Friday, 11 March 2011

Cause or effect?

I recently reported on the call from Alcohol Concern for alcoholic drinks to be confined to separate areas in shops. While I certainly wouldn’t say I’m in favour of this, taken in isolation it wouldn’t really make much difference to anything. Other countries (notably Australia) have similar restrictions and don’t seem to be notably abstemious societies. Indeed, we drank more beer, and had a society in which alcohol was more “normalised”, when most off-sales were through stand-alone off-licences, or separate counters in supermarkets, when pubs closed for three hours in the afternoon and at 10.30pm during the week, and across large areas of Wales were closed all day Sunday.

Rather than being a cause of increased at-home drinking, isn’t the rise of alcohol sales in supermarkets, and the increased use of price promotions, primarily a result of a growing market which has over the years become potentially much more valuable for retailers? Business reflects changes in society, it doesn’t in general drive them.

Another factor, of course, is that the rise of beer drinking at home is closely linked with the growth in ownership of cars and refrigerators. Regardless of price, the working man of 1955 would have struggled to get a slab of Carling home from the outdoor, and would have had nowhere to keep it cool once he had, whereas, even in those pre-lager days, most pubs had naturally cool cellars for their draught ale.

I’m also far from convinced that having alcohol on general display increases overall consumption to any significant extent. Yes, it may encourage people to buy particular wines or beers that are being promoted, but I doubt whether it very often persuades people to buy a bottle on impulse when they wouldn’t otherwise have bought any at all.


  1. Have you considered that the linear view of causality may in fact be flawed?

  2. A bit deep for a Friday afternoon that, Cookie ;-)

  3. "across large areas of Wales [pubs] were closed all day Sunday"

    I think that should read

    "across large areas of Wales [pubs] were officially closed all day Sunday"

    well according to my dad!

  4. Consumer behaviour might not be as simple and rational as you might think. (In fact, and I hate to break it to you, large areas of life mign not be). How about this: "Numerical signs, such as "2 for $2" or "Limit 12 per person" make you spend twice as much as you planned" - according to those smart folks at Cornell University (summary, full text )

    Are you so sure that supermarkets plan their displays to merely direct your choice? Rather than to trick you into buying stuff that you wouldn't have otherwise?

  5. "In fact, and I hate to break it to you, large areas of life might not be"

    You can patronise for England, can't you?

    But I would say promotions of various kinds overwhelmingly affect (a) product choice, (b) location of purchase and (c) timing of purchase rather than impacting on the total amount of stuff people buy. It varies between products, of course – promotions are unlikely to make me buy more washing powder over time, but they may encourage me to buy more chocolate.

    Clearly there is a constant interplay between information available and purchase decisions, and there are plenty of things people buy that they wouldn't really miss if they had never heard of them, but you are (again) being highly patronising if you believe that people are routinely being duped into buying stuff they neither want nor need.

  6. Patronising? No. Cynical - Yes.
    But whatever. You're surely not denying that impulse purchases over and above what's needed, are routinely made by most, if not all, shoppers?

  7. Life would be very dull if we only bought what we "needed" and never acted on impulse.

    But, as I said in the previous comment, for products that are not for immediate consumption, I doubt whether supermarket promotions over time have much effect on the total amount purchased.

  8. So doubt it. But let me get this right, you're saying that shoppers have (even if they don't know it) an annual (or lifetime) quota of booze (for instance) which they'll buy. And they'll buy this (and no more) regardless of what producers or retailers do to stimulate purchasing? Really?

  9. I never said that, did I? Obviously over a lifetime people's alcohol consumption will be affected by factors such as the general price level of alcohol, its social acceptability and whether they move in circles where alcohol consumption is commonplace.

    But, as I said in the original post, I doubt whether displaying beer and wine next to the veg makes much difference to overall consumption. And neither will the occasional "3 for 2" offer - people will just buy less next time.

    OK, if price promotions (which wasn't the subject of the original post anyway) lower the effective long-term price of alcohol, then more will probably be consumed. But it's generally accepted that the price elasticity of alcohol is relatively low.

    I don't think it's commonplace for someone to go into the shop to buy four cans to drink over two nights, see that they're on a BOGOF, buy eight and then drink four each night.

  10. Well, no, price promotions aren't the subject of your original post, but since you've brought the subject up it's worth mentioning that the Treasury has worked on an estimate of price elasticity for booze at around -1.0, i.e. a price reduction of 1% would be expected to increase demand by about the same amount. Although it's probably more complicated than that.

  11. I saw that re the Treasury, but it sounds like an overestimate to me. This review suggests that price elasticity for all alcohol categories tends to be well below 1, with beer lower than wine and spirits:

    • The demand for alcohol is price inelastic;
    • The own-price elasticity of wine and spirits are similar;
    • Beer is the most inelastic category, and the elasticity of beer is different to wine and spirits

    Of course alcohol is not a homogenous substance like oil, and if prices rise people have the option of shifting to cheaper drinks within the category, brewing their own or switching to illegal drugs.

    That document suggests that easier availability of illegal drugs tends to increase alcohol price elasticity.

  12. I saw that as well - isn't google great? I don't see much point in cherry-picking ping-pong. But I should point out that the mean of UK estimates quoted in that review are rather larger than 0 and not much larger than 1. I think this is what they usually call "relatively inelastic", rather than "inelastic" - there's an important difference there - but then economics was never a strong point for me. Bottom line - no, we wouldn't buy eight cans rather than four because of BOGOF promos on that basis of price alone , but we might well buy five. Or 9 rather than 8. If the promo was presented in the right way, however - with numerical signs, for instance - we might buy rather more than a simplistic view of the economics would suggest. Which is where, I think, I came in.


Comments, especially on older posts, may require prior approval by the blog owner. See here for details of my comment policy.

Please register an account to comment. To combat persistent trolling, unregistered comments are liable to be deleted unless I recognise the author. If you intend to make more than the occasional comment using an unregistered ID, you will need to tell me something about yourself.