Toward beneficial and practical standards for e-cigarettes

by Carl V Phillips

At the GTNF2015 conference I was on a panel discussing e-cigarette standards. The standards being discussed include manufacturing practices (e.g., clean rooms, hardware materials), specific technologies (e.g., whether a heating coil is capable of overheating, safe batteries), and ingredients (e.g., the perennial debates about whether some flavoring agents pose too much of a hazard). I decided to take the approach of addressing what the proper role for standards is, from a political economy perspective. What follows is the talk I gave (not word-for-word, and with a few additions to make it a document better suited for reading, taking advantage of the lack of time limit here). I follow that with some additional thoughts I voiced during the course of the Q&A and discussion.

1. The single proper and legitimate role of product standards is to protect consumers’ interests when they clash with merchants’ interests.

That is the most important single point here, so go back and read it twice. Whether we are talking about regulation or self-determined standards (the latter being the focus of this panel), if you cannot say why a particular standard is good for consumers then it is almost certainly not a proper standard. If it is not, it is probably either a protectionist measure by industry or an attempt by outside busybodies to harm consumers (or, as in the case of the Tobacco Control Act, both).

2. Merchants are legitimate stakeholders also, of course. The standards need to include the interests of merchants in terms of cost effectiveness. However, that is also a consumer interest: Any needless costs will be paid mostly by the consumers. That is simple economics, though it is often misunderstood. Almost every cost imposed on a producer is paid by the consumer.

Standards directly benefit merchants by reassuring consumers about the category and thus increasing total sales. Absent standards of some sort (which can take the form of industry rules, regulation, or product liability risk), there will be a race to the bottom, at least among some merchants. That is, there will be some merchants who cut every corner to save money, regardless of the effect on consumers. This is bad for consumers for obvious and immediate reasons. It also creates the risk of a “market for lemons” unraveling problem, whereby the low quality products drive out the higher quality products that necessarily cost more.

3. There are no other stakeholders than those I have mentioned. When government works properly regarding consumer products, it acts as an agent for consumers – it is never a stakeholder itself, merely an agent. That is, it is supposed to develop standards (in the form of regulations) that serve the public interest where consumer interests clash with those of merchants. But in the case of tobacco products, government has been captured by the “public health” special interest group that actively seeks to harm both consumers and merchants in pursuit of its special interest goals. In free societies, we do not normally consider such people — who have no direct stake in a matter, but merely an opinion about it — to be stakeholders.

4. Government – and more so, “public health” – cannot be trusted to contribute to standards that are in the public interest. Moreover, the next-most powerful entities in the space, the major tobacco companies, have a habit of using standards and regulation to benefit themselves at the expense of their competitors, thereby harming consumers by interfering with the market, or even directly at the expense of consumers.

This should not be overstated, as less sophisticated e-cigarette consumers often do when they blame the major companies for the fundamental problems facing e-cigarette producers and consumers. The majors are a partial bulwark against the worst excesses of captured government entities. They do a better job of serving as an agent for consumers than the government does in this case. However, even those among them that set out to develop standards and regulation in the public interest eventually run into the “scorpion and frog crossing a river” problem; big corporations cannot fully suppress what is built into their nature.

5. This combination of factors, particularly the fact that the government that is supposed to work on behalf of consumers has been captured by anti-consumer interests, has led to a backlash. There is general agreement among educated consumers – and those of us who represent consumer interests – that having no formal standards and no regulations is better than what governments would impose.

This has generated a simplistic libertarian buzz among consumers. Similarly, tobacco products, especially low-risk tobacco products, have attracted the interest of those who start with the simplistic libertarian view and look for issues to address. Therefore much of the policy chatter has a naive tone that suggests that unfettered free choice is optimal.

6. The problem is that consumers are never better off in a situation of unfettered liberty. It is true that simplistic economic analysis suggests that consumers are best off with a maximal array of choices (in the absence of externalities). That way each actor can choose which option generates the greatest net benefit. But this is trivially shown to be false in the presence of information costs and especially in the presence of important product characteristics that are not observable by the consumer.

Only the most sophisticated consumer can sort through the morass of available e-cigarette products right now. And even the most sophisticated cannot be sure that a particular manufacturer is not cutting some corners in manufacturing. It would be absurd to suggest that buyers of refrigerators or food should be free to choose – which really means forced to choose – among products of all possible quality levels. No one wants the “freedom” to choose a refrigerator that stands a good chance of catching fire or food that is infectious. No one would want to be forced to sort through information about products in order to find which products do not have those characteristics — if such information could even be determined with confidence, which is unlikely. Thus, the setting of muscular standards, by someone, is in the best interests of consumers.

7. So where does that leave us? Consumers depend on there being proper standards. Industry benefits from proper standards. But we cannot count on government to be part of the solution. Moreover, the most serious initiatives to create industry standards are heavily influenced by the majors, as would be expected; this is not necessarily bad, but it does create legitimate unease about it being done in the interests of consumers and the broader industry.

8. But we should back up a minute. This question of “who?” dominates the discussion, followed by specific details (“should we prohibit X”). These are terrible places to start the discussion. It really needs to start with Point 1, as well as recognizing that there are three fundamental categories for standards, in general and thus in the present case:

a) There are reasonable standards to reduce externalities (negative effects on third-parties), which possibly impose some cost on consumers. In the present case, this is basically just about child-safer packaging. It is difficult to think of any other externality from vaping that can be affected by product standards.

b) There are product characteristics like contamination, mislabeling, extremely high-risk ingredients, and mechanisms with high risk of failure that no consumer would choose. These can be flat-out forbidden without reducing anyone’s welfare. This goes back to that point about unfettered choice being a burden, not a benefit.

c) Then there is the hard part: Potential standards that would make the products better for most consumers, but would lower the welfare of a few informed consumers who accept the cost-benefit tradeoff that results from violating them. Though it is easy to forget if you are immersed in discussions among vaping enthusiasts, most consumers and potential consumers of e-cigarettes — as with most products — just want the product to work easily and be engineered to eliminate any unnecessary risks of failure or health hazard. They are best served by a product that adheres to standards that make the product as clean, as reliable, and, well, as boring as is possible for a reasonable cost, while still adequately meeting the consumer’s wants (nicotine delivery, aesthetics). That “boring” is not in any way derogatory; it describes most of us with respect to most products we buy. But there is a minority that would like to tinker or be able to make the informed choices to buy something that is not so clean and boring.

In the case of e-cigarettes, the obvious examples are flavoring agents or high-powered hardware that appear to perhaps create some extra health risk, but that some consumer might choose anyway. After all, we make consumption choices that pose small health risks every waking hour. In these cases, welfare is improved if there is a reasonable target standard set, but it is at a voluntary certification that consumers can choose to avoid if they want. The majority of consumers can just buy products that have A-level certification for being clean, reliable, and boring, without having to worry about the details, while the few can go some other way.

Needless to say, nowhere on this list are standards that would harm most consumers rather than benefiting them.

9. Standards will necessarily draw a bright line through what is a very fuzzy transition between the above point b), standards that basically everyone would agree to if they understood them, and c), those that a nontrivial portion of consumers would prefer to not be subject to. There is the further challenge of surmising what hypothetical fully informed consumers would choose if they understood everything. Addressing these challenges will inevitably be imperfect, but merely recognizing that they exist and should be explicitly considered will go a long way to improving the standards-setting process. It is not entirely clear that those in the standards-setting space have even thought this through. Failing to do so pretty much guarantees that they are headed in the wrong direction.

An interesting analogy can be found in the GothiaTek standards for smokeless tobacco. It is not overly cynical to say that these standards, while indeed making the products a bit cleaner, are a case of the dominant producer (for Swedish-style products) declaring that everyone else ought to adhere to the standards they have set for themselves, which has obvious competitive implications. In their own words, “it has now evolved into an industrial standard for all smokeless tobacco products in Europe.” But American smokeless tobacco consumers have demonstrated a clear and strong preference for products that do not meeting GothiaTek standards (the dominant American flavor profile results from production methods that are not GothiaTek compliant). Similarly, many vaping enthusiasts favor products that would not be likely to meet the standards created by a similar process for e-cigarettes. But then the question is, might American smokeless tobacco users actually prefer products that met, or at least came closer, to that standard if they understood what benefits they would get from it?

I closed the talk with : Of course that brings up the even trickier question of how to assess whether there actually are any such benefits. There is also the question of who can be trusted to come up with such standards. I would answer those, but I am out of time.

A few interesting points came up in the ensuing discussion. One was about what was the point of having voluntary standards? There are a few answers to that. The answer that undoubtedly dominates industry’s motivation is that such standards might mitigate governments’ imposition of bad standards via regulations. But there is also the point about a large portion of consumers wanting to see that “seal of approval” on their products — once they know that even looking for it is an option, of course. Finally, there is the fact that if there is an established minimum standard for good manufacturing practices and such, even if consumers do not understand it, any manufacturer who does not adhere to it will have a difficult time getting insurance and otherwise doing business-to-business transactions. (Aside: I believe this is mostly moot right now, and that everyone other than the traditional tobacco companies “insures” only via being judgment proof. I assume most smaller operators cannot even get liability insurance if they want it. Some of you reading this know more about this than I do, drop me a note to bring me up to speed.)

A related question was whether it might be good to have multiple standards created that compete with one another. I agreed that this might be a good idea in the short run, but the nature of these things is that in the longer run a single set of standards will emerge for major producers, though there is probably room for fringe alternative standards for the enthusiasts, as is the case for many products.

At one point during the discussion I pointed out that I had not been party to any of the various standards-setting discussions the industry has had, but I was guessing that none of them ever saw anyone say “wait, what is the benefit to consumers of this proposed standard?”, let alone discussing which side of the b)-vs.-c) divide something fell on. Judging from the sheepish looks from those who were principles in those conversations, I think I guessed right.

The most interesting thought I had during the discussion — something I would have put into my talk if it had occurred to me — was about the concept of incrementalism. A good rule of thumb in public policy making (and this is public policy making) is that if you are proposing a major departure from anything that has been done before, you are treading on dangerous ground. This is part of what the precautionary principle is about (the actual precautionary principle, not the absurd misuse of that term by “public health” people). Banning 99.99% of e-cigarette products, as the FDA deeming regulation might do, is a massive violation of this principle of incrementalism. But attempts to go from basically nothing to fully-formed detailed product standards for e-cigarettes also violates the wisdom of incrementalism.

The typical concern is that the chance of getting something badly wrong in an unforeseen is much greater with non-incremental changes. In this case, getting buy-in also becomes a much greater problem. It strikes me as far more promising to create an environment of trust and cooperation — and, yes, somewhat sticky buy-in — by establishing minimum standards for packaging, accurate labeling, basic sensible manufacturing processes, and a few basic sensible physical product limitations, and then seeing where to go from there. In other words, address category a) and what unambiguously falls into category b), get that solid and signed, and only then think about going into details beyond that. The create-it-whole approach pursed so far seems almost purpose-built to get a few manufacturers to sign on because it is merely a promise to do what they are already doing, while leaving the rest scoffing at the apparently cynical process. It is almost designed to fail to be aspirational. Getting to a set of standards that is conceptually simple and causes most educated observers to say “yeah, there is no way I would want to buy from anyone who does not meet that” — and thus contains nothing that reasonable people would say “I am not sure that is necessary” — would make a great deal of sense here.


2 responses to “Toward beneficial and practical standards for e-cigarettes

  1. Carl that is a very good explanation of why Top Down ‘ all at once town planning’ is dumb design.

    You are right to focus on the supply of a bog standard ‘ family car’ model e-cig, without restricting the enthusiasts from ordering the ‘sports model’ )or tinkering in the garage). Perhaps there could be a star system similar to that for cars? i.e if ‘its’ got temperature control , short circuit protection, certified pharma and food grade ingredients etc then it gets 5 stars?

    Regarding flavorings , a friend who is a engineer who has been making cosmetics, over the counter pharma, herbal remedies and the like, for 30 years, tells me that there are many thousands of flavorings (or more correctly aromas). Is it possible to create a list of say 30? that are known to have little risk?
    He also tells me that government systems that require all sorts manufacturers to keep proper records and invoices that prove that their output of product, matches their import of in this case ,certified pharma and food grade ingredients, are common and not too onerous.

    As far as the hardware goes, the batteries are or should be subject to the same standards as all other batteries, the electronics are a bit trickier- not sure how you would deal with that.
    While there are always going to be those who are naturally bent, greedy or plain crazy, at the moment, the profit margins for most look quite good. Therefore I think a more likely source of contamination problems is accidents; people who do not know what they are doing, trying ever more varied combinations.

  2. I have been deeply involved with standards (following them, not creating them, though I am close to colleagues who DO join IETF and other network standards bodies to set standards.) There are 2 main kinds I’ve seen. 1. Rigid Protocol standards. These may have options but they tend to be precise because they are designed for two disparate computers to talk to each other without the intevention of a being with common sense. 2. Security quality standards. THESE are graduated incremental standards. One vendor might meet 1-A, another might meet 2-b, for a lower cost. Customers who don’t need 1-A will buy a product that meets 2-b. An example for e-cigs might be that ‘L’ is for eliquids, ”P’ is for eliquid packaging, ‘DB’ is for mods, ‘DA’ is for atomizers. Each group might have levels 1-4 or 1-5. Level L-1 might be pg, nic (optional) vg only. L-2 might be pg, vg, nic (optional), flavorings that do not show any DA/AP at the 5ppm detection level. L-3 might allow up to 22 ppm DA/AP equivalent with disclosure, optional WTa or caffeine, optional NET. L-4 would have to be clean, have any DA/AP clearly labeled, pg, nic, vg, and nothing weird like other chemicals or oils (in other words, no “organic vegetable oil” or other nutso uneducated experiments.) Enforcement by whatever companies get a good reputation for their “Seal of Approval”, NOT by the standards-creation committee or bodies. Claiming to meet a standard you do not meet would A. lead to no seal, or loss of seal, and B. would probably violate normal Consumer Protection laws and possibly lead to lawsuits.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s