An earlier section (Changes in employment structure and education intensity) showed that the availability of secondary-educated workers has expanded far faster than structural change would require algebraically, so that education levels within almost every subsector of the economy have risen. It also finds many of the newly educated employed in sectors that are shown to have very low returns. The previous section (Returns to education) went on to show many bottlenecks due to the limited availability of workers with basic education easing. However, it is premature to conclude that more education is not necessary. It remains possible that rising education levels within subsectors result from technological changes.
Figures 3.2.3–9 show how the distribution of education and the wage returns have altered among employees within several occupations in the four countries. Table 3.2.8 provides estimates of the relative importance of these activities over time.
While much could be said about the details of these figures, three points stand out. First, within any given occupation, education levels differ tremendously across countries. Consider drivers (Figure 3.2.3), the median among whom have around 6 years of education in Thailand, 7 years in India, 9 in Indonesia, and 10 in the Philippines. Similar discrepancies are observed in all occupations, with Indian and Thai workers always having the lowest education levels, Indonesians occupying the middle, and Filipinos always being the most educated. Given the broad similarity of technologies utilized by drivers and household helpers across countries, it is difficult to conceive of good reasons why demand for education in these activities should vary so much. It seems likely that these are residual categories into which workers of any education category may fall, rather than face unemployment. The detailed examination of wages in these jobs, below, reinforces this interpretation.
Second, the wage returns align with the more aggregate results from the previous section. Education wage premiums in most of these services sector occupations have fallen in the Philippines and Thailand, but appear to have held firm in India.
Third, the first four occupations in Table 3.2.8 are clearly not what one has in mind when one speaks of the "knowledge economy. "Yet to be sure, they account for a substantial, and in the Philippines and India-a growing-share of nonagricultural employment. The case that transformation requires more education across the board must hinge on some notion that these rather significant sources of employment will be phased out soon.
With regard to the details of jobs, it can be seen that drivers (Figure 3.2.3 above) account for an arrestingly high percentage of nonagricultural male employment in both the Philippines and Thailand. For this large portion of the Southeast Asian male labor force, technology is not propelling rising education levels, as it simply has not changed. Jeepneys, buses, motorized tricycles, and taxis have not evolved much, if at all, over the sample period. Further, the returns to education among drivers are extremely low. Indeed, when one considers that less-educated drivers reside disproportionately in rural areas with lower costs of living, the real return to education for drivers in India, Philippines, and Thailand could be around zero. In other words, these men did not acquire their education in the aspiration of becoming drivers. Yet in these three countries, drivers are significantly more educated now than they used to be. Furthermore, if it is presumed that productivity differentials between drivers will be reflected in wage differences, then the zero premium on schooling indicates that there is nothing intrinsic to the technologies operated by drivers that requires education.
Female household helpers are considered next (Figure 3.2.4). Filipino maids earned essentially no wage premium on schooling in 1991 or in 2004, while returns to LS education and below for Indian maids have also flattened to zero. Yet education levels among maids of both nationalities have risen-especially in the Philippines. The interpretation is exactly the same as for drivers.
3.2.3 Education profile and wage indexes of male drivers
N = none; IP = incomplete primary; M = middle; P = primary; ILS = incomplete lower secondary; LS = lower secondary; HS = higher secondary; IT = incomplete tertiary; T = tertiary.
Note: Limited to commercial drivers and to those reporting wages.
Sources: India National Sample Survey Organisation, Socio-economic Survey, Schedule 10, 1993/94, 2004; Indonesia SAKERNAS 1994, 2004; Philippine Labor Force Survey, 1991, 2004, October rounds; Thailand Labor Force Survey, 1995, 2005, October rounds.
a % of total male nonagricultural employment. b % of total female nonagricultural employment. c % of total nonagricultural employment.
Sources: India National Sample Survey Organisation, Socio-economic Survey, Schedule 10, 1993/94, 2004; Indonesia SAKERNAS 2004; Philippine Labor Force Survey, 1991, 2004, October rounds; Thailand Labor Force Survey, 1995, 2005, October rounds.
In Thailand too, the wage premium for education among maids has fallen to nearly zero. Yet in contrast with the Philippine case, the education level of Thai maids has fallen, suggesting that the more-educated maids are finding better-paid employment.
Some tasks that underwent technological change are examined next, including retail sales, bookkeeping, and secretarial work. In each of these, economists would predict a return to education, because more-educated workers can negotiate new technologies more easily. As expected, in each of these professions, the wage premiums on secondary school are higher than those earned by drivers, maids, and security guards.
The above analysis suggests that workers of different education levels can be easily interchanged with each other, as education profiles of occupations vary greatly across countries. Countries with higher education levels in aggregate ended up with more-educated workers in every occupation. Some of these jobs, which are large employers, have not experienced technological change, and others do not even pay a premium for more-educated workers, implying that workers did not obtain their education in order to get these jobs. Finally, education indeed reaps a higher premium in those occupations that have encountered technological change.
Some will argue, rightly, that the analysis focuses disproportionately on lower-status or mechanical jobs, to the exclusion of more "cerebral" professions. One argument for looking at these professions is that the flowering of supply chains and just-in-time inventory management may have increased the cognitive complexity of management jobs. Casual empiricism also confirms that the value of education in these professions is climbing. However, the focus of the analysis was determined by a noteworthy statistical constraint. For most of the cerebral occupations scrutinized (e.g., bank managers, software technicians, pure engineers, and retail sales managers) the samples were too thin to be statistically useful for estimating education profiles or wage premiums. There is nothing intrinsic to the sampling schemes employed by the labor force surveys that would cause these professions to be undersampled. There are simply very few people employed in such activities in the four sample countries.
A further feature of many of the cerebral professions is that education requirements are mandated by law or regulations (doctors, lawyers, nurses, engineers, bureaucrats, etc.). In this context there is very little scope for examining education intensification or changes in wage profiles within these groups, as Figure 3.2.9 shows clearly.
3.2.4 Education profile and wage indexes of female household helpers
N = none; IP = incomplete primary; M = middle; P = primary; ILS = incomplete lower secondary; LS = lower secondary; HS = higher secondary; IT = incomplete tertiary; T = tertiary.
Note: Limited to those reporting wages.
Sources: India National Sample Survey Organisation, Socio-economic Survey, Schedule 10, 1993/94, 2004; Indonesia SAKERNAS 1994, 2004; Philippine Labor Force Survey, 1991, 2004, October rounds; Thailand Labor Force Survey, 1995, 2005, October rounds.
3.2.5 Education profile and wage indexes of male security guards
N = none; IP = incomplete primary; M = middle; P = primary; LS = lower secondary; HS = higher secondary; T = tertiary.
Note: Limited to those reporting wages.
Sources: India National Sample Survey Organisation, Socio-economic Survey, Schedule 10, 1993/94, 2004; Indonesia SAKERNAS 1994, 2004; Thailand Labor Force Survey, 1995, 2005, October rounds.
Top of page
