Lecture 10: Education(3): Educated for what? David Donaldson and Esther Duflo 14.73 Challenges of World Poverty
Introduction The Millennium Development Goals call for universal primary education by 2015, and equality between boys and girls at all levels. Substantial efforts have been devoted by the international community and by developing country governments to get children to enroll in school, using a combination of the methods we saw in the previous lectures. Rapid increase in enrollment: Between 1999 and 2006, countries in Sub-Saharan Africa increased their average net enrolment rates from 54 percent to 70 percent (e.g., Benin: 50% to over 80%). In South and West Asia, it rose from 75 to 88 percent. Is all this effort by international community, developing country government, parents, children, worthwhile?
The benefits of education: Individual benefits From Anne Case reading (The primacy of Education) What are the potential benefits to an individual of being educated? The returns to education: increase in wages or income due to one extra year of education. Ability to decode information (Schultz): understand better new technologies, new health ideas (germ theory), gain control over one s life Impact on Health, Fertility, life expectancy Could these benefits be affected by others things that happen in the country? Returns to education may be lower when there is nothing to do, and increase when new technologies appear in the country. Example in the reading:green revolution, India.
The benefits of education: Social benefits What are the potential benefits to society when someone is educated? Educated people may train others Educated people may adopt new technologies, and others may then imitate them. Are there potential cost to me if others around me are educated? Returns to education may fall if there are many educated people around: educated unemployed. Wages of the uneducated may fall if the educated take their jobs.
The benefits of education: Evidence Countries where people are more educated are much richer Graph Individuals who are more educated are also richer than those who are not Graph But the magnitude do not square very well together: From picture: wages about 10% higher for each extra year of education Countries in top decile of education distribution have about 8 more years of education than those in the bottom. They should have GDP 80% higher if private returns were the only part of the story. In fact they are about 15 times richer. Can the cross-country comparison be misleading? Easterly s arguments: Countries that are more educated may be different. People may chose to get educated when they see good prospects
Are there really any benefits to education? Easterly Critique: If education was really beneficial we should see that when years of education increase, GDP increase: we should see a correlation between change in education and change in GDP However: there is no such correlation. While countries that have more years of education are richer, it is not the case that countries where education has grown faster in the last few decades have become richer than countries where it has not grown as fast. Graph Easterly concludes: Education is another magic formula that failed us in our quest for growth.
Problems with this conclusion 1. Education is not very well measured at the country level: hard to measure enrollment rate, to account attendance, huge difference in quality from country to country. There is hardly one thing called education that we can compare from country to country. 2. In particular in the countries where education has grown fastest after 1960, the quality of education may also have decreased (as a result of the attempt to make it grow so fast): this tells us that rapid educational expansion is difficult, not that it is useless! 3. Some of the countries were education grew very fast also have had real problem (civil wars etc.). E.g. Angola, fought Portuguese colonialists from 1961 to 1975, and once they left, fought internally until 2002. Mozambique, fought the Portuguese from 1965 to 1975 and wound up in a civil war between 1981 and 1992. Senegal, Sudan.
What can we conclude from this discussion? With only cross-country data, it would be very difficult to settle the debate. The data is not very good, there are not that many countries, each country s experience is different, etc... It is more promising to look at the benefits of education at the individual level.
Estimating the returns to education We have seen that individuals who have more years of education earn higher wages. Graph However, it could not be the causal effect of education. Think of the counterfactual: are there reasons why the counterfactual wages of someone who received a higher education are higher? intelligence ability to pay We cannot easily do a randomized evaluation to solve this problem: it is not easy to randomly assign education! Instead, use educational expansion for a difference in difference strategy.
The Indonesian Inpres program 1973, oil shock: Indonesia is a oil producing country, used the money to build schools. School construction campaign started in 1973: affect cohort age 12 or younger in 1973. Those who were older did not benefit More schools were built in regions that were initially lagging behind in term of education Results: Impacts of the program on Education and on log(wages), both measured in 1995 (after the program is finished). We compare a young cohort (affected) to a old cohort (not affected), and we do the usual exercise of seeing whether the increase in education (and wages) between the young and the old cohort was faster in the regions that got more schools: Graph Graph increase in education and wages between old and young as a function of how many schools were built
Using the Indonesian INPRES program to measure the returns to education School building should have only increased education through wages: we can say that the 1.5% increase in wage that each new school caused for the young guys, is due to the fact that they got 0.20 years of education. If they had gotten one more year of education, what would have been their increase in ages? one year is 5 times more (1/0.20) than 0.20 year: effect would be five times larger. 1.5/0.20=7.5 This is our best estimate of returns to education: it only uses the part of the increase in
Education and Health Taiwan had a similar program for junior high school: rapid construction in 1968 to allow for universal compulsory junior high school education. A study by Grossman et al. applies the same technique and finds that women (and men) who were more educated thanks to the reform and the construction of schools in their regions had children who were less likely to be low birth weight, less likely to die within one month or within one year. Compulsory schooling is found to have saved one children 1 in 1,000!
Education and fertility Early fertility (teen pregnancy) is good neither for the child nor for the mother. Can keeping girls in school reduce the incidence of early fertility? Uniform distribution in Kenya: Reduced girls drop out in the next three years from 18% to 12%. Reduced teen pregnancy in the next 3 years from 14.4% to 9.8% The effect persist for older girls Graph Education is a powerful force to reduce early fertility.
Conclusion Individual data suggests strong effects of education in individuals Some mystery at the aggregate level: high correlation in levels, none in growth Part of the reason may be that the quality of education of the late arrivals (countries that increased education fast in the last few years) is lower. We d better have a look at the quality of education!!
Years of education and GDP log output per worker relative to the US 0-2 CAN ITA FRA NOR NET BEL GER SWI AUS SWE SPA AUT IS L UKG FIN ISR DEN SGP HKG IR E JPN VEN TRI SYR MLT JOR MEX ARG TWN GRE CYP BRB URS POR KOR ALG BRA URU IR N YUG HUN COL MRS SAF MLS CRI CHI FIJ GUA TUN TUR REU PER ECU POL PAN DOM CZE EGY SWZ PAR ELS THA SRL BOL PAK BAN CON NIC HON JAM PHL ID N ROM GUY BOT IND PAP BEN SUD SEN CAM SLE ZBW LES KEN HAI CHN GHA NGR GMB ZAM RWA MO Z TOG MAL CAF UGAZAI BRM MLW USA NZE -4 0 2 4 6 8 10 years of schooling
Years of education and individual wage 1.4 1.2 1 0.8 0.6 0.4 0.2 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Years of education Returns to Each Year of Education (OLS Estimate) Figure by MIT OpenCourseWare.
Growth in education and GDP growth From William Easterly, The Elusive Quest For Growth: Economists' Adventures And Misadventures In The Tropics, published by The MIT Press. Used with permission.
School building and growth and education and wages Educ.of young cohort Educ. of old cohort A1: Experiment of interest: education A2: Experiment of interest: log(wages) 5 0 4 3 2 1 0 1 1 0 2 4 6 8 10 0 2 4 6 8 10 Number of INPRES schools per capita Number of INPRES schools per capita Log(wages) of young cohort Log(wages) of old cohort 0.2 0.4 0.6 0.8 Slopes: 0.20 extra year and 0.015% higher wage for each school built per 1,000 kids
Uniform Distribution and Fertility Panel A: Fraction ever pregnant by end 2005 by age in 2005.25.2 No Uniforms Uniforms Fraction.15.1.05 0 13 14 15 16 17 Impact of uniforms program on teenage pregnancy, by age Figure by MIT OpenCourseWare.
MIT OpenCourseWare http://ocw.mit.edu 14.73 The Challenge of World Poverty Fall 2009 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.