Determinants of alcohol purchase and expenditure among Vietnamese households: Evidence from a Heckman selection model
Abstract
Aims: This study examines the determinants of household alcohol purchase decisions and expenditure intensity in Vietnam, addressing the potential sample selection bias arising from the high proportion of non-purchasing households.
Design/Setting/Participants: The analysis uses cross-sectional data from the 2020 Vietnam Household Living Standards Survey (VHLSS) with a sample of 9,389 households. A Heckman two-step selection model is employed to jointly estimate the probability of purchasing alcohol and the level of expenditure conditional on purchase.
Measures: The dependent variables are the probability of household alcohol purchase and the level of alcohol expenditure among purchasing households. Explanatory variables include demographic, socioeconomic, and geographic characteristics such as age, gender, marital status, ethnicity, household size, number of children, presence of elderly members, education, income, employment status, place of residence, and geographic region.
Results: The Wald test of independent equations confirmed significant sample selection bias (), validating the use of the Heckman selection framework. Male-headed households were 7.6 percentage points (pp) more likely to purchase alcohol and spent an additional 325.3 thousand VND annually (conditional on purchase) than female-headed households. Income demonstrated the strongest gradient: the richest quintile was 13.3 pp more likely to purchase alcohol and spent 2,608.0 thousand VND more annually among purchasers compared to the poorest group. Geographically, households in the Mekong River Delta were 16.8 pp less likely to purchase alcohol than those in the Red River Delta, yet purchasers in this region recorded the highest spending intensity nationwide (+1,649.9 thousand VND,), indicating a pattern of high-intensity consumption events.
Conclusion: The results highlight the importance of accounting for demographic and regional heterogeneity. We propose a multi-layered policy framework combining strengthened excise taxation to address income-driven affordability with structural interventions targeting gendered norms. Crucially, region-specific strategies are essential: a “formalization support” model via cooperatives is recommended for the Northern mountainous areas, while strict supply-side restrictions are critical for high-intensity markets in the Southern regions.
In consideration of publishing this article the authors transfer, assign, or otherwise convey all copyright ownership to the International Journal of Alcohol and Drug Research. By this transfer, the article becomes the property of the International Journal of Alcohol and Drug Research and may not be published elsewhere without written permission from the journal.
This transfer of copyright also implies transfer of rights for printed, electronic, microfilm, and facsimile publication. The author(s) will receive no royalty or other monetary compensation for transferring the copyright of the article to the International Journal of Alcohol and Drug Research. IJADR, in turn, grants each author the right to republish the article, without paying royalties to IJADR, in any book of which he or she is the author or editor, subject to the express conditions that (a) the author notify the International Journal of Alcohol and Drug Research in writing of this republication and (b) a credit line attributes the original publication to the International Journal Of Alcohol and Drug Research.
Licence:
Articles are licenced with a Creative Commons License Deed -- you are free to share articles but must give appropriate attribution, may not use for commercial purposes or distribute modified works. See CC/BY-NC/ND/4.0/.
Author Agreement:
As the submitting author, and on behalf of all of the manuscript authors I agree with the terms above relating to the copyright transfer of the manuscript to the International Journal of Alcohol and Drug Research.
