Thursday 31 January 2019

The art of worldly wisdom, smoking, and British American Tobacco

When students study either undergraduate finance or an MBA, they are taught various methods of valuing companies - primarily discounted cash flow models, utilising appropriate discount rates (or inappropriate ones - CAPM has been long debunked). The problem with this valuation education is that the mathematics involved in valuing a company - given certain cash flow and discount rate assumptions - is trivially easy. The hard bit is figuring out what the likely cash flows will be. And finance degrees/MBAs offer virtually no useful instruction in this regard.

That is because predicting what future cash flows will look like requires the use of mental models that draw on intellectual disciplines outside of finance, not to mention simple life experience. Predicting likely future cash flows requires not just a knowledge of economics and business strategy/competitive dynamics, but also an understanding of social/consumer psychology, the dynamics of political activism and organisational politics, and technology trends, amongst other things - not to mention the human psychological failings that drive market inefficiencies.

This is one reason Munger has described investing as really the art of worldly wisdom, and preached about the importance of having a wide array of mental models, acquired through a voracious appetite for reading. It is only through these avenues that investors can have any hope of exercising reasonable judgment and valuing companies correctly, but these interdisciplinary skills are not taught in business school.

An example at the moment are the shares of British American Tobacco (BATS LN), where a lot of the most important issues require insights into smoker behaviour/psychology, as well as the incentives of organised activism, that do not come naturally to finance wonks. I argue in this post that BATS is cheap at present, given my assessment of these issues, drawing mostly on intellectual interests and life experiences coming from outside the world of finance. I conclude that investors seem to be - much as they have for the past 30 years - misunderstanding smoker and activist behaviour.

The best performing industry over the past 35 years is still misunderstood

Over the past 35 years, the tobacco industry has been the best performing stock-market industry in the world, despite declining volumes in the developed world, lawsuits, and ever more stringent regulatory restrictions. There have been two reasons for this: one, the economics of the industry, and the favourable effect regulator behaviour has had on the industry's supply-side dynamics; and two, the fact that investors have continued to misunderstand smokers and smoking (and activist behaviour), not least because most people in the investment business do not smoke. I do (at least some of the time).

First, the easy bit on industry economics: Tobacco companies have prospered because regulators have essentially created formidable barriers to entry, and allowed the market to consolidate down to a small number of players with outsized market power, all of whom understand that the game is all about raising prices. Even today, BATS only makes about 2c per stick in operating profit. How hard will it be to increase that to 3-5c over the next decade? Not very. As is argued very eloquently in the excellent book 'Capital Returns', while most investors focus on the outlook for demand, in reality, industry profitability is driven by supply, and the tobacco industry's supply-side dynamics have been better than almost any other industry.

Most developed markets have tobacco advertising bans. This means that it is almost impossible to launch a new, competing tobacco brand. In many countries, you are not even allowed to engage in display advertising at the point of sale (cigarettes sometimes even need to be hidden from view behind the counter). Customers therefore wouldn't even know your new brand existed. Furthermore, because regulators want to reduce tobacco consumption, they are actually more than happy for tobacco companies to acquire each other and keep raising prices. Unlike most industries, antitrust concerns are therefore non-existent - quite the reverse. Meanwhile, the highly addictive nature of nicotine makes consumers not particularly price sensitive.

However, while all of the above is fairly obvious, it would be rendered irrelevant if smokers all just decided to stop smoking, or if regulators were to ban tobacco entirely, and it is the systematic over-estimation of these risks by investors that has driven the undervaluation of tobacco stocks over most of the past 30 years. This has allowed the companies to buy back tonnes of stock at cheap prices, or pay high dividend yields which investors have been able to reinvest at high returns (whereas most industries with great economics have traded at high valuations, preventing this opportunity).

Most investors have long believed that with so much regulatory pressure being brought to bear against smoking, and all the advertised health risks associated with smoking, that it was only a matter of time before everyone quit and the companies went out of business. They have been wrong. Complex spreadsheets and DCF models weren't of any assistance in understanding these issues.

Investors were starting to come around to appreciating and pricing in the power of the industry's economics in recent years, with the stocks being bid up to nearly 20x earnings, almost on par with other consumer staples behemoths such as P&G, Unilever, Colgate, and Coke. However, to a large extent this was due to the advent of next-generation products, including e-cigarettes and heated tobacco, which were seen as holding out the potential to stabilise industry volumes, or even return the industry to growth. But sentiment has recently turned down again, with the stocks heavily derating.

There have been two contributors to this sell-off. One has been a slowdown in the pace of next-generation product sales, both due to slowing market acceptance in some regions, and more importantly, tighter regulatory restrictions (if not outright bans in many countries). Secondly, news broke in November last year that the FDA is considering banning menthol-flavoured cigarettes in the US, which resulted in a punishing sell-off across the sector.

BATS has been one of the worst impacted by the latter. This is because BATS has recently acquired Reynolds, which has increased its exposure to the US market, while at the same time increased its debt levels to an elevated 4x EBITDA. The stock has fallen more than 50%, and de-rated to just 8x earnings. (Altria has also been appropriately punished for its overpriced acquisition of a stake in Juul).

Unfortunately for all the MBAs schooled in DCF math, assessing what the impact of a menthol ban will be on BATS has absolutely nothing to do with finance or valuation, but rather understanding likely smoker behaviour. The vast majority of investors are non-smokers, and so they tend to not understand what smoking is like, why people like it, and hence how smokers are likely to behave. I believe that, much as has been the case for the past 30 years, investors still misunderstanding smokers and their likely behaviour, and underestimating the resilience of demand.

Why have tobacco demand remained so resilient?

Smoking rates have declined over time in the developed world, but there remains a resilient minority of the population that continue to smoke, despite having to fork out higher and higher prices, and grapple with an ever widening variety of restrictions on where they can smoke, not to mention growing social stigma. And smoking rates still remain quite high in the developing world. Why has smoker behaviour been so resilient in the face of so many well-funded efforts to stamp it out? And why haven't regulators simply banned cigarettes altogether?

Let's start with the first issue, and how it pertains to the menthol ban. Here is the reality: smokers smoke for the nicotine, not the taste, and they smoke for emotional reasons rather than purely rational ones (and suffer terrible withdrawals when they stop). Because nicotine is highly addictive and the brain continuously craves more, smokers' bodies learn to ignore the foul odours and taste in order to get their fix. Furthermore, smokers learn to enjoy the taste of whatever source of nicotine they have become habituated to, as the taste becomes associated with the pleasant cognitive-emotional effects of nicotine. It happens in much the same way people come to like the taste of beer, despite often finding it distasteful when they first try it. If you drink beer, you probably don't notice the stench of booze either (non-drinkers, however, do).

Most smokers will tell you they enjoy the taste of their brand of cigarettes, and generally find other brands distasteful. However, this preference is a subjective illusion that reflects the specific tastes/blends to which the smoker has become habituated. If a smoker's preferred brand is unavailable, they will smoke an alternative brand instead of going through punishing nicotine withdrawal symptoms, 10 times out of 10 (with the exception of the minority on the verge of quitting anyway, where this becomes the final push for them to do so).

When smokers switch brands, they will initially not particularly like the taste. However, if they keep smoking that new brand long enough, they will eventually come to like it. And if they then try to go back to their old brand, they will find it distasteful. I am an on-again off-again smoker myself, having reacquired the 'habit' (read, addition) during my time living in Jakarta, which I've struggled to permanently kick. I smoked Indonesian A-Mild kretek (clove-tobacco cigarettes produced by Sampoerna, majority owned by Philip Morris) for a while, and enjoyed their sweet taste. Marlboro, by contrast, seemed harsh and distasteful.

However, somewhere along the line, associated with various failed efforts to quit, I switched to Marlboro (I probably stole a friend's Marlboro when I restarted). It wasn't long before I not only enjoyed Marlboro, but also found A-Mild almost unsmokeable. When I tried one occasionally, I was puzzled by why I liked them for so long. The answer is because you become habituated to a particular taste. If you like beer, what would you do if your favourite brews of beer were banned? Would you stop drinking beer? Or would you find a more distasteful brand to drink? Probably the latter, and I'm sure that given enough time, you'd come to quite like it as well.

In my judgment, if a menthol ban is implemented (and it is still just a proposal at the moment, and will also only impact the US market - at least for now), I believe the most likely outcome will be that the majority of menthol smokers will switch to non-menthol varieties in order to get their nicotine fix, and that in time, they will then come to prefer non-menthol cigarettes. By comparison, the market is currently pricing BATS as if most menthol smokers are likely to simply quit - an unlikely scenario in my view. I personally can't stand menthols, but if non-menthol cigarettes were banned, I would switch to menthols, and would come to enjoy them. If I'm right about this, it is quite likely that BATS's shares are heavily oversold at present, although they could initially fall further if the menthol ban is confirmed (currently it is just a proposal).

A lot of this also begs the question of why smoker behaviour is seemingly so resilient. Why don't people just quit, given punishing taxes, ever-more stringent restrictions, and widespread health warnings, and now their favourite flavours being banned? A big part of the problem is that non-smokers tend to focus only on the side-effects of smoking, while ignoring the main course - the benefits people get from nicotine. It's like focusing only on the hangover and asking, why to people drink?

Nicotine has complex effects on the brain and body. It releases dopamine and other neuro-chemicals that result in feelings of alertness, focus, motivation, and a sense of well-being. It also speeds up metabolism, and releases stored fat into the bloodstream. This increases energy levels and assists with weight loss. Nicotine, in short, is a very strong stimulant - a turbocharged form of caffeine. Despite what many people (most of whom have never smoked) believe, these are not illusions. It is not a placebo effect. The effects are real and measurable. Double-blind tests have been done which show groups dosed with nicotine outperform in both mental and physical tasks vs. control groups. Nicotine use is also proven to be associated with lower incidences of neuro-degenerative diseases such as Alzheimer's and Parkinson's disease (look it up; the science is undeniable - it's just not advertised).

Much like alcohol use, human beings have been smoking tobacco for thousands of years. There is a reason for this. The idea humans have smoked for millennia due to some collective delusion about non-existent psycho-cognitive effects is absurd. Nicotine has fallen out of favour in modern society due to a very-effective culture war waged by activists, and their PR offensive has been so successful that they have managed to convince most people there are no benefits associated with nicotine use - only costs. That is wrong. There are benefits (short term) and costs (long term). Nicotine makes you feel better in the short term, but living many decades heavily dosed on stimulants is not good for your long term health. It ages the body faster.

Unfortunately, we don't really know how bad the long term health consequences truly are (they are almost certainly negative, but how negative is less clear), as a lot of the 'science' around the health effects of tobacco use has become increasingly tainted with activist bias. Smokers putative 'black lungs' is a myth - those conducting autopsies cannot tell if someone is a smoker or not by looking at their lungs, unless they died of lung cancer (cancerous tissue turns black). The photo comparisons featured in anti-smoking ads and on smoking packets are between a smoker with lung cancer, and a non-smoker with healthy lungs - a dishonest comparison, and one where activism has clearly crossed the line into outright propaganda.

Lab experiments have also consistently failed to induce cancer/tumours in lab rates from heavy exposure to tobacco smoke, despite repeated attempts and huge research budgets. Consequently, claims that smoking 'causes' cancer and other health consequences are based purely on statistical correlation, rather than provable causation (i.e. smokers have, on average, higher rates of lung cancer than non-smokers; about one in eight lifetime smokers develop lung cancer; seven out of eight do not). But as is well known in the area of statistics, correlation is not causation.

Smokers likely, as a group, exercise less, and pay less attention to their long term health. They also skew poorer socio-economic groups these days, and the poor have lower life expectancies even controlling for tobacco use. The initial link between smoking and lung cancer established in the 1950s was based on correlation studies - lung cancer diagnoses had risen alongside rising smoking rates, but with a 20-30 year lag. But lung cancer rates have been rising in the developed world since the 1950s despite falling smoking rates, which is the opposite of what one would reasonably expect if smoking was the primary cause, suggesting that the initially observed correlation could well have been caused by something else, like rising emissions from automobiles alongside urbanisation, or other industrial emissions, for instance, growth of which happened around the same time. Diagnosis has also improved. Claims that smoking reduces life expectancy by 10-15 years also don't jibe with the fact that many countries with high smoking rates have above-average life expectancies. Smokers that exceed their life expectancy are still included in the 'smoking-related deaths' statistics.

This may seem like a digression, but the truth is that the subjective experience of many smokers is that the impact on their health is much more benign than the catastrophic consequences generally promised, and this creates an intuitive distrust of a lot of the ever-more-extreme health warnings, to the point where people just switch off and ignore them (by way of analogy, activists have recently succeeded in requiring cancer warnings be placed on coffee cups in California - if you're a coffee drinker, will this deter you?).

The point is, smokers smoke because of the psychological-emotional benefits they get from nicotine use, which they feel outweigh the risks they are taking. And it is hard to quit because there is a real cost associated with doing so (not just withdrawal symptoms, which are severe but short-lived, but a permanent reduction in energy levels and cognitive-physical performance, as you stop 'red-lining' your system on stimulants all day). If the bears are betting on a significant decline in smoking in the US following the menthol ban, I think they will be proven mistaken.

But where is all this regulation heading?

The other important question is, where is all this leading from a regulatory standpoint? Perhaps we are heading towards a complete ban on tobacco use - particularly now that there are campaigns to ban e-cigarettes as well? In my view, this is a textbook case of the toxic incentives associated with activist excess (discussed here more generally, and what has happened specifically with e-cigarettes here).

As is discussed in the former article, every hero needs a villain, and tobacco control activists need something to rail against, or they will be out of the activist business. For this reason, they have no incentive to have smoking completely banned, as it would put them out of a job and result in them being de-funded. This is probably why, despite well-funded anti-smoking activist groups throughout the world, we are yet to see a single instance of an outright ban being advocated for.

It is also likely why these activist groups are opposing e-cigarettes. They argue that there is an 'epidemic' of youth vaping in the US, but they fail to mention that unless and until there is any proof that vaping occasions harm, it cannot reasonably be emotively called an 'epidemic' (there is no such evidence at present, and the unbiased evidence that does exist suggests any harm is likely to be an order-of-magnitude less than conventional cigarettes). Nevertheless, these organisations have no incentive to base their activism on objective science. E-cigarettes are a threat to their basis for activism - if a 'safe' cigarette is invented, they will be out of the activist business. So they will continue to oppose e-cigarettes, and advocate for their heavy regulation and taxation (if not outright banning), regardless of what is fair and reasonable and what the science says.

Investors have viewed this regulatory impost on e-cigarettes negatively. New-generation products were seen as a potential growth opportunity for the industry, and a potential avenue to stabilise declining volumes, but regulators are now clamping down on that promising growth avenue. However, regulatory restrictions have been big tobacco's best friend over the past 30 years, and the actual truth is that vaping poses a major risk to big tobacco. Investors are making the typical mistake of focusing on the demand-side instead of the much-more-important supply-side.

Without regulatory restrictions, we would likely continue to see an explosion of new e-cigarette brands and varieties, as well as continuing device innovation. This would re-fragment the industry, and undermine the fabulous supply-side dynamics that currently exist. Furthermore, if marketing were allowed, you'd see a lot of marketing encouraging smokers to switch.

Fortunately for big tobacco, regulators are once again riding to the rescue. By clamping down on e-cigarettes, and taxing and restricting their use, they are likely to perpetuate the ongoing dominance of conventional cigarettes, and keep the industry highly consolidated and profitable for big tobacco (their activism will, of course, contribute to millions of unnecessary premature deaths, but that's ok, because they will get to keep their jobs). Consequently, my inclination is to be more positive on the outlook for the tobacco companies as these clampdowns happen, not less. But markets have taken the exact opposite approach, hammering the stocks on account of fading growth prospects.

One area where I do disagree with the bulls is that I have much less confidence any decision to ban or not ban menthol will be based on hard science (Morningstar, for instance, has argued that the science doesn't credibly support the notion it will reduce smoking rates, so there is a good chance the petition will fail). I believe they are right about the science, but tobacco industry regulatory policy has long since ceased being based on objective science. I think there is a good chance the ban will happen.


All told, I think investors are overestimating risks, and the insights outlined above draw as much from my experience as a smoker (and my research related to that), as well as trying to understand some of the grievance-study excesses we have been witnessing on US university campuses, than as they do from financial analysis. At least on the mental models I'm using, at 8x earnings, BATS looks oversold to me. I have been buying in the 25 GBP range.


DISCLAIMER: The above is for informational/entertainment purposes only, and should not be construed as a recommendation to trade in the securities mentioned in this article. While provided in good faith, the author provides no warranty as to the accuracy of the above analysis. The author owns shares in British American Tobacco and may sell this position or buy more shares at any time without notice.