2008 Research Consortium on Educational Outcomes and Poverty WP08/18 RECOUP Working Paper No.

18 Public Expenditure on Education in India: Recent Trends and Outcomes Anuradha De and Tanuka Endow Collaborative Research and Dissemination1 (CORD), India Abstract It is widely accepted that there is an acute shortage of resources in the educat ion sector in India. Economic reforms and associated requirements of fiscal discipline have ag gravated the situation. By contrast, however, official sources claim that significant progres s has been made in financing education. This paper examines whether, and in what ways, this is so. It analyses major trends in public financing of education in India, including expenditures by the central government, state governments, other local bodies and the NGO sector. Foreign aid, which is transferred primarily through central government budgets, is also included. The paper examines the level and composition of public expenditure on education and the mechanisms of resource sharing, allocation and utilization, in aggregate as well as separately for the centre and the states. It finds that while expenditure in real terms increased d uring the 1990s it has stagnated since then. As a proportion of GDP the share of public expenditure on education has been less than 4 per cent. But there have been major changes in the composition and m odalities of expenditure. Initially, education was the responsibility of individual states, b ut in 1976 it became the joint responsibility of both central and state governments. The analysis finds t hat the centre has been playing an increasingly important role in state education finance. Centrally spo nsored schemes, which are partly funded by external aid, have been a critical part of centre-tostate transfers. Expenditure trends in seven states are studied to explore the possible impact of expenditure on education outcomes. It indicates that for the less developed states recent chang es in education expenditure have improved access, but retention and learning achievements remain very low. Correspondence: CORD, G-18/1 Nizamuddin West, New Delhi 110 013, India, Tel: +91 11 24356085 Email: [email protected] Keywords: education, financing education, India, education outcomes. Acknowledgements:

Valuable research assistance was provided by Aftab Alam, Parwez Hussain, Shraban i Mukherjee and Debdulal Thakur from Collaborative Research and Dissemination. The authors are grateful to Dr. Praveen Jha for providing valuable guidance in conception of the paper, to Claire 1 Collaborative Research and Dissemination is the lead research organisation in India working with the Research Consortium on Educational Outcomes and Poverty (RECOUP), a DFID support ed research consortium. 1

Noronha for her insightful comments at different stages and to Professor Chris C olclough for his meticulous reading of several versions of the paper and his detailed comments an d suggestions. An earlier version of this paper was presented at the UKFIET conference Going for Growth? School, Community, Economy, Nation , 11 13 September 2007, Oxford. This paper form s part of the Research Consortium on Educational Outcomes and Poverty (RECOUP). Neither DF ID nor any of the partner institutions are responsible for any of the views expressed here. 2

1. Financing Education: Policy In India over the last fifty years there have been major changes in the level of financing of education and in priorities attached to different sectors within it. Education a t the primary level had been rather neglected by the early planners, who focused more on higher and tech nical education. The Constitution of India, under the original Article 45, had directed the State to provide, within a period of ten years from the commencement of this Constitution, for free and compulsory education for all children until they complete the age of fourteen years. But the goal of universal elementary education (UEE) was not addressed with urgency for four long decades possibly be cause Article 45 was placed in Directive Principles of State Policy and, therefore, was seen as n ot being justiciable.2 This situation changed as a result of the introduction of the Constitutional 83r d Amendment Bill in the Parliament (1997) and eventually the passing of The Constitution (Eighty-Sixth Am endment) Act, 2002 more than fifty years after India s independence. The relative neglect of elementary education has resulted in low literacy rates among adults in India and in majority of the poor children remaining out-of-school, being dep rived of even elementary education. There has been rapid improvement in school participation i n recent years but a national household survey commissioned by Government of India in 2005 (SRI, 2005 ) still reports 7.8 million out-of-school children in primary school going age. The Indian middl e and upper middle classes, on the other hand, benefited from subsidized higher education and have succeeded in creating a pool of highly knowledge able and skilled workers. Since the 1990s, however, d ue to the twin factors of awareness of these issues at the domestic level as well as pressure s temming from structural adjustment policies3 (SAP) the Indian policymakers have sought to correct the ba lance and there has been a concerted effort towards taking elementary education to all children, eve n those living in remote regions. We present in Table 1 an overview of education policies related to education fin ance as it evolved since India attained independence. Although several committees and commi ssions were appointed from time to time to deal with various issues, the education policy wa s shaped primarily by the Kothari Commission and the National Education Policy. After the UPA (United Progressive Alliance) government came into power in 2004, the Central Advisory Board of Educ ation (CABE) Committee was appointed to investigate the universalization of secondary educati on. While these committees had much to say about structure of the education system and prioritie s within it, their

recommendations regarding the level of education finance were somewhat similar. In 1966, the Kothari Commission had recommended that the public expenditure on education shou ld reach the level of 6 per cent of GNP by 1986. Subsequently, in 1996, the Saikia Committee examined the financial, among other, implications of the proposal to make free and compulsory education a fundamental right. It reiterated the need for an expenditure of 6 per cent of GN P on education with 50 per cent of it earmarked for primary education (an additional finance of Rs.4000 0 crores in next five years). In 1999, an Expert Group headed by Tapas Majumdar, made estimates for ad ditional fund requirements for UEE it was in the range of 137000 crores over the following 10 years (GOI, 1999). National Common Minimum Programme and CABE committee also had similar recommenda tions. However, policy statements and implementation do not always go together. Actual expenditures in the education sector have fallen far short of these targets (section 3). These r ecommendations repeatedly emphasize the need for higher investment in education and the importa nce of different levels of government in its financing. 2 In 1993 Supreme Court granted all children a fundamental right to free and comp ulsory education until they complete the age of fourteen years and stated that this right flows from Article 21 i.e. Right to Life. In 1996 the Government constituted the Saikia Committee of State Education Ministers (GO I, 1997) which recommended that the Constitution of India should be amended to make the right to free elementary education up to the 14 years of age, a fundamental right. 3 Following a severe balance of payment crisis, India took a $ 500 million struc tural adjustment loan from the World Bank in 1991. The loan was intended to help India tide over the BOP crisis and also to help it to liberalize its economy through economic reforms in the domestic economy as well as in the trade sector. The fiscal discipline that had to be followed by the centre and the states as a resu lt of SAP led to pruning of expenditure in social sectors like education. However, the renewed emphasis on e lementary education the world over affected India as well, and there was increased focus on education at the e lementary level, often at the cost of higher and technical education). 3

Table 1: Overview of Education Policies related to Financing of Education in Ind ia Education Policy / Committee Year Recommendations Kher Committee 1948-49 A fixed percentage of Central (10 per cent) and Provincial (20 per cent) revenues should be earmarked for education and that around 70 per cent of the total expenditure on education shou ld come from the local bodies and provinces Kothari Commission 1964-66 Public expenditure on education should reach the level of 6 per cent of GNP by 1 986 Vocationalization of secondary education Strengthening of centres of advanced study and setting up of small number of maj or universities of international standard. National Education Policy 1968 Investment on education to be gradually increased to reach a level of six per ce nt of national income as early as possible. Focus on science & technology and agriculture Provision of food and effective education at primary level (on a free and compul sory basis) Equality in education for rich and poor: common 10+2+3 education structure throu ghout India and eventually free schooling till class 10. Secondary Education Commission 1972 .. to assume certain direct responsibility for reorganization of secondary education and give financial aid for the purpose. Encourage private contribution through tax exemptions (income tax, property tax and custom duties) Industrial education cess should be levied for furtherance of Technical and voca tional education at secondary stage. 42nd Constitutional amendment 1976 Education transferred from list to concurrent list (School education under jurisdiction of both, the Centre

and the State). National Education Policy 1986 (with revisions in 1992) Resource support for implementing programmes of educational transformation, redu cing disparities, universalisation of elementary education, adult literacy, scientific and technol ogical research, etc. will be provided. For this actual requirements will be computed at regular inter vals and outlay on education will be stepped up so that more than six per cent of national income i s allocated from eighth plan onwards. While the role and responsibility of the States in regard to education will rema in essentially unchanged, the Union Government would accept a larger responsibility to reinforc e the national and integrative character of education, to maintain quality and standards (including those of the teaching profession at all levels) and to study and monitor the educational requirements of the country. Additional resources to be raised by mobilizing donations, asking beneficiary co mmunities to maintain school buildings, raise fees at higher levels of education and effectin g savings through efficient use of resources. 73rd and 74th constitutional amendment 1992 Statutory recognition of local governments, and inclusion of school educati on in the list of its responsibilities. Local bodies to play an important role in financing and implem enting education programmes. Saikia Committee 1996 Need for an expenditure of 6 per cent of GNP on education with 50 per cent of it earmarked for primary education. Recommended additional expenditure of Rs. 40000/ crores over next fiv e years on elementary education. Tapas Majumdar Committee 1999 Estimated additional fund requirements for UEE it was in the range of 13700 0 crores over the following 10 years. 86th Constitutional Ammendment 2002 Provide free and compulsory education of children between age 6 to 14 years , and provision of early childhood care and education for children below six years.

National Common Minimum Programme of present UPA Government 2004 Raise public spending in education to at least 6 per cent of the GDP with at lea st half this amount being spent on primary and secondary sectors. This will be done in a phased mann er. A Cess of two per cent on all central taxes to finance the commitment to univers alize access to quality basic education. A national cooked nutritious mid-may-meal scheme, funded mainly by the Central G overnment, will be introduced in primary and secondary school. The Integrated Child Development Services (ICDS) scheme will be universalized to provide a functional Anganwadi in every settlement and ensure full coverage for all childr en. all northeastern States will be given special assistance to upgrade and expand i nfrastructure. CABE Committee 2006 The additional financial requirement for universalising secondary education as p er cent of GDP works out to be around 0.18 per cent in 2003-04 and to 0.86 per cent 2019-20. With 6 per cent of GDP earmarked for education, the shares of elementary, second ary and higher secondary (as % of GDP) will be 3, 2 and 1 respectively. Source: From MHRD website; Relevant committee reports Moreover, education development in general and attainment of UEE in particular, depends not only on the quantum but also on the composition of expenditure on education. The process of allocation and disbursement also has a major impact on educational development. In order to understand the present state of public financing of education in India, we need to take a comprehensive look at the Indian fiscal system and the players involved. 4

This paper examines public financing of education in India in this context. It i s divided into six sections. There is an initial discussion of sources of finance and problems of estimation and interpretation of data. The trend and composition of aggregate public expenditur e in India is examined in the second section. The third section focuses on Centre-State relati ons in education finance and has contrasted their expenditure patterns. The mechanism of flow of funds and the role of centrally sponsored schemes are also discussed. Expenditure trends in several st ates and their possible impact on education outcomes are looked at in the fourth section. In th e next section the two major problems in public expenditure inadequacy of resources and inefficiency in allocation and utilization are analysed. The role played by foreign aid has been discussed in t his context. In the concluding section the recent changes in level of public expenditure and their i mplementation are reviewed. 2. Sources of Finance There are several important sources of education finance the public sector which includes expenditure by the central government, the state governments and the local bodie s, foreign aid which is transferred primarily through central government budgets and the private sect or financed largely by individual households and, to a smaller extent, non-profit and for profit privat e sector. The focus in this paper has been essentially on public expenditure on education by the centre and the states. Initially education was the responsibility of individual states, but in 1976 it was placed in the Concurrent List (denoting joint responsibility of both central and state governm ents). The total budget expenditure on education and the shares borne by Centre and States respectively are shown in Table A1 in Annexure While the States still bear the lion s share of expenses on educati on, their share in the total has been declining after 2000-01 (from 87.7 per cent in 2000-01 to 75.7 pe r cent in 2005-06 (Budget estimate), while Centre s share has shown a jump (from 12.3 per cent to 24 per cent during the same period) (Graph 1). Graph 1. Share of the Centre and the States in Education Expenditure 0 10 20 30 40 50 60 70 80 90

100 199293 199394 199495 199596 199697 199798 199899 19992000 200001 200102 200203 200304 200405 (RE) 200506 (BE) per cent Share of centre s Expenditure Share of States Expenditure It is evident that the expenditure patterns of both the central government as we ll as the state governments have to be examined to get an idea of the total expenditure. 5

Following two constitutional4 amendments in 1993, more power has been invested w ith rural and urban local bodies making elementary education a responsibility of these bod ies. These local bodies are largely funded by the respective state governments but they also spen d resources raised at their level. While in rural areas own resources raised by local bodies is a smal l proportion of their total revenue (3.6 per cent in 1997-8), it is much higher in urban areas (nearly two-thirds) (Rao, 2000). However the resources and expenditures of local bodies vary from state to state (Andhra Pradesh and Maharastra accounting for a major share) and though it is possible t o estimate their budgeted expenditure on education, it is not possible to estimate the extent to which this is funded from their own resources. So their expenditure has not been included in this exp enditure analysis. External aid is an important source of education financing. In relation to the t otal magnitude of education finance this had been insignificant before the introduction of econ omic reforms in the 1990s. Its role had changed since then. Earlier focused on technical and vocatio nal education, external resources are now increasingly directed at elementary education.

2.1 Problems of Estimation With several players involved in financing education and the various channels of transfer of funds, estimating total expenditure becomes quite complicated. Firstly it is dif ficult to ascertain the expenditure incurred on education directly because the expenditure for education by Department of Education is given under several major heads of account which includes general ed ucation , technical education , sports and youth services and art and culture . So expenditure on specific heads has to be separated to estimate expenditure on education department. Secon dly this expenditure does not take into account expenditure on education by other departments. For in stance, the department of Rural Development gives money for building schools, which should b e included as expenditure for education. Both in the central and the state governments, quite a large amount of expenditure on education is incurred by departments other than the department of education, and their proportion has increased over the years. These departments include several minis tries/ departments like Department of Women and Child Welfare, Ministry of Tribal Affairs and Minis try of Social Justice and Empowerment. Table 2: Expenditure on Education by Education and other Departments, Rs. Cr Year Total expenditure

on Education Expenditure by Other Departments on education Expenditure by Other Dept. as a % of Total Expenditure on education 1971-72 1007.8 85.3 8.5 1981-82 4288.8 498.7 11.2 1991-92 22393.7 3636.1 16.2 1996-97 43896.5 7524.9 17.1 2001-02 80505.9 15018.0 18.7 2003-04 95535.5 18265.2 19.1 Source : Financing Education in India, NUEPA , J B Tilak ; Analysis of Budgeted Expenditure on Education, Ministry of HRD, Govt. of India- various issues. Table 2 shows the increasing importance of other departments of the Central and State governments in budgetary expenditure on Education. It has increased faster than total expenditure and the share of other departments has increased from 8.5 per cent in early seventie s to 16 per cent in early nineties and around 20 per cent by the turn of the century. 4 By the 73rd and 74th Amendments to the Constitution, bodies of local self-gover nment Panchayati Raj bodies for rural and Municipal bodies for urban areas, respectively were accorde d Constitutional status, in 1993. The Eleventh and Twelfth Schedules of the Constitution provide illustrativ e lists of items which may, by law, be devolved on these local bodies. The Eleventh Schedule lists Education up to the Secondary level, Vocational Education, and Adult and Non-Formal Education, among others, for devo lution to Panchayati Raj bodies. Source: website of higher education department of Ministry of Human Devel opment, India (Education.nic.in) 6

2.2 Components of Expenditure While the aggregate expenditure reflects the priority accorded to education, in the Indian context it is important to distinguish between Plan and Nonplan expenditure. Pla n expenditure is that part of total budget expenditure, which is meant for financing the schemes and p rogrammes especially framed under the current Five-year plan or the unfinished tasks of th e previous Plans. So the Plan expenditure indicates the direction of changes in the education sector. Nonplan expenditure is the expenditure on operating and maintaining existing education infrastructur e. So at the end of a five year plan, the recurring parts of Plan expenditure on different programmes or schemes become part of Nonplan expenditure. So Nonplan expenditure is expected to increase stea dily over the years.5 The scope for decreasing this expenditure is very limited, as it involves mainta ining the stock of education infrastructure which has been determined by the policies in the past y ears and while savings through introduction of efficiency measures are possible that would be a one-time savings. Another distinction to note is between expenditure on Revenue account, and expen diture on Capital account. Expenditure on Revenue accounts constitutes the bulk of the budget expe nditure on education in India and very little is spent on the Capital account. But this doe s not imply that there is little or no asset creation in education. One of the main reasons for low expend iture on Capital accounts is that the entire grants-in-aid, including grants for capital works is booked under revenue account and not under capital account. Secondly expenditure on construction acti vity is often shown under the budget heads of other departments. The government also saves some of t he construction expenditure as often village panchayats donate land for construction of school b uildings and villagers provide voluntary labour and locally available construction material, and also m any states have private aided schools/colleges where initial land and buildings are provided by private individuals/organizations and only the recurring costs are covered by government aid6. 3. Public Expenditure on Education in India There have been several studies on education expenditure in recent times which p rovide valuable inputs to this paper7. Though the data presented in different papers br oadly agree with each other, there are minor differences in the estimates. So in this paper we have pr imarily used data from Analysis of Budgeted Expenditure on Education (Ministry of Human Resource Develo pment). The data presented in these reports take into account the expenditure incurred on ed ucation by all departments in addition to that by the education department, and is the standard source used for

analyzing public expenditure on education8. However much of the data used in thi s paper are expenditures by education department of the relevant governments on revenue acco unt. We next take a look at the budget expenditure on education, its share in GDP and the PlanNonplan distribution of public expenditure on education. It is seen from Table 3 that while expenditure in current prices has been growin g at the CAGR of 13.5 per cent p.a. for the period 1990-91 to 2003-049, the rate of growth has slowed down in the present decade10. Moreover expenditure in constant prices shows a much lower CAG R of only 6.5 per cent for the same period. Though the expenditure has almost doubled between 1990 -01 to 2000-01, it had stagnated and even shown a decline since then11 (Graph 2). 5 In the education sector, salary of teachers and administrative staff (other th an those recruited under Plan schemes), expenditure on repair and maintenance of schools, expenditure on repla cement of non-functional teaching equipment and similar items are Nonplan expenditure. Expenditure on con struction of new schools, additional classrooms, new toilets, salary of additional teachers and nonteachin g staff etc come under Plan expenditure. 6 Budgetary resources for education (1951-52 to 1993-94): document available at we bsite of MHRD 7 Jha, Das, Mohanty and Jha (2006), Tilak (2003), Srivastava, (2005), Dev and Mo oij (2002), Bashir, S (2000) 8 These estimates too suffer from some overestimation. For example grants made b y the Central Ministry to the State budgets are counted as expenditure of centre as well as states and so are double-counted (Jha et al, 2006). In spite of certain problems, these reports have been invaluable for studying ed ucation finance. Otherwise it would have been a very time consuming exercise to get details of education expen diture from all departments as it requires scrutiny of expenditure heads of several line ministries/departments , and totaling them. 9 A major chunk of the increase is the result of pay scale revisions in public s ector. 10 It even showed a decrease between 2000-01 to 2001-02. 11 The expenditure in 2004-5 may register as increase if the revised estimates c an be achieved. 7

In per capita terms, expenditure at constant prices shows a sharper decline. So education expenditure has stagnated and even declined in spite of the rhetoric on the part of the Government. This is also borne out by the trend in the share of public education expenditure in GDP (Graph 3 and Table A2). The share was around 4 per cent in 1990 but declined since then. It r ecovered to over 4 per cent only around the year 2000, but has been since been on a decline and nowhere near the oft quoted target of 6 per cent. As a proportion of total revenue expenditure, a decline in the present decade is observed - at the turn of the century more than 14 per cent was spent on educati on, but around 12 per cent is spent in the present decade. Table 3: Changes in Public Expenditure on Education in India Year Govt. expenditure on education by all Depts. in current prices (in Rs.crores) Per capita expenditure at current prices (in Rs.) Govt. expenditure on education by all Depts. In constant prices (in Rs Crores.)* Per capita expenditure at constant prices (in Rs.)* 1990-91 20491 242.12 27793 1995-96 38178 408.66 31396 2000-01 82486 814.77 52978 2003-04 89079 840.35 50642 CAGR (1990-1 to 2003-04) 13.5 6.5 CAGR (19992000 to 20034) 4.46 -0.41

328.40 336.07 523.30 477.74

*Price (1993-94) = 100 Source: Analysis of Budgeted Expenditure on Education, Ministry of HRD, Govt. of India-various issues. Note: GDP figures taken from National Income Statistics published by CSO Graph 2: Public Expenditure on Education by all Departments (in constant prices) 0 10000 20000 30000 40000 50000 60000

199091 199192 199293 199394 199495 199596 199697 199798 199899 199900 200001 200102 200203 200304 Rs. Crore 8

Graph 3: Share of Public Expenditure on Education in GDP (%) 3 3.5 4 4.5 5 199091 199192 199293 199394 199495 199596 199697 199798 199899 199900 200001 200102 200203 200304 per cent Graph 3: Share of Public Expenditure on Education in GDP (%) 3 3.5 4 4.5 5 199091 199192 199293 199394 199495 199596 199697 199798 199899

199900 200001 200102 200203 200304 per cent So while aggregate expenditure on education is increasing, the rate of increase is not necessarily keeping pace with budgetary expenditure, GDP growth, or rate of infl ation. Along with the trend in aggregate public expenditure on education, it is importa nt to look at the trends of Plan and Nonplan expenditures within education. This is because, a s already mentioned, Nonplan expenditure largely reflects the recurring expenses in the education sec tor while the Plan expenditure reflects new schemes and projects started by the government. Graph 4: Share of Plan Expenditure in Total Education Expenditure, % 0 5 10 15 20 25 30 35 40 45 50 199091 199192 199293 199394 199495 199596 199697 199798 199899 199900 200001 200102

200203 200304 In the recent years the changes in Plan expenditure show a somewhat different pi cture from changes in aggregate expenditure. Plan expenditure had increased slowly between 1996-7 and 2000-1 but since then has been increasing rapidly (Table A3 and Graph 4). This reflects the introduction of different schemes in elementary education sector. On the other hand aggregate ex penditure showed a reverse trend. As a result the proportion of Plan expenditure has increased. This brings us to the trend of different components of education expenditure lik e elementary, higher education, technical education and so on. Table 4 gives the breakdown of revenue expenditure by education departments on different sectors within it for selected years12. It shows that the 12 In the absence of data on education expenditure by all departments, we focus on composition of expenditure by the various education ministries/departments. This is a major limitation sinc e the contributions of other departments are quite significant in certain states. 9

proportions have remained similar over the last fifteen years: elementary educat ion is seen to increase gradually to receive around half of the total resources, secondary education rec eived around a third, the proportion spent on higher education fluctuated between 11 per cent to 15 pe r cent and technical education received a low and declining proportion. Expenditure on adult educatio n is small and declining, indicating change in priorities. This emphasis on elementary educatio n, together with low priority of education in aggregate public expenditure, has constrained the growt h of other sectors of education. Table 4: Inter Sectoral Allocation of Expenditure on Education in India(per cent ) Year Elementar y Adult Secondary Higher Technical Total expenditure in Rs crores 1990-91 46.27 1.59 32.17 13.45 4.38 17193.66 1995-96 48.30 0.82 32.80 12.29 4.10 31506.58 2000-01 47.61 0.36 31.59 14.71 4.04 62498.02 2004-05(RE) 51.45 0.43 30.13 11.67 3.82 85686.67 Source : Analysis of Budgeted Expenditure on Education, Ministry of HRD, Govt. o f India-various issues. Note: RE: Revised Estimates. Graph 5: Expenditure on Elementary Education by Education Departments (in constant prices) 0 500000 1000000 1500000 2000000 2500000 1990-1 1991-2 1992-3 1993-4 1994-5 1995-6 1996-7 1997-8 1998-9 19992000 2000-1 2001-2 2002-3 2003-4 Rs. crores Graph 5 reveals a remarkable situation. While allocations to the elementary sect or had indeed been increasing in the nineties, it has almost stagnated since then. Yet the Tent h Plan document (2002-2007) states The Central Government has introduced the 93rd Constitution Am endment Bill, 2001 for enacting the Fundamental Right to Free and Compulsory Education for chi ldren in the age group of 6-14 years. The enactment of a Central legislation would result in adeq uate provisioning of public resources for improving the accessibility of children to schools, quality up-gradation, and mitigating the costs of school attendance. The distance between policy statements and their implementations can be seen cle

arly. Different data sources (selected Education Statistics, DISE) bear evidence of ra pid increase in enrolment at elementary level. With near stagnation in aggregate expenditure it would be difficult to support initiatives to improve quality or mitigate costs of households. 4. Centre-State Relation 4.1 Mechanism of the Flow of Funds It will be useful at this point to discuss the actual flow of funds in the educa tion sector. As mentioned earlier, the expenditure on education in India comes from both the Cen tre and the States. The bulk of the tax revenue is collected by the centre while states have the mai n responsibility of 10

maintaining and developing the education sector. 13A part of the resource gap fa cing the state governments is met through transfers mandated by the Finance Commission from the central Government to the states. This includes States share in central taxes, non-plan g rants and some conditional grants. States also receive funds from the Planning Commission: dire ctly in the form of Central Assistance to States (used to support state plans) and indirectly throug h the central ministries in the form of Centrally Sponsored Schemes (CSS). The CSS funds reaches the dist rict level implementing society either through state budgets and relevant state ministry/de partments or via a direct route to DRDA (District Rural Development Agencies) or SIS (State Impleme nting Societies). Apart from the central funds, the States fund education directly. Its magnitude also depends on own tax revenues, own borrowings and other capital receipts. 4.2 Public Expenditure on Education by the Centre and the States Analysing aggregate education expenditure is of limited value in India. Disaggre gated analysis is needed because the Centre and the States have very different financial powers an d different sets of priorities and constraints.14 Table 5: Plan and Nonplan Expenditure on Education in Central and State Sectors Year Plan Expenditure Nonplan expenditure Share of the centre (per cent) Share of the states (per cent) Total (Rs. crore) Share of the centre (per cent) Share of the states (per cent) Total (Rs. crore) 1992-93 42.18 57.82 4009.00 5.53 94.47 21021.30 1995-96 44.99 55.01 8383.10 5.97 94.03 29795.00 1999-2000 48.22 51.78 13864.91 6.93 93.07 60951.18 2003-4 63.08 36.92 19092.33 7.23 92.77 69986.92 Source: Calculated from HRD, Govt. of Indiavarious issues. Analysis of Budgeted Expenditure on Education , Ministry of

Table 5 and Table A4 in the Annexure, shows that Centre s share in Plan expenditur e is very high and has increased from 42 per cent in 1992-93 to 63 per cent in 2003-04. On

the other hand, states have consistently accounted for around 92-94 per cent of non-plan expendi ture. This implies that while states are responsible for the bulk of expenditure centre has come to play a dominant role in shaping the country s education system. It has been noted earlier that the larg er part of education expenditure is Nonplan. The Centre has always played an insignificant part in it , accounting for 5 to 7 per cent of the total. But its role has assumed increasing importance in Plan ex penditure and calls for a closer look. The share of the Centre in Plan expenditure fluctuated between 25 to 30 per cent from the inception of Indian planning till the Fourth plan, a period when education was a state subject (Tilak 2003). Even after inclusion of education in the concurrent list in 1976, there w as no immediate increase in the share of Centre in plan expenditure it showed a decrease rather than increase15. However since then the Centre s share has been increasing over the years from arou nd 40 per cent in the early 1990s, to more than 60 per cent in the recent years. In absolute te rms the increase is particularly rapid since 2000-01. This implies an important shift in the role of Central government vis--vis state governments Plan expenditure through which changes are implemented in the sector is gradually being dominated by the Centre a sharp contrast from the government s professed emphasis on decentralization. State governments are also responsible for a large part of Plan expenditure. But in contrast to the Centre s expenditure pattern, absolute Plan expenditure by States has been flu ctuating around the same level since 1998-99, and so is accounting for a decreasing share of plan ex penditure. The Plan 13 In the early nineties net devolutions and transfers from the Centre contribut ed to more than a third of the total expenditure by the States. Over the decade this proportion has declined to a lit tle less than 30 per cent to meet the needs of fiscal discipline on the part of the Centre (Ramji et al, 2001) 14 In addition, each state has its own history of education development and its own sets of priorities and limitations. 15 For details see Tilak (2003). 11

expenditure by the States show a sharp increase for 2004-05 and 2005-06 (revised estimates and budgetary estimates, respectively)(Table A4) which may be indicative of more dec entralization in the education sector. Nonplan expenditures are almost exclusively the domain of the state governments. As this involves maintenance of existing infrastructure any decrease in its level would adversely impact school quality. But since 1999-2000, state Nonplan expenditure too has stagnated at the same level. The stagnation in expenditures of the state governments is clearly seen in table s A5 where education expenditures in constant prices are presented. It can be seen that in real terms , expenditure by states has decreased from the level reached in the year 1999-2000. Graphs 6 A and B and Table A6 compare the intra-sectoral allocation in expenditu re by education departments of the central and the state governments. Changes in their priorities are also noticed here. In the past the state governments focused on school education elem entary and secondary while Centre played a more important role in university and higher edu cation and technical education. But the central budget in the last fifteen years shows that the increase in expenditure has primarily come in the area of elementary education. In 1990-91, only 13.7 per cent was spent on this sector. But it increased rapidly since then and after 1995-96, the share of this sector exceeded 40 per cent and crossed the 50 per cent mark in 2003- 04. The increase in the share of elementary education has mainly been at the expense of secondary education. Afte r 2001 the share of university and higher education (the sector for which Centre had the primary res ponsibility) also declined sharply. The share of technical education has remained more or less unc hanged, varying within a narrow band (around 13-15 per cent). So the inter-sectoral budget alloc ation of the central government reflects the policy shift of priorities of public funding - towards e lementary education and away from the higher levels16. Graph 6 A 16 Tilak (2006a) has used recent union budget data to point out that the large i ncreases in budgetary expenditure in the last two years are confined primarily to Plan expenditures, and to elementar y education sector. This increase has been made possible by the collection from the introduction of education cess , three years ago. This is a separate dedicated non-relapsable fund for elementary education. These extra res ources are spent almost entirely on mid-day meal programme (which is a nutritional support programme run jointly by the central and state

government and entails serving of a hot cooked midday meal to all children enrol led in elementary stage) and Sarva Shiksha Abhiyan, a programme which attempts to combine all programmes for universalisation of elementary education in all states. 12

Graph 6 B The state budgets on the other hand do not show any shift. They have the respons ibility to maintain the existing education infrastructure and do not have the flexibility o f the central government. While about half of budgetary expenditure is allocated to elementary education, around a third goes to secondary education and the remaining largely to higher educatio n. Not much goes to technical education most states have encouraged private investment in this secto r in the recent years. 4.3 Centrally Sponsored Schemes (CSS) The difference in trends in public expenditure by the Centre and the States over the past fifteen years highlights the increasingly important role played by the Centre in States f inance, particularly in elementary education. A large proportion of transfers from the Centre are allott ed to the States for implementing Centrally Sponsored Schemes (CSS). These schemes are formulated by the Centre which also funds a major part of their implementing costs. States, which are the implementing agencies, fund the remaining portion. As a mode of transfer of resources direct central assistance is advantageous to the states from autonomy point of view. But the states have show n preference for CSS as it is given as grants and not loans which reduce their fiscal burden. Some prominent CSS in elementary education were designed after the introduction of New Economic Policy. In the nineties the major schemes in operation were Operation B lackboard (OB), Non-formal Education (NFE) and Teacher Education (TE), Midday Meal (MDM), Scheme for Free education for girls and District Primary Education Programme (DPEP)17. A few sma ller schemes Educational Technology (ET), Environmental Orientation of School education and I ntegrated Education of Disabled Children (IEDC) were for both elementary and secondary edu cation. The earlier schemes were targeted towards improving schooling infrastructure. The la ter schemes were incentive based. Centrally sponsored schemes have become critical players in education finance. I n a paper on Central transfers to States and centrally sponsored schemes prepared for discussio n on National 17 Non-Formal Education (NFE), launched in 1977-78, aimed at bringing out-of-sch ool children in the age group 6-14 years into the fold of education, targeted educationally backward states as well as hilly, tribal, desert areas and urban slums. Operation Blackboard (OB) was launched in 1987-88 for improving human and physical

resources in primary schools of the country and was extended in 1993-94. The Tea cher Education scheme, introduced in 1987 to create a suitable institutional infrastructure as well as an academic and technical resource base, has now been revamped with more emphasis on improving quality of teacher t raining institutions in partnership with the States. The DPEP, which was district focused, was a combina tion of several schemes including the existing ones. It started in 1994 and was a major initiative for u niversalizing primary education. Its major aims include: providing access to primary education for all children, redu cing primary drop-out rates to less than 10 per cent, increasing learning achievement of primary school students by at least 25 per cent, and reducing the gender and social gaps to less than 5 per cent. 13

Advisory Council, the pros and cons of CSS as a mechanism of resource transfers to states were examined by Saxena (2006). There he had pointed out that as a consequence of lib eralization, Centre s involvement with industry and energy sector was reduced, and so more Plan resources were left in its hands for state subjects. These Plan resources have been usually all ocated through CSS, and education being a concurrent subject, CSS has been one of the key resource trans fer mechanisms for that department. Initially there were a plethora of CSS in education18. A common trend noted in t he earlier years was that while many new schemes were launched over the eighties and nineties, th eir share in the total expenditure on CSS usually declined after the few initial years following their launch. In the paper referred to in the paragraph above, Saxena explained how these schemes giv e political mileage to the party which initiates them as their benefits can be identified easily19. So it is to the government s advantage to increase resources through these schemes rather than inc rease aggregate expenditure. But when the political party in power undergoes a change, the newco mer loses interest in their implementation for the success will be credited to the earlier governme nt. So rather than putting in more effort and resources for continuing the earlier schemes, new sch emes have been announced periodically. DPEP was an exception; except for a couple of years, it has maintained a fairly stable share of 20-25 per cent of the total expenditure. But an adverse s ide effect of introducing these schemes in selected areas has been the creation of islands of excellence w ithout proper integration into the mainstream. Accordingly, their positive effects tend to fad e away once they are over. The most recent development in this field is the introduction of Sarva Shiksha A bhiyan (SSA) in the year 2001-02, as a major tool for achieving UEE by 2010. It acts as an umbre lla scheme for elementary education insofar as it eventually intends to embrace all existing sc hemes in the area of elementary education. In SSA the States are required to contribute funds in a ra tio of 75:25 (centre: state) during the Tenth Plan period (2002-03 to 2006-07) and in a ratio 50:50 su bsequently. SSA is a combination of several schemes it gives some flexibility to the states to priori tize those schemes which are relevant to their local conditions. In 2001-02 it received 13 per cent of the expenditure on CSS (MHRD), and in the Tenth Plan it is expected to receive nearly 60 per cent o f that allocated for elementary education (Table A7). There have been evaluation studies (NIEPA 2005) which claim that the CSS have no

t only eased the resource constraint, but have had a positive impact on the outcomes in the e ducation sector. The Operation Blackboard scheme, along with DPEP and SSA had led to a decline in the number of single-classroom primary schools and improvement in average number of teachers i n primary schools. Larger number of female teachers had been appointed too. In elementary schools the percentage of female teachers in 2004 (42.98 per cent) was slightly higher than the average number for all types of schools (38.18 per cent). However other studies have questioned whether these schemes have added to available resources or substituted other sources. There is need for a systematic study on these schemes. Centrally Sponsored Schemes are a channel through which the central government h as been adding resources to the state education sector. However their coverage has varie d over time and there has been no uniformity in its distribution over different states (Table A8). Mah arashtra and Karnataka, two of the educationally better developed states, have received a maj or share of Operation Blackboard funds (17.4 and 10.9 per cent) while Madhya Pradesh and West Bengal r eceived very little (4.5 and 2.1 per cent). This brings us to an important aspect in the educ regional ation scenario imbalance in state education indicators and failure of budgetary mechanisms to c ompensate for it. Public Expenditure on Education in the States As discussed earlier, education had been a state subject in India till 1976, aft er which it became concurrent with the Centre. Therefore, state government policies would have an o ver-riding impact on the development of education in individual states. In this section we examine the trend in 18 During the Ninth plan there were 360 CSS in all, with allocation being about 60 percent of Central Assistance (Saxena, 2006) 19 Although caste and religious factors are prominent in elections, political pa rties also use social sector schemes and anti-poverty measures as instruments to woo the voters (Mooij & Dev 2002). 14

education expenditures and their possible impact on education outcomes in seven selected states. Given that the states in India have very different size, population, natural end owment and economic development, there is expectedly a lot of disparity in the education performance and the expenses incurred on education in all the states. To capture these variations we have loo ked at the education expenditure and outcomes of seven states in this section. The results obtained f rom such a small selective sample are at best indicative, and more reliable results can be obtain ed if changes in all states are studied20. The states have been selected from different levels of per capita Net State Domestic Product (NSDP) in the year 2004-5. According to this indicator, Haryana , Himachal Pradesh and Kerala are considered to be in the high income category. Meghalay is i n the middle income category and Rajasthan, Madhya Pradesh and Orissa in the low income category 21. 5.1 Trend of Per Capita NSDP in Selected States Table 6: Per Capita NSDP at constant prices (1993/94) Year Haryana H.P. Kerala M.P. Meghalaya Orissa Rajasthan 1990-1 10138 6473 5685 5607 6122 4158 5558 1995-6 11545 8801 8748 6790 7537 5204 7216 1999-2000 13308 11051 10178 8248 9003 5735 8555 2004-5 17465 14674 14441 7809 10450 6955 8368 CGR: 1990-1 to 2004-5 3.7 5.6 6.4 2.2 3.6 3.5 2.7 CGR:1990-1 to 1999-2000 2.8 5.5 6 3.9 3.9 3.3 4.4 CGR: 1999-2000 To 2004-5 5.6 5.8 7.2 -ve 3.0 3.9 -ve Source: Economic survey (various years), WPI from Office of the economic adviser , Ministry of Commerce and Industry Graph 7: Per capita NSDP at constant (1993-94) prices (Rs) 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 199091 199192 199293 199394 199495 199596

199697 199798 199899 19992000 200001 200102 20023 20034 20045 Year Rs Haryana H.P. Kerala M.P. Meghalaya Orissa Rajasthan 20 As already pointed out in FN11, data on sectoral composition of expenditure i s available for education department alone. Among the states discussed in this section, other departments in Madhya Pradesh play an exceptionally important role (30 to 50 per cent) in education finance. 21 Two of the high income states, Kerala and Himachal Pradesh, are educationally a dvanced states. They were included in the sample for in the early nineties their per capita NSDP were simi lar to other less developed states in the sample. 15

Table 6 and Graph 7 show the per capita NSDP (PCNSDP) figures for the seven sele cted states in the last 15 years. It is seen that in 1990-01, Haryana had a significantly higher PC NSDP relative to other states. All the other states except Orissa had similar figures in 1990-91. But K erala and Himachal Pradesh progressed steadily over these years (Kerala s PCNSDP nearly tripled betwe en 1990-91 and 2004-05, while for HP, it more than doubled) and their growth outstripped that o f the other states. Haryana had a lower growth rate over the 1990s, but accelerated since 2000-1. Me ghalaya had a medium growth rate and is at present in middle income category. The rest of the states also grew but at a much lower rate and MP and Rajasthan have shown a negative growth since 199 9-2000. The much-publicised high GDP growth rate in India in the last few years which is see n as a positive outcome of the economic reforms in the country, seems to have left the poorer st ates like Orissa, MP and Rajasthan relatively untouched. 5.2 Stagnation in Expenditure of Education Departments in Different States How has the education expenditure changed in the seven selected states over the last fifteen years? Graph A1 plots the aggregate education expenditure at constant prices in the seven states over the last 15 years. It shows that in the early nineties expenditure has increased in all the seven states but towards the latter half they show a constant or decreasing trend a trend sim ilar to that of aggregate expenditure in India as a whole (Table 3). Interstate comparisons are more meaningful in per capita terms. Table A9 and Gra ph 8 provides the per person education expenditure at constant prices. It appears that in the nineties per person expenditure had increased in all states, though at varying degrees, followed by a decline in the recent years. A part of this increase is likely to be accounted for by increase in teac hers salaries following The Fifth Pay Commission s recommendations for pay revision. So between 1997-98 an d 1999-2000, almost all the states show a spurt in per capita expenditure. Himachal Pradesh stands out as the best performer, showing steady rise in the in dicator over the years. Kerala, the other high income rapidly progressing state, has also shown g ood progress in terms of per capita education expenditure. But the state with the highest per capita r eal NSDP, i.e. Haryana, has given very little priority to education and per capita education expenditure for Haryana lies close to most of the low-income states throughout the period considered. Meghalaya, on the other hand, has shown considerable increase in per capita education expenditure despite being a middle income state22. However, for the low income states the increase in per capita expenditu re has been quite low.

The recent stagnation in aggregate education expenditure is reflected in the tre nd of per capita education expenditure as well Kerala with its decreasing population is the only state whose per capita education expenditure had continued to rise in the present decade. In all other states per capita education expenditure is non-increasing. 22 The seven North-Eastern states in India are the states along the North East b order which had been relatively neglected earlier. In the recent years there has been special emphasis on their education policy. Meghalaya is one of the seven states and it is possible that the changed policy is reflected in the higher expenditure. 16

Graph 8: Per capita Education Expenditure at constant (1993-94) prices 0 100 200 300 400 500 600 700 800 900 1000 199091 199192 199293 199394 199495 199596 199697 199798 199899 19992000 200001 200102 200203 200304 Rs Haryana Himachal Pradesh Kerala Madhya Pradesh Meghalaya Orissa Rajasthan Graph 8: Per capita Education Expenditure at constant (1993-94) prices 0 100 200 300 400 500 600 700 800 900 1000 199091

199192 199293 199394 199495 199596 199697 199798 199899 19992000 200001 200102 200203 200304 Rs Haryana Himachal Pradesh Kerala Madhya Pradesh Meghalaya Orissa Rajasthan In Table 7 the states are ranked according to their per capita NSDP and per capi ta education expenditure. It appears that these two variables do not always move in tandem. H aryana, in spite of high per capita NSDP, gives a low priority to education, in contrast to Himachal Pradesh. Although Rajasthan is a low-income state, its per capita education expenditure has increa singly outstripped that of Madhya Pradesh (another low-income state with comparable per capita NSDP leve ls) over the fifteen years under consideration. While Orissa is the poorest among the seven s tates in terms of PCNSDP, its education expenditure per head is not the lowest. While the low inco me states possibly do not have much scope to increase their expenditure on education easily, the di vergence in priority in Haryana and Himachal is probably due to divergence in state education policie s. Table 7: Ranking the States by Per Capita NSDP and Per Capita Education Expendit ure Per capita NSDP in 2001-2 (Rs.) Rank Per capita education expenditure in 2001-2 (Rs.) Rank Haryana 16773 1 416 4 Kerala 14441 2 471 3 HP 14246 3 867 1

Meghalaya 10450 4 592 2 Rajasthan 8368 5 369 5 Madhya Pradesh 7809 6 208 7 Orissa 6955 7 284 6 Source: Calculated from economic survey (various years) and Table A9 in annexure 5.3 Expenditure on Elementary Education Table 8 gives the proportions of expenditure on education allocated to elementar y stage in different states in selected years23. The proportions agree with the trend seen in aggregate expenditure 23 The usual sources of budgetary expenditure data give sectoral composition of expenditure of the education departments. This will be an underestimation of actual expenditure, particularly for those states where other departments spend a substantial amount on education. 17

of the state (Graph 6A). Though per capita expenditure on education have stagnat ed or even declined, the relative importance of elementary education has been maintained, and has eve n increased in some states (Orissa and Madhya Pradesh). It is only in the high NSDP states of Haryan a and Kerala that less than half of the expenditure is on elementary education. Table 8: Proportion of education expenditure spent on Elementary Education, % State Haryana H.P. Kerala M.P. Meghalaya Orissa Rajasthan 199091 45.90 56.62 52.40 59.38 55.46 54.98 54.38 199596 47.18 54.56 48.73 59.93 59.91 54.75 55.96 200001 45.13 56.67 46.28 68.46 61.18 60.62 57.83 200304 47.77 55.46 42.45 80.66 54.44 58.21 56.65 Source: Analysis of Budgeted Expenditure on Education , Ministry of HRD, Govt. of India-various issues. 5.4 A look at Some Outcome Variables The primary objective of expenditure on education is to educate students. So the re is a need to look at educational outcomes, rather than outlays. It is through the outcomes th at the efficiency and usefulness of education expenditure can be assessed. However it should be kept i n mind that the outcomes are not only influenced by educational expenditures in the immediate pa st but also by the educational development in the past years. These cannot be easily quantified but are important determinants. In the context of India s commitment to UEE, the outcome of financin g of elementary education is examined in this section. The analysis is limited by availability o f reliable data. Data on most of the outcome variables are collected by school teachers and they have a v ested interest in it as poor outcomes reflect on their work. So the reliability of the data is suspect. Much of the money spent on education in India, as has been pointed out, is accou nted for by school-teachers salary. This is particularly true for the States expenditure. But some of the States and a large share of the Centre s expenditure go for creating school infrastructur e, employing new teachers, improving access, providing teaching and learning materials, etc. So a n outcome of this expenditure would be physical indicators like improved access (more schools, mor e rooms and more teachers) and improved facilities at schools. The expenditure can also be reflected in improved enrolment (due to increased dr ive for UEE, for instance). The effect on pupil teacher ratio will depend on the increase in numb er teachers employed relative to enrolment. As long as the rate of adding new teachers keeps pace wit

h the rate of growth in enrolment, the pupil-teacher ratio (PTR) should remain the same. But if the numb er of teachers has not kept pace with rising enrolment, i.e., if PTR has been rising over time, the n clearly the quality of education imparted will be affected adversely. So a preferred outcome of expendi ture in the education sector should be a falling PTR till a stable class-size is reached in order to m aintain/improve the quality of education. The real objective of education expenditure is to improve learning achievement o f the pupils. In that sense the outcome that is most important is how much the students have impr oved their learning as a result of increased amounts being spent in the education sector. Since reli able learning test score data are not available for a period of time, state examination scores may be use d as proxy. The outcome variables are discussed in depth below.24 24 Tables A11 to A15 in Appendix give information on outcomes at national level. But as the aggregate picture hides a lot of interstate variations the discussion in this section is focused o n state level outcomes. 18

Table 9: School Facilities in Selected States in 2005/06 States % with Drinking water % with electricity connection % with girls' toilet % without Black Board in all classrooms Pupil teacher ratio Primary school Upper primary school Haryana (H) 90.0 (L) 41.8 (M) 70.7 (H) 5.1 41 24 HP (H) 88.7 (M) 54.2 (L) 29.7 (H) 5.6 20 15 Kerala (H) 93.7 (H) 79.7 (M) 71.5 (H) 4.7 26 26 MP (H) 84.7 (L) 19.4 (L) 21.4 (H) 9.8 41 32 Meghalay (L) 35.8 (L) 11.3 (L) 7.7 (M) 20.2 18 13 Orissa (M) 82.8 (L) 12.5 (L) 12.4 (H) 8.4 33 33 Rajasthan (M) 77.1 (L) 24.2 (L) 34.6 (M) 14.0 37 23 Source: Mehta (2006), State Report Cards. Note: H: High, M: Medium, L: low School facilities Reliable trend data for facilities at school are not available, so data for the year 2005-06 are examined in Table 9 to assess the status of infrastructure of the schools in the states under focus. Here, too, the higher income states, especially the ones with the higher educati on expenditure, fare the best. Kerala has the best indicators in terms of all aspects of infrastructu re. HP is deficient only in terms of electricity availability and girls toilet. Haryana also has fairly good facilities at its schools, except for lack of electricity. But availability of girls toilet is inadequate in all the states considered, except for Haryana and Kerala. The middle income state of Meghalaya shows poor indicators on all counts in spit e of its high education expenditure per capita. This is intriguing and needs a closer examinat ion since the higher amount spent should be reflected in better facilities at schools (unless the inf rastructure was in a very poor state in the early nineties, in which case the status in 2005-06 would be a n improved one). All the poorer states suffer from low power availability and inadequate number o f girls

toilets. But Orissa is better off than the other states in terms of availability of blackboard. It may be also recalled that Orissa, despite being the poorest among the seven states, had a per capita education expenditure higher than MP and was close to the expenditure of Rajasthan through out the period considered. In fact, if we consider elementary education expenditure per capita, then in Orissa it has not only been above that of MP, but has also crossed the expenditure in Rajastha n in several of the years. So there seems to be some link between education expenditure and school f acilities. School Participation: In Table 10 school participation indicators at elementary stage are compared wit h per student expenditure in that stage. It is seen that enrolment ratios at the primary stage are high for all states Haryana25 being the only exception. One needs to remember enrolment is not synon ymous with participation, and may include the nominally enrolled. At middle stage there is a closer correspondence between GERs and the level of finance. Meghalaya is the major exc eption where the high per student expenditure is not reflected in enrolment ratios. Table 10: Grading of Expenditure and Outcome indicators in 2003/04 State Per capita NSDP Per student expenditure on elem. Education GER primary (6 to 10 years) GER upper primary (11 to 13 years) Gender parity index Dropout (class 1 to 8) Boys Girls Boys Girls Primary Middle Haryana H M M M M M H H H 25 Haryana has a large number of private unaided and unrecognized schools (Aggar wal, 2000) at elementary stage and it is likely that the enrolments in these schools are not included in the calculation. This could be the explanation of the medium and not low GER at middle stage. 19

Himachal Pradesh H H H H H H H H H Kerala H M H H H H H H H Madhya Pradesh L L H H M L H M M Meghalaya M M H H L L H H L Orissa L L H H M L H M L Rajasthan L L H H L L M L L Note: 1. H: High; M: Medium; L: Low. 2. Dropout rates have been indexed in rever se as high dropout rate is a negative outcome. 3. Gender Parity Index is the ratio of girls' enrolment to boy s' enrolment in the relevant grades. Source: Derived from Table A10 in Annexure The dropout ratios show a close correspondence with expenditure levels states wi th middle or high per student expenditure have low dropout rates and vice versa. But here Madhya Pradesh in addition to Meghalaya shows contradictory results. In spite of low levels of exp enditure and low per capita NSDP, it shows medium levels of dropout. Other government departments in Madhya Pradesh spend substantially on elementary education, and the government had also introdu ced several innovative schemes in this sector. A more detailed analysis is required to under stand whether this contradiction is due to poor data or innovative strategies. Gender parity index is a good measure of UEE as girls participation has usually b een lower. India traditionally has a preference for male child, so that share of gir ls enrolment in total or existence of any gender bias in enrolment would indicate the areas where there i s need for an increased drive for UEE. In that sense girls enrolment share is an important outc ome variable for money spent on education in the country. It is a positive sign that there is no indication of a gender bias at the primar y level in any of the seven states. However, at the middle level, Rajasthan shows a considerable d ifference between boys and girls enrolment. In MP and Orissa, too, there is some gender bias. The practice of early marriage for girls and relatively less emphasis on their education in Rajasthan is well-known and would probably explain the bias. Poverty and low age of marriage for girls gener ally go hand in hand, and may provide an explanation for the bias in the other two states. Learning Achievement Higher enrolment, better school facilities, more teachers per class, all become meaningless without proper teaching activity in class and without proper learning on the par

t of the students. According to a MHRD (Ministry of Human Resource Development) Report a very high proportion (around 30 per cent) of enrolled students in India do not attend schools and tea cher absenteeism was also found to be quite high (20 per cent). It also found that more than half of upper primary school students cannot divide a three digit figure with one digit26. Table 11: Examination Results of Final Year Students in the Primary and the Uppe r Primary Stage in 2004/05, % Primary (class 4/5) Upper Primary (class 7/8) Boys Girls Boys Girls States passed Marks >=60% passed Marks >=60% passed Marks >=60% passed Marks >=60% Haryana 90.9 29.8 92.0 31.4 57.9 15.6 56.4 15.5 HP 96.6 54.1 97.2 57.9 78.7 20.6 78.6 22.3 Kerala 93.5 55.5 93.3 58.8 88.7 37.9 92.7 42.3 MP 78.1 25.3 77.0 24.2 58.2 18.3 58.3 19.1 26 A research study was commissioned under SSA, where assessments were made on t he basis of three unannounced visits to 6745 schools in 20 districts in 2006-7. MHRD, 2007. 20

Meghalay 87.7 24.3 88.8 24.6 87.1 22.9 86.5 22.8 Orissa 98.0 20.9 98.0 19.5 91.0 13.0 91.8 13.2 Rajasthan 96.7 52.6 96.6 49.5 86.7 46.3 88.0 46.8 Source: Mehta (2006): State Report Cards We look at the examination scores for primary and middle levels for 2004-05 in T able 11 for the seven states in focus. The pass percentages are quite high at primary level for most states. At middle stage there is a 20-25 per cent drop in Haryana, HP and MP. It is not cer tain whether higher pass percentages indicate better quality of education or that the other states h ave a better examination system and screens out weak students. It is not easy to rely on exam data as the dropout rates give a contrasting picture. As seen in the earlier table, the survival rate figures for these states and Kerala are comparatively higher while dropout rates of the other states are very high. From these different indicators it is seen that per student expenditure can expl ain a part of the variations in outcomes, but there remains a large unexplained part. Particularly at primary level indicators like school participation, gender parity and pass percentage seem to be quite high even when the levels of expenditure on education in several of these states are low. As there are limited opportunities of generating finance through budgetary sourc es the states have looked for assistance from nonbudgetary sources. Perhaps if expenditure from the se sources were also considered a better linkage between education and outcome could be observed. A common policy adopted by several states had been to encourage private unaided schools. The rationale behind this strategy has been that it reduced the financial burden on public sector as the better off go to private schools leaving the public sector to cater to a smaller number. Table 12 is taken from Sipahimalani (2000) where household spending has been estimated from household expenditure data collected by NSSO in 1995-96 and public expenditure is taken fr om budget documents. It is seen that more than a third of the expenditure in elementary ed ucation is made by the household sector. Since then privatization has possibly increased, both at eleme ntary level and at higher level. In recent years several states have invited private funds from individuals and t he corporate sector. Rajasthan Government had launched the Rajasthan Education Initiative (RE I), a new venture aimed at engaging global and local partners from the private sector, foundations and NGOs in innovative multi-stakeholder partnerships to support education in the State. Mad hya Pradesh have launched the scheme of fund a school which are exempted from both Foreign Contribu tion

Regulation act and income tax, and allows private funding of different component s in any school of their choice. Table 12: Contribution of Different Sector in Education Finance in India, 1995/0 6 Per cent of total expenditure funded by Total (Rs. Millions)Central government State /UT governments Households Elementary 4.9 57.3 37.8 244982 Secondary 4.9 67.7 27.4 143216 University and Higher Education 13.2 58.6 28.2 54406 Technical 32.9 67.1 -15001 Other 24.1 75.9 -8472 Total 7.1 61.3 31.6 466077 Source: Sipahimalani (2000) Finally, schooling outcomes depend not only on school factors but also on home f actors. Information from different rounds of NSSO and census survey shows that poverty a nd unemployment remains a major problem in India. Apart from income and expenditure, food insecur ity, illness, lack of choices and opportunities, forced livelihood options, vulnerability to crises and lack of access to basic services such as safe drinking water, school or health facilities are other important aspects of 21

deprivation, although many of these may be linked to income or expenditure-based poverty (Jha & Jhingran, 2005). Thus the concentration of poor people in a state is likely to affect the enrolme nt, and, particularly, the retention of children in schools, posing a serious challenge f or UEE. For many poor families, especially those from SC/ST families and some from minority communitie s and those from far-flung remote villages, the poverty is too crushing for children to be able t o take time out for studying and for conditions conducive for education to be brought about. The gir l child in a poor family is even more vulnerable because either she has to engage in income genera tion activities or has to take care of siblings27 and household chores. Since most parents are illitera te, even the enrolled children who attend school do not receive any help with their studies at home. M any enrolled children have to help their family for work on a seasonal basis and have irregular attend ance. All these factors lead some to argue that unless these problems are addressed and the resources di rected there, expenditure on school education will not be sufficient to achieve desired educat ion outcomes. The introduction of cooked midday meal in all primary schools and universal cove rage of children below 6 years under anganwadi facilities are too recent to evaluate the ir impact on outcomes. But the relevant departments have indicated prioritization of these pr ogrammes through financial allocations. It is very likely that as these policies directly target the children from economically and educationally deprived families, if implemented properly they w ould show positive outcomes. 5.5 Comparison of trend in Expenditure and Outcomes An attempt is made to look at changes in expenditure on elementary education and how it has impacted outcomes. Ideally one should compare changes in dropout rates and learn ing achievement tests with education expenditure. But in absence of relevant reliable data we ha ve looked at enrolment and pupil teacher ratios. The first could be a measure of improving access at el ementary stage while pupil teacher ratio is a useful index of school quality. Impact of Expenditure on Enrolment In the set of graphs in Graph A2 (A to G) in the Annexure, time series data on e xpenditure on elementary education at constant prices and enrolment in elementary stage is plo tted. To enable comparisons in the same graph, the units of enrolment figures are in certain sta tes in hundreds and in others in thousands. Expenditures are in rupees crores. The focus in these graph

s is not the absolute number of children enrolled, but on their changes over time. It is seen that the higher income states show a pattern different from that of t he low/middle income ones. The elementary enrolment graph is flat for HP and Kerala (graphs B and C) over the period considered and for the latter there is actually some decline reflecting i ts declining population. For Haryana, after an initial rise till 1993-94, enrolment hardly shows any grow th (graph A). For these high income states there is little correlation for expenditures show an in creasing trend in all three. This possibly indicates that almost universal enrolment has been achieved and there is little scope to increase enrolment28. In the other states (D, E, F and G) a strong correspondence in changes in these two variables is identified.29 In Meghalaya and Rajasthan, there is a steady rise in enrolment throughout. Enrolment has followed the pattern of expenditure during the nineties, but where expenditu re has stagnated/declined in the present decade, enrolment has continued to rise. The s ituation is similar in Madhya Pradesh; the sudden dip in both the indicators is a result of formation o f the new state Chattisgarh out of Madhya Pradesh. Orissa is the only state showing a poor growt h in enrolment. Unlike the other states, the increase in enrolment here has come more in the per iod after 1998-99. So in states where many children are still out of school, there seems to be a link between education expenditure and enrolment. 27 More than 90 per cent of BPL households are nuclear where both parents must w ork and there are no adults to take care of children (Jha & Jhingran, 2005). 28 These high income states also have a low fertility rate and child population is usually nonincreasing. 29 The sudden drop in both expenditure and enrolment in figures for Madhya Prade sh reflects bifurcation of the state and formation of the new state of Chattisgarh. 22

Impact of Per Student Expenditure on Pupil-Teacher ratio Graph 9 illustrates the increase in per student expenditure in the high income s tates and stagnation or decline in other states. This is in keeping with the trend in per capita education expenditure as given in Table A 9. Meghalaya started out with high per student e xpenditure (possibly because of low enrolment) but fell subsequently. All the other states started wi th similar (lowest were MP and Haryana) per student expenditure. It increased in Himachal Pradesh steadi ly thru 90's and stabilized at a high level. Kerala also showed continuing upward trend, even aft er 1999-2000. Madhya Pradesh has the lowest level of per student elementary expenditure and Ra jasthan has the poorest growth rate, a surprising finding considering the publicity given to ele mentary education by the state governments. It is remarkable that in spite of recommendations of seve ral committees, expenditure in education as a whole and that on elementary education have not sh own much progress in real terms. Graph 9. Per Student Elementary Education Expenditure at constant prices (Rs) 0 500 1000 1500 2000 2500 3000 3500 199192 199293 199394 199495 199596 199697 199798 199899 19992000 200001 200102 200203 200304

Year Ex pe nd itu re (R s) Haryana Himachal Pradesh Kerala Madhya Pradesh Meghalaya Orissa Rajasthan In the next set of graphs (Graph A3) in the Annexure we have looked at the chang es in pupil teacher ratios in primary and upper primary30 schools in these states. The seven states in focus show a clear divergence in the movement of PTR in upper primary schools depending on wh ether the state has high per capita NSDP or low/middle per capita NSDP. In the primary schools most states, irrespective of PCNSDP, show a flat graph si nce 1991-92 till around 2000-01 after that there are some sudden fluctuations. These do not correspond with the per student expenditure. However in upper primary schools, the states with increasing per student expendi ture (Haryana, HP and Kerala) show a declining trend in PTR. In view of the discussio n on the trend of enrolment, it is likely that this decrease in PTR reflects the stagnating enrolm ent with rising number of teachers. For the other four states at the middle level Meghalay and MP have trends slight ly different from the other two states. PTR for Meghalaya has fallen throughout, but for MP, the PTR started declining after 1993-94. Orissa and Rajasthan have witnessed a slightly increasi ng trend in PTR till around 2000-01, after which the trend was reversed. There appears no corresponde nce with the per student expenditure. In these states enrolment has been increasing while per stu dent expenditure on elementary education has been constant or declining per student expenditure. Thi s result should lead 30 Upper primary schools are different from upper primary stage. Upper primary s chools are any school whose highest class is between 6 to 8 usually it includes schools with classes 1 to 7/ 8 and schools with classes 6 8. 23

to an increasing pupil teacher ratio. This contradiction arose because during th is period most states introduced a new cadre of teachers teachers who were locally recruited on contra ct basis and paid a much lower salary and often had lower formal qualifications. The process started in Rajasthan with SIDA funded Shiksha Karmi project. But in the present decade it is Madhya Prades h which has led the way by mainstreaming these initiatives the cadre of assistant teachers (regul ar school teachers) is gradually being replaced by Shiksha karmis or samvida shikshaks (para teachers) (Leclercq, 2003). Factors influencing Public sector finance in India So far the paper has looked at the major changes in trend and composition of edu cation finance variables and some outcome indicators in the country. In this section we look at those factors which have impacted these variables and determined the direction of changes. 6.1 Underinvestment in Education It has been observed earlier (Graph 3) that expenditure on education as a tage of GDP has varied between three to four per cent over the period 1990-91 to 2004-05. s considerably short of the target of six per cent of GNP. And since the year 2000-1 there has decline in absolute expenditure in real terms. We look into the causes behind the decline and eps taken by the central and state governments to increase resources. Budgetary imbalance arising from resource sharing mechanisms There is an in-built imbalance between revenue and expenditure of the state gove rnments in the existing resource sharing system. In the revenue sharing design most of the residual powers accrue to the central government. Centralized planning in a mixed economy framew ork and nationalization of major financial institutions had led to concentration of fina ncial powers in the hands of the Centre (Rao, 2000). So the States have limited options of generatin g additional revenue and have to depend on transfers from the Centre for many of its routine expendit ures. Fiscal crisis following Structural Adjustment Following the structural adjustment programme (introduced in the early nineties) and accompanying economic reforms adopted at that time, both Centre and the States f aced a fiscal crisis. They were told to curtail expenditure and generate additional revenue. Despite m easures to control the size of the fiscal deficit, after an initial decline, it started increasing in the later part of the nineties. Implementation of recommendations of Fifth Pay Commission worsened the situation . percen This i been a the st

Several steps have been taken by the central and state governments towards fisca l reforms: (i) Following the fiscal reforms there has been a change in the relative contrib ution of direct and indirect taxes reflecting additional efforts by the governments to au gment their direct tax revenue to counter the impact of structural reforms on indirect tax collections. So revenue from indirect taxes as a proportion of GDP declined from 13 per cent in 1990-91 to ar ound 11 per cent from 1997-9831but the share of direct tax revenue increased from two to nearly f our per cent in the recent years. Unfortunately this was not sufficient to maintain tax revenue at e arlier level, the total tax collection as a proportion of GDP had declined from around 16 per cent in late e ighties and had begun to rise only since 2000. This indicates that the several measures taken by the respective governments to increase revenue receipts broadening of tax base, improved tax co mpliance as well as rationalising user charges were not sufficient to maintain aggregate tax reve nue at earlier levels.32. (ii) To meet fund shortages, both the Centre and the States turned to increased borrowing in the 1990s. Though this worked in meeting short term shortages, it led to high er interest payment burden. 31 There has been an increase in the proportion in 2004-5 budget estimates. 32 Notes from state Finances: a study of budgets of 2003-4: RBI February 2003 24

(iii) The need to limit revenue deficits in both the Centre and States budgets le d to limiting of public expenditure in the 1990s33. Several states subsequently propo sed some economy measures such as restriction on fresh recruitments/creation of new posts and lim iting the growth of administrative expenditure. Institutional reforms have been proposed through the Fiscal Responsibility and Management Act (FRBM). This act, at the Centre, requires the Centre to wipe out revenue deficits completely and to reduce the fiscal deficit to three per cent o f GDP by 2008-09. Fiscal Responsibility Acts passed by the States have similar requirements. This development implies that if revenue declines, government expenditure must also be pruned proportiona tely and here it is the soft social sectors like health and education that face budget cuts.34 The Indian government recently tried a different source of finance for meeting t he huge resource requirement for UEE. An education cess of two per cent was levied on al l central taxes in 2004, which yielded Rs 5000 crore in 2004-05. Indian government has established a fund titled as Prathmik Shiksha Kosh to be maintained by the Department of Elementary Education a nd Literacy (MHRD). The receipts will be utilized solely and exclusively for elementary educ ation, including Sarva Shiksha Abhiyan and Nutritional Support to Primary Education (Mid-day Meal Scheme). So this fund has played a crucial role in augmenting funds for SSA (Jha et al 2006) . The fund will be nonlapsable in nature and balances remaining un-utilised in a year will be avail able on a rollover basis for the said purposes. So it has the advantage that it is an additional su pplementary resource for financing elementary education and does not cut into existing sources of finance . Although many economists are not in favour of earmarking funds35, given that it has worked for many Asian, Latin American and African countries, this measure can be used for India provided spec ific safe-guards are built into the spending mechanism36 (Mehrotra, 2005). In general the steps taken to limit fiscal deficits, particularly the FRBM act, indicate that this will aggravate resource shortage. Since the central government is also under pre ssure to reduce its deficits it has restricted transfers to the states. The states in turn have a lo w tax base and the burden of meeting the nonplan costs. There is little scope in increasing tax revenue, so t he States have become increasingly dependent on the Centre to meet their education expenditure. 6.2 Inefficiency in resource utilization Expenditure on the education sector is lower than required in the majority of th e states. But that is only part of the problem. Present discussions and debates around education finan ce bring to the fore that it is not only the total allocation but its utilization and management whic

h determine the outcome. The problems may arise from improper sectoral allocation, from inabilit y to utilize allocated funds and from problems in implementing and monitoring plans due to la ck of coordination within the governments. Some of the possible reasons for inefficiencies in resource utilization are disc ussed below. Asymmetry in allocation within the sector The relationship between expenditure and educational development is a complex on e the development of education depends on the composition, efficiency and accountabili ty of the expenditure on education. 33 This was very difficult to maintain as implementation of the pay revisions re commended by the Fifth Pay Commission resulted in huge hikes in salary bills for the States; from 1995-96 t o 1999-2000 the average annual increase in per employee nominal expenditure, for 21 states, amounted to 59 per cent. 34 There have been some recent arguments indicating that fiscal management need not be an overriding concern for India since the debt situation has considerably improved. (Jha et al 2006) 35 World Bank and IMF are two main international organizations opposed to using such earmarked extrabudgetary funds. Their position is that using such funds can lead to loss of con trol on aggregate expenditure, distort resource allocation, etc. and also pose a risk that the sector the funds are earmarked for may not be the final recipient, since government funds are fungible. 36 For instance, DPEP funds were given to Central government on the condition th at states maintain at least the 1991-92 levels of expenditure on elementary education. 25

As discussed earlier the balance between Plan and Nonplan expenditures is import ant, and balance between recurring cost and asset creation is also desirable. Within each education sector a balance between the salary and non-salary inputs is required for efficient outco mes. Public expenditure on education in India suffers from disproportionately high teachers s alaries. Some attempts to estimate the proportion of salary expenditure in elementary educatio n showed that around 85-90 per cent of the total is spent on these items (Tilak, 2003). The high prop ortion of teachers salary is not only due to large number of teachers (the pupil teacher ratios ind icate the need for more teachers), it is also because of the high rates of government salaries. A study on UP teachers salaries showed that government school teachers earned 13.6 times the average per capita income in the state. For other states with comparable levels of per capita income, the teachers salary levels are also similar. And this ratio is very high compared to other developing countries (Meh rotra, 2005). Teachers have become politically powerful as a group in most states and so have good bargaining position. In contrast the salaries of teachers in large majority of private scho ols and also that of contract teachers in government schools (who are often with similar education le vels) are a fraction of the salaries of government teachers salary. Given the resource constraint insuffi cient amounts are allocated on non-salary inputs like Teaching Learning Material (TLM) and teacher s training. There is a lot of scope for improving efficiency in this area. SSA has tried to provide funds for non-salary expenses. It provides Rs 5000 for building repairs to schools in the different states. This amount will possibly be a welco me addition going by the urgent repairs needed in many of the schools, particularly in the north-east ern states. Under both DPEP and SSA, there is also a provision for a lump sum grant of Rs 2000 per annu m for school development for all elementary schools. There is also a TLM grant of Rs 500 per teacher per annum. One needs to assess the impact of these grants on schooling outcomes. Balance between different sectors within education is also required. Prioritisin g elementary education at the expense of secondary education has its positives and negatives. Few parents want their children to stop at class eight but are often forced to do so in the absen ce of sufficient government schools at that level. But public subsidization of free government sc hooling at the secondary level, in the absence of universal primary and upper primary education , hurts equity considerations by aiding the non-poor who can afford to pay for their education.

More than half the children at secondary/higher secondary level are in private s chools, many of which are aided by the government. As a result a 22 per cent of revenue expen diture of elementary schools and 50 per cent for secondary schools are on private aided schools37. Th e utility of the aided schools has been questioned by several researchers. If an unaided school starts being aided by the government it adds to the already squeezed public resources. And it is also iniq uitous as most of the students come from section which is able to pay. Three tiers of government, lack of coordination and resulting inefficiencies Though Indian federalism has been constitutionally a two-tiered system since 199 2, rural and urban local bodies existed before that, working as agencies for the state govern ment. With 73rd and 74th constitutional amendments in 1992 local governments received statutory reco gnition as elected bodies38. The amendments also contained illustrative lists of functions and sour ces of finance. Each state government was to appoint a State Finance Commission to assign taxes and f ees to local governments and to recommend tax devolution and grants (Rao, 2000). In addition state governments transfer the funds received from central ministries for central plans and CSS to the local bodies. In certain ways this system is parallel to Centre-State transfers. The funds come t hrough two separate strands, and the amounts are determined through separate norms. Moreover in central plans and CSS as the designs are made at central level, the state or the local bodies do not have any flexibility, and often lack motivation. If the plan is made by the higher level of government, inefficiency in its implementation is unavoidable even when it is partly or fully funded from the higher level. There were similar problems between the central an d state governments 37 Analysis of Budgeted Expenditure on Education (2003-4), MHRD, New Delhi 38 The rural local government is three tiered district panchayet, Taluk (Block) Panchayat and village panchayat. Urban local governments are municipal corporations in large cities, m unicipalities in smaller cities and towns and nagar panchayats in smaller towns. 26

and these inefficiencies existed even before the local government system was dev eloped (regarding implementation of NFEs and OBB). With the third layer there would be little impr ovement unless initiatives in planning from below are successful. Underutilization of funds: A major impact of the different tiers of government h as been underutilization of allocated funds. This is a serious problem that dogs many so cial sectors in the Indian economy, including education, labour and employment (Mooij and Dev, 2002) . The underutilization may be for certain items and not in others and thus starve priority items within the education sector, while maintaining funds for inessential items. (i) There is delay in utilization as the states need time to get acquainted with new guidelines and procedures relating to new schemes. (ii) There have been delays in releasing installments by the central bureaucracy resu lting in lower utilization in the present year and since allocation is based on previo us year s expenditure, this means lower allocation in the next year.39 (iii) For certain schemes, central funds are not released without matching grants from states and the poorer states may find it difficult to raise their share of resou rces. There are delays in release of funds on their part as well. (iv) Some schemes presuppose presence of infrastructure or institutions which may not be there. These schemes cannot be implemented in such cases. (v) Central Plan schemes and the centrally sponsored schemes are designed at the cen tre and have little flexibility. The states may not be interested in some of these s chemes due to political reasons or because they feel that the schemes are not very rele vant in their local contexts. So at the state level necessary plans for its utilization are not made40. Improving implementation of schemes and utilization of allocated funds w ill largely depend upon the effectiveness of the decentralized planning process in t he country. Some of these problems can be observed through the working of SSA, the latest in itiative in elementary education. As described earlier, this scheme was a combination of sev eral schemes and the state governments and local governments had some flexibility in its implemen tation. The planning started from the local government and was thereby expected to be more efficient in utilization of

resources. But it had its share of problems, some of which are discussed below: Although the Tenth Five Year Plan recommended that the outlay by the central gov ernment on SSA during 2002-03 to 2006-07 should amount to Rs 17000 crore, it remained sh ort of funds till 2004-05. The situation eased only when it received combined external support fro m the World Bank, DFID and EC committing Rs 4700 crore and a further Rs 5000 crore arising from th e imposition of two per cent Education cess on all central taxes. Table 13 shows how less than h alf of approved outlay of MHRD had been released under SSA. There is also a gulf between the planners in New Delhi and the activity of the s tate implementing societies. The SSA programme requires an Annual Work Plan and Budge t to be produced in each district on the basis of a perspective plan made from household surveys. Its basis is intended to be participatory and consultative. This plan should be the starting point for assessment of expenditure. But the plans are actually drawn up by the district officials and t he villagers are bypassed. 39 In the period 2001-2002 to 2004-2005 in as many as 66 cases first installment of grant was realeased in the month of September when the second installment should have been released. In hal f the cases the grant was released in March at the fag-end of the financial year (Rani, 2007). 40 For instance, lack of interest of the UP government in Total Literacy Campaig n and underutilization of large grants earmarked for the promotion of elementary education might have been due t o vested interest in social status quo, with education being perceived as a threat (Dreze & Gazdar, 1996). 27

The States allege that the Centre releases funds on a piecemeal basis, creating uncertainty, whereas the funds are supposed to be released twice a year. The uncertainty of t he time and amounts make it difficult for the SIS to timely release installments to the districts. I n Uttar Pradesh funds received by the SIS in September 2003 were released to the districts over the ne xt five months, Rani (2007). The proportion of funds utilized from those released varies for different states and is usually lower for the poorer states. The need to obtain matching grants and approval of the annual work plan and budget is useful for efficient resource utilization. However this puts the p oorer states at a disadvantage for their resource base is smaller and they also have lower institu tional capacity. The scheme is facing so many problems in its implementation that Jha et al (2006 ) have questioned the claim of Ministry of Human Resources and Development (MHRD) regar ding in the year 2003, within a period of just 9 spectacular progress made under SSA ( months, as many as 1 crore and 68 lakh out of school children .were enrolled in formal and al ternative schools )41. However many of these problems are expected to be solved as the SIS ( State Implementing Societies) get familiarized with the process. Table 13: Share of Funds Released by MHRD under SSA as a Proportion of Approved Outlay Approved outlay by MHRD under SSA (Rs. crores) Funds released by MHRD under SSA (Rs. crores) Funds released as % of approved outlay of MHRD 2002-3 1138 500 43.9 2003-4 3080 1558 50.6 2004-5 8335 2431 29.2 2005-6 11019 5051 45.8 2006-7 13608 7440 54.7 Source: Rani (2007) p 212 6.3 Increase in Foreign Aid Foreign aid has been considered as a possible additional source of education fin ance in many developing countries. However, in India as a proportion of aggregate public expe nditure on education, it appears to be quite low in 2002-03 when external aid was at its ma ximum level it was equivalent only to 1.5 per cent of education expenditure and three per cent of e

xpenditure on elementary education. However before the introduction of economic reforms in the 1990s its contribution to Indian education had been even lower. As pointed out by Tilak (2 008) foreign aid was felt necessary only in the case of foreign-exchange-intensive, capital-intensive sectors and for those sectors in need of foreign expertise. Education in India in general and school e ducation in particular do not fall in any of these categories. But the role of foreign aid in the educa tion sector has changed since then. In earlier years, education was financed primarily by domestic resources; the li mited foreign aid available was focused on technical and vocational education. But more aid resour ces from abroad flowed to India from the 1990s onwards, and became increasingly directed at elem entary education. The shift possibly came about because of the global awareness regarding the adve rse impacts of structural adjustment policies on social sectors like education and the launchin g of a compensatory social safety net programme funded by loans from World Bank/IMF. Following the i nternationally declared objective of achieving Education for All in the Jomtien Conference (1990) the international aid agencies were keen to increase their aid commitments to primar y education (Mehrotra et al, 2005)42. 41 Jha et al raise questions regarding associated requirements for this statisti c to be true: firstly, whether there was commensurate increase in the number of schools, teachers and classrooms and secondly, whether adequate re sources were available. 42 There is also a view that in the 1990s, there was a shift among the Indian po licymakers regarding the conceptualisation of poverty and the prioritization of social policies (Mooij & Dev, 2002). This included turning away from income and employment issues as solution to poverty and moving towards a broader concept of development, which includes education, health, housing roads, infrastructure, an d so on. 28

Level of Foreign Aid in Education It is difficult to estimate the total size of foreign aid in education through t he different budgets. It enters the flow of funds as a part of Plan expenditure through the c entral budget and is provided both for state Plan projects and for centrally sponsored schemes. For t he former, central government receives foreign aid and transfers a part through Additional Central Assistance to the State budgets. The major part of the aid comes through CSS which are given to Ce nter s Department of Education and are released either through state budgets or via a direct route to DRDA (District Rural Development Agencies) or SIS (State Implementing Societies). The rapid increase in foreign assistance from the Seventh to the Ninth Plans is evident from the data presented in Table 14. The allocation for the Tenth Plan has not shown any change in absolute terms. However, the data for actual expenditures indicate that a much l ower amount has been disbursed. Table 14: Allocation on Elementary Education during different Plans, Rs. Crores Five Year Plan Total Allocation External Aid 7th Plan 1987-1992 658.49 8.62 8th Plan 1992-1997 4006.55 613.62 9th Plan 1997-2002 14754 4093 10th Plan 2002-2007 28750 4904* *1 bn US. dollars Source: Annual Financial Statistics, GOI, relevant years Table 15 gives annual estimates from different sources. The second column figure s are given in Bashir (2000) and represent expenditure on externally aided projects in CSS. The estimates given in Tilak (2008) (col. 3) are for total foreign aid to education. Comparison of t he two series indicates the importance of CSS as an instrument for allocating foreign aid. Column 4 give s the estimates of external assistance to education in the present decade given in GOI s external bud gets. It is found to be much higher than the actual expenditure as given in column 3 since 2005-06. T his is perplexing as assistance for SSA has been given in the recent years but is not reflected in co lumn 3. Some estimate of the contribution of foreign aid as a proportion of central gove

rnment plan expenditure on education is also given in Tilak (2008). It is seen that its shar e has increased from five per cent in 1993-94 to 20 per cent in 2000-01. The proportion is higher for elem entary education it had increased from 10 per cent to 35 per cent. There had been a dip in the propo rtion in mid nineties. This should be seen in the context of decline in overall aid flow in the ninetie s - partly a fall-out of sanctions imposed by several donors in reaction to nuclear tests in mid nineties and partly because of completion of funding cycle from several sources. This absolute amount of extern al aid has not been increasing over the present decade, and its proportion has shown a declining tre nd primarily because of the sharp increase in GOI s allocation to elementary education. There is also a significant portion of foreign aid which does not flow through t he government budgets. These are funds both from foreign governments as well as from foreign p rivate nonprofit agencies (Oxfam, Action Aid, Save the Children s Fund etc), and these directly fun d education in the Indian non-government organisations. It is difficult to estimate the absolute si ze of this form of aid, as these often do not exclusively fund education interventions, but rather a combin ation of interventions of which education is an important component. As regards both the volume of external aid and the modalities by which it has be en provided, there have been two broad historical phases in external assistance. These are no w briefly discussed. 29

Table 15: Estimates of Foreign aid to Elementary Education in India, Rs. Crore Year External aid to CSS in elementary education External aid in elementary education Budget estimates for externally aided education projects in central plan (1) (2) (3) (4) 1992-3 4.00 1993-4 9.76 37 1994-5 103.33 122 1995-6 215.94 228 1996-7 201.62 219 1997-8 561.89 610 1998-9 585.3 597 1999-2000 808.25* 729 2000-1 948 2001-2 1210 1212.34 2002-3 1285 1383.09 2003-4 960 939.59 2004-5 683 683.45 2005-6 631 1996.5 2006-7 126 1647 2007-8 114 1677.6 2008-9 1584 *budget estimate Source- col (2): Bashir (2000), col (3): Tilak (2008), col (4): expenditure budg ets of GOI First Phase: Increasing Aid, Many Projects, Several Donors The first phase began with Andhra Pradesh Primary Education Project (1986) funde d by British Overseas Development assistance and some non-formal education projects f unded by UNICEF. The other major aided projects were the Swedish International Developmen t agency (SIDA) assisted Rajasthan Shiksha Karmi Project (1987), Mahila Samakhya Programm e in selected states (1988-90) with Dutch International aid agency assistance, Bihar Education Project (1990) supported by Unicef, UP basis education project (1991) supported by World Bank a nd Rajasthan Lok Jumbish project (1992) supported by SIDA. Initial projects were part of state pl ans but the later projects were part of the centrally sponsored schemes. External assistance to education started flowing in before a framework for it co uld be formulated. There were several donors in several projects in different districts working without any coordination. So DPEP, started in 1994, was made to serve as an umbrella program me for all aided education projects in primary education. The external assistance in this project in 2001 was about Rs

6716 crore comprising Rs 4951 crore as credit from IDA and the rest as grants fr om EC/DFID/UNICEF/Netherlands (MHRD, 2001). Second phase: Decline in Aid and New Modalities As discussed, the aid contribution as a proportion of total education expenditur e has decreased in recent years, though in absolute amount they remained at similar le vel. But there have been major changes in the nature of interactions between donor countries and Ind ian government. In 2003, India introduced a change in policy by short-listing only six countries from which it would accept bilateral aid (US, UK, EC, Japan, Russia and Germany). In addition, it decided not to accept any tied aid. The indication was a change in its status from aid-taker to ai d-giver (by announcing that it would stop taking bilateral aid from a number of small countr ies like Denmark, Netherlands, Norway, etc.) and many people have linked this decision either to t he fact that donors questioned the nuclear tests carried out by India in 1998 or to the more recent anti-Muslim pogrom in Gujarat. Along with forsaking this tied aid , the Indian government simultaneously announced that it was creating an India Development Initiative Fund (IDIF) for giving aid to less developed countries (DN, 2003). 30

At the same time there were changes in the international context of foreign aid. The lack of effectiveness of aid in promoting development in the recipient countries in the S tructural Adjustment era (Ranis, 2006) led to the donor countries rethinking on the aid modalities. New Aid Modalities were advocated, whereby the focus was placed upon the encouragement of recipient country ownership, aided by a sector-wide approach (SWA) rather than the earlier project-b ased approach to development. Poverty was to be tackled under the HIPC/PRSP initiative, led by the Bank and the Fund, and supported by bilateral donors. The Fast Track Initiative (FTI), anothe r new aid modality, had the achievement of the education MDGs (Millenium Development Goals) as its f rame of reference. Countries implementing policy and institutional reforms, and with the means to prove their performance record, are rewarded with additional aid and better coordinated exte rnal assistance for their educational plans (Subrahmanian, 2004). The impact of these developments showed in the reduced number of donors in India n elementary education. The World Bank, DFID and European Commission were the thre e donors who continued funding education. Given the Indian government s reluctance to let the d onor countries influence the path of development, and given the global shift away from a projec t-based approach of aid towards donors working closely with recipient countries national development programmes, the donors adopted an approach which accepted the lead role of the government. DFID India, in particular, developed independent relations with central and state governments f or better efficacy and also collaborated with other donors (including ADB, World Bank, UNICEF, etc.) in the form of trust fund agreements. DFID s programmes have been focused on four selected states (AP, MP, Orissa and WB) with a view to reducing inter-state disparities in achieving development goa ls. Between 2001-02 and 2005-06, the aid expenditure on India by DFID rose from 180 million pounds t o 247 million pounds. In the education sector, the expenditure rose sharply from 9.9 million p ounds in 1999-2000 to 90.6 million pounds in 2004-05. The Government of India had already launched the Sarva Shiksha Abhiyan without a ny financing from bilateral partners. In 2004 the Indian government requested IDA o f World Bank, DFID and EC for assistance. Out of a total project cost of $3.5 billion the dono rs jointly committed to spending around $1.046 billion during the Tenth Plan i.e. around 30 per cent of the total. So Centre had the responsibility of funding 45 per cent ($1.58) and the states 25 per cent ($875 million). World Bank was the major partner giving a loan of nearly half the amount (48 per cent)

while DFID (33 per cent) and EC s (19 per cent) aid was in the form of grants. The Centre passed on this aid to the States as a grant, irrespective of the form in which aid was received. For the Indian education sector those bilaterals not short-listed as donors coul d hardly be termed small in terms of their aid contribution. For instance, the Netherlands p ut in close to Rs 300 crore annually in programmes for women s education and empowerment (Mahila Samakhy a), social science research (Indo-Dutch Programme for Alternatives in Development, IDPAD), and local governance reforms in Kerala. Similarly, the Scandinavians had noteworthy collab oration in the areas of fishery and social forestry and in education projects like Lok Jumbish and Sh iksha Karmi. The annual support from the affected bilaterals, which are now allowed only to suppo rt civil society organizations, and from those UN and other multilateral agencies operating in In dia, is around Rs 700 crores (Sethi, 2003). These donor agencies were asked to fund NGOs directly in different states, albei t with many new restrictions facing NGOs receiving foreign funds. In the past any entity hav ing Foreign Contribution Regulation Act (FCRA) clearance could apply for and receive a grant from these bilaterals. Presently, however, each bilateral43 would have to submit a list of potential grantees providing details of amounts and purpose, which has to receive prior clearance f rom the Department of Economic Affairs, Government of India, before funds could be disbursed (Gover nment of India, 2005). 43 Small bilateral in the sense the bilaterals excluding the six countries mention ed above. 31

Impact on Education Finance Although foreign aid constituted a small proportion of total education finance, there is a need to analyse its net contribution. The usual argument for aid has been that availa ble finance from budgetary sources is spent on maintaining education structure namely salaries of teachers and staff and alternate non-budgetary sources are necessary for new developments. There have b een instances when Indian government went public in asking for aid for education44. Introduction of DPEP helped to relax resource constraints in planning for primar y education (Rani, 2003). In addition to the states own budget for each DPEP districts, the p rogramme made available about Rs.500 to 600 lakhs per year for a district. In 1994-95, the lev el of public spending per district was about Rs.6000 lakh per year in 1994-95, so that the additional resource amounted to an increase of about 10 per cent per year. However there are different opinions about whether foreign aid in education substituted for budgetary expenditure or provided a truly addit ional amount. Tilak (2008) questions the additional part whether it is additional to aggregate spending in base year or additional to normal growth in expenditure or simply additional programme and sc hemes. He argues that availability of foreign funds reduced pressure on both centre and the state government to mobilize additional resources for education. In recent years it is seen that wit h decreased reliance on foreign aid the government has tapped new resources in the form of education ces s. So foreign aid may have suppressed domestic resources to support the flow of education expendit ure. The inefficient utilization of resources in the education sector has been alread y discussed. Foreign funding, too, suffers from this problem. Despite overall progress, disbu rsement lags and other delays are fairly common. In the case of DFID India, for example, In the case of Shiksha Karmi, only half of the committed funds had been disbursed by the end of the project. I n the first three years of the twelve-year Andhra Pradesh District Primary Education operation less than 0.25m was spent of the 46m committed (Heath, 2006). DFID has of late started focusing on strengthening local institutions so that th ere is an improved utilization of funds, among other things. The Indian government has emp hasized the need for increased decentralization in its policy framework, but even with a transfer of responsibilities for primary education to the district level, no capacity-building exercise has been undertaken so that the districts can actually carry out their responsibilities.

A review of recent changes Financing education in India is not a simple process. Firstly there are many pla yers involved, both in its allocation and in its utilization these include several ministries in the central government and the state governments, local bodies, private agencies and foreign donors. Second ly, India is a densely populated country with 13 million out-of-school children between 6 to 13 years a ge even in 2005. So the resource requirement for even universalizing elementary education is massive , let alone for higher and technical education. Last, but not the least, India faces the challenging ta sks of overcoming poverty and unemployment for a large number of its citizens, so that education m ust vie for resources with other, more apparently pressing issues. The earlier discussions on the tren ds and composition of financing have pointed to some major policy changes in the recent years. In this concluding section a review of these changes is attempted. Low priority of Education The analysis of public expenditure in the present paper has focused on the centr al and the state governments. Aggregate education expenditure in current prices has been gr owing during the last fifteen years, but the rate of growth has slowed down in the present decade . Thus, in constant price terms, educational expenditure increased during the second half of the 199 0s, but it stagnated 44 Sadgopal (2003) talked about how Prime Minister Atal Bihari Vajpayee while in augurating the meeting of the high-level group on Education For All (EFA) expressed deep concern about the lack of funds for elementary education in India. Making a strong bid for additional external aid, h e reminded bilateral and multilateral aid agencies, including the World Bank, that in the year 2000 they had made a pledge in Dakar, Senegal, that "no country seriously committed to basic education will be thwarte d in the achievement of this goal by lack of resources 32

from the year 2000-01 onwards. As a proportion of GDP the situation looks worse: the ratio of education expenditure to GDP had declined from more than 4 per cent to 3.5 per c ent from 1990 to mid nineties and recovered to over four per cent around the year 2000. But since that date it has been decreasing again. So, at the macro-level, education has been struggling to maint ain its importance. Changing Priorities within Education Sector As seen in Graph 5 earlier, expenditure on elementary education in real terms ha s not shown a marked increase in real terms. This should be kept in mind when relative prior ities are discussed. However with declining aggregate investment in education, it is seen that elemen tary education has been gaining relative importance since the mid-nineties (from 45.2 per cent in 1 992-93 to 51.5 per cent in 2004-05 RE) at the cost of adult education, higher education as well as technical education. This is translated in a marked decline in resources allocated to these departmen ts in absolute terms (in constant prices). There is a need to rethink this policy of stressing elementary education at the cost of all other sections. Enrolment in Higher and Technical education has been increasing rapidl y in recent years and more than half of the students are enrolled in private institutions. Facilit ies for educational loan are still not developed and at present those enrolled in private institutions ha ve to depend on own resources for their education. CABE Committee Report recommends an increase in G ER at this level from present 12 per cent to 20 per cent. So if those from disadvantaged sections who have completed schooling are expected to be included in this 20 per cent more public resources need to be invested45. While the education subsidy for higher education may be reduced, fees hiked and more private financing encouraged for the upper income strata, there must be a clear governme nt policy that higher education will not be neglected. The same is true for technical education which is even more important for the poorer classes. Of course, additional resources alone cannot p rovide a solution: there is need to reform at the institutional level. However many argue that the vast pool of human resources in India can only be harnessed with a judicious combination of general and technical education, to allow the employment problems to be tackled at the same time as th e lack of elementary education. Increasing importance of the Central government in planning of elementary educat ion The state governments play a major role in financing education in India. But the

bulk of these expenses are on non-plan categories like salaries for teachers and repairs and m aintenance costs. So the thrust and direction for changes in the education sector come from the centr al government, although efforts are under way to provide more scope to the state governments. T he recent focus on the elementary sector has been spearheaded by the central government; the share of this sector in central government s expenditure has increased from 13 per cent in 1990-91 to more than 50 per cent in 2003-04. They have been able to augment resources through new sources of fina nce like education cess and foreign aid but the state governments had very limited scope to increas e their own budgetary revenue. This increase in central expenditure has been transferred to the states via the Planning Commission as Plan expenditure and is largely in the form of Centrally Sponsored Schemes. These CSS play a critical role in the financing process. While some of these schemes h ave been very innovative they had largely remained islands of excellence and were not integrat ed in the mainstream. They also had major weaknesses, particularly in their implementation. Very recen tly the umbrella scheme Sarva Shiksha Abhiyan has been introduced to replace all other schemes in elementary education. Weak Link between Expenditure and Outcomes While the amount of expenditure on education is important, the focus should be o n educational outcomes. It is a striking fact that compared to resources spent in the education sector, there is little to show at elementary level in terms of learning achievement. An excessive focus on outlay has resulted in assessing physical infrastructure creation, provision of teaching and learning materials, appointment of teachers, etc. rather than monitoring the learning pro cess to see how many children have learned what. So in educationally weaker states where new funds ha ve been spent on Prakash, Ved (2007) 33

improving access and infrastructure, positive linkage between enrolment and expe nditure is observed. But links between expenditure and quality indicators like dropout rates and pupi l teacher ratio was found to be nonexistent. The Balancing Act Attempts to limit expenditure and augment resources

The task of spreading education to villages and towns all over India is a major challenge given India s large size and population. The enormous resource requirement for thi s exercise has been aggravated by the decline in tax revenue with the states following adoption of S tructural Adjustment Programme and concomitant fiscal discipline. So it has become a balancing act fo r the governments on the one hand planning priorities demand increased expenditure on education, a nd on the other hand fiscal priorities demand decrease in expenditure. Both the central and the state government are seen to explore additional sources of funds. (i) Both central and state governments have tried - and have been partially successf ul - in increasing revenue collected from direct taxes (share increased from two per cen t to nearly four per cent of GDP). But this was not enough to offset the impact of structura l reforms on indirect taxes. (ii) To meet fund shortages, both the Centre and the States turned to increased borro wing in the 1990s which also have led to higher interest payment burden. (iii) Foreign donors have played a very important role, not so much in terms of quantu m of aid but in terms of nature of intervention. Their funds initially came as partial or ful l support for particular projects or schemes. Their funds were important inputs for many of th e successful CSS. Recently they have been revising their approach and are increasingly workin g through government budgets rather than on individual projects. (iv) The states have little scope to increase expenditure through budgetary sources a nd have attempted to encourage private sector finance in different ways. The norms of re cognition have been simplified and the barriers for setting up private unaided schools hav e been lowered. Private corporate sectors are increasingly being encouraged to support financing of government schools. Many of the newly implemented schemes require enormous suppo rt from the community. This has resulted in proliferation of private schools at all levels which goes against the commitment of free and compulsory elementary education.

(v) The latest in line is application of education cess in successive budgets. A ces s of two per cent on all central taxes was levied in 2004-05, yielding an estimated additiona l resource of Rs 5000 crores. This additional amount facilitated implementation of Midday Meal scheme and SSA. An additional one per cent cess was again imposed in the last budget to increase resources for secondary education. Limiting expenditure has been of first importance, both to the central and state governments. Governments at different levels have tried to improve the efficiency of utilizat ion. For this they had adapted a two-pronged approach of firstly replacing inputs with low cost alternati ves, and secondly by rationalizing fund flows and introducing accountability systems. The first ha s been achieved through adapting some of the successful innovative schemes. The schemes of Shiks ha Karmi, EGS (Education guarantee schools), and NFEs (Non formal education) were introduced i nitially to deal with schooling problems of children in remote areas and out-of-school children u nder difficult circumstances. But they have rapidly been adapted by many state governments as t hey were found to be low-cost and efficient alternatives. To deal with implementation problems the states have started involving local bod ies in school management and plan implementations. So far they have not been very successful. Lack of accountability has been cited as a key factor in failure of effective decentrali zation of primary education in rural India (Pritchett and Pande 2006)46. This is bound to happen i f planning and 46 Pritchett and Pande recommend unbundling the processes of delivering elementa ry education into constituent activities/functions; and unbundling various jurisdictions for service delivery. They use the first principles of public finance and accountability to decide which of the functions/activities sh ould be provided/financed/monitored by the state and, if so, which level of government s hould be responsible. 34

allocation remains centralized and implementation is decentralized. It is hoped that the situation will improve as the process of planning from below is successfully incorporated in SS A. While it is crucially important to mobilize resources and to improve the efficie ncy of spending, there are several pointers to the possibility that the real malady of education in India might lie elsewhere. In fact, it is possible that UEE for India may remain a distant d ream unless the problems of poverty and unemployment are addressed simultaneously with education . It is only when food security and unemployment-related issues have been much reduced in intensit y, and some urgent social issues are addressed that all children will be able effectively to gain a ccess to, and benefit from education. 35

References Aggarwal, Ya. (2000). Public and private partnership in primary education in ind ia - a study of unrecognised schools in Haryana. Operations Research and System management Unit, NIEPA Bashir, S. (2000). Ggovernment expenditure on elementary education in the nineti es. New Delhi: European Commission. Cordella, T. & G. D. Ariccia (1988). Budget support versus project aid. (IMF Wor king Paper, WP/03/88). Washington: International Monetary Fund. Dev, M. & Mooij, J (2002). Social Sector Expenditures In The 1990s: Analysis Of Centre And State Budgets. Economic and Political Weekly, 27 (9). Dreze, J. & H. Gazdar (2002). India: development and participation. New Delhi: O xford University Press. __________________ (1996). Uttar Pradesh: The burden of inertia. Indian developm ent: selected regional perspectives. New Delhi: Oxford University Press. DN (2003). Old Morality and New Realities. Economic and Political Weekly, June 1 4, 2003. Government of India (2007). Sarva Shiksha Abhiyan (SSA) 6th Review mission of Sa rva Shiksa Abhiyan, Aide Memoire. _________________ (2005). Guidelines for development cooperation with bilateral partners, F. No. 1/30/2003-PMU, Ministry of Finance. ___________________(1999). Expert group report on financial requirements for mak ing elementary education a fundamental right. New Delhi, MHRD (mimeo) ____________________(1995). Budgetary resources for education (1951-52 to 1993-9 4), Department of Education. New Delhi: Ministry of human Resource Development. ____________________(1986). National policy on education 1986. New Delhi: Minist ry of Human Development. ____________________(1968). National policy on education 1968. New Delhi: Minist ry of Human Development. ___________________ (various years). Selected educational statistics. New Delhi: MHRD. ____________________(various years). Analysis of budgeted expenditure on educati

on. New Delhi: MHRD. Heath, J. (2006). An evaluation of DFID s India programme 2000-2005. (DFID Evaluat ion Report EV 670 August 2006). Hindustan Times (2007, 27 August). Focus shifts to quality education. Hindustan Times. Jha, Praveen, Subrat Das, Siba Sankar Mohanty & Nandan Kumar Jha (2006). A study on financing of elementary education by the centre and states in India. Draft Report, PESLE. New Delhi: Aga Khan Foundation. Jha, J. & D. Jhingran (2005). Elementary education for the poorest and other dep rived groups. New Delhi: Manohar. Kamath, Sh. (1992). The failure of foreign aid: with special reference to the So vi. HKCER Letters, vol 12, January 1992. Leclercq, F. ( 2003). Education guarantee scheme and primary schooling in Madhya Pradesh, Economic and Political Weekly, May 10. Mehrotra, S. (ed.) (2006). The economics of elementary education in India: the c hallenge of public finance, private provision and household costs. New Delhi: Sage Publications. Mehrotra, S., Panchamukhi, P., Srivastava, R & Srivastava, Ra. (eds.) (2005). Un iversalizing elementary education in India: uncaging the tiger economy, New Delhi: Oxford Uni versity Press. Mehta, A. C. (2006). elementary education in India; Analytical Report 2006 _________ (several years): State Report cards. DISE, NIEPA ___________(several years): District Report Cards. DISE, NIEPA MHRD. (2007). Study of student attendance in primary and upper primary achools: abridged report, Edcil). 36

_______ (2005). Universalisation of secondary education. Report of the CABE Comm ittee, Ministry of Human Resources Development. ________ (2001) Working group report on elementary and adult education 2002-07. Mooij, J. & Mahendra S. Dev (2002). Social sector priorities: an analysis based on budgets and expenditures in India in the 1990s. (IDS working Paper 164). Sussex. Pritchett, L. & Pande, V. (2006). Making primary education work for India s rural poor: a proposal for effective decentralization. (Social Development Working Paper). Washington, DC: the World Bank. Panchamukhi, P.R. & Mehrotra, S. (2005). Assessing public and private provision of elementary education. In Mehrotra et al. (eds) Universalizing elementary education in India . New Delhi: Oxford University Press. Prakash, V. (2007). Trends in growth and financing of higher education in India. Economic and Political Weekly. Pratham (2007). Annual status of education report (Rural) 2007. Mumbai: Pratham Resource Centre. PROBE (1999). Public Report on Basic Education in India. New Delhi: Oxford Unive rsity Press. Ramchandran, V. (1999). External aid in elementary education: a double edged swo rd. Economic and Political weekly, December 11-17. Ramji, V., Sujatha, S. & V.K. Srinivasan (2001). a Study on Management of Public Expenditure by State Governments in India, Indian Institute of Economics. Rani, P. G. (2007). Every child in school: the challenges of attaining and finan cing Education for All in India. In Education for All, Global Promises, National Challenges. Internatio nal Perspectives on Education and Society, Volume 8. __________(2004). Growth and financing of elementary education in Uttar Pradesh: A province in India. Education Policy Analysis Archives, 12(25). __________(2003). Financing primary education and Sarva Shiksha Abhiyan. New Del hi: National Institute of Educational Planning and Administration. Ranis, G. (2006). Toward the enhanced effectiveness of foreign aid. (Centre Disc ussion Paper no.938). Yale University. Govinda, R. M. (2000). Fiscal decentralization in Indian federalism. Institute f

or Social and Economic Change, 2000. Govinda, R. M. & Singh, N. (2001). The political economy of center-state fiscal transfers in India. (Working Paper No. 107, Center For Research On Economic Development And Policy Reform). Sadgopal, A. (2003). Education for Too Few. Frontline, 20(24). Saxena (2006). Central transfers to States and centrally sponsored schemes: Draf t paper for National Advisory Council, retrieved from http://pmindia.gov.in/nac/concept.htm Sethi, H. (2003). Doing without aid? Economic and Political weekly, October 18, 2003. Sipahimalani,V (2000). Financing elementary education in India in the 1990s, Wor ld Bank, South Asia Education Sector Technical Working paper no. 2, New Delhi. SRI (2005). All India survey of out school in 6-13 age group. Social and Rural R esearch Institute Srivastava, R. (2005). Public expenditure on elementary education. In Mehrotra e t al. (eds) Universalizing Elementary Education in India. New Delhi: Oxford University Press . Subrahmanian, R. (2004). The politics of resourcing education: a review of new a id modalities from a gender perspective. Paper for the Beyond Access Seminar Series, Oxford. Swaroop, V., Shikha J. & Rajkumar, A.S. (2007). Fiscal effects of foreign aid in a federal system of governance: the case of India. Journal of Public Economics, 77, 207-330. Thomas, B. S. (1999). Foreign aid adjusts to new market realities. Third World E conomics. Tilak, J. B.G. (2008). Political economy of external aid for education in India. Journal of Asian Public Policy 1(1), 32-51. __________ (2006a). Cess-driven allocations for education. Economic and Politica l weekly, April 8. ____________ (2006b). On allocating 6 per cent of GDP to education, Economic and Political Weekly. ______________(2004). Public subsidies in education in India. Economic and Polit ical Weekly. ______________(2003). Public expenditure on education in India. In Tilak ed: Fin ancing Education in Indi., New Delhi: Ravi Books. 37

______________(1997). Five decades of underinvestment in education, Economic and Political weekly, September 6. ______________(1988). Foreign Aid for Education. International review of Educati on, XXXIV,. Varghese N.V. (2000). Reforming education financing, Sminaire Octobr. Paris: IIEP . World Bank (2006). From schooling access to learning outcomes: An unfinished age nda. An evaluation of World Bank support to primary education, Independent Evaluation Gr oup. Washington D.C.: the World Bank. 38

ANNEXURE Table A1. Share of Centre s and States Expenditure in Budgeted Expenditure on Educa tion Share of centre s Expenditure in Total (%) Share of All States Expenditure in Total (%) Total Budget Expenditure (Rs cr) 1992-93 11.40 88.60 25030.36 1993-94 12.77 87.23 28279.69 1994-95 11.74 88.26 32606.22 1995-96 14.54 85.46 38178.09 1996-97 14.38 85.62 43896.46 1997-98 14.67 85.33 48552.15 1998-99 15.71 84.29 61579.91 1999-2000 14.58 85.42 74816.09 2000-01 12.36 87.64 82486.48 2001-02 17.70 82.30 79881.92 2002-03 18.90 81.10 85507.30 2003-04 19.20 80.80 89079.25 2004-05 (RE) 20.22 79.78 104566.00 2005-06 (BE) 24.30 75.70 119029.90 Note : RE: Revised Estimates, BE: Budget estimate. Source : Analysis of Budgeted Expenditure on Education, Ministry of HRD, Govt. o f India- (various years). Table A2: Share of Public Expenditure on Education in GDP in India, Rs. crore Year Govt. expenditure on education by all Depts. In current prices Govt. expenditure on education by all Depts. In constant prices GDP at current prices(at factor cost) % of Education Expenditure to GDP 1990-91 20491 27792.61 510954 4.01 1991-92 22394 26704.34 589086 3.80 1992-93 25030 27120.04 673221 3.72 1993-94 28280 28279.70 781345 3.62 1994-95 32606 28957.64 917058 3.56 1995-96 38178 31396.46 1073271 3.56 1996-97 43896 34509.80 1243546 3.53 1997-98 48552 36560.35 1390148 3.49

1998-99 61579 43766.11 1598127 3.85 1999-00 74816 51490.77 1761838 4.25 2000-01 82486 52977.83 1902999 4.33 2001-02 79866 49513.76 2081474 3.84 2002-03 85507 51263.36 2254888 3.79 2003-04 89079 50641.98 2519785 3.54 200405( RE) 104566 60853.72 2830465 3.69 Source: Analysis of Budgeted Expenditure on Education, Ministry of HRD, Govt. of India, Various years. GDP figures taken from national income statistics published by CSO 39

Table A 3: Plan and Non-Plan Expenditure on Education (Revenue Account) at current Prices(Rs. Crores) % Distribution Plan Non-Plan Total Plan Non-Plan 1990-91 3247.9 17243.3 20491.0 15.85 84.15 1991-92 3746.9 18846.9 22594.0 16.58 83.42 1992-93 4009.0 21021.3 25030.0 16.02 83.98 1993-94 5181.1 23098.6 28280.0 18.32 81.68 1994-95 6538.7 26067.6 32606.0 20.05 79.95 1995-96 8383.1 29795.0 38178.0 21.96 78.04 1996-97 10305.0 33591.1 43897.0 23.48 76.52 1997-98 10388.0 38164.6 48552.0 21.40 78.61 1998-99 13280.0 48298.6 61579.0 21.57 78.43 1999-00 13864.9 60951.2 74816.1 18.53 81.47 2000-01 13274.3 69212.2 82486.5 16.09 83.91 2001-02 17902.6 61963.1 79865.7 22.42 77.58 2002-03 18627.5 66879.8 85507.3 21.78 78.22 2003-04 19092.3 69986.9 89079.3 21.43 78.57 200405( RE) 27350.0 77216.0 104566.0 26.16 73.84 2005-06 (BE) 37320.2 81709.6 119029.8 31.35 68.65 Note : RE: Revised Estimates, BE: Budget estimate. Source : Analysis of Budgeted Expenditure on Education, Ministry of HRD, Govt. o f India-(various years). Table A4: Plan and Non-plan Expenditure on Education in Central and State Sector s Year Plan Expenditure NonPlan expenditure Share of the centre (per cent) Share of the states (per cent) Total (Rs. crore) Share of the centre (per cent) Share of the states (per cent) Total (Rs. crore) 1992-93 42.18 57.82 4009.00 5.53 94.47 21021.30 1993-94 43.22 56.78 5181.10 5.94 94.06 23098.60 1994-95 38.08 61.92 6538.70 5.13 94.87 26067.60 1995-96 44.99 55.01 8383.10 5.97 94.03 29795.00 1996-7 44.44 55.56 10305.35 5.17 94.83 33591.11 1997-8 48.49 51.51 10387.59 5.46 94.54 38164.55 1998-9 47.92 52.08 13280.29 6.86 93.14 48298.62 1999-2000 48.22 51.78 13864.91 6.93 93.07 60951.18 2000-1 45.37 54.63 13274.32 6.03 93.97 69212.16 2001-2 53.40 46.60 17902.59 7.36 92.64 61963.11

2002-3 61.19 38.81 18627.51 7.11 92.89 66879.83 2003-4 63.08 36.92 19092.33 7.23 92.77 69986.92 2004-5(RE) 57.97 42.03 27350.03 6.84 93.16 77215.97 2005-6 (BE) 62.89 37.11 37320.21 6.68 93.32 81709.67 Source: Calculated from HRD, Govt. of Indiavarious issues. 40 Analysis of Budgeted Expenditure on Education , Ministry of

Table A5: Expenditure on Education at Constant Prices, 1993/04, Rs. Crores Year Expenditure by Education departments Expenditure by all departments on Education Centre All states/UTs Centre All states/UTs 1990-1 2228.72 21091.44 24511.12 1991-2 2044.02 20324.26 23745.06 1992-3 1947.39 20755.49 3092.54 24028.24 1993-4 2096.33 21316.78 3610.79 24668.80 1994-5 2276.21 21908.70 3398.54 25559.02 1995-6 2727.82 23693.07 4564.57 26831.88 1996-7 2886.88 25707.16 4964.10 29545.69 1997-8 3481.28 27474.53 5362.95 31197.39 1998-9 4494.20 31913.23 6876.73 36889.37 1999-2000 5046.54 37129.28 7506.44 43984.33 2000-1 5090.08 35049.99 6548.46 46429.37 2001-2 4982.63 35220.53 8753.58 40760.19 2002-3 5449.18 35654.85 9686.22 41577.15 2003-4 5785.94 35740.45 9721.98 40920.00 2004-5(RE) 6993.06 38685.52 11286.19 44541.90 2005-6 (BE) 9374.74 39822.94 14789.75 46063.97 Source: Calculated from Analysis of Budgeted Expenditure on Education , Ministry of HRD, Govt. of Indiavarious issues and Office of the Economic Adviser, Ministry of Commerce and Indu stry (price index from table 12.2, Index number of wholesale price average for months) 41

Table A 6: Sectoral Composition of Expenditure by Education Departments, % Year All state and UTS Elementary Secondary Higher Technical Other Total 1990-1 49.71 33.08 11.81 2.86 2.54 100 1991-2 49.29 33.98 11.43 2.90 2.40 100 1992-3 45.23 34.26 12.89 4.33 3.30 100 1993-4 46.22 33.14 13.26 4.35 3.04 100 1994-5 49.01 34.28 11.52 2.94 2.26 100 1995-6 49.62 33.51 11.41 3.03 2.42 100 1996-7 49.81 33.69 10.92 2.83 2.74 100 1997-8 49.76 34.44 10.75 2.83 2.23 100 1998-9 49.80 35.05 10.06 2.81 2.27 100 1999-2000 46.45 36.66 11.21 2.67 3.01 100 2000-1 48.82 34.05 12.66 2.61 1.86 100 2001-2 50.91 33.80 11.34 2.32 1.64 100 2002-3 49.12 34.91 11.95 2.42 1.59 100 2003-4 49.57 34.95 11.61 2.28 1.59 100 2004-5(RE) 50.86 33.75 11.04 2.53 1.81 100 2005-6 (BE) 51.01 33.53 11.02 2.72 1.73 100 Centre 1990-1 13.74 23.52 28.94 18.74 15.07 100 1991-2 16.50 23.77 28.92 18.43 12.38 100 1992-3 17.60 24.86 28.09 18.52 10.93 100 1993-4 18.59 26.72 24.53 19.33 10.83 100 1994-5 21.39 23.15 26.70 18.13 10.64 100 1995-6 39.55 19.93 19.89 14.04 6.59 100 1996-7 42.53 19.55 19.51 14.28 4.13 100 1997-8 48.37 15.08 20.29 12.77 3.49 100 1998-9 43.51 15.54 25.30 12.81 2.83 100 1999-2000 38.85 14.53 30.02 13.89 2.72 100 2000-1 39.35 14.63 28.84 13.94 3.25 100 2001-2 44.44 15.32 20.50 15.45 4.29 100 2002-3 46.87 14.17 19.27 15.19 4.50 100 2003-4 51.13 13.53 17.31 13.76 4.28 100 2004-5(RE) 54.64 10.28 15.13 10.90 9.06 100 2005-6 (BE) 61.19 8.68 11.50 8.73 9.91 100 Source: Calculated from HRD, Govt. of Indiavarious issues. 42 Analysis of Budgeted Expenditure on Education , Ministry of

Table A 7. Distribution Of Central Expenditure Across Centrally Sponsored Scheme s In Elementary Education, % 1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-2000 20 Teacher Training 14.9 26.4 28.4 21.8 8.9 6.4 3.9 5.8 5.4 NFE 14.1 14.3 46.6 37.0 12.7 10.1 8.1 5.8 5.4 OB 62.0 49.9 7.8 6.1 22.3 17.8 13.3 11.1 10.2 Midday Meal 0.0 0.0 0.0 0.0 36.7 51.0 47.2 51.0 51.1 SSA (non EAP) 0 0 0 0 0 0 0 0 0 DPEP (EAP*) 0.0 0.0 1.5 26.5 16.7 11.7 24.7 20.1 20.4 Other EAP 0.0 0.0 7.9 5.1 2.3 2.3 2.3 2.2 2.2 Other 8.9 9.4 7.8 3.6 0.4 0.7 0.5 4.0 5.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 1 Source: Srivastava (2005) in Mehrotra et al. Note: *Externally-aided programme 43

Table A8: Statewise Distribution of CSS Funds (1990/91- 1998/99), % State Operation Blackboard Non Formal education Teacher Education DPEP Maharashtra 17.4 0.5 1.4 6.1 Karnataka 10.9 0.1 8.5 6.1 Orissa 10.7 7.9 3.7 3.4 Andhra Pradesh 9.3 11.9 7.6 5.3 Rajasthan 8.6 8.8 12.0 1.5 Bihar 8.1 15.6 2.3 4.1 Uttar Pradesh 6.6 28.3 10.3 5.2 Assam 5.6 5.6 3.2 4.6 Madhya Pradesh 4.5 17.6 11.5 27.7 Gujarat 4.3 0.3 4.1 7.6 West Bengal 2.5 0.3 1.5 1.6 Himachal Pradesh 2.1 0.1 3.7 3.3 Tamil Nadu 1.3 0.8 10.2 11.1 Kerala 0.7 0.0 3.7 3.9 Haryana 0.5 0.2 2.0 5.8 Total for the period (in Rs crores) 1958.99 923.09 648.48 1307.53 Source: Bashir (2000) Table A 9: Per Capita Education Expenditure at Constant Prices, 1993/04, Rs State Haryana Himachal Pradesh Kerala Madhya Pradesh Meghalaya Orissa Rajasthan 1990-91 261 479 355 179 427 190 246 1991-92 256 452 337 176 453 203 235 1992-93 270 465 331 175 449 207 251 1993-94 262 463 379 177 490 210 265 1994-95 265 443 394 170 393 219 273 1995-96 293 492 441 187 451 232 290 1996-97 312 530 402 196 618 240 310 1997-98 333 635 414 195 454 258 315 1998-99 442 761 431 234 501 295 377 1999-2000 430 824 549 259 556 352 396 2000-01 425 793 509 214 558 301 377 2001-02 416 867 471 208 592 284 369 2002-03 411 790 604 506 301 353 2003-04 409 759 519 511 284 362 Source: Calculated from Analysis of Budgeted Expenditure on Education , Ministry of HRD, Govt. of Indiavarious issues, Office of the Economic Adviser, Ministry of Commerce and Industr

y (price index from table 12.2, Index number of wholesale price average for months), and population 44

Table A10: Expenditure and Outcome Indicators, 2003/04 State Per capita NSDP (Rs.) Per student expenditure on elem. Education (Rs.) GER primary (6 to 10 years) GER middle (11 to 13 years) Gender parity index Dropout rate (class 1 to 8)Boys Girls Boys Girls Primary Middle Haryana 16773 1303 73.5 77.3 68.2 62.4 0.91 0.93 21.26 Himachal Pradesh 14246 2758 106.1 106.9 99.9 96.5 0.92 0.92 14.28 Kerala 13924 1751 97.3 96.6 95.7 91.5 0.97 0.93 -9.54 Madhya Pradesh 7965 880 112.1 100.7 71.8 53.9 0.90 0.72 46.81 Meghalaya 10310 1515 104.2 106.9 60.3 62.0 1.03 1.11 71.13 Orissa 7043 959 120.2 109.4 74.3 47.2 0.92 0.83 68.5 Rajasthan 88042 937 114.2 107.4 58.1 44.7 0.86 0.54 61.72 Source: Selected Educational Statistics (2003-4), NIEPA Graph A1 Expenditure on Education at Constant Prices,1993/04 0 250 500 750 1000 1250 1500 1750 2000 2250 199091 199192 199293 199394 199495

199596 199697 199798 199899 19992000 200001 200102 200203 200304 year Rs. crores Haryana H.P. Kerala M.P. Meghalaya Orissa Rajasthan 45

Table A11: Growth of Recognised Educational Institutions in India, 1990/91 2004/05 Years Primary Upper Primary High/Hr. Sec/ Inter /Pre. Jr. Colleges Colleges for General Education Colleges for Profession al Education Universitie s/ Deemed Univ../ Instt of National Importance 1990-91 560935 151456 1991-92 566744 155926 1992-93 571248 158498 1993-94 570455 162804 1994-95 586810 168772 1995-96 593410 174145 1996-97 603646 180293 1997-98 619222 185961 1998-99 626737 190166 1999-00 641695 198004 2000-01 638738 206269 2001-02 664041 219626 2002-03 651382 245274 2003-04 712239 262286

79796 4862 886 184 82576 5058 950 196 84608 5334 989 207 89226 5639 1125 213 94946 6089 1230 219 99274 6569 1354 226 103241 6759 1770 228 107140 7199 2075 229 112438 7494 2113 237 116820 7782 2124 244 126047 7929 2223 254 133492 8737 2409 272 137207 9166 2610 304 145362 9427 2751 304

Source: Selected Education Statistics, 2004 Table A12: Gross Enrolment Ratios ( Ger), by School Type, 1990/91 Primary(I-V) Upper Primary (VI - VIII) Elementary(I-VIII) Year Boys Girls Total Boys Girls Total Boys Girls Total 1990-91 114 85.5 100 76.6 47 62.1 100 70.8 86 1991-92 113 86.9 100 75.1 49.6 63.1 101 73.2 87.7 1992-93 95 73.5 84.6 72.5 48.9 64.1 87.7 65.7 77.2 1993-94 90 73.1 81.9 62.1 45.4 65.1 80.2 63.7 72.3 1994-95 96.6 78.2 87.7 68.9 50 66.1 87.2 68.8 78.4 1995-96 97.1 79.4 88.6 67.8 49.8 67.1 86.9 69.4 78.5 1996-97 97 80.1 88.8 65.8 49.2 68.1 85.9 69.4 78 1997-98 99.3 82.2 91.1 66.3 49.7 69.1 87.4 70.7 79.4 1998-99 101 82.9 92.1 65.3 49.1 70.1 87.6 70.6 79.4 1999-00 104 85.2 94.9 67.2 49.7 71.1 90.1 72 81.3 2000-01 105 85.9 95.7 66.7 49.9 58.6 90.3 72.4 81.6 2001-02 105 86.9 96.3 67.8 52.1 60.2 90.7 73.6 82.4 2002-03 98 93 95 65 56 61 85 79 83 2003-04 101 96 98 67 58 62 88 81 85 2004/05

Source: Selected Education Statistics, 2004 46

Table A 13: Increase In Number Of Teachers ('000) School Type, 1990/91 Year Primary Upper Primary High / Hr. Secondary/ Intermediate Male Female Total Male Female Total Male Female Total 1990-91 1143 473 1616 717 356 1073 917 417 1334 1991-92 1152 492 1644 714 365 1079 931 450 1381 1992-93 1137 514 1651 709 376 1085 941 454 1395 1993-94 1110 513 1623 723 406 1124 953 492 1445 1994-95 1157 531 1688 746 410 1156 986 495 1481 1995-96 1176 558 1734 758 424 1182 1030 519 1549 1996-97 1190 566 1756 769 431 1200 1069 544 1613 1997-98 1226 597 1823 640 597 1237 1086 558 1644 1998-99* 1246 658 1904 814 464 1278 1168 579 1747 1999-00* 1236 683 1919 829 469 1298 1142 578 1720 2000-01* 1221 675 1896 820 506 1326 1184 577 1761 2001-02* 1213 715 1928 921 547 1468 1157 620 1777 2002-03 1167 746 1913 936 645 1581 1221 812 2033 2003-04 1260 837 2097 944 648 1592 1250 774 2024 * Provisional Source: Selected Education Statistics, 2004 Table A14: Pupil-Teacher Ration (PTR) by School Type, 1990/91 Year Primary Upper Primary Sec. /Sr. Secondary 1990-91 43 37 31 1995-96 43 37 32 1996-97 43 37 32 1997-98 42 37 32 1998-99 43 36 32 1999-00* 43 38 32 2000-01* 43 38 32 2001-02* 43 34 34 2002-03* 42 34 33 2003-04* 45 35 33 2004-05* 46 35 33 * Provisional Source: Selected Education Statistics, 2004 47 2004/05

2004/05

Table A 15: Drop-Out Rates at Primary, Elementary & Secondary Stages, 1990/91 04/05 Primary (I-V) Elementary (I-VIII) Secondary (I-X) Year Boys Girls Total Boys Girls Total Boys Girls Total 1990-91 40.1 46.0 42.6 59.1 65.1 60.9 67.5 76.9 71.3 1992-93 43.8 46.7 45.0 58.2 65.2 61.1 70.0 77.3 72.9 1995-96 41.4 43.0 42.1 56.6 61.7 58.8 66.7 73.7 69.6 1996-97 39.7 40.9 40.2 54.3 59.5 56.5 67.3 73.7 70.0 1997-98 37.5 41.5 39.2 53.8 59.3 56.1 66.6 73.0 69.3 1998-99 40.9 42.3 41.5 54.2 59.2 56.3 64.5 69.8 66.7 1999-00* 38.7 42.3 40.3 52.0 58.0 54.5 66.6 70.6 68.3 2000-01* 39.7 41.9 40.7 50.3 57.7 53.7 66.4 71.5 68.6 2001-02* 38.4 39.9 39.0 52.9 56.9 54.6 64.2 68.6 66.0 2002-03* 35.85 33.72 34.89 52.28 53.45 52.79 60.72 64.97 62.58 2003-04* 33.74 28.57 31.47 51.85 52.92 52.32 60.98 64.92 62.69 2004. 05* 31.81 25.42 29.00 50.49 51.28 50.84 60.41 63.88 61.92 *provisional Source: Selected Education Statistics, 2004 48

20

Graph A2 49

Graph A3 50

51