Structural Changes in Indian Economy since the 1950s: A Markov Chain Analysis

Understanding the evolution of the production structure of an economy has been one of the core macroeconomic issues for academicians, professional economists, policymakers and governments for decades. Within this context, this study investigated the structural changes in the Indian economy since the 1950s using a time-varying Markov chain (TVMC) model—a data-oriented non-parametric methodology—within a panel data framework. It aimed to understand whether India’s production structure followed the Classical–Chenery path of transformation or deviated towards a service-led trajectory. The production sector of the Indian economy was classified into three sectors–agriculture, industry and service. The data on the sectoral shares in the real gross value added (GVA) in the economy was taken from the National Statistical Accounts and deflated using the Wholesale Price Index (2011–12 as base year). The period under investigation was divided into three distinct phases: the dirigiste pre-liberalisation period (1951–1991), the neoliberal post-liberalisation period (1991–2011), and the digital integration period associated with the digital era (2011–2021). The findings showed that the Indian economy took a different growth path, experiencing direct service-led structural transformation of its production system. Its growth path showed (i) persistent decline in the GVA share of the agricultural sector throughout the study period, which was consistent with the agricultural transformation observed in developed economies; (ii) a stagnant GVA share of the manufacturing sector, indicating limited industrial deepening; and (iii) a sharp, sustained and premature expansion in the GVA share of the service sector. While projections about the production structure suggest that the service sector will sustain and improve its GVA share in the long-run, it would pave the path of jobless growth in the Indian economy. The novelty of this study lies in integrating historical data with a probabilistic model to analyse sectoral transformation, providing insights for future industrial and employment policies. However, its contribution to the literature is methodological, offering insight on how long-run historical data can enhance our understanding of structural change in an economy, which remains crucial for long-run policy framing.