Special‐interest Spillovers and Tobacco Taxation

Abstract

This study explores the effects tobacco special interest has had on cigarette tax rates. Using a spatial econometric model to control for tax competition and the endogeneity present in state tax rates, this study finds that tobacco special interest has played a significant role in influencing state cigarette tax rates. Pounds of tobacco grown in a state and tobacco industry campaign contributions are associated with lower cigarette tax rates. The empirical results show that the magnitude of the special-interest effect can be two to five times as large when the spatial spillover effects are included. (JEL C23, H71, H73) ABBREVIATIONS LM: Lagrange Multiplier LR: Likelihood Ratio MCMC: Markov Chain Monte Carlo OLS: Ordinary Least Squares SAR: Spatial Autoregressive SDM: Spatial Durbin Model SEM: Spatial Error Model SLX: Spatial Lag of X 146

State cigarette taxes and health expenditures: Evidence from dynamic spatial lag panel models

“…An important conclusion emerging from this literature is the spatial endogeneity of cigarette tax rates. Hoffer () explores the effect of tobacco special‐interest groups on cigarette tax rates using spatial econometric methods to control for tax competition and endogeneity in state tax rates and finds that cigarette tax rates are highly persistent and spatially interdependent and finds evidence of strong special‐interest effects. Hoffer and Lacombe () emphasize the importance of spatial dependence in cigarette tax rates for non‐health related reasons, and reinforce the idea that empirical analysis of the relationship between cigarette taxes and state health spending should account for spatial dependence in cigarette taxes.…”

“…Pounds of tobacco grown serves as a credible instrument candidate. Hoffer () describes how state cigarette tax policy cannot reverse cause agriculture yields but instead reflects the concentration of special interests in a state that lobby effectively for special treatment for the tobacco industry. This should be uncorrelated with unobservable factors affecting health and hospital spending.…”

“…The first stage regression model contained two instruments: pounds of tobacco produced in each state in each year and the Gini coefficient on household income in each state in each year. Cigarette taxes vary with state tobacco production because of the lobbying efforts by special interests in states with high tobacco production (Hoffer, ). The strength of the tobacco lobby in a state should be uncorrelated with unobservable factors affecting state health and hospital spending.…”

“…For health expenditures, most studies focus specifically on the expected reduction in health care expenditures attributable to smoking-related illnesses and conditions (Lightwood, Dinno, & Glantz 2008;Lightwood & Glantz, 2011, 2016Maciosek, Xu, Butani, & Pechacek, 2015). Medical expenses related to smoking, such as lung disease, stroke, and cancer, are expensive.…”

“…A related strand of literature focuses on the determinants of cigarette tax rates, including the influence of political factors and special-interest groups (Devereux et al 2007;Gallet, Hoover, & Lee, 2009;Golden, Ribisl, & Perreira, 2014;Hoffer, 2016). An important conclusion emerging from this literature is the spatial endogeneity of cigarette tax rates.…”