About the Project
The research project on Financing Human Development was entrusted to NIPFP as a part of the larger
UNDP - Planning Commission sponsored programme on " Strengthening State Plans for
Human Development" 2005-2009. Under this programme, apart from carrying out state - specific studies, NIPFP
produced a few Working Papers and Policy Brief and organised two conferences in 2006 and 2007, and a public lecture.
Concept Paper
It has been already demonstrated that India as a country has a long way to go before its human development can match even those of the more advanced Asian countries, leave alone those of Western nations more advanced in this aspect (like the Scandinavian countries). To provide a push for development in this area, the role of the government is crucial, since the areas (and the related services) involved – primary education, primary health, social welfare etc. – are not examples of classic public goods, but do have substantial elements of social externalities which calculations of private providers are not likely to take into account. Besides, between the alternative paradigms of human development, the adoption of ‘human rights pardigm’ – in which basic human development services is considered a citizen’s right – implies that the State has the responsibility of ensuring availability of the services concerned. Ensuring availability in practice in a country like India with low per capita income and a large number of poor would eventually translate into a large role of the government in the supply of the services in terms of production and distribution.
Given that social services are primarily State government responsibilities (more than 80 per cent of combined government expenditure in these areas are incurred by the States), the necessary public interventions would primarily be at the State level. Hence, the focus of this study will be on the State governments in general, and on eight selected States for closer examination. The role of the Central government in incentivizing the State governments to act, possibly with financial support, and provide directions, will be taken into account, but the main premise of the study is that the onus of human development rests squarely on the State governments.
India is a large country, and it is natural that there is substantial diversity both in the status of human development and in the financial health of the States that comprise it. While it is undeniable that better financial health should allow a State to spend more on human development, prima facie evidence does not indicate a very strong relationship, and the relationship between a State government’s financial health and the effectiveness of its interventions (as indicated by the status of human development) is even weaker. Obviously, both raising government expenditures and improving their effectiveness are important. While the broader connection between government expenditures and social sector outcomes also will be examined carefully in this study, the tentative findings immediately point towards the basic framework of the present analysis that should consist of the following parts:
- Finding the ground to be covered in various areas of human development and estimating (necessarily roughly) the costs involved in reaching given goals (MDG/ National Policy).
- Estimating additional resources that the States can generate, and further estimating that part which can reasonably be expected to be allocated to achievement of human development goals under the most favorable circumstances.
- Estimating resources that can be freed by reprioritizing expenditures in favor of human development.
- Assessing effectiveness of present interventions, and possible savings therefrom.
- Estimating Central assistance in various areas of human development.
Wherever possible, available findings will be used and even plugged into the empirical work, so that (a) we avoid reinventing the wheel and (b) economize on our efforts in this work which already has a very large canvas. Further details on the six analytical steps follow.
1. Costing the Transition to Stated Goals
The first issue that has to be considered here is what it is that is to which the costing exercise is to be applied. There are two alternatives here: one, costing the additional supply requirement of the service concerned, and two, costing the transition from the present state of human development to the desired state in terms of outcomes. This distinction is important, as can be briefly illustrated using an example. In the case of education, if the objective is universal literacy of a given age group of children, then one can simply consider the additional costs of making available free elementary education to all the children in the specified age group; this would correspond to the first alternative above. However, ensuring adequate supply alone does not guarantee achievement of the desired outcome, since demand side factors like income, location, parental education can also can influence the outcome, given the supply. Thus, in the second approach, determinants of actual outcomes other than the supply of the service also need to be considered and, if necessary, costed. In this study, we propose to conceptually separate out these two parts, with the costing exercise relating to the supply side only, and the other issues considered in the subsequent parts.
Even so, costing the needed improvements in social services is an elaborate exercise that is likely to be a major undertaking in itself. Often, overall estimates therefore simply accept available estimates of sub-sectors and aggregate (UNDP, 2004). Even so, (a) there may be choices among available estimates and (b) conversely, there may not be any available study for particular sub-sectors. To the extent possible, the present study would use official estimates (for example, for universal primary education, published estimates of a Committee set up by the Ministry of HRD). In the cases where the goal is in terms of a certain level of expenditure, as in the case of National Health Policy, the problem is obviated since the difference between the current level and the desired level would provide the estimate. In the cases where a costing exercise has to be undertaken by us, it would have to be a rough and ready method of estimated per unit cost multiplied by the gap in coverage in terms of the units concerned, on the lines of Delamonica, Mehrotra and Vandermoortele (2001). The main issue here is likely to be the per-unit costs of incremental services. Conceptually, there are two possibilities: one, of prevailing per-unit costs derived from actual level of services provided and the expenditure incurred thereon. The other possibility can be the ‘efficient’ cost, i.e., the per-unit costs that may be prescribed as normative ones on some basis. The latter obviously links up with issues 3 and 4 above. Further, there is the question of the extent of public intervention required, and this is likely to be different for different States, depending on various socio-economic characteristics including historical factors.
A general point here would relate to disaggregation of broad types of services into more specific services and carrying out the costing exercise in terms of the disaggregated services. For example, public health includes various types of programs, and the distance from goals would differ among States between from one program to another. Also, estimated per unit costs are likely to be more reasonable for disaggregated services than for a composite service.
2. Estimating Additional Resources from States’ Own Sources
To begin with, we distinguish between revenue (approximately equal to current) receipts and capital receipts (generally loans or various kinds of liabilities). Our estimates will only concern the former, since the latter for any sub-national unit cannot be limited a priori. The revenue receipts of States can be divided into own revenue receipts (tax and non-tax), shared taxes and grants. In general, there is no acceptable/feasible method of estimating ‘absolute’ revenue capacity of a State. This is why revenue capacity usually implies ‘relative’ revenue capacity, i.e., in relation to what other States achieve in terms of revenue-raising. This can be done in an aggregative manner (for example: doing it for just two broad categories – tax and non-tax revenues) or disaggregated. Given the data requirements, cross-State analyses will probably be better done with the more aggregated estimates, while State-specific studies can be done in a more disaggregated manner. Shared taxes can be estimated on the basis of projections of central tax revenues, and State shares mandated by the Finance Commission. Grants are discussed under section 5.
Once revenue capacities are estimated, the excess of the ‘capacity’ over actual realizations provide an estimate of possible additional resources mobilization from States’ own sources. While the borrowings also are possible sources of funds to be deployed in the human development sectors, in the current state of indebtedness of most of the States, it is best not considered. User charges, not resorted to so far to any significant extent, can provide additional resources too (subsumed under non-tax revenues), but there are certain limitations. First, these are advocated more in the context of economic services than social services, as some of the social services like social welfare simply do not admit of user charges, and second, for various reasons it may not be possible to resort to user charges are in any significant manner because of distributional considerations. Moreover, the rights approach to human development also militates against significant user charges (Prabhu, 2003).
To be sure, additional own source funds can also be generated through expenditure compression, but barring a full-fledged analyses of the expenditure side of the budget, it is likely to be difficult to work out reasonable estimates, except when there are some glaring issues. It may be more appropriate for this study to accept prevailing expenditure trends as given in general.
Even when additional resource mobilization is estimated, another consideration would be to estimate the part of it that can be expected to be available for human development areas. An a priori position on this can be to assume that the whole of the additional resources will be available, but an estimate based on actual trends may be more realistic.
This analysis will be anchored on a detailed assessment of each of the selected States’ finances to assess the feasibility of the conclusions regarding additional resource mobilization and its use for human development purposes. Since several States have or are in the process of legislating Fiscal Responsibility Acts, the associated rules should provide necessary benchmarks of fiscal and revenue deficits that can be used in our study.
3. Reprioritizing Expenditures
Reprioritization of expenditures can refer to both across functional categories and within functional categories. Examples of the former would be to privatize loss-making State Electricity Boards and State transport services, and channelize the savings of budgetary support to these organizations towards human development. Examples of the latter would include policy measures like converting bigger public hospitals into autonomous bodies, allowing them to determine user charges applicable, freezing or gradually reducing budgetary support, and using the savings to expand preventive health measures.
Several initiatives of this type have been taken by various States. An examination of such initiatives, their applicability in general and an assessment of the short and longer term considerations can lead to useful policy prescriptions. Besides, detailed examination of the expenditure pattern of the States can lead to some reprioritization possibilities for individual States for their consideration.
The issue of an appropriate role of the government in the supply of various services ties up here, since identification of areas for partial or complete withdrawal of the government – or change in its role as producer and provider of services – can automatically lead to reduced expenditure in some areas, freeing resources for human development. Policy recommendations of this type would generally be based upon some value judgments; we shall try to make these explicit.
4. Assessing Current Interventions and Suggesting Efficiency Enhancements
The present study proposes to examine this issue in both aggregated and disaggregated manner. Efficiency enhancement in the present context would essentially be defined as lower marginal unit costs. It would thus entail an analysis of expenditures, inputs, outputs and the outcomes to form an idea of efficiency. The two aspects of efficiency: allocative and technical will both have to be examined. Alternative techniques (like stochastic production frontier method and data envelopment analysis) for empirical analysis are available (see for example, Hensher, 2001 and Afonso and St. Aubyn, 2004); a suitable technique will be applied on a case by case basis, depending on the availability of necessary data.
However, it has to be recognized that social sector outcomes are results of several variables, of which financial inputs are only one set of factors. Any empirical analysis will therefore need to be adequately specified to take into account at least all the major determinants of the outcomes, and the contribution of the financial inputs will have to disentangled from the myriad of influences. But this exercise can point towards important policy insights with respect to desired outcomes. For example, if there are strong complementarities between two types of services, non-recognition of this factor can lead to major inefficiencies, and the proposed empirical analyses can unearth such policy lacunae and suggest efficiency enhancements with respect to human development outcomes. Care has to be taken to assess efficiency: a good example is that of teacher’s salary. In some quarters the notion that a large number of educated unemployed implies that teachers’ salaries are too high has gained acceptance, and policies have been introduced to lower the salaries (along with skill requirements) and increase the number of teachers to lower pupil-teacher ratio in the interest of efficiency. Quite the opposite policy is reported to have been followed in the more advanced Asian economies, again in the interest of efficiency (Mundle, 1998).
In this exercise, from the policy point of view, it will be important to examine the translation of inputs into output measures and then into outcome measures to identify the problem areas. For example, expenditure on immunization program can be matched with output indicators like number of persons immunized and outcome indicators like the incidence of the specific disease ( in this example, it should be noted that the translation of the output into outcomes can be with a lag, and these factors also need to be taken into account).
Another way of looking into this issue would be to examine the extent of financial allocations that eventually get spent on the service provision proper. This would require exercises in ‘expenditure tracking’ that can also be attempted. This can be done at two levels: one based on available details of budgetary expenditures (an example of this type of exercise is Sen and Diwan Chand, 2004), and the other based on an actual on-field tracking down of budget allocation for a specific scheme down the administrative ladder to the final spending unit. Obviously, the latter is a highly resource-intensive method, and can be used only on a limited scale. Other possibilities in the area of efficiency would include sub-state level targeting of expenditures based on human development indicators and flexible policy packaging (in terms of emphasis) according to specific requirements of different geographical areas. The equity of government expenditure is social sectors can also be important for efficiency, since these are areas where equity and efficiency are mutually consistent. For specific services, benefit incidence studies relating to India have already been conducted (for example, the study on benefit incidence of public expenditure on health at NCAER by Mahal et al, 2000), and the broad methodology, to a large extent dictated by the available data on consumption of the relevant services, is fairly well-settled.
5. Estimating Central Assistance for Human Development
This will essentially be based on available data on current levels and information on expected amounts of Central Assistance under various channels. Earmarked resources like the education cess will be incorporated into the amounts. Relevant Ministries of Government of India and Planning Commission should be able to provide some information in this respect. The Finance Commission’s specific purpose grants for human development areas will also be taken into account. However, estimation of general purpose transfers – both Plan and non-Plan – will have to be done on the basis of past trends and those mandated by the Finance Commission. How much of these will be available for human development purposes will be estimated in the same manner as in the case of additional resources raised by the States themselves
The mechanism of various types of Central transfers and their design will be critically reviewed to examine their suitability for the stated objectives in general and specifically for enhancing human development outcomes. This will include assessment of major Centrally Sponsored and Central Plan schemes in the area of human development.
It may be relevant here to note that some States among the selected ones could be receiving or likely to receive external funding also for human development purposes. For the sake of completeness, this information will be obtained from individual States for the State-specific studies.
It is likely that even after taking into account the various sources of additional resources, estimates of government expenditure on human development would fall short of the requirements. Other methods of financing the remaining gaps will then have to be resorted to. While out-of-pocket household expenditures tend to close the financing gap to the extent incomes permit, there are other sources: private expenditures like employers’ contributions, insurance, charitable institutions and other NGOs, corporate expenditures and community financing are some of them. While these will not constitute the focus of the study, some aspects of these non-government financing of human development services will be studied, particularly in the context of public-private partnerships. Another issue that is relevant, but not central to the main theme of the present study is decentralization and its impact on the provision of these services. This aspect is more in the area of effective governance than financing (although in certain contexts like community financing the distinction can get blurred somewhat), but is generally taken to be important in the Indian context. Hence, this is another topic that this study will examine, mainly with a view to distil good, replicable practices in service delivery for policy purposes.
References:
Afonso, António and Miguel St. Aubyn (2004), ‘Non-Parametric Approaches to Education and Health Expenditure Efficiency in OECD Countries’, Mimeo.
Delamonica, Enrique, Santosh Mehrotra and Jan Vandemoortele (2001), "Is EFA Affordable? Estimating the Global Minimum Cost of 'Education for All'". Innocenti Working Paper No. 87. Florence: UNICEF Innocenti Research Centre.
Gupta, Sanjeev, Marijn Verhoeven, and Erwin Tiongson (1999), ‘Does Higher Government Spending Buy Better results in Education and Health Care?’. IMF Working Paper No. 21, Washington DC: International Monetary Fund.
Hensher, Martin (2001), ‘Financing Health Systems through Efficiency Gains’, Working Paper No. WG3:2, Pretoria: Commission on Macroeconomics and Health, WHO.
Mahal, A., J. Singh, F. Afridi, V. Lamba, A. Gumber, and V. Selvaraju (2000),
‘Who benefits from public health spending in India?', Washington DC: Health,
Nutrition and Population, Human Development Network, World Bank.
Mundle, Sudipto (1998), ‘Financing Human Development: Some Lessons from Advanced Asian Countries’. World Development, Vol. 26, No.4, pp 659-672.
Prabhu, K. Seeta (2003), ‘Financing Human Development during Fiscal Stringency: Strategies and Options for Indian States’, Mimeo. Paper presented at conference on Critical reflections on State Human Development Reports, Goa.
Sen, Tapas K. and Diwan Chand (2004), ‘Public Expenditure for the Poor in Andhra Pradesh’. Mimeo. Delhi: National Institute of Public Finance and Policy.
UNDP (2004), The Economics of Democracy: Financing Human Development in Indonesia. Indonesia Human Development Report 2004. Jakarta: UNDP.