Until November 1997, the Republic of Korea was widely regarded as one of the most remarkable economic success stories among developing countries. Since the financial crisis of November 1997, however, the economy has experienced a sharp downturn: A recent government report estimated that Korea's per capital gross domestic product (GDP) fell to $6,823 in 1998 from $10,307 in 1997, retreating to the level of eight years ago; moreover, real wages are expected to decline for the first time in almost 20 years.
In light of these questions, the main purpose of this paper is to examine three questions. First, what has been happening in the Korean labor market and industrial and labor relations since the 1997 financial crisis? Second, what major government policies have emerged to cope with new developments, such as rising unemployment and labor disputes? Third, have these policies been effective? If not, would it be necessary to employ supplementary policies or to make a fundamental policy change? This paper examines these issues and makes alternative policy recommendations. Then it concludes by discussing the labor market outlook in Korea and drawing certain lessons from other emerging countries.
Chapter 1: Rising Unemployment and Policy Response
Overview
A rise in unemployment has been quite dramatic since the beginning of the economic crisis. The unemployment rate had been rather stable before the crisis, between 2% and 3%, but it rose sharply to 6.8% in 1998. The number of unemployed rose from about 560 thousand to 1.46 million, a net increase of about 0.9 million, as shown in Table 1-1.
The month to month increase is even more dramatic: In October 1997, one month before the economic crisis, the unemployment rate was 2.1%; by July 1998, it had risen to 7.6%; and in February 1999 to 8.7%. Nonetheless, these unemployment figures-also known as open unemployment-do not include discouraged workers or involuntary short-time workers, who are currently employed less than 18 hours a week despite their preference to work more. If these workers are included, the number of the underutilized (unemployed +discouraged workers + involuntary short time workers) will be about 2.26 million (1.46 + 0.61+0.19 force) as of 1998, equivalent to 10.5% of the total labor force.
One recent study calculated the severity of unemployment as a measure of social pain in Korea and compared Korea with other countries. It showed that the unemployment rate of 8.7 percent in February 1999 produced the same level of severity, as did a 13 percent unemployment rate in the United States and European countries. In other words, even the same unemployment rate generates more social distress in Korea than in other advanced countries primarily due to Korea's underdeveloped social safety net and relatively low labor force participation rates of women even as secondary bread winners.
The significance of unemployment trends lies in that the sharp rise in unemployment in 1998 was mostly due to the bankruptcy of small and medium-sized firms, not to the restructuring of large-scale firms. Small-sized firms have suffered disproportionately from the high interest rate policy imposed by the IMF as a conditionality of its rescue loans.
A further rise in unemployment is expected in 1999 as a byproduct of the downsizing and restructuring of the largest companies-especially the chaebol (family-owned large conglomerates) and the public enterprises.
The government's recent efforts to boost domestic demand may temper the sharply rising unemployment; nonetheless, its rising trend will continue through most of 1999.
Major Features of Unemployment
The sharp rise in unemployment has not been uniform among various types of workers. Rather, there have been noticeable changes in the composition of the unemployed.
First, the changes in employment by occupation illustrate that technicians and unskilled workers are most severely affected, losing approximately one million jobs in a year.
Professional and managerial workers appeared to gain somewhat, but in fact, they also lost some jobs considering the recent trend of rising employment in those occupations. Clerical and service workers jointly lost about 291 thousand jobs. By contrast, agricultural employment increased by one million, reflecting the role of the agricultural sector in Korea in buffering the repercussions of changing labor demand.
Second, the rise in unemployment is sharper for household heads, as shown in Table 1-3, suggesting that the likelihood of middle-aged unemployment is rising more rapidly than that of young or older workers.
In addition, male unemployment shows a steeper increase compared with female unemployment, partly because women tend to give up their job search and withdraw from the labor force in times of unpromising job prospects.
Third, Table 1-4 shows that temporary and day workers account for the major portion of the unemployed.
These marginal workers represented about 62.3 percent of the unemployed as of 1998. And this trend has accelerated over time, indicating a strong tendency towards marginalization of the underemployed.
Fourth, Table 1-4 reveals that the current unemployment situation is driven more by an economic recession than by corporate restructuring. Because the corporate restructuring of large-scale industries has taken place only since late 1998, its impact has not been fully felt yet. In fact, most of the unemployment generated during 1998 resulted from bankruptcies of small and medium-sized firms spawned by the IMF's high interest rate policy. Specifically, the high rates of unemployment in 1998 were mainly due to the economic recession triggered by the IMF's inappropriate policy package-deflationary package-which included policies of high interest rates and fiscal austerity.
Table 1-5 supports this observation. Among the unemployed during 1998, only 5.7% represented those who left jobs from firms with 300 or more employees.8
Those from firms with less than 20 employees accounted for more than 74.2 % of the unemployed. Therefore, it can be concluded that the unemployment effect of the 1998 recession has most severely hit the workers in smaller firms.
Falling Wages and Rising Income Inequality
After the crisis, the real wage level declined sharply from an annual growth rate of 2.4 percent in 1997 to a negative 9.3 in 1998, as shown in Table 1-6.
In terms of the quarterly changes during 1998, the wage level dropped precipitously until the third quarter of 1998 and then marked no further decrease.
Since the beginning of the economic crisis, it has been the third quarter of 1998 that most accurately illustrated the sharp reduction in wages before and after the crisis.
It shows that from the third quarter of 1997, nominal wages dropped by 8.1 percent and real wages plummeted by 14.2 percent. It exposes the degree of hardship that Korean workers have undergone during the year following the crisis. The reduction in nominal wages in 1998 has been unprecedented in Korea's economic history since wage data became available.
Not only has the general wage level worsened but also household income distribution has deteriorated sharply since the crisis. In Table 1-7, household income reduction during 1998 by income group shows that the crisis hit the lowest income group most severely.
For those in the lowest 20% income group, nominal incomes dropped by 17.2% and real incomes by 23.7%. In contrast, the highest 10% enjoyed the least reduction in household incomes. Indeed, nominal incomes rose about 4% for the highest group even though real incomes declined slightly by 2.5%. Since the crisis, income inequality has become more acute.
Recently, the press has begun to discuss the dissolution of the middle class in Korean society. One study showed that following the crisis, 64 % or 3.84 million out of the 6 million middle-class households, experienced downward mobility, while only 6% (360 thousand) moved up to the upper class, and about 30% (1.80 million) still remained in a highly unstable situation.
Another study found that those people below the absolute poverty level was 4.1% (1.64 million) of the total population in the third quarter of 1997, but jumped to 11.6% (4.64 million) at the end of 1998.
Unemployment Insurance Coverage
Against this rising unemployment and worsening income inequality, the government's major policy response has been a rapid expansion of unemployment benefits. The chief component of the policy has been an unemployment insurance program.
The government's approach, however, contains a number of problems. First of all, even though unemployment insurance has undergone a series of expansions, unemployment benefits still cover only a small fraction of the unemployed. The current program, moreover, tends to benefit relatively well-off workers dismissed from large-scale firms rather than those workers laid-off from small and medium-sized firms.
Initially, the program applied only to the firms with 30 or more regular employees. This ceiling has been lowered three times: First, it was reduced to 10 or more employees; then to five; and finally, no restrictions have been applied since October 1998.
Still, the unemployed become entitled to benefits only after a minimum insured period of six months. Those workers laid off from small-scale firms, therefore, could begin to receive benefits only after April 1999; the full impact of the benefits expansion will be more visible later on.
The current unemployment insurance disproportionately benefits the relatively well-off unemployed. As shown in Table 1-4, temporary and day laborers account for the majority of the unemployed, but currently are not covered by unemployment insurance.
Table 1-3 reveals that non-household heads constitute 54.7% of the unemployed. Many are young workers, seeking jobs upon graduation from university or high school. These newcomers to the labor market also are not covered by unemployment insurance. As a whole, unemployment insurance coverage is still quite limited. As of June 1998, among the 1.5 million unemployed, only 7% (about 105 thousand) received unemployment benefits.
The duration of unemployment benefits is remarkably short, and the level of benefits is quite meager, especially for those who earned low wages before dismissal. The duration of benefits depends on the worker's age at the time of unemployment and the length of his or her insured period. Although the average duration of benefits is only two months, the government granted special benefits between January and June of 1999, during which unemployed persons were entitled to a two-month equivalent of additional benefits. Similarly, the worker under the age of 25 who has been insured less than three years could receive only one month of benefits, but recently, the benefits have been extended to two months. The maximum duration of seven months is only available to those who have been insured more than 10 years and are over the age of 50.
This short duration can be extended for workers undertaking vocational training that has been approved by the government. But even their benefits are usually limited to three to six months.
Another problem with unemployment insurance lies in its minimum level of benefits. The level of benefits is set at 50% of the worker's average earnings during the preceding 12 months, up to a maximum of 1.05 million won (roughly USD 900) per month and a minimum of 250 thousand won (USD 200) per month. The minimum level of benefits- currently equal to 70% of the minimum wage-is highly insufficient for social protection. As a mere one-quarter of the average earnings in Korea, the current minimum wage and benefits based on it can hardly provide adequate insurance for the unemployed.
In sum, the unemployment insurance scheme in Korea is still in its infancy and is skewed very much toward preventing the potentially negative impact of unemployment benefits on work incentives rather than genuinely reducing the hardship of the unemployed. The government had been reluctant to provide comprehensive social protection in fear of the European welfare disease.
Workers Excluded by Unemployment Insurance
The social safety net for those who are not qualified for unemployment insurance is extremely limited. Only a few of those workers are eligible for public assistance. Korea's public assistance program does not require previous work experience, but does apply a stringent means-test. Regardless of their poverty, those who are classified as being able to work can not receive cash transfers and are eligible only for some in-kind transfers like medical and education subsidies and subsidized loans. In addition, to be qualified for public assistance, the beneficiary should not have income and assets exceeding a certain level and must not have family members capable of providing assistance.
Even for those who can receive cash transfers, the amount is extremely meager-122,000 won (about USD 100) per month, equivalent to less than 9% of the average monthly earnings. As of June 1998, about 8.8 % of the unemployed (about 143,000) were receiving public assistance. However, most were eligible for in-kind support only because they were classified as able to work.
Another example clearly illustrates the extreme underdevelopment of the social safety net in Korea. In September 1998, the Korean government conducted a survey to examine the structure of Korean unemployment. In that survey, the government asked how the unemployed household head was providing for his or her family. Table 1-8 reveals that as of September 1998, only 2.4 % of the unemployed lived off the social safety net, either by unemployment insurance or public pension or by public assistance.
It should be further noted that September 1998, the time in which the data was collected, was almost a year after the crisis. Even then, the social safety net covered only 2.4 % of the unemployed household heads. The remaining 97.6 % of the unemployed were all managing their livelihood through private safety nets. Among them, 52.5% were dependent on personal savings, 27.0% were living off the income of other family members, and 9.7% were receiving support from relatives and friends. Even those living off the loans from private sources constituted 7.0% of the unemployed, a much higher percentage than those covered by the social safety net.
As previously mentioned, about 7% of the unemployed were covered by unemployment insurance and about 8.8 % by public assistance programs as of June 1998. These figures were calculated based on the official government plan. On the contrary, the data directly collected from the unemployed reveals that even in September 1998, only 1.4% of the unemployed had received unemployment insurance and a mere 0.7% had received some kind of public assistance.
The inconsistency between the two findings indicates that there is still a substantial gap between what the government promised to do and what is actually taking place, between legal eligibility and practical implementation of social safety net.
The social safety net in Korea is still in its inception, and the majority of the unemployed rely private sources now more than ever before.
The insufficiency of public assistance puts tremendous strains on a private safety net. At the same time, the private source of support will be depleted eventually as joblessness continues. At issue is what should be done for those who are not covered by any kind of social safety net. There is a huge "blind spot" or "dead zone" of social safety net for the people that are covered by neither the unemployment insurance scheme nor public assistance programs. Faced with the enormous challenge, the government should give the highest priority to minimizing this blind spot as soon as possible.
Proposed Changes
To minimize the gap between legal eligibility and the actual implementation of social safety net, the following policy package bears serious consideration. First, policymakers should accelerate the expansion of unemployment insurance coverage and extend the duration of the benefits up to at least six months.
Second, policymakers should relax the restrictive conditions of the current public assistance program in order to help a broader range of people in need. Moreover, the benefit level should be raised to the poverty line, either one-half or one-third of the median income, depending on the poverty line used. These two policies will contribute to narrowing the gap between policy and practice.
The absence of an adequate social safety net renders direct job creation by the public sector ever more indispensable. Having recognized the importance of public job creation, the government initiated a large expansion of public works programs for the 1999 budget.
Still, the government's effort to expand public works programs should have the following features in order to be fully effective:
(1) Public works programs should focus on constructive works, such as environmental clean-ups, community service, and infrastructure development. Moreover, the programs will be more effective if they are geared toward improving the living conditions of the poor. Priority should be given to the cleaning up and rebuilding of social infrastructure, such as roads and sanitation, especially in the residential areas of the urban poor;
(2) The programs should target the workers who are genuinely unable to find work elsewhere and especially the workers with long-term unemployment records. In this regard, determining the level of wage rate at the public works programs is very important. If it is too high, approximate to the wage level of unskilled workers, substitution of employment will be likely to occur. Some workers, already employed, may leave the unskilled labor market and seek jobs in public works programs simply because the programs are rather loosely managed. Specifically, unskilled agricultural workers may want to switch to the urban public works. In this scenario, public works programs only generates a loose labor market in the agricultural sector without creating any additional employment in the economy as a whole. While the wage level should not be too low, it should be maintained at the level sufficient to make the programs unattractive to the non-poor workers. On the other hand, if it is too low, the very meaning of helping the unemployed poor gets lost;
(3) The programs should include some training features wherever possible, no matter how basi