The significance of education for economic growth and a progressive society needs no argument. India's performance in this area is indeed shameful. I do not want to delve on the past, but look towards the future. In this paper, I first discuss some of the myths about the problems of education in India, then evaluate recently proposed solutions, and at the end, offer my agenda for reform.

I. Challenging the Conventional Wisdom

Myth 1: The poor need their children to work/earn.

This is one of the most pernicious beliefs that pervade the education establishment. Several surveys including the recent PROBE study provide data to challenge this widely held belief. Nonetheless the belief controls much of education policy discussion. Its acceptance is taken as the sign of one's genuine concern for mass education, and its rejection as elitism or extremism.

In the past children were required to help the family, but the current realities are rather different. We must not be the prisoner of the past, especially in such a critical task of formulating national education policy.

Estimates of Child Labour (All-India, Rural)

Proportion (%) of children aged 5-14 who are in the workforce

	Female	Male
Census of India, 1981	8.8	10.0
National Sample Survey, 1993	7.8	6.9
NCAER Survey, 1994	3.5	4.4

Even in the PROBE states, the worst performer BIMARU states, among the out-of-school children in the age group of 6-12, only 5 percent of boys and 1 percent of girls engage in wage labour. The median work hours for boys are 3.3 and for girls 4.8, the average being 4 hours per day. However their cohorts who actually attend school also work on average 2 hours a day. Is it possible that parents keep their children out of school for the benefit of two extra hours of work?

The PROBE study concludes that only a small minority of children is full-time labourer and most of them work not as wage-labourers but as family helpers at home or in the fields. Moreover, the presumed causation that child labourers are unable to go to school because they have to work may actually be reverse, that is, they work because they have dropped out of school for other reasons (p. 16).

Myth 2: People are ignorant of the benefits of education.

Surveys to investigate why some children are never enrolled in school find that more than one-third of the children or parents are "not interested" in education. The finding is used to conclude that people do not really understand the importance of education and therefore cannot be relied upon to educate their children.

Work Patterns of Out-of-School Children (PROBE States) Boys Girls Proportion who performed wage labour on the day preceding the survey 5% 1% Average time of work on the day preceding the survey* 4.2 hours (3.3 hours) 5.1 hours (4.8 hours) Extra time of work, compared with children who are attending school 2.1 hours 2.2 hours *Median in brackets Source: PROBE survey (random sub-sample of 333 out- of-school children in the 6-12 age group)

Field researchers however have now come to the view that the supply and not the demand factors make people less interested in education. The quality of infrastructure (buildings, furniture, and supplies), teaching, and the concern for learning is so poor as to make the whole idea of education worthless, a waste of time. Whenever these drawbacks have been rectified, generally due to the initiative of some individual teacher, or principal, or bureaucrat, enrolments improve dramatically. (In the Total Literacy Programme, a few districts performed exceptionally well because of particular individuals.) People do not lack understanding of education's importance; they simply cannot be fooled into spending their time and money for a mediocre or a non-existent one. The PROBE study finds that almost 90 percent of the parents are keen to send their children to school, and that the gender discrimination in the desire to educate boys and girls is far lower than generally assumed.

Myth 3: People do not have money or are unwilling to spend on education.

The NCAER (1994) and NSSO (1991, 1993) studies suggests that in a given year people spend anywhere from Rs. 100 to more than Rs. 4000 per child on primary education. For the year 1986-87, the total household expenditure on primary education was Rs. 7388.5 million. This private spending on primary education is not just by the rich and the middle class in urban areas. Out of the Rs. 7388.5 million, Rs. 4202.5 million were spent by rural areas, more than half of the total private expenditure. In that year, the total government expenditure on primary education was Rs. 17,000 million. So the private expenditure is by no means an insignificant amount; it was more than 40 percent of what the government spent on primary education.

The NSSO study also points out that households in our villages on average spend about Rs. 104 on fees, Rs. 42 for books and supplies, and Rs. 107 for private coaching to provide primary education to their children. The poor have already been spending on education.

States in which governments spend more on education, people also spend more. If one interprets higher spending by governments as a little better quality of education, then it is apparent that parents are willing to spend when they are offered the kind of education they desire. In Kerala, for example, the poor spend a substantial portion of their income on primary education, one NSSO study suggests up to one-third of the income.

Myth 4: Basic education provided by the government is free.

The previous discussion clearly indicates that people, including the rural poor, spend a great deal of money on basic education of their children. A detailed analysis of the NCAER and NSSO studies leads Professor J.B.G. Tilak of NIEPA (1996, p. 363) to the following conclusions: "Households spend large sums of money on acquiring primary education; a sizable number of students do not receive primary education free, in contrast to the claims made by the government; a large number of students pay tuition fee, examination fee and other fees even in government primary schools in India."

It is extremely important to understand these myths and the ground reality in India so that we would not prescribe solutions for non-existent problems and overlook the real problems that confront suppliers and demanders of education. False solutions would not only delay achievement of our goals but also waste our scarce resources.

II. Proposed solutions: Would they be effective?

Solution 1: Make elementary education compulsory.

A lot of ink has been poured and throats have become sore demanding urgent passage of a law making elementary education compulsory. Proponents claim that it would force parents to send their children to school and the government to make schools available in all parts of India. But above all, the contention goes, it would signal society's sincere commitment to basic education for all.

Is such a law necessary to achieve universal education? Kerala has no compulsory-education law and has the highest literacy rate. The passage of the law would actually allow the government to become more passive, albeit with a self-righteous alibi. Our experience with other good-intention laws against child labour, child marriage, and other social ills does confirm this outcome. Those who want to practice these have rarely been deterred. And any positively development in these areas could hardly be attributed to these laws. The passage of law by itself barely fosters positive result.

The compulsory-education law would just create opportunities for more commissions, national and international seminars, and pious declarations. Sure, more money would be spent in the name of education, but not on education. If we are going to measure government's resolve by the size of the education budget, it is rather easy to spend more money that will keep all stakeholders in our education system happy-politicians, bureaucrats, and educationists. Except of course parents and children, in whose name the money will be spent. The experience all over the world clearly suggests that it is not so much the size of the government education budget but how the budget is spent that determines the efficacy of the education system. Our own state of Kerala stands witness to the significance of choice and competition in education.

How would the compulsory-education law be enforced? Would it fine parents for not sending their children to school? How would the parents who are too poor to send their children to school find money to pay the fines? Would it then imprison them? Just one parent or both? Would failure of the government in properly enforcing the law or in creating necessary school infrastructure so that the citizens could abide by the law result in its dismissal? Would the dismissal apply to the state as well as the central government? Or would it result in the dismissal of just education ministers? Along with the secretaries of the departments of education? Would the law, given these enforcement problems, ever be properly and consistently applied? Who would bear the burden of arbitrary enforcement? The fact that the proponents have not bothered to raise the issues of enforcement, let alone suggest any answers, reveals their lack of sincerity. They want to claim to have done something for the poor state of education in the country without really doing anything-just by passing a law.

Practical difficulties aside, we must not forget the distinction between laws and morals. No society has been able to legislate moral behaviour. We are no different. The world history of prohibition stands as a guide: Rise of the mafia, street violence to capture markets, corruption of the police, legal system, and the bureaucracy, untold deaths due to illicit liquor. In the final analysis, prohibition has done more damage, not just to the treasury, which is of the least consequence, but to families and communities, the alleged beneficiaries of the ban. The law of unintended consequences is far more potent in case of good-intention laws.

Education is undoubtedly critical for the child, but so is nutritious diet, healthcare, nurturing environment. Education, even a good education, without the other necessary ingredients would make the child only marginally better. Our concern is with all-round welfare of the child; we need sensible and active parenting for all the children. If we sincerely and genuinely believe in the effectiveness of the law to address social problems, shouldn't we then pass a law making good parenting compulsory?

Solution 2: Take education out of the Directive Principles and put it in the list of Fundamental Rights.

It is forgotten that the Articles 23 and 24 of the Constitution guarantee a fundamental Right against Exploitation. Article 23: Prohibition of traffic in human beings and forced labour-Traffic in human beings and beggar and other similar forms of forced labour are prohibited... Article 24: Prohibition of employment of children in factories-No child below the age of fourteen years shall be employed to work in any factory or mine or engaged in any other hazardous employment. Readers must judge how effectively these fundamental rights are protected by the government.

It is important to note that these Articles do not impose any positive obligation on the government; they require that the government simply prevent certain activities, namely, forced labour and employment of children in hazardous jobs. They do not mandate the government to provide alternative means of survival to those who are forced into labour (beggars for example) or to the children who are in hazardous jobs. They do not ask the government to provide for the care of these unfortunates, nor do they require the government to even offer safe, remunerative, non-hazardous employment. They mandate only prevention, not provision. Despite this limited mandate, the government has miserably failed in living up to the promise of Articles 23 and 24. The proponents of the fundamental right to education actually stipulate that government would be required to provide necessary schooling facilities. Is the government that cannot enforce Articles 23 and 24 capable of meeting the demands of the fundamental right to education? Again, the proponents do not properly address these questions.

Solution 3: Increase government spending on education to 6 percent of GDP.

The almost doubling of government expenditure on education is bound to have some positive impact. There is probably no government, particularly a democratic one, which is so inefficient and corrupt as to eat up all the increase in spending without any impact on the quantity and quality of education. The question is whether the impact would be in any way proportionate to the increase in spending.

Several of the countries that achieved high literacy rates in the post-war era, have rarely spent anywhere close to the 6 percent of their GDP. South Korea has spent about 3.2 percent, Japan spends around 3.8 percent, and China 2.6 percent. Student expenditure in the United States is one of the highest but student performance in mathematics, language, reasoning skill, and even geography is far below world standards. International evidence suggests that it is not how much the government spends but how it spends determines the quality of its education system. Filmer and Pritchett (1999), in their cross-section study of the states and union territories in India for the year 1992-93 to test the effect education expenditures on enrolment of students, especially from the lower income groups, conclude

"There is little evidence of a positive relationship between per student expenditures and enrolments of the students from the bottom 40 percent of family income.... This lack of a general effect is not surprising, as there is a huge literature that supports the proposition that, while additional spending has the potential to raise school quality, there is no necessary connection between school quality and school spending..." (p. 159).

A Tale of Two States

Kerala has the highest literacy rate in the nation, above ninety percent. The way Kerala spends its education money is also strikingly different from the other states. For illustration, I compare it with the state of West Bengal. Ideologically the governments of both states are equally committed to basic education and literacy. Both states have for long had popularly elected Marxist governments. The conclusions of the comparative analysis are however generally valid.

Table 1 highlights some of the crucial differences in the educational structure and the nature of government spending on education in the two states. Table 2 shows the effect of those differences on the performance of the education system in terms of literacy rate and the proportion of children never enrolled in school. (All data are from the NSSO 1991, 1993 and NCAER 1994; see also Tilak 1996. The data are for the year 1986-87 or 1991-92.) Kerala is one of the few states in the country were elementary education is not made compulsory by law. Both governments spend almost equal fraction of the total budget on education (about 25 percent). In West Bengal, 84 percent of rural children do not pay any fee for primary education but that number is only 48 percent in Kerala. Sixty percent of rural primary school children get free textbooks and supplies in West Bengal, only two percent in Kerala do. Households with less than Rs. 3000 in annual per capita income spend 25 percent of the income on elementary education in West Bengal but in Kerala it is 36 percent. The poor in Kerala spend the highest fraction of their income on their children's basic education compared to the poor in any other state in the country.

Table 1: State Commitment to Education

Characteristics	West Bengal	Kerala
Elementary Education Compulsory	Yes	No
Fee	Free
Primary Education	84%	48%
Free Textbooks and Stationary	60%	2%
Proportion of Income Spent on Primary Education by Households in the Lowest Income Quintile	25%	36%
Share of Education in the State Budget	26%	25%

Given these facts-more children get free education and supplies in West Bengal and the poor are asked to spend more of their own money in Kerala-one would expect that West Bengal would have a much higher literacy rate than Kerala. The facts speak otherwise

(Table 2). Kerala has 91 percent literacy rate while West Bengal has only 57 percent. Moreover, in West Bengal 46 percent of children in the age group of 6 to 14 have never enrolled in school, only two percent in Kerala suffer from that fate. What explains this vast difference in performance?

Table 2: State Performance in Education

Characteristics	West Bengal	Kerala
Literacy Rate	57%	91%
Children (age 6-14) Never Enrolled	46%	2%

Kerala and West Bengal: Unfair Comparison

Kerala undoubtedly has had a head-start: There have been strong education movements in the state since the pre-independence days and the government has consistently spent a much larger proportion of its budget on education since independence. It then seems unfair to compare the two states in terms of their educational performance. The cross-section comparisons at a single point in time do not control for variations over time. Kerala's current spending on education is almost the same as West Bengal, but since Kerala had a head-start, current literacy rates and the reach of education are likely to be different. Nonetheless it is instructive to examine the distribution of their education spending. Kerala and West Bengal have chosen to spend their education money rather differently. The difference in the nature of their spending is the real purpose of this comparison.

Table 3: Distribution of State Education Spending

Characteristics	West Bengal	Kerala
Free Primary Education in Government Schools	84%	48%
Free Primary Education in Private Schools	15%	48%
Grant of Scholarship	0.5%	10%
Transport Subsidy	2.3%	5.4%
Proportion of Private (aided) Primary Schools	11%	60%

It is surprising that in a thoroughly Marxist state like Kerala, 60 percent of the rural primary schools are private, as compared to only 11 percent in West Bengal. The proportion of private primary schools in Kerala is the highest in the country; the second highest is Maghalaya at 21 percent, and the national average is only five percent. The government of Kerala also pays expenses of almost half of the students enrolled in private primary schools. The number for West Bengal is 15 percent which is the third highest in the country (Tamil Nadu is at 20 percent); the national average is again about five percent.

Kerala has the highest proportion of private primary schools and it also subsidises the highest proportion of students in private schools. Both of these facts give the citizens of Kerala wider effective choice in selecting primary schools for their children. Many of the private schools are run by various religious groups in the state. They are generally more likely to be successful in exerting pressure on parents to send their children to school. The choices available to parents must increase attendance as well as retention rates in the state.

Kerala uses its public funds to encourage competition among schools. To avoid transportation costs, most parents generally send their children to the nearest school. The resulting "geographical clustering" of schools and their customers lessens competition among schools. Each school has a captured customer base. By subsidising transportation costs, Kerala helps parents send their children to the school they consider best, irrespective of the distance. This increases competition among schools. The provision of direct scholarship to students in Kerala also leads to the same result. With the scholarship money, students can go to any school of their choice. Among all the states in the country, the highest proportion of children in Kerala receives transportation subsidies and direct scholarships (Table 3).

The focus on how the two governments spend their education rupees indicates that Kerala by offering more choices to parents and increasing competition among schools actually practices market principles. Kerala's citizens have received far better educational service than those of almost any other state in the union. The Kerala model of education-of choice and competition-is unique in the country, and so is Kerala's educational performance. It is not just how much a state spends on education but how it spends that determines efficiency and effectiveness of the education system. The status of higher education in these two states is also worth comparing. State universities in West Bengal receive 91 percent of their budget from the government. In Kerala it is only 54 percent, the remaining amount is generated by fees, donations, endowments, and other sources. Again Kerala requires its universities to raise almost half of their budget from the customers and communities they serve. This fosters accountability and more attention to the needs of those who help finance state universities. This is one of the important reasons that Kerala performs better also in higher education than many other states in the union.

Source of Funding and the Nature of Spending	Central Universities	State Universities
Percent of Budget from the Government	90%	50%
Percent of University Budget Spent on Administration	41%	18%
Percent of University Budget Spent on Academic Programs	33%	55%

It may be pertinent to note that in general the higher the funding from the government, the lower the spending on academic programmes at universities. Central universities receive more than 90 percent of their funds from the central government and spend about 33 percent on academic programs and support and 41 percent on administration. The state universities on average get a little more than 50 percent of their money from state governments and spend 55 percent on academics and only 18 percent on administration. The state universities that are more dependent on non-government funds pay more attention to their students and less to their bureaucracy.

In Kerala, the government has been spending more on education but so do the people of Kerala. The poor in the state spend about 36 percent of their annual per capita income on elementary education-the highest proportion in the country (Table 1). (I find it difficult to believe these numbers; it is hard to imaging that poor parents spend more than one third of their income on elementary education. But in any case it must be some positive number even far lower than 36 percent). Contrary to the conventional wisdom, government spending is not a substitute for private spending. Both seem to grow together; they are complementary. Parents' financial commitment to their children's education is a crucial component of quality education. Moreover, as the empirical evidence suggests, schools and universities that depend on non-government funds manage their finances more responsibly and are more attentive and responsive to the needs of their customers.

These are some of the macro solutions that have been offered to improve education in India. Scores of micro solutions that deal with curricula, pedagogy, and such are also on the table. Many of them are relevant and useful but there is no particular pattern to them or any systemic theme unifying them. Since the focus of the paper is on broad structure of the education system, I do not discuss the micro reform issues.

III. Agenda for Reform

Reform 1: Liberalise, abolish the license-permit raj in education.

We have abolished the license-permit raj in the economy. Instead of government determining what and how much of goods and services should be produced, under market principles of choice and competition, we now allow businesses and consumers to make those decisions. In the economy, government sovereignty has been abolished and consumer sovereignty has been established. We need the same liberalisation in the education system.

It is more readily accepted that increased choice and competition are good for higher education. The central government had introduced a bill which would have allowed establishment of private universities. Private colleges already exist in professional fields such as medicine and engineering. But there is no full-fledged private university in the country. The bill would have helped increase choice in higher education. It would also have lessened the financial burden on central and state governments of subsidising higher education. As students who could afford to pay the tuition go to private universities, less subsidy would be necessary, or more resources would be available to now smaller number of students in government universities. University Grants Commission has allowed several colleges to become autonomous. According to one estimate (Rao 1999), there are about 121 autonomous colleges in eight states that have taken advantage of the permission. (Andra Pradesh 19; Gujarat 2; Himachal Pradesh 5; Madhya Pradesh 41; Maharastra 3; Orissa 5; Tamil Nadu 43; and Uttar Pradesh 3) A qualitative analysis of the autonomous and non-autonomous colleges suggest that former perform relatively better in terms of personnel and financial management, infrastructure facilities, teaching-learning processes, and student evaluation systems (Rao, Matthew, and Samantray 1999). Encourage of this trend towards autonomy should be the primary focus in the area of higher education.

Choice and competition to improve the quality and financial viability of primary and secondary education is actually more urgent. Though that proposition is not as readily accepted. It is argued that market principles may be good in higher education but they are inoperative, irrelevant, or even harmful in basic education. Many maintain that in achieving basic education in a poor and largely illiterate country like ours, government must control the education system as closely as possible. They grossly underestimate the stifling effects of the license-permit raj in education.

To establish a new school requires a license from education authorities. The barriers to entry are so numerous and diverse that they discourage most honest people who would like to serve the cause of education. Licensing has the same effect in education as it had on the economy. The government limits competition and arrogates the power to decide how many and what type of schools can serve educational needs of the people. Most importantly, it creates artificial scarcity of schools and allows existing schools to exploit their customers. To curb such exploitation, the government puts further restrictions on the amount of fees that can be charged for tuition, laboratory, and transportation and on the amount and purposes for which donations can be asked from parents. The restrictions are impossible to enforce because the artificially created scarcity compels parents to pay up; they are no alternative schools to send their children. This morass of controls is what the "potato chip theory of regulation" predicts. One restriction creates situation that demands further restrictions, which in turn require more restrictions. Once a bag of potato chips is opened, it's hard to stop at one or a few chips.

The need for liberalisation in education is no mere rhetoric. Take for example the Delhi School Education Act, 1973. Rule 44 of the Act requires "an intimation in writing to the Administrator of his or their intention to establish a school" with details about the management system, funds on hand, teachers and staff, design of the building. The school must obtain "Essential Certificate" by establishing that its existence serves the public interest. The Administrator decides by taking into account "the number and categories of recognised schools already functioning in that locality, and general desirability of the school with reference to the suitability and sufficiency of the existing schools in the locality and the probable effect on them (my emphasis)." In this environment the capture theory of regulation predicts permanent artificial scarcity of schools. The theory suggests that the existing businesses would take control of the licensing/regulatory agency and thereby create barriers to entry and reduce competition in the market. Not only opening of a new school but addition of a new class in an existing school also requires prior permission (Rule 45). Rule 46: Closing down of a school or any class in a school-No managing committee shall close down a recognised school, not being an unaided minority school, or an existing class in such school without giving full justification and without the prior approval of the Director, who shall, before giving such an approval, consult the Advisory Board.

Rule 8 tells that school cannot fire any employee without permission and due process. Terms and Conditions of Service of Employees of Recognised Private Schools, Clause 2: Subject to any rule that may be made in this behalf, no employee of a recognised private school shall be dismissed, removed or reduced in rank nor shall his service be otherwise terminated except with the prior approval of the Director. Clause 3: Any employee of recognised private school who is dismissed, removed or reduced in rank may, within three months from the date of communication to him of the order of such dismissal, removal or reduction in rank, appeal against such order to the Tribunal constituted under section 11.

Schools need an exit policy as badly as other businesses. The license raj in education however goes much further than it did in the industry. The government never required that the customer of a company must take the company's approval before switching to an another company to purchase a similar product or service. But, Rule 139: Admission on transfer certificate-No student who had previously attended any recognised school shall be admitted to any aided school unless he produces a transfer or school leaving certificate from the school which was last attended by him. Schools could deny their customers the freedom to switch; they have captive customers.

Rule 131 creates more corruption and arbitrariness. Power to Director to regulate manner of admission-Where the circumstances of a case so require, the Director may, notwithstanding anything contained in the admission plan, direct the admission of any student to an aided school, and, on receipt of the direction, the head of such school shall admit such student in the school.

Given the strict controls on entry, expansion, and closure, fee structure, collection of donations, government funding to schools is very minutely regulated. Chapter VI, Grant-in-Aid describes various heads under which grant is given.

Categories of aid 1. Aid shall be of two categories, namely- (a) maintenance grant; and (b) building grant 2. Maintenance grant shall be of two kinds, namely- (a) recurring maintenance grant; and (b) non-recurring maintenance grant 3. The recurring maintenance grants are- (a) staff grant; (b) provident fund grant; (c) pension and retirement benefit grant; (d) medical benefit grant; (e) benefits specified in Chapter X; (f) grants for the purpose of books and journals which are essential for the library; and (g) grants for the acquisition of essential equipments of the school. 4. Non-recurring maintenance grant Non-recurring maintenance grant shall be of the following categories, namely- (a) contingent grant; (b) rent grant; (c) depreciation grant for school; (d) hostel grant and depreciation hostel grant; (e) grant for equipment, furniture, games and sports materials and the like; (f) biennial or triennial grants for the purchase of books for the library and for the setting up of a book bank.

One cannot but be amazed that with all these paperwork schools actually manage to teach a little. It would indeed be far better to remove all these barriers and give schools an open grant so that they can decide how best to spend it to achieve quality education. Abolition of the license-permit raj in education would, as it has in the industry, increase choice and competition in the market for education and make schools more responsive to needs of students and parents.

Achievement Marks by School Management, Chennai, 1994-95 (Higher Secondary Level) Public Private Aided Private Unaided English All Boys Girls Number of Students Mathematics All Boys Girls Number of Students 47 48 47 351 49 48 51 192 61 62 59 602 55 52 51 424 77 77 77 595 76 74 79 470

Very few studies exist that compare the performance of private and government schools. However, P. Duraisamy and T. P. Subramanian (1999) compare the working of public, private aided, and private unaided higher secondary schools in the Chennai area. Their study (p. 43) indicates that in English and Mathematics, the two subjects they analysed, students of the private unaided schools did the best and those of the public schools the worst.

The specific reforms for education liberalisation are: A. Pass the private university bill. B. Repeal all barriers to entry and exit in lower and higher education. C. Grant autonomy to existing schools, colleges, and universities without reducing financial support. D. Remove departments of education and other government bodies from the provision of education, convert them in financing and monitoring agencies. E. Make choice and competition the central focus of education policy in this area.

Reform 2: Institutionalise accountability of government and private educational entities.

Once education is liberalised and all existing institutions are given autonomy, they will begin to compete for students. The decision-making powers, in all areas of management, would be handed down to schools, colleges, and universities-that is to parents, teachers, and local administrators. It is critical to institutionalise accountability, in managing finances as well as teaching and research, so that the competitive forces would work to improve, and not undermine, the overall experience of education.

A. Link government grants with performance

To facilitate the Schumpeterian process of creative destruction, government grants to all educational institutions must be linked with their academic performance. For completely private institutions, the link between revenues and performance does exist; parents would not continue to purchase their services unless the standards were met. But government grants to government as well as private academic institutions are largely automatic-based simply on the number of students and physical infrastructure of the past with an eye on the growth in the head count and the need for maintenance and expansion of facilities. (Refer to the earlier description of the multitude of headings under which grants are doled out.) Linking government grants with performance would help realign     self-interest and incentives of the institution with the needs of its customers. The institution's task is to educate;  its revenues should depend on the extent to which it meets its responsibility.

There are multitudes of ways to measure performance of institutions and the issue needs a deeper analysis. But  a simple way to gauge performance is to consider the number of students who pass the standardised board examinations. An institution's revenues could be linked with that number. Success in passing board     examinations is probably not the best way to judge the quality of education; we should think of better and yet  easier ways to assess the quality. The basic objective nonetheless is to make educational institutions earn their government grants by fulfilling their responsibilities.

B. Help establish independent certification, accreditation, and examination agencies.

Performance of the diverse institutions that would dot the educational landscape cannot be accurately measured by any single, uniform method. We would need as much experimentation in performance measurement as in the development of curricula and teaching pedagogy. Monopolies in the provision of education are harmful, so they are in the assessment of education. Competing sources of providers as well as assessors further the cause  of good education.

In place of the license-permit system, independent certification and accreditation agencies would help monitor the quality of education. Again, instead of a monopoly or oligopoly of such agencies, freely competing agencies would provide more objective and accurate information to parents and students to enable them to make     informed decisions. Significance and working of such independent institutions in various areas of the economy  are discussed in Desai 1998.

Reform 3: Empower students and parents.

In the current system, government grants are given to academic institutions-schools, colleges, and universities. For the market for education to work as efficiently and equitably as possible, it is critical that the support is given directly to students and parents instead of institutions. Free choices of empowered students and parents would then determine the educational landscape. Consumers of education would become sovereign.

The idea of empowering consumers of education is generally framed as separating provision/production/management from financing of education. The government does not run or control any academic institution, it simply provides support to students and parents so that they can purchase education in a competitive marketplace. There are several ways to implement this idea; two most common are scholarships and vouchers. Many countries, including those with high literacy rates, are moving towards a voucher system to improve the quality of their education. The education system of Chile is now almost completely based on vouchers. Singapore is moving towards Independent and Autonomous schools and colleges. Several school districts and states in the United States have been experimenting with a voucher system. College education in the U.S. has long been financed either directly by students and parents, or through scholarships given by colleges and foundations, or through government subsidised loans, or through outright grant from the government for those who meet the minimum family income criterion. (For details of the voucher system in several countries, see the special issue of Education and Urban Society, August1999.)

There are several types of voucher systems. The system is rather flexible so that it can be adapted to peculiar problems of a country. It also allows a great deal of freedom to institutions in managing their affairs. Through vouchers institutions receive a fixed some of money for educating students. Institutions then are free to decide how to spend that money so as to improve the quality of education, retain students (prevent dropouts), and attract more students to increase their voucher revenues. Depending on local conditions, institutions would design appropriate schemes and programmes to achieve above goals. In some areas, institutions may be able to attract and retain students by providing mid-day meals and uniforms, in other by offering a wider variety of extra-curricular activities, in still other areas by arranging for remedial classes, and so on. Institutions would be able to use funds to meet specific local needs without any directives from the government or urgings from the communities. It will be in their self-interest to discover and implement the most suitable schemes and programmes for their students.

A. Universal voucher system

In this system, government provides vouchers of a specified sum to all individuals in a given age group.  Individuals would then find institutions that meet their needs and aspirations. Netherlands and Chile have a very similar system.

B. Incentive voucher system

Generally vouchers cover the actual cost of providing education but do not pay for the opportunity cost of students' time or any extra incentive to acquire education. We in India, say, want to give special emphasis to girl education. An incentive voucher system would not only cover the cost of education but also provide some     money to girls in the specified age group/educational level. A larger amount of voucher would be given to these  girls and a part of the amount would then be given back by the institution to the girls as an incentive to enrol and attend school. The government could specify criteria (such as attendance, performance) that the girls should     meet in order to get the money.

C. Income-constrained voucher system

The government would specify a family income level above which students could not receive vouchers, in other words, no financial support for richer students for education. It could be a single cut-off (say, no voucher for the family income above Rs. 30,000 per year), or it could be a graduated cut-off (60 percent of the standard     voucher amount for the family income above Rs. 20,000 per year, 30 percent for above Rs. 30,000, and none above Rs. 40,000 per year).

What should be the actual amount of the voucher for elementary education? The expenditure estimates vary from survey to survey. Average cost of sending a child to a government primary school: (Rs/year) Source: PROBE survey (sub-sample of 831 children enrolled in government primary schools). Household Expenditures on Elementary/Primary Education (Rs per Student) Panchmuchi (1990) (Elementary Education, 1988-89) Government Private Aided Private Unaided Private Unrecognised Maharastra 382.25 4033.14 2452.18 2467.80 Karnataka 1194.46 1650.32 2530.39 Rajasthan 809.69 2185.46 7514.68 NCAER (1994) (Elementary Education, 1992-93) All Schools Bihar 290 Kerala 773 Delhi 1029 UNDP (1999) (Elementary Education, 1994-95) All Schools One District in UP 120 One District in Kerala 550 PROBE (1999) (Primary Education, 1996-97) All India Government Private 318 940 Tilak (1996) based on NSSO 1986-87 (Primary Education) All Schools India Rural Urban Total 84 117 109 Source: Manabi Majumdar (1999, p. 320) Average Cost of Sending a Child to School (Rs/year at constant 1996-7 prices) Primary Level NSS estimate, 1986-7* 212 PROBE estimate, 1996 318 Elementary Level NCAER estimate, 1994 478 *Excluding clothing expenses Source: NSSO, 1993; NCAER, 1996; PROBE survey

The government expenditure of elementary education varies from state to state in the rage of Rs. 500 per student per year to Rs. 1400. Private expenditures in government and government aided schools seem to rage from Rs. 300 to Rs. 1000. So the voucher amount could be in rage of Rs. 800 to Rs. 2400. Certainly additional work is required for more informed estimates of the voucher amount.

In conclusion, the three reform areas-liberalisation, accountability, and empowerment-provide a framework within which to design programmes and policies to give India an education system that could meet the needs of the 21st century. First and foremost, liberalise the whole education system from primary to higher. Allow free entry and exit; bury the license-permit raj education. Remove the archaic socialist restriction that education institution cannot operate for monetary benefit. Let the profit motive that produces all other goods and services so efficiently and abundantly play its role in the market for education.

Make all schools, colleges, and universities autonomous; let the teachers, parents, and local administrators have all the powers to make decisions. Link government grants to the performance of institutions. Incentives matter. We cannot make our education system wholly dependent on sudden emergence of crores of enlightened and wise teachers, professors, and administrators. Those who perform, must be rewarded, those who do not, must find a better use of their talents.

Design innovative ways to finance education of those who cannot afford it: Scholarships, voucher systems, loans. Encourage self-regulation of this autonomous education system by promoting independent certification, accreditation, and examination agencies. Transfer government institutions to trusts, societies, and private voluntary organisations (mislabelled as NGOs).

It must be acknowledged that many of our experts, including the Nobel Laureate Amartya Sen, argue for a larger role of the government in provision, financing, and management of the education system. They point out that government interventions have been critical in all countries that have high rates of literacy. That may be true. But we should learn from not what they did in the past but what they are doing today. Many of these countries are moving away from the government control of education through independent and autonomous institutions and voucher systems. We need not first repeat their mistake and then follow them on the new reform path. We must learn from their historic mistake and create our own path.

I look forward to a day when Sachin Tendulkar and Aiswariya Rai are paid crore rupees in an ad for a new college. If sellers of frivolous soft drinks can spend vast sums on ads, provides of a service as essential as education must at least be able to match. Only then we would be certain that we are not under-investing in education.

References Aggarwal, Yash, and Kusum K. Premi, Reforming School Education, Vikas Publishing House, 1998. Desai, Ashok V. Self-Regulation in the Civil Society, Centre for Civil Society, 1998. Duraisamy, P. and T. P. Subramanian, The Relative Effectiveness of Public and Private Higher Secondary Schools in an Urban Centre in India," Journal of Educational Planning and Administration, January 1999. Education and Urban Society, Special Issue on Charter Schools and Urban Education, August 1999. Filmer, Deon, and Lant Pritchett, "Educational Enrolment and Attainment in India: Household Wealth, Gender, Village and State Effects," Journal of Educational Planning and Administration, April 1999. Kothari, V. N. "State versus Market in Financing Education in India," Journal of Educational Planning and Administration, July 1999. Majumdar, Manabi, "Financing of Basic Education in India: Private Solution to Public Deficiency?" Journal of Educational Planning and Administration, July 1999. NCAER. "Non-Enrolment, Drop-Out and Private Expenditure on Elementary Education: A Comparison across States and Population Groups, 1994. NSSO. "Participation in Education," (NSS 42nd round), 1991 and 1993. PROBE Team, Public Report on Basic Education, Oxford University Press, 1999. Rao, K. Sudha, Management of Autonomy in Autonomous Colleges, Vikas Publishers, 1999. Rao, K. Sudha, George Matthew, and Sudhir K. Samantray, Autonomous and Non-Autonomous Colleges: Selected Case Studies, Vikas Publishers, 1999. Shah, Parth J. "New Education Policy: Choice and Competition," in Agenda for Change, March 1998. Shah, Parth J. "The Full Truth in Prof Sen's Half Truth," Business Standard, January 3, 1998. Swaminathan, Madhura, and Vikas Rawal, "Primary Education for All," in India Development Report, 1999-2000. Tilak, J.B.G. "How Free is 'Free' Primary Education in India?" Economic and Political Weekly, February 3 and 10, 1996. Tilak, J.B.G. "Education: Towards Improving Equity and Efficiency," in Vijay L. Kelkar and V.V. Bhanoji Rao edited India: Development Policy Alternatives, 1995. Varghese, N.V. "Private Schools in India: Presumptions and Previsions," in R.P. Singh edited Private Initiatives and Public Policy in Education, 1993.