Impact of trade liberalization on import demands in India: a panel data analysis for commodity groups

Abstract

The objectives of this study are to examine the impact of liberalization on price elasticities of imports at a commodity level for broad commodity groups and at an aggregate level for the Indian economy over the period 1993–1997. Liberalization is implemented by reducing tariff and non-tariff barriers and relaxing exchange controls. The impact of tariffs and market based exchange rate is introduced in the import price of the commodity. Relative price (import/domestic) elasticities for each commodity group using panel data of six countries over five years are estimated. For the Indian economy, relative price elasticities for each of the years at an aggregate level using data for twenty commodity groups as observations are obtained and these show that the price elasticity tends to rise with the increase in liberalization when tariff rates are reduced.

Acknowledgements

We are grateful to Mr David Bailey, Dr Kunal Sen and Professor Mark Taylor for their helpful comments. The second author wishes to thank the Leverhulme Trust for the award of an Emeritus Fellowship which facilitated the completion of the above study.

Notes

 We chose Japan as we have no gaps in observations and Japan's data are more reliable than European Union (EU) because exchange rate fluctuations are not the same for all EU countries and the extent of non-tariff barriers vary from country to country in the EU.

 In the case of strong separability, the utility functions are additive for each group and preferences are strongly or additively separable. If any subset of commodities appears only in a separable utility function, then quantities purchased within the group can always be obtained as a function of group expenditure and prices within the group alone.

 There are several advantages in the estimation of price elasticities of import using cross-country data: (a) it will take into consideration lower level of tariffs in our sample periods adopted by other countries other than India; (b) it will enable us to capture characteristics of recent vintage that is post-Uruguay round of trade negotiations and (c) it will provide a reasonable number of observations for the estimation procedure.

 The HS system was developed from Customs Cooperation Council Nomenclature, also known as the Brussels Trade Nomnclature system. The HS was implemented in 1986 for customs tariff and in 1987 for imports.

 It is possible to use the Flexible Generalized Least Squares provided the estimated covariance matrix is consistent.

 Data on ntms were collected from India's import policy (Government of India, 1992–1997) while for other countries they came from UNCTAD. UNCTAD definition was restricted to hard core non-tariff barriers only while India's definition included all types of ntms that is soft, medium and hard-core. As the above data series use the UNCTAD definition and UNCTAD source for India, ntms are zero for all commodities until 1996. No statistics was available for 1997. Given this limitation of data, we cannot estimate the impact of removal or relaxation of non-tariff barriers with the available information.

 This is expected given that a large amount of imports by developed countries is subject to multi-fibre arrangement quotas. Panagariya et al. (2001) estimated import demand elasticity for textiles and apparel and found that the import demand elasticity facing Bangladesh was about 26 which was considerably higher than many studies. They used disaggregate data and also the fact that exports under multi-fibre agreement are subject to binding quotas.