Demand System and Liquidity Constraints: Simple Methodology for Measuring Liquidity Constraint

This paper proposed a new methodology, which introduces a stochastic property in the field of demand analysis, for measuring the skewness of the consumption bundle distribution. We used the questionnaire data from the National Survey of Family Income and Expenditure, Japan, from 1989 to 2004. We regard two categories of variables such as the total expenditure of each household and the expenditure share of 10 commodity groups (e.g., food expenditure share, medical expenditure share, etc.) as random variables, and estimated the probability density function using the "Skew Normal Distribution." As a result, distortion in the same direction was detected in all of the 10 commodity groups. We also calculated that the measured skewness mutually relates to the major proxy of the liquidity constraint. As a proxy of households under liquidity constraint, here we employ the measurement of the fraction of Hand-to-Mouth (HtM) households. This paper was motivated by the need to provide a simple methodology for measuring the liquidity constraint even in countries with poor statistical literacy. Although several proxies have been proposed to measure the liquidity constraint, the methodology becomes complex for more rigorous measuring attempts. In contrast, our methodology only requires micro data on total consumption and a separate total for each commodity group.