Malaysian household expenditure on tobacco: evidence from pooled cross-sectional quantile estimations

Purpose This study aims to examine the factors associated with household expenditure share on tobacco at different ranges of the expenditure share among Malaysian households. Design/methodology/approach The analyses were based on pooled cross-sectional data of the Malaysian Household Expenditure Survey. A quantile regression was used to estimate the differentials in the share of monthly household expenditure on tobacco across different socio-economic, demographic and household groups at the 0.1, 0.25, 0.5, 0.75 and 0.9 quantiles of the tobacco expenditure share. Findings Characteristics of household heads (age, gender, ethnicity, marital status, education level, employment status) and household location were significant factors. Households headed by individuals aged 29 years or less spent a bigger share of their expenditure on tobacco than those headed by individuals older than 59 years. Households with heads who were employed and had primary-level education devoted a greater share of their expenditure for tobacco than households with heads who were unemployed and had tertiary-level education. The corresponding differentials were between 0.2% and 2.3%. Practical implications The results identify the target groups based on not only the socio-economic and demographic factors but also the distributional effects of tobacco expenditure share for the design of public policy to reduce the prevalence of smoking-induced illnesses. Originality/value This study represents new attempts to use pooled cross-sectional data and a quantile regression to take into account the heterogeneous tobacco expenditure behaviour according to different levels of tobacco spending among Malaysian households.