This paper analyses the structure of non-tax sources of the states. The major thrust is on presenting a detailed analysis of six select services drawn from social and economic services. While education, sports, arts and culture; medical and public health; and water supply and sanitation have been selected from social services, major and medium irrigation; minor irrigation; and roads and bridges have been chosen from economic services. The analysis of user charges at the disaggregated level for each of the services provided in different states is based on the data drawn from the state budget documents. A comparative analysis of the recovery rate over time is based on the results for two points of time, i e, 1993-94 to 1995-96 and 2001-02 to 2003-04.
SPECIAL ARTICLEjanuary 31, 2009 EPW Economic & Political Weekly54Mobilising Non-Tax Revenue: An Empirical Analysis of Trends in StatesMahesh C Purohit, Vishnu Kanta PurohitThis paper analyses the structure of non-tax sources of the states. The major thrust is on presenting a detailed analysis of six select services drawn from social and economic services. While education, sports, arts and culture; medical and public health; and water supply and sanitation have been selected from social services, major and medium irrigation; minor irrigation; and roads and bridges have been chosen from economic services. The analysis of user charges at the disaggregated level for each of the services provided in different states is based on the data drawn from the state budget documents. A comparative analysis of the recovery rate over time is based on the results for two points of time, i e, 1993-94 to 1995-96 and 2001-02 to 2003-04.The paper is an outcome of the study titled “Mobilising Resources through Reform of State Non-Tax Sources for Plan Development” submitted to the Planning Commission, New Delhi.Mahesh C Purohit (maheshpur@gmail.com) is with the Foundation for Public Economics and Policy Research, New Delhi, and Vishnu Kanta Purohit is with Indraprastha College for Women, Delhi.Augmenting resourcesis an exercise in balancing current revenues, economic growth and political economy. The political economy of mobilising resources, therefore, has to consider issues related to both, the economic effects and the fallout in terms of politics.1 IntroductionMobilising resources through reforms in non-tax sources of the states serves the twin purpose of having a rational non-tax structure and generating greater means to achieve economic growth. Irrational structure of non-tax sources has adverse economic effects that invalidate growth objectives. From the economic point of view, therefore, one has to keep in mind the objectives of equity, efficiency and neutrality; especially the impact of these on the economic growth of the economy. To fulfil these objectives, one often tends to adopt an econo-mically rational structure of non-tax sources that may not be palatable politically. 1.1 Objectives of the StudyThis study aims at presenting measures for mobilising resources via reforms in non-tax sources. In doing so, it has the objectives of (a) analysing the structure of non-tax sources of Indian states; (b) examining the structural reforms; and (c) suggesting a rational non-tax structure; which should be economically viable and should yield larger resources. Keeping these objectives in view, the study hypothesises that the non-tax revenues, which account for about 19% of the states own revenue, should play a major role in states’ own resources. This is imperative given the widening revenue-expenditure gaps in the states’ budgets and the demand to reduce the quantum of subsidy to meet the cost of public services through proper pricing. 1.2 Taxonomy of Non-Tax ResourcesNon-tax sources are defined as payment made to the government for which there is a quid pro quo.1 However, these non-tax sourcesdo not have similar features and are classified into three categories:First, there are some sources that are compulsory and requited payments. These sources include penalties (other than penalties on non-compliance of taxes) and fines (Musgrave and Musgrave 1973). The second sources comprise voluntary and unrequited payments. These payments include donations and contributions made to the government or any unclaimed funds lying with the government. The third category consists of voluntary and
SPECIAL ARTICLEEconomic & Political Weekly EPW january 31, 200955requited payments, including revenue earned from the resources owned by the government such as forest, marine, riparian habi-tats and wildlife. This category also includes revenue earned by sale of usage rights, admission fee and royalties and rental pay-ments received by the government. Income earned in the form of dividends and the interest receipts from investments made by the government also fall into this category. 1.3 Scope of the StudyThough the term “non-tax revenue” encompasses all the above components, to delimit the scope of this study and to have a thor-ough analysis of a few selected services, some of the non-tax sources given below have not been included in this study. First, payments made to the government in the form of subscription to a loan or to a long-term saving scheme, which isrepayable by the government in the future (that is, the payment made to the government is required by future payments orby transfer of other assets) is considered as “capital receipts” and receipts from the government borrowings, money creation or proceeds of disinvestments is not included in the scope of non-tax revenue. Second, revenue receipts of the government through the sale of goods and services (that are commercial in nature) are also not included. Third, in some non-tax receipts it is observed that not all receipts by the government are accounted for in its consoli-dated fund, e g, in case of education, although the students pay a variety of fees, only the tuition fees are credited to the treasury (Tilak 1993). Similarly, in medical and public health, it is observed that in some states the user charges paid by outdoor patients are not put into the consolidated fund. In Rajasthan, this is credited to the welfare of the hospital staff and in Tamil Nadu it is consid-ered to be a part of the maintenance expenditure of the hospital. Such variations hamper a comparative analysis of recovery rate of different services. Fourth, the revenue from lotteries is netted out because the purpose of this activity is not to provide any public service but to generate revenue. In many states, a major chunk of the revenue from lotteries is used to make payment for lottery prizes. Only a very small proportion of this revenue remains with the state government. The same holds true for other commercial acti-vities. Therefore, this study uses the concept of net non-tax revenue, i e,gross non-tax revenue minus expenditure for these commercial activities.Finally, notional receipts are excluded from the purview of this study, e g, interest on capital works in irrigation, which is matched by contra-entry in many states (signifying notional expenditure).The non-tax sources covered in the study include the following:(i) Administrative Non-Tax Receipts: This source accounts for about three-fourths of the states’ own non-tax revenue. In the future, this is likely to be the most productive and reliable sourceof non-tax revenues for the states. There are over 100 departmental sources of non-tax revenues in different states which are classified under three dozen heads. Thus, in attempt-ing a broad study of all the items under non-tax sources, the effort is to cover some aspects from each of the three broad components of administrative receipts, viz, gen-eral services, social services, and economic services.(a) Receipts from general serv-ices: These comprise receipts from the Public Service Commission, police, jails, supplies and disposals, stationery and printing, public works, other administrative serv-ices, contribution and recoveries towards pension and other retire-ment benefits, and other miscel-laneous general services.(b) Receipts from social services: Major items under this category include education, sports, arts and culture, medical and public health, family welfare, water supply and sanitation, housing, urban develop-ment, information and publicity, labour and employment, social security and welfare, and other social services.(c)Receipts from economic services: Major items under this class are crop husbandry, animal husbandry, dairy development, fisheries, forestry and wildlife, cooperation, other agricultural and rural programmes, special area programmes, major and medium irrigation, minor irrigation, village and small-scale industries, industries, non-ferrous mining and metallurgical industries, roads and bridges, tourism, and others. There are six major individual contributors to the states’ non-tax revenue. These are education, sports, art and culture; medical, public health and family welfare; water supply and sanitation; major and medium irrigation; minor irrigation; and roads and bridges. These account for a major share in the administrative component of the non-tax revenue in different states.Table 1: Composition of States’ Own Non-Tax Revenue (as % of total in each state)States General Services Social Services Economic Services 1993-94 2003-04 G* B* 1993-94 2003-04 G* B* 1993-94 2003-04 G* B*Andhra Pradesh 3.91 7.61 13.88 1.15 5.25 6.91 19.86 1.57 38.61 33.67 8.38 0.68Bihar 2.6527.0922.222.243.5528.0611.791.1490.1038.12-16.27-1.68Goa 2.841.2527.53 2.05 9.5110.47 21.27 1.3285.39 87.97 19.76 1.21Gujarat 3.819.1124.06 1.995.005.9616.76 1.3034.04 56.59 13.97 1.16Haryana 64.36 30.51 -12.56-1.10 3.0011.98 18.86 1.4723.88 35.82 9.410.76Karnataka 9.4760.30 24.16 1.737.234.179.67 0.82 36.64 31.20 12.19 0.99Kerala 27.55 38.05 11.74 0.9612.96 15.82 10.38 0.8149.73 39.64 3.97 0.35Madhya Pradesh 5.98 8.49 8.81 0.88 3.67 5.49 6.56 0.70 74.91 84.15 -0.26 0.05Maharashtra 9.6627.52 15.48 1.366.2510.76 10.77 1.0045.01 51.13 5.320.49Orissa 9.995.132.750.336.765.9011.631.0962.0061.336.600.67Punjab 16.3155.5827.342.4311.022.279.320.8654.0410.735.370.50Rajasthan 22.23 22.73 -5.64 -0.59 6.63 12.21 10.99 0.97 19.11 31.86 10.63 0.94Tamil Nadu 10.43 16.29 18.39 1.59 12.46 22.77 15.77 1.35 35.95 34.21 11.99 1.03Uttar Pradesh 49.79 12.46 -11.33 -1.46 3.74 15.66 19.73 1.94 25.52 42.69 8.24 0.81West Bengal 18.97 29.54 12.15 0.94 19.15 15.62 8.18 0.62 38.23 36.59 10.20 0.75Major 15 States 19.24 26.66 5.70 0.50 6.01 9.34 13.84 1.21 42.89 39.89 6.30 0.57G* = Growth rate for the period 1993-94 to 2003-04 and B* = Buoyancy for the period 1993-94 to 2003-04.
SPECIAL ARTICLEjanuary 31, 2009 EPW Economic & Political Weekly561.4 Methodology of the StudyTo analyse the efforts of the states in collecting appropriate user charges for the services provided, the study presents an estimate of the recovery rate (RR) from services as a percentage of revenue expenditure (RE) incurred on services. To facilitate comparison betweenRRi (recovery rate from the ith service) amongst the states, the study presents a normative approach to estimate “norms” of RR for each of the services by using panel data models.The study attempts to analyse differences between the actual and normative RR (estimated through the regression approach) to appraise the performance of the states in collecting RR for the select services. This is done by takingRR of a specific service as a function of related independent variables. The above exercise is based on data collected from the original source, viz, state budget documents for 15 major states.2 Also, it uses sources of data from Reserve Bank of India (RBI), Central Statistical Organisation, Bureau of Economics and Statistics, thevarious state departments dealing with different non-tax services, etc.2 A Theoretical PerspectiveGovernment intervention in providing goods and services for public welfare can be justified on the grounds of market imper-fection in provisions of these goods. The magnitude and the mix of these goods provided by the government, however, depend on the level of economic development and market imperfection.“Pure” public goods are the ones having features of non- excludability and non-rival consumption/joint consumption. In fact very few public goods are pure public goods. First, the joint consumption property might be partly violated. Although all individuals may consume the services in the same quantity, the benefit obtained may be reduced if more persons have a share in its consumption. This phenomenon is known as congestion or partial rivalness. Public goods can be distinguished on the basis of variability in their use, which occurs due to the opportunity cost of these public services (Oakland 1987). Second, goods pro-vided by the government may have joint consumption features but these may be excludable. Such goods and services may be termed as “price-excludable public goods” (Burns and Walsh 1981). Such services include television and radio transmission and services provided by transportation facilities, entertainment facilities (theatres, museum), recreational facilities (national parks, stadium) and information services.Samuelson stated that pure public goods have to be financed through taxes while the quasi-public goods could be subsidised or regulated by the government through a pricing mechanism. Mixed goods may be financed partially by user charges and partly through a combination of subsidy and user charges.In practice, however, considering government as a natural monopoly, various models of utility pricing have been developed. These are marginal cost (MC) pricing theory, average cost (AC) pricing theory and price discrimination theory based on a combi-nation of MC andAC pricing theories, viz, economic pricing, fully distributed cost pricing, demand-compatible pricing, uniform pricing, non-uniform pricing, multi-part tariff, two-part tariff, myopic pricing rules, spot pricing, Boiteux pricing, Ramsey pricing, etc.Clark applied the MC pricing principle to the public utilities keeping in view the objectives of equity and efficiency. However, there are both theoretical and practical problems related with marginal cost pricing. First, the assumptions on which the theory of allocative efficiency is based generally do not hold; and second, implementation may not be feasible.The earlier view of financing of government expenditure through budgetary resource was justified on the ground that the benefit of pure public goods are expected to accrue to the society as a whole. Therefore, the marginal cost pricing principle that is generally applicable for the pricing of private goods and services could not be applied to public goods. The government is expected to recover the cost either through taxes or public borrowings. The marginal cost pricing approach implies that in general (a) the Table 2: Yearly Average RR/RE of Social Services and Select Components(in %)States SocialServicesESACMPHWSS 1993-96 2001-04 1993-96 2001-04 1993-96 2001-04 1993-96 2001-04Andhra Pradesh 1.86 2.94 1.54 2.49 3.59 3.48 1.92 5.32Bihar 1.141.670.150.842.593.651.330.26Goa 8.0112.97 0.582.114.83 6.97 45.65 61.75Gujarat 2.523.291.111.618.465.720.140.5Haryana 3.816.592.31.849.1910.328.5111.13Karnataka 2.442.321.091.0166.220.810.38Kerala 2.402.222.012.335.823.760.11.2MadhyaPradesh 1.611.610.440.482.26 2.48 4.08 5.66Maharashtra 2.892.360.90.578.696.313.040.85Orissa 2.162.041.321.112.592.556.7111.23Punjab 3.732.870.790.838.686.5410.319.58Rajasthan 3.813.250.511.263.882.7223.0819.03Tamil Nadu 2.54 3.97 1.44 2.2 5.86 7.15 1.03 3.77Uttar Pradesh 1.73 3.32 1.4 3.38 2.24 3.05 0.04 0.26West Bengal 1.74 1.20 0.45 0.58 6.53 4.04 0.94 1.37Major 15 States 2.35 2.80 1.04 1.51 5.29 4.76 5 7.27ESAC = Education, Sports, Arts and Culture, MPH = Medical, Public Health and Family Welfare, WSS = Water Supply and Water Sanitation.Source: Reserve Bank of India,State Finance: A Study of State Budgets, various issues.Table 3: Yearly Average RR/RE of Economic Services and Select Components(in %)States EconomicServicesMMIMIR&B 1993-96 2001-04 1993-96 2001-04 1993-96 2001-04 1993-96 2001-04Andhra Pradesh 24.60 17.01 15.04 0.74 14.66 1.26 8.16 7.94Bihar 44.33 7.27 28.68 12.72 0.610.79 6.734.83Goa 80.15 99.41 17.24 108.3 7.63 66.67 24.15 12.07Gujarat 18.5023.316.0716.22.824.970.882.61Haryana 29.2527.927.1727.660.220.360.120.64Karnataka 14.18 15.35 4.8526.55 2.01 4.26 3.57 7.34Kerala 20.8410.147.096.591.351.888.364.76Madhya Pradesh 52.16 27.12 23.46 13.75 18.3 26.55 5.63 1.56Maharashtra 24.5827.518.3712.497.045.4771.350.12Orissa 45.15 36.29 17.74 32.16 2.213.155 9.38Punjab 30.5316.7221.216.120.940.260.810.41Rajasthan 14.98 21.41 6.83 4.28 32.82 34.29 1.811.94TamilNadu 9.5912.070.28 0.415.8612.92 5.6 7.12Uttar Pradesh 12.56 9.63 19.49 10.78 5.94 6.03 6.18 4.03West Bengal 7.75 9.45 3.51 2.45 4.78 4.02 7.24 8.43Major 15 States 23.31 18.08 8.67 7.57 6.53 6.13 5.25 4.57MMI = Major and Medium Irrigation, MI = Minor Irrigation, R&B = Roads and Bridges.Source: Reserve Bank of India,State Finance: A Study of State Budgets, various issues.
1993 96 2001 04
1993 96 2001 04
1993 96 2001 04
1993 96 2001 04
1993 96 2001 04
1993 96 2001 04
SPECIAL ARTICLEEconomic & Political Weekly EPW january 31, 200961It is due to these differences in the socio-economic factors across the states that there exist substantial variations in non-tax recoveries and the desired RR/RE. In the case of the services, where the desired RR/RE are higher than the actual RR/RE, scope for upward revisions exists. The difference in the actualRR/RE and the desired RR/RE indicates the inefficiencies in the recovery enforcement. In such cases, there is need to revise the rates upwardly. Given this framework of conceptual and practical limitations, this study adopts the method given below for estimating norms for desired RR/RE. The model for fixing norms of non-tax revenues is taken as follows:Let Yi (where i=1, 2, ..n) denote the non-tax revenue on a particular item collected by the ith state whose GSDP isXi. LetZi denote a vector of characteristics of the ithstate which are considered important for determining the level of collection of non-tax revenues for a particular service. We fit a model having the following functional form:(1) Yi = α + β Xi + γ Zi + ui where α, β, and γare constants and uiis a random variable that takes only positive (non-negative) values.The above function in equation (1) can be estimated from the data for different states for each component of non-tax revenue and the fitted value, viz:(2) Y^i = α^ + β^ Xi + γ^ Zi can be taken as the norm for that particular component of non-tax revenue.To estimate the norms for six select services, the following equations were set up: Equations for Regression1 recedu = {pcgsdp, urban} For Phase I (1993-94 to 1995-96) {pcgsdp, urban, sch, tch} For Phase II (2001-02 to 2003-04) 2 recmed = {pcgsdp, nodoc} For Phase I {pcgsdp, nodoc, beds, For Phase II hosp, dis} 3 recwss = {pcgsdp, urban} Same for both the phases4 recmajir = {pcgsdp, gsdpagr, Same for both the phases grsirrarea} 5 recmirr = {pcgsdp, gsdpagr} For Phase I {pcgsdp, gsdpagr, For Phase II grsirrarea} 6 recroad = {pcgsdp, gsdptran, rdl} For Phase I {pcgsdp, gsdptran, For Phase II dl, nov} where recedu = RR/RE for education, arts and culture; recmed =RR/RE for medical and public health; recwss =RR/RE for water and sanitation; recmajir =RR/RE for major irrigation; recmirr =RR/RE for minor irrigation; recroad = RR/RE for roads and tolls.In the above equations, for each of the six select services, the percentage share of recovery rate to the corresponding expendi-tures has been calculated for the two phases (1993-94 to 1995-96, termed as phase I) and 2001-02 to 2003-04 (termed as phase II), across the 15 major states. The average RR/RE is then obtained for each of the service for both the phases. The change in this average percentage share in the two phases indicates the efforts of the major states in mobilising resources through non-tax sources (Exhibits 1 to 6).These regression equations give estimated RR/RE for each select service by using the average of actual percentage share of revenue receipts of a service to the corresponding revenue expenditures for the concerned service provided by the govern-ment as the dependent variable. The independent variables are: per capita income (pcgsdp), proportion of relevant sectoral income in totalgsdp (gsdpcons for construction, gsdpagr for agriculture, gsdptran for transport), urbanisation (urban), availability of educational facilities (sch for schools, tch for teachers), availability of medical facilities (nodoc for number of doctors, beds for beds in hospitals, hosp for hospitals, dis for dis-pensaries), gross irrigated area (grsirrarea), road length (rdl), number of vehicles (nov), etc. Also, 15 state-wise dummy variables have been applied to get the state-wise estimated RR/RE for each service.The all-states norm is calculated by using the “normative approach” in the above regression equations. This method, thus, estimates the “norm” for RR/RE of each of the service by taking the deviation from the all-states average. The difference bet-ween the actualRR/RE for each service across the states and the desired RR/RE for each service from the all-states average gives the magnitude by which the states need to increase their existing recoveries. The results of the above regressions, fitted for all the services in the 15 major states, yield the norms for all the services.The study then adopts a method of comparing the adequacy of the actual average RR/RE with the estimatedRR/RE of the best performing state from amongst all the states. For each of the selected service, the norm can be the state with the highest RR/RE among the major states. Thus, in the case of education, the desired RR/RE of Haryana is highest in the first phase of our study and maximum for Uttar Pradesh in the second phase. These are taken as the norm for the education in this study. The same meth-odology is adopted for all the six services.The norms estimated on the basis of the related variables determine the relative efficiency of states when the state govern-ment is viewed as a multi-service provider and the user charges as price-like instruments, which restrict the use of the service to only the “paying” consumers.5 Conclusion and Policy ImperativesTo conclude, non-tax sources are not a fiscally significant sourceof revenue in the states’ budget and their growth is not keeping pace with other components of revenue receipts. Non-tax revenue can be a major source of budgetary receipts for the state governments if proper attention is paid towards pricing of the services. Its importance is now being realised in the context of bridging fiscal deficits of the states and the heavy financial requirements for upgrading and modernising basic infrastructure.
SPECIAL ARTICLEjanuary 31, 2009 EPW Economic & Political Weekly62Notes1 This is in contrast with the definition of tax as a compulsory payment made to the government for which there is no quid pro quo. See Dalton (1949).2 The major states include Andhra Pradesh, Bihar, Goa, Gujarat, Haryana, Karnataka, Kerala, Madhya Pradesh, Maharashtra, Orissa, Punjab, Rajasthan, Uttar Pradesh, Tamil Nadu and West Bengal. 3 Such services include education, health, and sani-tation, etc, Downing (1992).4In the literature on public finance, it is largely accepted that the marginal cost pricing principle is applicable to the provision of government serv-ices so long as (i) the benefits of the service accrue to the masses, and the marginal expenditure exceeds the marginal cost, and (ii) the marginal benefits of financing the service is equal to the possible marginal benefit from other competing investments. 5 In Berkeley, California (US), for example, a sur-vey was conducted to find out from the society whether the BART (train service) railway line be underground (requiring society to pay more) or overground (incurring less cost and consequently lower user charges). Society preferred that it be underground in the city areas. 6 The RBI presents a rich source of data on state finances. Due to various reasons of comparability, however, data on revenue from non-tax sources in RBI documents are not presented in sufficient details. Hence, this study uses RBI data for only a macro analysis. 7 From these documents, data related to each of the sub-items of revenue and expenditure on non-tax sources for the select services (at much disaggregated level) have been collected. Unfor-tunately, even in the budget documents, some of the items such as lotteries, interests, departmen-tal undertakings, etc, are not reflected correctly. Hence, necessary adjustments have been made, wherever required. 8 This is illustrated by the fact that Rajasthan, Bihar and Orissa are rich in minerals, whereas of Mad-hya Pradesh, Chhattisgarh and Uttaranchal have dense forest areas. Other states do not have an abundance of both these resources.ReferencesBurns, E Michael and Cliff Walsh (1981): “Market Provision of Price-excludable Public Goods: A General Analysis”,Journal of Political Economy, Vol 89, No 11.Dalton, Huge (1949): Principles of Public Finance (London: Routledge and Kegan Paul), p 32. Downing, Paul B (1992): “The Revenue Potential of User Charges in Municipal Finance”, Public Finance Quarterly, Vol 20, No 4, pp 512-27.International Monetary Fund (2006):Government Finance Statistics, June, Washington DC.Musgrave, Richard A and Peggy B Musgrave (1973): Public Finance in Theory and Practice (Boston: McGraw Hill).Tilak, J B G (1993): “Financing Higher Education in India: Principles, Practice and Policy Issues”, Higher Education, Netherlands, 26 (1), July: 43-67.Oakland, W H (1987): “Theory of Public Goods” in Auerbach, A J and M Feldstein (ed.), Handbook of Public Economics, Vol II (Amsterdam, North Holland).Reserve Bank of India:State Finance: A Study of State Budgets, various issues.In face of the severe revenue crunch for meeting revenue expenditure faced by all the states, a few policy prescriptions for select services can be given for mobilising additional resources for planned development of the states. As education plays a vital role in promoting socio-economic development of a country, it is of utmost importance that primary education is fully subsidised and user charges for secondary and higher education are so designed that these are progressive according to the income-group of the user. Also, credit market should be regulated for financing higher education by making procedures simple so that it is accessible to both poor and non-poor. This would also help in increasing tuition fee collection to state exchequer. Any increase in user charges for medical services can result in lower recourse to these services and higher rates of self-medication among the poor. Thus it is essential to differentiate between poor and non-poor people availing the medical facilities. One way could be to locate more facilities closer to the rural areas, as most of the poor people live in rural areas, and charge zero or lower fees from them. On the other hand, insurance status is a good indicator of those people who can afford the medical care and so full cost should be charged from insured persons. Also, as demand for this facility is price inelastic, a little increase in user charge from non-poor would not affect the demand for the service, provided that quality of services is also simultaneously improved.Water rate structure should be rationalised for better recovery of cost. Also, a reasonably accurate metering system must be installed and maintained for direct water users, and a timely bill-ing and collection system has to be in place. Also, there should be a lower water rate charge for non-domestic users as compared to domestic users. Among other things, accurate measurement of water on a volumetric basis is an important requisite for an effective pricing policy. With a view to improving resource use efficiency and sustainability of the environmental quality the following specific reforms could be useful: to increase the accountability of institutions to improve the reliability of service; having private sector participation; to improve financial sustain-ability of the service; to create financial incentives to reduceO&M losses and capital and financing costs; to improve the environ-mental sustainability of service; and to have participation of the community groups. In irrigation projects, the increase in water rates have been rather modest and states have not accepted the Irrigation Com-mission recommendations of reviewing and adjusting rates every five years. There is an urgent need to increase the user charges. Also, at present water rates are almost everywhere fixed crop-wise and with reference to area irrigated. However, many consid-erations like linking water rates to quality of irrigation services, rationalising rate structure and reducing cost of assessment and collection argue strongly for a system which makes water charges explicitly a function of volume and season. Thus, volumetric pric-ing should be adopted for better recovery of cost, though it can only be adopted in phased manner.To improve the maintenance of roads it is recommended that government start a system of electronic toll collection either through microwave technology or through infrared technology. This will be able to solve the problem of congestion due to toll collection considerably. For the Attention of Subscribersand Subscription Agencies Outside IndiaIt has come to our notice that a large number of subscriptions to the EPW from outside the country together with the subscription payments sent to supposed subscription agents in India have not been forwarded to us.We wish to point out to subscribers and subscription agencies outside India that all foreign subscriptions, together with the appropriate remittances, must be forwarded to us and not to unauthorised third parties in India.We take no responsibility whatsoever in respect of subscriptions not registered with us. MANAGER
SPECIAL ARTICLEjanuary 31, 2009 EPW Economic & Political Weekly54Mobilising Non-Tax Revenue: An Empirical Analysis of Trends in StatesMahesh C Purohit, Vishnu Kanta PurohitThis paper analyses the structure of non-tax sources of the states. The major thrust is on presenting a detailed analysis of six select services drawn from social and economic services. While education, sports, arts and culture; medical and public health; and water supply and sanitation have been selected from social services, major and medium irrigation; minor irrigation; and roads and bridges have been chosen from economic services. The analysis of user charges at the disaggregated level for each of the services provided in different states is based on the data drawn from the state budget documents. A comparative analysis of the recovery rate over time is based on the results for two points of time, i e, 1993-94 to 1995-96 and 2001-02 to 2003-04.The paper is an outcome of the study titled “Mobilising Resources through Reform of State Non-Tax Sources for Plan Development” submitted to the Planning Commission, New Delhi.Mahesh C Purohit (maheshpur@gmail.com) is with the Foundation for Public Economics and Policy Research, New Delhi, and Vishnu Kanta Purohit is with Indraprastha College for Women, Delhi.Augmenting resourcesis an exercise in balancing current revenues, economic growth and political economy. The political economy of mobilising resources, therefore, has to consider issues related to both, the economic effects and the fallout in terms of politics.1 IntroductionMobilising resources through reforms in non-tax sources of the states serves the twin purpose of having a rational non-tax structure and generating greater means to achieve economic growth. Irrational structure of non-tax sources has adverse economic effects that invalidate growth objectives. From the economic point of view, therefore, one has to keep in mind the objectives of equity, efficiency and neutrality; especially the impact of these on the economic growth of the economy. To fulfil these objectives, one often tends to adopt an econo-mically rational structure of non-tax sources that may not be palatable politically. 1.1 Objectives of the StudyThis study aims at presenting measures for mobilising resources via reforms in non-tax sources. In doing so, it has the objectives of (a) analysing the structure of non-tax sources of Indian states; (b) examining the structural reforms; and (c) suggesting a rational non-tax structure; which should be economically viable and should yield larger resources. Keeping these objectives in view, the study hypothesises that the non-tax revenues, which account for about 19% of the states own revenue, should play a major role in states’ own resources. This is imperative given the widening revenue-expenditure gaps in the states’ budgets and the demand to reduce the quantum of subsidy to meet the cost of public services through proper pricing. 1.2 Taxonomy of Non-Tax ResourcesNon-tax sources are defined as payment made to the government for which there is a quid pro quo.1 However, these non-tax sourcesdo not have similar features and are classified into three categories:First, there are some sources that are compulsory and requited payments. These sources include penalties (other than penalties on non-compliance of taxes) and fines (Musgrave and Musgrave 1973). The second sources comprise voluntary and unrequited payments. These payments include donations and contributions made to the government or any unclaimed funds lying with the government. The third category consists of voluntary and 
SPECIAL ARTICLEEconomic & Political Weekly EPW january 31, 200955requited payments, including revenue earned from the resources owned by the government such as forest, marine, riparian habi-tats and wildlife. This category also includes revenue earned by sale of usage rights, admission fee and royalties and rental pay-ments received by the government. Income earned in the form of dividends and the interest receipts from investments made by the government also fall into this category. 1.3 Scope of the StudyThough the term “non-tax revenue” encompasses all the above components, to delimit the scope of this study and to have a thor-ough analysis of a few selected services, some of the non-tax sources given below have not been included in this study. First, payments made to the government in the form of subscription to a loan or to a long-term saving scheme, which isrepayable by the government in the future (that is, the payment made to the government is required by future payments orby transfer of other assets) is considered as “capital receipts” and receipts from the government borrowings, money creation or proceeds of disinvestments is not included in the scope of non-tax revenue. Second, revenue receipts of the government through the sale of goods and services (that are commercial in nature) are also not included. Third, in some non-tax receipts it is observed that not all receipts by the government are accounted for in its consoli-dated fund, e g, in case of education, although the students pay a variety of fees, only the tuition fees are credited to the treasury (Tilak 1993). Similarly, in medical and public health, it is observed that in some states the user charges paid by outdoor patients are not put into the consolidated fund. In Rajasthan, this is credited to the welfare of the hospital staff and in Tamil Nadu it is consid-ered to be a part of the maintenance expenditure of the hospital. Such variations hamper a comparative analysis of recovery rate of different services. Fourth, the revenue from lotteries is netted out because the purpose of this activity is not to provide any public service but to generate revenue. In many states, a major chunk of the revenue from lotteries is used to make payment for lottery prizes. Only a very small proportion of this revenue remains with the state government. The same holds true for other commercial acti-vities. Therefore, this study uses the concept of net non-tax revenue, i e,gross non-tax revenue minus expenditure for these commercial activities.Finally, notional receipts are excluded from the purview of this study, e g, interest on capital works in irrigation, which is matched by contra-entry in many states (signifying notional expenditure).The non-tax sources covered in the study include the following:(i) Administrative Non-Tax Receipts: This source accounts for about three-fourths of the states’ own non-tax revenue. In the future, this is likely to be the most productive and reliable sourceof non-tax revenues for the states. There are over 100 departmental sources of non-tax revenues in different states which are classified under three dozen heads. Thus, in attempt-ing a broad study of all the items under non-tax sources, the effort is to cover some aspects from each of the three broad components of administrative receipts, viz, gen-eral services, social services, and economic services.(a) Receipts from general serv-ices: These comprise receipts from the Public Service Commission, police, jails, supplies and disposals, stationery and printing, public works, other administrative serv-ices, contribution and recoveries towards pension and other retire-ment benefits, and other miscel-laneous general services.(b) Receipts from social services: Major items under this category include education, sports, arts and culture, medical and public health, family welfare, water supply and sanitation, housing, urban develop-ment, information and publicity, labour and employment, social security and welfare, and other social services.(c)Receipts from economic services: Major items under this class are crop husbandry, animal husbandry, dairy development, fisheries, forestry and wildlife, cooperation, other agricultural and rural programmes, special area programmes, major and medium irrigation, minor irrigation, village and small-scale industries, industries, non-ferrous mining and metallurgical industries, roads and bridges, tourism, and others. There are six major individual contributors to the states’ non-tax revenue. These are education, sports, art and culture; medical, public health and family welfare; water supply and sanitation; major and medium irrigation; minor irrigation; and roads and bridges. These account for a major share in the administrative component of the non-tax revenue in different states.Table 1: Composition of States’ Own Non-Tax Revenue (as % of total in each state)States General Services Social Services Economic Services 1993-94 2003-04 G* B* 1993-94 2003-04 G* B* 1993-94 2003-04 G* B*Andhra Pradesh 3.91 7.61 13.88 1.15 5.25 6.91 19.86 1.57 38.61 33.67 8.38 0.68Bihar 2.6527.0922.222.243.5528.0611.791.1490.1038.12-16.27-1.68Goa 2.841.2527.53 2.05 9.5110.47 21.27 1.3285.39 87.97 19.76 1.21Gujarat 3.819.1124.06 1.995.005.9616.76 1.3034.04 56.59 13.97 1.16Haryana 64.36 30.51 -12.56-1.10 3.0011.98 18.86 1.4723.88 35.82 9.410.76Karnataka 9.4760.30 24.16 1.737.234.179.67 0.82 36.64 31.20 12.19 0.99Kerala 27.55 38.05 11.74 0.9612.96 15.82 10.38 0.8149.73 39.64 3.97 0.35Madhya Pradesh 5.98 8.49 8.81 0.88 3.67 5.49 6.56 0.70 74.91 84.15 -0.26 0.05Maharashtra 9.6627.52 15.48 1.366.2510.76 10.77 1.0045.01 51.13 5.320.49Orissa 9.995.132.750.336.765.9011.631.0962.0061.336.600.67Punjab 16.3155.5827.342.4311.022.279.320.8654.0410.735.370.50Rajasthan 22.23 22.73 -5.64 -0.59 6.63 12.21 10.99 0.97 19.11 31.86 10.63 0.94Tamil Nadu 10.43 16.29 18.39 1.59 12.46 22.77 15.77 1.35 35.95 34.21 11.99 1.03Uttar Pradesh 49.79 12.46 -11.33 -1.46 3.74 15.66 19.73 1.94 25.52 42.69 8.24 0.81West Bengal 18.97 29.54 12.15 0.94 19.15 15.62 8.18 0.62 38.23 36.59 10.20 0.75Major 15 States 19.24 26.66 5.70 0.50 6.01 9.34 13.84 1.21 42.89 39.89 6.30 0.57G* = Growth rate for the period 1993-94 to 2003-04 and B* = Buoyancy for the period 1993-94 to 2003-04.
SPECIAL ARTICLEjanuary 31, 2009 EPW Economic & Political Weekly561.4 Methodology of the StudyTo analyse the efforts of the states in collecting appropriate user charges for the services provided, the study presents an estimate of the recovery rate (RR) from services as a percentage of revenue expenditure (RE) incurred on services. To facilitate comparison betweenRRi (recovery rate from the ith service) amongst the states, the study presents a normative approach to estimate “norms” of RR for each of the services by using panel data models.The study attempts to analyse differences between the actual and normative RR (estimated through the regression approach) to appraise the performance of the states in collecting RR for the select services. This is done by takingRR of a specific service as a function of related independent variables. The above exercise is based on data collected from the original source, viz, state budget documents for 15 major states.2 Also, it uses sources of data from Reserve Bank of India (RBI), Central Statistical Organisation, Bureau of Economics and Statistics, thevarious state departments dealing with different non-tax services, etc.2 A Theoretical PerspectiveGovernment intervention in providing goods and services for public welfare can be justified on the grounds of market imper-fection in provisions of these goods. The magnitude and the mix of these goods provided by the government, however, depend on the level of economic development and market imperfection.“Pure” public goods are the ones having features of non- excludability and non-rival consumption/joint consumption. In fact very few public goods are pure public goods. First, the joint consumption property might be partly violated. Although all individuals may consume the services in the same quantity, the benefit obtained may be reduced if more persons have a share in its consumption. This phenomenon is known as congestion or partial rivalness. Public goods can be distinguished on the basis of variability in their use, which occurs due to the opportunity cost of these public services (Oakland 1987). Second, goods pro-vided by the government may have joint consumption features but these may be excludable. Such goods and services may be termed as “price-excludable public goods” (Burns and Walsh 1981). Such services include television and radio transmission and services provided by transportation facilities, entertainment facilities (theatres, museum), recreational facilities (national parks, stadium) and information services.Samuelson stated that pure public goods have to be financed through taxes while the quasi-public goods could be subsidised or regulated by the government through a pricing mechanism. Mixed goods may be financed partially by user charges and partly through a combination of subsidy and user charges.In practice, however, considering government as a natural monopoly, various models of utility pricing have been developed. These are marginal cost (MC) pricing theory, average cost (AC) pricing theory and price discrimination theory based on a combi-nation of MC andAC pricing theories, viz, economic pricing, fully distributed cost pricing, demand-compatible pricing, uniform pricing, non-uniform pricing, multi-part tariff, two-part tariff, myopic pricing rules, spot pricing, Boiteux pricing, Ramsey pricing, etc.Clark applied the MC pricing principle to the public utilities keeping in view the objectives of equity and efficiency. However, there are both theoretical and practical problems related with marginal cost pricing. First, the assumptions on which the theory of allocative efficiency is based generally do not hold; and second, implementation may not be feasible.The earlier view of financing of government expenditure through budgetary resource was justified on the ground that the benefit of pure public goods are expected to accrue to the society as a whole. Therefore, the marginal cost pricing principle that is generally applicable for the pricing of private goods and services could not be applied to public goods. The government is expected to recover the cost either through taxes or public borrowings. The marginal cost pricing approach implies that in general (a) the Table 2: Yearly Average RR/RE of Social Services and Select Components(in %)States SocialServicesESACMPHWSS 1993-96 2001-04 1993-96 2001-04 1993-96 2001-04 1993-96 2001-04Andhra Pradesh 1.86 2.94 1.54 2.49 3.59 3.48 1.92 5.32Bihar 1.141.670.150.842.593.651.330.26Goa 8.0112.97 0.582.114.83 6.97 45.65 61.75Gujarat 2.523.291.111.618.465.720.140.5Haryana 3.816.592.31.849.1910.328.5111.13Karnataka 2.442.321.091.0166.220.810.38Kerala 2.402.222.012.335.823.760.11.2MadhyaPradesh 1.611.610.440.482.26 2.48 4.08 5.66Maharashtra 2.892.360.90.578.696.313.040.85Orissa 2.162.041.321.112.592.556.7111.23Punjab 3.732.870.790.838.686.5410.319.58Rajasthan 3.813.250.511.263.882.7223.0819.03Tamil Nadu 2.54 3.97 1.44 2.2 5.86 7.15 1.03 3.77Uttar Pradesh 1.73 3.32 1.4 3.38 2.24 3.05 0.04 0.26West Bengal 1.74 1.20 0.45 0.58 6.53 4.04 0.94 1.37Major 15 States 2.35 2.80 1.04 1.51 5.29 4.76 5 7.27ESAC = Education, Sports, Arts and Culture, MPH = Medical, Public Health and Family Welfare, WSS = Water Supply and Water Sanitation.Source: Reserve Bank of India,State Finance: A Study of State Budgets, various issues.Table 3: Yearly Average RR/RE of Economic Services and Select Components(in %)States EconomicServicesMMIMIR&B 1993-96 2001-04 1993-96 2001-04 1993-96 2001-04 1993-96 2001-04Andhra Pradesh 24.60 17.01 15.04 0.74 14.66 1.26 8.16 7.94Bihar 44.33 7.27 28.68 12.72 0.610.79 6.734.83Goa 80.15 99.41 17.24 108.3 7.63 66.67 24.15 12.07Gujarat 18.5023.316.0716.22.824.970.882.61Haryana 29.2527.927.1727.660.220.360.120.64Karnataka 14.18 15.35 4.8526.55 2.01 4.26 3.57 7.34Kerala 20.8410.147.096.591.351.888.364.76Madhya Pradesh 52.16 27.12 23.46 13.75 18.3 26.55 5.63 1.56Maharashtra 24.5827.518.3712.497.045.4771.350.12Orissa 45.15 36.29 17.74 32.16 2.213.155 9.38Punjab 30.5316.7221.216.120.940.260.810.41Rajasthan 14.98 21.41 6.83 4.28 32.82 34.29 1.811.94TamilNadu 9.5912.070.28 0.415.8612.92 5.6 7.12Uttar Pradesh 12.56 9.63 19.49 10.78 5.94 6.03 6.18 4.03West Bengal 7.75 9.45 3.51 2.45 4.78 4.02 7.24 8.43Major 15 States 23.31 18.08 8.67 7.57 6.53 6.13 5.25 4.57MMI = Major and Medium Irrigation, MI = Minor Irrigation, R&B = Roads and Bridges.Source: Reserve Bank of India,State Finance: A Study of State Budgets, various issues.
1993 96 2001 04 
1993 96 2001 04 
1993 96 2001 04 
1993 96 2001 04 
1993 96 2001 04 
1993 96 2001 04 
SPECIAL ARTICLEEconomic & Political Weekly EPW january 31, 200961It is due to these differences in the socio-economic factors across the states that there exist substantial variations in non-tax recoveries and the desired RR/RE. In the case of the services, where the desired RR/RE are higher than the actual RR/RE, scope for upward revisions exists. The difference in the actualRR/RE and the desired RR/RE indicates the inefficiencies in the recovery enforcement. In such cases, there is need to revise the rates upwardly. Given this framework of conceptual and practical limitations, this study adopts the method given below for estimating norms for desired RR/RE. The model for fixing norms of non-tax revenues is taken as follows:Let Yi (where i=1, 2, ..n) denote the non-tax revenue on a particular item collected by the ith state whose GSDP isXi. LetZi denote a vector of characteristics of the ithstate which are considered important for determining the level of collection of non-tax revenues for a particular service. We fit a model having the following functional form:(1) Yi = α + β Xi + γ Zi + ui where α, β, and γare constants and uiis a random variable that takes only positive (non-negative) values.The above function in equation (1) can be estimated from the data for different states for each component of non-tax revenue and the fitted value, viz:(2) Y^i = α^ + β^ Xi + γ^ Zi can be taken as the norm for that particular component of non-tax revenue.To estimate the norms for six select services, the following equations were set up: Equations for Regression1 recedu = {pcgsdp, urban} For Phase I (1993-94 to 1995-96) {pcgsdp, urban, sch, tch} For Phase II (2001-02 to 2003-04) 2 recmed = {pcgsdp, nodoc} For Phase I {pcgsdp, nodoc, beds, For Phase II hosp, dis} 3 recwss = {pcgsdp, urban} Same for both the phases4 recmajir = {pcgsdp, gsdpagr, Same for both the phases grsirrarea} 5 recmirr = {pcgsdp, gsdpagr} For Phase I {pcgsdp, gsdpagr, For Phase II grsirrarea} 6 recroad = {pcgsdp, gsdptran, rdl} For Phase I {pcgsdp, gsdptran, For Phase II dl, nov} where recedu = RR/RE for education, arts and culture; recmed =RR/RE for medical and public health; recwss =RR/RE for water and sanitation; recmajir =RR/RE for major irrigation; recmirr =RR/RE for minor irrigation; recroad = RR/RE for roads and tolls.In the above equations, for each of the six select services, the percentage share of recovery rate to the corresponding expendi-tures has been calculated for the two phases (1993-94 to 1995-96, termed as phase I) and 2001-02 to 2003-04 (termed as phase II), across the 15 major states. The average RR/RE is then obtained for each of the service for both the phases. The change in this average percentage share in the two phases indicates the efforts of the major states in mobilising resources through non-tax sources (Exhibits 1 to 6).These regression equations give estimated RR/RE for each select service by using the average of actual percentage share of revenue receipts of a service to the corresponding revenue expenditures for the concerned service provided by the govern-ment as the dependent variable. The independent variables are: per capita income (pcgsdp), proportion of relevant sectoral income in totalgsdp (gsdpcons for construction, gsdpagr for agriculture, gsdptran for transport), urbanisation (urban), availability of educational facilities (sch for schools, tch for teachers), availability of medical facilities (nodoc for number of doctors, beds for beds in hospitals, hosp for hospitals, dis for dis-pensaries), gross irrigated area (grsirrarea), road length (rdl), number of vehicles (nov), etc. Also, 15 state-wise dummy variables have been applied to get the state-wise estimated RR/RE for each service.The all-states norm is calculated by using the “normative approach” in the above regression equations. This method, thus, estimates the “norm” for RR/RE of each of the service by taking the deviation from the all-states average. The difference bet-ween the actualRR/RE for each service across the states and the desired RR/RE for each service from the all-states average gives the magnitude by which the states need to increase their existing recoveries. The results of the above regressions, fitted for all the services in the 15 major states, yield the norms for all the services.The study then adopts a method of comparing the adequacy of the actual average RR/RE with the estimatedRR/RE of the best performing state from amongst all the states. For each of the selected service, the norm can be the state with the highest RR/RE among the major states. Thus, in the case of education, the desired RR/RE of Haryana is highest in the first phase of our study and maximum for Uttar Pradesh in the second phase. These are taken as the norm for the education in this study. The same meth-odology is adopted for all the six services.The norms estimated on the basis of the related variables determine the relative efficiency of states when the state govern-ment is viewed as a multi-service provider and the user charges as price-like instruments, which restrict the use of the service to only the “paying” consumers.5 Conclusion and Policy ImperativesTo conclude, non-tax sources are not a fiscally significant sourceof revenue in the states’ budget and their growth is not keeping pace with other components of revenue receipts. Non-tax revenue can be a major source of budgetary receipts for the state governments if proper attention is paid towards pricing of the services. Its importance is now being realised in the context of bridging fiscal deficits of the states and the heavy financial requirements for upgrading and modernising basic infrastructure. 
SPECIAL ARTICLEjanuary 31, 2009 EPW Economic & Political Weekly62Notes1 This is in contrast with the definition of tax as a compulsory payment made to the government for which there is no quid pro quo. See Dalton (1949).2 The major states include Andhra Pradesh, Bihar, Goa, Gujarat, Haryana, Karnataka, Kerala, Madhya Pradesh, Maharashtra, Orissa, Punjab, Rajasthan, Uttar Pradesh, Tamil Nadu and West Bengal. 3 Such services include education, health, and sani-tation, etc, Downing (1992).4In the literature on public finance, it is largely accepted that the marginal cost pricing principle is applicable to the provision of government serv-ices so long as (i) the benefits of the service accrue to the masses, and the marginal expenditure exceeds the marginal cost, and (ii) the marginal benefits of financing the service is equal to the possible marginal benefit from other competing investments. 5 In Berkeley, California (US), for example, a sur-vey was conducted to find out from the society whether the BART (train service) railway line be underground (requiring society to pay more) or overground (incurring less cost and consequently lower user charges). Society preferred that it be underground in the city areas. 6 The RBI presents a rich source of data on state finances. Due to various reasons of comparability, however, data on revenue from non-tax sources in RBI documents are not presented in sufficient details. Hence, this study uses RBI data for only a macro analysis. 7 From these documents, data related to each of the sub-items of revenue and expenditure on non-tax sources for the select services (at much disaggregated level) have been collected. Unfor-tunately, even in the budget documents, some of the items such as lotteries, interests, departmen-tal undertakings, etc, are not reflected correctly. Hence, necessary adjustments have been made, wherever required. 8 This is illustrated by the fact that Rajasthan, Bihar and Orissa are rich in minerals, whereas of Mad-hya Pradesh, Chhattisgarh and Uttaranchal have dense forest areas. Other states do not have an abundance of both these resources.ReferencesBurns, E Michael and Cliff Walsh (1981): “Market Provision of Price-excludable Public Goods: A General Analysis”,Journal of Political Economy, Vol 89, No 11.Dalton, Huge (1949): Principles of Public Finance (London: Routledge and Kegan Paul), p 32. Downing, Paul B (1992): “The Revenue Potential of User Charges in Municipal Finance”, Public Finance Quarterly, Vol 20, No 4, pp 512-27.International Monetary Fund (2006):Government Finance Statistics, June, Washington DC.Musgrave, Richard A and Peggy B Musgrave (1973): Public Finance in Theory and Practice (Boston: McGraw Hill).Tilak, J B G (1993): “Financing Higher Education in India: Principles, Practice and Policy Issues”, Higher Education, Netherlands, 26 (1), July: 43-67.Oakland, W H (1987): “Theory of Public Goods” in Auerbach, A J and M Feldstein (ed.), Handbook of Public Economics, Vol II (Amsterdam, North Holland).Reserve Bank of India:State Finance: A Study of State Budgets, various issues.In face of the severe revenue crunch for meeting revenue expenditure faced by all the states, a few policy prescriptions for select services can be given for mobilising additional resources for planned development of the states. As education plays a vital role in promoting socio-economic development of a country, it is of utmost importance that primary education is fully subsidised and user charges for secondary and higher education are so designed that these are progressive according to the income-group of the user. Also, credit market should be regulated for financing higher education by making procedures simple so that it is accessible to both poor and non-poor. This would also help in increasing tuition fee collection to state exchequer. Any increase in user charges for medical services can result in lower recourse to these services and higher rates of self-medication among the poor. Thus it is essential to differentiate between poor and non-poor people availing the medical facilities. One way could be to locate more facilities closer to the rural areas, as most of the poor people live in rural areas, and charge zero or lower fees from them. On the other hand, insurance status is a good indicator of those people who can afford the medical care and so full cost should be charged from insured persons. Also, as demand for this facility is price inelastic, a little increase in user charge from non-poor would not affect the demand for the service, provided that quality of services is also simultaneously improved.Water rate structure should be rationalised for better recovery of cost. Also, a reasonably accurate metering system must be installed and maintained for direct water users, and a timely bill-ing and collection system has to be in place. Also, there should be a lower water rate charge for non-domestic users as compared to domestic users. Among other things, accurate measurement of water on a volumetric basis is an important requisite for an effective pricing policy. With a view to improving resource use efficiency and sustainability of the environmental quality the following specific reforms could be useful: to increase the accountability of institutions to improve the reliability of service; having private sector participation; to improve financial sustain-ability of the service; to create financial incentives to reduceO&M losses and capital and financing costs; to improve the environ-mental sustainability of service; and to have participation of the community groups. In irrigation projects, the increase in water rates have been rather modest and states have not accepted the Irrigation Com-mission recommendations of reviewing and adjusting rates every five years. There is an urgent need to increase the user charges. Also, at present water rates are almost everywhere fixed crop-wise and with reference to area irrigated. However, many consid-erations like linking water rates to quality of irrigation services, rationalising rate structure and reducing cost of assessment and collection argue strongly for a system which makes water charges explicitly a function of volume and season. Thus, volumetric pric-ing should be adopted for better recovery of cost, though it can only be adopted in phased manner.To improve the maintenance of roads it is recommended that government start a system of electronic toll collection either through microwave technology or through infrared technology. This will be able to solve the problem of congestion due to toll collection considerably. For the Attention of Subscribersand Subscription Agencies Outside IndiaIt has come to our notice that a large number of subscriptions to the EPW from outside the country together with the subscription payments sent to supposed subscription agents in India have not been forwarded to us.We wish to point out to subscribers and subscription agencies outside India that all foreign subscriptions, together with the appropriate remittances, must be forwarded to us and not to unauthorised third parties in India.We take no responsibility whatsoever in respect of subscriptions not registered with us. MANAGER
