PSUs are back-but they should exercise caution when implementing artificial intelligence

The reason is that a PSU duality exists in India which aggravates the risks. The government’s public disavowal of PSUs exists cheek-by-jowl with heavy reliance on them.

While many ministers have publicly stated that PSUs drain national resources, official data demonstrates that the PSU comeback movie is playing out in India as well.

An annual survey of PSUs by the department of public enterprises—excluding banks, insurance companies and departmentally run companies—shows that the number of PSUs in India grew from 298 in March 2015 to 495 by March 2025, a 66% growth in 10 years.

Global data also shows a spike in PSU growth. According to the Organisation for Economic Co-operation and Development (OECD), a Paris-based multilateral think tank, both the number of PSUs and the value of their assets in real terms have roughly doubled between 2013 and 2024.

This indicates the growing significance of PSUs, especially their expanding role in the global economy. An OECD report, Ownership and Governance of State-Owned Enterprises 2026, further confirms the scale of PSU contribution to the global economy: 33,834 financial and non-financial PSUs had combined assets of $141 trillion in 2024, equivalent to 126% of global gross domestic product (GDP).

In terms of optics, the PSU resurgence may seem counter-intuitive when viewed against the Indian government’s recent disinvestment of majority shareholding in IDBI Bank in favour of Canada-based Fairfax Holdings.

While this disinvestment had been delayed for many years, its impending closure has now emboldened officials to announce a long list of assets for sale and monetization. At the same time, the government has both a monetary as well as strategic reliance on the public sector for achieving some of its targets.

The government is dependent on dividends from PSUs to meet a part of its non-tax revenue targets. Excluding surplus transfer from the Reserve Bank of India (RBI), the government received over ₹89,000 crore as dividend income from PSUs during 2025-26.

Even if we assume that RBI’s surplus transfer will remain unchanged during the current year ₹2.87 trillion, the government has budgeted to receive over ₹1.04 trillion as dividend income from PSUs during 2026-27.

But PSUs also have a larger strategic role. As oil companies, for example, they provide strategic price and supply stability during crises. Many manufacturing PSUs operate in sensitive areas where allowing private or foreign investment could end up endangering national security.

PSU banks, given their extensive network, provide the last-mile linkages for multiple government welfare schemes, including crediting direct benefit transfers to beneficiary accounts. In times of crisis, PSU banks act like conduits for liquidity or interest rate transmission.

Whenever RBI has to smoothen currency volatility, market trades are often coordinated through the foreign exchange desks of PSU banks. More importantly, funds for India’s defence imports are largely routed through PSU banks.

PSUs also meet another social objective by creating employment opportunities. In the public eye, Indian PSUs were always viewed as over-staffed and unproductive.

The truth is probably more nuanced. In terms of non-financial PSU employment as a percentage of total employment in the country, India does not figure among the top 20 countries in the OECD study. Surprisingly, Singapore tops the list with its PSUs having a 16% share in total employment.

This duality—publicized repudiation of PSUs while data reveals other truths—aligns well with India’s fuzzy PSU governance norms. The government has drafted clear and lucid guidelines for some issues but is vague on many critical areas. For example, there is a well-defined dividend policy framework for PSUs, but there seems to be no clarity on resolving the governance muddle inherent in PSU chairmen also acting as chief executives.

These grey areas may be advantageous when coercing PSU action or deploying moral suasion for strategic reasons. However, this same ambivalence could become a liability once multiple PSUs start implementing AI to improve productivity and efficiency standards.

The need for sharper and explicit regulatory guidelines has acquired some urgency, especially because the speed of implementation seems to have overtaken the need for caution and risk mitigation.

As a recent McKinsey article argued, PSUs are not private businesses and, hence, the risk of getting AI wrong through hasty implementation has multiple pitfalls.

The solution, therefore, is to have clear-cut controls put in place through transparent and well-reasoned guardrails and regulatory structures.

The author is a senior journalist and author of ‘Slip, Stitch and Stumble: The Untold Story of India’s Financial Sector Reforms’ @rajrishisinghal

Rajrishi Singhal

Rajrishi Singhal has been a senior journalist, a banker and a public policy analyst-cum-consultant. He has previously served as Executive Editor at The Economic Times, Executive Editor at Mint, Head (Policy, Research & Strategy) at a private sector bank, and Senior Fellow for Geoeconomic Studies at a Mumbai-based think tank. Rajrishi has a Master’s degree in Economics from Jadavpur University, Kolkata, and is the recipient of two prestigious fellowships: Gurukul Chevening Fellow at the London School of Economics (1997-1998), and the C.V. Starr Fellow at the Center for Advanced Study of India, University of Pennsylvania (2002). Rajrishi was an independent director on the board of self-regulatory organisation Advertising Standards Council of India (2021-25); he has also served on two government committees appointed to re-examine policy options in areas of financial services. Rajrishi's book on the long and inconsistent arc of financial sector reforms in India--called “Slip, Stitch and Stumble”--was published by Penguin Random House in 2024. He lives in Mumbai.