Abstract: This paper proposes a Bayesian factor-augmented bundle choice model to estimate joint consumption as well as the substitutability and complementarity of multiple goods in the presence of endogenous regressors. The model extends the two primary treatments of endogeneity in existing bundle choice models: (1) endogenous market-level prices and (2) time-invariant unobserved individual heterogeneity. A Bayesian sparse factor approach is employed to capture high-dimensional error correlations that induce taste correlation and endogeneity. Time-varying factor loadings allow for more general individual-level and time-varying heterogeneity and endogeneity, while the sparsity induced by the shrinkage prior on loadings balances flexibility with parsimony. Applied to a soda tax in the context of complementarities, the new approach captures broader effects of the tax that were previously overlooked. Results suggest that a soda tax could yield additional health benefits by marginally decreasing the consumption of salty snacks along with sugary drinks, extending the health benefits beyond the reduction in sugar consumption alone.
Joint with Michelle Sovinsky, Liana Jacobi, and Alessandra Allocca
Abstract: As illicit substances move into the legal product space, substitution patterns with legal products become more salient. In particular, marijuana legalization may have implications for the use of other legal “sin” goods. We estimate a structural model of multi-product use of illegal and legal substances considering joint use, limited access to illicit products, and persistence in use. We focus on a young person’s choice to consume marijuana, alcohol or cigarettes (and possible combinations), and we find that sin goods are complements. Furthermore, our findings emphasize the necessity of accounting for joint consumption and access to obtain correct price sensitivity estimates. Post-legalization, youth marijuana use would increase from 25% to 37%. However, counterfactual results show that a combination of (reasonable) tax increases on all goods along with enforcement against illegal use can potentially revert use to pre-legalization levels. The earlier the tax increases are implemented the more effective they are at curbing future use. Our results inform the policy debate regarding the impact of marijuana legalization on the long-term use of sin goods.
Joint with Dong-Hyuk Kim and Yong Song
Abstract: This paper introduces a new flexible specification for random coefficients discrete choice model, where the mean utility and instrumental variable (IV) first-stage equations are jointly modeled by a sparse finite mixture of normals, while observed and unobserved individual-level heterogeneity are, as in standard random coefficients multinomial (RCMNP) models, captured by observed consumer characteristics and random coefficients. Our flexible specification have three advantages. First, it relaxes the linear functional form assumption on the mean utility and IV first-stage equation, and provides a local approximation of nonlinear relationships. Second, it relaxes the Gaussian distributional assumption on the joint error of the mean utility (unobserved product characteristics) and first stage equation, which is commonly assumed in Bayesian demand estimation. Third, by limiting the mixture to the mean utility, our specification offers flexibility with parsimony. We provide four simulated examples to illustrate the specification is capable of capturing nonlinear relationships. We apply the new method to household milk demand data to illustrate its feasibility and applicability.
Joint with Liana Jacobi and Michelle Sovinsky
Abstract: Identification and estimation of complementarities across substances is essential to understand and assess intended and unintended consequences of (proposed) changes in substance market regulations, including impacts on use, health, and tax revenues. This paper proposes a multi-product choice model for the joint use of illicit substances when an individual’s choice set is restricted by the illegality of a substance and access to an illicit substance might be correlated with use decisions. We apply our model to data from the National Survey on Drug Use and Health, which contains information on substance use for US individuals aged 12 or older. We find that access into recreational marijuana is not random, which is consistent with earlier work. Furthermore, our preliminary results suggest that recreational and medical marijuana are independent in demand for consumers. Importantly, we also find that researchers would incorrectly conclude the products are complements in use if selection into access is not controlled for. Our results, therefore, shed light on why the literature has found conflicting results regarding the substitutable nature of the products.
Working paper coming soon.