Tobacco taxes are NOT the most effective tobacco control policy (as actually implemented)

Today NPR Morning Edition interviewed Ken Warner on the impact of the 1964 Surgeon General report.  During the interview Ken repeated the mantra that increasing taxes is  the most effective tobacco control policy.
There is a strong consensus that people smoke less as the price increases, with a price elasticity of -0.4 for adults and -0.65 for adolescents.  What this means is that a 10% increase in price leads to a 4% reduction in consumption by adults and a 6.5% reduction among youth.  Put another way, the effect of the price increase will depend on how big it is.
This also means that it is possible to compare the relative effects of different tobacco control policies to each other by computing the size of the price (tax) increase that it would take to achieve the same effect.

  • Smokefree workplaces reduce cigarette consumption by 29%.  Achieving the same effect with a price increase would require a tax large enough to increase the price by 73% (.29/.4).  Based on an average cigarette price of $6.03 per pack, it would, therefore, require a $4.37 tax increase to achieve the same effect (among adults).
  • Strong graphic warning labels would reduce smoking by 16%.  Achieving the same effect with price would require a $2.39 tax increase (among adults).
  • Strong media campaigns, as exemplified by the Legacy Foundation's "truth" campaign, reduced youth smoking by 22%.  Achieving the same effect with price would require a $2.04 tax increase (among youth).
  • An R rating for smoking in the movies would reduce youth smoking by 18%.  Achieving the same effect with price would require a $1.67 tax increase (among youth).

For comparison, the largest federal cigarette tax increase in history was the $0.62 increase in 2009 to fund the State Children's Health Insurance Program (SCHIP).  There have been larger state and local increases, but I am unaware of any that exceeded $1.67.
Another important issue about tax increases, especially small tax increases, is that the tobacco companies can adjust pricing (sometimes up, sometimes down) to adapt to the changes (Alamar et al, Chaloupka et al, Gilmore et al).
This is not to say that increasing tobacco taxes is a bad idea.  Tobacco tax increases are an important element of any tobacco control program and do reduce smoking and other tobacco use.  But, as implemented in the real world, they are not the "most effective" policy intervention.


Tobacco tax increase

Dear Dr Glanz,
This information is GREAT. 
To know that tax increase is not the most powerful tool but there are others which are even more in reducing initiation and increasing quitting. It is a very helpful finding. 
I have all along felt and also discussed with Prabhat that how can tax-increase ever work with bidi smokers, for example or Gutkha users since they are so cheap to begin with. How much tax increase can you do? 100%, 200%?. Even then price will stay so less although the poor will feel some pinch but those addicted will cut on other things like purchasing  less milk for the family or not buying milk at all. In stead would it not be more helpful to find it from the bidi or gutkha users themselves what will make them give up the use. We can also ask them specifically; what should be the price of bidi before someone like you will decide to give it up. 
Once again many thanks for this information.
mira aghi; height: 8px; opacity: 0.3; width: 20px; background-position: initial initial; background-repeat: no-repeat no-repeat;">

Stan, Where do you get your

Where do you get your stats on the impact of smokefree workplaces (reducing cig consumption 29%) and the other stats as well?
I wish to send this to others, but they will want to know your references on smokefree workplaces and pack warnings especially. Thanks!
Stephen Hamann
Bangkok, Thailand

Thanks, Stan; Sorry, I didn't see the links in the bullet points

Thanks, Stan; Sorry, I didn't see the links in the bullet points
S. Hamann


Click on the links for each intervention (at the beginning of each bullet).