Mahdiyeh Entezarkheir

Obesity, smoking, and cigarette taxes: Evidence from the Canadian Community Health Surveys

OBJECTIVES Recent studies suggest an ambiguous relationship between obesity and cigarette taxes. We employ Canadian data to evaluate the effects of cigarette taxes on smoking and obesity. METHODS We use a simple reduced form approach and exploit the significant cross-province differences that exist between Eastern and Western Canada to estimate the effects of higher cigarette taxes using aggregate health region and individual level data from the 2003 and 2005 waves of the Canadian Community Health Surveys (CCHS). RESULTS OLS estimates based on health regions data suggest that a 10% increase in cigarette taxes is significantly correlated with a 4-5% increase in the percentage of obese population. We also find cigarette tax elasticities of between -0.2 and -0.4 with respect to the percentage of smokers. Estimates from individual level data are similar. CONCLUSIONS In tandem, these results offer support to the possibility that health benefits from higher cigarette taxes and lower smoking, might be partially offset by a corresponding increase in obesity levels.

Mergers and innovation: evidence from a panel of US firms*

ABSTRACT Mergers lead to larger firms and a less competitive market structure, but their effects on innovation are not clear. Mergers may improve innovation incentives by promoting economies of scope and scale, R&D activities, and increasing the ability to deal with uncertainties. However, mergers may also discourage innovation by reducing competition, increasing costs, and decreasing production and R&D efficiencies. In this study, we investigate merger impacts on innovation using a panel data consisting of four different data sets on publicly traded US manufacturing firms from 1980 to 2003. Our proxy for innovation is based on citation-weighted patent stocks. In our estimation model, we control for endogeneity using instrumental variables and factors such as market share, size, industry, and time. We find that mergers are positively and significantly correlated with firms’ innovation. Our findings also indicate that merger effect on innovation is heterogeneous across industries, increases with market share, and is greater in the long run. Our findings are robust to alternative measures of innovation.

Is Innovation a Factor in Merger Decisions? Evidence from a Panel of U.S. Firms

The impact of innovation on mergers has been a subject of debate in merger enforcements. Firms may decide to merge because of increasing market share and expanding capacity. However, mergers may also be motivated by innovation since they provide resources for commercialization of innovation and allow for capturing knowledge spillovers. There are myriad studies on the drivers of mergers, but empirical evidence on innovation-induced changes of merger likelihood is limited. In this paper, we construct a panel data set of mergers among publicly traded US manufacturing firms from 1980 to 2003 and investigate the impact of innovation on merger decisions controlling for business cycles and proxies of neoclassical, behavioural and Q-theories of mergers. Our measure of innovation is based on the citation-weighted patent stock. We find that innovations are positively and significantly correlated with firms’ merger likelihood. Our results on the control variables, such as business cycles, Tobin’s q, and market shares, are also consistent with those reported in the literature. We also find that the magnitude of the merger impact of innovation varies across industries. Our main result is robust to alternative measures of innovation and different estimation methods.

Missing Observations on a Variable: When Do the Listwise Deletion and Indicator Approaches Work?

Econometric analyses widely use the listwise deletion approach - which is also called the complete case method - and the indicator approach for models with censored regressors and regressors with missing observations. Nevertheless, the existing literature provides few studies which examine the statistical properties of these methods and they suggest the listwise deletion method generates smaller bias than the indicator method. Using the Monte Carlo simulations this study shows sometimes the listwise deletion estimates have larger bias than the indicator estimates when missingness on a regressor is correlated with unobserved error terms and the value of the regressor. These conditions indicate that greater care is warranted in interpreting the estimates under each approach than the existing literature and the applications which generally employ these methods.