A unique feature in India’s federal architecture is the pivotal role played by the Union government in financing and monitoring social welfare programmes, and in ensuring that all states have adequate resources and are held accountable for meeting social policy goals. During 2000-2018, the Government of India (GoI) spent over Rs 14 lakh crores on social services.[i] A significant proportion of this expenditure is met through Centrally Sponsored Schemes (CSSs) – a specific purpose transfer from the Union to states, usually in the form of schemes.

While the practice of using specific purpose transfers dates to the pre-Independence era, over time, CSSs have emerged as the primary vehicle through which the GoI finances and directs state expenditure towards national priorities. Their dominance can be seen in their sheer numbers and the quantum of money flowing through them. During the 11th Five Year Plan (2007-2011), there were 147 scheme specific transfers accounting for over 40% of total central transfers to states.[ii] This increased signifcantly in the 12th Plan period. Of the total Rs 8.61 lakh crore transferred by the Union government to states between 2012 and 2015, Rs 5.88 lakh crore (68%) was released as assistance under CSSs.[iii]

The importance of CSSs as a fiscal instrument lies in the fact that they are the primary source of non-wage, uncommitted funds available to states. With a majority of states’ own resources tied to wages, pensions and other committed liabilities (sometimes over 80-90%[iv]), CSSs were designed as a top-up to augment state expenditure, allowing them to address infrastructure and human development deficits.

In principle, the rationale for CSSs is sound and in keeping with first principles: fiscal equalization to ensure that minimum standards of public services are provided to all citizens. Over time, however, the design and proliferation of CSSs have undermined this very rationale. Richer states with better administrative capacities have been able to capture a larger share of CSS funds, resulting in a significant misallocation of resources. Analysis by the Economic Survey 2016-17 of the six top CSSs – Pradhan Mantri Awaas Yojana (PMAY), Sarva Shiksha Abhiyan (SSA), Mid-Day Meal (MDM), Pradhan Mantri Gram Sadak Yojana (PMGSY), Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) and Swachh Bharat Mission (SBM) – found that under no scheme did the poorest district receive even 40% of the total resources. In fact, for the MDM and SBM, the share was under 25%.[v] Other studies of the SSA and National Health Mission (NHM) found similar results – that is, states with poorer health indicators did not necessarily get larger per capita transfers.[vi]

Moreover, the centralized nature of CSSs often makes them an inefficient tool to address state-specific needs and has undermined the autonomy of states to undertake expenditure decisions based on their local needs.

Recognizing these limitations, numerous attempts have been made to restructure schemes and restore them to their rightful place – the states. The last major impetus came with the adoption of the recommendations of the 14th Finance Commission, which increased state governments’ share in tax devolution by 10 percentage points. The resultant decrease in the fiscal space available with the GoI reiterated the need to significantly decrease and restructure CSSs. This led to the constitution of a committee of chief ministers under the aegis of the NITI Aayog. The committee made several recommendations including scheme rationalization, determination of a transparent criterion for interstate allocation, and greater flexibility in and creation of an institutional mechanism for Centre-state negotiation.

The changes that followed were minimal. While schemes were reordered under ‘umbrella’ programmes, within each umbrella programme, sub-schemes continued as before. As per the Union Budget 2016-17, even after the creation of 28 umbrella programmes, there were 950 Central Sector and CSS sub-schemes accounting for about 5% of the GDP and 9% of the total GoI expenditure.[vii] Three years later, in the interim budget for 2019-20, Central Sector Schemes constituted 12% of GoI expenditure, up from 9% in 2016-17; CSSs accounted for another 9%.[viii]

More importantly, there was no real change in the design or implementation of the schemes. Meetings for the planning of of education and health programmes continued as before, and the notification to allocate 25% as flexible, discretionary funds was not reflected in many of the scheme planning documents. Instead, the disbandment of the Planning Commission has resulted in an institutional vacuum with respect to planning. CSSs now fall under the domain of central ministries, leading to further centralization of social policy financing.

A call for rationalizing CSSs, however, has again gained momentum. The GoI recently committed to evaluating all CSSs before fresh appraisals are made and making scheme approval coterminus with the Finance Commission cycle. Accordingly, the Development Monitoring and Evaluation Office (DMEO) under the NITI Aayog has invited proposals to evaluate 28 umbrella CSSs under 10 sectors. Real change in social sector financing, however, will only be visible if the current design of CSSs is completely overhauled, in addition to scheme rationalization being carried out.

Before I offer some indicative steps on how this can be achieved, it is useful to highlight some of the main inefficiencies in the current design of CSSs. Broadly, these can be classified into four interrelated areas: planning failures, implementation failures, fiscal inefficiencies and administrative red tape. Each of these is described below.

Planning Design Failures

Budgets for CSSs are determined based on incremental plans prepared by the respective state governments and approved by a committee at the central level. This has given individual ministries significant discretion in determining scheme design and approving state-specific plans and budgets. There is often an inherent tension between central government priorities and states’ perceived needs. However, since the Centre controls the purse strings, central priorities dominate. To illustrate: in 2010, the Bihar chief minister had launched several state entitlement schemes for education, such as cycles, uniforms, etc. In its SSA budget, thus, the state proposed a low allocation for entitlement. However, the GoI’s own prioritization of entitlements meant that despite no demand, Bihar’s entitlement budget under SSA was enhanced by over 200%. In another example, in 2013–14, one state government wanted to use the SSA budget to provide vehicles for block-level officials to improve school-level monitoring. The approval board at the Centre, however, denied this request as purchase of vehicles was not permitted under SSA rules.[ix]

Implementation Failures

CSSs are typically designed by the central government but implementation rests with the state and local governments. Most CSSs come with rigid guidelines for execution which privilege a top-down, ‘one size fits all’ model with fixed norms and unit costs. For the NHM, for instance, the guidelines lay down fixed population norms to set up health facilities. These, however, underestimate requirements in states such as Rajasthan and Madhya Pradesh which have a population density lower than the national average.

More importantly, even granular implementation details such as hiring processes, training modules and schedules, communication strategies, etc. are laid down by the Centre. Consequently, states and local governments have very little flexibility in adapting implementation based on their specific jurisdiction. The problem is even more acute at the point of service delivery. In education, for instance, if a school wants to spend more money on buying teaching material rather than painting walls, the norms simply don’t allow it. Similarly, a survey conducted by Accountability Initiative in 2013 found that the pressure to meet RTE infrastructure requirements resulted in money for boundary walls being sent to all schools in Himachal Pradesh even though construction couldn’t be undertaken due to land unavailability.[x]

Fiscal Inefficiencies

Most CSSs are designed as a cost-sharing programme between the Union and the states. With the division of CSSs into ‘core’, ‘core of the core’ and ‘optional’, states are expected to contribute 50-60% of the total approved budgets from their own plan funds.[xi]

Within a scheme, however, the matching ratio is uniform across states irrespective of their fiscal capability. Release of funds by the GoI is contingent on states releasing their own share and meeting other conditionalities such as the submission of Utilisation Certificates (UCs). This has three important consequences with respect to distribution of resources. First, the uniform fund-sharing ratio often makes it difficult for the low-income states to put in their requisite share. As subsequent fund release is contingent on states submitting their share, this has an effect on the total quantum of money received by fiscally weaker states. Thus, while Karnataka may perform better than Bihar on most development indicators, it may also be able to avail of the CSS grant by making its matching contribution, while Bihar may find it difficult to put in its share. Second, the presence of conditionalities for fund release means that there is a considerable difference between the approved allocation and actual grants. In 2016-17, for instance, only 85% of total NHM approved budgets were released. These differences are amplified at the state level. Thus, while Bihar (one of the poorest states) received 79%, Gujarat and Haryana (fiscally stronger states) received over 100%.[xii] This creates uncertainty in implementing schemes and invariably states with greater shortfall in services levels suffer the most. Finally, the fixed fund-sharing ratios also creates perverse incentives for states which may not need the additional CSS fund to try and get it.

Layered Bureaucracy and Administrative Red Tape

Finally, detailed and rigid guidelines, complex paperwork and numerous conditionalities for fund release under CSSs have also created considerable administrative red tape, resulting in inefficiencies in approvals and fund flows. The situation is exacerbated by the fact that for some CSSs, the central government has set up parallel institutional structures responsible for CSS implementation in states, thereby creating a new stakeholder in the implementation process. Under the SSA and NHM, for instance, scheme planning and implementation rests with autonomous bodies known as State Implementation Societies.[xiii] The multiplicity of roles means that even simple tasks require approval and technical sanctions from different authorities. A study of the NHM in Uttar Pradesh conducted by Accountability Initiative found that it took a minimum of 22 desks through which the file had to pass for the release of funds from Treasury to the State Health Society (SHS). Other studies have found that the figures for Bihar and Maharashtra stood at 32 desks and 25 desks, respectively.[xiv] Possibly as a consequence, release of funds from the SHS to the Treasury took as long as five months in Maharashtra and over three months in Bihar and Uttar Pradesh.[xv] Delays at one level have a knock-on effect and often funds reach the last mile in the last quarter of the financial year.

Five-step Process in Reforming the CSS Design

These challenges highlight the need to move away from past reform efforts (which have focused on minor tweaks in CSSs) towards the first principles of the rationale behind specific purpose transfers. This will require a five-step process.

Moving from a Schematic to a Sectoral Approach

The first step is to limit the number of schemes. One way of doing this is to link finances to ‘national goals’. The committee of CMs on restructuring CSSs laid out nine key areas as part of the National Development Agenda for Vision 2022. It recommended that instead of the previous government’s strategy of bundling schemes under 22 umbrella programmes, funds could be released specifically for priority areas rather than multiple sub-schemes. This would give states the flexibility to plan activities within each priority area as per their own development needs. Steps in this direction have already been taken. The recently launched Samagra Shiksha – an overarching programme for school education extending from pre-school to class 12 – merged three previously independent CSSs. In theory the scheme allows states to prioritize interventions and sectors as per their need. Preliminary analysis of the scheme budget shows that indeed states are making decisions in keeping with their specific needs (albeit still guided by the GoI). Thus, while Uttar Pradesh and Bihar – which continue to lag behind in elementary education – allocated over 80% of their Samagra Shiksha budget for elementary education, states such as Haryana and Himachal Pradesh have focused on secondary education, allocating over 40% to the same. Similar steps should be taken in other sectors.[xvi]

Moving towards Block Grants

Having identified priority areas, the next step would be to ensure states have enough resources to fund these areas. Instead of allocations being linked to detailed and cumbersome planning and budgeting processes with restrictive, centralized guidelines, block grants could be given to states. This would allow for prioritization of different inputs and secure greater ownership by state governments. An example of this can already be seen in the Rashtriya Krishi Vikas Yojana (RKVY), a CSS established in 2007 to rejuvenate falling agricultural growth rates. Unlike most other CSSs, RKVY offers the flexibility to a state to choose activities under the scheme that most suit its requirements. Projects are prepared by the departments concerned and then scrutinized by a committee under the the state government’s Agricultural Production Commissioner. Most importantly, approval of the project is not done by the GoI but by the State Level Sanctioning Committee (SLSC), chaired by the Chief Secretary and with representatives from the Ministry of Agriculture and NITI Aayog as members.

Ensuring Equitable Interstate Distribution

Third, interstate distribution of the normative block grant portion of funding amongst states can be based on a formula that takes into account aspects like population, area and proportion of difficult areas, along with sector-specific needs. Differential cost-sharing norms that take note of the shortfall in service levels could further assist in ensuring that the distribution of funds fulfils the criteria of need and equality. Moreover, the formulaic nature of the grants will ensure predictability of fund flows and allow for better planning.

Reforming the Public Finance Management System

The fourth step is streamlining inefficiencies in the approval and fund flow process. This can be done by building a just-in-time Expenditure Information Network (EIN) which brings all expenditure units under one system. The first step in this process was undertaken in 2017, when the GoI mandated all CSS expenditure to be routed through the Public Finance Management System (PFMS). The system envisages each implementation unit to be under one system, thereby allowing the Centre and states to monitor funds at different levels. The problem, however, is that the system still functions as a push system, with funds being routed through multiple levels requiring approvals at every stage. By moving towards a pull system, each implementing unit could have the ability to automatically withdraw funds as needed. A defined resource envelope and appropriate access codes would ensure that funds are not misused.

Augmenting Capacity of the Evaluation Office

Finally, instead of focusing on monitoring the nuts and bolts of implementation, the GoI must build its capacity to develop a credible database on monitoring outcome indicators on a real-time basis. Currently, an inherent weakness in the CSS design is its focus on inputs. This creates perverse incentives for the entire administrative machinery to focus on ensuring adequate inputs, or at best, meeting output targets. Here, the DMEO’s role could be expanded by investing in systems to generate regular, credible and granular data on various outcome indicators and to conduct concurrent evaluations of key programmes. Over time, performance on outcomes could be linked to additional financial incentives available to states.

This piece was originally published on the Centre for Policy Research website as part of the ‘Policy Challenges 2019-2024’ series.