Popular methods include surveys, statistical analysis and rooting through rubbish
http://www.economist.com/blogs/economist-explains/2017/06/economist-explains-4
LAST month Britain joined a growing number of countries in which cigarettes can only be sold in plain packs. Tobacco companies claim that the move will boost the sales of contraband cigarettes by making them trickier to spot. There is one way to tell whether this actually happens: track how black-market sales change. But how can such sales be measured?
There are about a dozen ways to do it, of which three are the most commonly used, says Hana Ross of the University of Cape Town. The first is a comparison of the number of cigarettes sold legally (from records on cigarette taxes paid) with the number of cigarettes consumed (which is calculated from surveys asking people how much they smoke). The gap between the two figures is the estimated share of the black market. The second commonly used method is to ask smokers where they have bought cigarettes and how much they have paid; smokers might also be asked to show the most recent pack they have bought. A price lower than that of legally sold brands suggests a contraband sale; some smokers openly admit that they have bought contraband cigarettes, or show a telltale pack.
The third method is to look at discarded cigarette packs and count up each that looks like a black-market purchase, for instance by missing its tax sticker or displaying a brand that is not officially registered. Discarded packs can be collected from vendors who sell cigarettes by the stick, from litter in the streets, or by rummaging through rubbish bins or the hauls of refuse-collection trucks. (“We dress them as if they are going into space”, says Ms Ross about the recruits who rummage through the heaps.)
Each of these methods has its weakness. Smokers may, for example, be reluctant to mention purchases of cigarettes they know to be contraband. They may also claim to smoke less than they actually do (especially if researchers come round soon after a major anti-smoking campaign). Ideally, multiple methods should be applied to get a better estimate of the total black-market sales. And trends over time are best measured by applying the same method. Such studies are conducted in a growing number of countries. Just because a sale occurs in the shadows does not mean it is impossible to cast a smouldering light on it.