World Bank Reprint Series: Number 402 George Psacharopoulos and Ana Maria Arriagada The Educational Composition of the Labour Force An Internaltional Comprrison Reprinted with permission from the Internaltion1al Laultr Review (Geneva, International Labour Office), vol. 125, no. 5 (September-October 198oi, pp. 561-574. International Labour Review, Vol. 125, No. 5, September-October 1986 The educational composition of the labour force: An international comparison George PSACHAROPOULOS and Ana Maria ARRIAGADA * Policy-makers and analysts typically use differences between the school enrolment ratios of different countries or regions to justify educational investment. (For the example par excellence in this respect see Harbison and Myers, 1964.) In this article we argue that although the enrolment ratio is a readily available statistic, it may not be the most appropriate indicator for setting investment priorities in education. Instead we have compiled and present here statistics on the educational composition of the labour force, which we believe constitutes a more sensitive human capital index. Flows versus stocks At any given point in time, a country's productive capacity is largely determined by its factor endowments, such as the area of land under cultivation, the amount of installed physical capital, the size of the labour force and the amount of human capital embodied in the labour force. Such stocks of productive resources change over time as a result of investment flows, e.g. the stock of physical capital is augmented each year by the flow of net investment in new machines. In a similar fashion, the human capital stock is changed by the net additions to the labour force of the flow of graduates from the educational system. The difference between educational stocks and flows is perhaps more important than in the case of other factors of production for the following reasons: (a) the time-lag between investment in education (say, as measured by enrolments) and additions to the human capital stock is longer than in any other sector; (b) some investment in education may never translate into additions to the human capital stock because of losses in the process (e.g. graduates not * The World Bank. This paper would not have been possible without the data collection work of Motohisa Kaneko and subsequent work by Robini Horn. The views expressed are those of the authors and should not necessarily be attributed to the World Bank. Copyright © International Labour Organisation 1986 561 International Labour Review participating in the labour force, or school time wasted by students dropping out or having to repeat courses). It is true that enrolment ratios measure the effort a country is making to alter the human capital stock and will eventually be reflected, at least to some extent, in the stock. But enrolment ratios may give an inaccurate or distorted picture when used to assess relative priorities for educational investment. A more accurate measure is the stock of human capital embodied in the country's labour force. The data set Of course, the limiting factor in moving from enrolment to labour force statistics is that the latter are not so readily available, and, where they are, are seldom cross-classified by the educational level of the worker. Attempts to compile such data were made in the past by Layard and Saigal (1966), the OECD (1969) and Psacharopoulos (1973), but the number of countries examined never exceeded 14. With the results of the 1980 round of censuses now available, two fresh efforts have been made: first, an analysis of published censuses and special labour force surveys yielded usable data for 66 countries in all world regions; second, material available from the US Bureau of the Census (Maryland Office), refined to portray the educational attainment of the population at large by age and sex, yielded data for an additional 33 countries (Kaneko, 1986). Given that the majority of males over 15 years old participate in the labour force, we considered this second source an acceptable proxy of labour force educa- tional attainment. Thus our sample covers 99 countries in all. (For a detailed account of the labour force definitions and the exact sources of information used in each country case, see Psacharopoulos and Arriagada, 1986, Appendix B, available from the authors on request.) One major caveat is in order here. The data used in this article refer essentially to primary, secondary and higher formal education rather than to vocational training and other education and training obtained outside the ordinary educational system. The latter types can in fact account for a substantial part of the human capital stock embodied in the labour force. In many countries "corporate" education and training are, both quantitatively and qualitatively, very important sources of skills and knowledge. In *he United States, for example, it is estimated that almost as many people are enrolled in corporate educational programmes as in the country's four-year colleges and universities. Similarly, the education and training provided in firms are widely regarded as a major determinant of Japan's impressive post- war economic performance. As regards flows, too, the human capital stock is changing as a result of the educational upgrading of the existing labour force, of education and training given outside the formal educational system, and even of learning on the job. But since such data are not available on a global scale, this article is confined to the formal educational system. 562 The educational composition of the labour force Figure 1. Years of schooling of the labour force (regional averages) 3.7 |East Africa (N=12) 1.8 West Africa (N=6) East Asia and the Pacific (N=12) 2.5 |South Asia (N=5) 3.4 1Middle East and North Africa (N=14) 5.8 1Latin America and the Caribbean (N=25) 6.8 Southern Europe (N=6) 10.5 Eastern Europe (N=6) 10.0 Developed market countries (N=13) Data reduction Table 1 gives the distribution of the total labour force by level of educational attainment, together with unweighted regional averages. In view% of the wide variation between countries and regions, a method of data reduction must be employed. We used the mean years of schooling embodied in the labour force, estimated as: S = Z LiS, where L, is the percentage of persons in the labour force with the ith level of schooling; Si is the duration in years of the ith level of schooling; and i ranges over illiterates and persons with incomplete primary, completed primary, incomplete secondary, completed secondary, and university education. The appendix describes the methodology used in this calculation. As shown in figure 1, West Africa is the least educationally developed region with 1.8 years of schooling embodied in the labour force, followed by South Asia (2.5 years), the Middle East and North Africa (3.4), and East Africa (3.7). East Asia and the Pacific, and Latin America and the Caribbean score over five years of schooling, which is about half the value of the index in industrialised countries. Figure 2, showing the proportion of the labour force without even a primary education, gives a graphic summary of the differential level of educational development in the world. At this point it is worth contrasting the mean years of education embodied in the labour force with another popular index based on enrolment ratios. Harbison and Myers (1964), facing the same problem of data reduction, constructed a " composite index of human resource development " 563 Table 1. Educational composition of the labour force Region and Year Total Percentage of the labour force with - Mean country/area tabour years of : force No edu- Primary Secondary Higher schooling 2 (000) cation Incomplete Complete Incomplete Complete East Africa Botswana 1971 305 60.2 27.1 8.8 2.5 0.8 0.4 2.0 Ethiopia 1978 194 62.1 15.6 6.7 6.7 7.1 1.8 2.6 Kenva 1980 6 561 48.9 16.0 24.9 4.0 5.8 0.4 3.5 Lesotho 1976 357 38.1 26.0 22.7 12.3 0.8 0.1 4.3 Malawi 1977 1 433 35.4 31.7 18.0 13.1 1.5 0.3 2.9 Mauritius 1972 215 21.7 42.5 13.9 15.2 4.9 1.8 4.5 Mozambique 1980 2 917 57.7 28.2 11.4 2.5 0.2 0.1 1.2 Reunion 1982 157 19.8 28.4 9.4 25.1 13.1 4.2 5.4 Rwanda 1978 1 258 55.2 21.3 19.3 3.2 0.6 0.3 2.2 Sudan 1974 202 35.0 10.6 26.1 13.0 10.3 5.1 5.5 Swaziland 1976 108 38.7 17.8 22.8 14.8 3.3 2.6 4.4 Zambia 1979 51 20.2 38.4 13.8 6.3 20.2 1.1 5.5 Regional average 41.1 25.3 16.5 9.8 5.7 1.5 3.7 West Africa Cameroon 1976 2 748 61.7 17.6 14.3 0.7 5.0 0.7 2.2 Cote d'Ivoire 1978 435 59.2 3.1 18.0 14.4 5.3 0.01 3.2 Liberia 1974 433 82.9 5.1 3.2 4.3 2.5 1.9 1.3 Mali 1976 2 120 91.0 3.2 4.8 0.5 0.2 0.2 0.5 Nigeria 1967 24 074 86.2 3.8 9.1 0.5 0.3 0.01 0.8 Senegal 1976 1 395 37.0 43.6 10.9 5.4 1.8 1.2 2.9 Regional average 69.7 12.7 10.1 4.3 2.5 0.7 1.8 East Asia and the Pacific China 1982 521 378 28.3 13.1 21.3 25.8 10.7 0.9 4.5 Taiwan, China 1983 7 070 8.9 5.1 32.7 17.7 24.0 11.5 8.4 Fiji 1976 171 9.0 13.8 24.5 27.9 21.4 3.3 8.3 Hong Kong 1981 1 962 7.6 17.6 19.1 21.3 26.2 8.1 8.8 Indonesia 1978 53 097 31.6 23.1 35.7 5,3 3.8 0.5 3.9 Korea (Rep. of) 1980 12 682 14.8 1.1 33.2 18.5 23.4 9.1 8.0 Malaysia 1967 2 946 27.0 1.7 55.7 9.2 4.6 1.8 5.0 New Caledonia 1983 56 6.0 52.7 0.0' 36.6 0.01 4.6 5.9 Philippines 1980 13 599 7.8 21.3 27.4 15.1 12.7 15.7 7.0 Singapore 1974 823 40.3 4.9 21.9 16.0 8.3 8.5 5.3 Thailand 1974 13 56() 12.0 72.2 6.3 4.9 2.0 2.2 4.1 Tonga 1976 25 0.2 25.8 10.1 60.8 0.2 2.9 7.3 Regional average 16.1 21.0 24.0 21.6 11.4 5.8 6.4 South Asia Bangladesh 1981 23 599 62.4 15.9 4.1 10.7 5.5 1.5 2.4 India 1981 22 518 66.6 14.5 6.9 4.9 3.9 3.2 1.9 Maldives 1977 60 76.7 17.1 2.1 3.4 0.4 0.2 0.9 Pakistan 1975 28 775 75.8 11.2 7.7 3.7 0.6 '.0 1.2 Sri Lanka 1971 3 643 23.1 15.7 23.6 32.4 4.4 0.8 6.2 Regional average 60.9 14.9 8.9 11.1 3.0 1.4 2.5 Middle East and North Africa Afghanistan 1979 3 707 72.0 6.1 9.4 5.6 7.0 0.0' 2.1 n Algeria 1977 3 425 37.0 17.1 12.7 19.0 5.7 2.1 4.0 Bahrain 1980 70 43.9 10.6 8.0 15.2 15.5 6.8 4.9 Egypt 1976 10 265 53.2 19.3 10.1 5.6 6.7 5.1 3.3 Iran (Islamic Rep. of) 1976 8 925 63.5 10.1 10.9 9.1 3.2 3.2 2.7 Jordan 1975 305 37.0 10.3 12.9 20.7 11.0 8.1 5.6 o Kuwait 1979 489 33.8 24.8 10.9 9.2 13.1 8.2 4.5 : Lebanon 1970 572 29.5 35.2 15.2 9.5 6.3 4.3 3.9 Morocco 1971 4 042 81.9 6.5 0.7 8.4 2.5 0.01 1.2 ° Qatar 1970 48 66.8 11.0 7.5 5.5 6.0 3.2 2.5 C Syrian Arab Rep. 1975 1 839 33.6 31.9 17.3 5.2 8.3 3.7 4.1 ,, Tunisia 1975 795 73.0 3.1 12.3 5.4 4.3 1.9 2.2 o United Arab Emirites 1968 78 77.1 0.01 12.2 0.01 7.7 2.9 2.1 1 Yemen Arab Rep. 1981 1 202 0.01 80.3 10.9 2.4 4.9 1.5 4.1 Regional average 50.2 19.0 10.8 8.6 7.3 3.6 3.4 C Table 1. (continued) Region and Year Total Percentage of the labour force with - Mean country/area labour years of force No edu- Primary Secondary Higher schooling > (0(X)) cation Incomplete Complete Incomplete Complete Latin America and the Caribbean C Argentina 1960 7 840 7.0 35.4 38.2 5.7 9.9 3.7 6.2 Barbados 1970 82 0.6 7.2 15.8 60.1 5.2 11.1 8.9 Belize 1970 31 8.3 31.5 34.4 13.4 11.2 1.2 6.9 : Bolivia 1976 1 396 31.0 13.8 23.4 14.2 11.6 6.0 5.4 Brazil 1980 43 768 24.7 35.3 7.9 19.6 6.6 5.9 5.6 Chile 1981 3 592 4.1 32.2 18.8 24.4 12.2 8.3 8.1 Colombia 1978 7 660 16.3 31.3 23.6 9.2 14.9 4.7 5.0 Costa Rica 1973 575 10.6 34.7 34.7 8.4 6.5 5.1 6.4 Cuba 1981 3 404 4.2 23.5 22.6 29.0 8.9 11.9 8.2 Ecuador 1982 2 216 16.2 13.9 37.0 13.5 10.8 8.5 6.5 Guadeloupe 1982 1(03 8.7 31.8 10.6 21.4 22.3 5.3 6.5 Guatemala 1973 1 749 51.7 12.7 28.1 2.6 3.5 1. J 3.0 Guyana 1982 472 17.6 0.0' 34.2 41.7 1.2 5.3 6.8 Haiti 1982 1 869 72.9 15.5 3.4 3.5 4.0 0.8 1.6 Honduras 1961 553 53.3 33.3 8.2 2.0 2.6 0.7 2.1 Jamaica 1978 937 2.6 1.6 76.6 7.1 12.1 0.01 6.9 Martinique 1982 94 2.3 40.9 1.9 18.7 27.9 8.3 6.0 Mexico 1977 11 249 26.9 28.8 26.9 8.9 3.2 5.2 4.5 Nicaragua 1971 488 49.7 19.5 20.5 4.5 4.8 1.0 4.4 Panama 1970 471 34.2 19.5 19.5 11.2 10.5 5.1 4.8 Paraguay 1972 727 10.5 62.2 11.2 7.6 5.9 2.6 4.3 Peru 1981 4 927 13.5 20.7 26.2 12.9 14.6 12.1 7.0 Trinidad/Tobago 1980 336 3.5 21.7 40.4 22.3 9.1 2.9 6.6 Uruguay 1975 1 041 5.1 33.8 29.2 12.3 7.3 12.4 6.7 Venezuela 1979 4 105 15.9 24.8 24.8 17.1 11.1 6.3 6.2 Regional average 19.7 25.0 24.7 15.7 9.5 5.4 5.8 Southern Europe Greece 1981 3 570 3.9 12.5 44.5 12.8 15.1 11.3 7.9 Israel 1983 1 329 2.5 5.8 20.5 17.6 31.6 22.1 11.3 Portugal 1981 3 458 16.3 53.3 8.4 10.8 5.4 5.8 4.5 Spain 1981 12 919 13.5 9.5 47.6 10.8 8.8 9.7 6.2 Turkey 1980 13 615 16.1 10.3 50.4 9.8 8.9 4.5 5.1 Yugoslavia 1981 8 217 7.4 2.7 49.8 32.9 3.0 4.2 5.6 Regional average 10.0 15.7 36.8 15.8 12.1 9.6 6.8 Eastern Europe Bulgaria 1975 3 364 4.0 3.3 64.3 5.6 16.9 5.8 8.7 Czechoslovakia 1980 5 503 0.0 0.3 37.2 7.8 47.9 6.9 11.5 German Dem. Rep. 1981 6 188 0.2 0.01 19.2 0.01 63.9 16.7 11.9 Hungary 1980 3 983 0.7 6.1 22.1 30.2 32.5 8.4 10.4 Poland 1978 12 745 1.3 3.2 5.4 41.4 43.3 5.5 10.7 Romania 1977 10 794 0.0 3.3 38.8 29.9 23.1 4.9 9.8 Regional average 1.0 2.7 31.2 19.2 37.9 8.0 10.5 - Developed market countries , Australia 1981 5 155 0.7 2.9 3.4 67.8 21.9 3.3 11.1 c Belgium 1970 3 520 3.92 7.2 38.3 25.7 16.8 8.1 8.1 D; Canada 1981 12 054 0.0 3.1 10.2 26.9 23.9 35.9 11.7 o Denmark 1981 1 946 0.0 4.9 39.3 33.5 11.7 10.6 8.6 X Finland 1980 1 817 0.0 21.5 26.3 11.1 33.2 8.0 8.5 o France 1982 20 731 0.4 38.6 24.8 18.2 9.6 8.4 6.2 Germany (Fed. Rep.) 1982 24 167 0.7 0.01 12.5 53.9 19.4 13.5 10.4 Co Japan 1979 54 352 0.0 0.01 39.3 20.3 22.4 18.0 9.8 Netherlands 1975 5 188 0.0 1.7 31.3 44.9 11.2 10.9 9.1 0 New Zealand 1981 1 317 0.0 0.0 9.3 44.2 15.6 30.9 11.7 Norway 1980 1 537 0.0 2.1 0.01 67.7 16.7 13.4 11.3 Switzerland 1980 2 489 0.0 4.0 21.8 9.2 49.4 15,6 11.0 United States 1981 101 853 0.0 2.1 7.8 18.8 35.8 35.6 12.6 0 Regional average 0.4 6.8 20.3 34.0 22.1 16.3 10.0 Original data reporting precludes separation from the previous or following educational level. 2 Includes "unknown" educational level. (D Source: Psacharopoulos and Arriagada (1986), Appendix table A-1. International Labour Review Figure 2. Percentage of labour force with no schooling (regional averages) 41.1 East Africa (N=12) 69.7 West Africa (N=6) 16.1] East Asia and the Pacific (N=12) 60.9 South Asia (N=5) 50.2 Middle East and 5 .2 . .North Africa (N=14) 19 7 Latin America and 19.7. the Caribbean (N=25) 10.0 Southern Europe (N=6) 1.0 Eastern Europe (N=6) 0.4 Developed market countries (N=13) using the weighted sum of the enrolment ratios at the secondary and higher educational levels, i.e.: I = Es(1) + Eh(S), where Es and El, refer to the enrolment ratios at the secondary and higher educational levels respectively, and the number in parentheses to the arbitrary weights attached to each level of education. The authors justify the 5:1 weight of higher relative to secondary education as follows: "In our judgement, higher education should be weighted more heavily than second- level" (Harbison and Myers, 1964, p. 32). However, they are silent regarding the zero weight implicitly attached to primary education. By contrast, our index of mean years of schooling of the labour force explicitly incorporates primary education (Lp), and the relative weights are non-arbitrary and have a particular meaning. For example, a simple version of our index, to facilitate comparison with the Harbison and Myers one, is S = L0(O) + Lp(6) + L5(12) + Lh(16), where Li stands for the share of the labour force with a particular "ducational level, and the weight is the cumulative length in years of that cycle. The difference between using enrolment and labour force information is far from academic - it can influence decisions on educational policy, and this has indeed happened in practice. For example, the prescription for boosting a country's score of educational development on the Harbison and Myers index would be to increase university enrolments, since they carry a heavy weight in the value of the index. By contrast, the index of the mean years of schooling of the labour force is more responsive to primary education, not only because of its non-zero weight, but also by the fact that Lp constitutes the second largest share in the distribution of the labour force by educational level in developing countries.' 568 The educational composition of the labour force Longitudinal trends For 34 of the 99 countries in our sample, information on the educational composition of the labour force exists for more than one year. Table 2 demonstrates the slowness of the process of building human capital stock. In most developing countries the educational level of the labour force is being up- graded at a rate of about one-tenth of one schooling year per calendar year. Even spectacular increases in enrolment ratios, as in the case of West Africa, do not automatically translate into an improvement in the educa- tional composition of the labour force (see figure 3). The reason is that enrolment ratios provide no information on the internal efficiency of the system, e.g. its capacity to retain and effectively educate students.2 And even if the system were perfectly efficient, the labour force stock is some 30-40 times larger than the annual flow of new entrants. To illustrate the above point let us look at India. The primary-level enrolment ratio was 61 per cent in 1960, 73 per cent in 1970 and 72 per cent in 1980. These figures do not appear particularly low compared to other developing countries, and may lead one to think that primary schooling is proceeding satisfactorily. However, if we look at the educational attainment of the labour force in 1981, we find that 67 per cent are illiterate, while 21 per cent received only a primary education. Furthermore, of the 21 per cent with primary education, fewer than a third completed the cycle. As far as policy formulation is concerned, therefore, the conclusions reached by using one indicator rather than another can lead to different priorities for educational investment. Concluding comment The educational attainment of the labour force is a more accurate indicator of the return on educational investments than enrolment ratios. The former assesses the effective supply of human resources available for economic growth, and captures the degree in which flows translate into human capital stocks. Enrolment ratios alone are an insufficient basis for determining educational policies or priorities because they give a false (upwardly biased) impression of the educational attainment of a country's human resources. Enrolment ratios say little about what happens after enrolment. For example, do " enrolled " students in fact attend school ? Moreover, enrolment ratios do not provide a measure of the internal efficiency of the educational system because the numerator includes students who are over the age of the denominator cohort. Do students complete the courses in which they enrol? Do they drop out of school before they attain literacy? Do they repeat grades? Uncertainties such as these show that the use of enrolment statistics can be misleading when it comes to formulating educational policies, not only in regard to setting priorities for educational investment, but also in evaluating the internal efficiency of the system. 569 International Labour Review Figure 3. Flows versus stocks. A comparison of enrolment ratios' and labour force educational attainment2 [IZ Enrolment Labour force East Africa Primary ,.econdary 44.6 119650 l5.6 65 1970 13.1 82.3 L 1980 25 41.8 15;6 West Africa 37.4 1960 2 52.4 1970 |7 75.6 1980 17.6 22.8 6 East Asia and Pacific 102.6 |_ 1960 19.4 105 F 1970 23.4 113 | 1980 36 45.1 4 4 Y33 South Asia 56 1960 | 17.4 67.8 |_ |_1970 23.9 73.1 1980 25.2 23.8K 13.9 Middle East and North Africa 52 1960 12.8 69.7 1970 24.6 89.8 1980 41.9 29.8 L 15.9 Latin America and Caribbean 88.4 1960 14 945 1970 27.4 105 1980 38.6 49.8 -25.2 World Bank estimates. 2 Most recent estimates. 570 The educational composition of the labour force Table 2. Changes over time in the educational composition of the labour force Region and Year Percentage of the labour force with - Mean countrylarea years of No edu- Primary Secondary Higher schooling cation Incom- Com- Incom- Com- plete plete plete plete East Africa Mauritius 1972 21.7 42.5 13.9 15.2 4.9 1.8 4.5 1983 10.2 14.2 34.7 31.4 5.4 4.1 6.7 Zambia 1963 51.9 19.6 11.9 16.5 0.1 0.0' 3.3 1979 20.2 38.4 13.8 6.3 20.2 1.1 5.5 West Africa Nigeria 1963 90.0 5.9 2.6 0.9 0.4 0.2 0.5 1967 86.2 3.8 9.1 0.5 0.3 0.0' 0.8 East Asia and the Pacific Indonesia 1978 31.6 23.1 35.7 5.3 3.8 0.5 3.9 1980 26.1 18.9 33.4 11.4 8.9 1.2 4.9 Korea (Rep. of) 1969 44.9 9.1 30.2 7.3 6.1 2.4 3.9 1980 14.8 1.1 33.2 18.5 23.4 9.1 8.0 Malaysia 1967 27.0 1.7 55.7 9.2 4.6 1.8 5.0 1980 17.9 17.1 23.4 22.9 16.1 2.6 6.5 Singapore 1974 40.3 4.9 21.9 16.0 8.3 8.5 5.3 1980 21.9 3.0 46.4 18.4 6.3 4.0 6.0 Taiwan, China 1980 9.3 4.5 30.2 18.9 24.3 12.7 8.6 1983 8.9 5.1 32.7 17.7 24.0 11.5 8.4 Thailand 1960 37.4 55.6 1.1 3.5 2.0 0.4 3.3 1974 12.0 72.2 6.8 4.9 2.0 2.2 4.1 1980 10.1 64.2 7.0 11.2 4.1 3.4 4.6 South Asia India 1961 89.9 5.2 2.1 1.9 0.3 0.6 0.5 1981 66.6 14.5 6.9 4.9 3.9 3.2 1.9 Pakistan 1975 75.8 11.2 7.7 3.7 0.6 1.0 1.2 1981 65.9 7.4 5.2 16.4 2.4 2.7 2.5 Sri Lanka 1963 22.2 27.9 41.9 2.7 3.2 2.1 5.3 1971 23.1 15.7 23.6 32.4 4.4 0.8 6.2 1981 8.5 12.7 35.9 38.3 2.8 1.8 7.5 Middle East and North Africa Egypt 1960 63.6 27.7 2.1 4.7 0.3 1.9 1.7 1976 53.2 19.3 10.1 5.6 6.7 5.1 3.3 Morocco 1971 81.9 6.5 0.7 8.4 2.5 0.01 1.2 1982 56.4 12.9 8.8 14.9 5.1 1.9 2.9 Syrian Arab Rep. 1960 54.8 31.8 9.0 2.6 1.1 0.8 2.0 1975 33.6 31.9 17.3 5.2 8.3 3.7 4.1 571 International Labour Review Table 2. (continued) Region and Year Percentage of the labour force with - Mean countrlyarea years of No edu- Primary Secondary Higher schooling cation Incom- Coom- Incom- Com- plete plete plete plete Latin America and the Caribbean Argentina 1960 7.0 35.4 38.2 5.7 9.9 3.7 6.2 1980 4.7 24.9 34.8 17.7 9.5 8.4 7.4 Brazil 1960 48,2 45.1 3.4 2.3 0.4 0.5 2.4 1980 24.7 35.3 7.9 19.6 6.6 5.9 5.6 Chile 1969 18.6 36.0 20.5 11.3 11.2 2.3 5.9 1981 4.1 32.2 18.8 24.4 12.2 8.3 8.1 Colombia 1951 42.3 41.7 8.0 5.3 1.7 1.0 2.2 1964 28.9 48.5 12.1 5.8 3.3 1.3 2.8 1973 16.8 35.5 20.1 10.1 13.2 4.3 4.8 1978 16.3 31.3 23.6 9.2 14.9 4.7 5.0 Guatemala 1964 63.6 27.0 5.1 2.6 0.6 1.1 1.7 1973 51.7 12.7 28.1 2.6 3.5 1.4 3.0 hionduras 1961 53.3 33.3 8.2 2.0 2.6 0.7 2.1 1974 42.3 27.2 21.6 3.6 3.9 1.4 3.0 Mexico 1970 23.6 43.5 17.0 6.3 5.5 4.1 4.2 1977 26.9 28.8 26.9 8.9 3.2 5.2 4.5 Panama 1960 43.9 13.0 22.5 8.5 8.8 3.3 4.0 1970 34.2 19.5 19.5 11.2 10.5 5.1 4.8 1980 12.5 17.9 30.0 19.9 11.7 8.0 6.6 Paraguay 1972 10.5 62.2 11.2 7.6 5.9 2.6 4.3 1982 8.1 39.4 26.4 15.4 6.9 3.8 5.6 Peru 1961 31.1 33.7 20.7 6.1 5.4 3.0 3.9 1981 13.5 20.7 26.2 12.9 14.6 12.1 7.0 Uruguay 1963 8.8 44.8 24.5 14.7 3.7 3 4 5.1 1975 5.1 33.8 29.2 12.3 7.3 12.4 6.7 Venezuela 1979 15.9 24.8 24.8 17.1 11.1 6.3 6.2 1982 14.1 23.4 23.4 19.8 12.1 7.2 6.4 Southern Europe Greece 1961 42.0 29.8 17.4 5.1 2.8 2.9 3.2 1971 30.4 29.2 24.0 6.8 4.6 4.9 4.3 1981 3.9 12.5 44L5 12.8 15.1 11.3 7.9 Developed market countries Belgium 1970 3.9' 7.2 38.3 25.7 16.8 8.1 8.1 1977 3.62 7.8 43.0 20.8 13.9 11.0 8.3 Canada 1969 0.0 6.2 34.3 20.3 30.4 8.8 9.1 1981 0.0 3.1 10.2 26.9 23.9 35.9 11.7 572 The educational composition of the labour force Table 2. (continued) Region and Year Percentage of the labour force with - Mean cotintry/area years of No edu- Primary Secondary Higher schooling cation Incom- Com- Incom- Com- plete plete plete plete Denmark 1981 0.0 4.9 39.3 33.5 11.7 10.6 8.6 1983 0.M 2.2 17.6 15.2 47.9 17.1 10.8 Japan 1969 0.0 3.9 66.5 12.4 10.6 6.6 7.6 1979 0.() 0.0' 39.3 20.3 22.4 18.( 9.8 Netherlands 1967 0.0 2.9 54.1 8.8 2.2 11.0 6.1 1975 (.( 1.7 31.3 44.9 11.2 10.9 9.1 United States 1950 0.0 33.8 28.2 20.0 3.2 14.8 8.3 1969 0.0 12.3 23.3 10.1 35.1 19.2 10.6 1981 0.0 2.1 7.8 18.8 35.8 35.6 12.6 Original data reporting precludes separation from the previous or following educational level. 2 Includes "unknown' educational level. Source: Psacharopoulos and Arriagada (1986). Appendix table A-1 anid Appendix C. Appendix. Estimating the mean number of years of schooling of the labour force In order to estimate the mean number of years of schooling of the labour force in each country, the following formula was used: S= [(LP1 x YRSP/2) + (LP2 x YRSP) + (LS1 x (YRSP + YRSS/2)) + (LS2 x (YRSP + YRSS)) + (LH x (YRSP + YRSS + YRSH))] /100 where: S = mean number of years of schooling LP1 = percentage of the labour force with incomplete primary schooling YRSP = duration in years of primary education cycle LP2 = percentage of the labour force with completed primary schooling LS1 = percentage of the labour force with incomplete secondary schooling YRSS = duration in years of secondary education cycle LS2 = percentage of the labour force with completed secondary schooling LH = percentage of the labour force with completed and ipeomplete higher education YRSH = duration in years of higher education cycle and it is assumed that the workers with incomplete primary and secondary education attended for half of the years of the corresponding level (e.g. if primary level = 6 years, LPI = 3 years). 573 International Labour Review Example Applying the above to Cameroon (1976) we have: LPO = 61.7 per cent of the labour force (no education) LP1 = 17.6 per cent of the labour force YRSP = 6 years LP2 = 14.3 per cent of the labour force LS1 = 0.7 per cent of the labour force YRSS = 7 years LS2 = 5.0 per cent of the labour force LH = 0.7 per cent of the labour force YRSH = 4 years Thus the mean number of years of schooling is: S = [(17.6 x 3) + (14.3 x 6) + (0.7 x 9.5) + (5 x 13) + (0.7 x 17)] / 100 So S = 2.2 years. Notes I For a superb demonstration of how the application of different models can lead to different policy prescriptions in Nigeria and Greece, see Bowles (1969). 2 The enrolment ratios in figure 3 are "gross", the more widely available type. The gross enrolment ratio for a given level of education is derived by dividing the total enrolment for that level, regardless of age, by the population of the age grouip which, according to national regulations, should be so enrolled. Because the numerator includes over-age students, gross enrolment ratios can exceed 100 per cent (see UNESCO, 1985, p. II.5). References Bowles, S. 1969. Planninig educational systems for economic growth. Cambridge, Massachusetts, Harvard University Press. Harbison, F.; Myers, C. A. 1964. Education, manpower, and economnic growth: Strategies of human resoutrce development. New York, McGraw-Hill. Kaneko, M. 1986. The educational composition of tile world's population: A database. Washington, DC, The World Bank, Education and Training Department. Report No. EDT 29. Layard, P. R. G.; Saigal, J. C. 1966. "Educational and occupational characteristics of man- power: An international comparison ", in Briti.h Journal of Industrial Relations (London), July, pp. 222-266. OECD. 1969. Statistics of the occupational and educational structure of the labour force in 53 countries. Paris. Psacharopoulos, G. 1973. Returns to education: A)i international comparison. Amsterdam and San Francisco, Elsevier and Jossey-Bass. -; Arriagada, A. M. 1986. The educational attainment of the labor force: An international comparison, or moving beyond enrollment raiios for assessing priorities in educational investment. Washington, DC, The World Barnk, Education and Training Department. Report No. EDT 38. UNESCO. 1985. Statistical Yearbook. Paris. 574