Business Analytics Blog

Business Analytics Blog

Opinions expressed on this blog reflect the writer’s views and not the position of the Capgemini Group

Have smoking ban policies been effective across Europe?

Category : Figure It Out

When you live in a foreign country, you are constantly compared to the clichés associated to where you are from. Since I moved to the UK, people always act surprised when they realise I, a French person, don’t smoke. So I have wanted to look further into this habit: is it really a behaviour that can be associated to French people? Let’s have a look.

First I pulled data regarding proportion of daily smokers (over 16 years old) for European countries in 2002 and 2012 (or nearest years) – such data was available from the OECD for respectively 22 and 34 countries. The OECD created a heat map based on the 2012 data, which shows what the countries where people smoke the most are. The colour scheme is not great, yet we can see easily that France is in the norm.

The smoking rate of French people is just above the average compared to the rest of Europe (24.1% in France while the European average is 22.8%), far behind countries such as Greece or Croatia. But this map made me wonder, are we smoking less in France than in such other places thanks to the smoking ban?

Did banning smoking change behaviours in Europe?

To answer this question, I looked into which countries had enforced such a ban. Treated.com, a UK registered pharmacy, produced a timeline of when a smoking ban policy had been implemented in a country, with a distinction between a ban in public places, and a ban in restaurants and bars.

This highlights how recent this ban is, most countries having banned it over the last 10 years.

In order to assess whether there is a correlation between smoking ban policies and the proportion of smokers, I set up a three-step process, using R

1. I quantified my qualitative data, the smoking ban policies:

I was then able to compare this data to the ratio of daily smokers in each country

2. I then checked whether there was significant difference in the proportion of smokers due to the smoking ban policy (p-value of 0.01, thus less than 0.05 and significant)

3. Finally, I calculated the Pearson correlation coefficient

In other words, people tend to smoke more in countries with a more lenient smoking policy.

This result was intuitive, yet I was wondering what had been the actual impact of banning smoking: will the result be similar if we focus on how the proportion of smokers changed between 2002 and 2012 for the countries which enforced the ban within those dates?

Therefore, I reiterated this process with two different variables:

  • Change in the proportion of smokers
  • A redefined version of the smoking ban policy

1. I re-quantified my qualitative data, taking into account this timeframe:

I was then able to compare this data to the ratio of daily smokers in each country

2. This time, I could not conclude whether there was no significant difference between the change in the proportion of smokers and my “ban” variable (p-value of 0.59, thus greater than 0.05 and not significant).

3. Therefore, the correlation coefficient cannot be calculated in this case. This would suggest that the smoking ban policy can be used to highlight in which countries people smoke the least, but is not a driving force in reducing number of smokers.

 

This made me think, what else could drive this decrease? 

  • Advertising could be a cause for this, and there have been a number of studies regarding its effect on tobacco consumption, however they resulted in divergent opinions (Henry Saffer, ‘The Effect of Advertising on Tobacco and Alcohol Consumption’, 2004); therefore I decided to leave this element out of the equation;
  • The introduction of e-cigarette in Europe in 2006 (Consumer Advocates for Smoke Free Alternatives Association, ‘A historical timeline of electronic cigarettes’, 2016) could also drive this decrease in smoking behaviours. However, as of today, the effects of this recent phenomenon have not been entirely understood yet. For instance, a study published this year found out that, in England, there is no association between the use of e-cigarette and the rate of quit attempts (Beard, West, Michie and Brown, ‘Association between electronic cigarette use and changes in quit attempts, success of quit attempts, use of smoking cessation pharmacotherapy, and use of stop smoking services in England: time series analysis of population trends’, 2016).  This uncertainty surrounding the impact of e-cigarette explains why, as of today, it is hard to quantify how it affects smoking behaviours, thus why I did not consider this variable as part of this article;
  • The other key policy that could influence behaviour is tax; let’s see if this can be a driver of smoking behaviour.


Did tax change behaviours?

I decided to look into the overtime evolution of tax levy versus the proportion of smokers. This variable being more complex, the analysis has to be done country by country. Let’s see what happened in the UK.

1. To quantify the cigarette tax burden, I look into the annual retail price of cigarettes over time in the UK (provided by the Tobacco Manufacturers’ Association). Taxes have accounted for 73% to 80% of this price since 1990, which makes it the main driver of whatever impact the evolution of cigarette price had overtime.

Furthermore, I took into account the inflation for consumer prices, leaving taxes out of inflation as it does not depend on this – to give realistic comparison overtime.

2. Now the two variables are identified and quantified, we can check whether there is a significant difference.

3. This confirmation allows to perform a correlation analysis:

This strong negative correlation proves that a change in cigarette prices has a big impact on the proportion of daily smokers.

In conclusion, the smoking ban policy is useful in highlighting countries where people smoke less, but does not cause people to quit. Cigarette prices on the other hand, and therefore their main component the tax levy, is strongly correlated to people’s habits. Moral of the story? If you want people to stop doing something, don’t make it illegal, make it expensive.

About the author

Pierre Coupez
Pierre Coupez
As an Associate Consultant with a background in Business Analysis, I use a range of analytical techniques to help clients with better understanding their business and environment. I have experience spanning across Central Government, Consumer Products, and I am particularly interested in System Dynamics, optimisation and energy-related issues.

Leave a comment

Your email address will not be published. Required fields are marked *.