When You Remove the Alternative, People Go Back to Cigarettes

A paper published in the American Journal of Health Economics examined something policymakers often prefer not to look at too closely: what actually happens when access to safer nicotine alternatives is restricted. The study, E-cigarette Flavour Restrictions’ Effects on Tobacco Product Sales, analysed real-world sales data following flavour bans. Its core question was simple: if you take away flavoured vaping products, what do people do instead?

The answer is not complicated, but it is inconvenient. The researchers found that restricting e-cigarette flavours was associated with increases in cigarette sales. When the appeal or availability of vaping is reduced, some people don’t quit nicotine altogether. They switch back to smoking. This is substitution, a basic principle within Economics, and yet it remains almost absent from the way tobacco control policy is discussed in Australia.

That absence matters because substitution is not a niche or secondary effect. It is central to how people behave in constrained environments. When choices are limited, people do not simply stop wanting what they wanted before. They adjust. They look for the closest available alternative that satisfies the same need. In the context of nicotine, that need is both chemical and behavioural. It is addiction, habit, stress management, routine, and identity all at once. Policies that treat nicotine use as a switch that can simply be turned off tend to ignore that complexity, and the result is that behaviour flows around restrictions rather than disappearing.

Much of modern tobacco control rests on an assumption that if you make alternatives less attractive, people will simply stop. That assumption does not hold when tested against real-world behaviour. Nicotine use is not a binary choice between abstinence and the most harmful option. People move along a spectrum, often repeatedly, sometimes unpredictably. When lower-risk products are restricted, the higher-risk product becomes relatively more competitive again, not because it is preferred in isolation, but because it becomes the most accessible, familiar, and reliable option left standing.

That is what this study captures, not in theory or intention, but in actual purchasing behaviour. It reflects what people do, not what policymakers hope they will do. And that distinction is where much of the current disconnect lies.

Australia has taken this dynamic and intensified it. Rather than simply restricting flavours, policy has effectively removed retail access to nicotine vaping products, introduced prescription-only pathways that are difficult to navigate, and enforced supply-side crackdowns without eliminating demand. This is often framed as a decisive public health intervention, but from a behavioural perspective, it is an experiment in constraint. It reduces the availability of one category of product while leaving another, far more harmful category, widely accessible through legal channels.

The result is not the disappearance of nicotine use, but the reshaping of the market. If partial restrictions in the United States are associated with increases in cigarette sales, it raises an obvious and uncomfortable question about what happens when access is restricted far more aggressively. Demand does not vanish. It shifts, and it does so along predictable lines. Some people move back to combustible tobacco. Some turn to illicit vaping markets where product quality and safety are less certain. Others become dual users, maintaining smoking while intermittently accessing whatever alternatives they can find.

What makes this particularly significant is how rarely the substitution effect is acknowledged in Australian public health discourse. It is not just underemphasised, it is often excluded entirely. Discussions focus on uptake, prevalence, and risk, but rarely on displacement. The implicit model is one where reducing access to one product leads to a reduction in overall use, rather than a redistribution across products of varying risk. But when the most harmful product remains the most accessible, redistribution becomes the more likely outcome.

Even small shifts matter. Cigarettes are not just another nicotine product. They are uniquely harmful, responsible for the overwhelming majority of tobacco-related disease and death. A policy that inadvertently increases cigarette consumption, even at the margins, carries consequences that far outweigh equivalent changes in lower-risk product use. This is not a symmetrical trade-off. It is an asymmetrical one, where small behavioural changes can have disproportionately large health impacts.

There is also a deeper shift underlying this approach. Tobacco control in Australia has moved beyond harm reduction into something more absolute. The goal is no longer just to reduce disease and death, but to eliminate nicotine use. Within that framework, distinctions between products become less important. What matters is the presence of nicotine itself. Vaping, regardless of its relative risk, is positioned as a threat to that goal. And when something is framed as a threat, evidence that complicates that framing becomes difficult to incorporate.

This creates a kind of policy blind spot. Evidence showing that vaping can displace smoking sits uneasily alongside messaging that presents vaping primarily as a risk in its own right. Rather than integrating both realities, the system tends to prioritise one narrative over the other. The result is a policy environment where the risks of vaping are foregrounded, while the risks of displacing it with smoking are backgrounded or ignored.

What this paper shows is not that vaping is harmless, and it does not need to make that claim. What it shows is something more fundamental and harder to dismiss. Behaviour responds to incentives. Access shapes outcomes. Relative risk only matters if people are able to act on it. When a lower-risk option is constrained, some people will move toward a higher-risk one. That is not a theoretical possibility. It is an observed pattern.

The question for Australia is not whether vaping should be regulated. Regulation is inevitable and necessary. The question is whether current settings are reducing harm or redistributing it. If restricting access leads even a fraction of people back to smoking, then the policy is not neutral. It has a direction, and that direction carries measurable consequences.

Behind every percentage increase in cigarette sales is a person. Someone who might have switched but could not sustain it under current restrictions. Someone who tried to quit and found the available pathways too complex or unreliable. Someone who made a pragmatic decision based on what was accessible rather than what was ideal. Public health often speaks in aggregates, but those aggregates are made up of individual decisions shaped by constraints, incentives, and lived realities.

There is also a temporal dimension that is often overlooked. Behavioural shifts do not always happen immediately or cleanly. Someone who returns to smoking today may remain there for years. A missed opportunity for harm reduction is not just a momentary setback. It can become a long-term trajectory. When multiplied across a population, those trajectories accumulate into outcomes that are far harder to reverse than they were to prevent.

If you remove the alternative, you do not remove the behaviour. You change where it goes. That is the underlying lesson, and it is one that extends beyond this single study. Australia has not stepped outside that reality. It is operating within it, whether acknowledged or not. The only real question is whether policy is being shaped with that understanding in mind, or despite it.