Journal of Income and Wealth

  • Year: 2024
  • Volume: 46
  • Issue: 1and2

Estimation of equivalence scales using matching estimator

  • Author:
  • Prafulla Chandra Mishra1, Shrinivas Vijay Shirke2,*
  • Total Page Count: 35
  • DOI:
  • Page Number: 78 to 112

1(Retd. Additional Director General, National Accounts Division, Ministry of Statistics & Programme Implementation), New Delhi, India

2Director, National Accounts Division, Ministry of Statistics & Programme Implementation, New Delhi, India

Abstract

The Wellbeing and Sustainability Task Team (WSTT) of the 2008 SNA update, in its guidance note on “Distribution of household income, consumption and wealth”, has elaborated on the need to focus on equivalized results for the household sector, in order to account for the differences in household size and composition. The equivalized results are arrived at using equivalence scales, which are an effective tool in estimating the number of consumption units in each household. Equivalence scales help in arriving at comparable results across households for per capita income or consumption expenditure, taking into consideration the demographic differences among them. Though the OECD Equivalence Scale, OECD Modified Equivalence Scale, and Square Root Scale are popular due to ease of implementation, the fact that consumption patterns differ widely across countries cannot be ignored. While there are several empirical methods available for the estimation of equivalence scales, this paper is an attempt to construct an equivalence scale based on expenditure data using the matching estimator of household equivalence scale proposed by Szulc (2009). This approach has the advantage of being less computationally intensive, in terms of the estimation of complete demand systems required in other approaches. Since the equivalence scales are sensitive to the choice of matching method used, we explore different matching methods along with their application to Indian consumer expenditure data (NSS 68th round).

Keywords

Equivalence scales, Distributional accounts, Well-being, Matching estimator