What drives our Beer Consumption?---In Search of Nutrition Habits and Demographic Patterns

ABSTRACT

Conventional wisdom in Germany claims pork hocks with sauerkraut and beer. But is it really that simple? In an unbalanced cross-country panel covering 169 nations and time-series records of up to 52 years, we analyse drivers behind beer consumption. Based on data gathered from Worldbank and Faostat, we run multivariate panel regressions and test for the explanatory power of three categories of food and six macroeconomic and demographic variables. Indeed, we confirm most clichés of a typical beer drinker being a middle-aged urbanite with a strong desire for pork and potatoes, however, disliking cheese and wine.

KEYWORDS:

• Beer consumption
• nutrition habits
• demographics
• cross-country evidence

JEL CLASSIFICATION:

• J11
• J16
• Q11
• Q21

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Values provided by: https://de.statista.com/themen/2093/oktoberfest/.

² We provide an overview in the next section.

³ We do not consider the psychological perspective in terms of studies examining addiction to alcohol.

⁴ Colen and Swinnen (2011) conduct a closely related study, however, their focus is primarily the macroeconomic, religious and climatic drivers of beer consumption.

⁵ Although illnesses of the body and the mind due to alcohol, or alcohol as a source of these problems are a very important research field, we argue that they are not of high importance as general driving factors of beer consumption. Therefore, we will not include this part of the literature.

⁶ For survey and meta-studies on price elasticity and the impact of price and tax levels see Gallet (2007) and Wagenaar, Salois, and Komro (2009).

⁷ http://faostat3.fao.org/home/index.html.

⁸ http://data.worldbank.org.

⁹ Colen and Swinnen (2016) also state doubts that alcohol consumption for some countries might be misreported because of religious views or prohibitions. Because of the high number of countries in our sample and a conducted robustness test for only European countries we refrain from changing the original dataset.

¹⁰ For a detailed list, we refer to Appendix A1.

¹¹ Skog (1986) does not specify additional factors exactly; however, concludes that alcohol consumption and general private consumption are negatively related to various subsamples.

¹² In these equations we only depict the functional relationship between the variables. Which variables we include in levels (differences) can be seen from the result tables.

¹³ We test for redundancy of each/both fixed effects in every model using F- and likelihood-ratio tests, and find that the hypothesis can be rejected every time at the highest significance levels.

¹⁴ See Kesse et al. (2001), Männistö et al. (1997) and Herbeth et al. (2012), who support the influence of meat consumption and Toniolo, Riboli, and Cappa (1991) for counter-evidence.

¹⁵ See Kerr et al. (2004) demonstrating that men are more thirsty when it comes to beer than women.