment and Social Protection

Since the mid-1990s, the Kim Young Sam government (1993-1997) has tried to adopt reforms to labor law to enhance labor market flexibility. But it failed to bring about significant change due to fierce opposition from the KCTU and the FKTU, which staged the first and the biggest general strike in Korean history. (29) It was under the leadership of the next president, Kim Dae Jung, that substantial changes in labor market regulations were made. Against the backdrop of the 1997-1998 economic crisis, the president-elect initiated the Tripartite Commission and achieved a total of ninety historic agreements with labor and business on structural adjustments and burden sharing, emulating "competitive corporatism" in Europe. (30) The gist of the social pact was to strike a delicate balance between flexibility measures and improvements in labor rights and social safety nets.

As Table 4 illustrates, the 1998 tripartite agreement opened the door to a new welfare state and to a radical revision to the Labor Standards Act by allowing redundancy dismissal and employment adjustment for "urgent managerial needs." However, the rank-and-file corporate unions of the KCTU did not accept the deal and finally replaced the KCTU leadership with hard-liners at the eighth National Congress on February 9, 1998, just three days after signing the historic social pact. Since then, the new leadership of the KCTU, in a strategic move, had repeated withdrawal and participation in the Tripartite Commission, then finally withdrew from the commission for good on February 24, 1999.

Through a series of tugs-of-war, a number of preconditions for layoff had been added and reinforced. "Urgent managerial needs" were defined by a reference to "transfers, mergers and acquisitions." Also, prior to layoffs, employers must make every "verifiable" effort to avoid dismissal; apply fair standards in selecting employees for dismissal; and consult with trade unions or other worker representatives on efforts to avoid dismissal and on fair and reasonable selection criteria. Moreover, worker representatives have to be informed sixty days prior to planned dismissals for managerial reasons. Thus, the Korean labor market is still considered to be highly regulated. (31)

In a situation where collective dismissals are de facto impossible, labor market flexibility has been achieved through the expansion of nonregular workers. Employers who want to hire under fixed-term contracts face no restrictions in firms with fewer than five employees. Also, large firms are allowed to make fixed-term contracts of up to one year duration even without specifying an objective reason. Employment flexibility has been further enhanced by the introduction of temporary work agencies under the Dispatched Workers Act starting July 1998. Dispatching agencies were allowed to hire out workers to user firms for up to two years in twenty-six occupations that require special expertise and experience. In addition, an employee may be hired out in all industries and occupations in order to fill vacancies due to temporary absences of other employees. Since then, dispatch work has increased rapidly. At the end of 1998, only six months after the legislation, the Ministry of Labor estimated that the number of licensed agencies rose to about 800 and of hired-out personnel to 42,000. Other estimates give a much higher figure of dispatch workers, which include those hired out by illegally run agencies operating even before the adoption of the 1998 legislation. (32)

Social protection. Not surprisingly, the share of nonregular workers rose significantly and continuously to more than half of the employed, from 45.7 percent in 1997 to 56.6 percent in 2002. (33) Are these nonregular workers properly compensated and their (re)employability enhanced by labor market programs?

Following the social pact, the government first introduced extensive temporary measures, such as public works, while enacting a new social assistance program and extending social insurance programs to the previously excluded workers in the informal sector and to the self-employed. The coverage of the unemployment insurance system was drastically expanded. It was initially limited to workers in companies with more than thirty employees. The scheme rapidly expanded in four steps: in January 1998, to firms with ten or more employees; in March 1998, to companies with more than five workers; in October 1998, to enterprises with fewer than five workers--that is, to all companies; and finally, in July 1999, to temporary workers (employed at least one month per year), part-time workers (working more than eighteen hours a week), and day workers (who work less than 30.8 hours a week). (34)

In a similar vein, the state-administered national pension and work injury schemes also expanded to cover the entire working population. As state medical insurance had already expanded to cover the entire population in 1989, four major statutory social insurance programs have been put in place to cover all workers, in both the formal and informal sectors.

However, as Table 5 indicates, there is a huge discrepancy between regular workers and nonregular workers as regards effective coverage rates. (35) For instance, only 26 percent of nonregular workers are covered by the employment insurance scheme. Why is that? It is attributable to labor's organizational pattern. As mentioned before, nonregular workers are concentrated in small firms in which the organization rate is extremely low and collective bargaining rarely occurs. Moreover, even a considerable number of nonregular workers in large industries are not organized. In a word, workers' social rights are legally given, but they have not yet materialized for the unprivileged due to the lack of collective bargaining power.

The Korean government's long tradition of fiscal conservatism also makes things worse (see Table 1). Simply speaking, low-paid nonregular workers cannot afford to pay contributions--nor can small-business owners in the low-value-added marginal sector where a large segment of nonregular workers are employed. (36) Without government financial inducements, such as subsidized contributions or matching contributions, both small-business owners and nonregular workers remain reluctant to abide by social insurance laws. (37)

It is paradoxical that the social insurance programs introduced to protect the victims of labor market flexibility do not effectively cover them. Unfortunately, this paradox is replicated in ALMPs. Since ALMPs are run as a part of the employment insurance program, most nonregular workers and employees in unorganized small firms are excluded from the benefits and various services of ALMPs. So are new labor market entrants like young workers, who do not yet have a contribution history. Also, as Table 6 reveals, productive measures such as employment services for the unemployed and youth measures are not fully utilized. Expenditure on training for skill upgrading is also under-invested. In contrast, most ALMP expenditures are channeled into subsidized employment (67.2 percent) for insured firms, which is criticized for entailing dead-weight loss. (38)

Given the fact that the effectiveness of the ALMP differs significantly from one type of program to the other, the unproductive use of the ALMP in Korea contrasts with its use in small states in Europe, which emphasize employability-enhancing general measures such as employment services and training for regular and nonregular workers and more targeted measures for unprivileged workers such as the young and the disabled. Not to mention the lowest level of social expenditure (see Table 1), the combination of ALMPs in Korea makes it hard to improve the efficacy of the job-matching process and to enhance the work experience and skills of workers with the least stable employment histories.

Why has Korea failed to provide unprivileged workers with adequate security in the form of legal and social protection? In other words, why has Korean labor failed to secure flexicurity? Did they exert sincere efforts to achieve security for all when facing the inevitable pressure for labor market flexibility?

Korea's mighty corporate unions in big business are concerned with the vagaries of an increasingly competitive market. However, their concern seems confined to themselves. Figure 4 vividly demonstrates where their prime interests lie. From 1987, when democratic unionism took place following the collapse of the authoritarian regime, wage differentials started to grow and continued to widen even during and after the 1997 economic crisis, when solidarity among workers was most anticipated.

[FIGURE 4 OMITTED]

Not surprisingly, Korean union leaders worry about the self-interest of large corporate unions. Thus, they seek to level income and employment conditions by uniting fragmented corporate unions into industrial/sectoral unions and initiating collective bargaining at the industry level. Seminal industry-level collective bargaining was initiated in the metal and hospital sectors in 2004 and 2005. But their efforts faced the strong resistance of big corporate unions dissatisfied with the result of industry-level bargaining, which tended to be set in the middle. Finally, the labor union of Hyundai Heavy Industry, the biggest shipbuilder in Korea, withdrew from the Korean metal union and returned to firm-level collective bargaining. Likewise, the Big Three hospital unions of Seoul National University Hospital, Yonsei University Hospital, and Samsung Hospital followed suit. (39) It seems a futile attempt to start forming higher-level labor organizations that have collective bargaining power in order to overcome the self-interest of large corporate unions, since corporate unionism has been firmly entrenched in Korea.

From the discussions above, we can conclude that flexicurity for the entire working class is still a remote dream for Korea. Korean labor's great potential for solidarity and social justice seems to have been exhausted by the shortsighted self-interest of corporate unionism.

Conclusion

Explaining the diverse aspects of labor market flexibility is one of the central questions of comparative political economy. I have sought in this article to provide answers by focusing on labor's preference and behavior framed by union structure. The Korean case suggests that welfare state development is not just a function of power resources. Inasmuch as relative power tells, preference matters since it sets the direction of the power used. Despite the low unionization rate, the trade union's impact in the social and political arena is considerable in Korea, because the labor movement is led by large-enterprise unions situated in strategic exporting sectors.

And yet, Korean labor falls short of our expectation of it as a bulwark against the vagaries of the market and as a vehicle for a universal social program for all. Atomized corporate unionism prefers employment protection to flexibility, and wages to social compensation. Consequently, victims of labor market flexibility are given no appropriate opportunities for skills development, let alone social compensation. It may be too soon to properly assess the shadow Korean labor has cast. However, this study implies that without sincere efforts to construct disciplined labor unions at the national or industry level, it may take very long time to realize a genuine welfare state in Korea.

This study sheds light on the responses of organized labor, since labor consists of those affected directly by labor market reform. However important labor is, it is not the only actor in the reform process. Rulers (presidents or premiers), politicians (MPs), business (big business and small business), and bureaucrats (social and economic) are all influential actors. Therefore, a better theory of labor market politics requires an adequate understanding of actions and interactions among them. For this, a further study is needed to understand under what conditions rulers' strategic interests are activated; how different electoral laws affect elected officials' preferences and strategic interactions regarding labor market programs; how different economic structures and firm sizes have different influences on the formation of business interests; and how ontologically different missions and relative power between social and economic bureaucrats affect the labor policy making process.

Notes

This study was first presented at the workshop "Varieties of Capitalism in Asia" at Korea University in Seoul. I would like to thank Stephan Haggard, Byung-kook Kim, Moo-kwon Chung, and Joe Wong for their thoughtful comments and insights. I am especially grateful to the anonymous reviewers for their helpful comments and Heejung Chung and Yuiryoung Jung for their research assistance.

(1.) Marino Regini, "The Dilemmas of Labour Market Regulation." In Gosta Esping Andersen and Marino Regini, Why Deregulate Labour Markets? (New York: Oxford University Press, 2000); C. Crouch, Industrial Relations and European State Traditions (Oxford: Clarendon Press, 1993).

(2.) Ramesh Mishra, Globalization and the Welfare State (Cheltenham: Edward Elgar, 1999); Ramesh Mishra, The Welfare State in Crisis (Brighton: Wheatsheaf, 1984).

(3.) Nita Rudra, "Globalization and Decline of the Welfare State in Less Developed Countries," International Organization 56, no. 2 (2002); Myung-Sei Kang and In-Sub Ma, "Opportunities and Constraints of Tripartism and Welfare Reforms Under the Kim Dae Jung Government," paper presented at the meeting of the Korean Political Science Association, December, 2001.

(4.) Nicola Brandt, Jean-Marce Burniaux, and Romain Duval, "Assessing the OECD Jobs Strategy: Past Developments and Reforms," Economic Department Working Paper No. 429 (Paris: OECD, 2005).

(5.) Gilles Saint-Paul, "The Political Economy of Employment Protection," Journal of Political Economy 110, no. 3 (2002); Stephen J. Frenkel and David Peetz, "Globalization and Industrial Relations in East Asia: A Three-Country Comparison," Industrial Relations 37, no. 3 (1998); Dong-One Kim and Seongsu Kim, "Globalization, Financial Crisis, and Industrial Relations: The Case of South Korea," Industrial Relations 42, no. 3 (2003).

(6.) Dong-One Kim and Seongsu Kim, "Globalization, Financial Crisis, and Industrial Relations," p. 343.

(7.) Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge: Harvard University Press, 1971).

(8.) Heejung Chung, "Different Paths for Flexibility: Deregulated Employment Protection or Temporary Employment," paper presented at the ESPAnet conference "Making Social Policy in the Postindustrial Age," University of Fribourg, Switzerland, September 22-24, 2005; Mishra, Globalization and the Welfare State.

(9.) International Monetary Fund (IMF), Worm Economic Outlook (May 1994).

(10.) Organization for Economic Cooperation and Development (OECD), The OECD Jobs Study: Facts, Analysis, Strategies (Paris: OECD, 1994).

(11.) IMF, World Economic Outlook, p. 36.

(12.) This section draws heavily on Jae-jin Yang and Chung-in Moon, "South Korea: Globalization, Neoliberal Labor Reform, and the Trilemma of an Emerging Welfare State." In Joseph S. Tulchin and Gary Bland, eds., Getting Globalization Right (Boulder: Lynne Rienner Publishers, 2005); Gosta Esping-Andersen and Marino Regini, Why Deregulate Labour Markets? (New York: Oxford University Press, 2000); Marino Regini, "The Dilemmas of Labour Market Regulation." In Esping-Andersen and Regini, Why Deregulate Labour Markets?; Evelyne Huber and John D. Stephens, Development and Crisis of the Welfare State: Parties and Policies in Global Markets (Chicago: University of Chicago Press, 2001).

(13.) David Soskice, "Divergent Production Regimes: Coordinated and Uncoodinated Market Economies in the 1980s and 1990s." In Herbert Kitschelt et al., Continuity and Change in Contemporary Capitalism (Cambridge: Cambridge University Press, 1999); Esping-Andersen and Regini, Why Deregulate Labour Markets?; Peter Hall and David Soskice, Varieties of Capitalism: The Institutional Foundations of Comparative Advantage (New York: Oxford University Press, 2001).

(14.) Gosta Esping-Andersen, Social Foundations of Postindustrial Economie. (New York: Oxford University Press, 1999, ch. 5); Manuela Samek Lodovici, "The Dynamics of Labour Market Reform in European Countries." In Esping-Andersen and Regini. Why Deregulate Labour Markets? pp. 32-35; Torben Iversen and Anne Wren, "Equality, Employment, and Budgetary Restraint: The Trilemma of the Service Economy," World Politics 50, no. 4 (1998).

(15.) Superficially this seems like an echo of what Americas call "workfare." But workfare in the United States implies that social benefits are conditional on accepting work, while Nordic productivism implies that the welfare state must guarantee that all people have the necessary resources and motivation to work.

(16.) EPL version 2 is calculated on the basis of weighted average of indicators for regular contracts, temporary contracts, and dismissals. For detailed information, see OECD, Employment Outlook: Employment Protection Regulation and Labour Market Performance (Paris: OECD, 2004), Annex 2.A1.

(17.) This study uses OECD labor market statistics-indicators (updated July 15, 2005, www.oecd.org). Countries included in the analysis are the United States (USA), the United Kingdom (UK), Canada (CAN), New Zealand (NZL), Australia (AUS), Ireland (IRL), Switzerland (CHE), Sweden (SWE), Norway (NOR), Finland (FIN), Denmark (DEN), Germany (DEU), the Netherlands (NLD), Belgium (BEL), Austria (AUT), France (FRA), Spain (ESP), Portugal (PRT), Japan (JPN), and Korea (KOR).

(18.) Brandt, Burniaux, and Duval, "Assessing the OECD Jobs Strategy."

(19.) David R. Cameron, "The Expansion of the Public Economy: A Comparative Analysis," American Political Science Review 72, no. 4 (1978); Peter J. Katzenstein, Small States in World Markets: Industrial Policy in Europe (Ithaca: Cornell University Press, 1985); Geoffrey Garret and Peter Lange, "Political Responses to Interdependence: What's 'Left' for the Left?" International Organization 45, no. 4 (1991).

(20.) Olson, The Logic of Collective Action.

(21.) Margarita Estevez-Abe, Torben Iversen, and David Soskice, "Social Protection and the Formation of Skills: A Reinterpretation of the Welfare State." In Hall and Soskice, Varieties of Capitalism.

(22.) Korean labor was placed under the tight corporatist control of the authoritarian developmental state for economic and political reasons. First, the government's strategy for export-oriented industrialization was predicated on export price competitiveness in the world market. Thus, the economic imperatives were such that workers' collective action should be kept minimal and wages tightly controlled. Second, the security threat emanating from the Communist North led the South Korean government to be overly sensitive to union activities. For fear of the radicalization or "politicalization" of the labor movement, unions should be atomized and independent from outside "reddish" activists.

(23.) OECD, "Pushing Ahead with Reform in Korea: Labour Market and Social Safety-Net Policies" (Paris: OECD, 2000); Hagen Koo, Korean Workers: The Culture and Politics of Class Formation (Ithaca: Cornell University Press, 2001); Jae-jin Yang, "Corporate Unionism and Its Impact on the Korean Welfare State," Korean Political Science Review 39, no. 3 (2005).

(24.) OECD, "Pushing Ahead with Reform in Korea," p. 44.

(25.) Kwang-yung Shin, "Korean Industrial Relations and Political Turnaround" (in Korean), Quarterly Sasang 9, no. 2 (1997); Hagen Koo, "The State, Minjung, and the Working Class in South Korea." In Hagen Koo, ed., State and Society in Contemporary Korea (Ithaca: Cornell University Press, 1993).

(26.) Ministry of Labor (Korea), "Analysis of the Economically Active Population in 2004" (in Korean).

(27.) OECD, "Pushing Ahead with Reform in Korea," p. 53.

(28.) Ministry of Labor (Korea), "Analysis of the Economically Active Population in 2004."

(29.) It was the first general strike since the US military government outlawed the Korean Communist Party and leftist labor organizations in March 1947. The KCTU led the general strike, mobilizing 5 million workers for twenty-six days and costing a production loss of 3 trillion won--US$3 billion. See Jae-jin Yang, "The 1999 Pension Reform and a New Social Contract in South Korea" (Ph.D. diss., Rutgers University, 2000), p. 129.

(30.) Martin Rhodes, "The Political Economy of Social Pacts: 'Comparative Corporatism' and European Welfare Reform." In Paul Pierson, ed., The New Politics of the Welfare State (New York: Oxford University Press, 2001).

(31.) OECD, "Pushing Ahead with Reform in Korea"; see also Table 1.

(32.) OECD, "Pushing Ahead with Reform in Korea," pp. 63-64.

(33.) Yu-Sun Kim, Labor Flexibility and Nonstandard Workforce Increase (in Korean) (Seoul: Korea Labour and Society Institute, 2004), p. 17. Regular workers are defined as employees that work for more than one year and are paid "regular" wages, meaning that they are paid the standard wage plus overtime pay. Nonregular employment refers to temporary and daily work, plus atypical work such as part-time and on-call work. Temporary workers are employed for a determined length of time, usually longer than a month and shorter than a year. Daily workers are employed on a daily basis.

(34.) Bernard Gazier and Remy Herrea, "Escaping from the Crisis and Activating Labor Market Policies in Asia: Some Lessons from European and French Experience with a Special Focus on South Korea," paper presented at the international conference "Flexibility vs. Security? Social Policy and the Labor Market in Europe and East Asia," Seoul, November 30-December 1, 2000, p. 344.

(35.) The discrepancy is somewhat overblown, especially for health insurance, since means-tested health assistance is provided for low-income workers not covered by the health insurance program. Also, health insurance works on a household basis, whereby a considerable number of noncovered nonregular workers are protected through (male) breadwinners' health insurance.

(36.) As of 2000, only 10.8 percent of workers in workplaces with fewer than five employees are regular workers, while 89.2 percent are nonregular workers. By contrast, 86.4 percent of workers in firms with more than 300 employees are regular workers. See Kyungsoo Choi, "A Study on the Definitions of Employment Status to Measure Employment Structure Changes and Their Status" (in Korean), Nodongkyungjenonjip 24, no. 2 (2001).

(37.) Korea's social insurance schemes are basically funded by payroll contributions split between the employer and the employee--in the case of company workers--and by individuals, in the case of the self-employed. Without a contribution history, no one can claim benefits.

(38.) The effectiveness of wage subsidy programs has often been disappointing in terms of bringing the unemployed back into unsubsidized work, because subsidized jobs are created that would have been created even without the subsidy, and workers who qualify for a subsidy simply replace others who do not. See J. Martin, "What Works Among Active Labour Market Policies: Evidence from OECD Countries' Experiences," OECD Economic Studies, No. 30 (2000).

(39.) Hankook Daily, March 29, 2004; Hankyureh, September 20, 2004; Hankookgyungje, July 21, 2005.

Jae-jin Yang is associate professor of public administration at Yonsei University in Seoul. His areas of expertise are social policy and state bureaucracy. He serves on the editorial boards of Korean Political Science Review and Korean Public Administration Review. He has written many articles on structural adjustment and social policy, including "Democratic Governance and Bureaucratic Politics" (Policy and Politics, 2004) and "Skill Formation and the Origin of the Korean Welfare System (Korean Political Science Review, 2004).

Table 1 Global Integration, EPL, and Social Spending (2003)

                     Trade          Manufacturing (b)     EPL (c)
Country           Openness (a)         (% of GDP)        (Ver. 2)

Finland             58.66 (10)           24 (2)           2.1 (10)
Germany             56.18 (11)           23 (3)           2.5 (6)
Belgium            172.34 (1)            19 (8)           2.5 (6)
France              44.22 (14)           18 (9)           2.9 (3)
Spain               42.05 (16)           17 (11)          3.1 (2)
Portugal            51.69 (12)           18 (9)           3.5 (1)
Netherlands        108.87 (2)            15 (14)          2.3 (8)
Sweden              60.99 (6)            21 (6)           2.6 (4)
Denmark             59.07 (9)            16 (13)          1.8 (12)
Japan               19.87 (19)           21 (6)           1.8 (12)
Korea               61.56 (5)            23 (3)           2.0 (11)
Austria             76.72 (4)            22 (5)           2.2 (9)
Norway              48.43 (13)           11 (17)          2.6 (4)
Ireland             95.05 (3)            32 (1)           1.3 (1)
Switzerland         60.79 (7)            --               1.6 (14)
Australia           30.75 (18)           12 (16)          1.5 (15)
New Zealand         44.05 (150)          --               1.3 (16)
UK                  38.75 (17)           17 (11)          1.1 (18)
Canada              60.45 (8)            --               1.1 (18)
US                  18.51 (20)           15 (14)          0.7 (20)

                    Social
                 Expenditure (d)      ALMPs (e)      Passive LMPs (f)
Country           (% of GDP)         (% of GDP)         (% of GDPP)

Finland              24.8 (8)          0.91 (8)          2.09 (4)
Germany              27.4 (4)          1.14 (6)          2.31 (3)
Belgium              27.2 (5)          1.24 (4)          2.51 (2)
France               28.5 (3)          1.09 (7)          1.77 (6)
Spain                19.6 (13)         0.72 (11)         1.48 (7)
Portugal             21.1 (12)         0.67 (12)         1.28 (9)
Netherlands          21.8 (10)         1.83 (1)          1.86 (5)
Sweden               28.9 (2)          1.29 (3)          1.22 (10)
Denmark              29.2 (1)          1.74 (2)          2.68 (1)
Japan                16.9 (17)         0.32 (18)         0.46 (17)
Korea                 6.1 (20)         0.16 (19)         0.14 (20)
Austria              26.0 (7)          0.63 (13)         1.37 (8)
Norway               23.9 (9)          0.81 (9)          0.87 (13)
Ireland              13.8 (19)         1.17 (5)          0.91 (12)
Switzerland          26.4 (6)          0.77 (10)         1.02 (11)
Australia            18.0 (15)         0.39 (16)         0.74 (16)
New Zealand          18.5 (14)         0.52 (15)         0.80 (14)
UK                   21.8 (10)         0.53 (14)         0.37 (18)
Canada               17.8 (16)         0.37 (17)         0.77 (15)
US                   14.8 (18)         0.16 (19)         0.37 (l8)

Source: (a.) Imp.+Exp./GDP, http://devdata.worldbank.org/ (accessed
on Aug. 31, 2005).

(b.) Manufacturing output/GDP, World Bank, World Development
Indicators (2005).

(c.) Labor Market Statistics--Indicators, www.oecd.org (accessed on
August 30, 2005).

(d.) OECD Social Expenditure Database, www.oecd.org/els/social/
expenditure (accessed on Aug. 31, 2005).

(e.) Labor Market Statistics--Indicators, www.oecd.org (accessed on
August 30, 2005).

(f.) Expenditure on passive labor market measures, GDP, Labor Market
Statistics--Indicators, www.oecd.org (accessed on August 30, 2005).

Note: Number in the bracket is ranking.

Table 2 Trade Union Density and Collective Bargaining
Coverage (%)

                           Trade Union Density (TUD)

Country             1970 (b)    1980 (c)    1990    2000 (d)

Finland                51         69        72      76
Germany                32         35        31      25
Belgium                41         54        54      56
France                 22         18        10      10
Spain                  --         7         11      15
Portugal               --         61        32      24
Netherlands            37         35        25      23
Sweden                 68         80        80      79
Denmark                60         79        75      74
Japan                  35         31        25      22
Korea                  13         15        17      11
Austria                63         57        47      25
Norway                 57         58        59      54
Ireland                53         57        51      38
Switzerland            29         31        24      18
Australia              44         48        40      25
New Zealand            56         69        51      23
United Kingdom         45         51        39      31
Canada                 32         35        33      28
United States          27         22        15      13

                    Collective Bargaining (a)

Country             1980      1990      2000

Finland              90+       90+       90+
Germany              80+       80+       68
Belgium              90+       90+       90+
France               80+       90+       90+
Spain                60+       70+       80+
Portugal             70+       70+       80+
Netherlands          70+       70+       80+
Sweden               80+       80+       90+
Denmark              70+       70+       80+
Japan                25+       20+       15+
Korea                15+       20+       10+
Austria              95+       95+       95+
Norway               70+       70+       70+
Ireland              --        --        --
Switzerland          50+       50+       40+
Australia            80+       80+       80+
New Zealand          60+       60+       25+
United Kingdom       70+       40+       30+
Canada               37        38        32
United States        26        18        14

Source: OECD, Emplocment Outlook (2004), p. 145.

Notes: (a.) Figures with a + sign represent lower-bound estimates.
For the purposes of calculating ranking and averages, the indicated
value was increased by 2.5 percentage points.

(b.) 1971 for New Zealand.

(c.) 1982 for Australia and Portugal, 1984 for Canada.

(d.) 1997 for Portugal, 1998 for Spain, 2001 for Switzerland.

Table 3 Number of Unions and Membership by Firm Size (2003)

                                Less                  100-
                  Total      than 30      30-99        299

Unions             6,017       2,333      1,727      1,333
                (100.0%)     (38.8%)    (28.7%)    (22.2%)
Membership     1,536,846      26,292    101,691    220,806
                (100.0%)      (1.7%)     (6.7%)    (14.4%)

                                          1,000
                    300-        500-        and
                     499         999       over

Unions               243         200        181
                  (4.0%)      (3.3%)     (3.0%)
Membership        92,724     135,886    959,444
                  (6.0%)      (8.8%)    (62.4%)

Source: Ministry of Labor, "Analysis of Labor Unions in 2003"
(in Korean, 2004).

Table 4 Key Contents of the 1998 Social Pact

Management              * improvement of the corporate financial
transparency              structure
and corporate           * more responsible and more transparent
restructuring             governance corporate
                        * promotion of business competitiveness

Enhancing labor         * permission for employers to dismiss
market                    workers in cases of managerial need
flexibility             * permission for the establishment of
                          temporary work agencies

Policies to promote     * expansion and improvement of employment
employment stability      insurance
and combat              * livelihood support for the unemployed
unemployment            * expansion and improvement of the public
                          employment service
                        * expansion of vocational training
                        * job creation through public works and
                          business start-up subsidies
                        * consultation and rehiring requirements in
                          case of redundancy dismissals

Enhancing labor         * permission for public servants to form
rights                    workplace associations
                        * permission for teachers to join trade
                          unions
                        * permission for trade unions to engage
                          in political activities
                        * right of dismissed and unemployed workers
                          to join trade unions

Extension and           * integration of social partners in
consolidation             social security steering committees
of the social           * wage guarantee in bankruptcy cases
security system         * extension of social insurance coverage
                          to nonregular workers

Source: OECD, Pushing Ahead with Reform in Korea: Labour Market and
Social Safety-Net Policies (Paris: OECD, 2000), p. 49.

Table 5 Effective Coverage Rate for Regular vs. Nonregular
Workers (2003, %)

                                   Health      Employment    Severance
                       Pension    Insurance    Insurance        Pay

Regular workers           96.6         97.6          79.5         98.8
Nonregular workers
  Temporary               25.1         27.5          24.7         14.2
  Part-time                2.0          2.7           3.0          1.8
  On-call                  3.6          0.2           1.7           --
  Special employ          22.1         24.5          20.8         17.0
  Dispatch                52.0         55.1          52.0         46.9
  Subcontract             55.2         69.7          52.3         42.2
  Home work                4.8          4.8           4.8          4.2
  Total                   26.4         28.9          26.0         16.0
All Workers               57.7         59.5          49.8         52.9

Source: Yu Sun Kim, Labor Market Flexibility and Nonstandard Workforce
Increase (Seoul: Korea Labour & Society Institute, 2004, in Korean).

Table 6 Expenditures on ALMPs by Category (2000, %)

                   Employment                 Youth      Subsidised
Country             Service      Training    Measures    Employment

Finland                  11.2        32.7        17.8          29.9
Germany                  18.7        27.6         6.5          25.2
Belgium                  14.1        18.5          --          59.3
France                   12.5        20.6        30.1          30.1
Spain                     5.1        29.6         6.1          41.8
Portugal                 21.6        58.8        35.3          17.6
Netherlands              15.9        19.1         2.5          26.1
Sweden                   18.8        22.5         1.4          19.6
Denmark                   7.1        54.2         6.5          11.0
Japan                    39.3        10.7          --          46.4
Korea                     8.7        19.6         2.2          67.2
Austria                  26.5        34.7         8.2          20.4
Norway                   15.6        10.4         1.3           1.3
Switzerland              22.9        18.8         2.1          27.1
Australia                44.4         4.3        15.6          24.4
New Zealand              12.7        32.7        25.5          20.0
UK                       36.1        13.9        41.7           2.8
Canada                   39.2        33.3         5.9          15.7
US                       26.7        26.7          20           6.7

                    Measures
                    for the
Country             Disabled      Total

Finland                   8.4         100
Germany                    22         100
Belgium                   8.9         100
France                    6.6         100
Spain                     3.1         100
Portugal                  2.0         100
Netherlands              36.3         100
Sweden                   37.7         100
Denmark                  21.3         100
Japan                     3.6         100
Korea                     2.2         100
Austria                  10.2         100
Norway                   71.4         100
Switzerland              29.2         100
Australia                11.1         100
New Zealand               9.1         100
UK                        5.6         100
Canada                    5.9         100
US                         20         100

Source: OECD. OECD Emplm,meut Outlook (June), (Paris: OECD, 2001).