Abstract
This Selected Issues paper for South Africa presents a quantitative analysis of inflation dynamics in the country. The conduct of monetary policy has been complicated by a variety of unanticipated events that have had important effects on inflation. Exposed to exchange rate and other shocks, the model confirms that a delayed policy response to inflation shocks leads to persistently higher inflation rates and, subsequently, to a sharp real contraction of the economy.
IV. Unemployment in South Africa: The Role of Labor Market Regulations1
A. Introduction
1. Despite an increase in employment during 2003 and 2004, unemployment in South Africa remains stubbornly high. Brisk real GDP growth over the past few years has contributed to some decline in unemployment, to 26.5 percent in March 2005 from over 30 percent in 2002. During the past five years, the government has streamlined labor legislation and introduced active labor market policies to reduce unemployment and, in 2003, announced its goal of halving unemployment by 2014.
2. There are at least three views—not mutually exclusive—on why unemployment is so high. First, a lack of available skills, largely a legacy of apartheid, poses a key structural constraint to investment and labor demand. Second, labor legislation and employment protection policies have some unintended side effects; in this view, the labor market is not as efficient as it could be. Third, official statistics may underestimate employment, particularly in the informal sector.2
3. This paper seeks to shed some light on the potential role of labor regulations in explaining the high rate of unemployment in South Africa. To do so, we put developments in a broader perspective and address three questions. Why is unemployment so high in South Africa? How do labor regulations in South Africa compare with those in other countries and what is their impact? How does the authorities’ strategy compare with findings in this paper?
4. The paper is organized as follows. Section B presents some stylized facts on South Africa’s labor market. Section C looks more closely at the three views on why unemployment is so high. Section D presents some new empirical evidence focusing on the role of labor market inefficiencies. Section E reviews the government’s major reform initiatives, including skills development, public works program, and initiatives to foster small and medium-sized enterprises. Section F presents some policy conclusions.
B. Some Stylized Facts
5. Following a period of unemployment that began to rise in the mid-1990s, labor market outcomes started to improve in 2003 (Figures IV.1 and IV.2).3 Although the labor force grew by 4 percent annually during 1995–2002, employment increased by only about 11/4 percent annually. Changes in the composition of the economy appear to explain part of the slow growth in employment (Arora and Ricci, 2005).4 The unemployment rate rose to just over 30 percent in 2002, before declining to 26.2 percent in September 2004 and then slighlty edging-up to 26.5 percent in March 2005. While employment continued to grow it was not sufficient to offset the impact of a growing labor force. But the overall unemployment rate hides significant differences in terms of regions, race and gender, age group, and skills.
South Africa: Basic Labor Market Developments
Citation: IMF Staff Country Reports 2005, 345; 10.5089/9781451966763.002.A004
Source: South African Reserve Bank.South Africa: Unemployment—Some Stylized Facts in 2004
Citation: IMF Staff Country Reports 2005, 345; 10.5089/9781451966763.002.A004
Source: Statistics South Africa.The regional unemployment rate varies between 20 percent in Western Cape and close to 30 percent in Eastern Cape, Kwasulu-Natal, and Free State.
The unemployment rate is higher for black people (exceeding 35 percent for black women) than for any other population group. It is fairly low for whites, on the order of 5 percent.
Youth unemployment is particularly severe, with 70 percent of the unemployed under the age of 35.
Workers with some skills (with grade 11 as highest level of education) face a high probability of being unemployed, the unemployment rate of this cohort is about 40 percent. In contrast, the unemployment rate is more moderate (15 percent) for people with no schooling and much lower for highly skilled people (below 5 percent).
More than 50 percent of the unemployed have been seeking a job for at least one year, and more than a third of the unemployed have had no job during the last three years.
6. The unemployment rate in South Africa, which is high by international standards is more than twice as high as the average of a group of 25 comparator countries. However, South Africa’s informal sector appears to be much smaller than in other emerging market economies. Whereas employment in the informal sector accounts for 20 percent of total employment in South Africa, it is estimated to be above 50 percent in countries such as Brazil, Mexico, Indonesia and the Philippines.
Unemployment Rate: International Comparison
(2004, in percent)
Citation: IMF Staff Country Reports 2005, 345; 10.5089/9781451966763.002.A004
Sources: IMF, World Economic Outlook.1/ Countries with per capita income between US$1,000 and US$15,000, a population > 10 million, and a sovereign foreign currency rating of B or higher.7. Over the past decade, South Africa has reformed its labor legislation and the process by which wages are determined. South Africa’s post-1994 labor legislation aims to strike a balance between strengthening the rights of workers and avoiding increasing firms’ costs. The wage formation process centers on four different types of mechanisms: collective bargaining with extension of the agreement to parties not participating in collective bargaining, the setting of minimum wages on a sectoral level by a statutory wage-fixing body (sectoral employment standards), voluntary collective bargaining at the plant or enterprise level, and direct contracts (see Box 1).
C. Why Is Unemployment so High?
Data deficiencies
8. Data deficiencies, particularly in the data measuring informal employment, may possibly play a role in explaining the high rate of unemployment. A striking feature of the stylized facts is that informal employment is apparently much lower in South Africa than in other emerging markets (see Kingdon and Knight, 2004). Although there is no commonly agreed definition of “the informal sector” internationally, three criteria are used: (i) registration of a business; (ii) employee registration, which commits the employer to pay benefits; and (iii) size of businesses (South African Treasury, 2005, p. 42).5 Because informal employment is inherently difficult to measure, South Africa’s low recorded share of informal employment may, at least in part, reflect data deficiencies. But other reasons may also play a role in explaining low informal employment in South Africa. For example, some people may prefer to remain unemployed or inactive rather than take a job in the informal sector because of the high reservation wage. It is unclear why the reservation wage in South Africa should be higher than in other emerging markets, although, theoretically, generous transfers within households or high transportation costs from rural areas and former townships to cities, where most of the jobs are, could play a role.
9. Results of a recent IMF technical assistance mission suggest that South Africa does, indeed, suffer from data deficiencies, but it is unclear how important they are. The technical assistance mission examined the Labor Force Survey questionnaire, sampling methods, estimation methods, and adherence to International Labor Organization (ILO) recommendations for employment and unemployment statistics. Its findings suggest that improvements to the questionnaire and techniques are needed. Recommended improvements would tend to increase measured employment and reduce the unemployment rate. However, it is not possible to say that the magnitude of the changes to the estimates would be significant.
Labor Legislation and Wage Formation in South Africa
South Africa’s labor legislation rests on the following statutes:
The Labor Relations Act defines the relations between employers, employees, and unions.1 It manifests the right to participate in the formation of trade unions, promotes collective bargaining at the sectoral level, regulates the right to strike and the recourses to lockout, and outlines the procedures for the resolution of labor disputes through statutory conciliation, mediation, and arbitration (for which purpose the Commission of Conciliation, Mediation, and Arbitration has been established).
The Basic Conditions of Employment Act (BCEA) sets the minimum standards for employment, including the work week, annual leave, overtime, maternity leave, and rules regarding the termination of contracts. No terms in a formal employment contract can be less favorable. The BCEA also defines the institutional framework for sector-specific minimum wages.
The Employment Equity Act addresses discrimination issues and promotes fair treatment and equal opportunities for groups, that have historically experienced some disadvantages, such as blacks, women, and people with disabilities. It contains affirmative action measures for firms, exceeding a specific size.
The Skills Development Act aims to encourage workers, unemployed people, and previously disadvantaged people to acquire new skills.
There are four different wage-setting mechanisms in South Africa: collective bargaining with extension of the agreement to nonparticipative parties, minimum wages set on a sectoral level by a statutory wage fixing body (sectoral employment standards), voluntary collective bargaining at the plant or enterprise level, and direct contracts.
Collective bargaining, through bargaining councils, dominates in many sectors. The bargaining system is relatively centralized. Agreements on wages struck between trade unions and the largest employers are extended by law to small and medium-sized enterprises (SMEs), but exceptions can be granted.
Minimum wages are applied in sectors not covered by collective bargaining agreements, namely the agricultural sector, domestic services, private security, and retail sectors. Employment standards and minimum wages are applied on the basis of recommendations of the Employment Standards Commission. There is no single national minimum wage in South Africa. Minimum wages differ across sectors for identical jobs.
Collective bargaining at the enterprise level takes place over and above sectoral collective bargaining, with the terms typically being more favorable for workers and specific to the enterprise.
Direct contracting plays the most important role in the upper market.
Skills
10. Under the apartheid regime, the non white population had great difficulty pursuing an education. Although the low level of workers’ skills is likely to contribute to the high rate of unemployment, it is difficult to determine by how much. It appears to constrain labor demand, but employers and employees have different perceptions of its importance. Survey results from South Africa’s Bureau of Economic Research indicate that the shortage of skilled labor is the second most important constraint for the manufacturing sector.6 However, other survey results suggest that less than 10 percent of the unemployed attribute their inability to find a job to a lack of skills.
South Africa: Reasons for Not Working
South Africa: Reasons for Not Working
| Reason | Unemployment thousands |
In Percent total |
|---|---|---|
| Has found a job, but is only starting at a definite date in the | 34 | 0.8 |
| Seasonal | 17 | 0.4 |
| Lack of skills or qualifications for available jobs | 357 | 8.6 |
| Cannot find any work | 3,345 | 80.9 |
| Cannot find suitable work | 152 | 3.7 |
| Recently retrenched | 110 | 2.7 |
| Other reason / other | 120 | 2.9 |
| Total | 4,135 | 100.0 |
South Africa: Reasons for Not Working
| Reason | Unemployment thousands |
In Percent total |
|---|---|---|
| Has found a job, but is only starting at a definite date in the | 34 | 0.8 |
| Seasonal | 17 | 0.4 |
| Lack of skills or qualifications for available jobs | 357 | 8.6 |
| Cannot find any work | 3,345 | 80.9 |
| Cannot find suitable work | 152 | 3.7 |
| Recently retrenched | 110 | 2.7 |
| Other reason / other | 120 | 2.9 |
| Total | 4,135 | 100.0 |
Constraints in South Africa’s Manufacturing Sector: Survey Results
Citation: IMF Staff Country Reports 2005, 345; 10.5089/9781451966763.002.A004
Source: Bureau of Economic Research.The role of labor market institutions
11. Labor market institutions play an important economic role in that they rebalance bargaining power and provide insurance to workers. Unions can help balance the power between large firms and individual workers during wage and other labor negotiations. Such institutions as minimum wages, social security benefits, and unemployment insurance schemes are intended to protect individual workers. Social security benefits are important in ensuring a minimum pension and health coverage, and unemployment insurance helps reduce an individual’s loss of income in the case of dismissal.
12. But poorly designed or overly generous labor market institutions may have unintended side effects and hamper employment creation. Overly generous minimum wages and generous social security benefits lead to a rise in the tax wedge, increase the cost of labor and tend to reduce employment. Widespread collective bargaining limits the ability of individual firms to negotiate wage agreements that reflect their specific circumstances.
D. Empirical Cross-Country Analysis
13. In this section we compare (i) the degree of flexibility of labor market institutions in South Africa with those in other countries, both industrial and developing; and (ii) assess the extent to which labor market inflexibility may contribute to unemployment. Based on standard indicators for industrial countries, previous staff work (Alleyne, 2000) suggests that South Africa’s labor market flexibility is roughly in line with that of countries in the Organization for Economic Cooperation and Development (OECD).
Panel analysis
14. We compare labor market institutions in South Africa with those in up to 75 industrial and developing countries over the period 1970–2000. We use five-year period averages and the following indicators for labor market institutions: (i) minimum wages, (ii) trade union density, (iii) employment protection and benefits, and (iv) size of the government sector.7
15. Labor market institutions in South Africa are similar to the average of a large group of countries. In principle higher values for each of the institutional indicators signal less flexible labor markets. For the 1995 to 2000 period, South Africa’s institutional indicators were somewhat below the world averages. However, it is noteworthy that whereas South Africa has no single national minimum wage, it does have sector-specific minimum wages, which are not captured here. The sector-specific minimum wage structure is fairly complex. Minimum wages differ significantly for similar types of jobs across sectors and this is likely to limit mobility across sectors. Minimum wages also vary for urban and rural areas.8
16. In the following panel estimation, the total unemployment rate is used as dependent variable. The rigidity of labor market institutions is measured by an index that combines the variables from each of the categories described above.9 The index is a simple average of the normalized values of the variables.10 It ranges between 0 and 1 for each period, with higher values indicating less flexible labor markets. As additional control variables, we include indicators for macroeconomic performance, trade openness (which can be interpreted as an indicator of competitiveness), and demographic characteristics of the labor force.
Labor Market Outcomes and Labor Market Institutions
Unweighted averages across groups of countries : 1995–2000
There is no national minimum wage in South Africa
Labor Market Outcomes and Labor Market Institutions
Unweighted averages across groups of countries : 1995–2000
| Variable | All | Industrial | Developing | ZAF | ||
|---|---|---|---|---|---|---|
| Labor market outcomes | ||||||
| Unemployment | ||||||
| Total | 9.1 | 8.7 | 9.4 | 29.3 | ||
| Female | 11.0 | 9.8 | 11.8 | 38.0 | ||
| Youth | 18.1 | 15.9 | 19.1 | 45.6 | ||
| Labor force participation | ||||||
| Total | 70.0 | 71.0 | 69.8 | 66.0 | ||
| Female | 54.7 | 60.3 | 53.5 | 49.4 | ||
| Labor market rigidity indicators | ||||||
| National minimum wage (Percent of GNI per capita) | 55.6 | 47.0 | 60.5 | n.a 1/ | ||
| Union membership (Percent of Labor Force) | 18.8 | 24.8 | 16.2 | 16.9 | ||
| General government employment (Percent of total employment) | 10.2 | 20.9 | 9.6 | 9.4 | ||
| Social security contribution (Percent of GDP) | 4.7 | 10.4 | 2.2 | 0.5 | ||
| Monthly unemployment benefit (Percent of monthly job earnings) | 49.9 | 55.0 | 47.3 | 45.0 | ||
There is no national minimum wage in South Africa
Labor Market Outcomes and Labor Market Institutions
Unweighted averages across groups of countries : 1995–2000
| Variable | All | Industrial | Developing | ZAF | ||
|---|---|---|---|---|---|---|
| Labor market outcomes | ||||||
| Unemployment | ||||||
| Total | 9.1 | 8.7 | 9.4 | 29.3 | ||
| Female | 11.0 | 9.8 | 11.8 | 38.0 | ||
| Youth | 18.1 | 15.9 | 19.1 | 45.6 | ||
| Labor force participation | ||||||
| Total | 70.0 | 71.0 | 69.8 | 66.0 | ||
| Female | 54.7 | 60.3 | 53.5 | 49.4 | ||
| Labor market rigidity indicators | ||||||
| National minimum wage (Percent of GNI per capita) | 55.6 | 47.0 | 60.5 | n.a 1/ | ||
| Union membership (Percent of Labor Force) | 18.8 | 24.8 | 16.2 | 16.9 | ||
| General government employment (Percent of total employment) | 10.2 | 20.9 | 9.6 | 9.4 | ||
| Social security contribution (Percent of GDP) | 4.7 | 10.4 | 2.2 | 0.5 | ||
| Monthly unemployment benefit (Percent of monthly job earnings) | 49.9 | 55.0 | 47.3 | 45.0 | ||
There is no national minimum wage in South Africa
17. Our results suggest that labor market rigidities are associated with a higher rate of unemployment. The index of labor market rigidity has the expected positive sign and is statistically significant in the presence of a number of controls. Values for its estimated regression coefficient are relatively stable throughout various specifications.
Unemployment and Labor Market Institutions: Panel Regression Results 1/
denotes ignificant at the 10 percent level;
denotes significant at the 5 percent level;
denotes significant at the 1 percent level.
Dependent variable is total unemployment. t-statistics in parentheses. Regressions include time-effects.
Excludes government employment.
Unemployment and Labor Market Institutions: Panel Regression Results 1/
| Variable | (i) | (ii) | (iii) | (iv) | (v) |
|---|---|---|---|---|---|
| Constant | 2.758 *** | 3.059 * | 24.959 *** | 15.738 *** | 17.57 *** |
| (3.37) | (1.83) | (3.42) | (3.64) | (3.91) | |
| Index of labor market rigidity | 5.311 *** | 12.988 *** | 10.184 ** | 10.587 ** | |
| (2.77) | (3.27) | (2.56) | (2.71) | ||
| Index of labor market rigidity modified 2/ | 8.692 * | ||||
| (1.80) | |||||
| Productivity growth | -0.019 * | -0.025 ** | -0.027 ** | -0.030 ** | |
| (-1.67) | (-2.17) | (-2.38) | (-2.63) | ||
| Change in CPI inflation rate | -0.001 *** | -0.001 | -0.001 ** | -0.0004 ** | |
| (-3.27) | (-2.37) | (-2.15) | (-1.97) | ||
| Real interest rate | 0.010 | 0.040 | 0.042 | 0.043 | |
| (0.14) | (0.54) | (0.58) | (0.56) | ||
| Trade openness | -0.145 *** | -0.18 *** | -0.19 *** | ||
| (-2.85) | (-3.54) | (-3.60) | |||
| Growth, ages population 15–64 | 0.102 | 0.12 | |||
| (1.24) | (1.36) | ||||
| R2 adjusted | 0.21 | 0.25 | 0.33 | 0.40 | 0.37 |
| Number of cross-sections (countries) | 75 | 49 | 48 | 45 | 45 |
| Number of observations | 313 | 123 | 120 | 114 | 114 |
denotes ignificant at the 10 percent level;
denotes significant at the 5 percent level;
denotes significant at the 1 percent level.
Dependent variable is total unemployment. t-statistics in parentheses. Regressions include time-effects.
Excludes government employment.
Unemployment and Labor Market Institutions: Panel Regression Results 1/
| Variable | (i) | (ii) | (iii) | (iv) | (v) |
|---|---|---|---|---|---|
| Constant | 2.758 *** | 3.059 * | 24.959 *** | 15.738 *** | 17.57 *** |
| (3.37) | (1.83) | (3.42) | (3.64) | (3.91) | |
| Index of labor market rigidity | 5.311 *** | 12.988 *** | 10.184 ** | 10.587 ** | |
| (2.77) | (3.27) | (2.56) | (2.71) | ||
| Index of labor market rigidity modified 2/ | 8.692 * | ||||
| (1.80) | |||||
| Productivity growth | -0.019 * | -0.025 ** | -0.027 ** | -0.030 ** | |
| (-1.67) | (-2.17) | (-2.38) | (-2.63) | ||
| Change in CPI inflation rate | -0.001 *** | -0.001 | -0.001 ** | -0.0004 ** | |
| (-3.27) | (-2.37) | (-2.15) | (-1.97) | ||
| Real interest rate | 0.010 | 0.040 | 0.042 | 0.043 | |
| (0.14) | (0.54) | (0.58) | (0.56) | ||
| Trade openness | -0.145 *** | -0.18 *** | -0.19 *** | ||
| (-2.85) | (-3.54) | (-3.60) | |||
| Growth, ages population 15–64 | 0.102 | 0.12 | |||
| (1.24) | (1.36) | ||||
| R2 adjusted | 0.21 | 0.25 | 0.33 | 0.40 | 0.37 |
| Number of cross-sections (countries) | 75 | 49 | 48 | 45 | 45 |
| Number of observations | 313 | 123 | 120 | 114 | 114 |
denotes ignificant at the 10 percent level;
denotes significant at the 5 percent level;
denotes significant at the 1 percent level.
Dependent variable is total unemployment. t-statistics in parentheses. Regressions include time-effects.
Excludes government employment.
18. The following additional findings emerge from our analysis.
Several macroeconomic shocks have a statistically significant impact on unemployment (column ii, see also Blanchard and Wolfers, 2000). Lower labor productivity growth can result in higher unemployment to the extent that real wages keep rising relatively fast. Disinflation can also cause an increase in unemployment, particularly if it is not fully anticipated and if the wage formation process is backward-looking. Higher real interest rates tend to restrain economic activity and may therefore contribute to higher unemployment. However, the coefficient on real interest rates is not significant at conventional significance levels.
Trade openness contributes to a decline in the unemployment rate (columns iii and iv).11 Our results indicate that additional trade liberalization should be conducive to a reduction in unemployment. Openness may stimulate competition, which, in turn, may lead to the use of more efficient production methods, with positive effects on growth and employment. Moreover, Ahmed, Arezki and Funke (2005) show that additional trade liberalization could help increase South Africa’s attractiveness as a destination for foreign direct investment, which could also stimulate employment and growth.
Although demographic changes may also have an impact on unemployment, the coefficient for the growth in the working-age population, presented in column iv is statistically insignificant, though with a positive sign.
19. Results are also relatively robust with respect to changes in the calculation of the labor market rigidity index. Theoretically, the size of the government sector may have an ambiguous effect on unemployment, although there is some empirical evidence that a larger government sector is associated with higher rates of unemployment.12 Column v presents estimation results for a modified labor rigidity index that excludes the government employment component. The results are similar to those in column iv.
20. Overall, these results suggest that a reduction in labor market inflexibilities can be expected to lower the unemployment rate. But, as measured by these indicators, South Africa’s labor market institutions are similar to those in many other countries. The indicators used so far are fairly broad and may not capture some important constraints. Anecdotal evidence suggests that the costs of dismissing workers are relatively high in South Africa (see also Levy, 2003); and these are not captured in the above indicator.
The role of the cost of hiring and dismissing workers
21. In this section, we compare labor market inflexibilities, using a new data set that focuses more specifically on the cost of hiring and dismissing workers. The World Bank’s 2004 Cost of Doing Business Database measures the flexibility of labor laws on the basis of alternative indicators: (i) the difficulty of hiring a new worker13, (ii) restrictions on expanding or contracting the number of working hours, (iii) procedural difficulties of dismissing a redundant worker, (iv) an average of the three indices (rigidity of employment index), and (v) the pecuniary cost of dismissal (“firing costs”).
22. On the basis of these new indicators, South Africa’s labor laws are somewhat more inflexible than those in other regions but less inflexible than, for example, Brazil’s (Figure IV.5, Table IV.4) 14 This is true in particular, of the difficulty of hiring and dismissal. The costs of dismissing and retrenching workers represent a significant proportion of the total labor costs facing businesses and, therefore, act as a deterrent to the recruitment of new workers. Legislative amendments passed in 2002 have helped streamline reconciliation and arbitration procedures, but some bottlenecks remain.
Labor Market Inflexibilities
Citation: IMF Staff Country Reports 2005, 345; 10.5089/9781451966763.002.A004
Labor Market Rigidity 1/
A larger value of the indicator implies more cumbersome or costly hiring and dismissal procedures. The index in the last column is an average of the other three indices.
Labor Market Rigidity 1/
| Difficulty of Hiring Index |
Rigidity of Hours |
Difficulty of Dismissal Procedures |
Rigidity of Employmen Index |
||
|---|---|---|---|---|---|
| East Asia and | 20 | 30 | 22 | 24 | |
| Europe and Central | 31 | 51 | 42 | 41 | |
| Latin America and Caribbean | 44 | 53 | 34 | 44 | |
| Middle East and North Africa | 22 | 52 | 40 | 38 | |
| OECD: High | 26 | 50 | 26 | 34 | |
| South Asia | 37 | 36 | 53 | 42 | |
| Sub-Saharan | 53 | 64 | 50 | 56 | |
| South Africa | 56 | 40 | 60 | 52 | |
A larger value of the indicator implies more cumbersome or costly hiring and dismissal procedures. The index in the last column is an average of the other three indices.
Labor Market Rigidity 1/
| Difficulty of Hiring Index |
Rigidity of Hours |
Difficulty of Dismissal Procedures |
Rigidity of Employmen Index |
||
|---|---|---|---|---|---|
| East Asia and | 20 | 30 | 22 | 24 | |
| Europe and Central | 31 | 51 | 42 | 41 | |
| Latin America and Caribbean | 44 | 53 | 34 | 44 | |
| Middle East and North Africa | 22 | 52 | 40 | 38 | |
| OECD: High | 26 | 50 | 26 | 34 | |
| South Asia | 37 | 36 | 53 | 42 | |
| Sub-Saharan | 53 | 64 | 50 | 56 | |
| South Africa | 56 | 40 | 60 | 52 | |
A larger value of the indicator implies more cumbersome or costly hiring and dismissal procedures. The index in the last column is an average of the other three indices.
23. The lack of time-series data makes it difficult to estimate properly the effect of these rigidities on unemployment. Results from a simple cross-section analysis suggest that a 10 percentage point reduction in the labor market rigidity index may, on average, be associated with a reduction in the unemployment rate of about ½ of 1 percentage point. At first glance, this effect may appear small. However, the results refer to an average country in the sample, with an unemployment rate of below 10 percent. Moreover, this index focuses on hiring and dismissal procedures and does not capture the effects of other labor market institutions.
E. Ongoing Initiatives to Reduce Unemployment
24. The authorities are using several instruments to reduce unemployment. For example, they are implementing a National Skills Development Strategy, and in the 2005 budget, they introduced measures to support small business development. Other initiatives focus on young people.
25. In March 2005, the government launched the Nationals Skills Development Strategy (NSDS) for the period 2005 to 2010. Similar to its predecessor (NSDS 2001-05), the new NSDS aims to enhance people’s skills through training initiatives and learnership programs. The strategy also has equity targets, and all social partners have agreed to aim at accelerating Broad-Based Black Economic Empowerment and Employment Equity. More than three million workers have received training since 2001. By October 2004, the government has already met its March 2005 target of having 80,000 people under the age of 35 in learnerships. The NSDS is funded through the skills development levy, a 1 percent payroll tax. Employers offering training receive a partial refund of the levy. Moreover, the 2005 budget proposes widening the relief for small businesses from this tax.
26. An expanded public works program, launched in 2004, is designed to create one million temporary jobs within the next five years. This labor-intensive program seeks to promote economic growth and development by offering people temporary jobs while they acquire skills. Employment opportunities are created in government-funded infrastructure projects, environmental and cultural programs, and social programs involving home-based care and early childhood development.
27. Among other measures in the 2005 budget are several designed to foster small businesses: tax relief, streamlined filing obligations, and administrative assistance. More specifically, the 2005 budget proposes to put in place (i) community tax helpers to educate and help small businesses with tax administration; (ii) a small business help desk in the form of a call center; (iii) accounting and payroll packages for small businesses in the form of software and web-based systems; (iv) a small retailers value-added tax (VAT) package, which provides small businesses for a simplified method of accounting for the VAT; and (v) VAT education. The promotion of small businesses may be expected to generate additional employment.
28. The government has set up other initiatives with a view, in particular, to tackling youth unemployment. In 2001, it established the Umsobomvu Youth Fund to address the challenges facing the youth in the country. Its mandate is to promote job creation and skills development among young people between the ages of 18 and 35 by making strategic investments to contact, inform, and counsel youth groups; and by creating opportunities for young people to acquire skills and pursue meaningful self-employment opportunities. In a broader perspective, the National Youth Service is a program that engages young people in providing services to the community. It helps participants obtain on-the-job learning while providing a service to the community.
F. Policy Conclusions
29. The evidence presented in this paper suggests that low skills and labor market inflexibilities contribute to South Africa’s high rate of unemployment. It is apparent that a multifaceted strategy is needed to tackle the problem. Because of possible statistical deficiencies, employment in the informal sector may not be measured accurately. Yet it is difficult to gauge the contribution of each of these factors to unemployment. The relatively small size of the informal sector remains somewhat of a puzzle and requires further investigation. Although improvements to the Labor Force Survey—to the questionnaire and techniques—may shed further light on this issue, it is not possible to say that the magnitude of the changes to the unemployment estimates would be significant.
30. The authorities’ strategy for reducing unemployment—a top official priority—comprises a number of useful initiatives: the programs to enhance workers’ skills and provide temporary jobs in infrastructure and other projects, as well as measures to foster small business development.
31. The empirical evidence presented in this paper suggests that an easing of labor market inflexibilities tends to lower unemployment. Ongoing initiatives need to be reinforced with labor market reforms that reduce the cost of labor.
Although South Africa’s labor market institutions are, in some ways similar to the OECD average, South Africa scores lower than comparator countries in some areas, such as how difficult is to dismiss workers. The government could stimulate job creation by further streamlining dismissal procedures, as well as by educating stakeholders on the legal requirements for dismissal; reportedly some of the costs arise from misinterpretations of the law.
Reforms should include reducing the scope of centralized collective bargaining. Although reducing the administrative and regulatory burden for small businesses, as announced in the 2005 budget, is a step in the right direction, job creation is most likely constrained by collective wage bargaining. The extension principle, by which agreements on minimum wages struck between unions and employers are extended to small and medium-sized companies (SMEs), reduces wage flexibility. Although labor laws allow for exemptions, which have often been granted, the process of applying for an exemption can be expected to increase costs. International experience confirms that, in a number of countries with a relatively low rate of unemployment, the wage formation process for SMEs is characterized by considerable flexibility, for example in the form of bargaining at the enterprise level.
The minimum-wage structure should be reviewed and simplified. Although there is no single national minimum wage in South Africa, the sector-specific minimum wage structure is fairly complex and likely to limit mobility between regions and sectors. Simplifying the minimum wage structure and moderating periodic minimum-wage adjustments to reduce labor costs would stimulate job creation and facilitate mobility across sectors and industries.
32. Given South Africa’s recent strong economic performance and generally favorable outlook, it is an opportune time to move ahead quickly with critical reforms of the labor market.
G. References
Adrogué, Ricardo (2004), “Labor Markets in Brazil—Understanding the Rise of Informality”, Selected Issues Paper (Washington: International Monetary Fund).
Ahmed, F., R. Arezki, and N. Funke (2005), “The Composition of Capital Flows: Is South Africa Different”, IMF Working Paper, WP05/040, (Washington: International Monetary Fund).
Alleyne, Trevor (2000), Labor Legislation in South Africa, 1995–98, Selected Issues Paper. (Washington: International Monetary Fund).
Arora, V. and L. Ricci (2005), “Unemployment and the Labor Market”, in: Post-Apartheid South Africa: The First Ten Years, ed. by Michael Nowak and Luca Antonio Ricci (Washington: International Monetary Fund), forthcoming.
Bertola, G., D. Blau and L.M. Kahn (2002), “Labour Market Institutions and Demographic Employment Patterns”, NBER Working Paper 9043, July.
Bhorat, H. 1999, “The October Household Survey, Unemployment and the Informal Sector: A Note”, The South African Journal of Economics Vol. 67, no. 2, pp. 320–326.
Bhorat, H. (2004), “Labor Market Challenges in the Post-Apartheid South Africa”, South African Journal of Economics, Vol. 72, no. 5, pp. 940–975.
Black, P. A., and Rankin, N., 1998, “On the Cost Increasing Effect of the New Labour Laws in South Africa”, South African Journal of Economics, Vol. 66, no. 4, pp. 452–463.
Blanchard, O. and J. Wolfers 2000, “The Role of Shocks and Institutions in the Rise of European Unemployment: The Aggregate Evidence”, Economic Journal, 110, pp. 1–33.
Botero, J, S. Djankov, R. La Porta, F. Lopez-de-Silanes, and A. Shleifer (2004), The Regulation of Labor”, Quarterly Journal of Economics, Vol. 119, pp. 1339–1380.
Casale, D., C. Muller, and D. Posel, 2004, “A Reconsideration of the Rise in Employment in South Africa, 1995—2003”, South African Journal of Economics, Vol. 72, no. 5, pp 978–1002
Chandra, Vandana, L. Moorty, B. Rajaratnam, and K. Schaeffer, 2000, “Constraints to Growth and Employment in South Africa: Statistics from the Large. Manufacturing Firm Survey”, The World Bank Southern Africa Department, Discussion Paper, no. 14 (Washington: World Bank).
InterAmerican Development Bank (IDB) 2004, “Good Jobs Wanted: Labor Markets in Latin America” (Washington: Inter-American Development Bank).
International Monetary Fund 2003, “Unemployment and Labor Market Institutions: Why Reforms Pay Off?”, World Economic Outlook, Chapter IV, pp. 129–163 (Washington: IMF).
Kingdon, G. and Knight, J., 2004, “Unemployment in South Africa: The Nature of the Beast”, World Development, Vol. 32, no. 3, pp. 391–408.
Levy, Andrew, 2003, “Some Notes on South African Labour Legislation and its Effect on the Cost of Employment”, mimeo.
Nickell, S.J 2003, “Labour Market Institutions and Unemployment in OECD Countries”, CESifo DICE Report 2/2003 (München: Institute for Economic Research).
Rama, M., and R. Artecona. 2002. “A Database of Labor Market Indicators Across Countries” Washington, DC: World Bank.
Scarpetta, S. (1996), “Assessing the Role of Labor Market Policy and Institutional Settings on Unemployment: A Cross-Country Study”, OECD Economic Studies, Vol. 26, no. 1, pp. 43–98.
South Africa, National Treasury, 2005, Budget Review 2005.
World Bank, 2005. “World Development Report 2005: A Better Investment Climate For Everyone”, (Washington: World Bank).
Prepared by Norbert Funke and Victor Lledo (both AFR).
Moreover it is widely acknowledged that structural reforms that would lead to an increase in South Africa’s growth potential should also be conducive to job creation.
Bhorat (2004) analyzes the South African Labor Market challenges since 1994 in greater depth.
Casale, Muller, and Posel (2004) argue that some portion of the increase in employment may be the result of changes in data definitions.
In South Africa, “the informal sector consists of those businesses that are not registered in any way. They are generally small in nature, and are seldom run from business premises. Instead, they are run from homes, street pavements or other formal arrangements” (Statistics South Africa Labor Force Survey, September 2004, p. xxv).
“The survey results are obtained from questionnaires completed by senior executives in the trade, manufacturing and building sectors during the final month of each quarter. The constraint indices are calculated as 2/3 (gross percent of respondents reporting a certain constraint to be a serious problem) + 1/3 (gross percent reporting a slight problem”). See Bureau of Economic Research (2005), “Manufacturing,” University of Stellenbosch.
Other variables, such as the tax wedge, are also expected to play an important role. However, comparable time-series data are not available for this large sample. Data are from Rama and Artecona (2002). Scarpetta (1996), IMF (2003) and Nickell (2003) review the empirical literature for OECD countries, and IDB (2004) covers Latin American countries. For a microeconomic analysis based on enterprise survey data for South Africa, see Chandra and others, (2000, 2002).
For example, minimum wages in the domestic sector for workers who work more than 27 hours in rural areas are about US$115 (R754.65) per month and in urban areas US$140 (R930.15) per month.
Ideally, one would like to include in the regression all institutional indicators simultaneously so as to identify their individual effects without incurring an omitted variable bias. However, the observed high colinearity among some of the institutional indicators (for instance, the correlation between trade unions and public employment is above 0.5) would produce estimates with high variance, thus compromising their accuracy.
The choice of indicators was motivated by data availability. The normalization procedure consists in transforming each of the indicators in a variable between 0 and 1 that ranks countries for each period in this continuum. Let xi, xmax, xmin be, respectively, the variable x for country i along with the maximum and minimum across all countries in any given period. Its normalized value x0i therefore
Trade openness is defined as the sum of total exports and imports as a share of GDP.
This index comprises information on the availability and flexibility of part-time contracts and fixed-term contracts as well as information on minimum wages.
The labor indices were substantially revised between 2003 and 2004 following comments in various presentations and the publication of the background paper (Botero and others 2004). So while there was no reform of labor legislation in South Africa during that period, the 2004 indicators made it look worse than in 2003.
