Over the last five years, India has taken important steps toward significantly reforming its welfare architecture ranging from direct benefit transfers (DBT), Ayushman Bharat, income support (PM-Kisan) to the implementation of the fourteenth finance commission recommendations. However, underlying these reforms are important unresolved and deeply contested questions about the architecture of the welfare state. In particular, the questions revolve around centralization and capacities of various levels of government to deliver. Welfare policy under the new government will necessarily have to confront these questions and the opportunities and challenges they present. The ability of the new government to navigate this terrain will determine its effectiveness and capability to deliver high quality public services to India’s poorest.
Technology, income support, citizens and bureaucracy
Technology has been at the heart of the welfare reform project over the last decade. In 2014, when the National Democratic Alliance (NDA) first rode to power, it embraced the Aadhar and DBT (effort and accelerated progress. In March 2014, only 28 schemes used DBT to transfer funds. By May 2019 this had increased to over 400. In January 2019, the first national attempt was made, with the launch of PM-Kisan to use the DBT architecture, to introduce a basic income support program in India. However, excessive reliance on technology to implement DBT has exposed three crucial limitations of the system – the last mile problem, lack of accurate data to identify the beneficiaries and alienation of the citizenry.
The key rationale for scaling DBT and moving toward direct cash transfers through income support programs is its ability to curb payment leakage and improve efficiency. In making the case for a Universal Basic Income (UBI), the 2017 Economic Survey argued that by moving resources directly in to beneficiary accounts, income transfers have the potential of cutting down bureaucratic layers. This could curb discretion, simplify monitoring and therefore reduce corruption.
However, recent studies show that rather than reduce bureaucracy, getting the DBT architecture right, requires significant bureaucratic intervention. From opening accounts to promoting financial literacy and facilitating bank transactions, local bureaucrats are critical to DBT. This is best highlighted in a recent, Niti Aayog commissioned process monitoring exercise of using DBT to access the Public Distribution System scheme in 3 Union Territories (Chandigarh, Dadra & Nagar Haveli, and Pudducherry) by Muralidharan et al. In their study, Muralidharan et al found that 20% beneficiaries reported non receipt of payment even though official records indicate a transfer failure rate of less than 1%. This gap is not a consequence of leakage, rather the study attributed gaps to lack of beneficiary awareness/knowledge of transfers and administrative issues including payments being made into bank accounts beneficiaries do not access, or processing errors.
This problem of administrative preparedness is also highlighted in a forthcoming World Bank report. The report argues that DBT requires sophisticated financial management and technical skill sets that are beyond the current bandwidth of the local, sub-district bureaucracy. In various states, some of the key financial management tasks have been outsourced to private players and state IT cadres however, capacity, just in terms of sheer human resources remains a problem. Moreover, anecdotal evidence suggests that that DBT has at least in the short run increased rather than reduced the workload at the frontline. Frontline officers are now responsible for tackling citizen claims and disputes on personal authentication, financial address information, payment settlements but without any improvements in human resource capacity and with skill sets remaining unchanged. This has caused significant, non-trivial disruption at the last mile. 
Importantly targeted programs like PM-KISAN require bureaucrats to identify eligible beneficiaries. To do this critical data like land records and Socio Economic Caste Census surveys need to be regularly updated and disputes between citizen claims and official records need negotiation. Doing this right, requires bureaucrats to dialogue with citizens, coordinate across departments and absorb feedback – a skill that Indian bureaucrats simply do not posses. Countries like Brazil and Mexico have invested in large cadres of social workers at the local government level to do just this. But in the rush to bypass bureaucrats through DBT and transfer cash directly in to bank accounts, this crucial investment has been ignored. Strengthening basic capacity and capability of frontline bureaucracy – notably in terms of its human resources, even if its only task is to move money, will require empowering local governments with skills and resources to be genuinely responsive to citizen needs.
However, responsive governments require active citizen participation. Digitized efficiency risks casting citizens’ as passive recipients of government largesse rather than active claimants of rights. This is not mere romantic activism. Technology by its very nature creates centralized systems that are distanced and bewildering for ordinary citizens in ways not different to the frustrating everyday encounters readers of this document have had with call center agents. Digitized welfare systems genuinely risk closing off spaces for citizens to complain, protest and demand accountability when rights are denied. The point here is not to argue against administrative efficiency, rather to highlight risks that need to be addressed. A balance needs to be struck between efficiency gains through centralized control and responsiveness through decentralized, citizen-centric governance. Striking this balance will be a critical challenge for the new government. One important way through which this challenge can be met is through strengthening the implementation of the Right to Information Act, particularly mandatory requirements for proactive disclosure of information to citizens. But for information to be empowering, it must be relevant for citizens. This requires citizens to be actively engaged with government in the process of generating information. Some governments, notably Rajasthan, have begun experimenting with ways of identifying relevant information through regular dialogues with civil society and building fora for making information available to citizens’ offline. These experiments ought to be studied and replicated.
Regulation vs. Public Provision
The launch of Ayushman Bharat in October 2018 marked the beginning of a significant architectural transition in India’s welfare system – a transition from direct provisioning (government running hospitals and schools) toward financing citizens (through income support and health insurance) and regulating private providers. But this transition poses a critical challenge. How does a state that struggles with routine tasks build capability to regulate a sector as complex as health care?
A functioning health insurance system must ensure that patients are not under-treated; nor over-treated or over-charged. Ensuring this requires adaptive price setting, strict regulation, third-party monitoring and, quality improvements in public sector hospitals.
Pricing. Getting prices right is the central dilemma in any insurance program and one that all countries struggle to solve. This is because prices need to fulfill the dual function of ensuring “neither too much, nor too little”. But costs for the same procedure are likely to differ across hospitals because of quality, location and capacity. Therefore, a single price can never ensure that both constraints are effectively met, and in fact, it is certain that these prices will never be the “right” prices. Moreover, if the price is too low for a hospital, it will either choose not to enroll in the scheme, or it will deny services. When the price is too high, the hospital will make additional profits, or worse, try to convince patients to receive services when they are not needed.
The one thing that countries implementing large-scale insurance programs have in common is a large analytical and data center that continuously examines procedures, procedure coding and charges from the insurance scheme. Prices have to be frequently negotiated and updated based on the data, and this is a job for specialized teams of hundreds in each state.
Regulation and Insurance Fraud: In tandem, insurance schemes require creating a strong regulatory framework for fraud control. India’s current regulatory environment is seriously weak. Ila Patnaik et al have shown that all 17 insurance ombudsman offices in India are currently vacant with a backlog of 9000 complaints. Gaps in the current regulatory framework imply that there is no established procedure for settlement of claims, redress of consumer behavior against rejection of claims or even penalties for rejecting claims in in violation of existing regulations. This in turn creates incentives for regular violation of norms by insurance companies. Not surprisingly, the complaints rate in India is orders of magnitude higher than comparable jurisdictions across the globe. The success of PMAY is now intrinsically tied not only to the functioning of the health department, but also the criminal justice and court systems. A new, stronger legislative framework for regulation and insurance fraud is urgently needed.
The only way to ensure that these conditions for implementation success are met is through massive investments in a skilled workforce. In the United States, the (largely) single-purchaser Medicare scheme employs 6000 people to cover 44 million beneficiaries. These are all highly trained administrative staff handling insurance audits, pricing and medical records; dealing with anti-trust cases and fraud and examining billing issues in each state. India has nowhere near this scale of staff. Consider U.P where the scheme may cover 50% of the population, or 100 million people. That would imply that the administrative staff to run a single purchaser scheme should be above 10,000. Across India most state insurance schemes and the RSBY were run by trusts and offices the employed fewer than 100 staff. In UP, the RSBY headquarters had only 42 staff. The point is simply this. Running a scheme as complex as a large-scale health insurance program requires people. Since the expertise currently does not exist (at least at this scale) PMAY will have to develop the necessary institutions that will train these professionals.
Finally, and this is a lesson that applies more broadly to the dilemma of public vs private provisioning, there is no getting around the critical need to strengthen public systems. Specifically in health care, in the long run well-functioning public hospitals will provide a much-needed backstop against predatory practices, denial of service and overcharging in the private sector. Especially in districts where competition is limited, public hospitals will limit the monopoly power of the private sector, flush with the new money from the scheme. A framework for transferring resources from the scheme to help government hospitals improve their quality is just as important as funding flows to the private sector.
The Centralization vs. Decentralization tug of war
A critical underlying issue that influences the dynamic of welfare provisioning is that of financing India’s welfare scheme. Scholars of federalism in India have characterized India as a “quasi-federal” or a federal system with “centripetal bias”. In this context, the union government has historically played an important role in financing welfare related schemes and the dynamic of fiscal federalism and center-state relations is a critical ingredient that influences implementation of welfare policy.
In the last five years, India has transformed its fiscal federal architecture, devolving unprecedented levels of independence and resources to it’s states. However, in doing so it has created a new dilemma – one of disparities between state governments to be able to deliver welfare and other services. Effectively this has meant that in absence of central support, the states with lower capacity to design and implement continue to fall further behind high-income states widening regional disparities across the country.
This has not to say that the recent changes in its fiscal federal architecture should be rolled back. (These included the dismantling of the planning commission; the establishment of the NITI Aayog and the GST council as new institutions for negotiating center-state relations; and the implementation of the recommendations of the 14th finance commission (FFC) that sought to enhance fiscal decentralization to States by reducing central government control over state spending). At one level, these changes were long overdue. Historically, India’s fiscal federal architecture has been extremely centralized. The delegation of fiscal powers and responsibilities specified in the constitution as Singh and Rao argue reflects a “centripetal bias with the Centre having ‘overwhelming and overriding’ economic powers.” To put it in perspective, States incur 60 percent of government expenditure but collect only 40 percent of revenue. All state borrowing is subject to approval from the union.
The Planning commission, with its mandate of centralized planning emerged as a critical instrument for centralizing India’s fiscal system. To illustrate, calculations by the Fourteenth Finance Commission reveal that between 2005 and 2012, central government spending on State subjects increased from 14 to 20 per cent and its spending on Concurrent Subjects increased from 13 to 17 per cent. The bulk of this spending took place through specific-purpose transfers or Centrally Sponsored Schemes (CSS) financed and monitored through the Planning Commission’s plan funds, making CSSs one of the most important vehicles of central transfers to states. To illustrate, during the 11th Five Year Plan (2007-2011), scheme specific transfers accounted for over 40% of central transfers to states. Importantly, these schemes, as argued comprehensively by Avani Kapur in this volume, were designed and implemented in an extremely centralized, tightly controlled, one-size fit all architecture that undermined state flexibility.
States have long complained about fiscal centralization. From as far back as the 1969 National Development Council meeting to consultations with the Fourteenth Finance Commission in 2013, State governments’ argued strongly against the centralized nature of CSS and encroachment by the Centre on the constitutional mandate of state governments. Chief Ministers’ in particular, resented having to seek Planning Commission approval for state plans. Several commissions noted the negative consequences, as articulated by state governments, of the proliferation of CSSs and recommended reduction in their quantity and greater flexibility in their design. This deep centralization coupled with the fact that the post-liberalization era had rendered the centralised planning process irrelevant made the need to reforming (and even dismantling) urgent and critical.
The setting up of the Niti Aayog and the implementation of the recommendations of the 14th finance commission recommendations to enhance fiscal devolution to states marked an important juncture in fiscal federal relations in India. However, it also brought to the fore an important new fault-line in the centralization vs. decentralization tug of war that has shaped the dynamics of fiscal federalism in India. The future shape of federalism in India will depend significantly on how the new government navigates this fault line and the institutional space it creates for re-negotiating center-state relations.
The primary challenge for fiscal federalism in India comes from India’s growing regional disparities or what Vijay Kelkar has recently identified as India’s “developmental imbalance”. Governance capability is the primary driver behind these expanding disparities. To illustrate, a study by Mundle et al which ranked governance performance, defined as public service delivery, of 19 major States in 2001 and 2011highlights the large and persistent development distance between States in terms of per capita income and service delivery outcomes. In infrastructure, the density of state highways in Karnataka at 10.8 kms. per 100 sq. kms. was five times that in Odisha at 1.95 kms per 100 sq.kms in 2011. Power availability in Bihar in 2011 at 117 k was three times higher than 36 kWh in 2001, but it was still only 1/15 that of Gujarat at 1559 kWh. In the social sectors the Bihar literacy rate in 2011 at 47% was only about half that of Kerala at 91%, while the infant mortality rate of 91 per 1000 newborns in Odisha in 2001 was nearly nine times the rate in Kerala. Importantly and unsurprisingly, governance (bureaucratic capacity and decision-making processes) is the critical ingredient for moving the needle on service delivery improvements in low-income states like Bihar and Madhya Pradesh. This exposes, as Mundle argues, a critical tension between high-income states and low-income states. High-income states have the capacity to design and implement their own schemes and can better leverage fiscal decentralization. Low-income states need CSS with their centralized design and implementation structure, simply because they lack planning capabilities.
The Planning Commission with its Plan funds and CSS played an important (albeit imperfect) role in responding to these governance deficits by designing and financing (through plan funds) schemes linked to core public services. And for all its flaws, States through the National Development Council, had access to an institutional space for debating and exercising some leverage over Plan funds. Moreover, the Planning Commission played a critical co-ordination function. CSS, linked to Plan funds, were administered by line-departments but the Commission played a critical policy coordination function linking schemes to state plans and determining resource availability. Further, the five-year plans and plan allocations, enabled states to broadly predict the quantum of plan funds they were likely to receive, ensuring somewhat predictable expenditure over a five year cycle. In the absence of the planning commission, this role has been taken over the Finance Ministry and line departments. As a result, the last five years have arguably witnessed an even greater centralization of schemes. In fact, recent analysis of state budgets highlights that in the last few years, the share of central schemes as a percentage of overall central expenditure has increased 9% (for CS and CSS respectively) in 2016-17 to 12% (CSS) and 9% (CS) respectively.
The response to the institutional vacuum left behind by the dismantling of the Planning Commission isn’t one of simply re-introducing the planning and budgeting function of the commission albeit with greater flexibility to states. This is because despite their intent, CSS as an instrument have been unable to effectively respond to the challenge of regional disparity. Low income and weakly governed regions are simply unable to utilize CSS funds. As central schemes increased in the last two decades, central spending increased in richer rather than poorer states. As Rathin Roy highlights, in 199-91, CSS and CS spending accounted for 1.22% of total expenditure in Gujarat and 13.9% in Rajasthan. This reversed in 2013-14. Spending in Rajasthan dropped to 3.2% and increased in Gujarat to 13.12%. This trend is even more acute at the district level. Based on an analysis of six key schemes, the 2016 economic survey highlighted that India’s poorest districts received barely 40% of the total funds allocated to a state.
The primary fault line in India’s fiscal federal architecture is this: fiscal centralization, arguably critical for poorer states, has in fact benefited richer states. Addressing this challenge will require a significant overhaul of existing financial instruments available with the central government. Specifically, it will need a careful rebalancing of the centralization vs. decentralization dynamic in ways that provides necessary governance support to poorer states while ensuring fiscal autonomy. Crucially, the center needs to reorient its role from being a micro-manager of schemes and programs to playing a far greater strategic role, building national policy frameworks and providing technical capacity to States to enable and empower them to plan, design and implement social policy programs. Balancing these tensions and re-shaping the role of the center requires the creation of institutional spaces for center-state deliberations. This can be achieved through a revitalized Inter-state council (ISC) tasked specifically with creating a deliberative space for centre-state dialogues on welfare policy and negotiating the tensions that have come to shape center-state fiscal relations. The fourteenth finance commission had recommended the creation of such a body. The new government must implement this recommendation with urgency.
Welfare policy in India is poised at a critical juncture. Reforms to India’s welfare architecture over the last five years present new challenges and fault-lines that need resolution, if India is to move toward creating a new 21stcentury welfare architecture. India doesn't need new schemes rather it needs consolidation and balancing between competing welfare strategies. Getting this right will require significant investments in state capacity. This is the welfare task for the new government.
 DBT leverages Aadhar and banking penetration to move money directly in to beneficiary bank accounts)
 Universal Basic Income: A Conversation With and Within the Mahatma (pp. 173-212). New Delhi: Govt. of India, Ministry of Finance, Economic Division.
 Muralidharan, K., Niehaus, P., & Sukthankar, S. (2017, September). DIRECT BENEFITS TRANSFER IN FOOD: Results from One Year of Monitoring in Union Territories (Rep.). Retrieved https://econweb.ucsd.edu/~kamurali/papers/Other Writing/20170905_UT_DBT_Report.pdf
 Aiyar, Y., Bhattacharya, S., & Pritchett, L. (2019, March 14). The Solutions State: Why the digital needs the human. The Indian Express. Retrieved from https://indianexpress.com/article/explained/the-solutions-state-why-the-...
 Narayanan, R., Swamy, R., & Dey, N. (2019, April 4). The politics of information. Indian Express. Retrieved from https://indianexpress.com/article/opinion/columns/the-politics-of-inform...
 Patnaik, I., Roy, S., & Shah, A. (2018, May 21). The rise of government-funded health insurance in India (Working paper No. 231). Retrieved https://www.nipfp.org.in/media/medialibrary/2018/05/WP_231.pdf
Rao, M. G., & Singh, N. (2006). The political economy of federalism in India. New Delhi: Oxford University Press.
 Reddy, Y. V., & Reddy, G. V. (2019). Indian fiscal federalism. New Delhi, India: Oxford University Press.
 Kelkar, V. (2019). Towards India's New Fiscal Federalism. Seminar 717: The Union And The States,717, 34-38.
 Mundle, S., Chowdhury, S., & Sikdar, S. (2016). Governance performance of Indian states 2001-02 and 2011-12 (No. 16/164).
 Mundle, S. (2019). The Case For A Radical Transformation. Seminar 717: The Union And The States,717, 38-43.
 Aiyar, Y., & Tillin, L. (2019). The Problem. Seminar 717: The Union And The States, 717, 10-13.
 Roy, R. (2019). Changing Fiscal Dynamics. Seminar 717: The Union And The States, 717, 18-22.
 Ibid, 22.
 Eight Interesting Facts About India (pp. XI-XIV). (n.d.). New Delhi: Govt. of India, Ministry of Finance, Economic Division.