Toolkit for Assessing the Unrecorded Alcohol Market

What is the size and composition of the unrecorded market?

field research approaches

Population–Based Surveys

One approach to sizing the unrecorded market is through population-based surveys (typically household surveys) that inquire into respondents’ unrecorded purchase and consumption habits. In addition to estimating total unrecorded consumption, survey instruments can also provide estimates of the prevalence, frequency, and distribution of consumption across different categories of unrecorded alcohol by including specific questions that inquire directly about the different unrecorded beverage types. These questions are likely to yield more reliable data for types of unrecorded alcohol that are easier for consumers to identify ― namely, homemade, surrogate, and products acquired through cross-border shopping (as compared to contraband, counterfeit, and tax leakage alcohol). Although factors such as price may signal to a consumer that a product is illicit or unrecorded in nature, it may be hard for the consumer to differentiate, for example, between contraband and tax leakage products. Counterfeit alcohol may be easier to differentiate from other types, depending on how well it imitates the genuine product in packaging and taste and the nature of the counterfeiting. For example, refills that occur on premise are typically harder for consumers to identify as counterfeit.

The strengths and weaknesses of survey approaches to measuring unrecorded consumption have been well summarized by others (e.g., Bloomfield, 2011; Greenfield & Kerr, 2008; Sobell et al., 1995). Links to existing survey items and instruments used to investigate unrecorded alcohol consumption and/or alcohol consumption more generally are also provided in Annex: Existing Survey Instruments.

Although surveys can provide valuable information by going directly to the consumers themselves, a major limitation is that some respondents will not be forthcoming in their responses. This is especially true for behaviors that are stigmatized within a given socio-cultural context, such as the consumption of certain types of unrecorded alcohol (e.g., surrogate alcohol, illegal alcohol) or consumption of alcohol more generally by members of certain groups (e.g., women, underage individuals). One strategy for combatting the underreporting of stigmatized behaviors is to embed questions about general alcohol consumption and unrecorded consumption, in particular, within a larger survey containing less sensitive questions. Researchers may also use uplift factors to adjust consumption estimates derived from surveys in order to correct for underreporting. These are often taken from studies that seek to estimate the magnitude of underreporting of alcohol consumption in a given population.

Another challenge for population-based surveys is that heavy drinkers, who often consume more unrecorded alcohol relative to the rest of the population, tend to be underrepresented. Typically, these surveys are conducted at the household level, an approach that fails to capture individuals who are homeless or institutionalized. This is particularly problematic when the special populations excluded from the sample exhibit different drinking patterns ― for example, contain higher numbers of heavy or problematic drinkers and / or consumers of unrecorded alcohol ― than the rest of the population. Heavy drinkers may also be less likely to agree to participate in surveys that include questions about alcohol consumption. Thus, survey researchers should always carefully evaluate the potential for undercoverage and put in place strategies aimed at minimizing or eliminating it.

When looking at change over time, the challenges posed by underreporting of consumption or undercoverage of specific consumer populations are somewhat mitigated if it is assumed that the error in estimates obtained from population-based surveys due to underreporting, undercoverage, and other sources of bias remains constant over time. This simplifying assumption has been used by researchers to justify the utility of population-based surveys for identifying trends in unrecorded consumption.