Abstract
This Selected Issues paper focuses on some of the key stylized facts of Korean business and export cycles over 1960–2001, and calculates a chronology for the classical cycle in these series by applying a variant of the Bry and Boschan (1971) cycle-doling algorithm. It highlights that the Korean classical business cycle and exports cycles are extremely asymmetric, as they exhibit long-lived expansions and much shorter-lived contractions. The results also indicate that the probability of ending a contraction or expansion phase in Korean industrial production and Korean real exports is independent of their duration.
II. Labor Market Developments in Korea Since the Crisis1
The adjustment to the economic crisis involved major changes in the labor market, including the end of lifetime employment, increased reliance on nonregular (i.e., temporary and daily) workers, increased flexibility of compensation through performance-based schemes, and higher participation and employment of women. These changes have made Korea’s labor market more flexible. However, continuing rigidities may be inducing an excessively high level of nonregular employment, with negative implications for productivity and equity.
A. Introduction
1. The economic crisis in 1997–98 led to changes not only in the areas of macroeconomic policies, corporate governance, and financial structures, but also to policy reforms in the labor market and adjustments by employers and employees. Most visibly, the crisis put an end to lifetime employment, which had been made possible by the previous era of sustained fast economic growth and tight labor markets. Other important changes in the labor markets included: increased reliance on nonregular employment, increased flexibility of employment compensation, and higher participation and employment of women.
2. This paper is organized as follows. Section ? briefly reviews the immediate impact of the financial crisis on unemployment and various policy responses. Section ? discusses labor market adjustments since the crisis.2 The last section points out some remaining areas to be addressed to further improve the functioning of the labor market in Korea.
B. Immediate Impact of the Crisis and Policy Responses
3. The Korean labor market showed significant flexibility in response to the crisis, both in terms of prices and quantity. The immediate impact of the crisis on the labor market was a sharp rise in unemployment. The unemployment rate rose from 2.6 percent at the onset of the crisis in November 1997 to peak at about 8½ percent in February 1999.3 This was a severe shock to the labor market, considering that, from the 1980s to the mid-1990s, the annual average unemployment rate never exceeded 4 percent and was typically in the region of 2–3 percent. Unlike the past, when unemployment was concentrated among young people, this time unemployment among those aged 30–50 rose substantially. Unemployment would have been even higher if labor force participation had not decreased, mainly due to the postponement of job search by younger workers and withdrawal by female workers.
4. Earnings adjusted flexibly to lower labor demand. Average monthly earnings fell by 2½ percent in 1998, or by 9 percent in real terms. As the OECD (2000) notes, pay cuts are rare in OECD countries, even during crisis situations. The brunt of the adjustment in Korea was borne by overtime and bonuses, which fell by 15 percent and 17 percent, respectively.4 In contrast, regular salaries rose by about 3 3/4 percent. Wages responded very flexibly due to the decentralized nature of wage bargaining in Korea. Compensation is determined on an annual basis at the enterprise level; hence, outcomes reflect primarily firm-specific factors. Further, the statutory minimum wage has little impact, having fallen to less than 25 percent of the average manufacturing wage (OECD, 2000).
5. The initial policy response to the rise in unemployment included the expansion of the social safety net. Measures adopted included:
temporary wage subsidies to firms that retained redundant workers, and further assistance through vocational training and job placement;
an expansion of the coverage of the employment insurance system (EIS) in 1998, first to include all enterprises with five or more employees, and later, to cover all enterprises;5
given the limited coverage of EIS at the beginning of the crisis, public works programs were also initiated and employed about ½ million people in 1998 and 3/4 million in 1999.
With the implementation in 2000 of the National Basic Livelihood Security Law, a comprehensive institutional framework for welfare provision was put in place.
6. For longer-term restructuring, the Tripartite Commission was formed in January 1998 to forge agreements on layoffs, pay cuts, and reductions in overtime and bonuses. Under the auspices of the Commission, two major changes were enacted to the labor laws in 1998:
Dismissal of employees was made easier, by allowing layoffs of redundant workers in cases of “urgent material need,” including mergers and acquisitions.
Temporary work was legalized, and temporary employment agencies were allowed to operate.
These changes signaled that employment security could no longer be taken for granted, and opened the way for new forms of employment arrangements. Although firms that were downsizing resorted mainly to “honorable” retirement, spinoffs, early retirement, and attrition, rather than outright dismissal (Park, Duck Jay, et al, 2001), the change of laws that made dismissal easier likely facilitated the process of employment adjustment.
C. Labor Market Adjustments Since the Crisis
7. As the economy recovered, unemployment dropped briskly. By December 1999, unemployment had halved from its peak in February 1999. Since then, it has continued to fall steadily, and stood at 3 1/4 percent (714,000 individuals) by end-2001. The improvement has to some extent been due to “discouraged workers” dropping out of the labor market, as the labor force participation rate has fallen, from about 63 percent in 1996 to 61 percent by end-2001. Thus, in absolute numbers, employment has grown by only about 1 1/4 percent (or about 250,000 jobs) since 1997.6
8. Flexible compensation schemes have grown. The pickup in labor demand was accompanied by improved earnings, which grew by 11 percent in 1999. Strikingly, however, regular pay grew by only 6 percent, while overtime and bonuses both grew by more than 25 percent (see Figure II.3). The same pattern continued in 2000 and 2001, with the flexible component of monthly earnings growing faster than regular pay. Evidently, employers have remained cautious, increasing compensation through means that could be easily reversed if business conditions were to worsen. Thus, in 2001, as the economy slowed, overtime pay grew by only ½ percent, keeping overall earnings growth down to 7 percent, about the same as in 2000.
Employee Growth, by Type of Work (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employee Growth, by Type of Work (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employee Growth, by Type of Work (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Unemployment Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Unemployment Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Unemployment Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employment (annual average: 1997–1998)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employment (annual average: 1997–1998)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employment (annual average: 1997–1998)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
9. The share of temporary and daily workers in total employment, which was already high by OECD standards, has increased further. In fact, most employment gains have been in temporary and daily work. Between 1997 and 2001, the number of regular workers fell by 9 percent.7 By contrast, the number of temporary and daily workers grew by 9½ percent and 17 percent, respectively. Consequently, the share of temporary workers rose from 32 percent to 34 percent of the workforce; and daily workers from 14 percent to 17 percent (Figure II.6). Thus, more than half of all employees now hold “nonregular” jobs, and Korea now has the highest share of nonregular jobs among OECD countries. Nonregular workers are highly concentrated among women, older workers, and the less-educated (Figure II.7).
Employment, By Type of Warlear (In percent of total)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employment, By Type of Warlear (In percent of total)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employment, By Type of Warlear (In percent of total)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Nonregular Employment (In percent of total employment)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Nonregular Employment (In percent of total employment)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Nonregular Employment (In percent of total employment)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
10. The growth of nonregular employment is also reflected in the significant variations in working hours (Figure II.8). The number of regular hours has not changed much but overtime has been highly variable. Evidently, firms have been eliciting increased overtime instead of hiring more workers. Thus, labor costs would not be ratcheted up during an upturn and could be easily reduced during a downturn.
Growth of Hours Worked, By Type (In percent per appear)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Growth of Hours Worked, By Type (In percent per appear)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Growth of Hours Worked, By Type (In percent per appear)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employment, By Productive Sector (annual average: 1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employment, By Productive Sector (annual average: 1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Employment, By Productive Sector (annual average: 1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Average Earnings, By Productive Sector (1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Average Earnings, By Productive Sector (1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Average Earnings, By Productive Sector (1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Labor Productivity and Unit Labor Cost (1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Labor Productivity and Unit Labor Cost (1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Labor Productivity and Unit Labor Cost (1997=100)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
11. The shift to the service sector has accelerated. The shift to services in terms of both output and employment has been going on since the 1980s, and in this respect, Korea is becoming more like other OECD countries. Services now account for the bulk of employment. Between 1997 and 2001, the share of manufacturing employment fell from 23 percent to 19½ percent, while the share of services rose from 66 percent to 70 percent. In 2001, the strong demand for labor in the service sector, which tends to be more labor-intensive than manufacturing, helped keep unemployment low despite the economic downturn. Further, earnings growth has been slower in the service sector due to the fragmentation and small size of firms and the relative weakness of trade unions. Between 1997 and 2001, overall earnings grew by 5 percent, which comprised earnings growth of 6 percent in manufacturing and 3½ percent in services.
12. Labor productivity growth has been robust. Although productivity growth slowed in 2001 in connection with the downturn, over the longer period 1997–2001, the annual growth of labor productivity has averaged 13 percent. With wages having grown modestly, unit labor costs fell by an average of 5 percent per annum.
13. The role of women in the labor force has grown.8 Female labor force participation fell during the crisis, but has regained its previous level (see Figure II.2). The unemployment rate for women has always been below the rate for men, but the gap has widened (see Figure II.1). Consequently, the proportion of women in the work force has risen by about 2 percentage points. This increase, however, has taken the form of more women being hired as temporary and daily workers (see Figure II.7). Thus, among regular workers, the proportion of women has fallen by 3 percentage points since 1997. By contrast, among temporary and daily workers, the ratios have risen by 1 percentage point and 3 percentage points, respectively.
Unemployment Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Unemployment Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Unemployment Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Labor Farce Participation Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Labor Farce Participation Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
Labor Farce Participation Rate (In percent)
Citation: IMF Staff Country Reports 2002, 020; 10.5089/9781451822069.002.A002
14. Unionization has fallen. The onset of the crisis fostered closer cooperation between unions and employers under the auspices of the Tripartite Commission, due to the realization that the severe downturn required joint efforts. As the economy recovered in 1999, the number of labor disputes and days lost also rose. However, the percentage of workforce that is unionized has fallen from about 14 percent at the beginning of 1990s to about 12 percent after the crisis, and is concentrated in the larger firms and those in manufacturing. The unionization rate in Korea is now well below the OECD average of 40 percent.
Indicators of industrial relations
Indicators of industrial relations
| Labor disputes | Unionization | |||||
|---|---|---|---|---|---|---|
| No. of cases | Workers Days lost (in thousands) | Members (in thousands) | Rate (in %) | |||
| 1990–96 | 176 | 108 | 1,909 | 1,709 | 14.3 | |
| 1997 | 78 | 44 | 445 | 1,484 | 11.2 | |
| 1998 | 129 | 146 | 1,452 | 1,402 | 11.5 | |
| 1999 | 198 | 92 | 1,366 | 1,481 | 11.8 | |
| 2000 | 250 | 178 | 1,894 | … | … | |
Indicators of industrial relations
| Labor disputes | Unionization | |||||
|---|---|---|---|---|---|---|
| No. of cases | Workers Days lost (in thousands) | Members (in thousands) | Rate (in %) | |||
| 1990–96 | 176 | 108 | 1,909 | 1,709 | 14.3 | |
| 1997 | 78 | 44 | 445 | 1,484 | 11.2 | |
| 1998 | 129 | 146 | 1,452 | 1,402 | 11.5 | |
| 1999 | 198 | 92 | 1,366 | 1,481 | 11.8 | |
| 2000 | 250 | 178 | 1,894 | … | … | |
15. In sum, the main trends that have emerged in the aftermath of the crisis are:
The share of nonregular employment, which was already high by OECD standards, has increased further. The shift of employment to the service sector has also sped up.
Employers have increased their use of the flexible components (overtime and bonuses) of compensation. Working hours have also become more variable, due to increased reliance on overtime.
Female participation in the labor market has grown, especially among nonregular workers.
The increased flexibility of the labor market has been accompanied by robust growth of labor productivity.
D. Remaining Issues
16. The standard workweek is set to be reduced from 44 hours to 40 hours.9 The reduction has been agreed in principle by the Tripartite Commission. However, the actual date for implementation has not yet been set. The example of France, which recently reduced its workweek, shows the importance of giving employers enough flexibility to accommodate the reduction, in order to minimize the negative impact on productivity and costs.10
17. Although labor market flexibility has increased in Korea since the crisis, the high level of nonregular employment and its continuous rise has raised a number of questions. Is this development linked to the crisis, or is it a more structural trend? If the latter, what are the driving forces behind it? The evidence seems to support the view that the high share of nonregular workers is a more fundamental issue in Korea. The shift from regular to nonregular workers had already been happening prior since the early 1990s. The crisis, and the legalization of temporary work agencies, may therefore have accentuated the situation. Thus, during 1990–99, regular employment grew only by 2 percent, while the number of temporary and daily workers increased by 32 percent and 24 percent, respectively.
18. The large proportion of nonregular employment may be leading to excessively rapid turnover of workers and insufficient training. As the OECD (2000) notes, average tenure in Korea (six years for men and four years for women in 1997) is significantly below the corresponding OECD averages (10½ years and 8½ years, respectively, in 1995). Short tenures, in turn, are likely to reduce the incentive for employers to provide training, with negative implications for productivity growth.
19. The continuing impediments to workforce restructuring seem to be inducing the substitution of nonregular workers for regular workers.
A large portion of regular workers’ compensation come from bonuses, which are not generally paid to temporary or daily workers. In addition, severance pay, applicable to workers with at least one year of tenure, is quite generous by OECD standards.
Dismissal of regular workers, although allowable under law, must meet strict tests. First, employers must show that they have made every reasonable effort to avoid dismissals. Second, they must consult with workers on criteria for selecting those to be laid off. Third, they must give laid-off workers 60 days advance notice and the Ministry of Labor 30 days in cases involving large numbers of workers. Consequently, dismissal of regular workers is rarely used by employers.
20. Further reforms, therefore, are needed to level the playing field for both types of workers, and to strike an optimal balance between job protection and job creation. The OECD (2000) suggests the following measures:
Employment protection for regular workers could be further eased to reduce employers’ incentives for preferring for nonregular workers.
The cost of regular workers could be further reduced by gradually phasing out the separation allowance required for such employees.
Restrictions on temporary employment agencies could be reduced to enhance labor market flexibility. Currently, these agencies cannot participate in manufacturing, are limited to certain firms and only for one year (with a one-time extension), and need to renew their licenses frequently.
21. Further corporate restructuring will require the strengthening of social safety nets. Although Korea has achieved much in restructuring the corporate sector, many firms continue to have excessive leverage and low profitability. Their downsizing will necessarily entail further shocks to the labor force. The adjustment process would be eased if there were more efficient and effective safety nets in place. In this connection, the government plans to extend the coverage of the EIS to more categories of workers in the near future, as the system currently covers regular and temporary workers in all enterprises, but not part-time or daily workers.
APPENDIX Data Sources
This appendix describes the main sources of the data used in this paper.
The following data series are compiled by the Ministry of Labor and are available on their website (www.molab.go.kr) or the Korean National Statistics Office (www.nso.go.kr). They can also be downloaded from the CEIC© database by subscribers.
Population, labor force
Employment, unemployment (including seasonally adjusted)
Employment and earnings, by productive sector
Data on industrial relations (unionization and labor disputes) can be downloaded from the National Statistics Office’s website.
The following were obtained from the Report on the Monthly Labor Survey, which is published by the Ministry of Labor.
Components of earnings
Type of employment and hours worked
Data on labor productivity and unit labor costs were obtained directly from the Korea Productivity Center (www.kpc.or.kr). They are not currently available on the English version of the website.
References
Betcherman, Gordon, Amit Dar, Amy Luinstra, and Makoto Ogawa, 2001, “Labor Adjustment, Non-Standard Work, and Employment Programs: Korea in an OECD Context,” in Park, et al (Eds.).
Choi, Jungkiu, Wookjin Chung, and Sukyeong Kim, 2001, “Mobilizing South Korea’s Women,” McKinsey Quarterly Issue Number 4.
Kim, Dae II, 2001, “Labor Market Changes in Korea After the Crisis,” Paper Presented to the IMF-KIEP Conference on the Korean Crisis and Recovery, May 17–19, Seoul.
Min, Seung-kyu, 2001, “Preconditions for the Introduction of a Five-Day Work Week in Korea,” Korea Economic Trends, Samsung Economic Research Institute, September 1,
OECD, 2000, Economic Survey of Korea.
Park, Duck Jay, Jonghee Park, and Gyu-Chang Yu, 2001, “Assessment of Labor Market Response to the Labor Law Changes Introduced in 1998,” in Park et al (Eds.), 2001.
Park, Funkoo, Young-bum Park, Gordon Betcherman, and Amit Dar 2001, Labor Market Reforms in Korea: Policy Options for the Future. Seoul: World Bank and Korea Labor Institute.
Park, Yoon-bae, 2001, “Government Urged to Create More Jobs to Narrow Widening Income Gap,” Korea Times (December 11).
This paper was prepared by Hong Liang and Henry Ma (both APD).
A survey of labor market developments since the crisis is also provided in Kim (2001).
In Korea, those who have looked for work during the previous one week are counted as unemployed, while the ILO and OECD definitions include those who have looked for work in the preceding two weeks and four weeks, respectively. Hence, measured unemployment would be higher in Korea if the ILO or OECD definitions are used.
These components account for about a third of total compensation.
As a result, the proportion of wage workers covered by EIS rose from 33 percent to 70 percent. Daily workers and part-time employees working less than 80 hours a month remain ineligible.
If the labor force participation rate had stayed at 62 percent (the average for 1995–97), unemployment would have peaked at 14 3/4 percent in February 1999 (instead of 8½ percent), and would have been about 4 percent in end-2001.
Regular workers are those who work more than one year at a firm and are paid standard wages, plus bonuses and overtime. Nonregular workers are those who work for a set period of time, which can be extended or terminated at the discretion of the employer.
Also see Choi, et al., 2001, “Mobilizing South Korea’s Women.”
See France: 2001 Article IV Consultation—Staff Report, IMF Country Report No. 01/199, October 2001.
