By Amit Kapoor and Sai Sruti Panda
The importance of cities in driving India’s growth is well established, with urban centres accounting for close to 60 per cent of the country’s GDP. Yet the story of what drives this urban economy has often been easier to tell for organised and registered factories than for the vast network of enterprises operating across manufacturing, trade and services. Through data sets such as the Annual Survey of India, India has long had a structured view of registered factories in the manufacturing sector. Since 2021-22, the Annual Survey of Unincorporated Sector Enterprises (ASUSE) has strengthened understanding of the unincorporated non-agricultural enterprise economy, other than that captured in ASI. But it has remained harder to observe how this enterprise economy differs across cities. The ASUSE 2025 million-plus-cities report marks an important shift in that direction.
According to ASUSE 2025 data, India has an estimated 7.92 crore unincorporated non-agricultural establishments which employ around 12.8 crore workers, and for the first time, the Ministry of Statistics and Programme Implementation (MoSPI) has also released separate estimates for million-plus cities. The report is significant because it moves beyond national and state-level aggregates and shows how enterprise density, sectoral composition and value generation vary sharply across cities.
The latest city-level estimates based on ASUSE 2025 provide separate data for 46 million-plus cities, offering a more granular view of how small enterprises are distributed across urban India. According to the report, India’s 46 million-plus cities account for around 13% of the country’s unincorporated establishments but contribute nearly 21% of the sector’s Gross Value Added. Among the 46 million cities, 6 cities, Kolkata, Surat, Greater Hyderabad, Delhi, Ahmedabad and Greater Mumbai contributed to around 40 per cent of the estimated establishments in 2025. The report also suggests that an establishment’s scale in a city doesn’t translate into higher productivity. Of these 6 cities, Kolkata has the largest estimated enterprise base, at about 8.84 lakh establishments, but when we examine establishment productivity, its GVA per establishment is Rs 2,91,107. On the other hand, among other million-plus cities, Pune has about 2.44 lakh establishments, but its GVA per establishment is Rs 4,61,116. Whereas Greater Hyderabad has an estimated 6.08 lakh enterprises, its GVA per establishment is also high, i.e., Rs 7,14,113. As the report notes, these numbers vary across city-level estimates and are subject to sampling variability; however, they still offer an overall landscape view of establishments operating in these cities when interpreted with caution.
The sectoral composition of establishments across manufacturing, trade and services reveals the distinct economic structure of India’s million-plus cities, showing how differently urban enterprise systems are organised across the country. For instance, Ludhiana shows how older industrial capabilities continue to shape the unincorporated sector, with roughly one-third of its establishments engaged in manufacturing. Coimbatore and Ahmedabad also show a significant manufacturing presence. Manufacturing-oriented cities require stronger support for technology adoption, logistics, production networks and cluster upgrading. In contrast, cities such as Faridabad, Bengaluru and Nagpur are dominated by other services. These service-heavy cities need interventions focused on skills, digital adoption and market access.
Trade-dominant cities such as Srinagar and Vijayawada represent a different pattern. In these cities, the unincorporated sector is less about production and more about commerce, distribution and local market intermediation. Such cities may have large numbers of establishments, but their productivity depends heavily on market infrastructure, logistics, storage, credit access and the ability of traders to connect with wider regional demand. City-level enterprise data provides a basis for more differentiated and targeted policy design.
In urban unincorporated areas across most million-plus cities, proprietary and partnership establishments account for 95% to 100% of total establishments. However, the report also reveals considerable variation in the share of these establishments employing hired workers. In cities such as Srinagar and Delhi, more than 40% of establishments employ hired workers, indicating a greater capacity to move beyond self-employment. In Gwalior, the share is below 8%, as most establishments remain confined to owner-operated activity. This distinction is important because hired-worker establishments are more likely to reflect some level of scale, market demand and business expansion. Cities with very low shares of such establishments may have many enterprises, but a large proportion of them remain closer to subsistence activity than growth-oriented entrepreneurship.
The data also show that ownership, which is the operational characteristic of an establishment, reveals a sharply uneven geography of women’s participation in India’s enterprise economy. Several cities in Gujarat and Maharashtra report some of the highest shares of female-owned proprietary establishments at around 40%, including Surat, Vadodara, Ahmedabad and Rajkot in Gujarat, and Pune, Pimpri-Chinchwad, Nagpur, Navi Mumbai and Chhatrapati Sambhajiagar in Maharashtra. At the other end, Srinagar records a share below 10%, while Patna and Varanasi remain at around 12% to 13%. This variation shows that women’s participation in urban enterprise ownership is highly uneven and shaped by local social, financial and labour-market conditions.
The ASUSE city-level estimates should therefore become a regular part of India’s urban economic measurement. Since cities were treated as a separate sampling stratum, with samples drawn independently, the report offers a more grounded view of enterprise patterns across cities. For researchers and policymakers, this adds a new layer of evidence for understanding how enterprise density, sectoral composition, employment structure, and value generation vary across million-plus cities.
A more productive urban economy cannot be built only through infrastructure and large firms. It also requires data to better understand the urban economy and what needs to be done to further improve the performance of millions of small establishments that form the operating base of Indian cities. ASUSE 2025 insights on cities shed light on the future of urban productivity, which will depend not only on how many enterprises cities contain but also on how much value those enterprises generate.
(Amit Kapoor is Chair & Sai Sruti Panda is Researcher at Institute for Competitiveness. X: @kautiliya).
The article was published with The Sunday Guardian on August 9, 2026.
