Robust and internally consistent economic statistics are a precondition for effective industrial policy. Comment with reference to the debate over manufacturing GVA estimates in India.

Industrial policy is only as sound as the numbers it is built on. India's manufacturing Gross Value Added (GVA) is estimated by the National Statistical Office (NSO) from two very different streams, and the recent debate over whether the official figure is overstated shows why statistical consistency is a policy precondition, not a technical footnote.

Why statistics condition industrial policy

• Target-setting: the "Make in India" goal of raising manufacturing to 25% of GDP, pursued through the ₹1.97 lakh crore Production Linked Incentive (PLI) outlay across 14 sectors, is measured against the GVA series itself — an inflated baseline flatters progress [4].
• Resource allocation: incentive design, sectoral subsidies and credit targeting depend on knowing which segments are actually growing.
• Credibility: investors and multilateral lenders discount data that cannot be reconciled, raising the cost of policy signalling.

The manufacturing GVA debate

• Two measurement universes: the Annual Survey of Industries (ASI) covers the organised factory sector (10+ workers with power, 20+ without) [3], while the Annual Survey of Unincorporated Sector Enterprises (ASUSE), launched in 2021, covers informal units where manufacturing contributes about a fifth of GVA [2].
• Divergence: economists have argued that an "alternative estimate" built bottom-up from ASI plus ASUSE yields a lower manufacturing GVA than the top-down National Accounts figure, implying the sector's GDP share may be overstated.
• Methodological roots: differing reference periods, sample-based versus near-universe coverage, and imputation for the informal sector make the two streams hard to splice.
• Institutional response: the NSO's new GDP series with base year 2022-23, released in February 2026, updates sources and weights — a step towards closer alignment [1].

Rebasing alone will not settle the question; what is needed is transparent publication of reconciliation methods, faster ASUSE-ASI turnaround, and independent scrutiny through the National Statistical Commission. Statistics are ultimately an instrument of accountable governance: getting the manufacturing number right is the first condition for getting the manufacturing strategy right.