Abstract
This article describes a methodology to calculate a Spatial Cost of Living (SCOL) index using the Colombian data for 2006 that considers the microeconomic behaviour of households. Estimating an Almost Ideal Demand System (AIDS) to recover the expenditure function of the 23 main Colombian cities, the index is compared to the fixed basket approach and to an alternative methodology proposed by Romero (Citation2005). This comparison suggests that the index is biased when the substitution effects are not considered, and this bias increases when the difference in expenditure levels between cities increases. According to our calculations, Bogota is still the most expensive city in the country.
Acknowledgement
The authors acknowledge the Nucleus of the Scientific Millennium Initiative Regional Science and Public Policy. The authors acknowledge and appreciate the wise comments of two anonymous referees and the financial support of Chilean Fondecyt no. 1111061 ‘Evaluación Económica de la Política Regional en Chile’ and the Nucleus of the Scientific Millennium Initiative ‘Regional Science and Public Policy’.
Notes
1 The demographic variables included are the socioeconomic status of households, number of persons in the households and education of the household's head.
2 The complete code and results for the AIDS are available upon author request.
3 Generally, the median consumer is used in the demand system estimation literature.