India: A middle class too small to supplant China’s economic role

At a time when Chinese consumption is slowing significantly, could India become a driver of global consumption? For now, this prospect presents considerable difficulties, with many obstacles to overcome. Despite dynamic growth and the Indian economy’s expanding share of global output (8.2% of worldwide GDP in 2025), the middle-class consumer population remains modest and is nowhere near the size of China’s.

This is due to the concentration of employment in low-value-added and largely informal sectors. The few manufacturing sectors with high value-added and strong job potential — electrical equipment, electronics, automotive, chemicals and pharmaceuticals — continue to face constraints from an unfavourable business environment and a skills shortage that is holding back their move into premium market segments.

Furthermore, the highest-paying service sector jobs, particularly in information technology, are now threatened by the rise of artificial intelligence. Without concrete measures, the “demographic window” — which is expected to remain favourable for another 15 to 25 years — could close before India establishes a solid industrial base and a sufficiently large middle class.

India is struggling to establish a robust middle class

With an annual growth rate of 7.3% over the past decade (excluding the pandemic), India has one of the highest growth rates among the world’s major economies. Yet, its middle-class consumer base remains surprisingly small. When measured in current U.S. dollars, total consumption is very high in India (USD 2,241 billion in 2025), but this figure primarily reflects the country’s overall population size. By comparison, this consumption level falls 3.5 times short of China’s corresponding figure. Furthermore, per capita consumption adjusted for purchasing power parity (PPP) is 1.3 times lower than that of Indonesia, the Philippines and Vietnam. When evaluated at constant prices, consumption per capita has increased only slightly over the past thirty years compared to China and even Vietnam (Chart 1).

According to our estimates, based on the research conducted by López-Calva and Ortiz-Juarez (2014)[1], India’s middle-class consumer segment accounts for just 1.6% of the total population (see Table 1), or fewer than 24 million people. This is less than in Thailand and Vietnam (more than 30 million each) and far below China (approximately 490 million, according to the same definition).

The majority of households identified as middle class by standard Indian research institutions (the Centre for Monitoring Indian Economy [CMIE] and the National Council of Applied Economic Research [NCAER]) would, in fact, be categorised as “vulnerable” according to World Bank criteria: households that have officially emerged from poverty yet lack the financial stability necessary for sustained economic security.

Highly concentrated consumption and limited upward mobility

The structure of Indian household consumption reveals significant disparities between urban and rural areas. In 2024, average monthly consumption per capita amounted to 6,996 rupees in urban areas (approximately USD 84.5), compared with 4,122 rupees in rural areas (less than USD 50). The wealthiest 5% of urban consumers spent an average equivalent of USD 245 per month per person. Consequently, discretionary spending power remained concentrated among the highest income brackets.

The structure of consumption has changed over the past decade. The proportion allocated to food purchases (34.8% of total spending on average) declined by 4.3 percentage points between 2011–12 and 2023–24, while allocations to transportation and communications increased to 12.4% (+3.8 pp). Although the allocation to durable goods has increased, it remains below 7% of total household spending. This increase was driven more by a rise in the use of credit than by income growth. Over the past ten years, consumer loans have been the main source of household debt. Although still modest relative to GDP, household debt rose by 12.2 pp to reach 47.8% of GDP by the end of 2025, according to the Bank for International Settlements (BIS).

The labour market suffers from structural weaknesses

The limited size of the middle class and the low level of household consumption in India can be attributed, in particular, to the enduring structural weaknesses of the labour market. The labour market faces two major problems: insufficient job creation and the fact that most of the jobs created are low-skilled and low-paid.

Insufficient job creation

Despite the Indian economy’s robust growth, its employment rate remains substantially lower than in comparable Asian economies: 52.7% in 2025 according to the International Labour Organization (ILO), compared with 65.6% in Indonesia and 71.3% in Vietnam. This disparity is mainly due to low labour force participation among women (34% compared with 53.4% and 68%) and young people (40.8% of those aged 20-24 compared with 66.7% and 69.4%).

Over the past ten years, employment growth has averaged just 2.4% annually, whereas GDP growth was close to 6% per year.

The services sector (representing 33.2% of employment in 2025) accounted for the bulk of job creation during this period (+3.2% per year), far ahead of the manufacturing sector (+1.4%) and, above all, agriculture (+0.1%), which nevertheless remains the country’s largest employer (41.5% of the workforce). This is not enough to absorb the 10 to 12 million new entrants to the labour market each year.

A limited number of sectors have significant potential for creating high-quality, well-paid jobs

To identify sectors with high potential for job creation, we calculate employment elasticity both for aggregate economy (KLEMS data) and for the formal manufacturing sector alone (Annual Survey of Industries, 2025).

Over the period 1981–2023, aggregate employment elasticity remained low at a virtually unchanged level, close to 0.3. This stability, however, masks significant sectoral divergences: elasticity remained stable in the services sector throughout the period (0.5), but it declined sharply in the manufacturing sector (from 0.4 during 1981–1990 to just 0.1 during 2012–2023[2]): at the same growth rate, India now creates far fewer industrial jobs per unit of growth than it did forty years ago (Table 2). In the formal manufacturing sector alone, employment elasticity reached 0.7 over the past ten years, driven by very robust job creation. However, this positive trend has not been sufficient to offset the decline in informal sector employment, which accounts for 73.4% of total manufacturing employment but only 38% of the sector’s value added. Labour productivity in the formal manufacturing sector is thus estimated to be 5.4 times higher than in the informal sector.

In capital-intensive sectors, employment elasticity ranged from 0.5 to 0.6 (including electrical equipment, transportation, chemicals and metals), whereas in labour-intensive sectors — specifically wood, textiles and apparel, and manufacturing food — employment elasticities remained particularly low, or in certain cases, negative. In other words, growth in value added in these sectors with high labour absorption potential is primarily driven by productivity gains rather than job creation. Furthermore, even among companies operating in the formal sector, employment elasticity remains low at an average of 0.4. Consequently, the traditional mechanism facilitating labour mobility from the agricultural sector to the labour-intensive manufacturing sector has been very limited over the past forty years.

The only manufacturing subsectors that combine strong growth, high employment elasticity, and high productivity — thereby offering real potential for creating high-quality jobs — are electronics and information technology, electrical equipment, the automotive industry, the chemical industry, and, to a lesser extent, the pharmaceutical industry. In these subsectors, the formal employment elasticity ranges from 0.8 to 1.2 (Chart 2). However, in 2025, these jobs accounted for just 1.4% of total employment and 5.9% of GDP. The potential therefore remains negligible at the macroeconomic level.

Several structural constraints limit job creation in India:

- The high rigidity of the labour market: it consists mainly of small, informal businesses — those with fewer than 50 employees account for 65% of the productive sector — largely due to the stringency of labour legislation (for example, companies with more than 300 employees cannot carry out mass layoffs without authorisation).

- Numerous obstacles impede the growth of industrial and job-creating businesses: an unfavourable business environment (particularly difficulties in accessing land acquisition to build factories); high costs of accessing capital (particularly for small and medium-sized enterprises); insufficient export competitiveness in labour-intensive sectors (textiles, clothing), which encounter disadvantages, including less advantageous tariff conditions[3] than competing Asian economies.

- The mismatch between the skills of the available workforce and the needs of businesses. This mismatch constrains wage growth and the development of household purchasing power, obstructs industrial advancement, and, consequently, the creation of high-quality jobs.

Inadequate job quality due to a lack of appropriate skills

A predominantly informal and low-paid labour market

According to the ILO, 87.2% of total employment in India is informal, with 73.4% concentrated in the manufacturing sector alone, predominantly engaged in low-value-added activities that require few skills. Furthermore, the Periodic Labour Force Survey (PLFS) 2023/24 indicates that 58.4% of Indian workers are self-employed, 19.8% are day labourers, and only 21.7% are regular employees — the highest-paying status, with a wage gap of more than 60% compared to the other two categories. According to the ILO, 20% of the population captures 79% of aggregate labour income. The agricultural sector, which still employs 41.5% of the labour force while contributing less than 15% to GDP, is 3.7 times less productive than the services sector, with a median hourly wage that is 40 to 50% lower.

A widespread skills shortage

This concentration of earnings reflects the low skill levels of the jobs held: according to the PLFS, 88.2% of workers hold jobs requiring little or no skill. The proportion of skilled jobs is just 11.2% and has declined by 4.6 pp since 2018. The underlying challenge is not a shortage of graduates but rather a misalignment between what the labour market needs and the educational preparation provided to graduates[4], which results in their gradual downward mobility into positions far below their formal qualifications.

Unemployment, the downward mobility of graduates, and a shortage of technical skills

This disparity between skills and business needs directly contributes to unemployment among graduates, which reached 13.5% in 2025 according to the ILO, compared with 1.3% for individuals without a degree. Within the 25- to 34-year-old demographic, the unemployment rate stands at 15.2% for graduates (19.2% for those with a master’s degree), compared with an average of 6.1% across this age group. Only 8.3% of graduates secure positions that genuinely correspond to their educational attainment; 53% of bachelor’s degree holders and 36% of master’s degree holders are overqualified for their positions.

In low-skilled jobs, the lack of technical training represents a significant barrier to youth employment: according to the World Bank’s Enterprise Survey (2022), 17.8% of traditional service companies and 14% of “intermediate” manufacturing companies cite this as a major obstacle. In India’s manufacturing sector, 32.9% of positions are reportedly held by underqualified workers, according to the ILO, compared with 28.8% in Indonesia and just 16.5% in Vietnam.

A direct impediment to industrial advancement

A survey conducted by the Ministry of Skill Development & Entrepreneurship across seven key sectors — automotive, textiles, agriculture, livestock, IT, commerce and green energy — reveals that the pace of workforce training development significantly lags behind industry requirements. This forces companies to train a considerable number of young graduates themselves before they are ready for the workforce. According to the Telecom Talent in the 5G Era report by the Telecom Sector Skill Council, only 40% of IT and telecommunications graduates are actually immediately employable.

This disparity in skills particularly constrains the transition towards higher-value-added manufacturing activities within the sectors identified above (automotive, electronics and chemicals). Their growth trajectory depends on technological advancement — electric vehicles, semiconductors, specialty chemicals — and therefore on intermediate to advanced technical skills that are currently lacking. Should training provision fail to align with industry demand, employment elasticity and growth in these sectors will stagnate or decline, undermining the only source of high-quality job creation currently available — and consequently, the Indian economy’s ability to expand its middle class.

These structural limitations affecting national industrial development diminish its appeal to foreign investors, despite its size. The stock of foreign investment in India stood at USD 560 billion in Q2 2026, representing just 15% of GDP (compared with 20% in China, more than 52% in Vietnam, and around 67% of GDP in Thailand and Malaysia).

Artificial Intelligence: a new risk to job creation

Artificial intelligence (AI) may also limit the expansion of high-paid employment opportunities in the service sector. Its deployment poses a significant threat to employment in India, particularly in the highest-paying service sectors — the same sectors that possessed significant potential for expanding India’s middle-class consumer base.

Agriculture (41.5% of the labour force), construction (12%), and trade/transportation (18%) remain largely immune to the direct replacement of workers by AI, which reduces the overall macroeconomic scale of the coming shock. The service sectors offering the most lucrative employment — IT, finance and insurance — account for just 1.3% and 1.2% of total employment, yet these remain the primary targets of automation. In IT services companies, the adoption of AI is expected to reduce the need for entry-level roles (quality assurance, software testing and support roles) but increase demand for positions centred on platform development and management oversight. Current sector-specific evidence indicates a notable slowdown in job creation within IT services but no widespread contraction. A study by the Indian Council for Research on International Economic Relations (ICRIER) on the IT sector confirms a decline in job creation for routine and entry-level roles.

While new professions are emerging such as data centre operators, AI DevOps engineers who serve as intermediaries between software development and systems operations teams, and ethical AI specialists, the volume of employment opportunities generated remains substantially lower than the number of jobs at risk. Furthermore, the skill requirements for these newly created positions exceed those possessed by the existing cohort of graduates.

Reforms are a step in the right direction but do not go far enough

Since 2019, the Modi government has implemented a series of measures to develop industry, attract foreign direct investment, and gain greater integration in global trade (after decades of protectionist policies).

Corporate subsidy programmes (Production-Linked Incentives) have helped support the development of those sectors with the greatest potential for job creation (electrical, electronics, chemicals, and automotive). Similarly, labour market reform (which came into force in late 2025) may alleviate impediments to business expansion and the creation of formal jobs. However, its actual impact will depend on its adoption by the states, which retain considerable discretionary autonomy over labour law, and on its effective implementation on the ground. The free trade agreement signed between India and the European Union in late January 2026 (which is expected to take effect in 2027 at the earliest) should mitigate the tariff disadvantage experienced by labour-intensive industries: the EU has pledged to remove tariffs, which currently range from 4% to 12% on textiles and clothing.

However, these measures alone will be inadequate for creating high-quality jobs if the education system fails to meet the needs of Indian businesses. It is therefore imperative that India continues to improve its business environment and, even more importantly, reform its education system to promote job creation in high-value-added and well-paying sectors. Without concrete measures, the “demographic window” — which is expected to remain favourable for another 15 to 25 years — could close before India establishes a robust industrial foundation and a substantial middle class.

*****

While India’s economic growth has demonstrated considerable dynamism, this growth has not translated into the development of the middle class. The real medium-term concern involves the emergence of a “neo-precarious” middle class — substantial in terms of numbers but lacking the foundations necessary for advancing towards a high-income economy (households with stable employment, the ability to save, and social protection). The creation of formal, high-quality jobs is currently inadequate to sustain industrial expansion and its progression towards higher-value segments. The cultivation of certain niche activities, while promising, remains very limited and is far from benefiting the entire population. Finally, the development of AI poses more of a risk than a genuine opportunity for the country’s development. India lacks both the infrastructure and the educational system necessary to deploy AI in a way that would yield tangible benefits for its economy and labour market.

[1] These authors define the middle class as individuals for whom the risk of falling back into poverty is limited (less than 4% over four years), corresponding to daily per capita income ranging from 17 to 98 2021 PPP dollars (with vulnerability thresholds at 6.85 and 14 2017 PPP dollars, according to the World Bank). In the absence of reliable national income data collection mechanisms in India, our study applies these thresholds to the World Bank’s consumer expenditure data, an approach aligned with the objective of assessing actual purchasing capacity deployed in the market (although this tends to slightly underestimate the middle class in terms of income, as household savings remain unaccounted for in expenditure measurements).

[2] Employment elasticity remained relatively stable between 1990 and 2012.

[3] Historically, tariff rates imposed on Indian products exported to Europe have been higher than those applied to certain other Asian countries (Bangladesh, Vietnam, Pakistan).

[4] According to the World Bank and the NCAER, educational and training programmes are too theoretical and fail to develop cognitive and interpersonal skills. Vocational education and training are not designed to meet the needs of high-demand sectors. Collaborative relationships between educational institutions and employers — particularly regarding internships and on-the-job training — remain inadequately developed.