The Decline of Cigarette Usage, Explained by Network Effects
Once ubiquitous in America, cigarettes today are becoming an increasingly rare sight. If you ask anyone other the tobacco industry itself, this is a good thing—smoking is the number one preventable cause of premature death in the United States, while costing tens of billions per year in medical expenditures and loss of productivity. To analyze how best to discourage smoking, we must both understand the incentives that lead people to begin smoking as well as how methods of deterrence work. In his paper, Network Effects in the Demand for Cigarettes, Jeffrey Nehls attempts to explain these questions as network effect phenomenons.
The most basic and obvious network effect associated with cigarettes is that peer cigarette use influences increases demand. As more people around you begin to smoke cigarettes, the appeal of cigarettes grows. Nehls defines a utility function of smoking, U = s(x) + αη, which is very similar to the reservation price function we’ve worked with in class—r(x)f(z), where r(x) is the intrinsic value of the thing to person x and f(z) is a function of z, the fraction of people who use or associate with that thing. In Nehls’ formula, s(x) is the direct individual utility of smoking, where x is made up of direct variables such as price, advertising and other factors. This is quite similar to r(x)! On the right hand side, f(z) is analogous to αη, which represents peer effect, where α describes the intensity of the network effect and η is the number of people in the network.
Another different, less obvious network effect described by Nehls is one that was created by the tobacco industry. Similar to the airline’s industry use of frequent filer miles unique to so called ‘alliances’ of a few airlines, tobacco companies offer points for purchasing tobacco products unique to their company, redeemable for various merchandise. Both the airline and tobacco points are examples of switching costs, where by switching airlines or companies (or quitting!), individuals will lose out on the value of their points. Thus there is a deterrent to switching or quitting.
So why is cigarette usage in a steep decline? From dedicated public service announcements warning of the health risks of smoking to the prohibition of smoking in a vast amount of public areas, there are many variables that contribute. We can hypothesize that due to this general push against smoking, the amount of smokers has dipped below one of the tipping points, or unstable equilibria, as described in Chapter 17 of Easley and Kleinberg’s book Networks Crowds and Markets. Once below this point, a kind of cascade drives adoption further down. A specific example is given by Nehls. Imagine that a new policy makes cigarettes harder for youths to obtain cigarettes. Imagine that the policy causes an older sibling to not smoke, causing a younger brother also not to start smoking. We can imagine that at a certain point, the declining amount of smokers reaches a point, where once passed, drives demand ever downward, as more would-be smokers abstain from smoking since barely anyone else is, in a snowball effect. So if the adoption of smoking can be modeled as a simple network effect model with equilibria and tipping points, the goal of anti-smoking advocates should be to run commercials, information campaigns, and pass anti-smoking laws to simply reach the unknown tipping point, under which the tobacco industry would presumably be forced to downsize greatly.
Source
http://www.pugetsound.edu/files/resources/1359_NetworkEffectsInTheDemandForCigarettes.doc