Erik Nesson

Estimating the Relationship between Alcohol Policies and Criminal Violence and Victimization

Abstract Violence is one of the leading social problems in both Europe and the United States. The development of appropriate public policies to curtail violence is confounded by the relationship between alcohol and violence. In this article, we estimate the propensity of alcohol control policies to reduce the perpetration and victimization of criminal violence. We measure violence with data on individual-level victimizations from the U.S. National Crime Victimization Survey. We examine the effects of a number of different alcohol control policies in reducing violent crime. These policies include the retail price of beer, drunk driving laws and penalties, keg laws and serving and selling laws. We find some evidence of a negative relationship between alcohol prices and the probability of alcohol- or drug-related assault victimizations. However, we find no strong evidence that other alcohol policies are effective in reducing violent crimes.

Crystal Clear? The Relationship Between Methamphetamine Use and Sexually Transmitted Infections

Public health officials have cited methamphetamine control as a tool with which to decrease HIV and other sexually transmitted infections, based on previous research that finds a strong positive correlation between methamphetamine use and risky sexual behavior. However, the observed correlation may not be causal, as both methamphetamine use and risky sexual behavior could be driven by a third factor, such as a preference for risky behavior. We estimate the effect of methamphetamine use on risky sexual behavior using monthly data on syphilis diagnoses in California and quarterly data on syphilis, gonorrhea, and chlamydia diagnoses across all states. To circumvent possible endogeneity, we use a large exogenous supply shock in the US methamphetamine market that occurred in May 1995 and a later shock stemming from the Methamphetamine Control Act, which went into effect in October 1997. While the supply shocks had large negative effects on methamphetamine use, we find no evidence that they decreased syphilis, gonorrhea, or chlamydia rates. Our results have broad implications for public policies designed to decrease sexually transmitted infection rates. Copyright

The Impact of Tobacco Control Policies on Adolescent Smoking: Comparing Self-Reports and Biomarkers

This paper provides new evidence on how tobacco control policies affect adolescent smoking as measured by self-reported measures and serum cotinine levels, a biomarker of nicotine exposure. I use data from the National Health and Nutrition Examination Surveys covering 1988–1994 and 1999–2012. Higher cigarette excise taxes lead to statistically significant decreases in smoking prevalence as measured by both self-reports and serum cotinine levels. Among self-reported smokers, cigarette excise taxes are associated with statistically significant reductions in serum cotinine levels but not in cigarette consumption. Among likely smokers, as defined by serum cotinine levels, I do not find that cigarette excise taxes reduce serum cotinine levels on the intensive margin, but I do find that tobacco control policies, especially policies directed at minors, may increase the incidence of misreported smoking status.

Interaction Terms in Poisson and Log Linear Regression Models

This paper develops a difference-in-semielasticities (DIS) interpretation for the coefficients of dichotomous variable interaction terms in nonlinear models with exponential conditional mean functions, including but not limited to Poisson, Negative Binomial, and log linear models. We show why these interaction term coefficients cannot be interpreted as a DIS or semielasticity in the same manner as continuous coefficients, which has been overlooked by some empirical researchers. Then we show how interaction terms can be easily transformed into a DIS and derive the asymptotic distribution of this estimator. We illustrate the discrepancy between the interaction term coefficient and the DIS using an empirical example evaluating the relationship between employment, private health insurance and physician office visits. Our results can be applied in treatment effect models when the outcome variable is logged and the dichotomous variables indicating treatment participation and the post-treatment time period.

Looking Down the Barrel of a Loaded Gun: The Effect of Mandatory Handgun Purchase Delays on Homicide and Suicide

We exploit within‐state variation across time in both the existence and length of statutory delays – both explicit wait periods and delays created by licensing requirements – between the purchase and delivery of a firearm to examine the effect of purchase delays on homicides and suicides. We find that the existence of a purchase delay reduces firearm‐related suicides by between 2% and 5% with no statistically significant increase in non‐firearm suicides. Purchase delays are not associated with statistically significant changes in homicide rates.

The relationship between cigarettes and electronic cigarettes: Evidence from household panel data

We use the Nielsen Consumer Panel to investigate the impact of tobacco control policies on purchases of electronic cigarettes (e-cigarettes), cigarettes, and smoking cessation products. We measure product quantity, product type, nicotine content, and liquid volume of e-cigarettes, and product quantity and nicotine content of cigarettes. Higher cigarette excise taxes decrease both cigarette and e-cigarette purchases, suggesting that cigarettes and e-cigarettes are complements, and higher cigarette excise taxes reduce the aggregate amount of nicotine purchased from cigarettes and e-cigarettes. Cigarette smoke-free air laws decrease cigarette purchases, while e-cigarette smoke-free air laws do not affect cigarette or e-cigarette purchases.

INTERACTION TERMS IN POISSON AND LOG LINEAR REGRESSION MODELS: Interaction Terms in Poisson and Log Linear Regression Models

This paper develops a difference‐in‐semielasticities (DIS) interpretation for the coefficients of dichotomous variable interaction terms in nonlinear models with exponential conditional mean functions, including but not limited to Poisson, Negative Binomial, and log linear models. We show why these interaction term coefficients cannot be interpreted as a DIS or semielasticity in the same manner as continuous coefficients, which has been overlooked by some empirical researchers. Then we show how interaction terms can be easily transformed into a DIS and derive the asymptotic distribution of this estimator. We illustrate the discrepancy between the interaction term coefficient and the DIS using an empirical example evaluating the relationship between employment, private health insurance and physician office visits. Our results can be applied in treatment effect models when the outcome variable is logged and the dichotomous variables indicating treatment participation and the post‐treatment time period.

Stock prices and inflation hedged firms

ABSTRACT As the Federal Reserve continues its near-zero rate policy, the threat of inflation remains a concern among both policymakers and businesses. This article uses over 30 years of accounting data and stock returns to examine how publicly traded firms respond to increasing inflation expectations. We first examine whether firms make balance sheet adjustments in response to expected inflation. We then examine whether these activities have a positive effect on stock prices. We find that firms increase inventory, increase capital expenditures and reduce long-term debt when there is an increased expectation of inflation. We then find that firms that increase inventory in this economic regime are rewarded in the market. Markets also reward firms that increase their cash positions and reduce long-term debt possibly suggesting investor flight to safety.

Do Pimples Pay? Acne, Human Capital, and the Labor Market

We use data from the National Longitudinal Study of Adolescent to Adult Health to investigate the association between having acne in middle to high school and subsequent educational and labor market outcomes. We find that having acne is strongly positively associated with overall grade point average in high school, grades in high school English, history, math, and science, and the completion of a college degree. We also find evidence that acne is associated with higher personal labor market earnings for women. We further explore a possible channel through which acne may affect education and earnings.

The Measurement of Health and the Connection between Health Inequality and Income

We examine the extent to which self-reported health measures suffer from reporting bias and then characterize how this reporting bias affects the estimation of income-related health inequality as measured by the concentration index. We run a comprehensive set of tests of reporting bias using several self-reported health measures and several clinical measures of health from the National Health and Nutritional Examination Surveys. Our results confirm the existence of significant, positive, income-related reporting bias and also suggest that higher income individuals react more strongly to a change in objective, clinical health measures. We find that self-reported health measures significantly overstate the degree of income-related health inequality relative to clinical health measures. Parallel to and in support of the analysis described above, we propose the use of a multidimensional measure of clinical health in the context of measuring income-related health inequality.