Consumption and voluntary savings response to service tax

What is it about?

Using unique and proprietary data from a large bank in India, we analyze the effect of a tax policy change on households' consumption and retirement savings behavior. We find that households reduce their consumption by 14% and increase their voluntary retirement savings (Public Provident Fund, aka PPF) by 16% in response to an increase in the service tax policy. Individuals close to their retirement age (55 years of age and above) and without any withdrawal restrictions from their PPF account tend to reduce their expenditures more and save more. Individuals with financial constraints and withdrawal restrictions do not reduce their expenditure significantly. Households’ saving decisions (especially tax-exempted instruments) and the age of the household (Agarwal, Pan and Qian(2020)) have significant policy implications and an important research area in public finance. Most of the existing empirical studies are limited to investments in retirement savings or tax-exempted retirement savings. The behavior of retirees when there is an increase in consumption or service tax is an unexplored area. An increase in service tax increases the amount spent on present consumption, while people close to their retirement expect additional future consumptions post-retirement due to life-changing events. Therefore, households may increase their retirement savings in tax-exempted schemes by reducing current expenditures after a change in the service tax. Therefore, the saving behavior of households close to retirement is an important area of research. This paper uses a novel and detailed dataset of household debit card spending and a voluntary savings program aimed at retirement saving to jointly test the impact of service (consumption) tax policy on consumption and retirement saving. We use a policy change as announced in the Indian annual budget as a natural experiment to analyze the causal impact. We show that households save more and spend less once they attain 55 years of age or close to their retirement age when there is an increase in the service tax. We also document that households do not cut down their consumption when they are further away from their retirement in response to the service tax increase. Our results are robust to various alternative specifications and heterogeneity tests.

Featured Image