Archives: Publication

Should social registries be used to decide who receives cash transfers for women? The answer depends on the purpose of the transfer. If the scheme is income support, a registry can help identify eligible households. If the transfer is meant to recognise unpaid domestic and care work, or strengthen women’s control over money, household poverty filters weaken the case for the scheme.

Social registries and changing poverty

The World Bank defines social registries as information systems that support outreach, intake, registration and assessment of potential eligibility for social programmes. They are entry points into welfare schemes, not merely databases. Leite et al. describe them as systems with a social policy role and an operational role. The distinction is useful because a registry built only for verification can become an exclusion device. One built for inclusion must allow registration, correction, appeal and outreach.

This is especially relevant in India. Desai et al., using India Human Development Survey data from 2004-05, 2011-12 and 2022-24, show that households move in and out of poverty as the economy grows. A welfare database fixed at one point in time will miss some who fall into distress later and retain some who have moved out. That argues for registries that can be updated, corrected and used during shocks.

Social registries are now common. The World Bank says 62 countries had operationalised them by 2024, covering 1.1 billion people. Brazil’s Cadastro Único, linked to Bolsa Família, remains the best-known example. It shows both the promise and the limits of such systems. A registry can improve targeting and coordination, but it cannot decide the moral purpose of a welfare transfer.

India’s social registries

India already has registry-like systems. The Union government launched e-Shram in August 2021 as an Aadhaar-seeded national database of unorganised workers. By December 19, 2024, it had more than 30.48 crore registrations. Workers aged 16-59 can register if they meet the eligibility criteria; the official FAQ also says the database may be used in emergency or pandemic-like situations.

Rajasthan’s Bhamashah scheme began in 2008 and was relaunched in 2014. A World Bank note says around 50 lakh women were enrolled in the initial phase and 29 lakh accounts were opened. The scheme later became an end-to-end platform for cash and non-cash benefits. Karnataka’s Kutumba takes another route. It is an entitlement management system being developed into an integrated social information system. Its base is the Public Distribution System database, with other government databases linked to it.

The contrast is useful. Rajasthan relied on a large enrolment exercise. Karnataka built on existing administrative data and allowed updates through the system. For states, the second route is usually cheaper and less disruptive, especially when PDS, pensions, land, caste, disability and income records already exist. The risk is poor data quality. That risk does not disappear through a survey; it has to be handled through correction, verification and appeal.

Cash transfers to women need a purpose

Unconditional cash transfers to women have spread across state politics. Kotiswaran and Jana estimate that about 118 million adult women in 12 states receive such transfers, with more states promising similar schemes. Their work also finds no evidence, from research across five states, that such transfers discourage women’s education or paid work, or reinforce the gendered division of labour.

There are three broad arguments for these transfers. The first is income support. The second is recognition of unpaid domestic and care work. The third is women’s agency through direct control over a predictable sum of money. These are different claims. They cannot all be handled through the same eligibility logic.

If the transfer is income support, social registries fit the job. Household income, landholding, employment status, ration card status, disability, age and other markers can help identify who should receive the money. The registry should still be open to revision because poor households are not a fixed population.

If the transfer is payment for unpaid domestic and care work, the case for targeting is weaker. Women across income classes perform such work. Poor women may face harsher constraints, but household poverty is not the basis on which the work exists. A transfer justified as recognition should therefore be broad-based, with few exclusions. If it is filtered through household assets, income certificates or land records, the design changes the purpose of the scheme.

The same applies to agency. A woman’s control over money is shaped by marriage, inheritance, work, mobility and household bargaining power. These do not map neatly onto a poverty line. A scheme designed for agency should avoid rules that make access depend on male relatives, local certification or repeated documentation.

Maharashtra, West Bengal and women’s cash transfers

Maharashtra and West Bengal show how state governments are now resolving this question in practice. Maharashtra made e-KYC compulsory for beneficiaries of the Mukhyamantri Majhi Ladki Bahin scheme in 2025. The scheme pays ₹1,500 a month to eligible women. A later report said the beneficiary count fell from 2.4 crore to nearly 1.7 crore after the April 30 e-KYC deadline, though officials said the reduction also reflected eligibility violations such as income-tax status, age and overlap with other schemes.

The Maharashtra case is not a simple story of digital cleaning. Reports from Pune described Aadhaar-bank linking problems, incorrect entries and e-KYC errors that delayed payments for eligible women. The state then asked district collectors to use anganwadi workers for field verification so genuine beneficiaries were not excluded because of procedural mistakes.

West Bengal has moved in a similar direction through its socialregistry.wb.gov.in portal. Reports say the Annapurna Bhandar scheme, which replaces Lakshmir Bhandar, offers ₹3,000 a month to eligible women through DBT, with registration and status checks routed through the state portal. The eligibility filters include age, government employment, pension and income-tax status.

Both states are treating women’s cash transfers as targeted income support. Once that decision is made, a social registry is a defensible tool. The deeper issue is that the design has already chosen one justification over the others.

Building social registries without exclusion

States should build social registries, but they should not confuse database construction with welfare design. The cheaper route is to begin with existing administrative records, as Karnataka has done, rather than wait for a fresh census-style exercise. Voluntary enrolment alone will leave out the most vulnerable. Migrant workers, homeless women, widows, newly married women, disabled persons and those without stable documents will not always find their way to a portal.

A credible registry must therefore have four features: regular updating, assisted enrolment, field verification and a working appeals process. It should show beneficiaries why they were rejected or removed. It should allow correction without forcing people to start from the beginning. It should use local workers for outreach, but pay them for the work.

Cash transfers should also sit inside a wider policy for women across the life cycle. Kotiswaran and Jana argue that transfers need support from care infrastructure, financial literacy, predictable payments, grievance systems, school-level education on the division of labour, subsidised kitchen technologies, community kitchens and crèches. That is the right way to read the limits of cash. A monthly transfer can ease a constraint. It cannot, by itself, change the distribution of unpaid work.

Social registries can improve delivery when states know what they are trying to deliver. They cannot rescue a confused scheme. If women’s cash transfers are income support, targeting through a registry is consistent. If they are recognition for unpaid care work or a claim to economic agency, governments should say so and design them with far fewer gates.

Bornali Bhandari is a Professor and Co-Director of the Centre for Gender and Macroeconomy at NCAER.Views are personal.

While India has a large institutional network for vocational education and training (VET), there are systemic issues such as poor perception, weak industry linkages, inadequate quality, and low public investment. Studying VET systems of Germany, Canada, and Singapore – with diverse features that shape worker productivity and employment outcomes – Afridi and Chandna draw policy lessons for overhauling India’s vocational education system.

India’s vocational education and training (VET) system is central to improving labour productivity, yet a key paradox persists: while the share of unskilled workers has declined, the proportion of formally trained workers has stagnated (Periodic Labour Force Survey (PLFS), 2017-18 to 2023-24), and productivity growth remains modest (Afridi et al. 2025). The cross-country evidence suggests that the quality of skill training in India may not be adequate enough to translate into high-skilled employment (Afridi and Chandna 2026). This reflects not just a shortage of skills, but also poor quality and weak alignment of training with labour market needs.

India’s skilling challenge 

Despite a large institutional network of over 14,000 Industrial Training Institutes (ITIs) and significant seat capacity, utilisation is low at around 48%, and employment outcomes are extremely weak, with placement rates as low as 0.09% (NITI Aayog, 2023). These outcomes point to systemic issues, including poor perception of vocational education, weak industry linkages, inadequate training quality, and low public investment. India spends only about 3% of its education budget on VET in 2021 (Ministry of Education, 2022), whereas countries such as Germany, Canada, and Singapore allocate around 10-13% of their education expenditure to vocational training in 2020 (Organisation for Economic Co-operation and Development (OECD), 2023; Singapore Department of Statistics, 2024).

International experience

We study three diverse VET systems that vary in the timing of VET integration and institutional design, which shape their employment and labour productivity outcomes.

Early Hybrid VET (Germany): Germany introduces VET at the upper-secondary level, combining classroom instruction with structured apprenticeships in firms. This model is characterised by strong employer involvement in training, financing, and curriculum design, along with nationally standardised certifications that ensure high skill portability (OECD, 2023).

Tertiary-Hybrid VET (Canada): Canada has a VET model at the post-secondary level, where community colleges offer applied diplomas and certificates integrated with work-based learning such as internships and co-op programmes. Employer involvement is more limited and largely advisory, and certification portability varies across provinces due to decentralised governance (OECD, 2023, Skolnik 2021).

Tertiary VET (Singapore): Singapore adopts a tertiary VET model through a dual-track system (Institute of Technical Education (ITE) and Polytechnics), supported by strong State funding and industry advisory mechanisms. It features clearly defined progression pathways, allowing students to move from the ITE to Polytechnics and further to universities, along with opportunities for lifelong learning (UNESCO-UNEVOC, 2021).

Figure 1 shows that countries with a higher share of VET-trained adults are associated with higher employment rates, and notably, those where VET is concentrated at the upper-secondary level lie above the ‘fitted line’, indicating stronger employment outcomes for countries that introduce VET at earlier stages of education. This is reinforced by evidence on productivity in Figure 2, where countries with upper-secondary (early) VET also exhibit higher average labour productivity than those where VET is concentrated at the post-secondary level, in 2021.

Figure 1. Cross-country comparisons: Employment rate and VET share (2021)

Source: Data on the share of adults with vocational education or training and employment rates are taken from International Labour Organization (ILO), 2026 and OECD, 2023. Authors’ calculations.

Notes: (i) ILO data are used to compute the share of youth (15-54 years of age) with vocational education or training and to derive employment rates (from unemployment rates, 15+ age group) for 2021. (ii) Countries are classified by dominant VET level among adults aged 25-34 based on ‘Education at a Glance, 2013’. (iii) Sample size includes 29 OECD countries and Singapore.

‍Figure 2. Average labour productivity across countries at the dominant level at which VET Is offered (2021)

Source: Data on productivity are from ILO (2025) and VET classification is based on OECD (2023). Authors’ calculations.

Notes: (i) Labour productivity is measured as output per hour worked (GDP (gross domestic product) in constant 2021 international dollars, PPP (purchasing power parity)), based on ILO modelled estimates. (ii) Blue bars represent countries with upper-secondary dominant VET (Level 1), while green bars represent post-secondary/short-cycle tertiary dominant VET (Level 2).

International experience indicates that effective VET systems share three core features: strong public investment, deep employer engagement, and clear pathways for skill portability.

We estimate the benefit-cost ratio (BCR), which evaluates expected earnings gains from vocational training relative to its costs across the three countries with diverse VET systems, to assess the returns from education systems which formally embed skilling.

Our analysis shows high returns from these systems. The BCR is around 6.9 in Germany, 5.1 in Singapore, and 4.2 in Canada, indicating that earnings gains are several times higher than the training costs in each case, as shown in Table 1.

‍Table 1. Benefit-cost ratios (BCRs) per person for VET programmes 

‍Sources: Eurostat (2024): Germany earnings; Statistics Canada (2023): Canada earnings; World Bank: PPP conversion factors; OECD (2023): Relative earnings (ages 25-34), employment rates, and cost data for Germany and Canada; Ministry of Manpower, Singapore (2024): Earnings; Ministry of Education, Singapore (2024): Cost data. Authors’ calculations.

Notes: (i) Calculations are on a per-person basis using the most recent available data (2020-2023), assuming labour market entry at age 25 after completing VET. (ii) Benefits are measured as expected earnings between ages 25 and 34, assuming no real annual earnings growth, and calculated as earnings × employment rate, (iii) Germany: hourly median wages (2022) are annualised (38 hours/week, 52 weeks/year) and adjusted using relative earnings indices. (iv) Canada: median annual earnings (2020) for ages 25-34 are derived using OECD relative earnings ratios. (v) Singapore: starting salaries of Polytechnic graduates (2023) are used as a proxy for earnings at age 25. (vi) Total costs are calculated as annual VET programme costs multiplied by average programme duration.

‍Policy implications for India

For India, the policy implications are clear in terms of strengthening the VET system.

First, education reforms should prioritise the early integration of VET into the general education system. While the National Education Policy (NEP), 2020 recommended such integration, progress in this direction has been slow.

Second, improve the relevance and perception of VET, especially ITIs, through state-level branding campaigns, better alignment of ITI course offerings with local labour market demand, and strengthening the quality of training, infrastructure, and instructors within ITIs.

Third, enable portability across vocational tracks by operationalising the National Credit Framework and establishing clear progression pathways between ITIs, polytechnics, and higher education institutions. This would involve assigning credits to vocational qualifications based on learning outcomes and training hours, and establishing rules for the recognition and transfer of these credits across institutions and programmes.

Fourth, strengthen public-private partnerships by actively involving industry in curriculum design, training delivery, certification, and apprenticeship programmes across the VET ecosystem.

Recent policy initiatives such as the Employment Linked Incentive (ELI) scheme, PM Internship Scheme, and ITI upgradation efforts indicate growing attention to employment, but they remain fragmented and insufficient. They focus more on hiring incentives and infrastructure rather than systemic reforms like formal training, certification, and long-term skill development pathways (Press Information Bureau, Government of India, 2024).

Overall, India’s skilling challenge is not merely about expanding access but about transforming the quality, structure, and labour market relevance of vocational education. Comprehensive reforms, aligned with global best practices, are essential to unlock the full potential of India’s workforce.

The views expressed do not necessarily reflect that of the authoRs’ organisation or the I4I Editorial Board.

A well-designed social registry, containing comprehensive household information, can address the ever-changing state of poverty and simultaneously drive critical human development goals.

Should social registries be mixed with cash transfers to women? The answer depends on the policy objectives. But first, what are social registries? These are defined as information systems that support outreach, intake, registration, and determination of potential eligibility for one or more social programmes (Leite et al. 2017). Ideally, social registries should be designed to be dynamic, open, and provide continuous access to registration, typically through on-demand applications with active outreach to vulnerable populations. This matches India’s experiences of poverty, where people transition in and out of poverty (Desai et al., 2025). Dynamic social registries are particularly suited to address the dynamic nature of poverty in India.

The use of social registries is a well-established global practice. According to the World Bank, as of 2024, 62 countries had operationalised these systems. The earliest and most prominent example is Brazil’s social registry database, Cadastro Único, which is closely linked to the country’s landmark cash transfer programme, Bolsa Família.

Closer home, the national e-Shram database functions similarly to a social registry. Several States have also developed their social registries, such as Rajasthan’s Bhamashah programme and Karnataka’s Kutumba initiative.

The e-Shram portal was developed after the pandemic to tackle any future national crisis like Covid-19. Registration was voluntary and unconditional for unorganised workers aged 16-59 years, who were not members of EPFO, ESIC, or NPS. Rajasthan’s Bhamashah scheme, initiated in 2008 and relaunched in 2014, relied on a survey-based method to enrol 50 lakh in its initial phase.

Karnataka’s Kutumba is an Entitlement Management System which is being developed into an Integrated Social Information System, consisting of a social registry, integrated beneficiary management systems, payments platform and a grievance redressal system. Starting 2019, it leveraged the existing Public Distribution System (PDS) database alongside other State and Central government registries, requiring no separate survey. Kutumba also accepts on-demand applications from households and individuals through an online portal. The database aggregates individual attributes (such as name, gender, education, occupation, and disability status) and family attributes (caste, income, land holdings, deprivation status, and Priority Household status under the NFSA). Both the Bhamashah and Kutumba databases are linked with Aadhaar and utilise unique family identifiers. The latest State to adopt this approach is West Bengal, where residents have been asked to submit a range of demographic and economic data, either online or offline, on a voluntary basis.

Why give cash transfers to women?

Cash transfers (UCTs) to women have become immensely popular with States. Setting political economy incentives aside, the literature (like Kotiswaran and Jana 2026) suggests three primary justifications: income support, recognition of unpaid domestic and care work, and the promotion of female agency.

Income Support: It is highly practical to use social registries to identify women who require targeted income support. Social registries in India are family databases and have a range of information on income, assets, land holdings, etc., which can then be used to determine eligibility.

Payments for unpaid domestic & care work and/or improving women’s agency: If cash transfers are distributed to women for these reasons, the transfers should be truly universal rather than targeted. Because the majority of women, regardless of socioeconomic class, perform disproportionate unpaid care work and benefit from enhanced individual agency. Using a social registry to filter eligibility based on household poverty metrics defeats the core purpose of the programme.

Policy alignment in States

States like Maharashtra and West Bengal are currently reforming their cash transfer schemes for women by requiring them to complete the electronic Know Your Customer (e-KYC) verification (Maharashtra) or requiring them to register on the State’s new social registry portal (West Bengal). Essentially, both States are treating cash transfers to women as targeted income support programmes. Consequently, their reliance on social registries is logical and aligned with their policy goals.

States are encouraged to build dynamic social registries. The point is how to build these comprehensive databases. One is the survey-based approach of Rajasthan versus Karnataka’s Kutumba model of using available databases to build one. Given that the latest census is already underway and States themselves have a lot of administrative data, the latter method may be a more cost-effective way of building social registries. Good practices also involve reaching out to the vulnerable population on an ongoing basis. Of course, the data has to be validated and verified. And they have to be dynamic to allow for the dynamic nature of poverty prevailing in India.

Cash transfers, whether universal or targeted, are just one tool for addressing women-related concerns. A well-designed social registry, containing comprehensive household information, can simultaneously deliver targeted income support to women and drive critical human development goals.

While data, technical tools, and proven models are readily available, State governments must align their policy priorities and design dynamic systems which maximise the economic and social returns on their investments.

The writer is a Professor and Co-Director of the Centre for Gender and Macroeconomy at NCAER. Views are personal.

Joining the CPTPP could integrate India into global supply chains, boost exports and support its Viksit Bharat 2047 ambitions before tougher entry conditions emerge.

India means to be a developed country by 2047. The slogan, Viksit Bharat, is everywhere but the arithmetic is quite daunting. No large economy has travelled from being poor to prosperous in a couple of generations without accessing world markets and the supply chains that run through them. Japan, South Korea, China and the entrepôts of Southeast Asia all did so. India, by contrast, still finds the main entrance blocked.

The multilateral trading system is broken. The World Trade Organization’s Doha round expired years ago without ceremony. Its dispute settlement machinery — long the protection of small countries against big ones — has been allowed to seize up. Rules are now being written in regional clubs that do not wait for Geneva. India can either mourn the old order or join the new one.

Perhaps the most attractive of the new clubs lies in the Pacific. The Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) binds 12 economies — among them Japan, Canada, Australia, Vietnam and, since December 2024, Great Britain. Together they generate nearly $13.5 trillion, about 13.5 per cent of world output, and house 500 million consumers. Tariffs on most of the goods traded within the bloc have been eliminated. Commerce among members has grown by more than 7 per cent a year since 2019, briskly outpacing their trade with everyone else.

It is the composition of trade, as much as its size, that should interest India. CPTPP exchange is well diversified — primary goods, processed products, components, capital equipment — and tightly enmeshed in the value chains that organise modern manufacturing. Membership buys not merely cheaper access to a dozen markets but also a place in the vast networks that connect them.

As it stands, India’s presence is modest. The bloc buys around $62 billion of Indian goods a year, under 1.5 per cent of its imports. Even what India sells — pharmaceuticals, refined fuel, gems, textiles, metals — its share is thin. Across its 50 largest export lines into the group, the median share is only about 5 per cent. Measures of trade intensity confirm what the eye suspects, that India ships less to the CPTPP members than the size of those economies would lead one to expect. The shortfall reflects barriers, not incapacity.

Sceptics will recall why India withdrew from the Regional Comprehensive Economic Partnership in 2019: Fear of Chinese goods arriving through the front door and drowning out our own manufacturing sector, especially small and medium firms. The worry was not unfair but the CPTPP answers it twice over. China is not a member, and the agreement’s rules of origin are specifically built to stop outsiders from free-riding through token processing. And the pact India would enter is a gentler thing than the American-designed original. When America walked out of the Trans-Pacific Partnership in 2017, the remaining members went on to form the CPTPP and suspended 22 of the most onerous clauses: On intellectual property, investor-state arbitration and public procurement — precisely those that would have pinched Indian generics makers and cramped policy. The clauses are dormant, not dead.

Therein lies a case for haste. The suspended provisions can be revived by consensus. Should Washington return — and American trade policy has recently been nothing if not changeable — the hard terms would snap back and India’s price of entry would rise. The door stands open and the barrier to entry is low largely because the bloc’s weightiest potential member is loitering outside it. That state of affairs may not last.

A domestic bonus comes attached. India’s exporters are generally concentrated in the West and North. The East and South, for all their promise in electronics, food processing and textiles, have trailed behind. Several CPTPP members lie directly off India’s eastern coast. Tying trade to that map could draw money and infrastructure towards the laggard regions and diversify India’s export map spatially.

The familiar objection, that the club’s labour, environmental and intellectual property standards lie well beyond India’s reach, is overdone. India already has trade deals with Britain, Japan, Australia and Singapore, which comprise a third of the membership. Indeed, free-trade agreements have lately become the centrepiece of India’s commercial strategy; joining the CPTPP is the natural next step. Accession to the CPTPP is not automatic and may not even be easy, but the alternative of assembling equivalent access one bilateral deal at a time will certainly be a lot harder.

India is right to pursue agreements with America and the European Union — and those talks will proceed at their own pace. The CPTPP is finished, road-tested and waiting, with many years of growth behind it and rules that India, for now, can stomach. In a world where the certainties of trade and geopolitics are dissolving, a ready-made bargain on tolerable terms is a rare prize. India should try and claim it before the moment passes entirely.

The authors are, respectively, with Johns Hopkins University, Washington DC, and NCAER, New Delhi