I see this exchange as sharing the challenging conversations Jonathan Liberman and I have had on this topic. I hope his critique¹ provides further opportunity to challenge some of the reasons a timid government might use to reject a regulated market model (RMM).

Liberman argues that an RMM is probably no better than a conventional regulatory scheme that leaves tobacco companies’ marketing ability intact, and because it is less politically acceptable, it is a distraction. I accept that the potential benefits of an RMM I argued² might not all be achieved, but maintain that it can deliver better outcomes, is worth further development, and perhaps even fighting for.

INCENTIVES AND DISINCENTIVES

Disincentives (punishment) suppress behaviour, and while there are competing incentives to act, there are incentives to break or avoid rules. If incentives are large enough, finding proportionate penalties that effectively discourage action can be problematic. This seems to be the case with tobacco. Currently, corporation law mandates profit maximisation for tobacco companies. While there is no viable mechanism for profitability to be pursued through genuinely less harmful products, there is inevitable conflict with tobacco control goals. This creates problems, and requires huge and sometimes futile enforcement efforts. We need clever regulation because the solution of less regulation is seen as naïve.³ Laws should be about creating frameworks for people and organisations to act in ways society values, not just about creating disincentives for undesirable actions. The RMM proposal is about changing the playing field so that incentives to act contrary to the social interest are largely removed, making the task of ensuring compliance with the aims of policy much easier. The proposal is only unusual in that there are few situations where such a simple solution has been identified as possible. I simply do not understand an argument that a regulatory model based on disincentives is to be preferred to one which holds the promise of gaining a high degree of self-compliance.

A NON-PROFIT MARKETER OF TOBACCO PRODUCTS

The essence of the RMM is that a non-profit-maximising entity (called the Tobacco Products Agency or TPA) with a charter to maximise public health and social cohesion should be the sole agent marketing harmful tobacco products. The core idea is that the TPA as a monopsony would source cigarettes from many different manufacturers and could sell them under a common, generic label. Monopsonies are normative in many areas of commerce. A form of regulated market has been used for alcohol in Scandinavia, Canada, and parts of the USA (in the latter two cases, focusing on distribution not marketing). Monopsonies have existed for a long time for the sale of agricultural products, whether they be single government controlled authorities or small numbers of private companies, each buying the produce of large numbers of farmers. It happens already for tobacco growers. More recently, monopsonistic structures have been expanded to manufacturing. It is now common for product marketers (large, typically transnational companies) to source their product from a number of small producers, typically in the developing world, and to market them under their own brands. Marketing benefits more from large size, thus marketing centralises to fewer companies, while manufacturing becomes decentralised with more players. Neither farmers nor these small producers spend money on promotion: they cannot link their efforts to their products over those of others. The novelty in the RMM is in suggesting that a monopsonistic structure be mandated for tobacco marketing.

ACHIEVING BALANCED AND ACCURATE COMMUNICATION

Commercial entities have incentives to make the most positive possible claims about their products because this typically increases sales and profit. This can justify huge marketing budgets. Even where advertising is largely restricted, industry can make use of unregulated channels and notionally regulated channels (when they can do so in ways that minimise their risks of getting caught—that is, of the disincentives becoming real). Can a conventional regulator really control communication from tobacco companies to consumers? Where will they place the regulatory policemen? In the store to stop the retailer passing on messages from the company?⁴ At the news desk in a media outlet to trace the origins of every press release and story relevant to the issue? There is a huge difference between legislating control over communication and maximising the likelihood that companies will comply with the spirit as well as the letter of the law. How do you regulate innuendo? Can Liberman point to an example of where this has been achieved? Marketing is built on brands which are the mechanism for linking the marketer to the product. Packaging product from all manufacturers under a single brand (for example, type A cigarettes) can take away this link. Any branding, however prosaic (for example, the number 747), can be used to market, once it is promoted. This is the fundamental limitation of regulating for less attractive brands.

PROBLEMS WITH THE RMM

Liberman argues that products need to be identical if they are branded in the same way. Not so. Do apples need to be identical to be in the same bag? We expect some similarity, but the extent varies. The same brand of cigarettes can vary widely across markets.⁵ If there is too much variation, the consumer might seek another brand that offers more consistency. If consumers were sufficiently upset, then the TPA might need to intervene to protect itself in the marketplace. That said, I must ask, what consumer values are being served to have a mechanism that stops two cigarettes that are equally good at killing you tasting slightly different?

He is also concerned that the TPA would be motivated to seek the cheapest possible cigarettes. Cheaper tobacco is no less harmful. To the extent it is less consumer-attractive, it should reduce demand and thus harm.

INNOVATION

Liberman argues that conventional regulation can facilitate innovation. This is true, but a market-based solution has a far better track record. The biggest problem for the current system is likely to be the ongoing opposition of the industry to any innovations that threaten their profitability, and their huge capacity for misinformation. They have beaten us before. They deluded the community into believing they had genuine harm-reduced products when they did not, as with the Lights and Mild fiasco. Liberman is correct, a TPA might be corrupted. However, is it likely that a government controlled agency, however corrupted, could be as effective at stimulating tobacco use as companies like Philip Morris and British American Tobacco? We would all be better off if they were taken out of the equation.

IS THE RMM PROPOSAL A DISTRACTION?

Would advocacy for an RMM set us back by distracting from achieving more attainable goals? I don’t think so. Taking the worst case scenario: we advocate for it and governments say “no”. What have we lost? This depends on whether the RMM is positioned as a means of achieving an agenda or the agenda itself. In my opinion, it needs to be seen as a means. This is likely to lead to greater progress than if we adopted a less ambitious agenda, and I cannot see a plausible scenario by which we would get less. We are a very long way from the regulatory solution Liberman advocates, yet we have had lots of time to get there. Is it really going to be more difficult to sell a novel solution, than beg for more policemen? Why not offer both as options for achieving the agenda of needed reforms?

WHERE ARE WE?

When I wrote my 2003 article² I saw the RMM as both a means of better controlling tobacco marketing and of more effectively engineering moves towards less harmful tobacco products. Since then, I have become pessimistic about the potential of a smoked tobacco harm reduction strategy. We should aim to phase out smoked tobacco as quickly as possible. Murray Laugesen and colleagues from New Zealand have such a proposal (www.smokeless.org.nz). They propose to do this by phasing out nicotine in smoked tobacco, but allowing it in other forms. It demands critical scrutiny and, where possible, empirical testing. Even if we could rapidly phase out virtually all smoked tobacco, a TPA-like mechanism would still be desirable for other nicotine products, and to provide any residual smoked tobacco that might be allowed, but it would be less important.

In my opinion there are only three defensible positions for resolving the tobacco pandemic: prohibition of all nicotine use (impractical in most countries); rapid phasing out of the “dirty syringe” (smoked tobacco); or, the strongest possible strategies for discouraging smoked tobacco use. Both more realistic strategies will be much harder to achieve while the fox is in charge of the hen house. I think we should confront the fox of the tobacco industry rather than Liberman’s approach of fixing holes in the fence of regulation. We should be seeking the strongest possible framework for discouraging cigarette use.