Information technology and the Indian economy.

Abstract

Using a multi-sector growth model this thesis attempts to capture the impact of the information technology sector on the growth of the Indian economy. I focus attention on three structural features. Specifically a movement towards services and a decrease in agriculture as a share of GDP, a sharp increase in merchandise and services trade and a sharp increase in GDP growth rate from the period 2004 to 2009. I ask the question whether the four sector Ramsey model can capture these structural features? An important input into the growth model is the econometric work on firm level panel data. Using firm level data obtained from NASSCOM and the CMIE database I carry out production function estimation and create an aggregate productivity index using the Bailey, Hulten and Campbell or the BHC index. The parameters or coefficients of capital, labor and intermediate goods are obtained using the Levinsohn-Petrin method in order to obtain consistent estimates and to account for the endogeneity problem arising from OLS estimates. These coefficients are then used to evaluate aggregate levels of capital and labor used in the IT sector. These aggregate measures of inputs are used in the social accounting matrix constructed using the GTAP database for India for the base year 2001. The growth model is characterized by three final goods agriculture, manufacturing and services and one intermediate good stylized as information technology. Using a social accounting matrix (from GTAP) for the base year 2001 as input into the growth model, we compare the output of the baseline to WDI data. We observe a close fit of GDP and sectoral output for the period 1991 to 2003, however the model output severally undershoots thereafter. We apply a parametric experiment on our original baseline model and observe a similar trend. This shows that information technology accounts for growth trends for the period 1991 to 2003 but cannot account for the sharp acceleration in the GDP growth rate for the period thereafter. The model however does account for the increase in services as a share of GDP, an increase in the contribution of services in the growth of GDP and the sharp increase in exports of IT. Using firm level panel data we simulate shocks to the TFP parameter and observe the above trends repeating for every case.