Nelson Jon P
Department of Economics (JPN), Pennsylvania State University, University Park, Pennsylvania.
Alcohol Clin Exp Res. 2014 Apr;38(4):1167-75. doi: 10.1111/acer.12345.
Relatively little is known about cross-country differences in alcohol affordability or factors that determine differences in affordability over time. This information is potentially important for alcohol policy, especially policies that focus on higher taxes or prices to reduce total alcohol consumption. This study estimates cross-country alcohol consumption relationships using economic models incorporating income and prices and alternative models based on alcohol affordability. The data and analysis are restricted to higher income countries.
Data for alcohol consumption per capita (ages 15+) are analyzed for 2 samples: first, 17 countries in the Organisation for Economic Co-operation and Development for the period 1975 to 2000; second, 22 countries in the European Union for the period from 2000 to 2008. Panel data models are utilized, with country and time fixed-effects to control for confounding influences. In economic demand models, covariates are real per capita income and real alcohol price indices. In affordability models, income is divided by prices to yield an index of alcohol affordability.
Analysis of data trends reveals that much of the increase in affordability is due to rising real incomes, and not falling real prices. Economic models of demand perform slightly better statistically, but differences are not substantial as income and affordability are highly correlated. For both samples, exogenous rates of growth of alcohol consumption are negative. Price and income elasticities, on average, are within the range of prior estimates. Affordability elasticities are between 0.21 and 0.25.
Although alcohol affordability is a valid concept statistically, its use in policy discussions tends to hide underlying causes of changes in affordability. A better approach is a comparison and analysis of trends and cross-country differences in real incomes and real alcohol prices together with the affordability index. Country-level analysis of income and price elasticities also is required.
关于不同国家间酒精饮料可承受性的差异，以及决定可承受性随时间变化的因素，人们了解相对较少。这些信息对酒精政策可能具有重要意义，尤其是那些旨在通过提高税收或价格来减少酒精饮料总消费量的政策。本研究使用纳入收入和价格的经济模型以及基于酒精饮料可承受性的替代模型，估计不同国家间酒精饮料消费的关系。数据和分析仅限于高收入国家。
对两个样本的人均酒精饮料消费量（15岁及以上）数据进行分析：第一个样本是经济合作与发展组织的17个国家在1975年至2000年期间的数据；第二个样本是欧盟22个国家在2000年至2008年期间的数据。采用面板数据模型，并通过国家和时间固定效应来控制混杂影响。在经济需求模型中，协变量是实际人均收入和实际酒精饮料价格指数。在可承受性模型中，收入除以价格得出酒精饮料可承受性指数。
对数据趋势的分析表明，可承受性的大幅提高主要是由于实际收入的增加，而非实际价格的下降。需求的经济模型在统计上表现稍好，但由于收入和可承受性高度相关，差异并不显著。对于两个样本，酒精饮料消费的外生增长率均为负。价格和收入弹性平均处于先前估计的范围内。可承受性弹性在0.21至0.25之间。
尽管酒精饮料可承受性在统计上是一个有效的概念，但在政策讨论中使用它往往会掩盖可承受性变化的潜在原因。更好的方法是对实际收入、实际酒精饮料价格以及可承受性指数的趋势和跨国差异进行比较和分析。还需要对国家层面的收入和价格弹性进行分析。
