The fourth volume of the India Policy Forum features papers on schooling inequality, the duration of microfinance groups, sub-national fiscal flows, and reform of the power sector, land policies, and higher education. Suman Bery, Barry Bosworth, and Arvind Panagariya edited the volume. The editors’ summary appears below, and you can download a PDF version of the volume:
Download individual articles:
- The Political Economy of the Indian Fiscal Federation
- Can Schooling Policies Affect Schooling Inequality? An Empirical Evaluation of School Location Policies in India
- Mortgaging the Future? Indian Higher Education
- Microfinance Lifespans: A Study of Attrition and Exclusion in Self-Help Groups in India
EDITORS’ SUMMARY The India Policy Forum held its fourth conference on July 17 and 18 of 2007 in New Delhi. This issue of the journal contains the papers and the discussions presented at the conference. The first paper examines the fiscal relationship between the Central Government and the states of India. The next two papers focus on the Indian educational system, specifically the social implications of government policies governing access to primary and secondary schools, and the challenges facing the country’s system of higher education. The fourth paper evaluates the performance of an important component of India’s microfinance system. Finally, the fifth paper provides an assessment of recent efforts to reform the distribution segment of the electric power industry. In addition to the working sessions of the conference, T.N. Srinivasan of Yale University, a member of the advisory panel, delivered a public lecture on the topic of: “Economic Reforms, External Opening and Growth: China and India.”
The Political Economy of the Indian Fiscal Federation
Despite massive unfulfilled need and repeated rhetorical commitment to increase public spending, public expenditure in India on education and health has never exceeded more than 3.3 and 1.3 percent of GDP, respectively. Implementing such spending, and to a large degree paying for it, is the responsibility of India’s states. In her paper, Indira Rajaraman argues that an important explanation for this persistently low level of spending lies in the nature of fiscal transfer arrangements in India’s federal structure, particularly the unpredictable and discretionary nature of significant components of these transfers.
The assignment of expenditure responsibilities and revenue rights in India gives rise to a vertical fiscal gap at the sub-national (state) level. The closure of this gap is provided for by the appointment, every five years, of a constitutional body called the Finance Commission. The report of each Commission, once accepted by the government, prospectively defines the formula for statutory flows from the national government (the “Center”) for the succeeding quinquennium. Such statutory flows from the Center to the states are predictable in relation to the underlying tax base, are pre-defined both in aggregate and in their distribution between states, and are unconditional. In Rajaraman’s view, these are all desirable properties to permit states to make multi-year expenditure commitments of the kind needed for provision of primary education and health.
However, such statutory flows represent only part of the story. In the years before 2005, statutory flows never exceeded 60 percent of the total flow. The remaining Center–state transfers took place under a range of nonstatutory mechanisms, largely under the control of an extra-constitutional body called the Planning Commission, and were unpredictable in aggregate from year to year.
While initially entirely discretionary, in 1969–70 the inter-state allocation of a portion of these “Plan” transfers was in turn subjected to a periodically revised formula (commonly referred to as the “Gadgil Formula”). However, this formulaic distribution was accompanied by a shift from a full grant basis to one comprising 70 percent loans and 30 percent grant. This shift to borrowed funds rather than grants implicitly altered incentives away from health and education state-level spending, which were unable to bear the ensuing interest burden. This disincentive, associated with the loan component, led to a gradual reduction in the share of this formulaic component in overall non-statutory flows.
Against this policy and institutional background, the paper performs three empirical exercises to determine the year-to-year changes in the share in grants from the Center received by states in aggregate that was not subject to formula and therefore open to bargaining by the states. The first empirical exercise quantifies the non-formulaic bargaining margin within aggregate flows for each year of the period 1951–2007, and estimates it to have varied inversely with an index of political fractionalization in the federation. As fractionalization increased, the formulaic share rose. The system thus fluctuated in response to changes in the political situation. This instability is inappropriate for funding requirements of basic developmental services.
The second exercise tests whether the control over aggregate state borrowing from the financial markets (constitutionally vested at the national level, and an important force for macroeconomic stability) represents opportunistic behavior influenced by the national electoral cycle. The difference between the consolidated fiscal imbalance, or deficit (aggregated across national and state levels), and the imbalance for the Central Government alone, provides a proxy measure for measuring the extent of sub-national borrowing from financial markets.
The consolidated fiscal imbalance is shown to have risen in years preceding Parliamentary elections. This is in contrast to the fiscal imbalance at the Center, which was not dictated by the electoral cycle. Taken together, the two sets of specifications strongly suggest that aggregate Central limits on state borrowing from financial markets were raised in pre-election years.
This inter-temporal variability, together with the spatial distortions implicit in the opaque system for allocating borrowing entitlements across the states in all years, further adds to the fiscal uncertainty faced by states, and inhibits orderly and sustained planning.
The third empirical exercise deals with a major initiative that commenced in 2005 to reduce the accumulated debt burden of the states. The proposal to reduce this debt originated from the Finance Commission, and addressed debt owed by the states to the Center arising from the loan component of Plan transfers mentioned earlier. The debt relief was to be granted in exchange for promises of fiscal adjustment.
The Finance Commission took the view, later endorsed by Parliament, that the differences in initial conditions across states should be taken into account in setting such conditionality. However the conditionality actually imposed by executive action at the Center envisaged a common terminal year deficit level for all states, implying a difference in the magnitude of adjustment that varies by as much as 10 percent of state GDP, with presumed adverse consequences, once again, for the stable provision of essential state level developmental services.
Starting in 2005–06, there has been a regime change with the replacement of direct Central lending to states for Plan expenditure, with a more inflexible system of caps on state borrowing as part of the conditionality for the above-mentioned debt concessions. Thus, the kinds of uncertainties and patterns in aggregate borrowing limits on states will not be visible for a while longer.
Rajaraman further notes that there has been a fall over the last ten years in the share of state expenditure in overall public spending on health and education because of the huge new Central expenditures on primary education and mid-day meals in schools, which are not routed through states. Thus, the policy response has been to alter the pattern of functional responsibility, rather than restoration to the states of their constitutionally assigned functions, with correction of the adverse incentives that became embedded in the de facto structure of sub-national funding.
Finally, Rajaraman also uses the empirical exercises to draw implications for the nature of dialogue between the Center and the states regarding fiscal matters. She notes the absence of a dispute-resolution forum where the de facto functioning of fiscal arrangements can be subjected to continual examination and monitoring by all partners to the federation. Within such a forum, major issues spanning Central transfers, revenue rights, expenditure externalities, and unfunded mandates, could be resolved in a participatory framework. Its need is likely to become even more urgent as India moves to an integrated nation-wide goods and services tax (GST), where the direct role of the states in revenue collection would be even more restricted, and the need for a broad review of fiscal federal arrangements even more urgent.
Can Schooling Policies Affect Schooling Inequality? An Empirical Evaluation of School Location Policies in India
Over the past several decades, a primary tool used by the Government of India to improve school enrollments, particularly those of the Scheduled Castes (SCs), has been the expansion of access to schools. To this end, the government has long embraced the objective of providing a school within easy walking distance from each rural household. In her paper, Anjini Kochar argues that in implementing this policy, scant attention was paid to the fact that targeting access to schools as a primary objective may constrain the government in addressing other critical aspects of schools, particularly those related to school quality. This is because decisions regarding the location of schools determine more than just access to schools; they combine with the residential structure of a society to define the school community, and hence school characteristics known to affect schooling attainment.
According to Kochar, it is the nature of residential communities in rural India that makes this trade-off between access and quality likely. Rural India resides in habitations—distinct residential settlements within a village— which vary in size but are, on average, fairly small. Because habitations are generally organized along caste lines, the rural economy is thus characterized by a considerable degree of caste-based segregation. The stated policy objective of providing a school within easy walking distance of each household, in conjunction with the geographic distance across habitations, requires the government to adopt a policy that provides schools to relatively small habitations and frequently results in multiple schools within a village.
Therefore, the paper argues that the current school location policy does not permit an optimal allocation of schools based upon enrollment or size. Because school enrollment determines the availability of inputs such as the number of teachers, there is a corresponding variation in the number of teachers per school. To the extent that this attribute of schools affects schooling attainment, Kochar argues that the policy generates schooling inequality across regions, with schools in smaller habitations being of generally lower quality than those in larger habitations.
School location policies also affect the caste composition of the student population. When schools are provided in SC habitations as well as in the other habitations of a village, the residential segregation that characterizes the village gets translated into a corresponding system of de facto schooling segregation. The corresponding difference in the caste composition of students across village schools is also likely to affect schooling attainment.
The paper explores these hypotheses empirically, examining the relationship between school enrollments and availability of schools within habitations, as well as the effect of the number of teachers and the prevalence of schooling segregation. To identify the effect of these school attributes, Kochar uses the policy rules that determine whether a school can be placed in a habitation and the number of teachers assigned to a school. These rules are specified at the district level, and are implemented by the government based on district level data on habitations collected in the All India Education Surveys (AIES). The paper uses this same data that guides policy decisions, and relates it to household data from the Government of India’s National Sample Surveys. The use of policy rules specific to the attributes in question, and the availability of the data that guides current policy decisions, provides a compelling source of identification. To assess the effects of school segregation, Kochar uses the insight that schooling segregation exists only when schools are provided in the SCs/STs (Scheduled Castes/Scheduled Tribes) habitations. Because the AIES data also provide information on the size distribution of SC/ST habitations, it is possible to identify the probability of schools being located in SC/ST habitations (a proxy for schooling segregation) separately from the overall effect of school availability.
The paper has two principal findings. First, based on the size distribution of habitations within a district, the author finds that the current policy rules do affect access, but they also affect teacher numbers and schooling segregation. The regression analysis shows that schools with two or fewer teachers experience reduced enrollments. The results on teacher availability suggest that the decision to provide schools even to relatively small habitations generates a source of schooling inequality: children who reside in small habitations with schools attend schools of poorer quality than those who reside in larger habitations.
Second, the author finds that school location policies also perpetuate caste-based inequalities. Since the SC habitations are generally smaller than others, this means that SC schools are of lower quality, as measured in terms of the availability of teachers. The empirical results show an asymmetric effect of schooling segregation by caste: children of upper castes benefit significantly while segregation has little effect on the SCs. The benefits of living in districts with widespread access to schools therefore vary by caste.
The results of the paper suggest that improvements in school quality cannot be affected without re-considering the government’s school location policies. Kochar admits, however, that improving school quality along the dimensions considered in the paper is no easy task. She suggests an alternative policy that consolidates habitation schools to provide one school in each village, which would enable an optimal number of teachers in each school and thereby improve schooling attainment. While the greater distance to school implied by such a consolidation, particularly for children from the SC/ST habitations, may reduce access, the paper argues that the savings generated by the consolidation could be used to implement a system of cash transfers to children from the SC and the ST conditional on their school attendance records. The positive effects from increased teachers and economies of scale are enough to provide cause for a reconsideration of school location policy in India.
Mortgaging the Future? Indian Higher Education
The higher education system in India also faces troubling distortions and suboptimal outcomes. In their paper, Kapur and Mehta argue that the vast majority of institutions of higher learning are incapable of producing students with skills and knowledge. Attendance does not serve as a screening system for the vast bulk of students, nor does it prepare students to be productive and responsible citizens. The current system is highly centralized, politicized, and militates against the production of general intellectual virtues. It may come as no surprise then, that the last few years have witnessed a rapid rise in skill premiums in India despite the country’s huge population.
Kapur and Mehta maintain that the poor state of the sector and the recent rise in skill premiums can be largely explained by the regulatory bottlenecks facing Indian higher education. Despite impressive reforms elsewhere, Indian higher education remains one the last bastions of the “license control raj”—with troubling implications for India’s future. The paper argues that the result is a state of crisis in Indian higher education notwithstanding the success of a few professional schools. The fact that the system produces a noticeable number of high-quality students is largely the result of Darwinian selection mechanisms and very little because of pedagogic achievements.
According to the authors, the most acute weakness plaguing India’s higher education system is a crisis of governance, both of system and of individual institutions. Because the prevailing political ideological climate views elite institutions as anti-democratic, there is a natural response in political circles to influence admissions policies, internal organization, and the structure of courses and funding. The paper provides data to show that there has been a massive increase in both private higher education and the flight of elites to foreign educational institutions. However, the private sector also suffers from regulatory obstacles and governance weaknesses, raising doubts as to its ability to address the huge latent demand for quality higher education in the country.
From the perspective of the three key suppliers of Indian higher education—markets, the state, and civil society (philanthropy)—the authors elaborate on six significant distortions. First, the process of regulatory approvals diminishes the capacity of private investment to respond to market needs. Second, the regulatory process produces an adverse selection in the kind of entrepreneurs that invest since the success of a project depends less upon the pedagogic design of the project and more on the ability to manipulate the regulatory system. Third, there are significant market failures in acquiring physical assets that are necessary for educational institutions, especially land. Fourth, regulatory approvals are extremely rigid with regard to infrastructure requirements (irrespective of costs or location) and academic conformity to centrally mandated course outlines, degree structures, and admissions policies. Fifth, a key element of a well-functioning market — competition—is distorted by restricting foreign universities from setting up campuses in India, which limits benchmarking to global standards. Sixth, another central element of a well-functioning market, informational transparency, is woefully inadequate.
The university system in India is the collateral damage of Indian politics. As the paper demonstrates, the dismal educational outcomes are not the result of limited resources. For politicians, the benefits of the license-control raj extend beyond old-fashioned rent seeking by manipulating contracts, appointments, admissions, and grades in government-run colleges and universities to the use of higher education for vote-banks, partisan politics, and as a source of new entrepreneurial activities.
The authors identify three key variables that help to clarify the political economy of India’s higher education: the structure of inequality in India, the principal cleavages in Indian politics, and the nature of the Indian state. India is an outlier in the extreme degree of educational inequality, which has led to a populist redistributive backlash. However, the specific redistributive mechanisms are conditioned by the principal cleavages in Indian politics and the nature of the Indian state. The growth of identity politics has sharply enhanced political mobilization around two key cleavages in Indian society: caste and religion. Consequently, redistributive measures follow these two cleavages rather than other possibilities such as income, region (urban–rural), or gender. Thus, the focus on redistribution helps explain why Indian politicians have obsessed over reservations (that is, quota-based affirmative action) in elite institutions of higher education rather than improvements in the quality of primary and secondary schooling, and the thousands of colleges of abysmal quality.
The consequences of the preceding political economy are onerous. One, a diminished signaling effect of higher education; two, an ideological entrapment between what the authors call half-baked socialism and halfbaked capitalism, with the benefits of neither; and three, a pathology of statism wherein higher education policy is being driven foremost by the state’s own interest (or perhaps its own ideological whims). Much of what goes in the name of education policy is a product of the one overriding commitment of the education bureaucracy—namely state control in as many ways as possible.
The paper also highlights the role of the Indian judiciary in higher education reforms, arguing that it has done as much to confuse as to clarify the existing regulatory framework. Although there has been a distinct shift in the Supreme Court’s stance in the past decade, its primary response does not always center on what will enable the education system to adequately respond to demands. Rather, it has uneasily and often confusingly attempted to reconcile disparate principles, be it the dichotomy between education being a charitable or commercial enterprise, or the inherent tension between institutional autonomy and equitable access in higher education.
Kapur and Mehta conclude with a few options for change moving forward. Market failure in higher education means that substantial public investment will continue to be critical in this sector. However, since there are few clear analytical criteria to address the central question of what is “good” higher education, the paper argues that a regulatory system that emphasizes diversity, flexibility, and experimentation is in the long run most likely to succeed. Such a system will also need a different conception of accountability than the one currently prevailing in the Indian system, where resource allocation decisions are centralized to an extreme degree in the Planning Commission, the Ministry of Human Resource Development, and the University Grants Commission. Its quality depends entirely upon the informational resources of a very small group of decision makers and presumes an omniscience that few decision makers can have. Instead India needs to move to a regulatory system with increased horizontal accountability that empowers students to make better informed decisions. Finally, Indian policy makers need to recognize that the competition for talent is now global and that only a combination of a flexible and supple state system that enlists the energies of the market as well as a committed non-profit sector will be able to meet the challenges and the vast scale of demand for higher education in India.
The expansion of rural credit through the “formal” financial system has been a major goal of Indian policy since independence. While a number of initiatives (including nationalization of the country’s major commercial banks) have been taken over the years, success of these initiatives has been only partial.
In 1992, the Reserve Bank of India (RBI), India’s central bank and banking regulator, issued guidelines to the public sector commercial banks (which still dominate Indian banking) encouraging them to lend to small preformed groups called “self-help groups” (SHGs). These groups are almost always composed of rural women, and are often assisted by non-governmental organizations (NGOs) in their formation and their subsequent growth and development.
While the scheme, sometimes called the “commercial bank–SHG linkage scheme”, was in part inspired by the success of Bangladesh’s Grameen Bank in sustainably widening access to financial services in that country, the Indian SHG scheme differs in several respects from the Bangladesh model, and therefore needs to be assessed in its own right. One such difference is the provision of subsidized refinancing to the commercial bank by the National Bank for Agriculture and Rural Development (NABARD) (a publicly-owned affiliate of the RBI). The RBI reports that over 2.5 million of such groups have borrowed from commercial banks since 1992, and loan disbursements by commercial banks to SHGs were 29 percent of all direct bank credit to small farmers in 2004–05.
Microfinance Lifespans: A Study of Attrition and Exclusion in Self-Help Groups in India
However, in spite of the growing importance of SHGs as a source of credit to the poor, there is little systematic evidence on their internal functioning. The paper by Baland and Somanathan attempts to fill this informational gap by using survey data on SHGs created during the period 1998–2006. It does so by describing the survival of groups and members within groups, documenting group activities, and estimating the determinants of group and member duration using an econometric survival model.
The data comes from a survey of 1,102 rural SHGs and the 16,800 women who were members of these groups at some point during the period 1998– 2006. It considers all groups formed by PRADAN (an NGO that has actively promoted SHGs since the start of the NABARD program) in the districts of Keonjhar and Mayurbhanj in northern Orissa, and the Raigarh district in the newly formed state of Chhattisgarh in central India. Although the group members are engaged in a variety of collective activities, saving and credit do seem the most important. Almost all groups surveyed had made small loans to their members and 68 percent of them had received at least one loan from a commercial bank.
For those members who do borrow from the group the average size of the loan, provided from internal group funds, is Rs. 2,200 per year. For groups with at least one bank linkage, 83 percent of members in the group received some part of this loan, and the average amount received by these members is Rs. 2,189 per year. Although loan sizes provided by some specialized microfinance institutions are often larger, these SHG loans are sizable as a fraction of local earnings and, for women who received both group loans and bank loans, it corresponds to roughly two months of labor earnings at the minimum wage in these areas.
The group members in many SHGs appear to be collectively involved in activities not directly related to credit. About 10 percent of the surveyed groups are involved in the preparation of school meals, 3 percent administer state programs that distribute subsidized foodgrains, and about half of them get involved in family or village conflicts or help members during periods of personal distress. These groups therefore seem to play a role in promoting solidarity networks in the community.
The paper then estimates models of both group and member duration. It finds that factors behind group survival are quite different from those affecting member longevity. With respect to group survival, the highest attained level of education in the group is important for its survival, perhaps because some educated members are needed to facilitate transactions and ensure that group accounts are accurate. The presence of other SHGs in the area also has a positive effect on group duration. It may be that a dense cluster of groups allows for the sharing of costs, provides each group with ideas for successful activities, or simply instills in members the desire to survive, compete, and be part of a larger network.
Drawing upon on a large literature pointing to the importance of social heterogeneity in collective action, the paper then explores whether such heterogeneity matters for the average duration of groups and the members within groups. For each member surveyed, the paper records both their individual caste group (or jati) and the “official” caste category to which they belong—ST, SC, Other Backward Castes (OBC), and a residual category often termed General Castes that we refer to as Forward Castes (FC).
The particular question explored is whether heterogeneity matters for group functioning when members belong to different jatis in the same official caste category. The paper finds that commonly used measures of fractionalization and social heterogeneity based on these classifications do not have systematic effects on group survival, but that they do help explain the departure of individuals from groups. Even within broad caste categories, heterogeneity matters. This suggests that the “official” classifications fail fully to capture the relevant social hierarchy.
The members from traditionally disadvantaged groups, especially from the ST, are more vulnerable to group heterogeneity. In addition to group heterogeneity, lower levels of education, lower landholdings, and fewer relatives within the SHG are also associated with higher rates of member exit.
The paper also finds that the bulk of the difference in the duration of membership in a SHG observed between Chhattisgarh and Orissa can be attributed to characteristics of groups in these areas; the authors find that state-level variations in performance are negligible once these characteristics are incorporated in their model.
The results suggest that it is problematic to evaluate the success of microfinance interventions based on conventionally reported coverage figures because they do not account for attrition. The authors’ concern is not with overall attrition rates but with the selectivity they exhibit. It is predominantl