TWO‐CHOSEN‐ONE TAXATION: EXAMINING ITS POTENTIAL EFFECTIVENESS TO REDUCE DRINKING INITIATION AND HEAVY ALCOHOL CONSUMPTION IN LOW‐ TO MIDDLE‐INCOME COUNTRIES

We would like to thank the commentators for their thoughtful comments in response to our for debate contribution [1]. All commentators were in agreement that ‘two-chosen-one’ (2C1) taxation may have the potential to reduce alcohol consumption and drinking initiation in low- and middle-income countries (LMIC) which have a high prevalence of abstainers [1–5]. However, as noted in the paper and by the commentators, 2C1 taxation may have limitations in terms of unrecorded consumption, tax rate implementation and potential changes in beverage preferences. Medina-Mora raises the concern of a potential shift in alcohol consumption towards unrecorded alcoholic beverages as taxation increases [5]. Such a shift may be associated with all taxation increases [6,7], and is not specific to 2C1 taxation. Our evidence base for control of unrecorded alcohol consumption is limited, but some measures exist and should be the subject of further study [7]. Sarntisart notes that specific taxation rates should be linked to an inflation index [3,8]; otherwise, the relative cost of the tax will decrease with inflation. 2C1 taxation has an advantage over specific (only) taxation in that the ad valorem component of 2C1 taxation will act as a taxation floor, which creates inflation-binding taxation. None the less, 2C1's specific tax rate should be linked with inflation. Sarntisart also raises the issue of companies reporting abnormally low ex-factory prices to reduce the ad valorem tax rate [9]. To resolve this problem, governments should implement measures to verify the accuracy of the ex-factory prices reported by the manufacturing companies and/or base ad valorem taxation on the retail price. We agree with Österberg regarding the need for consistent taxation rates for similar beverages [4]; otherwise, the effectiveness of taxation will be less due to substitution effects [10]. In Thailand, differential tax rates have been the result of the political influence of alcohol companies [11], and such influences in general often hinder the implementation of best practices around the world. Österberg also raises the concern that changes in beverage preferences will decrease the effectiveness of the 2C1 taxation system. As 2C1 taxation levies a specific tax based on alcohol content on the cheapest alcoholic beverages which heavy consumers of alcohol purchase, 2C1 taxation will be effective in decreasing alcohol consumption among heavy drinkers. We disagree with Cook's conclusion that age restrictions may be better than taxation in reducing drinking initiation in Thailand, as the US data upon which this conclusion was based examined the effects of taxation on 30-day abstinence, which is different from life-time abstention. Life-time abstention is an embedded value in the culture of Thailand and other LMIC countries, and half of drinkers in Thailand do not consume alcohol before the age of 20 years [12]. Although higher tax rates on beverages preferred by youth and on high-alcohol content beverages can be achieved through methods such as minimum pricing [13], this may lead to fair trade violations [14] and has the potential downside of increased profits going to the alcohol industry. In summary, because LMICs typically have a high prevalence of life-time abstainers, an alternative view of alcohol control policies may be required which addresses simultaneously the issues of drinking initiation prevention among youths and of harmful alcohol consumption among heavy drinkers [15]. 2C1 taxation is a system which may accomplish both objectives. However, to implement 2C1 taxation effectively, governments need to implement equal tax rates among similar beverages, bind specific taxation rates to inflation, and either verify the accuracy of ex-factory price declarations or tie ad valorem taxation to alcohol retail prices. None.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.