Sorry, but Anna Gilmore is still lying about economics

by Carl V Phillips

Anna Gilmore, an anti-tobacco activist and “researcher” at the University of Bath (they must be so proud!) is primarily known for doing junk epidemiology but has also branched into junk economics.  (The definitive collection about her can be found at Chris Snowdon’s blog, running a search for her name.)  It is not entirely clear whether she is in the subset of lairs who know they are disseminating disinformation or the subset who claim to be experts but simply have no idea what they are doing.  My assessment, based on her previous foray into economics (see my analysis of it) is that while she seems to be intentionally misleading about epidemiology, in economics she probably never read so much as a first-semester textbook before writing her unintentional comedy.

Her new “contribution” is “Understanding tobacco industry pricing strategy
and whether it undermines tobacco tax policy: the example of the UK cigarette market, by Anna B. Gilmore, Behrooz Tavakoly, Gordon Taylor, Howard Reed, about the impact of cigarette tax increases on retail prices.  It was widely laughed at by commentators because the part of her press release that was usually highlighted in the news was that smokers react to price increases by shifting to cheaper products (that is, from premium brands to brands with lower wholesale, and thus after-tax, prices).  The observations were along the lines of “hey, look, Anna Gilmore seems to have learned basic economics!”  But she actually did not.

I do have to give credit to the health reporters on this one.  Not only did they emphasize the one simplistic observation that she got right, but some fixed her false claim “tobacco tax increases are the most effective means of reducing tobacco use” (by “tobacco” she means cigarettes, of course — the standard casual anti-THR lie is thrown in there), making it accurate with “one of the most effective”.  (Basic education about the risks from smoking is, by a huge margin, the most effective way to convince people to not smoke.  Taxes come in third, after promotion of harm reduction.)  Unfortunately, further down in the articles, most of the reporters went on to report her innumerate claims also.

The core of her innumeracy — failure to understand first-semester economics before trying to write about economics — is claiming that the setting new prices after a shock to the market (in this case, a new tax increase) is a matter of complicated volition rather than simple market forces.  She attributes manufacturers’ actions to various nefarious motives, which is perhaps just political lying, but probably is because neither she nor her coauthors understood what they were writing about (and did not stop to think that perhaps University of Bath has an economics department, and that any decent student in it, let alone any professor, could have explained it).

The key is the following observation :  When taxes are increased, the retail price of premium brands increases by more than the tax increase (i.e., there is a wholesale price increase also) while the price of low-end brands increases by less than the tax increase (i.e., there is a wholesale price reduction).  I am going to assume that this simple factual claim was true — not a foregone conclusion due to Gilmore’s history of publishing out-and-out false claims, but since it is exactly what an economist would predict it seems like a safe assumption.

To predict that the manufacturers would lower wholesale prices (i.e., eat some of the tax themselves) on the highly price-competitive cheaper brands is just a matter of applying the most common supply and demand curve analysis (see, e.g., Wikipedia).  When there is an increase in the cost of supply (a tax has exactly the same effect as, say, an increase the price of the tobacco leaf), it is shared between the consumer (higher prices) and supplier (prices not quite so much higher as to fully compensate for the cost increase).  Usually most of the cost is borne by the consumer, but typically not all of it.  Basic stuff.

[For those who want to delve a bit deeper:  The consumer pays most of the cost because the supply curve is much closer to horizontal than in the typical picture like the one in the link.  For those who want to delve even deeper, and to argue that long-run supply curves do not actually slope upward at all, despite the usual picture, it gets a bit more complicated.  We still observe that cost shocks are shared by consumer and supplier.  The easiest explanation is that in a highly price-competitive market, at least one supplier is going to have an upward-sloping short-run demand curve over the relevant range — because they have already invested the capital in their manufacturing and distribution capacity and so are better off lowering prices to somewhat offset the reduction in demand, just like in the standard graph.  If this is true for even one supplier in the competitive market, the others will also have to lower their prices or lose revenues to the cheaper competitor.]

The extra price increase on premium brands is even easier to explain.  The manufacturer of a premium brand has some “market power” because consumers have loyalty and will not switch to a competitor to save a few pennies (and there are relatively few competitors, so they are all doing it).  This lets the supplier charge more without losing so much business that the reduced quantity erases the increase in net revenues per unit.  The ultimate in market power is a monopoly; with no competitors at all, the only constraint on a monopolist’s prices is consumers reducing their consumption, and so they can command a lot of net revenue per unit.

What happens in a market power situation when there is a cost shock like a tax?  The consumer becomes less sensitive to any given increase in net revenue.  It is easy to see:  Imagine a good with a production cost of $1 that the manufacturer charge $1.20 for because that is the price that maximizes total net revenue; raising the price to $1.21 would drive more than 5% of the customers to cheaper competitors, and so the net effect would be a loss.  Now imagine a new tax of $1 on everyone, including the competitors.  A price rise to $2.20 maintains the old margin, but now an increase to $2.21 is a much smaller change in the base price, and so consumers will react less.  Less than 5% will be lost due to such an increase, but the impact on the supplier (making an additional 5% cent per unit) is still exactly the same, so the additional increase is a good idea.  So it is perfectly predictable that a tax increase would lead to an additional wholesale price increase.  (Note that the total net revenue of the supplier might still drop due to the reduced demand that the big price increase causes, but this does not affect how to optimize the situation as it is.)

Snowdon discovered that Gilmore’s close allies have been failing to understand this since at least 2008:

“The hypocrisy of the industry knows no bounds,” said Deborah Arnott, the director of Action on Smoking and Health. “While complaining bitterly about tax increases, these companies have been raising the price of cigarettes to fill their own coffers while hiding behind the screen of tax rises.”

Undoubtedly in the last five years, in response to such silly claims, someone explained the basic economics to them.  Of course manufacturers do not like tax increases — they decrease demand.  Of course they raise their prices (on premium brands) when there is such an increase — it only takes the two paragraphs above to explain why.  So perhaps the current claims are intentional disinformation rather than mere unforgivable ignorance after all.

It is hard to tell exactly what they want people to believe, since it is hard to be precise when arguing nonsense.  But they try to suggest that these simple predictable effects of supply and demand, observed in a market with multiple competitors who want to take business from each other, are some complicated unified plot to manipulate their customers.  Somehow every bit of it, both the competitive price decreases and the premium price increases, is part of a grand scheme to do something other than simply make as much profit as possible given the taxes and what consumers want.

Ah, to have the easy life of working in the tobacco control industry.  Archeology is made much easier if you just take the Jewish Bible as fact and do not have to bother with all of that pesky carbon dating and digging in the ground.  Economics is similarly easy if your bible lets you avoid bothering with all the theory and empirical evidence from economics.

So why did I cover this junk science about cigarettes in the anti-THR lies blog?  The answer is that basic economics offers the strongest arguments in favor of THR.  Therefore any attempt to mislead people about the basic economics, and to claim that the tobacco/nicotine market is somehow mysteriously different from markets for all other goods that consumers want, tends to aid the anti-THR agenda.  The main message of this analysis is that the market was working just like it does for any other consumer good, with price changes reflecting exactly the same profit-maximizing forces that we would expect in normal consumer markets for phones or almonds or coffee.

So long as we get the economics right, the case for THR is even stronger than the usual arguments that are based on health effects alone.  More about that soon when I widely release the paper that I am now circulating.