...the report says “the smoking ban is demonstrably the most significant cause of pub closures”.
“While there is significant variation in the trajectories of the pub estates before the ban there is an almost total correlation between the three GB lines after the ban. This indicates that they are affected by a strong common factor — the smoking ban.
“The correlation is in fact so close that the trend line for the three countries is identical.”
The report predicts pub numbers will continue to fall, with another 1,700 to close in England before the fourth anniversary of the ban in July 2011.
Using CGA’s widely-accepted statistics for the net figure of pubs closing, the common trajectory shows closures accelerating after the first year of the ban in each country — from between 0.5% and 1.2% in the first year to between 3.7% and 4.4% in the second year.
Scotland lost a further 3.7% of its estate in the third year.
CR Consulting blames the decline on the loss of sociability in pubs where smoking has been banned.
“With smokers being moved outside, the price premium [in pubs] can no longer be justified [by drinkers] so more people drink at home,” it maintains.
Isn't it interesting that exactly the same sort of statistical analysis that was used to show that smoking causes lung cancer, and that passive smoking poses a significant health threat, is now being used to show that smoking bans cause pub closures? Correlation equals causation, it seems.
I suppose that, if nothing else, it throws down a challenge to the antismoking statisticians: If you believe that the statistical association between smoking and disease is very strong, then you must also believe that the association between smoking bans and pub closures is at least as strong.
But, who knows, maybe antismoking campaigners might say, "Yes, we accept that smoking bans are driving pubs out of business. But that's a price worth paying if a few lives are saved."
They might even go on to say, "It's a good thing that pubs are going out of business. It's not just the smoking that's killing everyone. It's also the drinking. And sitting around talking all evening. The death of the British pub will bring a tremendous improvement in the health of the British population."
After all, if health is all that matters, who cares how many pubs close?
Needless to say, antismokers have yet to say any such thing. They blame the closure of pubs on everything but the smoking ban. It's the consequence of supermarkets selling cheap alcohol, and undercutting pubs. Or it's the result of the credit crunch
and economic recession. Or it's persistent bad weather that's been keeping the punters away.
In the absence of a causal mechanism, it's perhaps understandable that research on the ill-effects of smoking relies heavily on correlations between the incidence of smoking and cancer, or smoking and heart disease.
But is the same approach necessary when considering the effect of smoking bans on pub trade? For while it is not possible to interrogate a cancerous lung to learn how and when the disease started, it is possible to interrogate pubgoers to learn when and why they stopped going to pubs. And either a great many smokers will say that they stopped going to pubs because they could no longer smoke inside them, or they will not. Why doesn't anyone simply ask the erstwhile pubgoers themselves?
There might be a few practical problems with this. How does one locate ex-pubgoers? If current pubgoers are surveyed, they might well all turn out to be composed of happy non-smokers, delighted with the smoking ban.
From such a survey, or series of surveys, it will emerge that some number of smokers ceased going to pubs for some variety of reasons, and that some number of non-smokers also stopped going for a variety of different reasons, and that some number of non-smokers started going to pubs.
My own guess is that those results would show that a large fraction of smokers stopped going to pubs after smoking was banned in them, because they no longer enjoyed the pub experience. And I suspect that a substantial fraction of non-smokers subsequently stopped going to pubs because their smoking friends no longer went to them. And maybe some fraction of non-smokers started going to pubs because they liked the new smokefree pubs. But that's just my guess, based upon my experience as a smoker, and knowing quite a few smokers.
If I was the CEO of some chain of pubs, I would have commissioned such a survey a year or two ago, because I would have wanted to know why I was losing so many customers.
There's no need for any fancy statistical analysis of trends over time. Just ask the customers.