Peer-reviewed article

US state cigarette tax increases and smoke‐free legislation in relation to cigarette expenditure across household socio‐economic circumstances: a quasi‐experimental study

Summer Sherburne Hawkins, Melissa A. Kull, Christopher F. Baum · 21 Nov 2018

Abstract

BACKGROUND AND AIMS: While research has focused on outcomes of tobacco control policies, less is known about the mechanisms by which policies may affect tobacco use. We estimated the associations of changes in cigarette taxes and smoke-free legislation with (1) any household cigarette expenditure and (2) the level of household expenditure on cigarettes, as well as (3) tested interactions with socio-economic circumstances. DESIGN: Difference-in-differences regression models to estimate the associations between changes in US state cigarette taxes and smoke-free legislation with changes in household expenditure on cigarettes. SETTING: Forty US states and District of Columbia. PARTICIPANTS: From annual, cross-sectional surveys (with a longitudinal component) between 2000 and 2014, 128 138 households interviewed quarterly in the Consumer Expenditure Survey. MEASUREMENTS: Dependent measures included any household cigarette expenditure, expenditure in real dollars and budget share of cigarette expenditure. Policy measures included state cigarette taxes and 100% smoke-free legislation. Covariates included respondent age, race/ethnicity, sex; household education; poverty level; family structure; and number of children and adults. FINDINGS: Every $1.00 cigarette tax increase was associated with a 1.5 percentage point (-0.028, -0.002) reduction in any cigarette expenditure and an increase of 0.1% (0.1%, 0.1%) budget share and $10.11 ($8.38, $11.84) absolute expenditure. The association with absolute expenditure was stronger among smoking households above poverty level ($10.73; $8.94, $12.51) than below ($4.72; $2.37, $7.07). The enactment of smoke-free legislation was associated with $2.33 (-$4.56, -$0.10) less expenditure, but not with any expenditure (0.1%; -1.6%, 1.8%) or budget share (-0.1%; -0.1%, 0.1%). The association with absolute expenditure was stronger among households above poverty level (-$2.62; -$4.95, -$0.29) than below (-$0.34; -$4.27, $3.58) CONCLUSION: Cigarette tax increases in the United States between 2000 and 2014 may have reduced smoking prevalence due to an absolute and relative increase in household tobacco expenditure while smoke-free policies appear to have led to a reduction in expenditure. Although tax increases had a stronger impact on absolute expenditure among households above the poverty level, impact on relative expenditure was similar, and consequences for socio-economic inequalities in smoking will vary based on the broader financial situation of households.