A Spatial cost of living index for Colombia using a microeconomic approach and censored data

This paper describes a methodology to calculate a spatial cost of living index using Colombian data for 2006 that takes into consideration the microeconomic behavior of households. Using the Almost Ideal Demand System and recovering the expenditure functions for the 23 main Colombian cities, the index proposed is compared to the traditional methodologies used to calculate the regional basket of goods in the country and to an alternative methodology proposed by Romero (2005). This comparison suggests that when the substitution effects are not considered, and the same basket of goods is evaluated in every city, the index is biased, and this bias increases when the difference between cities increases. For reducing the bias, we use a microeconomic approach that keeps the households' level of utility constant and allows substitution among different baskets of goods. According to our calculations, Bogota is still the most expensive city in the country followed by Armenia, Cali, Bucaramanga and Ibague.