Nathan Tefft

Taxing Soft Drinks And Restricting Access To Vending Machines To Curb Child Obesity

One of the largest drivers of the current obesity epidemic is thought to be excessive consumption of sugar-sweetened beverages. Some have proposed vending machine restrictions and taxing soft drinks to curb childrens consumption of soft drinks; to a large extent, these policies have not been evaluated empirically. We examine these policies using two nationally representative data sets and find no evidence that, as currently practiced, either is effective at reducing childrens weight. We conclude by outlining changes that may increase their effectiveness, such as implementing comprehensive restrictions on access to soft drinks in schools and imposing higher tax rates than are currently in place in many jurisdictions.

Mental Health and Employment: The SAD Story

This paper explores the relationship between health-related quality of life (HRQOL) measures and employment status in light of a constructed index related to Seasonal Affective Disorder that depends only on latitude and day of year. In models including demographic covariates and indicators for state, year, and quarter, more hours of darkness is associated with poorer HRQOL, which in turn is associated with a lower likelihood of employment. The relationships between the darkness index and HRQOL measures are stronger overall for women than for men. Inclusion of both the darkness index and the HRQOL measures in models of employment status determinants provides some evidence that the former operates through the latter in predicting a lower likelihood of employment. When specifying the darkness index as an instrument for HRQOL, each additional day of poor mental health per month leads to a 0.76 percentage point increase in the probability of unemployment among women.

Decomposing the Relationship between Macroeconomic Conditions and Fatal Car Crashes during the Great Recession: Alcohol- and Non-Alcohol-Related Accidents

Abstract This paper investigates to what extent and in what ways conditions related to the 2007-2008 recession reduced fatal crashes. It hypothesizes that the reduction in fatal automobile accidents operates through both the quantity of driving and changes in behaviors associated with driving. Using state-by-quarter fixed effects models, the study shows that unemployment rate increases significantly reduce fatal accidents. Decomposing the fatal accident rate into accidents per mile traveled and miles traveled per capita reveals that higher unemployment is significantly associated with fewer accidents per mile, and also reveals that fatal accidents associated with alcohol are more responsive to unemployment rate changes than are accidents overall. These results suggest that the recession’s “lost” fatal accidents occurred in areas hit harder by the recession and were in the form of fewer alcohol-related accidents per mile traveled rather than fewer miles traveled overall.

Non-Linear Effects of Soda Taxes on Consumption and Weight Outcomes

The potential health impacts of imposing large taxes on soda to improve population health have been of interest for over a decade. As estimates of the effects of existing soda taxes with low rates suggest little health improvements, recent proposals suggest that large taxes may be effective in reducing weight because of non-linear consumption responses or threshold effects. This paper tests this hypothesis in two ways. First, we estimate non-linear effects of taxes using the range of current rates. Second, we leverage the sudden, relatively large soda tax increase in two states during the early 1990s combined with new synthetic control methods useful for comparative case studies. Our findings suggest virtually no evidence of non-linear or threshold effects.

Fast food prices, obesity, and the minimum wage.

Recent proposals argue that a fast food tax may be an effective policy lever for reducing population weight. Although there is growing evidence for a negative association between fast food prices and weight among adolescents, less is known about adults. That any measured relationship to date is causal is unclear because there has been no attempt to separate variation in prices on the demand side from that on the supply side. We argue that the minimum wage is an exogenous source of variation in fast food prices, conditional on income and employment. In two-stage least-squares analyses, we find little evidence that fast food price changes affect adult BMI or obesity prevalence. Results are robust to including controls for area and time fixed effects, area time trends, demographic characteristics, substitute prices, numbers of establishments and employment in related industries, and other potentially related factors.

The Dow is Killing Me: Risky Health Behaviors and the Stock Market.

We investigate how risky health behaviors and self-reported health vary with the Dow Jones Industrial Average (DJIA) and during stock market crashes. Because stock market indices are leading indicators of economic performance, this research contributes to our understanding of the macroeconomic determinants of health. Existing studies typically rely on the unemployment rate to proxy for economic performance, but this measure captures only one of many channels through which the economic environment may influence individual health decisions. We find that large, negative monthly DJIA returns, decreases in the level of the DJIA, and stock market crashes are widely associated with worsening self-reported mental health and more cigarette smoking, binge drinking, and fatal car accidents involving alcohol. These results are consistent with predictions from rational addiction models and have implications for research on the association between consumption and stock prices.

Substitution Patterns Can Limit the Effects of Sugar- Sweetened Beverage Taxes on Obesity

Dramatic increases in obesity and sugar-sweetened beverage consumption over the past several decades have become major public health and clinical concerns. Obesity rates tripled in 30 years, and sugar-sweetened beverage consumption among children more than doubled in the last 2 decades of the twentieth century (1). Many children drink more sugar-sweetened beverages than milk, and sugar-sweetened beverages represent the largest category of daily caloric intake (7%–12%) for many demographic groups (1). Emerging evidence suggests that increasing consumption of sugar-sweetened beverages raises weight and obesity rates. These trends and the evidence that modest but persistent reductions in calories (approximately 1 can of sugar-sweetened beverage per day) could halt the obesity epidemic for 90% of the population or more (2) has focused attention toward enacting policies to curb consumption of sugar-sweetened beverages, especially in children. Most prominent among these policies has been the push to increase the prices of sugar-sweetened beverages through increases in state and local taxes on these items. A recent Institute of Medicine (IOM) report suggests that governments implement a sizeable tax to reduce the overconsumption of sugar-sweetened beverages (3). Originally proposed more than a decade ago (4), the idea follows basic economic reasoning and the compelling success story of the reduction in tobacco use from tobacco taxation. The economic intuition is that higher prices discourage use, so enacting a “sin tax” on carbonated soft drinks (or sugar-sweetened beverages in general) could lower consumption; subsequently, the effect of decreased sugar-sweetened beverage consumption on obesity rates could also have an effect on the high social costs of obesity. Although U.S. states have taxed carbonated soft drinks for nearly 100 years as a means of raising revenue, only recently has this policy been evaluated for its potential effect on reducing obesity rates. Evidence on this issue is available primarily in 2 varieties. A host of studies have used data on household consumption and grocery store prices to show that households that encounter higher prices reduce carbonated soft drink purchases (5). These estimated price responses are often used by other researchers to calculate the reductions in calories attributable to lower carbonated soft drink consumption and to then calculate the implied reduction in weight from these reductions in calories (6). A typical estimate is that a 10% price increase on carbonated soft drinks would reduce consumption by approximately 10%, leading to a 20 kcal per day reduction (7). These studies do not fully address several complex behavioral and biological responses, however. These responses include the possibility that consumers would substitute other caloric beverages (eg, orange juice, chocolate milk) or foods for sugar-sweetened beverages if only the latter are taxed. Another possibility is that manufacturers would respond to higher taxes by lowering prices (“strategic pricing”). A third possibility is a dynamic response of weight to a change in the rate of sugar-sweetened beverage consumption. A second variety of studies addresses these issues by directly linking existing state-level carbonated soft drink tax rates to information about both daily beverage consumption and measured weight to estimate actual (as opposed to hypothetical) tax effects. These studies have found little or no evidence that people faced with higher carbonated soft drink taxes have lower weight (8). How could these 2 sets of studies come to such different conclusions on the effects of carbonated soft drink taxes on weight? A key distinction between these similar approaches is in how substitution patterns are incorporated — that is, if consumers drink fewer carbonated soft drinks, what, if anything, do they drink instead? Studies of the first variety typically assume that consumers respond only partially to the reduction in carbonated soft drink calories by increasing intake of other caloric beverages such as juice or whole milk. For example, some researchers (6) assume only one-third of each calorie reduction in carbonated soft drinks from taxation is replaced by substitution of other high calorie beverages. Rather than make these assumptions, studies of the second variety estimate actual substitution patterns and have shown that consumers fully offset all calorie reductions in carbonated soft drinks from taxation by drinking other high-calorie beverages. These studies find no reduction in population weight from carbonated soft drink taxes. On the basis of substantial increases in both obesity and sugar-sweetened beverage consumption in the past decades and the links between them, the effectiveness of sugar-sweetened beverage taxation in reducing rates of obesity is essential to consider in any discussion of public health measures available to reduce obesity rates. Evidence suggests caution in enacting sugar-sweetened beverage taxation legislation with a core purpose of obesity reduction. “Big taxes” may have different effects than the small taxes currently in place, but these big taxes would need to dramatically shift the substitution patterns we see when small taxes are implemented. Big taxes would need to effectively shift people toward water and other low-calorie drinks in a way that small taxes do not; taxes alone would not likely cause such a shift. However, what if sugar-sweetened beverage taxes are ineffective in reducing obesity? Does this mean that we should not consider implementing them as public health measures? Not necessarily — the broader benefits of tax increases on sugar-sweetened beverages must be considered. Most studies suggest that such taxes will likely reduce consumption of sugar-sweetened beverages, and researchers (8) find that the reduction in calories from soft drinks is equivalent to the increase in calories from whole milk, which may be a good public health outcome, irrespective of any obesity reduction. Associated high levels of added sugar may also lead to other poor health outcomes, such as diabetes, cardiovascular risk, and poor dental health. Sugar-sweetened beverage taxes could successfully combat these poor health outcomes even if taxation does not lower obesity rates. In addition to encouraging other potential health benefits, sugar-sweetened beverage taxes may be helpful in reducing obesity rates if they are used as one aspect of a larger, more comprehensive policy approach that aims to redirect consumers away from caloric sweeteners and toward more healthful alternatives such as water or food without added sweeteners. For example, a tax combined with a subsidy for water could be more effective than a tax in isolation. Additionally, a recently proposed extension of a sugar-sweetened beverage tax that instead taxes all caloric sweeteners at the manufacturer level would limit consumers’ ability to easily switch to foods with caloric sweeteners or other unhealthful beverages (9). A key feature of this broader approach is that it limits the possibility for substitution to unhealthful alternatives, ensuring that these policies effectively achieve the goal of reducing and stemming the rise in obesity.

Alcohol-impaired motor vehicle crash risk and the location of alcohol purchase

Motor vehicle crashes involving alcohol impairment are among the leading causes of mortality and morbidity in the U.S. In this study, we examine how the probability of driving after a binge-drinking episode varies with the location of consumption and type of alcohol consumed. We also investigate the relationship between the location of alcohol purchase and the number of alcohol-impaired fatal motor vehicle crashes. Using multiple datasets that are representative of the U.S. between 2003 and 2009, we find that binge-drinkers are significantly more likely to drive after consuming alcohol at establishments that sell alcohol for on-premises consumption, e.g., from bars or restaurants, particularly after drinking beer. Further, per capita sales of alcohol for off-premises consumption are unrelated to the rate of alcohol-impaired fatal motor vehicle crashes. When disaggregating alcohol types, per capita sales of beer for off-premises consumption are negatively associated with the rate of alcohol-impaired fatal motor vehicle crashes. In contrast, total per capita sales of alcohol from all establishments (on- and off-premises) are positively related to the rate of alcohol-impaired fatal motor vehicle crashes and the magnitude of this relationship is strongest for beer sales. Thus, policies that shift consumption away from bars and restaurants could lead to a decline in the number of motor vehicle crashes.

Education, maternal smoking, and the earned income tax credit

Abstract We estimate and explore mechanisms of the impact of the Earned Income Tax Credit (EITC) expansions on the smoking behavior of women. Differential increases in federal EITC benefits by family size in the mid-1990s allow for a comparison of smoking status changes between mothers with one and more than one child. We exploit these changes in a difference-in-differences framework using data from the 1993-2001 waves of the Behavioral Risk Factor Surveillance System (BRFSS) and show that the increase in EITC benefits yielded a significant decline in the likelihood of being a current smoker among unmarried mothers with less than a college degree. Although women with a high school degree or less and women with some college education received similar benefit increases on average and exhibited similar labor supply responses, the reduction in the likelihood of smoking was concentrated among those with some college.

Financial Conflicts of Interest in Medicine

We explore financial conflicts of interest faced by doctors. Pharmaceutical firms frequently pay physicians in the form of meals, travel, and speaking fees. Over half of the 334,000 physicians in our sample receive payment of some kind. When a doctor is paid, we find that he is more likely to prescribe a drug of the paying firm, both relative to close substitutes and even generic versions of the same drug. This payment-for-prescription effect scales with transfer size, although doctors receiving only small and/or infrequent payments are also affected. The pattern holds in nearly every U.S. state, but it is strongly and positively related to regional measures of corruption. * We have benefited from discussions with Anirban Basu, Doug Conrad, Gordon Dahl, Raymond Fisman, Dan Hamermesh, Ryan Hansen, Jon Karpoff, Ian Larkin, Sheridan Titman and seminar participants at the University of Washington. ‡ Contact: Joseph Engelberg, Rady School of Management, University of California at San Diego, (Email) jengelberg@ucsd.edu (Tel) (1) 858-822-7912; Christopher A. Parsons, Rady School of Management, University of California at San Diego, (Email) caparsons@ucsd.edu (Tel) (1) 858-534-8782; Nathan Tefft, School of Public Health, University of Washington, (Email) tefft@uw.edu (Tel) (1) 206-221-5897.We use the geographic distance between a doctor’s office and drug company headquarters to instrument for the likelihood of pecuniary transfers, such as meals or speaking fees. Doctors tilt prescriptions in favor of the paying firm’s drugs, shifting away from both branded and generic substitutes. Larger transfers cause larger shifts in prescriptions. We explore two potential explanations: 1) information flow (or its perception), and 2) rent seeking. Payments increase prescriptions of branded drugs over generic equivalents, situations where information cannot play a large role. However, doctors residing in states known to be corrupt in other ways (e.g., electoral fraud) are much more sensitive to payments from the drug industry, as are male doctors.