Growth in India’s Service Sector: Implications of Structural Breaks

Abstract

The paper is an attempt to identify the multiple structural breaks in India’s GDP, as well as its main growth enhancing sector i.e., services and its components and subsequently calculate the growth rate in different regimes. The paper uses the Bai-Perron (Econom 66(1), 1998, J Appl Econom 18(1), 2003) methodology of estimating multiple endogenous structural breaks (both pure and partial) in India’s service sector and its components and GDP during 1950–2010. Further, the paper uses the Boyce (Oxf Bull Econ Stat 48:385–391, 1986) methodology of estimating kinked exponential model of the growth rate, and further uses the Banerjee, Lumsdaine and Stock (J Business Econ Stat 10:271–287, 1992) test and the Lumsdaine and Pappel (Rev Econ Stat 79:212–218, 1997) test to check for the stationarity in the presence of structural breaks. The data used in this paper are the components of subsectors of services GDP and GDP at factor cost (with 2004–2005 as base). It is found that there is very little difference between the estimation of pure and partial structural break dates in India’s services GDP and its subsectors and four such breaks have been identified with help of Bai-Perron (Econom 66(1), 1998, J Appl Econom 18(1), 2003) methodology. The Boyce methodology of estimation of growth rates finds that mainly in the third and fourth regimes, the growth rates are highest in the subsectoral as well as at the aggregate levels of services GDP. The Banerjee, Lumsdaine and Stock test (J Business Econ Stat 10:271–287, 1992) and the extended Lumsdaine and Pappel test (Rev Econ Stat 79:212–218, 1997) cannot negate the presence of unit root in the data, irrespective of the presence of multiple structural breaks. The paper concludes with the identification of four broad regimes of growth of India’s services GDP and in the subsectors with possible explanations thereof.