A background note can be accessed here: MoSPI’s Index of Service Production
The proposed Index of Service Production (ISP) relies heavily on GST data and administrative records to generate monthly indicators for the formal services sector. To what extent does this data architecture risk systematically under-representing India’s large informal services economy?
The proposed ISP is an important institutional step because India’s services sector now contributes more than half of the economy, yet lacks a high-frequency indicator comparable to the Index of Industrial Production (IIP). However, its underlying data architecture does create a significant risk of under-representing the informal services economy.
The index relies heavily on GST records, which are available monthly and therefore attractive for real-time measurement. But GST primarily captures formal economic activity and excludes service providers below the registration threshold, along with GST-exempt activities. In a structurally dual economy such as India’s, where formal and informal sectors coexist but do not always move together, this limitation becomes analytically important.
The challenge extends beyond GST. The proposed Annual Survey of Incorporated Services Sector Enterprises (ASISSE) also draws its frame from GST-network entities, which again limits informal sector coverage. Administrative datasets collected by ministries face similar constraints. As a result, a substantial segment of informal services activity may remain outside the index.
This issue becomes particularly important for sectors such as health and education, which are GST-exempt and together account for roughly 10 percent of services output. Since ASISSE data are still under development, these sectors are currently proposed to remain outside the ISP framework. Interpretation of the index, especially over time, would therefore require caution.
The methodology converts nominal GST turnover into volume indices using deflators such as Service Producer Price Indices (SPPIs). How robust is this approach in capturing real service sector output, given the heterogeneity of services and pricing dynamics?
The robustness of the ISP will depend significantly on how nominal GST turnover is converted into volume-based measures of output. GST data capture sales value rather than quantities, making deflation central to the credibility of the index.
Internationally, Producer Price Indices (PPIs) are commonly used for such exercises. India, however, does not yet have a comprehensive PPI framework for services. In its absence, the working group has proposed using Service Producer Price Indices (SPPIs), sector-specific consumer price indices where available, and the non-food CPI as a broader fallback deflator.
The larger concern lies in the highly heterogeneous nature of the services sector itself. Services ranging from telecommunications and finance to hospitality, transport, education, and healthcare exhibit very different pricing structures, productivity dynamics, and demand conditions. Applying broad deflators across such diverse activities risks oversimplifying underlying economic movements.
This challenge is arguably more complex in services than in manufacturing, where the IIP operates within a relatively more homogeneous production structure. The choice of deflator therefore becomes analytically crucial. Even small mismatches between price behaviour and actual output dynamics can distort volume estimates for specific sub-sectors.
Much will depend on the eventual operational methodology, particularly on how granular the deflation exercise becomes across different service categories and whether sector-specific price indicators are progressively strengthened over time.
The ISP adopts a “tri-source” model combining GST data, administrative datasets, and enterprise surveys across over 40 service sub-sectors. How does this integration affect statistical consistency and reliability across sectors?
The ISP’s proposed “tri-source” architecture, combining GST data, administrative datasets, and enterprise surveys across more than 40 service sub-sectors, is both ambitious and institutionally significant. It reflects an attempt to build a comprehensive, high-frequency statistical system for a highly diverse sector of the economy.
The integration model can improve coverage and reduce dependence on any single source. GST data offer monthly frequency, administrative datasets provide sectoral depth, and enterprise surveys can fill information gaps that transactional databases may not capture. Aggregation through a Laspeyres index using sectoral weights linked to economic contribution also provides a structured statistical framework.
At the same time, integrating multiple datasets introduces substantial methodological challenges. Different sources may vary in definitions, reporting standards, periodicity, coverage, and quality. Reconciling these differences while avoiding overlaps or double counting will require continuous harmonisation.
The issue is especially complex in services because many activities are intangible, rapidly evolving, and institutionally fragmented across ministries and regulators. Maintaining consistency across sectors would therefore require regular revisions, careful metadata management, and sustained coordination between statistical agencies and line ministries.
Despite these challenges, the exercise represents a major institutional undertaking by MoSPI. Its eventual credibility will depend not only on the initial methodology but also on transparency, revision practices, and the system’s ability to adapt as India’s services economy evolves.