The Schelling Point for being on the Discord server (see sidebar) is Wednesdays at 10 PM EST

Thank you for doing something ambiguously between smoking and not smoking

“Funge” is a funny word. It refers to the thing which fungible things are able to do, sort of along the lines of what an extra unit of a good is going to replace. I don’t think it’s a real word and I’ve only heard it used by people connected to the Center For Applied Rationality. This is too bad, as it prevents everyone else from understanding, let along generating, important sentences. Like “Be careful what you’re funging against.”

Maybe this is why so many people are so careless what they’re funging against. Consider our recent discussion of the minimum wage. The minimum wage means no one has to work for below minimum wage. Its desirability depends a lot on whether below-minimum-wage funges against above-minimum wage jobs or against unemployment. That is, if we ban 100 below-minimum wage jobs, do we get 100 above-minimum-wage jobs, 100 more unemployed people, or a mixture of both?

This was also part of the thrust of my argument about drone warfare – it’s not funging against peace, it’s funging against much worse types of warfare. The same piece cited the status quo bias and indeed these two ideas are probably related.

This, combined with a complicated regulatory environment and sheer bad luck, seems like the best explanation for the trials (both metaphorical and literal) of e-cigarettes.

E-cigarettes (the “e” is for electronic) are pseudo-cigarettes that contains nicotine without tobacco. They don’t smell bad, they don’t produce secondhand smoke, and they don’t cause cancer. They are strictly better than regular cigarettes in every way. The governments of several countries are doing their best to ban them.

The governments’ position is that they are a stealth attempt to trick people who have successfully avoided regular cigarettes into smoking anyway. There’s some merit to this. Some of them have nice fruity flavors that might appeal to children. And because they don’t produce smoke, they’re legal to use indoors, where some people might not be allowed to use the real thing. There might be this tiny contingent of non-smokers who were just waiting for a flavorful and indoor-useable way to get the addictive expensive chemical they have no reason to want.

And yes, this would be bad. Nicotine is addictive no matter how you get it. There are some claims – and I don’t yet know how seriously to take them – that the other chemicals in tobacco inhibit monoamine oxidase which further perverts dopamine levels and makes cigarette smoking more addictive than nicotine alone would be, but this is different from saying nicotine isn’t addictive at all. Even if nicotine has few ill effects – and in fact this seems to be the case – there is a strong economic and convenience-based argument for not getting addicted to it.

I should clarify that “few ill effects” claim. A massive overdose of nicotine can kill you (so can a massive overdose of caffeine, Tylenol, or vitamins). Nicotine is a stimulant which raises your heart rate and blood pressure a bit (so is caffeine). It may increase the risk of diabetes, but it may treat cognitive impairment. Overall, it seems to have a complicated mix of minor bad and minor good effects, about the same as anything else in health. And like everything else in health, tomorrow three labs will come out with studies proving it causes cancer, and a fourth will come out with a study proving it prevents cancer, and one of them may even be right.

There are some studies that show that e-cigarettes have “toxic additives”, but these seem to be in ridiculously tiny trace amounts, don’t seem to make it into the vapor or the body of the user, and the entire problem could be solved by regulation anyway if anyone had a desire to regulate them. This entire issue struck me as a red herring and I bet you can buy fish at any market in the country with more toxic additives than the worst e-cigarette on the market.

So let’s accept that using e-cigarettes will get you addicted and set you back a lot of money and otherwise be annoying but probably not deadlier than anything else you do on a daily basis. What then?

Well, in that case, it’s worse than not smoking but much much better than smoking. And whether or not their existence is a good thing depends on what they funge against. Do they funge against smoking tobacco or not smoking at all?

I would have liked to get the CDC’s opinion, but their webpage on the issue is missing and from commentary I gather it didn’t have the information I wanted anyway. But I did find this:

A June 2011 national study conducted under the supervision of ECH Research of Cincinnati, in conjunction with Opinionnaire, surveyed more than 200 smoker households that use electronic tobacco products and found that 99% of e-cigarette users are either current or past users of multiple forms of tobacco. Approximately 70% of survey respondents said they intended to quit smoking before starting e-cigarettes.

I can’t find the original or even so much as a description of what “ECH” stands for. A sketchy online survey claims that 70% of e-cigarette smokers were former smokers.
So let’s just say “probably some high number”. This seems quite plausible to me. How many non-smokers think “You know, I want a product with all of the addictiveness and expense of cigarettes, but none of the coolness? In fact, I want to look like a chronically uncool recovering addict inexplicably smoking a glowstick.”

That same Etter and Bullen paper says that 96% of the ex-smokers said the e-cigarette helped them quit or reduce smoking, and 79% felt they might relapse to smoking again if they didn’t have e-cigarettes. Randomized trials seem to confirm this result, with the average smoker in the trial dropping from 19 cigarettes per day to 2 cigarettes per day after trying e-cigs.

It’s not really surprising that e-cigarettes work. My current model of cigarette addiction is that it consists of the interaction between (1) nicotine, (2) smoking-associated behaviors which have become associated with the rush from nicotine over time and might have more complicated components like “oral fixation”, and maybe (3) a contribution from MAOIs in the tobacco. Normal cigarettes have all three. E-cigarettes have (1) and (2). Nicotine patches have (1) only. Therefore, e-cigarettes should be more useful in quitting than nicotine patches, albeit not a perfect replacement for regular cigarettes. This does indeed seem to be what has been observed.

Some people argue that the effects of e-cigarettes haven’t been perfectly studied, that they might be unsafe in some unclear way. And that as a smoking-cessation device, they’re technically a medical device and therefore need to undergo as much study and regulation as any other medical device before being given to the public.

I know there’s constantly a debate between the people who want to evaluate each new medical intervention for safetly and the people who want to use exciting new potentially life-saving technologies now, and I know that sometimes the former group do turn out to be right. Varenicline is a popular antismoking drug that was eventually discovered to drive people insane in various way and sometimes lead to suicide; although it is still used on people with both an extreme desire to quit smoking and impressive mental fortitude, it’s nice that people paid careful attention to the side effects and didn’t just give it out like candy.

But e-cigarettes are literally the exact same thing as something that’s given out to anyone who asks in convenience stores, except without the cancer. To suddenly hold them to an extremely high standard of safety seems like a fallacy of fungibility.

The worst are the people – one of whom has so far appeared in every article I have read on the subject – who say that we should be careful because “Big Tobacco” is pushing them as a “solution” to the problem of declining cigarette sales. First of all this is just factually wrong; most e-cigarettes are made by alternative companies in direct competition with Big Tobacco. Second, if your reasoning strategy is identifying the Evil People and then minimizing their utility, you probably shouldn’t be making public policy.

So I’m against banning e-cigarettes, and I’m even against things like taxing them or prohibiting their use in public places, on the grounds that the more smokers are encouraged to switch to e-cigarettes, the better. Like, if a public e-cigarette ban reduces the number of smokers who switch to e-cigarettes by 2%, you’ve just killed an extra 9000 people per year – about three 9-11 attacks, or twice the number of US soldiers who died in the Iraq War.

(this is why public health is about a hundred times more important than any other political issue, and even tiny little marginal issues in public health are more important to get right than, say, anything you will see people changing their profile pictures about on Facebook.)

So obvious conclusion is obvious and almost too boring to discuss. I got interested in e-cigarettes because a friend asked me whether he should start taking them even though he didn’t smoke as a way to get the cognitive enhancement effects of nicotine.

I guess that depends on what you value. I personally wouldn’t do this because I’m terrified of addiction. Then again, I avoid coffee and I drink like three glasses of alcohol per year in order to avoid addiction, which most people would consider sort of excessive. And I frequently take weird psychoactive Mexican herbs in order to achieve lucid dreams and would use LSD in a heartbeat if it were legal. So I guess my answer is that my feelings on the costs vs. benefits of various substances aren’t likely to generalize across the population.

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

42 Responses to Thank you for doing something ambiguously between smoking and not smoking

  1. Andrew Rettek says:

    I remember talking to you about this a few months ago, and I remember you had said that you couldn’t find evidence that nicotine alone was addictive. Am I misremembering or did you find something after that conversation?

    • Scott Alexander says:

      I don’t remember saying that, but if I did say it I was wrong. I might have said it wasn’t *as* addictive, or that some people were claiming it wasn’t addictive, but those people seem to be wrong.

  2. Doug S. says:

    I wonder… how hard would it be for a Mad Scientist Lab to make a virus that could wipe out the world’s tobacco crop?

  3. Berry says:

    Isn’t there some sort of argument to the effect that we *want* people to smoke cigarettes because it saves the government money or something like that? Anyone remember the details?

    • Benquo says:

      If your sole goal is to minimize health care expenditures, and the only policy decision you can influence is whether to discourage or encourage smoking, then you should encourage it because it will make people die earlier.

      That doesn’t mean we want people to smoke. There are plenty of ways to reduce health care spending with less adverse effect on healthspan.

      • Fnord says:

        Even if you REALLY want to reduce healthcare costs and REALLY don’t care about peoples lifespans, my impression is that you’re better off encouraging people to drive recklessly and possibly overeat and never exercise. Cancer is expensive; trauma and heart attacks are are cheap.

        • gwern says:

          Well gosh, no one says you can’t do all of those! Every sane supervillain knows not to put all your eggs in a basket (you want a whole roulette of options).

  4. ari says:

    (this is why public health is about a hundred times more important than any other political issue, and even tiny little marginal issues in public health are more important to get right than, say, anything you will see people changing their profile pictures about on Facebook.)

    You know it’s not health policy alone that does that. Basically any public policy that does anything at all, over a large enough population, is going to affect how and when people in that population die. Given enough people, someone’s going to be right next to just about any margin you can imagine, and the tiniest change in incentives is going to push them over into making different decisions.

    Example: EU bans sex discrimination in car insurance. Car insurance used to be cheaper for young women than young men, owing to the fact that the latter are generally more dangerous on the road. So what’s an insurance company that wants to be competitive do when they can’t discriminate by sex? They’re going to pick a price for both sexes that’s somewhere in between the earlier prices. In a large enough population, there’s going to be some poor young men who happened to be right near that margin and go from not driving to driving, and being young men they’re more likely to drive like assholes, and boom, you’ve got a body count.

    I suppose the difference is that with health policy you can at least sort of expect that a good economist might be able to track the effects of a given policy. If e-cigarettes are less deadly than conventional ones and mostly only used by people who used to smoke conventional ones, and that’s the whole story (or close enough to being the whole story), then sure enough, e-cigarettes will save lifes. Non-health-related public policy can change incentives all over the board or only change incentives which seem to have nothing to do with health – but the effects should be there regardless. It’s just that it’s infeasible to actually figure them out.

    For instance, suppose (suppose) that the obvious simple economic analysis on minimum wages applies and any increase in the minimum wage knocks people off the labor force. And furthermore that unemployed people are more likely to die young than even miserably employed people. In that case, raising the minimum wage = corpses. An increase in the income tax should have a very similar effect. In a large enough population, someone’s going to be the marginal employee who gets fired when there’s less money to pay.

    In public health, you could kinda more or less stop there in a lot of cases and make a sane decision based on that analysis. In non-health public policy, well, maybe a higher minimum wage has other useful effects that make up for the direct effect, or maybe a higher income tax allows the government to fund programs that more than make up for the effects of the tax. Figuring out exactly what happens to people’s behavior when you make sweeping changes to the economy, even if they might be small ones in numeric terms, is just really, really hard.

    How politicians making policy nations with many millions of people deal with this, I don’t know. Some just go with the flow of identity politics, I suppose, but politicians aren’t stupider than the general population, they have to be aware that even decisions that seem like simple morally good things (let’s not discriminate based on sex in car insurance!) can have a body count.

    • Scott Alexander says:

      But sex discrimination and minimum wage issues don’t kill 400,000 people a year. Smoking does. That makes a small percentage-wise effect on smoking matter more. Even your minimum wage example was only able to work by turning it into a public health one (unemployed people are less healthy).

      • Federico says:

        This graph of the 20th century’s biggest killers seems pertinent.

      • Fnord says:

        Well, if what you care about is body count, then obviously only things that turn into public health matters are important. Because anything that causes a body count is a public health issue. But if you don’t accept that economic issues have value, we’re banning cars to prevent car accidents.

        Total yearly wages in the US are about 6 trillion dollars (source). So, to a first approximation, a 1% discrimination-driven pay gap between men and women is effectively depriving women of 30 billion dollars each year, collectively. Statistical values of human life used by government agencies seem to be between 5 and 10 million dollars, so the harm here is 3000-6000 death-equivalents for a 1% change in wage gaps. Whereas using the figures in the OP, a 1% change in smoking is 4500 deaths.

        All this is first-approximation, order-of-magnitude stuff, of course. But it does look like the orders of magnitude are comparable.

  5. wallowinmaya says:

    Same with nicotine gums. You can only buy them in pharmacies (at least that’s the case in Germany) and they are also quite expensive which probably prevents many people from utilizing them.
    In fact, I use e-cigarettes because they’re much cheaper than nicotine gums.

  6. Douglas Knight says:

    if a public e-cigarette ban reduces the number of smokers who switch to e-cigarettes by 2%

    That is an inherently ambiguous phrasing. Percentages are proportion. 2% is 2 out of 100. 100 what? What is the denominator? 100 smokers or smokers-who-would-have-switched? If 10% of smokers switch to e-cigarettes, then there is an order of magnitude difference between 2% of smokers and 2% of switchers. Your phrasing suggests the latter, but your deaths suggests the former. One way to clearly indicate that you’re talking about 2% of smokers rather than 2% of switchers is to say “reduces by 2 percentage points” (or just “2 points”), meaning subtracting 2% from 10%, rather than taking 2% of 10%.

  7. Phil says:

    “The minimum wage means no one has to work for below minimum wage. ”

    No, the minimum wage means means no one can work for below minimum wage.

    • “No, the minimum wage means means no one can work for below minimum wage.”

      Not even that– the minimum wage means no one can legally work for below minimum wage.

  8. Anonymous says:

    What kind of Mexican herbs? Calea Zacatechichi?

    • Scott Alexander says:

      Yes. I’ve tried it in capsule form and it has done absolutely nothing. Debating whether to get the raw leaves and make it in the tea to see if that helps or if this is a waste of time and of having-to-try-really-disgusting-tea.

  9. Vilhelm S says:

    Since you mention addiction as a negative, it might be interesting to note that gwern and wedrifid on Less Wrong both mention the habit-forming property as a something potentially desirable. They suggest you take a dose as you do something like exercise or pomodoro, in order to create a habit.

  10. Deiseach says:

    Personal axe-grinding going to be going on all over the place, so beware me ranting on this one.

    If I were Absolute and Sole Empress of the World in the morning, I would ban cigarettes and every form of tobacco in whatever form possible. I’ve had a close family member die of lung cancer solely due to cigarette smoking, and I can tell you from personal witness it is a wretched, miserable, horrible, awful way to die. If you have the choice between dying of lung cancer and getting eaten by a shark, go for the shark. It’ll be faster and less painful, believe me.

    Now, my relative tried unsuccessfully for years to give up cigarettes. Tried, failed, went on low-tar cigarettes as a substitute (at least these wouldn’t be doing as much harm, right?) and over the course of a few years with those, increased the amounts she smoked (because, as far as I can make out, the low-tar ones didn’t deliver the same whatever it is you get from cigarettes so she needed to up her intake). Sounds like long-term addiction to me, yes?

    So – e-cigarettes. Now, nicotine patches and gums are perfectly legal and even sold in chemists’ shops (drugstores for you Americans) and are advertised as cigarette-substitutes when giving up smoking. I don’t know about e-cigarettes. I think an outright ban would be a bad idea (I can see a court case brought by a manufacturer arguing if nicotine patches are legal, why not these?) but I wouldn’t let them be sold in ordinary shops alongside cigarettes.

    And I do see the dangers that these might be a way of “smoking without the smoking”. Same way that alcopops were a worry for underage drinking (and let’s face it, these were invented and marketed specifically to attract younger and female drinkers), these kinds of ‘not really cigarettes’ might be abused by young smokers or those who want to start smoking – and kids do want to start smoking; at the school where I worked, I’ve seen twelve year olds (yes, I knew they were twelve because that’s the age they start in secondary school) smoking at the school gates before they started class in the morning, or at lunch break.

    So my instinct is “Ban! Ban everything!”! but if these are introduced, they need to be controlled because human nature will always abuse something – and if you can get a legal drug fix without the risk of cancer, why wouldn’t you jump at it (unless you’re a weirdo like me, who doesn’t even like taking prescription medicines if she can get away with not doing so, and refused an offer by a legitimate and not a pill-pushing doctor to be prescribed Xanax)?

    • But there’s no real lung cancer risk to smoking e-cigarettes, so if lots of people start smoking them I don’t see why it matters.

      Alcopops vs normal alcoholic drinks is a different matter, because alcopops have the same negative effects in the short and long run as alcohol.

      • Deiseach says:

        I don’t think nicotine addiction is any better for people, and I do think there is a risk of nicotine addiction. Maybe I’m prejudiced from reading all the Golden Age murder-mysteries where the deceased was done in by nicotine poisoning which the murderer hoped to pass off as a result of too much tobacco or else as a natural-causes, if sudden, death 🙂

        • Max says:

          If you don’t think nicotine addiction without risk of lung cancer is any better, then would you consider it equally bad if your relative were still alive today, never having gotten cancer or any other smoking related disease, but were still shelling out the same money that they used to for cigarettes?

        • Of course nicotine is addictive!

          However, nicotine addiction is much better for people if they get it through e-cigs than if they get it through cigarettes. If all smokers changed to e-cigs there would be far fewer deaths. The cost of banning e-cigs is many more deaths.

  11. Anonymous says:

    “and would use LSD in a heartbeat if it were legal”

    Why is it being illegal stopping you? This is a genuine question – I can’t understand why someone would take this stance.

    • Joe says:

      It’s probably more dangerous to take illegal drugs

      • anon1 says:

        You don’t know how much you’re getting, and unless you sacrifice some for testing you don’t know whether it’s LSD at all rather than some other chemical with much less history of safe human use. Knowing the dosage is kind of important.

  12. Romeo Stevens says:

    >So let’s accept that using e-cigarettes will get you addicted and set you back a lot of money and otherwise be annoying but probably not deadlier than anything else you do on a daily basis. What then?

    e-cigarettes are a really cheap nicotine delivery system. Like pennies per cigarette equivalent cheap if you mix juice yourself. I don’t see how taking advantage of the effects of nicotine is any worse than caffeine. I started vaping while I study and have seen huge productivity improvements from the reduction in ugh fields.

  13. BenSix says:

    I ended my unhappy love affair with tobacco this year and I wouldn’t use e-cigs except for curiosity. They don’t have the organic aesthetic of cigarettes (which somehow exists despite their being crammed with noxious chemicals).

    And that as a smoking-cessation device, they’re technically a medical device and therefore need to undergo as much study and regulation as any other medical device before being given to the public.

    And yet, antipsychotics are distributed like candy despite their long-term effects being somewhat mysterious but liable to involve brain tissue loss. The simultaneous phenomena of the prohibition and popularising of drugs are very odd.

    • Scott Alexander says:

      Antipsychotics do, in fact, require years of study and regulation to be approved, and a prescription to be distributed. They’re at least as hard to get than nicotine gum, and I don’t think anyone is really itching to get their hands on them.

  14. Aaron Brown says:

    But e-cigarettes are literally the exact same thing as something that’s given out to anyone who asks in convenience stores, except without the cancer. To suddenly hold it to an extremely high standard of safety seems like a fallacy of fungibility.

    “To suddenly hold it” would read better as “To suddenly hold them”, since “e-cigarettes” is plural. (I wouldn’t nitpick like that if your essay weren’t so great.)

  15. Mary says:

    Its desirability depends a lot on whether below-minimum-wage funges against above-minimum wage jobs or against unemployment. That is, if we ban 100 below-minimum wage jobs, do we get 100 above-minimum-wage jobs, 100 more unemployed people, or a mixture of both?

    To be sure, you also have to know what the motives of those urging the minimum wage increase are. If it is improve the lot of the poor, these questions are relevant. If it is to give themselves moral egoboo about doing good for the poor without expense or effort on their own part, they are irrelevant to them.

    • Scott Alexander says:

      Seems like genetic fallacy.

      • Jonas says:

        Seems unlike genetic fallacy to me.

        I think you, Scott, are assuming a definite (individually preferred social?) utility function in terms of which you evaluate the policy based on its projected consequences.

        I think Mary’s point is that other people may have different utility functions, including one where self-satisfaction is more important than the lot of the poor. Whether advocating the policy is desirable depends on your desires (and not just its consequences).

  16. Joe says:

    As a former smoker I can tell you that no one will ever be addicted to an e-cigarette. I used one for about two days and they are not worth the money. If you want to quit use the lozenges. They have enough kick but also cause the hiccups which discourage long term use.

  17. Army1987 says:

    I remember a discussion about whether to allow homosexual couples to adopt children in meatspace; after a while, it turned out that Alice thought adoptions by homosexual couples would funge with adoptions by heterosexual couples, Bob thought they would funge with children staying in orphanages, and since “children are better raised by straight couples than by gay ones” and “children are better raised by gay couples than by orphanages” don’t actually disagree, the discussion ended there.

    • Scott Alexander says:

      Oh, cool, in retrospect that does explain the disagreement much better than I’d been able to do before!

  18. Doug S. says:

    “Displacing” might be the word you’re looking for, as in “be careful what you’re displacing”.

  19. Douglas Knight says:

    I don’t like this use of “funge.” It doesn’t seem to have much to do with “fungible.”

    Usually people say that money is fungible. The best I can do of making coherent your use of the word is as people said that debt is fungible, but really I don’t think there is any coherence to your usage, let alone connection to the usual usage.

    • Douglas Knight says:

      I meant to say “as if people said that debt is fungible,” the point being that people don’t.

  20. Justen says:

    I’m actually one of those tiny fraction of people who were waiting for nicotine to become available in a non-tobacco form (I don’t like lack of dosing control with patches and I loathe chewing gum, so those options were right out). So yes, we do exist. I use it for the mild cognitive enhancing effects. It’s cheaper than caffeine, a habit I keep up mainly for personal pleasure, so the financial problem isn’t that big a deal in my view. Also, thanks to the slight work performance enhancement it might be a net benefit – especially since I’m a freelancer, so higher productivity leads directly toward more pay.

    Anyway, food for thought. Great article, thanks.