Catalog

Work and joblessness

Labor force survey data

The latest labor force survey data from each country are used to analyze the profile of their young populations. Labor force surveys are nationwide investigations of households conducted by the national statistical agency of each country. The surveys cover work-related, demographic, and socioeconomic characteristics of the population and are primarily geared to providing information on labor conditions in the country. Using sophisticated random sampling techniques (with national censuses as sampling frames), unbiased estimates of the national structure of employment, unemployment, education attainment, and wages are calculated. For each country two sampling periods are used, 10–15 years apart, one in the 1990s and the other in the early years of this century: India (1993/94 and 2004/05), Indonesia (1994 and 2006), Philippines (1991 and 2006), and Thailand (1995 and 2005). Sample sizes are large and range from about 140,000 people in the Philippines in 1991 to about 600,000 in India in 2004.

Before the results are given, several potential weaknesses of the data should be elucidated. The labor force survey data capture statistics of sample households, hence the results may be subject to sampling errors. The sampling in each country is undertaken at different frequencies and durations. In the annual (or less than annual) surveys, the timing of the survey may impact the results, especially for young people, since the "seasonal" component of youth behavior is particularly strong (compared with older age groups), as, for example, if the labor force survey is undertaken during a school holiday. Similar effects could arise if the survey is carried out during a harvest period or tourist season.

The questionnaires used in the labor force surveys not only differ among countries, but also change over time within each country. To maintain comparability, all questions and variable definitions were mapped carefully against each other using applicable and related International Labour Organization (ILO) definitions. There were also considerable changes in industrial and occupational classifications, but as most of these are based on different versions of the International Standard Industrial Classification of all Economic Activities and International Standard Classification of Occupations, a concordance was made to link the datasets. These results should be read with these caveats in mind. Given the differences across countries between the two sampling periods, the comparisons should be seen in qualitative terms, and in relative rather than absolute magnitudes.

Youth unemployment

Figure 2.1.5 presents estimates of male and female unemployment rates for the four countries based on their latest labor force survey, and compares those rates with averages for developing Asia, industrial countries and the world. It would seem that while youth unemployment rates in developing Asia are high (11.3%), they are marginally below broad averages for industrial economies (12.7%), and for the world (13.2%) as of 2006. But in Indonesia and the Philippines the unemployment rates for young men and more so for women, are far higher than those averages. On this indicator, the young people of India and Thailand fare better.

Unemployment versus joblessness

Commentators have questioned the adequacy of the youth unemployment rate as an indicator of youth labor market conditions (for example, Rees 1986, Bowers et al. 1999, Ryan 2001, Fares et al. 2006, ILO 2006, World Bank 2006b, O'Higgins forthcoming). A lower unemployment rate, for instance, may stem from a contraction in the labor force participation rather than from an increase in employment. Specifically, attention has begun to focus on discouraged young workers who are excluded from youth unemployment statistics, that is, young people who are neither in education nor employment and who are not actively searching for work. Young people who become discouraged when jobs are difficult to find may drop out of the labor force. (This group includes all those who are not in education or employment.) In other words, young people who are not engaged in a "useful" or "productive" activity—so-called jobless youth.

In this chapter, the jobless rate is defined as the percentage of all young people who are neither in education nor employment. This is a useful indicator because it takes account of discouraged young people, who are the most in need of support in terms of education, training, and counseling to minimize the risks of them becoming entirely detached from the labor market. The jobless rate gives a sense of the size of youth labor market problems in relation to the youth population as a whole.

It is perfectly possible for youth unemployment rates to be very high but, if labor force participation is very low, to affect only a very small proportion of the young. Using the jobless rate as an indicator also reduces the problems of comparison over time and among countries, as the definitional requirements are much less strict than for the unemployment rate. The jobless rate may be bigger (or smaller) than the unemployment rate according to whether the proportion of the inactive population that does not participate in education is greater (or less) than the proportion of the active population that is unemployed.

As evident from Figure 2.1.6, unemployment rates do indeed gloss over the presence of large numbers of discouraged workers in all the four Asian countries. Rates of joblessness are much higher than conventionally defined rates of unemployment. The difference is particularly stark for India, where rapid economic growth has not translated into equivalent job creation. Though youth joblessness has declined there, the reduction has been marginal. As explained later, the decline appears to have come mainly from a reduction in joblessness among young women. In the three Southeast Asian countries, joblessness among young people has worsened, and has recently been highest in Indonesia. Thailand has maintained the lowest youth jobless rate among the four countries.

The jobless rate also gives a rather different picture regarding the relative position of different categories of young people. This is discussed in the following sections.

Young women versus young men

Young women normally face significantly more difficult labor market conditions than young men. This is captured by their greater jobless rates (Figure 2.1.7). Although the difference is less severe in Southeast Asia than in India, jobless rates have risen among men and women. In contrast, young Indian women—though still highly disadvantaged in comparison with men—have enjoyed a fall in joblessness as the country tapped into their rising potential, especially in urban areas. For instance, nearly 50% of young software workers in 2004/05 were women. Cultural perceptions of gender roles influence women's opportunities to plan a career. A wide literacy gap between boys and girls is partly responsible for higher joblessness among the latter. In Asia, parents, particularly in low-income families, prefer to send sons to school first and poor young girls are often left with the option of getting married, resulting in a movement from one poor household to another. Rigid working arrangements also affect female labor supply.

Urban versus rural young people

Relatively more urban young people used to be jobless than their rural counterparts in the 1990s. But the early years of this decade have brought a reversal in the pattern in three of the four countries. Joblessness is now usually higher among the young in rural areas than in urban areas (Figure 2.1.8). The Philippines is an exception where internal migration is an essential element of rural livelihood strategies and rural transformation, not just a way to escape rural areas (Quisumbing and McNiven 2005). Family networks in the countryside support large flows of migrants attracted by prospects of better education and higher wages in the towns and cities of the Philippines, where 75–80% of first-time migrants live with relatives and acquaintances.

Young adults versus teenagers

Joblessness is much higher among young adults (aged 20–24) than teenagers (aged 15–19) (Figure 2.1.9). This is most probably because of the higher proportion of teenagers who are still in full-time education. Both these age groups in Indonesia, Philippines, and Thailand experienced an increase in joblessness following the financial crisis. Job-hopping by young adults is one cause of their higher joblessness.

The inexperience of young workers is reflected in the lack of satisfaction in their initial job choices. For example, Topel and Ward (1992) find that a typical young man changes jobs seven times in the first 10 years of his career, and that transition to a higher-wage job stabilizes employment. Similarly, Neal (1999) finds that the probability of high-school graduate boys leaving their first job in a given career is 70% for workers with fewer than 7 years of experience. Once they exceed this period, the probability drops to 45%.

Educated versus uneducated

As the less educated grow older, their jobless rate climbs higher than that of the more educated. On the whole, there is an inverse relationship between education level and joblessness for those aged 25–34 (Figure 2.1.10). The most marked effects are for those with a tertiary level of education and those who have not completed even the most elementary level. The relationship is strongest in Thailand, where those without any education are five times as likely to be jobless as those with a bachelor's degree (or higher).

School-to-work transition and opportunity cost of waiting for a job

For those who have finished school, the transition to work takes time. Before finding a job, young people have to wait for an appropriate opening, because of their limited skills and their inexperience. The duration of transition indicator of the Organisation for Economic Co-operation and Development is the difference between the median ages of school-leaving and entry into employment, and these can be calculated from labor force survey data (Figure 2.1.11).² A lower difference does not mean that the typical young person takes less time to find a job but that the incidence of joblessness is much lower. A comparison of data between the two data time points shows that the number of schooling years has generally increased, culminating in average school-leaving ages of 17–19 years. The age at which young people start finding jobs has either increased or remained around the same, at 20–22 years. As a result, the estimated duration of waiting time in the job market varies between 1 and 5 years across the four countries. This is despite the fact that labor forces have generally received more education over the past 10 or 15 years, and more so in the case of younger workers.

As young people leaving school typically do not immediately find employment, joblessness increases. However, high joblessness need not always be a sign of labor market difficulties if flows into and out of work are high and spells of unemployment short. But equally, low joblessness does not rule out problems. If flows are low and spells are long, low joblessness may be associated with chronic problems (Ryan 2001).

Remaining unemployed while searching for a job results in a loss of earnings. This loss can be calculated as the product of the median length of waiting time in the job market and the real median wage for young workers. This loss is calculated for young workers with 0–5 years of experience and those with 3–6 years of experience (Table 2.1.1). The loss has increased over time, reflecting the higher opportunity cost of searching for a job. The comparatively high losses for more experienced young people in India and less experienced ones in the Philippines echo changes in the median wages.