As a lifelong Iowan, studying public health policy and cancer prevention in a state with one of the nation’s fastest-rising cancer rates, I care deeply about reducing smoking. Cigarettes are still the leading cause of preventable cancer, disease and death in the United States. Harms of this magnitude warrant critically assessing policies related to cigarettes and safer alternatives—and that includes taxation.
Over the last decade, a substantial body of economic research has found that when elected officials make nicotine vapes more expensive or less accessible, people will switch back to lethal combustible cigarettes.
The logic is straightforward and intuitive, yet often ignored. If two products satisfy similar demand, raising the price of one increases demand for the other. In economic terms, cigarettes and vapes function as substitutes. This substitution effect has appeared repeatedly in real-world studies (16 of 18) rigorously examining vape taxes or regulations across the US.
This substitution effect is especially concerning given emerging evidence that the adverse consequences weigh heavily on vulnerable populations, such as youth, older adults and parents with young children.
So when a recent study, published in Health Economics in April, concluded that higher vape taxes are unlikely to increase cigarette consumption among adults who vape—thereby garnering plenty of media coverage—I took notice.
The paper’s central policy implication rests on a major limitation: measuring what consumers say they would do, not what consumers actually do.
Policymakers should, too—but not by jumping to the conclusions reached by the study. Before enacting policies that further restrict access to vapes, we should first assess the validity of these claims. We owe that to people whose lives may be on the line.
The study itself is thoughtful and methodologically sophisticated. Researchers conducted a choice experiment, asking adults who vape how they would allocate purchases across various vaping products and cigarettes under different tax and price scenarios.
The paper is particularly useful in examining how consumers respond differently to pods, disposables, tank systems, and e-liquids—an important issue because states tax and regulate these products in dramatically different ways, and different population subgroups show different purchasing patterns across these products.
But the paper’s central policy implication rests on a major limitation: measuring what consumers say they would do, not what consumers actually do.
That distinction matters.
Real-world evidence points overwhelmingly in one direction. Across high-impact economic studies, the vast majority find substitution to cigarettes after vapes become more expensive.
Economists refer to this as the difference between “stated preferences” and “revealed preferences.” Stated-preference research can be valuable when policymakers are evaluating genuinely novel products or regulations that do not yet exist in the marketplace. If no real-world evidence exists, hypothetical experiments may be the best available tool.
But that is no longer true for vape taxes. These taxes have been implemented across the country for more than a decade. Policymakers no longer need to rely on hypothetical purchase experiments, because they can instead directly observe how consumers actually behave when taxes raise prices.
The real-world evidence points overwhelmingly in one direction. Across high-impact economic studies examining vape tax changes, the vast majority find substitution to cigarettes after vapes become more expensive.
Researchers have observed this effect using both retail sales data and population survey data: When vape prices rise in the real world, cigarette smoking often rises, too.
Once real-world evidence becomes available, the value of choice experiments ends.
Rather than being utilized to justify expanding regressive vape taxes, the Health Economics study should be taken as a warning about the inherent limitations of hypothetical-choice experimental research.
In the same study, consumers reported that they would not reduce cigarette smoking if faced with higher cigarette taxes. This specific result is quite difficult to reconcile, given decades of tobacco-economics research consistently showing that higher cigarette prices reduce smoking. With any study that struggles to reproduce one of the most foundational realities in tobacco control, policymakers should be leery of drawing broader policy conclusions.
Do not take this as me saying that choice experiments are useless. On the contrary, they can provide valuable insights into how consumers substitute across different products, especially in a rapidly evolving marketplace.
But policymakers should only rely on choice experiments to investigate hypothetical behavior change for policies that don’t yet exist. Once real-world evidence becomes available, the value of choice experiments ends. It doesn’t matter what people say they will do; what matters is what people actually do.
Photograph by Lindsay Fox via Flickr/Creative Commons 2.0