(Aip a1 3 I POLIcy RESEARCH WORKING PAPER 2731 An Alternative Technical The results of this paper appear to indicate that Education System in Mexico CONALEP is a highly cost- effective program. CONALEP has had spillover effects on A Reassessment of CONALEP the rest of the technical education system in Mexico Gladys Lopez-Acevedo by stimulating other educational institutions to be more efficient. The World Bank Latin America and the Caribbean Region Poverty Reduction and Economic Management Sector Unit December 2001 I POLIcY RESEARCH WORKING PAPER 2731 Summary findings Using matched pair methods, L6pez-Acevedo reevaluates training. CONALEP graduates earn 20-28 percent more the labor market performance of graduates of Mexico's than the control group. And employers invest more in Colegio Nacional de Educaci6n Profesional Tecnica training CONALEP graduates than they do in training (CONALEP), the country's largest technical education individuals in the control group. system. She also assesses the impact of innovations L6pez-Acevedo shows that the innovations introduced introduced by CONALEP in 1991. by CONALEP increase graduates' probability of finding a The analysis shows that individuals in the control job and shorten their job search. A cost-benefit analysis group find jobs faster than CONALEP graduates do, but appears to show that CONALEP is an effective training a larger share of CONALEP graduates work in an system. occupation consistent with their field of specialization or This paper-a product of Poverty Reduction and Economic Management Sector Unit, Latin America and the Caribbean Region-is part of a larger effort in the region to reduce poverty and inequality through human capital investment. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Michael Geller, room 14-046, telephone 202-458-5155, fax 202-522-2112, email address mgeller@worldbank.org. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at gacevedo@worldbank.org. December 2001. (74 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Policy Research Dissemination Center An Alternative Technical Education System in Mexico: A Reassessment of CONALEP Gladys L6pez-Acevedol The World Bank aacevedo(kiworldbank.org Key Words: Technical education and matching methods. JEL Classification: 128; J24 and N36. ' This research was completed as part of the "Training Mechanisms Reform" Project at the World Bank. We are particularly grateful to the Human Development Sector Team, Eduardo Velez Bustillo, Anna Maria Sant'anna, Indermit S. Gill, Xiaolun Sun, and Joseph S. Shapiro who provided valuable support. Valuable research assistant was provided by M6nica Tinajero. Publication assistance was provided by Erica Soler. These are views of the author and do not necessarily reflect those of the World Bank, its executive directors, or the countries they represent. Comments were received from government officials attending the seminar organized by the Bank and the Council for Standardization and Certification of Labor Competencies (CONOCER) to review the studies sponsored by the World Bank Training Mechanisms Reform Project. Table of Contents I. Background 5 II. Labor Market 6 III. Education and Training 10 IV. CONALEP 13 V. Evaluation of the CONALEP System: Students and Graduates 16 VI. Data 22 VI.1 Methodology 24 VI.2 Results 26 VII. Benefits from CONALEP's Reformed Program 39 VII. 1. Methodology 39 VII.2. Results 39 VIII. Cost-Benefit Analysis 45 IX. Conclusions 45 Selected References 48 Annex 1 51 Annex 2 58 Annex 3 65 Annex 4 69 Annex 5: List of Tables and Figures 73 Acronyms CBET Competency Based Model CECATI Non-professional, Elementary Vocational Training CENEVAL Centro Nacional de Evaluaci6n CONALEP Colegio Nacional de Educaci6n Profesional Tecnica CONOCER Council for Standardization and Certification of Labor Competencies COSNET Council of the National System of Technological Education DGETI Technical-professional schools EAP Economically Active Population ENE National Employment Survey ENECE National Employment, Training, and Salary Survey ENEU National Urban Employment Survey NAFTA North American Free Trade Agreement OECD Organisation for Economic Co-operation and Development PMETyC Education Modernization Project SEIT Under Ministry for Technological Education and Research SEP Ministry of Public Education SESIC Under Ministry for Tertiary Education and Scientific Research STPS Ministry of Labor I. Background The period spanning from the second half of the 1980s until the late 1990s is important for the Mexican economy, as it encompasses a major structural change from a protected, public-sector driven economy to a globally integrated, private-sector led one. For all its merits, this change seems to have produced an increasingly unequal distribution of the fruits of economic growth. The World Bank Report "Earnings Inequality after Mexico's Economic and Educational Reforms" (2000) showed that the most plausible hypothesis for the worsening in earnings inequality in Mexico is the increased rate of skill-biased technological change brought about by trade liberalization. This World Bank Report also found that Mexico is experiencing increasing returns to higher education, and that the skill composition of employment in manufacturing and other export sectors has moved toward demanding a higher proportion of skilled workers, particularly in industries that are most open to international competition. When rising demand for skills is not met by supply, the result is a persistent shortage of skilled labor and constrained growth. The excess demand also forces firms to pay above market- clearing wages in order to retain the workers they train. On the supply side, the roots of the shortage problem can be traced to thrqe main factors. The first is low educational attainment- particularly among the poor. The second is insufficient financial support to those students who are academically qualified but who are financially needy. The third is the persistence of antiquated and unresponsive training mechanisms-vocational and technical systems are not providing new entrants with appropriate skills.2 Effective technical training is Mexico's primary tool for reaching an equilibrium in the market for skilled labor. 2 Evidence on the low educational achievement in technical education is drawn from the Council of the National System of 5 Several attempts have been made to evaluate technical education programs in Mexico.3 Using a rigorous impact evaluation method, this paper re-examines the performance and evolution of the College of Professional Technical Education (CONALEP) system. CONALEP is the backbone of Mexico's skills training structure and has become the most important government technical education system. This paper is organized as follows. Section II reviews briefly the Mexican labor market. Section III describes the Technical Education System in Mexico and the place of CONALEP within this system. Section IV discusses the evolution of the CONALEP system. Section V reviews CONALEP's past evaluations. Section V also introduces the CONALEP graduate tracer survey, the National Employmept Survey (ENE), and the National Employment, Training, and Salary Survey (ENECE) used in this study. Section VI discusses the CONALEP benefit results compared to a well-designed control group. Section VII discusses the CONALEP benefits of the reformed program (the introduction of the modular course, among others). Section VIII presents the cost-benefit analysis. Section IX offers conclusions. The annexes at the end of this paper include the most relevant quantitative results that support the paper's findings. H. The Labor Market Crisis and change have marked the past twenty years of Mexico's economic development. Many crises have had important impacts on labor markets. In the early 1980s, Technological Education (COSNET). This Council applies other tests in the SEIT schools to measure students' formal reasoning and the ability to learn mathematics. In addition, each institution designs its own proficiency examination. The "technological" area uses as criteria 7 points in the learning examination (in a 0 to 10 scale), a minimum of 18 correct answers out of 32 in the over-all knowledge examination and 12 correct answers out of 24 in the test to assess capacity for learning mathematics. SEP, in the "Informe de Labores 1997-1998," reports that 234,925 students took this exam. Of them 3,231 (1.3 percent) were rejected from upper-secondary education, not having the knowledge and capacities for entry requested by the educational institutions. 3 See World Bank, 1997, Mexico: Training Assessment Study. Carnoy B. et. al., 2000, "Aprendiendo a trabajar: Una revisi6n del Colegio Nacional de Educaci6n Profesional Tecnica y del Sistema de Universidades Tecnol6gicas de Mexico." 6 Mexico and the rest of Latin America plunged into a major recession, brought on by over- borrowing in the 1970s as a result of extremely low real rates of interest, and by excessive reliance of some countries on oil as an export commodity. When the United States drastically increased interest rates to fight its own inflation, Latin America and other developing countries were caught with high foreign debt to gross national product (GNP) ratios and major interest repayments. Moreover, the steep decline in oil prices worsened the crisis for Mexico and other oil exporting countries (World Bank 1998, 1999a). The economic downturn in the early-1980s increased underemployment rates and lowered real income and wages sharply. The crisis also ended Mexico's (and Latin America's) import-substitution industrialization and forced the restructuring of Mexico's economy. The debt crisis and restructuring turned Mexico's manufacturing and agriculture sectors toward exporting and away from a protected domestic market. Mexico's average economic growth rate in the period 1959-1981 was about seven percent annually, or approximately four percent per capita. However, from the slow-down in 1983 onwards, growth rates have been much lower, about 2.6 percent annually (a 0.3 percent per capita growth rate). Nevertheless, in the past four years (1997-2001) the rate of economic growth has increased to five percent annually, or three percent per capita. The peso crisis of 1994 was no different. The crisis caused sharp rises in unemployment, a slowing of employment growth, and a drop in real wages. Real wages did not return to 1985 levels until 1998. Large numbers of workers moved to the informal sector and to rural areas, with establishments of fewer than six employees growing by 6.3 percent in 1995 and establishments of more than six employees growing by only 0.6 percent (World Bank 1999b). At the same time, 7 the North American Free Trade Agreement (NAFTA) led to a rapid growth in export industries-Mexico had US$21.5 billion growth in exports from 1994 to 1998, compared to just US$6 billion in export growth from 1991 to 1994 (World Bank 1999). According to the latest available national employment survey (ENE99), the economically active population (EAP), defined as the sum of the employed population and the open unemployed population, numbered nearly 40 million people. The average net participation rate was nearly 56 percent. From 1995 to 1999, the open unemployment rate decreased from 4.7 percent to 1.7 percent.4 Mexico's labor force grew at an average rate of 2.8 percent per year from 1995 to 1999. This means that nearly 1,113,000 new entrants were added to the labor force every year. Women's labor force participation, while still low compared to the level in developed economies, rose significantly in the 1990s. Data from the Organisation for Economic Co- operation and Development (OECD) shows that the rate for women 25-54 years old increased from 37 percent in 1990 to 44 percent in 1998 (OECD 1999). The other important feature of the late 1990s, according to the International Labour Organisation (ILO) data, is that labor force growth and declining open unemployment were accompanied generally by rising real salaries and wages in manufacturing. This was after a more than 30 percent decrease in real manufacturing wages from 1982 tol988. 4 Mexican wages are likely to increase in real terms for the third consecutive year in 2000, by three to five points above inflation. The improvement fits with the pattem of booming economic growth in Mexico in 2001, coupled with a steady curbing of inflation. The latest government figures show that the economy grew by 7.5 percent during the first seven months of the year 2000, compared with 3.7 percent in the whole of 1999. Twelve-month accumulated inflation was down to 9.10 percent at the end of August, compared with 12.3 percent at the start of the year. Nevertheless, experts agree that, with inflation under control, wage increases during 2001 must be backed by increases in productivity in order to prevent a renewed increase in inflation. 8 The private sector accounts for about 88 percent of total employment in Mexico, a much larger share than that prevalent in other OECD countries. The Mexican private sector shows a growing duality: a large traditional sector coexisting alongside an expanding modem sector. The former, which consists primarily of micro-enterprises and small and medium-size enterprises (defined to include firms with up to 250 workers), employs a large fraction of the labor force but accounts for a small portion of output and exports. Roughly, these enterprises account for 71 percent of total employment, 53 percent of employment in manufacturing, 95 percent of employment in the retail sector, and 73 percent of employment in services. From 1988 to 1996, annual output per worker was low in the service sector. While some studies have shown that the manufacturing sector has become more efficient as a result of trade liberalization, with gross labor productivity increasing at an annual rate of 3.1 percent during the 1988 to 1996 period, this rate was still low compared with that in other developing countries, and was about the same as in the United States (World Bank 1998). One plausible explanation for this slow growth in labor productivity is the lower educational level of Mexican workers and the resulting deficiency in the on-the-job human capital accumulation compared to elsewhere. The increase in wages associated with an additional year of work experience for Mexican men is low compared to the increase for men with similar educational attainment in other countries (3.8 percent in Mexico compared with 8.1 percent in United States, 8.4 percent in Japan, and 9.1 percent in France). This rate is low even when compared with the rate in countries at a similar level of development and with comparable education indicators, such as Brazil (6.2 percent) and Colombia (5.8 percent). Given the well- documented correlation between wage growth, on-the-job training, and productivity observed in 9 many countries, these differences are consistent with the hypothesis that in Mexico post-school investment in human capital results in lower productivity growth. The observed low level of investment in human capital could also be explained by the incentive structure of labor regulations. In practice, as has been well-documented, firms appear to enjoy more flexibility than a strict interpretation of the law would suggest (World Bank 1999b). III. Education and Training The structure of Mexico's educational system has the following main characteristics. Basic education is the Mexican government's highest priority. The basic education system consists of: a) early childhood education (or pre-school), which is optional for children from 3 to 5 years old; b) mandatory primary education, ideally for children aged 6 to 12, but due to late enrollment and grade repetition it is targeted at children aged 6 to 14, and c) mandatory basic secondary school education, consisting of a 3-year cycle, and intended for children aged 12 to 16. Upper-secondary education in Mexico is divided into a) bachillerato general (general baccalaureate), b) bachillerato tecnico (technical baccalaureate) and c) bachillerato bivalente (bivalent baccalaureate). The bachillerato general education system is administered by the Sub- secretariat for Tertiary Education and Scientific Research (SESIC), while the technical baccalaureate system is administered by the Sub-secretariat for Technological Education and Research (SEIT) (OEDC 1997). The bachillerato tecnico-training is provided through a range of institutions that include CONALEP, offering programs aimed at mid-level careers in the work force. Students graduate 10 with the qualification of professional technician, technical professional, or base level technician, depending on the type of institution they attend and the program they undertake. CONALEP is unique in that it offers the opportunity for students to gain access to higher education as they can opt to take more courses per semester and to take a separate high school diploma exam. The bachillerato bivalente training institutions also offer the opportunity to study for a technical middle level career, while at the same time qualifying students for entry to higher education. Programs in this stream are available in the areas of agriculture, fishery, manufacture, and services. The complexity of the arrangements at the upper-secondary level are readily seen in Table Al.1. In a parallel way, the national education system also offers skills training programs in a formal classroom format, with courses ranging from a few hours to several months. These courses have no academic prerequisites and provide job skills training for entry-level technical positions (Capacitaci6n para el Trabajo). Most students in these training programs have a primary education background. The system also covers adult education, including non-traditional job skills training, self-instructional formats, special education, education for indigenous and rural populations, and open education at all levels. Training in Mexico is given at four levels: a) job skills training with no formal academic requirements, b) upper-secondary level training which requires middle school to have been completed, c) undergraduate university level training, and d) graduate level training. The Mexican educational system expanded rapidly at the secondary and university levels even during the economic crisis years of the 1980s and early 1990s (OECD 1997). In the 1990s, 11 the total number of students at primary level hardly rose at all, increasing from 14.4 million in 1990-91 to just 14.6 million in 1998-99. Yet terminal efficiency, the percentage of students finishing sixth grade with the group they started school with, increased from 70 percent in 1990- 91 to 86 percent in 1998-99. Basic secondary education has expanded very rapidly in the past 20 years, increasing from three million students in 1980-81 to more than five million in 1998-99. In 1980, only 58 percent of 13-15 year-olds were in basic secondary school; in 1998-99, 80 percent of that age group were enrolled. Even so, dropout rates continue to be high (and they are still rising) at the basic secondary level, so that despite basic secondary being compulsory, at the end of the 1990s only 65 percent of 18 year-olds had completed basic secondary (SEP 1999a). These data include both rural and urban areas. In urban areas, the dropout rates are higher than in rural areas. Besides the rapid expansion of basic secondary in the 1980s and 1990s, the key change in Mexican education in the past two decades has been the rapid increase in enrolment in post-basic education, and the rise in the percentage of basic secondary graduates who go on to upper- secondary. In 1990-91, only 75 percent of those who finished basic education continued on to upper- secondary; in 1998-99, the proportion rose to 95 percent (SEP 1999a). Table 1 shows that of all the students who attended upper-secondary in 1999, 7.96 percent went to CONALEP, 0.76 percent attended schools offering the bachillerato general, and 21.19 percent attended schools offering the bachillerato t&cnico. Tables Al.2 and Al.3 show the main differences between these educational systems. 12 Table 1. Enrollment in Upper-secondary by Type of School 1997 % 1998 % 1999 % Federal (SEIT. SESIC) 1,015,636 38.97 1,032,059 38.03 1,035,960 36.93 General Upper-secondary (Bachillerato General) 20,781 0.80 20,373 0.75 21,375 0.76 Upper-secondary by cooperation 68,441 2.63 67,262 2.48 66,788 2.38 Upper-secondary (COBACH) 83,946 3.22 89,369 3.29 88,016 3.14 Technical Upper-secondary 597,416 22.92 594,762 21.92 594,581 21.19 Technician (CETIS and CBTIS) 45,073 1.73 38,947 1.44 40,154 1.43 Technician CONALEP 197,906 7.59 218,884 8.07 223,273 7.96 Technician (Others) 2,073 0.08 2,462 0.09 1,773 0.06 State 703,515 26.99 773,195 28.49 815,421 29.06 Autonomous (University) 374,201 14.36 369,992 13.63 367,960 13.12 Private 512,743 19.67 538,651 19.85 586,193 20.89 Total 2,606,095 100.00 2,713,897 100.00 2,805,534 100.00 Source: SEP, "Compendio Estadistico por Entidad Federativa 1999," DGPPP. IV. CONALEP In December of 1978, the Mexican Government created CONALEP as a public decentralized body of the Ministry of Public Education (SEP). CONALEP was intended to provide a national network of upper-secondary schools that would prepare young people to become technicians at the upper-middle educational level. At this skill level 4 in the ISCED international classification (upper-secondary), there was a gap that was growing with the increasing demands for skilled labor. With the establishment of CONALEP, the Government also wanted to strengthen and rationalize the complex provision for technical secondary education in Mexico. In 1979, the first ten CONALEP schools were opened, offering training in seven careers to 4,100 students. Not surprisingly, five of these careers focused on manufacturing, while the other two careers dealt with medical assistant and nursing professions. By 1982 the number of students enrolled in courses in CONALEP leading to technical qualifications increased to 72,000 and by 1989-1990 the total was 155,300. Since 1983, in addition to its career programs for technicians, CONALEP has also offered short courses for industry. This program was expanded 13 in 1986 through the introduction of mobile training facilities. By 1990, the number of students enrolled in these courses had increased to 61,300. The major growth in student numbers during this period was facilitated by a rapid growth in the number of CONALEP schools, from 10 in 1979 to 239 in 1986, by which date all 31 states in Mexico had CONALEP schools. However, the distribution of students by state was uneven, with about one-third of all students attending schools within the metropolitan zone of Mexico City. The size of the individual CONALEP schools was also uneven. The number of careers expanded substantially from the original seven to 146 by the beginning of the 1990s, although these careers were reduced to 29 between 1993 and the beginning of 1997. The rapid growth during the 1980s and the beginning of the 1990s coincided with a shift toward white-collar occupations in commerce, administration, computing, and accounting, which now comprise more than half of the students in CONALEP. The educational services at CONALEP schools were expanded in 1991-1992 by the introduction of the modular program, which was the forerunner of the competency-based education and training (CBET). In 1994, as part of the Education Modemization Project (PMETyC) financed by the World Bank, CONALEP introduced a competency-based model (CBET) for nine careers, to bring the CONALEP education program closer to the needs of industry. The initial pilot project to introduce competency-based education and training effectively in CONALEP demonstrated the challenges of this new way of teaching. This project helped the institution to understand the complexities of its significant role as a player in the forthcoming standards-based approach to education and training, and the need- for major reforms to its administration and educational practices. 14 The CONALEP decision to move to CBET was a direct consequence of Mexico's decision to develop national competency standards as part of PMETyC, coordinated by the SEP and the Ministry of Labor (STPS). This new approach is run by the Council for Standardization and Certification of Labor Competencies (CONOCER), which is organized as a trust (fidecomiso) governed by a tripartite board of directors consisting of labor representatives, entrepreneurs, and government. The SEP budget finances the trust. Established in 1995, PMETyC is intended to strengthen the links between formal education, training, and the needs of the labor market. Different countries are coming to terrns with the requirements of work-based training in different ways (Ahier, 1999). Learning can take place in a range of settings, including on the job, off the job, in a technological institution, and at home. The skills required for employment involve lifelong learning to upgrade skills, preparing people for higher levels of employment, or providing opportunities to develop life skills that make people more valuable as citizens. This last aim sparks much debate, and different countries weigh programs differently depending on local perspective. European countries have always placed considerable emphasis on the general education component of formal vocational courses; Mexico has done the same (Boud and Garrick 1999). Countries such as the United Kingdom, Australia, and New Zealand have put much less emphasis on these broader consider tions, concentrating more in their vocational courses on developing the technical skills needed in the workplace. There is now a move away from such an instrumental approach toward a more balanced curriculum. This new direction emphasizes more 15 generic skills and seeks not to cut off the range of students' options too early, allowing them to move more easily to higher levels of learning in the same field or a new one (Hobart 1999). The importance of career programs that allow students to develop general skills alongside technical ones has been acknowledged in many countries (Frantz 1998). These skills have different names in different countries-they are called key competencies in Australia, strategy for prosperity in Canada, process independent qualifications in Denmark, crossing or transferable skills in France, key qualifications in Germany, essential skills in New Zealand, core or conmmon skills in the United Kingdom, and workplace know-how in the United States (Hobart 1999). In light of the increased need for more generic skills, Mexico has started to re-examine its own strategy, as specific technical skills can quickly become outdated. V. The Evaluation of the CONALEP System: Students and Graduates The socioeconomic and academic level of CONALEP students varies according to location. Data from the National Evaluation Figure 1 Center (Centro Nacional de Evaluaci6n, Family Income of Students at Selected lnstitutions in 1999 CENEVAL) suggests that CONALEP most 500-CONLLEP 'S 4a0 -uw--Col.Bach. frequently serves students from a lower 0 EdlOMeX. socioeconomic status at the upper-secondary = -m DGETI school level in Mexico City. The results of a 8 ooo\ random sample of those who took the entrance a Y/ \eag 0 examination to upper-secondary school in the Frequercy in Population metropolitan area of Mexico City in 1999 Note: This graph assumes normal population distribution. 16 suggest that CONALEP students come from families with the lowest average income and the lowest parental education (Figure 1). The parents of an average CONALEP student have about two years less formal education than the parents of a student attending a Colegio de Bachilleres, and three years less formal education than the parents of a student attending the high schools of the Instituto Politenico Nacional. Students attending CONALEP do not necessarily do poorly on the entrance test, nor do they all come from low educated or low-income parents. About 20 percent of CONALEP students in this sample scored higher than the average student attending the Colegio de Bachilleres. Approximately 35 to 40 percent of the parents of CONALEP students have higher levels of education than the parents of an average student at the Colegio de Bachilleres. Nonetheless, on average, CONALEP students come from the lower socioeconomic categories and generally have lower scores in the CENEVAL examination than students in the other streams of upper- secondary education. Only students attending other technical-professional schools (DGETI) are comparably low on these indicators. 17 Table 2. CONALEP Students Compared to Students from Selected Institutions Centro Nacional de Evaluacion All' Option2 Global3 Family4 GPA In' Mother's Father's Private Lower Institutions Number Test Score Income Lower- Schooling6 SchooUng6 Sec -1 secondary (years) (years) CONALEP Mean 2.27 54.2 2271.2 7.627 7.1 8.2 8.70E-03 N 460 460 460 460 460 460 460 SD 2.13 15.3 2269 .6992 4.8 5.2 9.29E-02 Col.Bach. Mean 3.05 66.4 3132 7.658 9.0 10.0 3.33E-02 N 421 421 421 421 421 421 421 SD 2.28 13.2 2845 .75722 5.2 5.1 .18 Edo.Mex. Mean 2.41 64.6 2721 7.931 8.452 9.9 1.76E-02 N 1192 1192 1192 1192 1192 1192 1192 SD 2.09 16.56 2436 .7600 4.9 5.2 .13 DGETI Mean 2.71 59.6 2610 7.7205 7.700 9.2 2.20E-02 N 682 682 682 682 682 682 682 SD 2.32 15.6 2488 .7271 5.0 5.4 .15 IPN Mean 1.97 80.7 3315 8.1865 9.8 11.3 5.81E-02 N 430 430 430 430 430 430 430 SD 1.61 13.9 2652 .7871 4.7 5.0 .23 UNAM Mean 1.46 88.1 3967 8.3935 9.8 11.4 9.41E-02 N 510 510 510 510 510 510 510 SD .83 11.6 3385 .7864 5.212 5.0 .29 Other Mean 1.38 82.9 3896 8.5417 11.969 12.9 8.33E-02 N 48 48 48 48 48 48 48 SD .96 15.8 3164 .8124 3.676 4.2 .28 TOTAL Mean 2.33 67.9 2945 7.9248 8.6 10.0 3.50E-02 N 3743 3743 3743 3743 3743 3743 3743 SD 2.03 18.5 2693 .7982 5.1 5.2 .18 1. Col.Bach. refers to Colegio de Bachilleres, the local answer to over-demand; Edo.Mex. to the Estado de Mexico, state-centralized high school system; DGETI is the Direcci6n General de Educaci6n T6cnica Profesional, a centralized institution; IPN is the Instituto Polit6cnico Nacional -centralized-; and UNAM is the Universidad Nacional Aut6noma de Mexico -autonomous-. 2. This is the average preference number toward each institution from students who applied and got in. 3. Out of 128 questions. 4. In net pesos per month. 5. Grades go from 5 (fail) to 10. 6. Years of schooling. CONALEP's Past Evaluations The CONALEP system has been evaluated several times in the past. The first evaluation was done by CONALEP (1994) and CONALEP (1999) using graduate tracer surveys. These data sets are described in the next section. The other evaluations were done by Lane and Tan (1996) and by Lee (1998). CONALEP also hired international consultants (Camoy and others 2000) to assess the evolution of the CONALEP system. For this purpose the consultants used a different data set as is explained below. 18 The CONALEP (1994) and (1999) tracer studies had several problems, one of the most important being the lack of a well defined control group. A control group was expected to be added later, using data from the National Urban Employment Survey (ENEU). However, the studies neither include in-depth information on how the analysis was performed nor do they provide useful information on how CONALEP graduates perform relative to a control group. Lane and Tan (1996) also encountered several problems in their evaluation. The first was the construction of a non-arbitrary control group. The ENEU sample is representative of metropolitan areas while the CONALEP graduate tracer survey is representative nationally. The difference in geographical coverage of the two groups makes comparison difficult. Second, the control groups were constructed ad hoc. The control groups included individuals between the ages of 17 and 30: (a) those who have completed lower-secondary education; (b) those who have completed non-professional, elementary vocational training (CECATI), and (c) those who have completed one to three years of general academic (non-vocational) high school. Some doubts remain with respect to the second group, since the ENEU survey does not distinguish between formal and informal training/technical courses. Lee (1998) compares the individuals from the Encuesta de Egresados 1994 (the treatment group) with two other groups. One group comprises all 1991 graduates from upper-secondary diversified technical education programs; this group's labor force participation and employment performance in January 1994 was compared with that of CONALEP graduates of 1991, and of 1991-93 combined. The first comparison group was created from a mail survey of all graduates, with a 45 percent response rate, and therefore is likely to be biased toward those who were either employed, studying, or had a higher level of earnings. The second comparison group was made 19 up of employed workers aged 20 to 24, as reported in the aggregates of the ENEU of January 1994. The results of these evaluations concluded that CONALEP graduates actively participated in the labor market at a much higher rate than the similar age cohort of the general population, and at a much higher rate than graduates from traditional technical high schools. On average, CONALEP graduates found jobs faster than control individuals, and about two-thirds of CONALEP graduates worked in jobs related to the specialization they had studied. Using cross- cohort comparison, these evaluations also suggested that CONALEP graduates' earnings increased rapidly within the first two to three years of employment. These conclusions are as expected, although the magnitudes of the participation rate and the increase in earnings in comparison to the magnitudes in traditional technical high schools and the general population are surprising-thirty percent in Lane and Tan, and forty percent in Lee. The results should be considered with caution, since these studies failed to control for possible self-selection bias that could account for different labor market outcomes between the CONALEP group and the comparison groups. In addition, some of these evaluations do not fully explain how the control groups were constructed. A fourth evaluation, aimed at understanding the background experience and goals of CONALEP students, conducted a survey with five percent of the senior students (ready to graduate) and freshmen students, the control group. The sample was 4,930 third year students and 725 first year students who, on the basis of their responses, were then divided into three groups using a socioeconomic status indicator. The results confirm the assumption that close to one-third of the students from CONALEP come from a low socioeconomic background. Another 20 40 percent come from a middle socioeconomic range. About 18 percent have parents with basic secondary school or more, own their own home with four or more rooms and have either a car, a phone, or both. The average entry test scores for the sample show several important trends in social class, gender, and cohort, as described below. Girls in both cohorts enter CONALEP with slightly lower scores than boys. The first year (1999) cohort entered with higher scores than the third year (1997) cohort. We would assume that a higher fraction of those in the 1997 cohort who had lower entry scores would have dropped out by the third year. Thus, we could conclude that CONALEP student entry scores have actually risen more than suggested by the data. In the third-year cohort, entry scores positively correlated with rising socioeconomic indicators for both boys and girls. However, there seems to be little relationship between socioeconomic status and entry score in the 1999 cohort, except for higher-class girls. In sum, CONALEP students come from relatively low socioeconomic backgrounds and tend to score at the lower passing end of the higher secondary school entry test. About half have general basic secondary education, with another third coming from basic technical secondary schools. Somewhat less than half of the third year students indicate that the CONALEP option was their first choice of higher secondary school, and somewhat more than half of the first year cohort say it was their first choice. A second questionnaire was given to firms that hired CONALEP graduates from regular courses or training courses. In general, the interviewed firms who hire students from CONALEP and use its training services think highly of the organization. Approximately 72 percent of firms (public lower, private higher) think that the academic level attained by CONALEP students is 21 high or very high. About 55 to 60 percent of companies said that the technological level of a CONALEP education is high or very high, with large public companies giving the lowest ranking (46 percent). VI. Data The CONALEP Graduate Tracer Surveys This paper re-evaluates CONALEP's effectiveness using the CONALEP graduate tracer surveys conducted in 1994 and 1998.5 The first CONALEP graduate tracer survey was conducted in February 1994 (CONALEP, 1994) on the basis of a random sample of 1500 former CONALEP students who graduated between June 1991 and June 1993. The surveyed graduates were selected to represent the profile of the graduates in each of the three years in terms of all 13 major occupational groups of careers and the six geographical regions of the country. However, the sample is dominated by 1992 graduates who comprise 50 percent of the sample; 1991 and 1993 graduates each represent 25 percent. The sample selection is probabilistic and statistically representative of the universe of graduates in each cohort. For each graduate (M), two substitutes were chosen from the same career and school (S and T). Table 3. Distribution of the 1994 Sample by Cohort Graduation Year Planned Selection % Actual Selection % Cohort 1991 375 25 346 24.7 1992 750 50 704 50.3 1993 375 25 349 24.9 Total 1500 100 1399 100 Source: CONALEP (1994). 5A third CONALEP graduate tracer survey was conducted in January of 2001. The data are expected by mid-2001. 22 Table 4. Actual Sample Selection (original and substitutes by cohort) Selected Substitutes Total % vs 1,500 Graduation Year Cohort M S T Z 346 23.1 1991 268 53 20 5 704 49.9 1992 560 96 42 6 349 23.3 1993 286 46 15 2 1,399 93.3 Total 1,114 195 77 13 Cumulative percentage 74.3% 87.3% 92.4% 93.3% 93.3% Source: CONALEP (1999). The second CONALEP Graduate Tracer Study (CONALEP 1999) was conducted between May and June of 1998 on the basis of a random sample of individuals who graduated between June 1993 and June 1997. The sample is representative of geographical regions, all 29 careers and all cohorts. The difference between the actual sample of 5,574 individuals and the planned sample of 10,000 was due to exogenous factors such as changes in address (3,590 cases); addresses that belonged to different states (651 cases); differences between the number of graduates officially registered and those found in the administrative records (229 cases), and technical careers that had never been offered (7 cases). CONALEP (1998) extensively reviews the sample frame of the second CONALEP Graduate Survey as described by LEVANTA, the consultant firm which designed the sample process. The distribution of the 1998 CONALEP survey was as follows. The table shows that the response rate is high. Table 5. Distribution of the 1998 Sample by Cohort Cohort Interviewed Graduates % Completed Number % Interviews 90-93 779 14.0 59.0 91-94 951 17.1 72.0 92-95 1,127 20.2 85.4 93-96 1,268 22.7 96.1 94-97 1,449 26.0 109.86 Total 5,574 100.0 84.5 Source: LEVANTA C. 6 This value, as listed in CONALEP data sets, appears to exceed 100 percent because the number of responding graduates exceeded the goal number. 23 The ENE98 and ENECE99 Surveys Two other surveys are used in this paper, The National Employment Survey (ENE) and the National Employment, Schooling, and Training Survey (ENECE). The first is representative at a national level and by urban and rural areas. It has rich information on individual labor market characteristics. The ENE98 has a sample size of nearly 200,000 individuals. The second survey is a module of the National Employment Survey. The 1999 sample size was 164,550 individuals. The ENECE is also representative at the national level and has useful additional information on the professional profile of the individuals and the training status, such as type of training received, training time, date of training, place of training, etc. VI.1 Methodology In order to compare CONALEP graduates to a control group, this paper examines labor force participation, employment status, earnings, training and hours worked for both the CONALEP group and the control group. To construct the control group, this paper uses the statistical approach of propensity score matching. As discussed by Ravallion (1999) and Todd (1999), the idea behind matching is to find a comparison group that is as similar as possible to the treatment group in terms of the relevant observable characteristics such as age, sex, education, region of residence, as summarized by the propensity score. In calculating the propensity scores, we followed Ravallion's methodology (1999) and Gill and Dar (1995). First, we chose two representative sample surveys of eligible non-participants as well as one of the participants. The two surveys of eligible non-participants are The National Employment Survey of 1998 (ENE98) and the National Education, Training, and Employment 24 Survey of 1999 (ENECE99). Both surveys have the advantage of a large number of eligible non- participant respondents, which ensures good matching. The participant survey used is the 1998 CONALEP graduate tracer study. Although the participant and non-participant data come from different surveys, the surveys are comparable since some of the questions are identical, all are from similar survey periods, and all are nationally representative. Next, the two samples were pooled and a logit model of CONALEP participation as a function of the variables that are likely to determine participation was estimated. The variables included were age, sex, education, region of residence, and the location where training was under taken. The predicted values of the probability of participation were created from the logit regression -the propensity scores. There was a propensity score for every sampled participant and non-participant.' The goodness of fit and the models estimations are shown in Tables AA.4, A1.5 and A1.6. These models consistently classified correctly 99 percent of the non participant group cases and 72 percent of the participant group cases. The overall percentage of correctly predicted cases is 98 percent. Then we calculated propensity scores of the three and five nearest neighbors. This means that for each individual in the CONALEP group, the three and five observations in the non participant sample that have the closest propensity score were found, as measured by the absolute differences in scores. Alternatively, another transformation was used, the lag-odds ratio log (p/(1-p)), where p is the propensity score for matching. Heckman and others (1998) have proposed an alternative method for the nearest neighbor. Instead of relying on the nearest neighbor, they use all the non-participants as potential matches but weigh each according to its proximity. 7Those individuals staying at home, in an education program, or with zero hours of work were excluded from the sample. 25 The mean values of the outcome indicators for the three and five nearest neighbors were computed using labor market status, hourly earnings, earnings, economic sector, and training. The difference between the mean and the actual value for the treated observation is the estimate of the gain due to the program for that observation. The mean of these individual gains was computed to obtain the average overall gain. VI.2 Results In order to assess CONALEP's effectiveness, we examine CONALEP graduates versus the control group in terms of labor force participation, status in the labor market, sector, further training at work, wages, and hours worked. Interpretation and tabular data of each area are presented in the following subsections. Labor Force Participation' Figure 2. Irrespective of distance criteria or nearest Percent of Individuals Seeking Jobs neighbors, the proportion of individuals seeking 94 - 97 employment in the CONALEP group is higher than in the control group. It is unclear whether labor force 9 0 participation of the CONALEP group has declined with 91 - 94 respect to the control group over time. Additionally, the 90-93 percent of individuals who are searching for a job is 0% 5% 10% 15% Percent higher in the CONALEP group than in the control group. *CtrI Group mConalep 8 Data for this section are presented as follows. Table 6 shows the labor force participation of the CONALEP graduates compared to the ENE98 control group. Table 7 shows the labor force participation of the CONALEP graduates compared to the ENECE99 control group. Both tables were calculated using the three nearest neighbors' distance. Tables A2. 1 and A2.2 show the results using the five nearest neighbors criteria. 26 It is difficult to interpret why this proportion increased substantially for the cohort graduating in 1996, a crisis recovery year. It appears that the peso crisis, from which Mexico recovered in 1995-6, had a much larger effect on CONALEP graduates than it did on control group individuals (Tables 6-7). The labor force participation rate of CONALEP graduates is shown in Table 6.9 Contrary to previous studies, the results indicate that the share of CONALEP graduates in the working population is lower than the control group. Moreover, the CONALEP job search share is higher compared to the control group. Further analysis might be needed to explain the greater percent of CONALEP graduates who are searching for a job. Results also suggest that between 2 and 3.5 percent more control individuals worked without pay than CONALEP graduates did (Tables 8- 9). Although between 3.9 and 5.6 percent more control individuals are employed than CONALEP individuals are, CONALEP individuals earn between 20 and 27.5 percent more per hour than control individuals do (Tables 6-7, 15-16). It appears, then, that the lack of employment of CONALEP graduates relative to the control group does not translate into a lack of income. 9 Only those working or searching for a job were considered in the matching exercise. 27 Table 6. Labor Force Participation by Cohort Matching group: Age 17-65. Three nearest neighbors based on propensity scores Working people Searching for ajob Cohort CtrL Group CONALEP Difference Ctr/ Group CONALEP Difference 90- 93 94.1 93.0 -1.0 5.9 7.0 1.0 91 - 94 96.4 93.6 -2.8 3.6 6.4 2.8 92 - 95 95.2 89.9 -5.3 4.8 10.1 5.3 93-96 94.7 88.9 -5.8 5.3 11.1 5.8 94 - 97 93.1 90.5 -2.6 6.9 9.5 2.6 Total' 94.8 91.2 -3.6 5.2 8.8 3.6 ENE 982 97.5 2.5 ENE 98, LS' 94.5 5.5 ENE 98, US 95.7 4.3 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). Table 7. Labor Force by Cohort Matching group: Age 17-65. Three nearest neighbo based on propensity scores Working people Searching for ajob Cohort Ctrl. Group CONALEP Difference Ctrl. Group CONALEP Difference 90 - 93 97.0 94.5 -2.5 3.0 5.5 2.5 91 - 94 95.7 93.1 -2.7 4.3 6.9 2.7 92 - 95 96.3 88.3 -8.0 3.7 11.7 8.0 93 - 96 94.7 88.8 -5.9 5.3 11.2 5.9 94 - 97 95.7 87.9 -7.8 4.3 12.1 7.8 Total' 95.9 90.8 -5.1 4.1 9.2 5.1 ENECE 992 98.1 1.9 ENVECE 99, LS3 95.7 4.3 ENECE 99, US' 98.4 1.6 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). 28 Employment Status'" In general, there are not substantial differences between the employment status of CONALEP graduates compared to the control groups using either ENE98 or ENECE99. A large proportion of both CONALEP graduates and the control group individuals are employees. Albeit, the proportion of CONALEP graduates that are employees or wage earners (84.3 and 83.8) is less than in the control groups (86.5 and 84.6). The proportion of self-employed is higher among CONALEP graduates (9.8) than it is in the ENE98 control group (7.5). There is also no clear pattern of this proportion through time. Figure 3. Interestingly, the proportion of self-employed in the Employment Status, Conalep v. Control 1991-1994 cohort (5.3) is higher compared to the self- employed in the 1993-1996 cohort (2.6). This might indicate that self-employment increases as graduates gain more work experience." In relation to employment sectors, commerce, .3 -2 -1 0 2 3 restaurants, hotels, personnel, communications, and Percent by which government have the highest percent of CONALEP graduates (33.8, 24.1 and 31.9 respectively). Unsurprisingly, these sectors also employ the largest share of individuals in the control groups. In Mexico, both manufacturing and services employ close to 80 percent of the labor force. Few ' Data for this section are presented as follows. Table 8 shows the employment status of the CONALEP graduates compared to the ENE98 control group. Table 9 shows the employment status of the CONALEP graduates compared to the ENECE99 control group. Tables A2.3 and A2.4 show the employment status using the five nearest neighbors criteria Table 10 shows the proportion of CONALEP graduates and the proportion of ENE98 individuals in the control group by economic sector. Table II shows the proportion of CONALEP graduates and the proportion of ENECE99 individuals in the control group by economic sector. Tables A2.5 and A2.6 show the results using the five nearest neighbors criteria. " Maloney (2000) asserts that some Mexican workers are joining the informal sector voluntarily at the prospect of higher incomes. Furthermore, at least for some workers, especially those with limited educational achievements, leaving formal sector employment represents a desirable professional move which entails more responsibilities and higher pay. 29 CONALEP graduates work in the primary sector, the extraction (mining) sector or the electricity and gas sectors. With respect to overall patterns of employment, considering both sector and labor market status, the results for the CONALEP group are very similar to those obtained for the control groups. An important feature, however, is that CONALEP offers careers that are demanded in the manufacturing and service sectors. Due to the ENE98 limitations, it is not possible to assess in detail the type of job obtained by the individual. However, the CONALEP graduate tracer survey allows us to infer whether there is congruency in the CONALEP graduate professional profile. Among the employed CONALEP graduates, more than half reported that they were working in the occupational category congruent with their field of specialization. Close to 70 percent of employed graduates consistently reported that CONALEP training or specialization was "very useful" or "useful" in their current occupation. This high rate of congruency might be comparable to the high rate among apprentices in Germany, but it is significantly higher than in other developed countries (OECD 1997). 30 Table 8. Employment Status. Matc in oup: Age 17-65. Three nearest neighbors base on propensity scores Employer Self-employed Employee Cooperative membership Worker without pay Control CONALE Control CONALE Control CONALE Control CONALE Control CONALE Diference Cohort Gop Difference Gop Difference Gop Difference Gop DiffrneGop p Dfeec 90 - 93 1.0 2.8 1.9 8.0 9.3 1.4 85.6 85.8 0.2 0.0 0.0 0.0 5.4 2.0 -3.4 91 - 94 1.2 2.9 1.7 7.3 12.5 5.3 88.2 81.4 -6.8 0.0 0.7 0.7 3.4 2.5 -0.9 92 - 95 1.6 2.3 0.7 7.4 9.4 2.0 85.5 84.8 -0.8 0.0 1.2 1.2 5.4 2.3 -3.1 93 - 96 2.4 3.6 1.1 8.1 10.7 2.6 85.1 83.0 -2.1 0.0 0.8 0.8 4.4 2.0 -2.5 94 - 97 0.9 1.3 0.4 6.7 6.4 -0.3 88.1 87.2 -1.0 0.0 1.3 1.3 4.2 3.8 -0.4 Total' 1.4 2.6 1.2 7.5 9.8 2.3 86.5 84.3 -2.2 0.0 0.8 0.8 4.6 2.5 -2.0 ENE 982 4.3 24.1 60.2 0.04 11.4 ENE 98, LS' 0.2 4.8 77.9 0.02 17.1 ENE 98, US4 2.5 11.0 77.8 0.00 8.7 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). Table 9. Employment Status. Matching group: Age 17-65. Three nearest neighbors based on propensity scores _ Employer Self-employed Employee Cooperative membership Worker without pay Cohort Control CONALE Diference Control CONALE Difference Control CONALE Difference Control CONALE Diference Control CONALE Diference Group P ifeec Group P feec Group p ffeec Group p feec Group p 90 - 93 1.2 4.0 2.8 11.0 13.3 2.3 84.0 82.0 -2.0 0.0 0.0 0.0 3.7 0.7 -3.0 91 - 94 2.1 1.8 -0.4 7.0 13.6 6.7 86.1 81.7 -4.4 0.0 1.2 1.2 4.8 1.8 -3.0 92 - 95 1.2 2.9 1.6 8.6 10.0 1.4 85.2 85.7 0.5 0.0 0.7 0.7 4.9 0.7 -4.2 93 - 96 2.7 6.9 4.2 8.1 6.2 -1.9 81.2 84.6 3.4 0.0 0.0 0.0 8.1 2.3 -5.7 94 - 97 0.9 1.0 0.2 6.9 7.1 0.2 86.2 86.7 0.5 0.9 1.0 0.2 5.2 4.1 -1.1 Total' 1.7 3.3 1.7 8.4 10.5 2.1 84.6 83.8 -0.7 0.1 0.6 0.5 5.3 1.7 -3.5 ENECE 992 4.0 24.4 60.8 0.03 10.7 ENECE 99, LS' 0.2 4.4 79.6 0.00 15.9 ENECE 99, US' 2.9 9.2 81.4 0.02 6.5 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). 31 Table 10. Economic Sector. Matching group: Age 17-65. Three nearest neighbors based on propensity Agriculture, fishing, etc. Extraction Manufacturing Construction Electricity, gas, and water Control CONAL Control CONAL Dif Control CONAL . Control CONAL Control CONAL Diference Cohort Group EP Group EP Group EP Group EP Group EP 90 - 93 2.3 0.8 -1.4 0.1 0.0 -0.1 22.2 29.8 7.6 3.0 1.2 -1.8 0.8 0.8 0.0 91-94 1.3 2.8 1.5 0.1 0.4 0.2 20.4 33.3 12.9 2.5 2.1 -0.4 0.4 0.4 -0.1 92-95 2.1 0.4 -1.7 0.1 0.8 0.7 24.4 35.7 11.3 2.4 3.1 0.7 0.2 0.4 0.2 93-96 2.5 0.4 -2.1 0.1 0.4 0.3 25.7 32.1 6.4 2.6 4.0 1.3 0.7 0.4 -0.3 94 - 97 1.5 0.9 -0.6 0.1 0.0 -0.1 25.8 38.5 12.6 2.5 3.0 0.5 0.1 0.4 0.3 Total' 1.9 1.1 -0.8 0.1 0.3 0.2 23.6 33.8 10.2 2.6 2.7 0.1 0.5 0.5 0.0 ENE 982 20.3 0.4 18.1 5.51 0.5 ENE 98, LS' 17.0 0.1 31.8 4.57 0.2 ENE 98, US' 6.3 0.5 19.7 2.22 1.3 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience ( 8 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). Table 10. (cont.) Commerce, Restaurants, and Transportation and Financial Services Personnel, Common, and Hotels Communications Social Services - Cohort Control CONAL Diffe Control CONAL Difference ro Dffe Group EP Diference 90 - 93 27.8 29.8 2.0 4.9 4.9 0.0 2.4 4.1 1.7 36.5 28.6 -7.9 91 - 94 29.2 23.9 -5.3 4.9 2.5 -2.4 2.7 2.5 -0.3 38.5 32.3 -6.2 92-95 29.0 22.5 -6.5 5.6 3.5 -2.1 3.1 3.1 0.0 33.2 30.6 -2.6 93 - 96 27.0 22.2 -4.8 4.5 1.6 -3.0 2.0 1.6 -0.4 34.8 37.3 2.5 94-97 27.1 22.2 -4.9 5.3 2.6 -2.7 1.0 2.1 1.1 36.6 30.3 -6.2 Total' 28.1 24.1 -4.0 5.0 3.0 -2.0 2.3 2.7 0.4 35.9 31.9 -4.0 ENE 982 22.4 4.4 1.0 27.5 ENE 98, LS' 26.0 3.6 0.2 16.5 ENE98, US' 31.6 8.1 2.9 27.6 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). 32 Table 11. Economic Sector. Matching group: Age 17-65. Three nearest neighbors based o propensity Agriculture, fishing, etc. Extraction Manufacturing Construction Electricity, gas, and water Control CONAL . Control CONAL Control CONAL . Control CONAL Control CONAL Cohort Group EP Group EP Group EP Difer Group EP D r Group EP 90 - 93 1.2 0.0 -1.2 0.0 0.0 0.0 26.8 30.4 3.6 3.6 0.7 -2.9 0.0 2.0 2.0 91 - 94 1.5 1.2 -0.4 0.0 0.6 0.6 24.7 33.9 9.2 3.6 1.8 -1.8 1.0 0.6 -0.4 92 - 95 2.4 0.7 -1.6 0.0 1.5 1.5 18.9 27.7 8.8 3.0 1.5 -1.5 0.6 0.7 0.1 93-96 2.6 0.0 -2.6 0.6 0.0 -0.6 23.1 33.8 10.8 3.8 2.3 -1.5 0.0 0.8 0.8 94-97 3.3 1.0 -2.3 0.0 0.0 0.0 24.0 .35.4 11.4 3.3 3.1 -0.2 1.7 0.0 -1.7 Total' 2.1 0.6 -1.5 0.1 0.4 0.3 23.5 32.1 8.6 3.5 1.8 -1.7 0.6 0.9 0.3 ENECE 992 21.0 0.3 18.8 5.53 0.5 ENECE 99, LS' 17.2 0.1 30.5 4.71 0.1 ENECE 99, US' 4.1 0.2 18.2 2.29 0.3 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). Table It. (cont.) Commerce, Restaurants, and Transportation and Financial Services Personnel, Common, and Hotels Communications Social Services Cohort Control CONAL Difference Control CONAL Difference Control CONAL Difference Control CONAL Difference Group EP Group EP Group EP Group _EP 90-93 22.6 23.0 0.4 4.2 5.4 1.2 3.0 4.1 1.1 38.7 34.5 -4.2 91-94 28.4 25.6 -2.8 6.2 4.2 -2.0 2.1 2.4 0.3 32.5 29.8 -2.7 92 - 95 30.2 25.5 -4.6 7.1 2.2 -4.9 1.8 4.4 2.6 36.1 35.8 -0.3 93 - 96 28.8 20.8 -8.1 4.5 0.8 -3.7 0.0 3.1 3.1 36.5 38.5 1.9 94 - 97 27.3 21.9 -5.4 4.1 2.1 -2.0 1.7 0.0 -1.7 34.7 36.5 1.7 Total' 27.5 23.6 -3.9 5.3 3.1 -2.2 1.7 2.9 1.2 35.6 34.6 -1.0 ENECE 992 21.5 4.5 0.9 27.0 ENECE 99, LS' 22.1 2.1 0.2 22.9 ENECE 99, US' 32.8 8.8 2.9 30.5 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). 33 Further Training'2 About 39 percent of CONALEP workers receive Figure 4. further training at work, a significantly higher level than Rceived Trairngr the 37.2 percent of control group individuals that do (Table 12). It appears that government investment in CONALEP training for an individual leads to additional investment by firms in training for the same individual. Moreover, a significant proportion of CONALEP 35 : 6 graduates (89.7 percent) report that their training was PactSt d FbtRds related to their current employment or work activity. Figure 5. Trahing related tD wk Although the 1994-1997 cohort shows a ..E. considerable increase in the proportion of CONALEP graduates receiving training related to ! . ! il | !work, a lower proportion of this cohort reported Oortd W 3 I Ireceiving further training than older cohorts did. In response to a question asking the purpose of further 84 85 85 8 8 89 9) Powt d Pespmderts training, nearly 60 percent of CONALEP graduates said that they received training in order to update their technical knowledge (Table 14). Compared to the ENECE99 control group rate (near 32 percent), the CONALEP rate is quite high. This could indicate that employers invest more in training CONALEP graduates than they do in the control group because investment in the 12 Data for this section are presented as follows. Table 12 shows the proportion of CONALEP graduates compared to the ENECE99 control group that receive further training at work. Table 14 indicates the reasons for further training among CONALEP graduates as well as individuals from the control groups. 34 CONALEP group is more profitable. Compared to the 1994-1997 cohort, the rate of further training is slightly higher for the 1991-1994 cohort and much higher for the 1990-1993 cohort, which could mean that individuals in older cohorts need to update their skills in order to work efficiently. The proportion of CONALEP graduates who Figure 6. undergo training because it is a job prerequisite Earnings per month increases over time. This suggests two possible C p explanations: (a) employers' expectations of CONALEP graduates rise as they become more familiar with them, Control and (b) the technical complexity of jobs held by 1800 2000 2200 CONALEP graduates increases over time. 1998 Pesos Earnings and Hours Worked3 On average, CONALEP graduates earned 17 percent more than the ENE98 control group and 22 percent more than the ENECE99 control (Tables 15 and 16). Controlling for hours Figure 7. worked, CONALEP graduates earn close to 20 Earnings per hour percent more than the ENE98 control group and Conalep 27.5 more than the ENECE99 control group. Even after the 1994 crisis, CONALEP graduates obtained Control I higher earmings than individuals in the control 10 11 12 13 group did. This might indicate that CONALEP has 1998 Pesos been a good alternative for low income individuals seeking a lasting job. Furthermore, it seems that in downturns such as in the 1994 crisis, 3 Tables 15 and 16 show average earnings, average earnings per hour, and average hours worked for CONALEP graduates and 35 CONALEP served as an altemative to other programs. The 1997 results are surprising, but as anomalous data this cohort may be treated as an outlier. Table 12. Training Received at Work Matching group: Age 17-65. Three nearest neighb rs based on propensity Yes No Cohort Ctrl. Group CONALEP Difference Ctrl. Group CONALEP Difference 90 - 93 38.8 45.8 7.0 61.2 54.2 -7.0 91 - 94 38.2 34.3 -3.9 61.8 65.7 3.9 92 - 95 39.9 43.4 3.5 60.1 56.6 -3.5 93 - 96 34.4 38.5 4.1 65.6 61.5 -4.1 94 - 97 33.3 30.7 -2.6 66.7 69.3 2.6 Total' 37.2 38.9 1.7 62.8 61.1 -1.7 ENECE 992 22.1 77.9 ENECE 99, LS' 12.1 87.9 ENECE 99, US' 39.1 60.9 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). Table 13. Training Related to Work Matching group: Age 17-65. Three nearest neighbors based on propensity YES NO Cohort Ctrl. Group CONALEP Difference Ctrl. Group CONALEP Difference 90 - 93 90.9 95.7 4.7 9.1 4.3 -4.7 91 - 94 84.2 88.3 4.1 15.8 11.7 -4.1 92 - 95 85.5 85.5 0.0 14.5 14.5 0.0 93 - 96 87.3 88.2 1.0 12.7 11.8 -1.0 94 - 97 81.4 90.0 8.6 18.6 10.0 -8.6 Total' 86.1 89.7 3.6 13.9 10.3 -3.6 ENECE 992 86.1 13.9 ENECE 99, LS' 74.2 25.8 ENECE 99, US' 77.7 22.3 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). the ENE98 and ENECE99 control groups respectively. 36 Table 14. Reasons for Training. Ma tching group: Age 17-65. Three nearest neighbors bas ed on propensity Prerequisitefor entering a job Update Self-interest Was Requested Other Cohort Control CONAL Diferece ontrol CONAL Difference Control CONAL Differece ontrol CONAL Dffeence ontrol CONAL ff GroupDifeEnceGroup EP fern Group EP Dee Group EP Gop E 90-93 9.1 7.1 -1.9 31.8 60.0 28.2 22.7 15.7 -7.0 27.3 10.0 -17.3 9.1 7.1 -1.9 91-94 9.1 8.3 -0.8 40.3 53.3 13.1 24.7 18.3 -6.3 20.8 11.7 -9.1 5.2 8.3 3.1 92 - 95 7.4 6.5 -0.9 39.7 53.2 13.5 22.1 24.2 2.1 26.5 8.1 -18.4 4.4 8.1 3.7 93 - 96 17.0 17.3 0.3 35.8 51.9 16.1 13.2 9.6 -3.6 30.2 7.7 -22.5 3.8 13.5 9.7 94-97 9.1 16.7 7.6 31.8 43.3 11.5 27.3 30.0 2.7 27.3 6.7 -20.6 4.5 3.3 -1.2 Total' 10.1 10.2 0.2 36.4 53.6 17.3 22.1 18.6 -3.5 26.0 9.1 -16.8 5.5 8.4 2.9 ENECE 992 10.7 37.9 22.0 24.6 4.8 ENECE 99, LS3 30.7 11.4 27.7 25.4 4.8 ENECE 99, US' 13.9 27.2 27.5 26.9 4.5 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). 37 Table 15. Average Earnings, Earnings per Hour, and Hours Worked by Sample Group Matching group: Age 17-65. Three nearest neighbors based on propensity (1998 pesos) Earnings Earnings per hour Hours worked per week Cohort Control CONALE Difference Control CONALE Diffe Control CONALE Difrence Group - ru ru 90 - 93 1910.5 2660.0 749.5 10.6 14.3 3.7 44.5 46.4 1.9 91 - 94 1851.1 2170.4 319.4 11.0 12.4 1.3 43.8 45.6 1.8 92 - 95 1883.6 2262.3 378.7 10.7 14.9 4.2 43.5 46.0 2.4 93 - 96 1980.1 2288.8 308.8 10.7 12.3 1.6 45.3 46.9 1.6 94 - 97 1826.9 1655.1 -171.9 10.1 9.6 -0.5 44.3 45.0 0.7 Total' 1890.4 2208.7 318.3 10.6 12.7 2.1 44.3 46.0 1.7 ENE 982 2046.0 11.6 43.5 ENE 98, LS' 1266.3 6.6 44.8 ENE 98, US' 2088.5 11.2 45.8 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). Table 16. Average Earnings, Earnings per Hour, and Hours Worked by Sample Group Matching group: Age 17-65. Three nearest neighbors based on propensit (1998 Pesos) Earnings Earnings per hour Hours workedper week o Control CONALE Control CONALE Control CONALE Cohort Group Difference Group P Diference Group P Diference 90 - 93 2135.9 2878.9 743.0 11.5 15.3 3.8 44.2 45.3 1.1 91 - 94 1860.4 2177.5 317.1 10.4 12.5 2.1 44.4 46.0 1.6 92 - 95 2003.5 2757.3 753.8 11.1 16.5 5.4 43.9 44.7 0.8 93-96 2081.3 2561.7 480.4 11.5 13.6 2.1 43.7 46.8 3.0 94 - 97 1772.5 1733.5 -39.0 9.7 11.0 1.3 45.4 45.8 0.3 Total' 1977.1 2454.5 477.3 10.9 13.9 3.0 44.3 45.7 1.4 ENECE 952 1984.4 11.2 43.8 ENECE 99, LS' 1253.1 6.6 44.8 ENECE 99, US' 2142.2 11.4 46.0 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and 1-5 years of experience (22-26 years old). 38 VII. Benefits from CONALEP Reformed Program Given that the CONALEP graduate tracer surveys of 1994 and 1998 are comparable, this section analyzes the impact of modular courses and reforn programs, innovations implemented by CONALEP after 1992. The cohorts from the survey of 1994 are considered to be graduates of the pre-reform program including cohorts who graduated in 1991, 1992, and 1993. Additionally, cohorts from the survey of 1998 are considered to be from the post-reform program including cohorts who graduated in 1994, 1995, 1996, and 1997. The first subsection describes the methodology used in the analysis, and the second discusses the results. VH.1 Methodology There are several methods for estimating duration models. The Kaplan Meier and the proportional hazard model were calculated to analyze whether graduates from the cohorts of the 1998 survey of the reformed program found a job faster than those from the cohorts of the 1994 survey of the pre-reformed program. In addition, multinomial models were estimated to assess if the reformed program increased individuals' probability of studying further after completing CONALEP. They also permitted estimation of the probability of having a certain status in the job market. Regression models were estimated to assess if the reformed CONALEP program increases CONALEP graduates' earnings. See Annex 3 for details of the methodology used. VII.2 Results Table 17 shows the Kaplan-Meier estimates of the mean and median time of job search after graduation. The median time for cohorts graduating before the reforms were introduced in CONALEP (Survey 94) is 4 months, while for those cohorts in the reformed program (Survey 39 98) the median time is 3 months. The preliminary finding based on the Kaplan Meier estimates is important because it shows that graduates from the 1998 Survey found jobs faster than those from the 1994 Survey. If only a simple average of values for surveyed individuals had been taken, the mistaken conclusion would be reached that graduates of the 1998 Survey search for a job longer than individuals of the 1994 survey do. Table 17. Kaplan Meier Estimates Job Search after graduation from CONALEP (Months) Time Estimated Probability (T>t) Survey 94 Survey 98 0.25 12.0 7.0 0.5 4.0 3.0 0.75 1.6 1.0 Mean 9.2 6.8 Median 4.0 3.0 Cases censored 169 (20.4%) 124 (3.8%) Total number of cases 827 3273 Table A4.1 shows the hazard ratio or risk of finding a job and a respondent's region."4 Graduates from the 98 Survey have a 45 percent greater probability of finding a job than graduates from the 94 Survey do. Graduates from the North or Center of Mexico have a higher probability of finding a job (between 60 and 45 percent) than graduates from the South (22 percent) do. The 1993-1996 cohort had a 4 percent higher probability of finding a job than the other cohorts did. Also, the higher the level of schooling of the household head, the higher the chance of the CONALEP graduate of finding a job. 14 Since employment is not conventionally a risk, we shall refer to the technical "risk" of finding employment as the "probability" or "chance." 40 In Table 18, some scenarios were calculated. Given a base category (male, living in the Center region, age, etc.), the median time for a male graduate to find a job in the 94 Survey is 4 months, and in the 98 Survey 2.8 months. The mean length of time that female graduates search for a job is longer than that for male graduates. Furthermore, female graduates in the 98 Survey found jobs faster than those from the 94 Survey did. Not surprisingly, the job search is longer for graduates without job experience while studying. However, the difference in job search time between individuals with work experience and those without is small. Table 18. Cox Regression Model. Job Search (time) after finishing CONALEP Covariate** Survey 94 Survey 98 Difference Male 4.0 2.8 1.2 Female 4.9 3.0 1.9 Age (mean=21.5 years) 4.0 2.8 * 1.2 Age (22.5 years) 4.0 2.8 * 1.2 Center Area 4.0 2.8 1.2 South-East Area 2.9 1.9 1.0 Center-South Area 2.9 1.9 1.0 North-East Area 2.0 1.9 0.1 North-West Area 2.8 1.9 0.9 Center-North Area 2.8 1.9 0.9 Pacific Area 2.8 1.9 0.9 91 Cohort 4.0 92 Cohort 4.0 93 Cohort 5.0 94 Cohort 4.9 96 Cohort 2.8 97 Cohort 2.9 None - Primary HH 4.0 2.8 1.2 Lower-Secondary HH 4.0 2.8 * 1.2 Upper-Secondary HH 4.0 2.8 * 1.2 University - + HH 4.0 2.0 * 2.0 Don't know HH 4.0 2.0 * 2.0 GDP per capita (mean=38.8) 4.0 2.8 * 1.2 GDP per capita 3.0 2.0 1.0 Worked when studying 4.0 2.8 1.2 Did not work when studying 5.0 3.0 2.0 ** The other covariates are in the base category or at the mean. Base categories: men, Center area (D.F. and Mexico), 91 cohort, none or primary school education of household head (HH), and had worked when he was studying at CONALEP. Means: 21.5 years old, 38.8 thousands of 1998 pesos GDP per capita. *Not significant at 5 percent. 41 Given the base category, 18 percent more of 98 Survey respondents were working than 94 Survey respondents were (Table 19). In the 94 Survey, the North, Center, and Pacific regions correlate with an increased probability of working. The probability of searching for a job is 17 percent more for the 1994 graduates than it is for the 1998 graduates. For 1994 female graduates, the probability of working is 16 percent higher than for 1998 female graduates. Also, the probability of searching for a job is slightly higher for the 1994 graduates than it is for 1998 graduates. The probability that a graduate continues to study is three percent higher in the 1994 than in 1998 Survey. For female graduates the probability of working at home as housewives is 16 percent higher in the 98 cohort (Table 19). The 94 Survey respondents earn higher hourly wages than the 1998 cohorts do (Tables A4.2, A4.3). A plausible explanation is that since 1994, real wages in Mexico have declined by almost 40 percent. 42 Table 19. Marginal Effects of Having a Certain Type of Activity af er Completing CONALEP Prob[Acdvity=j I X, dummykl Marginal Prob[ActivityV-j I X, dummy=kJ Marginal Variable k=1 kre Effect Variable k=1 kSv9 Effect Survey 98 Survey 94 Survey 98 Survey 94 j = Work j = Searching forjob 98 Survey 0.77 0.59 0.183 98 Survey 0.08 0.26 -0.174 Women 0.64 0.80 -0.162 Women 0.10 0.11 -0.012 Age 0.004 Age 0.000 South-East Area 0.78 0.74 0.040 South-East Area 0.09 0.11 -0.027 Center-South Area 0.81 0.74 0.076 Center-South Area 0.08 0.11 -0.030 North-East Area 0.85 0.73 0.122 North-East Area 0.08 0.12 -0.039 North-West Area 0.82 0.73 0.085 North-West Area 0.06 0.12 -0.055 Center-North Area 0.84 0.73 0.111 Center-North Area 0.06 0.12 -0.055 Pacific Area 0.84 0.73 0.116 Pacific Area 0.04 0.12 -0.081 92 Cohort 0.75 92 Cohort 0.10 93 Cohort 0.76 93 Cohort 0.09 94 Cohort 0.46 94 Cohort 0.37 96 Cohort 0.66 96 Cohort 0.20 97 Cohort 0.43 97 Cohort 0.41 Lower-secondary HH 0.73 0.75 -0.016 Lower-secondary HH 0.11 0.11 0.003 Upper-secondary HH 0.68 0.75 -0.069 Upper-secondary HH 0.11 0.11 0.001 University - + HH 0.73 0.75 -0.012 University - + HH 0.08 0.11 -0.035 Do not know HH 0.74 0.75 -0.003 Do not know HH 0.09 0.11 -0.021 GDP per capita 0.062 GDP per capita -0.057 Base categories for independent variables: 94 Survey, men, Center area (D.F. and Mexico), 91 cohort, none or primary school education of household head (HH). Units for GDP thousands of 1998 pesos. 43 Table 19. (cont.) Prob[Activity--j I X, dummy'kl Marginal Prob(Activi%=j I X, dummy=k< Marginal Variable k=1 k=0 Effect Variable k=1 k=O Effect Survey 98 Survey 94 Survey 98 Survey 94 E j = Student = Housework 98 Survey 0.06 0.09 -0.031 98 Survey 0.08 0.04 0.034 Women 0.07 0.06 0.003 Women 0.17 0.01 0.160 Age -0.005 Age 0.001 South-East Area 0.06 0.06 0.000 South-East Area 0.06 0.07 -0.009 Center-South Area 0.05 0.07 -0.014 Center-South Area 0.05 0.07 -0.021 North-East Area 0.03 0.07 -0.040 North-East Area 0.03 0.07 -0.040 North-West Area 0.04 0.07 -0.029 North-West Area 0.06 0.07 -0.005 Center-North Area 0.04 0.07 -0.033 Center-North Area 0.05 0.07 -0.025 Pacific Area 0.05 0.07 -0.014 Pacific Area 0.05 0.07 -0.023 92 Cohort 0.07 92 Cohort 0.07 93 Cohort 0.06 93 Cohort 0.07 94 Cohort 0.10 94 Cohort 0.04 96 Cohort 0.07 96 Cohort 0.05 97 Cohort 0.10 97 Cohort 0.03 Lower-secondary HH 0.08 0.06 0.022 Lower-secondary HH 0.06 0.07 -0.011 Upper-secondary HH 0.12 0.06 0.063 Upper-secondary HH 0.07 0.07 0.003 University - + HR 0.13 0.06 0.063 University - + HH 0.05 0.07 -0.020 Do not know HH 0.04 0.06 -0.028 Do not know HH 0.11 0.07 0.041 GDP per capita -0.016 GDP per capita 0.014 Base categories for independent variables: 94 Survey, men, Center area (D.F. and Mexico), 91 cobort, none or primary school education of household head (HH). Units for GDP thousands of 1998 pesos. 44 VIII. Cost-Benefit Analysis Campos (2001) and Carnoy and others (2000) provide a very detailed discussion on the unit costs of CONALEP, the general bachillerato, and the media superior schools. Unit cost data are provided for 1992, 1994, 1995, and 1998. Cost items are divided into two classes: investment in infrastructure and equipment, and operational expenses. Operational expenses include, among other things, salaries of teachers and administrators, security services, and utilities (electricity, telephone, water, etc.). The cost data refer to the three year program. The control group's unit cost per year is $11,512.90, or 7.4 percent higher than CONALEP's unit cost of $10,719.98 (in 1998 pesos). As shown in Section V, the control group's average earnings are lower than CONALEP's average earnings ($26,504.40 vs. $22,684.8, 1998 pesos). It follows that CONALEP's present value is always positive. An alternative scenario was estimated assuming that the control group's unit cost is unknown, that there is a discount rate of 5 percent, and that earnings differences remain constant over the next 30 years. The breakeven year, when the discounted present value of accumulated benefits equals costs, is 12 years in the alternative scenario. If opportunity costs are added, the breakeven year is 18 years. IX. Conclusions The Mexican government introduced CONALEP as an alternative technical education system to the traditional upper-secondary education. CONALEP has undergone significant structural changes in the past decade. A major transformation took place in 1991, when CONALEP reduced the number of careers offered from 146 to 29 careers and introduced modular courses, the forerunner of the competency based education and training model (CBET) now adopted in Mexico. 45 The first part of this paper re-examines CONALEP's performance compared to a well- designed control group. Contrary to previous evaluations, this paper shows that CONALEP graduates search longer for a job but that job congruency is higher compared to the control group. In agreement with previous evaluations, this paper shows that CONALEP increases graduates' earnings. However, the order of magnitude of earnings increase differs greatly from previous studies. This paper finds that on average, CONALEP increases graduates earnings by 22 percent -not the 30 or 40 percent found in other studies- compared to a control group. The second part of this paper evaluates the benefits of the 1991-1992 CONALEP reforms. Results indicate that graduates from the pre-reformed program (94 Survey) search longer for a job compared to those of the post-reformed program (98 Survey). Moreover, graduates from the post-reformed program have 45 percent more probability of finding a job than those from the pre-reformed program. Furthermore, the 94 Survey cohorts earned higher hourly earnings than the 98 Survey cohorts. A plausible explanation is that since 1994, real wages have decreased in Mexico by almost 40 percent. The third part of this paper examines CONALEP's cost-effectiveness. The results indicate that CONALEP is a highly cost-effective program. In addition, as mentioned by other authors, CONALEP has had spillover effects on the rest of the technical education system by stimulating other educational institutions to be more efficient and to adapt to a changing economic and social situation (Carnoy and others 2000). It is difficult to discern the relative contribution of the different factors responsible for the good overall performance of CONALEP, but it is safe to conclude that the special features of CONALEP as a whole have made it possible. These are as follows: autonomous national 46 organizational structure, decentralized operation, strong link to industry, industry-experienced instructors, and modular courses. However, further challenges remain, notably curriculum adjustment to changing market circumstances and improvement of external and internal efficiency. 47 Selected References Ahier, J. (ed.). 1999. Education, Training and the Future of Work. London: Routledge. Carnoy, B., and others. 2000. Aprendiendo a trabajar: Una revisi6n del Colegio Nacional de Educaci6n Profesional Tecnica y del Sistema de Universidades Tecnol6gicas de Mexico. Processed. Campos, M. 2000. Estudio de Costos del CONALEP. Processed. Boud, D., and J. Garrick. 1999. Understanding Learning at Work. London: Routledge. CONALEP (Colegio Nacional de Educaci6n Profesional Tecnica). 1994. Encuesta de Empleo a Egresados del CONALEP, Cohorts 1991, 1992 and 1993. Final Report. ---------. 1999. Encuesta de Empleo a Egresados del CONALEP, Cohorts 1991, 1992, 1993, 1994 and 1995. Final Report. Gill, I., and A. Dar. 1995. "Costs and Effectiveness of Retraining in Hungary." Internal Discussion Paper, Europe and Central Asia Region. The World Bank. Frantz, N. 1998. "Identification of National Trends and Issues for Workplace Preparation and their Implications for Vocational Teacher Education." Journal of Vocational and Technical Education 14(1). Fall 1998. Blacksburg, Va. Heckman, J., and others. 1998. "Matching as an Econometric Evaluation Estimator." Review of Economic Studies 65(2). April. Hobart, B. 1999. "Globalization and its Impact on VET." Review of Research. Adelaide. NCVER. 48 Kye, L. 1998. "An Alternative Technical Education System: A Case Study of Mexico." Staff Working Paper No. 554. The World Bank. Lane, J., and H. Tan. 1996. Evaluacion del Programa DGETI. Processed. Lee, K. W. 1998. "An Alternative Technical Education System: A Case of Study of Mexico." International Journal of Educational Development. Oxford. L6pez-Acevedo, G. 2000. "Teachers' Salaries and Professional Profile." HD Working Paper No.64. The World Bank. Maloney, W., and G. L6pez-Acevedo. 2000. A Comprehensive Development Agenda for Mexico: Note on Labor Markets in Mexico. SEP (Secretaria de Educaci6n Puiblica). 1997. Informe de Labores. ---------. 1998. Informe de Labores. ----. 1999a. Informe de Labores. ---------. 1 999b. Compendio Estadistico por Entidad Federativa. OECD (Organisation for Economic Co-operation and Development). 1997. Reviews of National Policies for Education: Mexico Higher Education. Paris. ---------. 2000. Education at a Glance. Paris. Todd, P. 1999. A Practical Guide to Implementing Matching Estimators. Processed. Power, C. 1999. "Technical and Vocational Education for the Twenty-First Century." Prospects: Quarterly Review of Comparative Education. Vol. XXIX. No. 1. pp. 29-36. Paris. UNESCO. 49 Ravallion, M. 1999. "The Mystery of the Vanishing Benefits: Ms Speedy Analyst's Introduction to Evaluation." Handbook on Evaluating the Poverty Impact of Projects. The World Bank. Sellin, B. 1999. European Trends in the i)evelopment of Occupations and Qualifications. Luxembourg. CDEFOP. Smith, P. 1999. "The Internationalization of Vocational Education and Training." Review of Research. Adelaide. NCVER. World Bank. 1997. "Mexico: Training Assessment Study." White Cover Draft. -----. 1998. "Enhancing Total Factor Productivity Growth." Report No. 17392-ME (Gray Cover). ---------. 1999a. "Export Dynamics and Productivity: Analysis of Mexican Manufacturing in the 1990s." Report No. 19864-ME (Green Cover). ---. 1999b. "Mexican Labor Markets; New Views on Integration and Flexibility." Volume Two: Technical papers. Poverty Reduction and Economic Management Unit. Mexico Department. ---. 2000. "Earnings Inequality after Mexico's Economic and Educational Reforms." Report No. 19945-ME (Gray Cover). December. Mexico Department. 50 ANNEX 1 Table A1.1. Institutions that provide Upper-secondary Education in Mexico General upper-secondary Technical professional education Technological upper-secondary Bachilleres Colleges (CB) College of Professional Technical Centers for Industrial and Education (CONALEP) Services Technological Studies Preparatoria Schools (CETIS)2 State Institutes for Work Training Science and Humanities (ICATIS)' Centers for Industrial and Colleges (CCH) Services Technological State Colleges for Scientific and Bachillerato (CBTIS)2 Incorporated Bachillerato Technological Studies (CECyTE)' Centers for Technical Industrial Centers for Industrial and Services Studies (CETI)4 Technological Studies (CETIS)2 Centers for Scientific and Centers for Industrial and Services Technological Studies (CECyT)5 Technological Bachillerato (CBTIS)2 Centers for Technological Studies (CET)5 Nursing and Obstetrics School (ESEO)3 State Colleges for Scientific and Technological Studies (CECyTE)1 Centers for Ocean Technological Studies (CETMar)6 Centers for Continental Water Studies (CETAC)6 Centers for Farming and Agricultural Technological Bachillerato (CBTA)7 Centers for Forestry Technological Bachillerato (CBTF)7 1. ICATIS and CECyTEs are operated by state Governments. 2. CETIS and CBTIS are coordinated by the General Directorate of Technological Industrial Education (DGETI). 3. ESEO is part of the National Polytechnic Institute (IPN). It is the only modality in which graduates are professional technicians. 4. CETI offers technical programs. 5. CECyT and CET are coordinated by [PN. 6. CETMar and CETAC are coordinated by Department of Scientific Education and Ocean Technology (UECyTM). 7. CBTA and CBTF are coordinated by the General Directorate of Farning and Agricultural Education (DGTA). Source: Infomme de Labores. Several years. SEP. 51 Table A1.2. Hours of Education for Work and Study Hours of Theory and Practice In Hours of Theoretical Institutbon Workshops and/or Companies Study Colegio de Bachilleres 6 in 3'd and 4" Semester (Upper-secondary College) 10 in 5'h and 6'h Semester 27 average CONALEP 17 average in 1st Semester, up to 70% 33 average class hours weekly Centros de Estudios de Bachillerato 14 in *5h and 6th Semester 26 to 29 CBTA 11 average 23 average CBTIS Y CETIS (Upper-secondary) 11 average 23 average CBTIS Y CETIS (Technical) 24 average 10 average CECYT (IPN) 13 27 Colegio de Bachilleres (Upper-secondary College - State of Mexico) 14 17 to 20 CECYT (State of Mexico) 15 average 21 average Centros de Bachillerato Tecnol6gico (Technological Upper-secondary Centers - State of 13 to 14 26 to 27 Mexico) Enferneria (Nurse Training School) 32 12 (Technical UNAM) Preparatory School (UNAM) 30 Colegio de Ciencias y Humanidades (College of 28 Sciences and Humanities) (UNAM) Preparatoria, Universidad Aut6noma del Estado de Mexico (Preparatory from the Autonomous 37 University of the State of Mexico) (UAEM) Preparatorias Oficiales y Anexas a las Normales (Official Preparatories and Attached to the Teaching 36 to 38 Schools) (State of Mexico) Source: COMIPEMS 1998. CONALEP; CBTA; CBTIS; CETIS; CECYT. Table A1.3. Number of Specialties by Institution Institution Number of Type of Studies Specialties CONALEP 29 Technical Professional DGETI (CETIS, CBTIS) 42 Technical Professional 12 Bivalent Upper-secondary CETI. Techno-Industrial Teaching Center (Centro de Ensefnanza 12 Bivalent Upper-secondary Tecnico Industrial) UECYTM (CETMAR and CETAC) 5 Technical Professional DGETA (CBTA, CBTF) 18 Bivalent Upper-secondary CECyTE'S. Scientific and Technological Studies' Center in the States (Centros de Estudios Cientificos y Tecnol6gicos en los 48 Bivalent Upper-secondary Estados) Source: COSNET 1997. 52 Table A1.4. Goodness of fit using ENE 98, ENECE99, and 98 CONALEP Graduates Predicted Cases % Correct Not CONALEP CONALEP Observed Not CONALEP 108,086 1,096 99.0 Cases CONALEP 1,366 3,315 70.8 Independent variables included in the probit model to find the matching group (5 nearest neighbors and probability scores) Cn,trnl Crnmin rONAI.F.P Education Technical complete with lower-secondary 81.4 81.2 Technical incomplete with lower-secondary 0.1 0.0 Technical complete with upper-secondary 12.4 12.6 University incomplete 4.9 4.8 University complete or more 1.2 1.4 Total 100.0 100.0 Sex Men 51.4 52.7 Women 48.6 47.3 Total 100.0 100.0 Age Mean 23.9 23.6 Median 23 23 Std. Deviation 4.7 4.6 Minimum 17 17 Maximum 53 53 Percentiles 20 20 20 40 22 22 60 24 23 80 27 26 State Aguascalientes 4.8 2.4 Baja California 2.4 2.8 Coahuila 6.3 4.6 Chiapas 3.2 2.4 Chihuahua 1.6 1.5 Distrito Federal 5.6 4.7 Guanajuato 5.0 14.2 Guerrero 1.5 1.4 Hidalgo 2.7 3.8 Jalisco 6.8 3.4 M6xico 12.6 15.2 Morelos 2.4 2.4 Nayarit 3.8 5.7 Nuevo Le6n 5.1 3.9 Oaxaca 2.3 2.6 Puebla 4.1 4.5 Queretaro 2.7 1.7 Quintana Roo 1.6 1.4 San Luis Potosi 3.2 3.0 Sinaloa 3.8 5.2 Sonora 5.4 2.8 Tabasco 2.7 2.8 Tamaulipas 6.1 3.8 Veracruz 4.7 3.6 Total 100.0 100.0 53 Table A1.5 Goodness of fit using ENECE99 Predicted Cases % Correct Not CONALEP CONALEP Observed Not CONALEP 41180 772 98.16 Cases CONALEP 574 4107 87.74 Independent variables included in the probit model to find the matching group (3 nearest neighbors and probability scores) Control Group CONALEP Education Technical complete with lower-secondary 78.7 76.5 Technical incomplete with lower-secondary 12.5 14.8 Technical complete with upper-secondary 6.3 6.6 University incomplete 2.4 2.0 University complete or more 100.0 100.0 Total Sex Men 52.6 52.9 Women 47.4 47.1 Total 100.0 100.0 Age Mean 24.4 24.2 Median 24 23 Std. Deviation 4.8 5.0 Minimum 17 17 Maximum 47 53 Percentiles 20 20 21 40 22 22 60 25 24 80 28 27 Place of training Not training 63.0 61.4 Work 25.1 25.2 Other institution 11.9 13.4 100.0 100.0 54 Independent variables included in the probit model to find the matching group (3 nearest neighbors and probability scores) cont. Control Group CONALEP State Aguascalientes 3.7 1.5 Baja California 0.1 Baja California Sur 3.3 1.5 Coahuila 7.1 6.5 Chiapas 2.9 3.2 Chihuahua 1.7 2.7 Distrito Federal 3.5 4.3 Guanajuato 6.0 4.3 Guerrero 1.2 0.6 Hidalgo 2.1 7.9 Jalisco 6.0 12.3 Mexico 11.7 11.8 Morelos 2.9 0.9 Nayarit 3.7 2.9 Nuevo Le6n 5.9 3.7 Oaxaca 2.2 5.4 Puebla 3.9 6.9 Queretaro 2.8 2.4 Quintana Roo 1.9 3.6 San Luis Potosi 3.0 5.9 Sinaloa 4.5 0.5 Sonora 5.0 4.0 Tabasco 2.3 2.7 Tamaulipas 7.8 1.4 Veracruz 4.8 2.9 100.0 100.0 55 Table A1.6. Goodness of fit using ENECE99 Predicted Cases % Correct Not CONALEP CONALEP Observed Not CONALEP 41180 772 98.16 Cases CONALEP 574 4107 87.74 Independent variables included in the probit model to find the matching group (5 nearest neighbours and probability scores) Control Group CONALEP Education Technical complete with lower-secondary 77.9 74.4 Technical complete with upper-secondary 12.5 17.1 University incomplete 7.3 5.7 University complete or more 2.2 2.8 Total 100.0 100.0 Sex Men 53.0 57.5 Women 47.0 42.5 Total 100.0 100.0 Age Mean 24.5 24.2 Median 24 23 Std. Deviation 4.8 5.1 Minimum 17 17 Maximum 46 53 Percentiles 20 20 21 40 22 22 60 25 24 80 29 27 Place of training Not training 64.5 59.1 Work 24.1 25.1 Other institution 11.4 15.8 56 Independent Variables included in the probit model to find the matching group (5 nearest neighbours and probability scores) cont. Control Group CONALEP State Aguascalientes 3.2 Baja Califomia 0.1 Baja Califomia Sur 3.1 1.5 Campeche 0.0 Coahuila 6.6 6.3 Chiapas 2.9 3.9 Chihuahua 2.0 3.1 Distrito Federal 3.4 3.7 Guanajuato 7.0 5.3 Guerrero 0.9 Hidalgo 2.0 11.2 Jalisco 6.1 13.4 M6xico 10.5 11.4 Morelos 3.1 0.6 Nayarit 3.7 2.6 Nuevo Le6n 5.6 2.2 Oaxaca 2.0 7.4 Puebla 4.1 5.2 Queretaro 2.9 0.6 Quintana Roo 2.0 4.4 San Luis Potosi 2.8 7.7 Sinaloa 4.3 Sonora 6.4 4.2 Tabasco 2.9 2.9 Tamaulipas 7.2 1.1 Veracruz 5.1 1.3 Total 100.0 100.0 57 ANNEX 2 Table A2.1. Labor Force by Cohort Matching group: Age 17-65. Five nearest neighbors based on propensity scores. Working Searchingfor a job Cohort Ctrl. Group CONALEP Difference Ctrl. Group CONALEP Difference 90 - 93 95.4 93.6 -1.8 4.6 6.4 1.8 91 - 94 95.0 93.4 -1.7 5.0 6.6 1.7 92 - 95 95.5 91.4 -4.1 4.5 8.6 4.1 93 - 96 94.1 90.4 -3.8 5.9 9.6 3.8 94 -97 94.6 91.6 -3.0 5.4 8.4 3.0 Total' 94.9 92.1 -2.8 5.1 7.9 2.8 ENE 982 97.5 2.5 EIVE 98, LS' 94.5 5.5 ENE 98, US' 95.7 4.3 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). Table A2.2. Labor Force by Cohort Matching group: Age 17-65. Five nearest neighbors based on propensity scores. Working Searchingfor ajob Cohort CtrI Group CONALEP Difference Ctrl Group CONALEP Difference 90 - 93 96.2 94.3 -1.9 3.8 5.7 1.9 91 - 94 93.4 90.9 -2.5 6.6 9.1 2.5 92 - 95 95.9 89.4 -6.5 4.1 10.6 6.5 93 - 96 95.6 87.7 -7.9 4.4 12.3 7.9 94 - 97 96.2 86.1 -10.1 3.8 13.9 10.1 Total' 95.4 89.9 -5.5 4.6 10.1 5.5 EWECE 992 98.1 1.9 ENECE 99, LS' 95.7 4.3 ENECE 99, US' 98.4 1.6 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). 58 Table A2.3. Employment Status. Match ing group: Age 17-65. Five nearest neighbors based on pr ensity. Employer Self employed Employee Cooperative membership Worker without pay Cohort Control CONALEP Diference ColCONALEP DiffGerence Contro CONALEP DiGoerence Control CONALEP Difference CGotro CONALEP Difference 90 - 93 0.9 3.0 2.1 7.8 9.6 1.8 85.3 84.8 -0.4 0.0 0.0 0.0 6.0 2.5 -3.5 91 - 94 2.1 2.8 0.7 7.1 13.0 5.9 86.3 81.0 -5.3 0.0 0.5 0.5 4.6 2.8 -1.8 92 - 95 1.8 3.0 1.2 8.6 8.6 0.0 84.2 85.4 1.2 0.0 1.0 1.0 5.4 2.0 -3.4 93 - 96 2.1 4.3 2.2 7.1 9.6 2.5 87.0 83.2 -3.8 0.0 0.5 0.5 3.8 2.4 -1.4 94 - 97 1.0 1.1 0.1 7.4 7.1 -0.2 86.7 87.4 0.7 0.0 1.1 1.1 4.9 3.3 -1.6 Total' 1.6 2.9 1.3 7.6 9.7 2.1 85.9 84.2 -1.7 0.0 0.6 0.6 4.9 2.6 -2.3 ENE 982 4.3 24.1 60.2 0.04 11.4 ENE 98. LS' 0.2 4.8 77.9 0.02 17.1 ENE 98, US' 2.5 11.0 77.8 0.00 8.7 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). Table A2A. Employment Status. Matching group: Age 17-65. Five nearest neighbors based on prol ensity. Employer Self-employed Employee Cooperative membership Worker without pay Cohort Control CONALEP Difference Control CONALEP Difference Control CONALEP Difference Control CONALEP Difference Control CONALEP Difference Group Group Group Group Group 90 - 93 1.0 4.1 3.2 7.7 15.5 7.8 85.6 79.4 -6.2 0.0 0.0 0.0 5.8 1.0 -4.7 91 - 94 3.3 2.8 -0.5 7.4 8.5 1.1 82.6 84.9 2.3 0.0 0.9 0.9 6.6 2.8 -3.8 92 - 95 0.8 2.8 2.0 6.6 10.3 3.7 86.9 86.0 -0.9 0.0 0.9 0.9 5.7 0.0 -5.7 93 - 96 1.8 6.3 4.5 10.5 5.2 -5.3 82.5 85.4 3.0 0.0 0.0 0.0 5.3 3.1 -2.1 94 - 97 0.0 0.0 0.0 6.3 9.4 3.0 84.8 85.9 1.1 0.0 1.6 1.6 8.9 3.1 -5.7 Total' 1.5 3.4 1.9 7.8 9.8 2.0 84.4 84.3 -0.2 0.0 0.6 0.6 6.3 1.9 -4.4 ENECE 992 4.0 24.4 60.8 0.03 10.7 ENECE 99, LS' 0.2 4.4 79.6 0.00 15.9 ENECE 99, US' 2.9 9.2 81.4 0.02 6.5 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). 59 Table A2.5. Economic Sector. Matching group: Age 17-65. Five nea rest neighbors based on prope sity. Agriculture, fishing, etc. Extraction Manufacturing Construction Electricity, gas, and water Cohort Control CONALEP Difference Control CONALEP Difference Control CONALEP Difference Control CONALEP Difference Control CONALEP Difference Group Group Group Group Group 90-93 2.8 0.5 -2.3 0.0 0.0 0.0 20.7 31.3 10.6 1.8 1.0 -0.8 0.9 1.0 0.1 91 - 94 2.1 2.7 0.7 0.4 0.5 0.0 20.3 30.9 10.6 3.3 2.3 -1.0 0.4 0.5 0.0 92 - 95 2.3 0.5 -1.7 0.0 1.0 1.0 23.0 34.0 11.0 2.7 3.0 0.3 0.0 0.5 0.5 93 - 96 0.8 0.0 -0.8 0.4 0.0 -0.4 25.2 30.4 5.2 3.4 4.9 1.5 0.4 0.5 0.1 94-97 2.0 1.1 -0.9 0.0 0.0 0.0 27.2 32.8 5.6 2.5 3.8 1.3 0.5 0.5 0.1 Total' 2.0 1.0 -1.0 0.2 0.3 0.1 23.2 31.8 8.6 2.8 3.0 0.2 0.4 0.6 0.2 ENE 982 20.3 0.4 18.1 5.51 0.5 ENE 98, LS3 17.0 0.1 31.8 4.57 0.2 ENE 98, US' 6.3 0.5 19.7 2.22 1.3 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). Table A2.5. (cont.) Commerce, Restaurants, and Transportation and Financial Services Personnel, Common, and Hotels Communications Social Services Cohort rol CONALEP Difference Co CONALEP Difference CONALEP Difference Con CONALEP Difference Group Group Group Group 90 - 93 27.2 24.7 -2.4 5.5 5.6 0.0 2.8 5.1 2.3 38.2 30.8 -7.4 91 - 94 28.2 26.4 -1.9 5.4 2.7 -2.7 2.1 1.8 -0.3 37.8 32.3 -5.5 92 - 95 28.8 23.9 -5.0 5.9 3.0 -2.8 3.2 3.6 0.4 34.2 30.5 -3.8 93-96 26.5 21.1 -5.4 4.6 1.5 -3.2 2.1 2.0 -0.1 36.6 39.7 3.2 94 - 97 23.8 23.0 -0.8 5.4 2.2 -3.3 1.0 2.7 1.7 37.6 33.9 -3.7 Total' 27.0 23.9 -3.1 5.4 3.0 -2.4 2.2 3.0 0.8 36.9 33.4 -3.4 ENE 982 22.4 4.4 1.0 27.5 ENE 98, LS3 26.0 3.6 0.2 16.5 ENE 98, US' 31.6 8.1 2.9 27.6 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). 60 Table A2.6. Economic Sector. Matching group: Age 17-65. Five nearest neighbors based on propen ity. Agriculture, fishing, etc. Extraction Manufacturing Construction Electricity, gas, and water Cohort CGntrou CONALEP Difference CGontrou CONALEP Difference CGontrou CONALEP Difference ContrlCONALEP Difference Con CONALEP Difference Group Group Group ~~~~~~~Group Group 90-93 1.9 0.0 -1.9 0.0 0.0 0.0 23.8 31.3 7.4 2.9 1.0 -1.8 0.0 2.1 2.1 91 - 94 3.1 0.9 -2.2 0.0 0.9 0.9 24.8 37.0 12.2 4.7 2.8 -1.9 2.3 0.9 -1.4 92-95 2.4 0.0 -2.4 0.8 0.9 0.1 21.3 26.2 4.9 3.1 1.9 -1.3 0.0 0.0 0.0 93 - 96 3.3 0.0 -3.3 0.0 0.0 0.0 21.5 33.7 12.2 5.0 3.2 -1.8 0.0 1.1 1.1 94 - 97 3.8 1.6 -2.2 0.0 0.0 0.0 25.3 32.8 7.5 1.3 4.7 3.4 0.0 0.0 0.0 Total' 2.9 0.4 -2.4 0.2 0.4 0.2 23.2 32.1 9.0 3.6 2.6 -1.0 0.5 0.9 0.3 ENECE 992 21.0 0.3 18.8 5.53 0.5 ENECE 99, is3 17.2 0.1 30.5 4.71 0.1 ENECE 99, US' 4.1 0.2 18.2 2.29 0.3 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). Table A2.6. (cont.) Commerce, Restaurants, and Transportation and Financial Services Personnel, Common, and Hotels Communications Social Services Cohort Control CONALEP Diference Contr CONALEP Difference Control CONALEP Difference Control CONALEP Difference Group CNLPDern Group COAE PfrneGroup Group 90 - 93 29.5 19.8 -9.7 4.8 4.2 -0.6 1.0 5.2 4.3 36.2 36.5 0.3 91-94 25.6 25.0 -0.6 5.4 5.6 0.1 2.3 1.9 -0.5 31.8 25.0 -6.8 92 - 95 28.3 27.1 -1.2 6.3 2.8 -3.5 2.4 4.7 2.3 35.4 36.4 1.0 93 -96 29.8 24.2 -5.5 3.3 1.1 -2.3 0.8 1.1 0.2 36.4 35.8 -0.6 94- 97 30.4 21.9 -8.5 3.8 3.1 -0.7 1.3 0.0 -1.3 34.2 35.9 1.8 Total' 28.5 23.8 -4.7 4.8 3.4 -1.4 1.6 2.8 1.2 34.8 33.6 -1.1 ENECE 992 21.5 4.5 0.9 27.0 ENECE 99, LS' 22.1 2.1 0.2 22.9 ENECE 99. US' 32.8 8.8 2.9 30.5 1. Sample: Labor force in the matching groip. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). 61 Table A2.7. Training Matching group: Age 17-65. Five nearest neighbors based on propensity. Training Not training Cohort Ctrl. Group CONALEP Difference Ctrl. Group CONALEP Difference 90 - 93 35.8 49.0 13.2 64.2 51.0 -13.2 91 - 94 30.1 39.1 8.9 69.9 60.9 -8.9 92 - 95 42.2 46.4 4.2 57.8 53.6 -4.2 93 - 96 36.1 36.0 -0.1 63.9 64.0 0.1 94 -97 34.1 31.3 -2.8 65.9 68.7 2.8 Total' 35.7 41.1 5.4 64.3 58.9 -5.4 ENECE 992 22.1 77.9 ENECE 99, LS3 12.1 87.9 ENECE 99, US4 39.1 60.9 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). Table A2.8. Relationship with Work Matching group: Age 17-65. Five nearest neighbors based on propensity. YES NO Cohort Ctrl. Group CONALEP Difference Ctrl. Group CONALEP Difference 90 - 93 92.3 93.9 1.6 7.7 6.1 -1.6 91 - 94 87.5 88.4 0.9 12.5 11.6 -0.9 92 - 95 85.5 84.3 -1.1 14.5 15.7 1.1 93-96 88.6 88.9 0.3 11.4 11.1 -0.3 94 - 97 82.1 85.0 2.9 17.9 15.0 -2.9 Total' 87.4 88.4 1.1 12.6 11.6 -1.1 ENECE 992 86.1 13.9 ENECE 99, LS' 74.2 25.8 ENECE 99, US' 77.7 22.3 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). 62 Table A2.9. Reasons for Training. Matching group: Age 17-65. Five nearest neighbors based on prpensity. Prerequisite for entering ajob Update Self-interest He was asked to Other Cohort CGontro CONALEP Difference Grou CONALEP Difference Cont CONALEP Difference Control CONALEP Difference G CONALEP Difference Group GrouP Group Group Group 90-93 10.3 2.0 -8.2 43.6 63.3 19.7 20.5 18.4 -2.1 20.5 8.2 -12.3 5.1 8.2 3.0 91-94 9.8 7.0 -2.8 36.6 58.1 21.6 17.1 18.6 1.5 31.7 9.3 -22.4 4.9 7.0 2.1 92 - 95 9.4 5.9 -3.6 35.8 56.9 21.0 24.5 21.6 -3.0 28.3 5.9 -22.4 1.9 9.8 7.9 93 - 96 8.9 19.4 10.6 35.6 55.6 20.0 20.0 8.3 -11.7 26.7 2.8 -23.9 8.9 13.9 5.0 94-97 7.7 15.0 7.3 34.6 40.0 5.4 30.8 30.0 -0.8 26.9 10.0 -16.9 0.0 5.0 5.0 Total' 9.3 8.5 -0.8 37.3 56.8 19.5 22.1 18.6 -3.5 27.0 7.0 -19.9 4.4 9.0 4.6 ENECE 992 10.7 37.9 22.0 24.6 4.8 ENECE 99, LS 30.7 11.4 27.7 25.4 4.8 ENECE 99, US' 13.9 27.2 27.5 26.9 4.5 1. Sample: Labor force in the matching group. 2. Sample: Labor force. 3. Sample: Labor force with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Labor force with upper-secondary complete and 1-5 years of experience (22-26 years old). 63 Table A2.10. Average earnings, earnings per hour and hours worked by sample group Earnings Earnings per hour Hours worked per week Cohort Gro CONALEP Difference Control CONALEP Difference Gro CONALEP Difference Group Group Group 90 - 93 1924.7 2844.4 919.6 11.0 15.5 4.4 43.9 46.2 2.3 91 - 94 1984.2 2238.2 254.0 12.5 12.8 0.3 43.2 45.5 2.3 92 - 95 1898.5 2310.7 412.3 10.7 15.3 4.6 43.5 46.4 2.8 93 - 96 1992.9 2255.4 262.5 11.1 12.3 1.1 44.1 47.2 3.1 94 - 97 1749.5 1654.7 -94.9 9.4 10.1 0.7 45.0 44.3 -0.7 Total' 1914.1 2262.7 348.6 11.0 13.2 2.2 43.9 46.0 2.0 ENE 982 2046.0 11.6 43.5 ENE 98, LS3 1266.3 6.6 44.8 ENE 98, US' 2088.5 11.2 45.8 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and I - 5 years of experience (22 - 26 years old). Table A2.11. Average earnings, earnings per hour and hours worked by sample group Earnings Earnings per hour Hours worked per week Cohort ro CONALEP Difference CONALEP Difference CONALEP Difference Group -_Group Group 90 - 93 2047.0 3146.0 1099.0 11.4 16.6 5.2 44.0 45.6 1.5 91 - 94 2055.7 2397.8 342.1 11.5 13.8 2.3 44.2 46.8 2.5 92 - 95 2117.1 2872.7 755.6 12.0 16.7 4.7 43.8 45.0 1.2 933-96 1968.0 2181.0 213.0 11.0 11.2 0.2 45.1 47.6 2.5 94 - 97 1697.2 1750.9 53.7 9.5 12.1 2.6 44.8 46.0 1.2 Total' 2000.8 2528.8 527.9 11.2 14.3 3.0 44.4 46.2 1.8 ENECE 992 1984.4 11.2 43.8 ENECE 99, LS' 1253.1 6.6 44.8 ENECE 99, UJS' 2142.2 11.4 46.0 1. Sample: Workers in the matching group. 2. Sample: All workers. 3. Sample: Workers with lower-secondary complete and 3 years of experience (18 and 19 years old). 4. Sample: Workers with upper-secondary complete and I - 5 years of experience (22 - 26 years old). 64 ANNEX 3 Methodology Modelsfor Duration Data or Survival Analysis The variable of interest in this duration analysis is the length of time to fmd a job after graduating from CONALEP, conditional on being unemployed or searching for a job. The functions of interest are the survival and the hazard function: Survival function: P{T 2 t} P{tTt+AjTt} Hazard rate: A(t) = lim P A-+w A where: T is the random variable associated with survival time. Kaplan Meier estimate of the survivalfunction Kaplan Meier is a strictly empirical approach (non-parametric), but it does not consider the influence of covariates. The estimators are given by: Sn1T d, k f n,- X(T ) di n. where: nk is the number of individuals whose observed duration is at least Tk, and dk is the number of observed drop-outs at time Tk. 65 Cox Regression Model or Proportional Hazard Model The Cox model allows exploration of the relationship between the survival experience of an individual and a set of explanatory variables or covariates. In this analysis, the hazard rate is the risk to find a job after being unemployed. The model specifies that the hazard is given by: AQt) = el'x' AO (t) where: AO (t) is the baseline hazard, or the hazard for an individual with X = 0. X, is the vector of explanatory variables for individual i. Xi includes variables such age, gender, schooling, region, GPP. The parameter estimates, a1; are obtained maximizing the partial likelihood. The estimated survival function for the individual i is: s, (t) = [so (t)xP((X') where: S0 (t) is the estimated baseline survival function (for an individual with X = 0). The Cox regression model is semi-parametric because no particular probability distribution is assumed for the survival times, although the model is based on the assumption of proportional hazards. The adequacy of the fitted model was also tested throughout the residuals. Residuals were calculated for each individual in the sample. Their behavior is approximately known when the fitted model is satisfactory. A number of residuals have been proposed, among them the martingale residuals, which take values between - X and 1; they have properties similar to those of linear regression, but they are not symmetrically distributed around zero. Another method that was used to verify the assumption of proportional hazards between the groups of interest was the Kaplan Meier estimate of the survival function for each group. For this estimate, we plotted log (H (t,)) against log( t;) which yielded parallel curves across the different groups, thereby providing evidence that the assumption of proportional hazard was correct. 66 Multinomial Logit Model The multinomial models estimated have a response variable with categorical outcomes 0, 1, 2, ..., J. In this analysis, these variables are status in the labor market and the type of occupation. The model also has K explanatory variables, Xi = [XilI Xi2 ,..., XjK ] such as age, region, schooling, gender, cohort, and GDP. There are K parameters of the model for the outcome j, B) = [fi(j), i,...,1tj . In the multinomial logit model, the set of coefficients B(), B(",...B(J) corresponding to each outcome category is estimated. Assuming that B() =0, where y = O is the category base, the probability that the variable y takes the value "j" is: pi(O) = Pi {y =}= 1 + eX,B(J) j=le(I) X,B( j) p(i) = Piby = j} = Je i - ~~~~~J J=' , j=1,2, J The above-estimated equations provide a set of probabilities for the J+l choices faced by an individual with characteristics X, . The marginal effects of the characteristics on the probabilities are obtained by differentiating (1): ME(i) = kP 6k= - P k ] '9Xk j-0 The marginal effect for a categorical explanatory variable k can be estimated by: n EME (kj) MEk(J) = 1=1 n where: ME() = Pi {y =ix, with Xjk 1}- Pj {y= jIX with Xlk O} 67 Hypotheses about coefficients were tested using a likelihood ratio test, which is based on the statistic: 2 = -2(L - L) which under Ho has the distribution X(do-d.)- where: L. is the log-likelihood associated with the null hypothesis (constrained model). La is the log-likelihood associated with the alternative hypothesis (full model). d is the number of degrees of freedom for the constrained model. da is the number of degrees of freedom for the full model. 68 ANNEX 4 Table A4.1. Cox Regression Model for Time to Find a Job After CONALEP Confidence Interval Variable Hazard Ratio Std. Err. L U 98 Survey 1.45 0.15 1.2 1.8 Women 0.84 0.03 0.8 0.9 Age 0.99 * 0.00 1.0 1.0 South-East Area 1.22 0.08 1.1 1.4 Center-South Area 1.36 0.08 1.2 1.5 North-East Area 1.60 0.09 1.4 1.8 North-West Area 1.44 0.08 1.3 1.6 Center-North Area 1.45 0.08 1.3 1.6 Pacific Area 1.40 0.08 1.3 1.6 92 Cohort 1.00 * 0.11 0.8 1.2 93 Cohort 0.78 0.10 0.6 1.0 94 Cohort 0.61 0.12 0.4 0.9 96 Cohort 1.04 * 0.09 0.9 1.2 97 Cohort 0.90 * 0.18 0.6 1.3 Lower-secondary HH 0.95 * 0.04 0.9 1.0 Upper-secondary HH 0.99 * 0.06 0.9 1.1 University - + HH 1.09 * 0.10 0.9 1.3 Don't know HH 1.00 * 0.12 0.8 1.3 GDP per capita 1.17 0.07 1.0 1.3 Did not work when studying 0.82 0.03 0.8 0.9 No. of subjects 4072 No.. of failures = 3781 Log likelihood= -28403.93 LR chi2(2)= 294.67 Prob > chi2= 0.000 Event (failure): To find a job. Censure: Not to find ajob until survey time. Base categories for covariates: 94 Survey, men, Center area (D.F. and Mexico), 91 cohort, none or.priimary school education of household head (HH). Graduate worked when studying at CONALEP. Units for GDP thousands of 1998 pesos. * Not significative at 5 percent 69 Table A4.2. Probability (Position in Occupation of CONALEP's Graduates). Marginal Effects Estimated. Prob[Posidon=j I X, dummy=kJ Marginal Prob(Position=j I X, dummyrkc Marginal Variable k=1 k'0 Effect Variable k'=1 k-'O Effect Survey 98 Survey 94 Survey 98 Survey 94 j = Employer, self employed = Employee 98 Survey 0.09 0.06 0.021 98 Survey 0.893 0.903 -0.010 Women 0.06 0.09 -0.029 Women 0.920 0.885 0.035 Age 0.002 Age 0.001 South-East Area 0.08 0.08 -0.001 South-East Area 0.89 0.90 -0.004 Center-South Area 0.10 0.08 0.024 Center-South Area 0.87 0.90 -0.029 North-East Area 0.03 0.09 -0.063 North-East Area 0.96 0.88 0.082 North-West Area 0.05 0.09 -0.035 North-West Area 0.94 0.89 0.049 Center-North Area 0.08 0.08 -0.006 Center-North Area 0.91 0.89 0.016 Pacific Area 0.08 0.08 -0.004 Pacific Area 0.90 0.89 0.009 Years after graduation 0.010 Years after graduation -0.013 Not working when graduated 0.07 0.10 -0.037 Not working when graduated 0.91 0.87 0.047 Lower-secondary HH 0.08 0.08 -0.001 Lower-secondary HH 0.90 0.89 0.003 Upper-secondary HH 0.09 0.08 0.005 Upper-secondary HH 0.88 0.90 -0.020 University - + HH 0.11 0.08 0.035 University - + HH 0.87 0.90 -0.022 Don't know HH 0.11 0.08 0.027 Don't know HH 0.88 0.89 -0.015 GDP per capita -0.001 GDP per capita 0.001 Base categories for independent variables: 94 Survey, men, Center area (D.F .and Mexico), working when graduated from CONALEP, none or primary school education of household head (RH). Units for GDP thousands of 1998 pesos. 70 Table A4.2. cont. ProbIPosidon=j I X, dummy=*c Marginal Prob[Position=j I X, dummyv=k Marginal Variable kSv9 k=O Effect Variable k1 k=O Effect Survey 98 Survey 94 Survey 98 Survey 94 j Cooperative's member j = Worker without payment 98 Survey 0.00 0.01 -0.009 98 Survey 0.02 0.02 -0.002 Women 0.01 0.01 -0.001 Women 0.01 0.02 -0.006 Age 0.000 Age -0.003 South-East Area 0.01 0.01 0.003 South-East Area 0.02 0.02 0.002 Center-South Area 0.01 0.01 0.002 Center-South Area 0.02 0.02 0.002 North-East Area 0.00 0.01 -0.007 North-East Area 0.01 0.02 -0.011 North-West Area 0.00 0.01 -0.004 North-West Area 0.01 0.02 -0.011 Center-North Area 0.00 0.01 -0.006 Center-North Area 0.01 0.02 -0.004 Pacific Area 0.01 0.01 0.000 Pacific Area 0.01 0.02 -0.005 Years after graduation 0.000 Years after graduation 0.003 Not working when graduated 0.01 0.01 -0.002 Not working when graduated 0.01 0.02 -0.008 Lower-secondary HH 0.00 0.01 -0.003 Lower-secondary HR 0.02 0.02 0.001 Upper-secondary HH 0.01 0.01 0.003 Upper-secondary HH 0.03 0.02 0.012 University - + HH 0.00 0.01 -0.007 University - + HH 0.01 0.02 -0.006 Don't know HH 0.00 0.01 -0.007 Don't know HH 0.01 0.02 -0.006 GDP per capita -0.001 GDP per capita 0.000 Base categories for independent variables: 94 Survey, men, Center area (D.F. and Mexico), working when graduated from CONALEP, none or primary school education of household head (HH). Units for GDP thousands of 1998 pesos. 71 Table A4.3. Regression Estimated Coefficients. Dependent Variable: Log(Earnings per hour) Confidence Interval Variable Coefficient Std. Err. T P>ItI L U 98 Survey -0.29 0.024 -11.9 0.00 -0.34 -0.24 Women -0.10 0.019 -5.4 0.00 -0.14 -0.07 Age 0.01 0.002 6.3 0.00 0.01 0.02 South-East Area -0.20 0.034 -5.9 0.00 -0.26 -0.13 Center-South Area -0.08 0.031 -2.7 0.01 -0.14 -0.02 North-East Area -0.04 * 0.029 -1.4 0.17 -0.10 0.02 North-West Area -0.03 * 0.030 -0.9 0.36 -0.09 0.03 Center-North Area -0.09 0.028 -3.2 0.00 -0.14 -0.03 Pacific Area -0.05 0.029 -1.7 0.09 -0.10 0.01 Years after graduation 0.07 * 0.043 1.6 0.12 -0.02 0.15 Years after graduationA2 0.00 * 0.007 -0.5 0.62 -0.02 0.01 Lower-secondary HH 0.09 0.020 4.3 0.00 0.05 0.13 Upper-secondaryHH 0.11 0.037 3.0 0.00 0.04 0.19 University - + HH 0.15 0.045 3.5 0.00 0.07 0.24 Don't know HH -0.02 * 0.060 -0.3 0.77 -0.14 0.10 GDP per capita -1.00 0.368 -2.7 0.01 -1.72 -0.28 Did not work when studying -0.10 0.018 -5.5 0.00 -0.13 -0.06 Employee -0.24 0.032 -7.4 0.00 -0.30 -0.18 Cooperative's member 0.04 * 0.111 0.4 0.69 -0.17 0.26 Worker without payment -0.83 0.109 -7.6 0.00 -1.04 -0.61 Agriculture, fishing -0.28 0.068 -4.1 0.00 -0.41 -0.14 Construction -0.09 0.055 -1.7 0.10 -0.20 0.02 Commerce, rest., and hotels -0.22 0.023 -9.6 0.00 -0.26 -0.18 Other activity sectors -0.06 0.020 -2.7 0.01 -0.10 -0.02 Constant 6.19 1.391 4.5 0.00 3.47 8.92 Number of observations 4534 F( 25, 4508) 26.96 Prob > F 0.00 R2 0.13 AdjustedR2 0.12 Base categories for covariates: 94 Survey, men, Center area (D.F. and Mexico), none or primary school education of household head (HH), working when graduated from CONALEP, employer or self employed, manufacturing. Units for GDP thousands of 1998 pesos. * Not significative at 10 percent 72 ANNEX 5 List of Tables Page 1 Enrollment in Upper-secondary by Type of School 13 2 CONALEP Students Compared to Students from Selected Institutions 18 3 Distribution of the 1994 Sample by Cohort 22 4 Actual Sample Selection (original and substitutes by cohort 23 5 Distribution of the 1998 Sample by Cohort 23 6 Labor Force Participation by Cohort 28 7 Labor Force by Cohort 28 8 Employment Status, ENE98 31 9 Employment Status, ENECE99 31 10 Economic Sector, ENE98 32 11 Economic Sector, ENECE99 32 12 Training 36 13 Training Related to Work 36 14 Reasons for Training 37 15 Average Earnings, ENE98 38 16 Average Earnings, ENECE99 38 17 Kaplan Meier Estimates 40 18 Cox Regression Model 41 19 Marginal Effects of Having a Certain Type of Activity 43 AL.1 institutions that provide Upper-secondary Education in Mexico 51 A1.2 Hours of Education for Work and Study 52 A1.3 Number of Specialties by Institution 52 A1.4 Goodness of fit using ENE98, ENECE99, and CONALEP98 53 A1.5 Goodness of fit using ENECE99, 3 nearest neighbors 54 A1.6 Goodness of Fit using ENECE99, 5 nearest neighbors 56 A2.1 Labor Force by Cohort, ENE98 58 A2.2 Labor Force by Cohort, ENECE99 58 A2.3 Employment Status, ENE98 59 73 A2.4 Employment Status, ENECE99 59 A2.5 Economic Sector, ENE98 60 A2.6 Economic Sector, ENECE99 61 A2.7 Training 62 A2.8 Relationship with Work 62 A2.9 Reasons for Training 63 A2.10 Average Earnings, ENE98 64 A2. 11 Average Earnings, ENECE99 64 A4. 1 Cox Regression Model for Time to Find a Job after CONALEP 69 A4.2 Probability, Position in Occupation of CONALEP's Graduates 70 A43 Regression Estimated Coefficients 72 List of Figures Page I Family Income of Students at Selected Institutions 16 2 Percent of Individuals Seeking Jobs 26 3 Employment Status, CONALEP vs. Control 29 4 Received Training 34 5 Training Related to Work 34 6 Earnings Per Month 35 7 Earnings Per Hour 35 74 Policy Research Working Paper Series Contact Title Author Date for paper WPS2717 Bridging the Economic Divide within Raja Shankar November 2001 A. Santos Nations: A Scorecard on the Anwar Shah 31675 Performance of Regional Development Policies in Reducing Regional Income Disparities WPS2718 Liberalizing Basic Carsten Fink November 2001 R. Simms Telecommunications: The Asian Aaditya Mattoo 37156 Experience Randeep Rathindran WPS2719 Is There a Positive Incentive Effect Truman G. Packard November 2001 T. Packard from Privatizing Social Security? 89078 Evidence from Latin America WPS2720 International Migration and the Global Andres Solimano November 2001 A. Bonfield Economic Order: An Overview 31248 WPS2721 Implications for South Asian Countries Sanjay Kathuria November 2001 M. Kasilag of Abolishing the Multifibre Will Martin 39081 Arrangement Anjali Bhardwaj WPS2722 Japan's Official Development Masahiro Kawai November 2001 J. Mendrofa Assistance: Recent Issues and Shinji Takagi 81885 Future Directions WPS2723 Using Development-Orineted Equity Helo Meigas November 2001 S. Torres Investment as a Tool for Restructuring 39012 Transition Banking Sectors WPS2724 Tropical Bubbles: Asset prices in Santiago Herrera November 2001 R. Izquierdo Latin America, 1980-2001 Guillermo Perry 84161 WPS2725 Bank Regulation and Supervision: James R. Barth November 2001 A. Yaptenco What Works Best? Gerard Caprio Jr. 38526 Ross Levine WPS2726 Applying the Decision Rights Florence Eid November 2001 A. Santos Approach to a Case of Hospital 31675 Institutional Design WPS2727 Hospital Governance and Incentive Florence Eid November 2001 A. Santos Design: The Case of Corporatized 31675 Public Hospitals in Lebanon WPS2728 Evaluating Emergency Programs William F. Maloney December 2001 A. Pillay 88046 WPS2729 International Evidence on the Value Luc Laeven December 2001 R. Vo of Product and Geographic Diversity 33722 Policy Research Working Paper Series Contact Title Author Date for paper WPS2730 Antidumping as Safeguard Policy J. Michael Finger December 2001 R. Simms Francis Ng 37156 Sonam Wangchuk