Poor Economics

by Abhijit Banerjee

  WHY SCHOOLS FAIL

  Because in many developing countries, both the curriculum and the teaching are designed for the elite rather than for the regular children who attend school, attempts to improve the functioning of the schools by providing extra inputs have generally been disappointing. In the early 1990s, Michael Kremer was looking for a simple test case to perform one of the first randomized evaluations of a policy intervention in a developing country. For this first attempt, he wanted a noncontroversial example in which the intervention was likely to have a large effect. Textbooks seemed to be perfect: Schools in western Kenya (where the study was to be conducted) had very few of them, and the nearuniversal consensus was that the books were essential inputs. Twenty-five schools were randomly chosen out of 100, and textbooks (the officially approved books for those classes) were distributed. The results were disappointing. There was no difference in the average test scores of students who received textbooks and those who did not. However, Kremer and his colleagues did discover that the children who were initially doing very well (those who had scores near the top in the test given before study began) made marked improvement in the schools where textbooks were given out. The story started to make sense. Kenya’s language of education is English, and the textbooks were, naturally, in English. But for most children, English is only the third language (after their local language and Swahili, Kenya’s language), and they speak it very poorly. Textbooks in English were never going to be very useful for the majority of children.35 This experience has been repeated in many places with other inputs (from flip charts to improved teacher ratios). As long as they’re not accompanied by a change in pedagogy or in incentives, new inputs don’t help very much.

  It should now be clear why private schools do not do better at educating the average child—their entire point is to prepare the best-performing children for some difficult public exam that is the stepping- stone toward greater things, which requires powering ahead and covering a broad syllabus. The fact that most children are getting left behind is unfortunate, but inevitable. The school Abhijit went to in Calcutta had a more or less explicit policy of expelling the bottom of the class every year, so that by the time the graduation exam came around, it could claim a perfect pass record. Kenyan primary schools adopt the same strategy, at least starting in sixth grade. Because parents share these preferences, they have little reason to put pressure on the schools to behave otherwise. Parents, like everyone else, want schools to deliver what they understand to be an “elite” education to their child—despite the fact that they are in no position to monitor whether this is what is actually being delivered or give any thought to whether their children will benefit from it. For example, English-language instruction is particularly popular with parents in South Asia, but non-English-speaking parents cannot know whether the teachers can actually teach in English. The flipside of this is that parents have little interest in the summer camps and the evening classes—kids who need those classes are not going to win the lottery, so what is the point?

  We can also see why Pratham’s summer schools worked. The public-school teacher seems to know how to teach the weaker children and is even willing to put some effort into it during the summer, but during the regular school year this is not his job—or so he has been led to believe. Recently, also in Bihar, we evaluated a Pratham initiative to fully integrate remedial education programs into government schools, by training the teachers to work with their materials and also by training volunteers to work as teacher’s assistants in these classrooms. The result was striking. In those (randomly chosen) schools that had both the teacher training and the volunteers, the gains are substantial, mirroring all the Pratham results we saw above. Where there was just teacher training, on the other hand, essentially nothing changed. The same teachers who did so well during the summer camps completely failed to make a dent: The constraints imposed by the official pedagogy and the particular focus on covering the syllabus seem to be too much of a barrier. We cannot just blame the teachers for this. Under India’s new Right to Education Act, finishing the curriculum is required by law.

  At the broader, societal level, this pattern of beliefs and behavior means that most school systems are both unfair and wasteful. The children of the rich go to schools that not only teach more and teach better, but where they are treated with compassion and helped to reach their true potential. The poor end up in schools that make it very clear quite early that they are not wanted unless they show some exceptional gifts, and they are in effect expected to suffer in silence until they drop out.

  This creates a huge waste of talent. Among all those people who drop out somewhere between primary school and college and those who never start school, many, perhaps most, are the victims of some misjudgment somewhere: Parents who give up too soon, teachers who never tried to teach them, the students’ own diffidence. Some of these people almost surely had the potential to be professors of economics or captains of industry. Instead they became daily laborers or shopkeepers, or if they were lucky, they made it to some minor clerical position. The slots that they left vacant were grabbed, in all likelihood, by mediocre children of parents who could afford to offer their children every possible opportunity to make good.

  Stories about great scientists, from Albert Einstein to the Indian math genius Ramanujam, both of whom did not make it through the educational system, are of course well-known. The story of the company Raman Boards suggests that this experience may not just be limited to a few extraordinary people. A Tamil engineer named V. Raman started Raman Boards in Mysore in the late 1970s. The company made industrial-grade paper products such as the sheets of cardboard used in electrical transformers. One day, V. Raman found a young man, Rangaswami, outside the door of the factory, asking for a job. He was from a very poor family, he said, and he had some engineering education, but just a diploma, not a proper college degree. Compelled by his insistence that he could do good work, Raman gave him a quick intelligence test. Impressed by the results, he took the young man under his wing. When there was a problem, Rangaswami would be assigned the task, and working initially with Raman, but increasingly on his own, he would come up with a creative solution to it. Raman’s firm was eventually bought up by the giant Swedish multinational, ABB—it is now the most efficient of the many plants that ABB runs the world over, including in Sweden. Rangaswami, the man who could not get an engineering degree, is the head of engineering. His colleague, Krishnachari, another of Raman’s finds—an ex-carpenter with little formal education—is a key manager in the components division.

  Aroon, Raman’s son, who ran the company before it was sold, now runs a small R&D unit with a few people who were with him at Raman Boards. His core research team of four includes two people who never completed high school, and no qualified engineers. They are brilliant, he says, but at the beginning the problem was that they didn’t have the confidence to speak up, so how could one know? It is only because it was a small firm, and yet one that did a lot of R&D, that they were discovered. And even then it took a lot of patient work to discover their capabilities and they needed constant encouragement.

  This model is obviously not easy to replicate. The problem is that there are no straightforward ways to identify talent, unless one is willing to spend a lot of time doing what the education system should have been doing: giving people enough chances to show what they are good at. Yet Raman Boards is not the only firm that thinks there is a lot of undiscovered talent out there. Infosys, one of India’s IT giants, has set up testing centers where people, including those without much formal qualification, can walk in and take a test that focuses on intelligence and analytical skills rather than textbook learning. Those who do well get to become trainees, and successful trainees get a job. This alternative route is a source of hope for those who fell through the gaping holes in the education system. When Infosys closed its testing centers during the global recession, it was front-page news in India.

  A combination of unrealistic g
oals, unnecessarily pessimistic expectations, and the wrong incentives for teachers contributes to ensure that education systems in developing countries fail their two basic tasks: giving everyone a sound basic set of skills, and identifying talent. Moreover, in some ways the job of delivering quality education is getting harder. The world over, education systems are under stress. Enrollment has gone up faster than resources, and with the growth in the high-tech sectors, there is a worldwide increase in the demand for the kind of peoples who used to become teachers. Now they are becoming programmers, computer systems managers, and bankers instead. This is going to be a particularly serious issue for finding good teachers at the secondary level and beyond.

  Is there a way out, or is the problem simply too difficult?

  REENGINEERING EDUCATION

  The good news, and it is very good news indeed, is that all the evidence we have strongly suggests that making sure that every child learns the basics well in school is not only possible, it is in fact fairly easy, as long as one focuses on doing exactly that, and nothing else.

  A remarkable social experiment from Israel shows how much schools can do. In 1991, 15,000 more or less indigent Ethiopian Jews and their children were airlifted out of Addis Ababa in a single day and dispersed into communities all over Israel. There, these children, whose parents had had on average between one and two years of schooling, entered elementary schools with other Israeli children, both long-term settlers and recent immigrants from Russia, whose parents had had on average 11.5 years of schooling. The family backgrounds of the two groups could not have been more different. Years later, at the point when those who entered school in 1991 were about to graduate from high school, the differences had narrowed considerably. Sixty-five percent of the Ethiopian children had reached twelfth grade without grade repetition, compared to the only slightly higher 74 percent among the Russian emigrants. It turns out that even the most severe disadvantage in terms of family background and early life conditions can largely be compensated for, at least in Israeli schools, where the right conditions are met.36

  Successful experiments have given us a number of ideas on how to create these conditions. A first factor is a focus on basic skills, and a commitment to the idea that every child can master them as long as she, and her teacher, expends enough effort on it. This is the fundamental principle behind the Pratham program, but it is also an attitude that is encapsulated by the “no excuse” charter schools in the United States.37 These schools, such as the Knowledge Is Power Program (KIPP) schools, the Harlem Children’s Zone, and others, mainly cater to students from poor families (particularly black children), with a curriculum that focuses on the solid acquisition of basic skills and continuous measurements of what children actually know: Without such diagnosis, it is impossible to evaluate their progress.

  These schools have been shown, in several studies based on comparing those winners and losers of the admission lotteries, to be extremely effective and successful. A study of charter schools in Boston suggests that expanding fourfold the capacity of charter schools and keeping the current demographic profile of students the same would have the potential to erase up to 40 percent of the citywide gap in math test scores between white and black children.38 The mechanism at play is exactly what we see in Pratham’s programs: Children who are completely lost in the regular school system (their test scores are way behind those of other children when they enter charter schools) are given a chance to catch up, and many take it.

  A second piece of good news from Pratham’s work is that it takes relatively little training to be an effective remedial teacher, at least in the lower grades. The volunteers who had such dramatic effects were mostly college students and other people with a week or ten days of training in pedagogy. Moreover, this extends beyond teaching only reading and basic arithmetic. The same program in Bihar that put volunteers in classrooms also had them teach the children who could read well to use their reading skills to learn—Pratham calls this Reading to Learn, the sequel to its more basic Learning to Read—and the learning gains were substantial. Charter schools mainly use young, enthusiastic teachers, and they are able to significantly help both primary-school and middle-school children.

  Third, there are large potential gains to be had by reorganizing the curriculum and the classrooms to allow children to learn at their own pace, and in particular to make sure the children who are lagging behind can focus on the basics. Tracking children is a way to do that. In Kenya, the study mentioned earlier compared two models to assign first-grade students to two separate classes. In one model, children were randomly assigned to a classroom. In the other, they were split up based on what the children already knew. When students were assigned according to their initial level, so that the teachers could address the children’s needs better, students at all levels of initial achievement did better. And the gains were persistent: At the end of third grade, students who had been tracked in first and second grades were still doing better than those who had not been tracked.39 Alternatively, one could find other ways to tailor the teaching to the needs of individual students. One possibility is to make the boundaries between the grades more fluid, so that a child whose age puts him in fifth grade but who needs to take second-grade classes in some subjects can do so without additional stigma.

  More generally, a lot could be done to change the unrealistic expectations that everyone has. A program in Madagascar that simply told parents about the average income gains from spending one more year in school for children from backgrounds similar to theirs had a sizable positive effect on test scores, and, in the case of parents who found out that they had underestimated the benefits of education, the gains were twice as large.40 An earlier study in the Dominican Republic produced similar results with high school students.41 Since it is essentially free to have teachers simply pass on information to parents, this is so far the cheapest known way to improve test scores, among all the interventions that have been evaluated.

  It may also be a good idea to try to set more proximate goals for both children and teachers. That way everyone can stop focusing so much on that one elusive outcome at the end of many years. A program in Kenya that offered a $20 USD PPP scholarship for the next year to girls who scored in the top 15 percent on an exam not only got the girls to do much better, but it also put pressure on the teachers to work harder (to help the girls), which meant that boys did better, too, even though there was no scholarship for them.42 In the United States, rewarding children for achieving long-term goals (such as getting high grades) was not successful, but rewarding them for effort on reading proved extremely effective.43

  Finally, given that good teachers are hard to find and information technology is getting better and cheaper by the day, it seems rational to use it more. The current view of the use of technology in teaching in the education community is, however, not particularly positive. But this is based mainly on experience from the rich countries, where the alternative to being taught by the computer is, to a large extent, being taught by a well-trained and motivated teacher. As we have seen, this is not always the case in poor countries. And in fact, the evidence from the developing world, though sparse, is quite positive. We did an evaluation of a computer-assisted learning program run in collaboration with Pratham in the government schools in Vadodara in the early 2000s. The program was simple. Pairs of third- and fourth-graders got to play a game on the computer. The game involved solving progressively difficult math problems; success in solving them gave the winner a chance to shoot some garbage into outer space (this was a very politically correct game). Despite the fact that they only got to play for two hours a week, the gains from this program in terms of math scores were as large as those of some of the most successful education interventions that have been tried in various contexts over the years, and this was true across the board—the strongest children did better, and so did the weakest children. This highlights what is particularly good about the computer as a learning tool: Each child is able to set
his or her own pace through the program.44

  This message of scaling down expectations, focusing on the core competencies, and using technology to complement, or if necessary substitute for, teachers, does not sit well with some education experts. Their reaction is perhaps understandable—we seem to be suggesting a two-tier education system—one for the children of the rich, who will no doubt get taught to the highest standards in expensive private schools, and one for the rest. This objection is not entirely unwarranted but unfortunately, the division exists already, with the difference that the current system delivers essentially nothing to a very large fraction of children. If the curriculum were radically simplified, if the teacher’s mission were squarely defined as making everyone master every bit of it, and if children were allowed to learn it at their own pace, by repeating if necessary, the vast majority of children would get something from the years they spend in school. Moreover, the gifted would actually get a chance to discover their own gifts. It is true that it would take some work to put them on the same footing as those who went to elite schools, but if they had learned to believe in themselves, they might have a chance, especially if there is a willingness in the system to help them get there.45 Recognizing that schools have to serve the students they do have, rather than the ones they perhaps would like to have, may be the first step to having a school system that gives a chance to every child.

  5

  Pak Sudarno’s Big Family