Archives: Publication

India faces a persistent challenge in generating formal, productive jobs for its growing workforce. The four new Labour Codes rolled out in November 2025, seek to address the structural imbalance in the workforce by simplifying 29 existing labour laws into a more coherent framework. 

Over the next five weeks, based on their new NCAER working paper, Farzana Afridi and co-authors will review these reforms. Each of the explainers (one per week) on the four new Codes – Code on Wages, Code on Social Security, Industrial Relations Code, and Occupational Safety, Health and Working Conditions Code – will delve into the timeline of the reform and its legislative provisions, and present a critical review of the theoretical and empirical evidence on the potential effects on employment and labour productivity, including formalisation of the workforce. The final post in the series will synthesise the possible implications of all the four codes on India’s labour market.

In this introductory post, Afridi lays out the structural weaknesses in India’s workforce and provides a global perspective on the country’s employment and labour productivity landscape. 

India’s economic growth over the past three and a half decades has been striking but the headline numbers carry a persistent and increasingly urgent challenge of addressing the structural weaknesses in India’s labour force.

First, India’s labour force participation rate sits below the trend line for economies at comparable economic levels (Figure 1). This reflects a structural failure to productively engage a young population that is growing rapidly (NCAER Report, 2025).

Figure 1. Labour Force Participation Rates (LFPR) by GDP Per Capita (2024)

Source: Authors’ calculations. World Bank database (2025).
Notes: (i) The per capita GDP (gross domestic product) series is measured at constant 2015 US$ prices. (ii) LFPR is for ages 15+. (iii) Per capita GDP classifications follow the World Bank’s FY2025 income group thresholds. (iv) Sample consists of 169 countries.

‍Second, India’s workforce continues to exhibit disproportionate concentration in agriculture and self-employment, and limited expansion in salaried work (Periodic Labour Force Survey (PLFS) Annual Report, 2025). As seen in Figure 2, India’s agriculture sector continues to house the lion’s share of employment (about 45%), despite contributing less than 20% to total GVA (gross value added). The expected transition of surplus agricultural labour into formal wage employment remains incomplete. Labour market adjustment is taking place through informalisation, with workers entering the labour force in low-productivity and precarious forms of employment.

‍Figure 2. Sectoral shares in total GVA (gross value added) and employment in India

Source: Authors’ calculations based on data from PLFS 2017-18 to 2023-24, Ministry of Statistics and Programme Implementation (MoSPI)
Note: GVA is in 2011-12 constant prices.

India’s employment and labour productivity challenge within a global perspective

Not surprisingly, India’s share of wage and salaried workers lies below the international average (Figure 3). This structural imbalance gets reflected in India’s anomalously higher share of self-employment relative to its per capita income (PLFS). This formalisation deficit remains the central labour market challenge as a large and increasingly educated work force is unable to access stable contract-based work at scale.

Figure 3. Proportion of wage and salaried workers by GDP per capita (2024)

Source: Authors’ calculations based on World Bank database (2025).
Notes: (i) Proportion of wage and salaried workers is given as percentage of total employment. (ii) Income classifications follow the World Bank’s FY25 thresholds. (iii) Sample consists of 169 countries.

Third, the informal structure of the workforce is accompanied by a consistent decline in India’s labour-capital ratio as income (Figure 4). The Penn World Table data (v11.0, 2023) tells us that India’s capital stock expanded by over seven times between 1990 and 2023, while employment roughly doubled in size. Capital is substituting for labour at a dramatic pace, even in sectors where India’s comparative advantage should favour the opposite.

‍Figure 4. Labour intensity of production by GDP per capita (1990-2023) 

Source: Authors’ calculations based on data from Penn World Table 11.0 (2023), Groningen Growth and Development Centre and World Bank database (2025).
Notes: (i) L/K (labour/capital) ratio is the number of persons engaged (in millions) divided by capital stock measured at constant 2021 US$ prices (in millions), (ii) The per capita GDP series is measured at constant 2015 US$ prices. (iii) The colour scheme distinguishes the World Bank Income Groups (FY2025 thresholds) with the countries arranged in decreasing order of per capita GDP after India (right-hand side graph). (iv) Sample (left-hand side graph) consists of 171 countries. (v) 95% confidence interval bands are shown. A 95% confidence interval means that, if you were to repeat the experiment with new samples, 95% of the time the calculated confidence interval would contain the true effect.

‍The predominant informal structure accompanies low worker productivity. India’s output per worker is slightly above global labour productivity levels but still has a significant bridge to cross to reach the per capita GDP of high-income countries, even by 2047 (Figure 5). Cross-country regressions further indicate India’s weak relationship between productivity growth and income expansion, reflecting a dual nature of narrow concentration in capital- and skill-intensive sectors, and a larger share of the workforce engaged in low-productivity activities (Figure 5).

Figure 5. Labour productivity by GDP per capita (1990-2018)

Source: Authors’ calculations based on data from Economic Transformation Database (ETD; 1990-2018), Groningen Growth and Development Centre and World Bank database (2025).
Notes: (i) Output per worker is GVA (measured in constant 2015 US$) divided by persons employed. (ii) Per capita GDP is in constant 2015 US$. (iii) Income classifications follow the World Bank’s 2017-18 thresholds. (iv) The colour scheme distinguishes the World Bank Income Groups with the countries arranged in decreasing order of per capita GDP (2018) after India. (v) Coefficients (right-hand side graph) are plotted with 95% confidence interval bands. (vi) Sample (left-hand side graph) consists of 51 non-OECD (Organisation for Economic Co-operation and Development) countries, as available in the ETD.

‍India’s new Labour Codes: What do they entail and what is the likely impact?

In summary, India faces a persistent challenge in generating formal, productive jobs for its growing workforce, with only a small share employed in the formal wage sector. The four new Labour Codes rolled out in November 2025, seek to address the structural imbalance in our workforce by simplifying 29 existing laws into a more coherent framework. This marks a shift from a fragmented legal system to a unified structure aimed at reducing regulatory complexity and improving compliance.

The writer is professor of Economics, ISI (Delhi), and visiting professor, NCAER.

Underemployment and overwork coexist in India’s youth labour market, revealing how work is unevenly distributed across people and places.

India’s youth labour market presents a puzzling picture. Fewer than half of young adults aged 20–29 are employed. Yet among those who are employed, long workdays are common, often extending well beyond a standard eight hours once commuting is included.

This contrast is not easy to reconcile. If jobs are scarce, why are many workers overburdened? And if work is available, why do so many remain outside the labour market?

The answer lies in how work is distributed across people, types of jobs, and places.

Who Gets Access to Work and Who Is Left Out

The unevenness is most visible in who gets access to work. The estimates based on data from India’s Time Use Survey 2024 reveals that nearly 79 percent of young men are in paid employment, compared to just about 18 percent of young women. In several large states such as Bihar and Uttar Pradesh, fewer than one in ten young women are employed.

This gap largely explains why overall participation remains low and varies sharply across states. Where more women work, employment rates rise; where they do not, the demographic dividend remains unrealised.

In effect, India’s employment challenge is as much about expanding access across the nation as it is about creating jobs.

Where Work Is Created and Why It Remains Fragmented

Even among those who do find work, it remains unevenly distributed across the type of employers. Informal (household) enterprises remain the default entry point into work – even for those with higher education. Only about 9.5 percent of young adults are employed in formal enterprises – corporate sector, government and NGOs – compared to roughly 37 percent in household enterprises. In rural India, this imbalance is even more pronounced.

The consequence is not just lower job quality, but uneven labour utilisation. Employment in household enterprises are often fragmented, limiting both the number of hours people can work and the consistency of that work. This reflects the small scale of such enterprises and unstable demand.

Even when jobs exist, they do not absorb labour evenly – leaving many underemployed while others remain excluded from work altogether.

Work Is Not Shared Evenly

This unevenness is clearest in the number of hours young people work.

On average, employed young adults spend about 6 hours and 55 minutes a day at work. This appears close to a standard workday. But the average conceals a sharply divided reality. At one end, about one-quarter of young workers put in more than 8 hours a day. At the other end, one-third of workers in informal enterprises work less than 6 hours.

Long hours reflect strong demand and inefficient job design, while short hours indicate underemployment – people working less than they would prefer because adequate work is unavailable. The result is a labour market that overuses some workers while leaving many others underutilised.

The Clock Reveals What Employment Data Misses

Looking only at paid work captures only part of the picture; how people spend a full day offers a fuller view of how labour – paid or unpaid – is actually used.

Young men work significantly longer hours in paid employment – about 7 hours a day compared to roughly 5 hours for young women. Yet once unpaid household and care work is included, the pattern reverses: young women, on average, work about 9.5 hours a day, compared to roughly 8 hours for young men.

This issue is not how many people work, but how work is distributed across activities. It shows that the issue is not only insufficient employment, but how time is allocated – often in ways that limit participation in paid employment.

Where You Live Shapes How Long You Work

Even for those in work, where people live shapes how they experience work. When commuting time is included, the length of the workday changes sharply. The share of young workers in formal employment spending more than 8 hours a day on work-related activities rises from about one-quarter to over 60 percent. More than one in three spends over 9 hours.

This reflects a deeper spatial mismatch between jobs and housing. Employment opportunities are often concentrated in specific urban locations, while affordable housing is located far from them. The result is a longer effective workday – driven not by work, but by getting to work.

Labour market outcomes are shaped as much by urban design as by employment policy.

India Is Not One Labour Market

These patterns are not uniform across the country. Some states absorb a majority of their young workforce, while others struggle across both rural and urban areas. For instance, employment participation among young adults ranges from about 58 percent in Gujarat to around 37 percent in Bihar – a gap of over 20 percentage points.

This variation goes beyond participation. States differ not just in how many young people work, but in how work is organised – across formal and household enterprises, across hours worked, and across distances travelled. In more urbanised and economically advanced states, gender gaps in work intensity tend to be narrower, and labour markets are better able to match workers to jobs.

These differences suggest that labour market outcomes are shaped less by national constraints alone and more by state-level institutions, infrastructure, and economic structures. India’s labour market is not one system, but many operating under different constraints.

Fixing Allocation, Not Just Supply

If work is unevenly allocated, expanding employment alone will not resolve the imbalance.

First, access constraints shape who can enter the labour market. A large part of the challenge lies in improving access to work, particularly for women, by addressing mobility, safety, and the transition barriers such as care work burden. Without this, even expanding labour demand will leave a significant share of the workforce underutilised.

Second, the structure of firms determines how labour is absorbed. A labour market dominated by informal enterprises limits the effectiveness with which workers can be absorbed. Moving more workers into formal enterprises is not only about job quality; it also increases the intensity and stability of work, allowing labour to be used more productively.

Third, spatial organisation affects the efficiency of work. When jobs and housing are poorly aligned, commuting time expands the effective workday without increasing productive work. Improving transport systems and aligning affordable housing with employment centres can reduce this distortion and improve both productivity and well-being.

Finally, labour policy must account for how time is allocated beyond paid work. Unpaid care responsibilities continue to shape who can participate in employment and how much they can work. Expanding access to childcare and related support systems is therefore not only a social objective, but an economic one.

From Population to Utilisation

India’s demographic dividend is often understood in terms of numbers – a large and growing working-age population. But numbers alone do not generate growth. What matters is how effectively that population is utilised.

The challenge, therefore, is not only to bring more people into work, but to ensure that work is distributed and organised to use time and effort more fully. India’s demographic dividend will be realised not through more jobs alone, but through better use of work – across people, places, and time.

Vidya Mahambare is Union Bank Chair Professor of Economics and Director (Research and Fellow Programme in Management) at Great Lakes Institute of Management (GLIM). Poonam Munjal is a Professor at the National Council of Applied Economic Research (NCAER). Palash Baruah is a Fellow at National Council of Applied Economic Research (NCAER). Niharika Gowthaman is a Research Associate at the Great Lakes Institute of Management (GLIM). Ayushi Mishra is a Graduate Student at the Great Lakes Institute of Management (GLIM). Views are personal.

The Periodic Labour Force Survey (PLFS), launched in 2017, addressed this by providing annual estimates for rural and urban areas and quarterly estimates for urban areas, replacing the episodic framework with continuous measurement. 

India’s struggle with timely labour statistics dates back decades. The National Sample Survey Office’s (NSSO’s) Employment and Unemployment Surveys operated on thick and thin rounds, with comprehensive data collected only roughly every five years — a significant inadequacy for a large, fast-changing labour market. The Periodic Labour Force Survey (PLFS), launched in 2017, addressed this by providing annual estimates for rural and urban areas and quarterly estimates for urban areas, replacing the episodic framework with continuous measurement. However, PLFS itself retained limitations in frequency, geographical granularity, and international comparability. The 2025 redesign directly addresses these gaps.

The PLFS, revamped from January 2025, pursues three objectives: monthly estimates of key labour indicators — worker population ratio (WPR), labour force participation rate (LFPR), and unemployment rate — at the all-India level for both rural and urban areas; extended quarterly coverage to rural areas for the first time, enabling combined national estimates; and continued annual estimates in both usual and current weekly status on a calendar year basis.

Sampling methodology has been substantially strengthened. The first stage units (FSUs) surveyed have increased from 12,800 to 22,594, while households per FSU have risen from eight to twelve, yielding a total sample of approximately 270,472 households — a 2.65 times increase. Critically, the district now replaces the NSS region as the basic stratum, making PLFS estimates representative at the district level and significantly improving the granularity for sub-state labour market analysis.

The rotational panel design has been restructured significantly. Previously, urban households were visited once per quarter with no revisit provision for rural households. Under the new design, all selected households are visited four times across four consecutive months, ensuring 75% matching between consecutive months and 50% between successive quarters. The selection method has shifted from PPS with replacement to simple random sampling, with sub-stratification introduced in urban areas.

Beyond sampling, several additional items have been incorporated into the first visit schedule, including five education-related questions on class completed and years of schooling; a question on the nature of the vocational training certifying body; two questions on land possessed and leased out; and four questions on household income from rent, interest, pension, and remittances. These additions make the new PLFS schedule considerably richer — effectively combining labour force measurement with household income, asset, and education profiling in a single instrument.

Recognising that the July-June framework made India’s labour statistics difficult to compare with international databases — all of which operate on calendar year cycles — and to provide a meaningful transition between the old and new series, MoSPI has released a back series of PLFS estimates on a calendar year basis. The back series begins in 2021 and not earlier for two reasons. First, 2020 was severely disrupted by Covid-19, making a clean, fully canvassed calendar year impossible. Second, the PLFS schedule was expanded around this period to incorporate a consumption block — household usual monthly consumer expenditure collected through five questions mirroring recent NSS rounds — making pre-2021 unit-level data structurally incompatible with later years.

A critical distinction must be noted. The calendar year back series through 2024 was constructed using the old sample design re-cut to a calendar year window; January 2025 marks the first data collected entirely under the new design. When multiple design elements change simultaneously — sample size, stratification, selection method, panel structure, and schedule content — differences in estimates may reflect genuine labour market shifts, seasonal effects, or design effects, making clean separation difficult. MoSPI itself has cautioned that post-January 2025 results cannot be straightforwardly compared with earlier publications.

India now has better labour market data than it has ever had. The revamp of PLFS comes at a particularly consequential moment — the four new labour codes covering wages, industrial relations, social security, and occupational safety have been brought into force in November 2025, reshaping hiring practices, formalisation incentives, and social protection coverage across the economy. Whether these reforms are delivering — whether formalisation is increasing, whether wage floors are binding, whether social security coverage is expanding — requires high-frequency, district-level data. India’s five trillion-dollar ambition and the Viksit Bharat vision both rest on assumptions about who is working, in what conditions, and with what protection. For too long, the data arrived too late and at too aggregated a level to be of real policy use. That constraint has now been lifted. The question is whether policymakers and researchers will use what they now have with the rigour the moment demands.

Jyoti is an associate fellow, and Dhruv is a research analyst at NCAER, New Delhi. Views are personal.