POLICY RESEARCH WORKING PAPER 2404 The Distribution of Public spending on tertiary education in Mexico is Mexico's Public Spending strongly regressive, benefiting on Educati n .mainly the nonpoor in urban on Education areas. To give the poor a chance at higher education, Gladys Lopez-Acevedo student loan programs or Angel Salinas means-tested financial aid and scholarship programs (though rarely devoid of subsidy) are preferable to free education services, because loan and aid programs target the students who suffer from the financial market's failure to provide long-term loans for higher education. The World Bank Latin America and the Caribbean Region Economic Policy Sector Unit and Mexico Country Office July 2000 POLICY RESEARCH WORKING PAPER 2404 Summary findings Research shows that education has played a crucial role education is neutral and higher levels of instruction are in raising levels of earnings and that returns to education moderately regressive. In the central region primary in Mexico have increased, particularly in higher schooling is very progressive, while lower secondary education and in the upper tail of the conditional schooling is almost neutral. Upper secondary and tertiary earnings distribution. instruction strongly benefit the richest income deciles. In Lopez-Acevedo and Salinas examine patterns of public the southern region basic (primary and lower secondary) spending on education in the face of further increases in education is very progressive, upper secondary education earnings inequality. is neutral, and tertiary education is highly regressive. In They analyze the incidence of benefits using two sets Mexico City all levels of education except primary are of data: data on unit costs per student by state and by strongly regressive. education level, and data from surveys on household Lopez-Acevedo and Salinas show that public spending income and spending. Among their findings: at the tertiary level is more regressive than household * Nationally, the poorest income groups get most of spending. So much of public spending on tertiary the national and state subsidy for primary education. At education favors nonpoor families in urban areas that to higher education levels the poor get progressively smaller reallocate the spending so that poor students have a subsidies. chance to participate would require developing credit * For all Mexico, government spending on primary markets for higher education. The government's role education is very progressive. In lower secondary should be to help overcome market failures in the education it is neutral. And in upper secondary education financial sector, which limit the availability of long-term it benefits mainly the middle and upper classes. Tertiary financing for higher education. These failures can be education is strongly regressive, benefiting mainly the corrected through student loan programs or means- richest deciles and mainly in urban areas. tested financial aid and scholarship programs. Such * But those government patterns vary by region. In programs are rarely devoid of subsidy but are preferable the central region average total spending is more to the direct, cost-free provision of services because the uniformly distributed than the national pattern. In the subsidy is targeted more closely to the source of market northern region the subsidy is progressive. Primary failure. This paper-a product of the Economic Policy Sector Unit and Mexico Country Office, Latin America and the Caribbean Region-is part of a strategy to reduce poverty and inequality in Mexico. The study was part of the research project "Earnings Inequality after Mexico's Economic Reforms." Copies of this paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Michael Geller, room I4-142, telephone 202-458-5155, fax 202- 522-2093, email address mgeller@worldbank.org. Policy Research Working Papers are also posted on the Web at www.worldbank.org/research/workingpapers. The authors may be contacted at gacevedo@worldbank.org or asalinas@worldbank.org. July 2000. (14 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 The Distribution of Mexico's Public Spending on Education Gladys Lopez-Acevedo (LCSPE) and Angel Salinas (LCClC)1 Abstract2 Research has shown that education in Mexico has played a crucial role in the process of earnings formation and that returns to education have increased only in the higher levels of education and in the upper tail of the conditional earnings distribution. This paper examines the public educational expenditure patterns in the face of possible further increases in earnings inequality. Several benefit-incidence analysis are carried out bringing together two important and unique sources of information unit cost per student by state and by educational level as well as data from the households income and expenditures surveys. Some of the most interesting results are: i) at national level the poorest income groups receive the bulk of primary education subsidy (federal plus state expenditures), while at higher levels of education they receive progressively smaller subsidies. ii) Government's educational expenditure pattern changes across regions. That is, in the North Region primary education is near equality line and regressive for other levels of instruction. In the Central Region, primary schooling lies above the equality line while lower secondary is very close to it. Upper secondary and tertiary instruction benefit the richest income deciles. In the South Region, basic education is very progressive, upper secondary is at the equality line and tertiary education level lies below the 45-degree line. In Mexico City, the cumulative distribution at all levels of education, except for primary, is far below the 45-degree line. It is also shown that public expenditures at the tertiary level is more regressive than the pattern of household expenditure. A large share of public resources given to this level of education tends to favor non-poor students in urban areas. This paper argues that a strategy to reallocate the education public expenditures from a higher to a lower level of instruction in order to favor the poor groups, would have to involve the development of higher educational credit markets. Meaning that, the government's appropriate, role could be to help overcome market failures in the financial sector, which limit the availability of long- term finance for investments in higher education. These failures can be corrected through student loan programs or means-tested financial aid and scholarship programs. These programs are rarely devoid of subsidy components, but they are preferable to a direct, cost-free provision of services because the subsidy is more closely targeted to the source of market failure. The paper is part of a comprehensive work meant to build a poverty and inequality strategy for Mexico. This research was completed as part of the "Earnings Inequality after Mexico's Economic and Educational Reforms" study at the World Bank. We are grateful to INEGI and SEP (Ministry of Education) for providing us with the data. These are views of the authors, and need not reflect those of the World Bank, its Executive Directors, or countries they represent. 2 This paper was prepared with research support from Monica Tinajero. INTRODUCTION Research has shown that education has played a crucial role in the process of earnings formation and that the returns to education have increased only in the higher levels of education and in the upper tail of the conditional earnings distribution3. In this context, it is essential to analyze the impact of the public educational expenditure on school enrollment, the groups that have been benefited with the public expenditure, and the public expenditure trends. Thus, this paper investigates government's educational expenditure patterns in the face of possible further increases in earnings inequality. In doing so, a benefit-incidence analysis is carried out using unit cost per student by state and by educational level. This paper is structured as follows: Section 1 presents a short description of the data. Section 2 has a brief review of the educational system in Mexico. Section 3 discusses the two elements of the benefit-incidence analysis: enrollment and educational expenditures in Mexico, also examines the distribution of total subsidies allowance for each state, across the levels of education and income deciles. Section 4 compares the education subsidies by levels of schooling for 1994 and 1996. The last section presents the concluding remarks. I DA TA This paper uses the data from the National Household Income and Expenditures Survey (ENIGH) for 1996. The ENIGH is collected by the Instituto Nacional de Estadistica, Geografla e Informdtica (INEGI). This survey is available for 1984, 1989, 1992, 1994 and 19964. Each survey is representative at national level, urban and rural areas. For 1996, the ENIGH is also representative for the states of Mexico, Campeche, Coahuila, Guanajuato, Hidalgo, Jalisco, Oaxaca and Tabasco. The survey design was stratified, multistage and clustered. The final sampling unit was the household and all the members within the household who were interviewed.5 In each stage, the selection probability was proportional to the size of the sampling unit. Then, it is necessary to use the weighs6 in order to get suitable estimators. The available information can be grouped in three categories: • Income and consumption: the survey has monetary, no monetary and financial items. * Individual characteristics: social and demographic, i.e., age, school attendance, level of schooling, position at work, economic sector, etc. * Household characteristics. In addition, data from the Direccion General de Planeacion, Programacion y Presupuesto (DGPPyP, Ministry of Education) regarding educational government expenditures (Federal plus state) assigned to the different levels of schooling for each state is used in order to calculate the unit costs. 2 PUBLIC EDUCATioNAL SYSTEM The structure of Mexico's educational system has the following main characteristics described as follows. First, there is basic education, which is the government's priority. The basic education system consists of: i) early childhood education (or pre-school), which is optional for children 3 3 Lopez-Acevedo, Gladys et al.(1999) 'The sample in a given year is independent from another. ' The sample size for 1996 is as follows: households 14,042 and individuals 64,359. 6 The weights should be calculated according to the survey design and corresponds to the inverse of the probability inclusion. 2 to 5 years old and ii) mandatory primary education where the official entry age is 6 and ideally should be completed in 6 years. In fact, due to late enrollment and grade repetition, however, the target population is 6 to 14 years; iii) mandatory lower secondary school consist of a 3-year cycle, and it is intended for children ages 12 to 16. At this level, the structure is divided in two areas: general and vocational/technical. In parallel, the system also includes the telesecundaria, a distance education program designed to reach remote areas through the transmission of recorded lessons via television network supported by face to face assistance from tutors. The next level, following basic education, is middle level education with options available to students who may choose technical schools and upper secondary education. The duration of these programs is 3 years. A high percentage of the students go for bachillerato also called upper- secondary which allows them to pursue tertiary instruction. On the other hand, a demand for technical studies has been increasing steadily in recent times. Finally, there is tertiary education. This level of education encompasses three lines of study: a system of federal technological institutes, state and autonomous universities, and teacher-training institutes. There is at least one university for each state, and the large universities have campuses in various cities. 3 ENROLLMENTAND PUBLICEXPENDITURES IN THE BENEFITINCIDENCEANALYSIS The benefit-incidence methodology, which is applied in this paper, ranks individuals into groups by income deciles. It then draws information on individual public school enrollment by state and decile to tally up numbers of beneficiaries of each group. These numbers are then multiplied by the government's unit cost of provision allowance for each state and educational level. This provides a profile of distribution for a specific category of educational public expenditures throughout the distribution of income or the "benefit incidence". Thus this technique assumes that the benefit derived from education is equal to the government cost of providing this service. The incidence analysis brings together two sources of information. First, data from income- expenditure surveys (ENIGH) used to construct the deciles. The ENIGH surveys identify the educational level, type of school and total income/expenditure. Second, government expenditures (Federal plus state) on education assigned to the different levels of schooling for each state from the Direccion General de Planeacion, Programacion y Presupuesto, DGPPyP, (Ministry of Education) used for calculating unit costs. Equity issues are then analyzed using the Lorenz Curves based on the pattern of government subsidies to education received by different population groups, highlighting the results of changes in the use of educational services and changes in government's expenditures for education by levels and by state.7 3.1 ENROLLMENT RA TES As shown in table 1, variability of enrollment between poor and non-poor individuals is not substantial at the primary educational level. However, urban areas show slightly larger primary enrollment rates than in rural areas, which might be explained by higher accessibility and affordability to the private system. Enrollment rates for the educational levels beyond primary and probable lower-secondary levels decrease dramatically, particularly for the extremely poor, thus resulting in an increase in the educational gap between poor and non-poor. Tables l.A and 2.A in the Annex show enrollment by educational level and types of schools used in the benefit incidence analysis. 7 For a review see Dominique Van de Walle and Kimberly Nead (1995). 3 Table 1. Total and Public Enrollment Rate by Poverty Status, Location and Level of Education 1996. INEGI/CEPFAL poverty line Urban j Rural Total Poverty Status All Public All Public All Public Primary (6-11 years old) Extreme 93.2 93.2 93.5 93.5 93.3 93.3 Moderate 96.4 96.4 94.6 94.6 96.0 96.0 Non-poor 96.1 95.7 96.4 96.3 96.1 95.7 Total 95.4 95.2 93.9 93.9 94.9 94.7 Lower Secondary (12-14 years old) Extreme 49.1 48.9 29.0 28.8 37.9 37.6 Moderate 68.7 68.8 51.0 51.2 64.8 64.9 Non-poor 81.4 81.3 59.5 59.8 79.1 78.8 Total 68.5 67.7 36.8 36.6 58.4 57.4 Upper Secondary (15-17 years old) Extreme 23.5 21.4 6.9 5.9 14.5 12.9 Moderate 39.6 36.8 22.2 21.7 36.0 33.5 Non-poor 61.7 54.0 24.5 21.8 58.0 50.1 Total 45.7 39.8 12.8 11.7 36.4 31.2 University (18-24 years old) Extreme 3.4 2.9 0.4 0.4 1.8 1.6 Moderate 7.4 7.0 2.3 2.2 6.4 5.9 Non-poor 24.0 17.6 5.9 5.4 22.0 16.1 Total 15.3 11.5 2.0 1.8 12.0 8.9 Source: Own calculations based on ENIGH, 1996 Given that coverage at primary level and the first years of lower secondary is already sizable and decreasing due to demographic factors which cause the population in this group to stagnate and start to shrink at the beginning of the next century.8 This in turn frees some resources so that coverage may be increased at the upper-secondary level. 3. 2 PUBLIC EDUCATIONAL EXPENDITURES Total public educational spending per student in Mexico increased steadily up to 1994 and peaked in 1998, even though the total student population increased from 26 million in 1994 to 28 and a half million in 1998. By 1998, total spending in education increased by 5.2 percent of GDP, less than a full percentage point above the 4.9% of GDP reached in 1995. The federal government currently accounts for close to 80% of total sector spending. 8From 1973-1994, there was a change in the population structure: the population ages between one year through 14 dropped 36%, between those 15 and 64 increased 59.8% and the age group over 65 rose 4.2%. 4 Figure I Figure 2 Education Spending per Student Distribution of Education Ekpenditure by 5000 Source 4000- 3000 -0 O\ | OP | Source: IV Informe de Gobierno, 1998 Source: IV Informe de Gobierno, 1998 A desegregation of public expenditures in education by instruction level for 1994 and 1996 is shown below. Public expenditures in primary and lower secondary absorb a large proportion (59% in 1996) of federal budgetary resources for formal education services. Yet, public expenditures in upper secondary and tertiary level were 13.7% and 27.3% each respectively. Another observation about the evolution of educational public spending evolution is that it has become more egalitarian in per-capita terms across different schooling categories.9. In the early 1980s, the amount of federal spending per university student was 10 times the amount spent per primary student. This ratio fell to arounld 7 times in the early 1 990s. Federal spending on the other levels relative to thie primary level indicates a similar decline, even though the absolute amounts increased at all levels. In 1996, upper-secondary received 1.5 as much as each primary school student and each university student received five times as much as a primary student, compared to 2.1 and 6.8 in 1994, respectively (tables 2 and 3). Table 2. Feea_n tt xedtrso ulcEuain 94(huad fcretpss P rimary Lower Secondary Upper Secondary Tertiary Federal Expenditure 17,947,229 8,603,383 6,610,913 13,141,420 State Expenditure N/A N/A N/A N/A Total Expenditure 17,947,229 8,603,383 6,610,913 13,141,420 Enrollment 13,593,797 4,661,522 2,386,758 1,461,189 60% State 40edtr ,2,2947747186,1 ,1,6 Subsidy per Student (pesos) 13206 1,864 2,7708 86,9947 1500~~~~~~~~~~~ Prmr Suet qialne1.014 2.0 6.8 Souces EN9GH 940 and1 D992 (931994 9), SEP19 99 7 Na,G ~C TAbl 3.Fdeseralgatind StatbieExpenditures on PulidEucation,b 1996ucio (Thousr94andso1curen pesos shownbelo. Pubic e pedtrsiprimary Lower werSecondary *Uperseonda Tareprortia on Federa Expnditure feea bdetr esuce o 33,328,32313,394,89810,884,85 21t,651,986 SaeExpendituresiupeseodrantetaylvlwe 8,90,29 474,40n1,69710 2,210,962tiely Tnotalexpobendtreaio42,248,572tonofedcaioa 18,142304s12754,56 23,862,948hatitha benrolment r gltaini e-cpt em 13,802,39i4,972,11 2,767,993 1,459,820 Intherl Sub0sid th muto eeapnigper Snvrsttudent (pesos)ti3,061h3,649u4,608e16,347 primary student.e hiqrtiivlltoarunenties 1.00e1.19 19 .51 Feealseni 34nte te TabeI 2 Fore derGobieno, 1998. edtue nPbi Euain 94 Tosnso cretpss Pimay oerSeon;y pprSeonaryTetir Source: ENIGH 96 and DGPPyP (1999), SEP 3.3 BENEFIT INCIDENCE ANALYSIS Next, a comparison was made between the cumulative distribution of the various education sub- sectors and the distribution of per capita annual total and federal public educational expenditures. Beforesaid, in order to derive the cumulative distribution for various educational levels, individual public school enrollment by state and decile is multiplied by the government's unit cost of provision allowance for each state. This is also done subsequently by region and state. Figures 3 and 4 show the cumulative distribution by total andfederal educational expenditures for all of Mexico. One of the main messages is that the poorest income/expenditures deciles receive the bulk of the primary education subsidy. This same group, at higher levels of education receives progressively smaller subsidies. This indicates that primary education is very progressive and lower-secondary education is basically neutral. Upper-secondary schooling, benefits the middle and upper classes. Finally, the tertiary level is strongly regressive in that it mainly benefits the richest deciles. At national level, public expenditures seem quite equal, as shown by the fact that the expenditure line lies very close to the 45 degree diagonal. Figure 3 Figure 4 Cumulative Distribution of Total Education Cumulative Distribution of Federal Education Expenditures, 1996 National Fxpenditures, 1996 National 100 o 90 9 80 -8 70 -7 6-0 -. 65 0~~~~~~~~~~~~~~~ 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 --Primary - Lower Sec ---- -Upper Sec -Primary Lower Sec -----Upper Sec Tertiary Total expend --Eq u ality - .Tertiary Fed expend ---- Equality Source: ENIGH 96 and DGPPyP, SEP Source: ENIGH 96 and DGPPyP, SEP When desegregated by region, see figures 5 and 6, it becomes evident that the educational inequality in the Central Region of Mexico leads the national pattern. Still, in the Central Region, the curve for total and federal schooling expenditures lies above the equality line. This implies that on average total schooling expenditures for that region are more uniformly distributed than the national pattern. 6 Figure 5 Figure 6 Cumulative Distribution of Total Education Cumulative Distribution ofFederal Education Expenditures, 1996 Center Region Expenditures, 1996 CenterRegion I00.0 10.0loow- 900 090.0 80.0 8 70.0 / 70.0 . 600 60 600 'o 400 o 40.0 0.0 30.0 ,._ 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 to --Primary . Lower Sec - Upper Sec -_-_ Primary ....... Lower Sec -.- - Upper Sec - - Tertiary Totalexpend ---- Equality |_ - Tertiary - Fedexpenditure --.--Equality Source: ENIGH 96 and DGPPyP, SEP Source: ENIGH 96 and DGPPYP, SEP The distribution of the average subsidy in the South Region and Tabasco State lies above the average distribution for the North Region (figures 7 through 10). One plausible explanation is the higher concentration of the enrollment in the lower deciles (mainly in primary) in the South Region and Tabasco compared to the concentration in the North, where the students are in the medium and top deciles. In the South, public enrollment is highly progressive particularly for primary school, as shown by the fact that public school enrollment is above and far from the 45- degree diagonal. It should also be mentioned that public education spending in upper-secondary in Tabasco is basically neutral at high level of income, while progressive at the bottom of the distribution. Figure 7 Figure 8 Cu mulative Distribution of Total Education Cumnulative Distribution ofFederal Education Expenditures, 1996 South Region Expenditures, 1996 South region 100.0 - 100.0 90.0 - 90.0 ' so.o , - . / 80.0 - - / 70.0 - V 70.0 - - 60.0 ." 60.0 _ ' 'S 40.0 2 3 4 40.0 9 .~30.0 * ~30.0 d 20.0 - ~20.0 7 to00 - 10.0 0.0 , , , . ,0.0 1 2 3 4 5 6 7 8 9 10 2 3 4 5 6 7 8 9 10 |-Primary . Lower Sec -.-.- Upper Sec | - --- Primary Lower Sec - - Upper Sec -|. Tertiary Total expend --+--Equality -----Tertiary - Fedexpenditure e Equality Source: ENIGH 96 and DGPPyP, SEP Source: ENIGH 96 and DGPPyP, SEP 7 Figure 9 Figure 10 Cumulative Distribution of Total Education Cumulative Distribution of Federal Education Expenditures, 1996 Tabasco State Expenditures, 1996 Tabasco State 100.0 - 100. 90.0 90.0 80 0 _~-" 80.0 / 8 70.0 70.0 ; t560.0 0o { > ,/ 8 60.0 - 0 20.0 '0.0 , 0.0 -# 40.0 ._ _. . . . . . 1 2 3 4 5 6 7 9 9 10 1 2 3 4 5 6 7 8 9 10 Primary Lower Sec - --U-pper Sec _ -_primar .....y Lower Sec ---- -Upper Sec - Tertiary - Total expend -* Equality -----Tertiary - Fed expendilure + Equality Source: ENIGH 96 and DGPPyP, SEP Source: ENIGH 96 and DGPPyP, SEP In the North Region (figures 11 and 12), the cumulative distribution of educational subsidy lies below the 45 degree diagonal, except for primary schooling, which is near the equality line. In general, this can be explained by both larger populations in the medium and top deciles and higher enrollment rates in higher levels, which probably reflect higher incomes in the North Region and easier access to schools. Figure 11 Figure 12 Cumulative Distribution ofTbtal Education Cumulative Distribution ofFederal Education Expendcitures, 1996 North Region Expenditures, 1996 North Region 100.0 100.0 90.0 90.0 .. 80.0 * 80.0 -- i '~70.0 * I *70.0 - *S600 * 9 60.0 / , 50.0 500 / X 43oo 8~~~~~' ;* 0 3o.50 40.0 0*' 40.0 / 30.0 3 - 30.0 20.0 2 , , 20.0 -- - --' 10.0 -*~010.0 - - 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 ] Primary . LouerSec - - UpperSec --Primary - Lower Sec UpperSec -- --- Tertiary - Total expend _ Equality - - Tertiary - Fed expenditure ----+ Equality Source: ENIGH 96 and DGPPyP, SEP Source: ENIGH 96 and DGPPyP, SEP The distribution of per capita public expenditures in Mexico City (figures 13 and 14) is far below the 45 degree diagonal indicating that it is very regressive. Public expenditures in primary level are progressive for the high-income deciles, in that the primary curve lies above the 45 degree axis and it is much more progressive than the distribution of per capita expenditures, reflecting the fact that fewer higher income children attend public primary schools. Spending at 8 the lower and upper secondary level is more progressive than the public expenditures, although the curves still lie below the 45 degree diagonal. Only university instruction is more regressive than the average distribution of total expenditures. Interestingly, public expenditures in education in Nuevo Leon (see figures 15 and 16) are far below the 45-degree diagonal following a pattem similar to Mexico City. Figure 13 Figure 14 Cumulative Distribution oflbtal Education Cumulative Distribution of Federal Education Expenditures, 1996 Distrito Federal Expenditures, 1996 Distrito Federal 100.0 100.0 90.0 90.0 80.0 80.0 es 70.0 9' 70.0 600 r00 .o- 50.0 o40.0 -4. 30.0 , .' 030.0 ~- 20.0 - 2-~*-0.0 - - 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 |____Prinnary ---- LowerSec -.-- UpperSec |----Priary -- LowerSec -----UpperSec - - - --.Tertiary - Total expend e-4-Equality -----Tertiary - Fed expenditure ---- Equality Source: ENIGH 96 and DGPPyP, SEP Source: ENIGH 96 and DGPPyP, SEP Figure 15 Figure 16 Cumulative Distribution of Total Education Cumulative Distribution of Federal Education Expenditures, 1996 Nuevo Leon State Expenditures, 1996 Nuevo Leon State 100.0 100.0 90.0 90.0 80.0 , : 80.0 :1 E 70.0 ' 700 60000 WU / X 50.0 i 50.0 - o 40.0 / / -. , 640.0 V Y 30.0 -/ ',.; 30.0 .-, ; 20.0 , _ 20.0 10.0t 10.0 , 0.0 - 0.0 ' 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Primaly ------ Lower Sec --- Upper Sec ---- Primnary - Lower Sec - - -Upper Sec _ Tertiary - Total expend ----- Equality -----Tertiary Fed expenditure ----Equality Source: ENIGH 96 and DGPPyP, SEP Source: ENIGH 96 and DGPPyP, SEP The evidence presented suggests that public subsidies for education, particularly at the tertiary level, are regressive. A large share of public resources is given to the high-income level students. A strategy to reallocate public expenditures from tertiary to secondary level in order to favor the poor would involve a comprehensive agenda that would meet the challenges posed in upper- secondary level such as financing and quality of education. 9 4 COMPARISON OF THE FEDERAL SUBSIDYDISTRIBUTION FOR 1994 AND 1996 This section compares the year 1994 and 1996 to assess a change over time in the targeting of education spending. Figures 17 through 20 show the subsidy received by students in each income decile for all levels of education in 1994 and 1996. As indicated, the subsidy in primary level increased from 1994 through 1996. Such increment was slightly higher for the bottom income decile as compared to the top deciles. In contrast, subsidies decreased from 1994 to 1996 for the lower secondary level. Such reduction had a higher impact on deciles 6 through 9. For upper secondary educational level, per capita subsidies decreased on average by 200,000 thousand pesos for students in deciles 7 through 9. Finally, the tertiary level also experimented a reduction of approximately 1,000,000 thousand pesos in the ninth decile. In both years, the pattern is progressive for the primary level, as it was found in the previous section, meaning that the subsidy is higher for the poor. On the other hand, the subsidy for upper secondary and tertiary levels is still regressive, benefiting mainly the non-poor, although, the distribution of subsidy has become more egalitarian in 1996 compared to 1994. For lower secondary educational level, the middle income groups receive most of the subsidy. Figure 17 Figure 18 Upper Secondary Level Subsidy Tertiary Level Subsidy 2,000,000 1600,000~~~~~~~~~~~~~~~~600,0 1600,000 * 1 994 03199 6 Deci l *194096Deci l S,ource: ENIGH 96 anld DGPPyP, SEP Source: ENIGH 96 and DGPPyP, SEP Figure 19 Figure 20 Primary level Subsidy N Lwer Secondary level Subsidy 6,000,000 2,000,000 5 0,0,000 A600,0002 |n l .l 1 800,000 a~~~~~~~~~~50,0 a a 2000,00 400,000~~~~~~~~~~~~~~~~040,0 i'000,000 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~' 1 2 3 4 5 6 7 8 9 10 1 2 3 4 S 6 7 8 9 10 * 1994 01996 Decil _ C1994 M 1996 Decil Source: ENIGH 96 and DGPPyP, SEP Source: ENIGH 96 and DGPPyP, SEP 10 5. CONCLUSIONS Enrollment rates for the educational levels beyond primary and probable lower-secondary levels decrease dramatically, particularly for the extremely poor, thus resulting in an increase in the educational gap between poor and non-poor. Given that coverage at primary level and the first years of lower secondary is already sizable and that demographic pressure is decreasing, the population of this group is virtually stagnated and will start to shrink at the beginning of the next century. This in turn frees some public resources, which can eventually be used to increase coverage at the upper-secondary level. Government spending per student steadily increases until 1994 and stays the same until 1995, peaking again in 1998. On the other hand, after 1994, government spending per student becomes better distributed. Nevertheless, government spending still favors tertiary education. Spending on education continues to be concentrated in the federal sector, which accounts for over 80 percent of total sector spending. Another noteworthy observation about the evolution of public spending on education in Mexico is that it has become a little bit more egalitarian in per-capita terms across different schooling categories. By moving towards a more evenly distribution of per capita spending across different levels, equity seems to have improved. At the same time, the external environment changed in a manner that raised the relative return to higher education, thereby tending to make more efficient what had initially been an inefficient allocation of resources. With respect to the public educational expenditures by income strata, the results indicate that at national level the poorest income groups receive the bulk of primary education subsidy (federal plus state expenditures). This same group, at higher levels of education receives progressively smaller subsidies and the pattern changes across regions. In the North Region, primary education is near equality line and regressive for other levels of instruction. In the Central Region, primary schooling lies above the equality line while lower secondary is very close to it. Upper secondary and tertiary instruction benefit the richest income deciles. In the South Region, basic education is very progressive, upper secondary is at the equality line and tertiary education level lies below the 45 degree line. In Mexico City, the cumulative distribution at all levels of education, except primary, is far below the 45-degree diagonal. At national level, public subsidy for primary education slightly increases from 1994 through 1996. By contrast, subsidies for all other levels of education decreased. The pattern of primary subsidy for both years is progressive meaning that the subsidy is higher for the poor. On the other hand, subsidies for tertiary education are regressive, benefiting primarily the non-poor. Overall, the distribution is slightly more egalitarian in 1996 than in 1994. Federal educational expenditures on upper secondary level tend to be regressive. In 1994, the pattern was more regressive than in 1996. For lower secondary, the middle income groups receive the bulk of the subsidy. The public educational system can improve its targeting to the poor by increasing its focus on the secondary (lower and upper) levels versus university levels. Public expenditure at the tertiary level is more regressive than the pattern of household expenditure. A large share of public resources given to this level of education tends to favor non- poor students in urban areas. A strategy to reallocate the educational public expenditures from a higher to a lower level of instruction in order to favor the poor groups, would have to involve the development of higher educational credit markets. Meaning that, the government's appropriate role could be to help overcome market failures in the financial sector, which limit the availability of long-term finance for investments in higher education. These failures can be corrected through student loan programs, or means-tested financial aid and scholarship programs. These programs are rarely devoid of subsidy components, but they are preferable to a direct, cost-free provision of services because the subsidy is more closely targeted to the source of market failure. 11 ANNEX Table L.A Enrollments by type of school, 1994 Education level Public Private Others Total Primary 13,593,797 895,913 40,689 14,530,399 Lower Secondary 4,661,522 388,806 12,004 5,062,332 Upper Secondary 2,386,758 778,587 49,385 3,214,730 Tertiary 1,461,189 530,754 1,503 1,993,446 Source: ENIGH 94 Table 2.A Enrollments by type of school, 1996 Public Private Others Total Primary 13,802,395 768,748 1,746 14,572,889 Lower Secondary 4,972,116 326,229 4,153 5,302,498 Upper Secondary 2,767,993 875,129 15,782 3,658,904 Tertiary 1,459,820 580,962 7,680 2,048,462 Source: ENIGH 96 12 REFERENCES Bouillon, Cesar, Arianna Legovini and Nora Lustig (1998), "Rising Inequality in Mexico: Returns to Household Characteristics and the Chiapas Effect". Mimeo. Bourguignon, Francois, Martin Fournier, and Marc Gurgand (1998), "Labor Incomes and Labor Supply in the Course of Taiwan's Development, 1979-1994". Mimeo. Consejo Nacional de Poblaci6n. Proyecciones de la Poblaci6n de Mexico 1996-2050. Mexico: Diciembre 1998. De la Torre, Rodolfo (1997), "Desigualdad, Pobreza y Polarizaci6n Social en Mexico". Mimeo. Diario Oficial de la Federaci6n. Ley Federal de Educaci6n. Mexico: Secretaria de Gobernaci6n 1993. Fiszbein (1991), "A Essays on labor markets and income inequality on less developed countries". Berkeley, Ph.D. Dissertation. University of Califomia. "Facing Up to Inequality in Latin America Economic and Social Progress in Latin America". Inter-American Development Bank. 1998-1999 Report Londofio, Juan Luis (1996), "Poverty, Inequality and Human Capital Development in Latin America, 1950-2025". World Bank Latin American and Caribbean Studies, (June). Lopez-Acevedo, Gladys and Angel Salinas (1999) "The Evolution and Structure of the Rates of Returns to Education in Mexico (1987-1997): An Application of Quantile Regression", The World Bank, mimeo. Lopez-Acevedo, Gladys et al. (1999) Earnings Inequality and Education Attainment After Mexico's Economic Reforms. The World Bank, Report No. 19945-ME. Organizaci6n para la Cooperaci6n y Desarrollo Econ6micos. Examenes de las Politicas Nacionales de Educaci6n: Mexico. Educaci6n Superior. Francia: OCDE, 1997. Presidencia de la Repuiblica. Cuarto Informe de Gobiemo, 1° de Septiembre, 1998. Mexico: Septiembre 1998. Psacharopoulos, G., Morley, S., Fiszbein, A., Lee, H., Wood, B. (1992), "Poverty and income distribution in Latin America: the story ot the 1980s". The World Bank. Psacharopoulos, George (1993), "Returns to Investment in Education: A Global Update". World Bank Policy Research Paper No. 1067, (January). SEP. Subsecretaria de Educaci6n e Investigaci6n Tecnol6gicas. Consejo del Sistema Nacional de Educaci6n Tecnol6gica (COSNET) Evaluaci6n de la Educaci6n Tecnol6gica, Vol. I, Evaluaci6n del Ingreso al Nivel Medio Superior de la Educaci6n Tecnol6gica. Tomo III, Fundamentaci6n Te6rica del Examen. Mexico: Consejo del Sistema Nacional de Educaci6n Tecnol6gica (COSNET), Octubre 1994. SEP. Subsecretaria de Educaci6n e Investigaci6n Tecnol6gicas. Consejo del Sistema Nacional de Educacion Tecnol6gica (COSNET) Evaluaci6n de la Educaci6n Tecnol6gica, Vol. I, Evaluaci6n del Ingreso al Nivel Medio Superior de la Educaci6n Tecnol6gica. Tomo V, Informe de Resultados, 1992-1993 y 1993-1994. Mexico: Consejo del Sistema Nacional de Educaci6n Tecnol6gica (COSNET), Junio 1994. 13 SEP. Subsecretaria de Educaci6n e Investigaci6n Tecnol6gicas. Consejo del Sistema Nacional de Educaci6n Tecnol6gica (COSNET) Evaluaci6n de la Educaci6n Tecnol6gica, Vol. II, Evaluaci6n Institucional del Nivel Medio Superior de la Educaci6n Tecnol6gica. Tomo IV, Informe de Resultados 1992-1993. Mexico: Consejo del Sistema Nacional de Educaci6n Tecnol6gica (COSNET), Octubre 1994. Secretaria de Educaci6n Puiblica. Programa de Desarrollo Educativo 1995-2000. M6xico: 1995. SEP/ANUIES. Coordinaci6n Nacional para la Planeaci6n de la Educaci6n Superior. Datos Basicos de la Educaci6n Universitaria, Mexico, 1996. Mexico: Direcci6n General de Educaci6n Superior, 1997. SEP. Direcci6n General de Planeaci6n, Programaci6n y Presupuesto. Compendio Estadistico por Entidad Federativa 1997. Mexico: Noviembre 1997. SEP. Subsecretaria de Educaci6n e Investigaci6n Tecnol6gicas. Estadistica Basica. Sistema Nacional de Educaci6n Tecnol6gica, 1997-1998, Inicio de Cursos. Mexico: DGETI, 1998. Secretaria de Educaci6n Puiblica. Informe de Labores 1997-1998. Mexico: Octubre 1998. SEP. Direcci6n General de Planeaci6n, Programaci6n y Presupuesto. Estadistica Basica de Educaci6n Superior. Inicio de cursos 1997-1998. Mexico: Enero 1999. Secretaria de Hacienda y Cr6dito Piiblico. Plan Nacional de Desarrollo, Informe de Ejecuci6n 1998. Mexico: Marzo 1999. Secretaria de Educaci6n Puiblica. Programa del Sector Educativo para 1999. Mexico: 1999. Van de Walle, Dominique and Kimberly Nead (1995), "Public Spending and the Poor: Theory and Evidence". The World Bank. World Development Report (1996). The World Bank. Oxford University Press. World Bank (1999). Government Programs and Poverty in Mexico, Green Cover. World Bank (1998). Country Economic Memorandumn, "Factor Productivity and Growth". 14 Policy Research Working Paper Series Contact Title Author Date for paper WPS2391 Are the Poor Protected from Budget Martin Ravallion July 2000 P. Sader Cuts? Theory and Evidence for 33902 Argentina WPS2392 What Factors Appear to Drive Private Dipak Dasgupta July 2000 S. Crow Capital Flows to Developing Countries? Dilip Ratha 30763 And How Does Official Lending Respond? WPS2393 Will the Euro Trigger More Monetary Patrick Honohan July 2000 A. Yaptenco Unions in Africa? Philip R. Lane 31823 WPS2394 Tax Evasion, Corruption, and the Waly Wane July 2000 H. Sladovich Remuneration of Heterogeneous 37658 Inspectors WPS2395 Decentralizing the Provision of Health William Jack July 2000 H. Sladovich Services: An Incomplete Contracts 37698 Approach WPS2396 Aid Dependence and the Quality of Stephen Knack July 2000 P. Sintim-Aboagye Governance: A Cross-Country 38526 Empirical Analysis WPS2397 Verifying Exchange Rate Regimes Jeffrey Frankel July 2000 E. Khine Eduardo Fajnzylber 37471 Sergio Schmukler Luis Serven WPS2398 Determinants of Current Account Cesar Calder6n July 2000 H. Vargas Deficits in Developing Countries Alberto Chong 38546 Norman Loayza WPS2399 Managers, Investors, and Crises: Graciela Kaminsky July 2000 E. Khine Mutual Fund Strategies in Emerging Richard Lyons 37471 Markets Sergio Schmukler WPS2400 Child Care and Women's Labor Force Monica Fong July 2000 P. Sader Participation in Romania Michael Lokshin 33902 WPS2401 Telecom Traffic and Investment in Scott J. Wallsten July 2000 P. Sintim-Aboagye Developing Countries: The Effects 38526 Of International Settlement Rate Reductions WPS2402 Debt Management in Brazil: Afonso S. Bevilaqua July 2000 S. Bery Evaluation of the Real Plan and Marcio G. P. Garcia 85178 Challenges Ahead Policy Research Working Paper Series Contact Title Author Date for paper WPS2403 Can the World Cut Poverty in Half? Paul Collier July 2000 E. Khine How Policy Reform and Effective David Dollar 37471 Aid Can Meet International Development Goals I,