Understanding the determinants of excessive alcohol consumption, especially binge drinking, is important for informed alcohol policy and evaluation (Gilmore et al., 2013; Xu and Chaloupka, 2011). High-intensity drinkers who consume several drinks within a short time- period increase risks of serious health, safety and social problems for themselves and others (Anderson, 2008; USDOJ, 2004). For the United States, binge drinking accounts for more than half of an estimated 80,000 annual deaths and three-quarters of $224 billion in economic costs resulting from excessive alcohol consumption (Bouchery et al., 2011; CDC, 2012). Binge drinking is strongly associated with alcohol-impaired driving, and alcohol-related fatalities are 20% of underage fatal accidents (Naimi et al., 2003; NHTSA, 2012). Binge drinking patterns vary importantly by age group and gender. In the US, prevalence (28.2%) and intensity (9.3 drinks per episode) are highest among young adults aged 18-24 years (CDC, 2012), and then decline with age. However, frequency (5.5 episodes per month) is highest among older adult bingers. Binge prevalence among men (23.2%) is more than twice the rate for women (11.4%), and intensity and frequency also are much higher for men. For persons under 18 years, binge drinking is a special concern since excessive use of alcohol and intoxication by youth are closely associated with similar problems in adult populations (Green and Ross, 2010; Nelson et al., 2005; USDHHS, 2012). Approximately 22% of US high school seniors engaged in binge drinking in 2011 (Johnston et al., 2012). Similar drinking patterns and costs are reported for other countries (Plant et al., 2010). For example, Anderson (2008) reports a binge prevalence of 28% for the European Union (EU), with frequency highest among persons aged 15-24 years. One in six (18%) EU youth report bingeing three or more times in the last month, and one in eight (13%) have been intoxicated more than 20 times in their life. Binge drinking is more common in northern European countries, but high prevalence rates also are reported for some southern countries such as Spain (Ahlstrom and Osterberg, 2004/2005; Anderson, 2008; Soler-Vila et al., 2013). Overall, binge drinking in the United States is estimated to account for 90% of alcohol consumed by youth and young adults and 50% of alcohol consumed by adults (USDOJ, 2004). Country-level data for Europe indicate that high-intensity drinking is more strongly associated with alcohol-related problems compared to high per capita consumption (Bloomfield et al., 2003).
During the past several decades, economists have devoted considerable research to alcohol prices and taxes as determinates of drinking and drinking patterns, including binge drinking. Economic studies that incorporate prices or taxes fall into three general categories: first, population-level econometric studies for average per capita consumption based on aggregate data that include drinkers and non-drinkers alike, regardless of age, gender, or drinking patterns. A majority of studies contained in several recent meta-analyses fall into this category (Fogarty, 2009; Gallet, 2007; Nelson, 2013a; Wagenaar et al., 2009). Second, individual-level studies of alcohol use (participation, number of drinks per month), which do not include specific measures of heavy or binge drinking. Some price studies based on survey data, such as the Harvard College Alcohol Study, are informative for determinates of drinking generally but not necessarily for bingeing or other heavy drinking activity (An and Strum, 2011; Farrell et al., 2003; Picone et al., 2004; Williams, 2005). Third, individual-level studies of binge drinking that incorporate alcohol price or tax variables. These studies provide a stronger evidence base for effective alcohol policies that address abusive and high-intensity drinking. Most price-binge studies reviewed below use individual survey data for the United States, but coverage for other countries can be expanded by including available natural experiments and field studies.
Despite its importance, I found no previous reviews that address in a comprehensive manner the effects of alcohol prices and taxes on binge drinking by age group. Previous reviews either cover only a few early studies for youth (Chaloupka et al., 2002; Grossman et al., 1994; de Walque, 2014); omit prices and taxes as evidence (Ham and Hope, 2003; Kuntsche et al., 2004); or review relatively few economic studies for binge drinking. For example, a meta-analysis by Wagenaar and colleagues (2009) includes only 10 individual-level studies for heavy drinking, while Elder and colleagues (2010) cover 10 studies for excessive drinking, including two natural experiments. A systematic review by Patra and colleagues (2012) focuses on alcohol-related harms, but binge drinking studies are limited to only three economic studies and several natural experiments. Results by age or gender also are not reported in past reviews or apply mostly to early studies. In contrast, the present review examines 56 economic studies for binge drinking divided equally among three age groups. Results by gender are reported. Eleven natural experiments and field studies also are reviewed. As discussed below, discrepancies in prior reviews arise in part due to different methods required to search the economics literature on alcohol use. Further, several widely-cited studies have attempted to draw a general policy link between alcohol prices (or taxes) and excessive alcohol consumption, but evidence cited is mostly drawn from aggregate econometric studies (Anderson et al., 2009; Babor et al., 2010; Edwards et al., 1994; Nelson et al., 2013). This is incomplete and potentially misleading, since price and tax elasticity estimates for general populations may not apply equally to binge drinkers and other excessive drinkers (Ayyagari et al., 2013; Cook and Moore, 2000; Nelson, 2013b). Further, aggregate estimates tend to be biased away from zero (Manning et al., 1995; Nelson, 2013a). A comprehensive survey is required to address effects of prices and taxes on prevalence, intensity, and frequency of binge drinking for different age groups. To fill this gap, I conducted a systematic review of individual-level studies designed to better understand the potential role of economic incentives for reduction of binge drinking.