Monday, 11 March 2013

Rationale for Booze Taxes (wonkish)

The joys of teetotalism become apparent in Manitoba?
 This city's rightwing tabloid, The Winnipeg Sun, has had quite a few cover stories about the horror of a rise in Manitoba booze taxes. Some writers, like hard-right columnist Tom Brodbeck, are even suggesting going the route of Alberta by privatizing liquor stores. The desire for cheaper beer from Sun writers is not too surprising given the effects of intoxication on political beliefs. Heck, the provincial Conservatives might've won a landslide majority last election if only people showed up to the polls hammered! 

Kidding aside, I can't really "feel your pain" as a nondrinker who buys pop at bars. But the rise in the booze tax does represent an opportunity to talk about the economic theory of Pigouvian ("sin") taxes and the prospect of a tax shift.

The basic idea is that production/consumption of some good, like booze, has a negative effect on others (economists term this a "negative externality") - like injury from drunk driving or bar fights. A study from the Ontario Ministry of Finance details this, as well as noting that Alberta (with a privatized liquor system and lowered booze taxes) has the "highest rates of alcohol-related problems, such as drunk driving fatalities" of Canada's provinces. 

 The solution some economists propose to this "external cost" is to "internalize" the cost with taxes.

This simple idea gets trickier when presented graphically or mathematically. Neoclassical economists (that is to say, mainstream ones) have a strong desire to formalize stuff like this and frame it in terms of marginal analysis.

The idea is that in a private market, the production (and consumption) of liquor is determined at the level where the "private marginal cost" equals the "marginal benefit" of liquor sales. The "marginal benefit" is the joy ("utility") one gets from drinking one more bottle/glass/case (unit) of beer, whereas the "marginal cost" is the price to consumers or producers of buying/making one more unit of beer.

Free Market Liquor Output.

Image Constructed by "The Analyst"
The "marginal damage" is the cost of one more unit of beer to outside parties. So this is the cost of drunk driving accidents or people kept up at night by folks who've had a few too many pints at the local bar.

Negative external of beer

Image Constructed by "The Analyst"
This, in turn, raises the true cost to society of producing/consuming one more unit of beer ("social marginal cost"). The idea goes that the "efficient" level of output for society ("socially efficient output") would result from a higher price.

"Socially efficient" level of
beer production/consumption.

Image constructed by "The Analyst"
A well-designed tax, however, can shift the output level to a "socially optimal" level.

"Socially optimal" level of output
generated by Pigouvian tax.
Image constructed by "The Analyst"
Now, I know of a conservative-leaning U of W economics alumni who's argued (in the case of carbon taxes) that the "socially optimal" level of output isn't known. This kind of perfectionist quibbling is quite common among those with an bachelors degree in economics, perhaps because not enough time is spent examining how utterly imperfect the real world is next to shiny models and how one needs to work off of very rough estimates/"good enough" judgment in real life policy-making.

But the idea that you can limit harmful activities, from consumption or production of goods, seems like the key, policy-related point one get out of this elaborate economic model. A lot of economists even propose shifting from income taxes to Pigouvian taxes under the notion that "we'd like more income" and like less of various goods with negative externalities.

The problem is that Pigouvian consumption taxes are regressive (like user-fees). They hurt those with less the most

Which is where an idea I'd support for tax reform in Manitoba comes in. Our province doesn't have the best record when it comes to environmental sustainability. Instituting a carbon tax, of let's say $100/tonne (guesswork - with more studies, a more "rigorous" number could come up) and reducing income taxes for the bottom 60% would be an interesting way to make the tax system more steeply progressive and more facilitative of a greener, less fossil fuel dependent, future.

Would you like to support this blogger? Consider making a donation or checking out their shop!

Also consider liking this blog on Facebook!  


  1. I have no issue with government's collecting taxes via general sales tax or even a special sin tax.

    However, the issue is: Do we need the government to be the actual retailer of liquor? If the answer is yes, what are those reasons? And if alcohol must be sold this way, why not tobacco?

    1. Might answer question more fully in future post. Answer to "why not tobacco" is that the provincial Gov't doesn't want to spend the political capital fighting private vendors.

  2. They already fight the demand from private vendors of alcohol who want a bigger slice of the market with no political loss of capital.

    The government makes $250 million for sales from the Liquor Commission. Hard not to think that tobacco might not be higher numbers and you could make a claim to be doing for health reasons.

    1. It almost always requires infinitely less political capital to "fight" (ignore) calls for change from business interests than it does to fight against vested interests by introducing new changes.

  3. @ Dobbs - Beer & smokes go together! :)

  4. I have no problem with MLCC employees making a decent wage. Prices in Alberta aren't much different than here, factoring out the PST, and unless you're shopping at a Superstore, the selection can be very limited in most private retailers.

  5. The ban on smoking at the beaches this past week was billed as good for the kids. Why not only allow tobacco sold with alcohol.

    & double: if the goal is higher wages, then tobacco sold by workers at a government store fits the bill as well.

    And if doesn' argument about what the difference is between alcohol and tobacco remains.

    Or is this an example of a stand pat government, looking to get more capital from the their store without explaining and contrasting why the government needs to be the retailer?

  6. I don't understand your argument. Why do you insist on claiming that alcohol and tobacco are the same product, or share some historically similar marketing process? Or should be sold in a similar fashion? Why don't you insist that alcohol and big mac's be sold in a similar fashion?

  7. Big Macs can't be sold to youth nor are subject to other restrictions in how they are sold.

    Interestingly, the U.S. has a law enforcement branch called Alcohol, Tobacco and Firearms. Big Macs not included.

    Might not be the same product but they are an area of special concern to the government.

    In Canada firearms are regulated federally. Alcohol and tobacco are a shared regulation but how they are sold lies with the provinces.

    I want to know why and still haven't heard the reason the province thinks it has to be the retailer of alcohol. If it is a good reason, it probably also applies for tobacco.

  8. Should have read Big Macs can be sold...

  9. OK, I agree with you. Smokes should be sold in gov't run stores. Less chance that kids will have access. People might be inclined to quit or cut back if the purchasing process is a bit more cumbersome. See, that was easy!