India’s Sectoral Policymaking Needs to Move Beyond the Investment Race

BY DEVANSHU JHA

India’s states have become remarkably sophisticated at competing for investment. Across sectors from textiles and electronics to IT, data centres, automobiles and renewable energy governments offer capital subsidies, interest subventions, electricity concessions, tax reimbursements, land support and regulatory relaxations. The logic is straightforward: attract investment, create jobs and accelerate growth.

But there is a deeper question that India’s sectoral policies rarely asks with sufficient rigour: what does the government receive in return for the incentives it provides?

The problem is not that investment is unimportant. It is indispensable. The problem is treating investment itself as the outcome, rather than as the beginning of an economic transmission mechanism.

A 10,000 crore investment is not automatically equivalent to 10,000 crore of public value. Its development impact depends on how much of that investment is genuinely additional, how many productive jobs it creates, whether it raises local productivity, whether domestic firms enter its supply chain, whether technology and managerial capabilities diffuse, whether exports increase, and whether the fiscal and infrastructure costs borne by the state are justified.

Consider data centres. A state may celebrate the arrival of a large hyperscale facility and the associated capital investment. Yet data centres are highly capital-intensive and can generate relatively few direct jobs compared with the size of investment. Their larger economic value may lie elsewhere in enabling cloud computing, artificial intelligence, digital services, startups and productivity gains across other industries. If those spillovers do not materialise, the state may have subsidised electricity, land and infrastructure without capturing the full developmental benefits it anticipated.

The same distinction applies to textiles. A policy that rewards investment in machinery may increase installed capacity without necessarily improving total factor productivity, moving firms into higher-value products or strengthening export competitiveness. In IT and global capability centres, counting the number of units attracted can obscure a more important question: are firms locating only routine operations, or are they bringing research, product development and high-value decision-making functions? For MSMEs, the number of enterprises receiving support says little about whether those firms subsequently scale, formalise, export or become more productive.

This points to a fundamental shift that India’s industrial policy needs to make: from investment attraction to productivity-linked public value creation.

A useful framework could begin with a simple question: What is the Value for Government (VfG) generated by each rupee of public support?

Conceptually:VfG = Incremental Fiscal Return + Employment Value + Productivity Gains + Capability Accumulation + Ecosystem Spillovers + Strategic Value + Sustainability Benefits − Incentive Cost − Public Infrastructure Cost − Externalities.
But even this is incomplete without one crucial concept: additionality.If a company would have invested in a state even without a subsidy, the subsidy may have created little incremental investment. The relevant counterfactual is not the investment attracted, but the investment that would not have occurred without public intervention. This suggests a new architecture for industrial incentives. Every major incentive should be evaluated through four lenses :

First, Investor Value: What does the firm gain?

Second, Government Value: What fiscal and economic returns accrue to the state?

Third, Public Value: What benefits accrue to citizens through jobs, services, affordability and environmental outcomes?
Fourth, State Value: Does the intervention build capabilities that strengthen the state’s long-term economic position: skills, technology, supplier networks, innovation and strategic resilience?

These values can diverge sharply. A data centre may create enormous investor value, moderate direct employment and substantial strategic value but only if it catalyses an ecosystem beyond its physical walls. A textile factory may create jobs, but its long-term state value depends on whether it upgrades the capabilities of the surrounding industrial cluster.

The objective should be to maximise additional productive capacity per rupee of public support.

This would also change how policies are evaluated. Instead of asking only how much investment was attracted, governments should publish metrics such as incentive cost per incremental job, incentive cost per rupee of additional GVA, productivity gains, export additionality.If, local procurement, private R&D, supplier development and fiscal return. Large projects should be subject to ex-post evaluation to determine whether promised spillovers actually materialised.

Such a framework would not mean abandoning investment promotion. It would make it more intelligent. The competition among Indian states should no longer be a race to offer the largest subsidy package. It should become a race to demonstrate the highest economic additionality.If and public value generated per rupee of state support.

India has entered an era in which capital is increasingly mobile, states are increasingly competitive and fiscal resources remain scarce. The next frontier of industrial policy, therefore, is not simply to attract more investment.

It is to ensure that every rupee of public support buys something that markets alone would not have delivered and that the value created for investors is matched by enduring value for the government, the state and the public.

The question India’s sectoral policies must now ask is no longer: “How much investment did we attract?” It is: “How much additional productivity and public value did we create with the investment we subsidised?”

Devanshu Jha is a public policy expert and thought leader .He is an alumnus of London School of Economics, Lee Kuan Yew School of Public Policy and IIM RANCHI.