26/08/2013 § 2 Comments
The title of the post was going to be ‘Beer taxes cause child abuse’, but that over-states the case (and is inflammatory). Instead, let me just refer you to a paper I stumbled across and make a gentler point: taxes don’t mechanically change behaviour. They don’t pull a lever over here and have exactly the effect you think they will over there. Life is messy, people are complicated, and prices are just another variable.
The paper is here (pdf). The authors look at the impact of beer taxes on child abuse. The dependent variable is the probability of severe violence against children, and the main independent variable is the state tax on beer in the United States. Several other variables, including other alcohol policies, were included. The key results are in Table 1.
They find that the tax on beer significantly and sizeably reduces the probability of abuse by women. For men, on the other hand, the parameter is nearly zero for 1976 data (in the full model) and is positive and large in 1985. That is, higher beer taxes lead to child abuse.
Yes, this is just one paper. Yes, the results are ambiguous when you take into account gender and years and model specification.
But, it serves as a warning. I can take these results and tell a plausible story. Some father just wants to go home and have a few beers and get a nice buzz on. The price has gone up, so he has to ration his beers a bit more. He can’t afford that fourth or fifth or sixth one, the one that gives him the click*. Or, when he does, he doesn’t want that now-expensive buzz ruined by some snotty, whining kid. A few beers at home has turned into worrying about money, and he takes it out on the children.
Sin taxes aren’t mechanical. Raising prices does reduce consumption (of alcohol or cigarettes or fat or sugar or whatever). But what drives the social interest in personal behaviour is the harm caused, not the consumption itself. The link between prices and harm is not as clear-cut, not as mechanical. In fact, as this article suggests, it may sometimes be the opposite of our expectations.
Brick: It’s like a switch, clickin’ off in my head. Turns the hot light off and the cool one on, and all of a sudden there’s peace.
09/08/2013 § 3 Comments
I have been discussing with colleagues a recent article in the American Journal of Agricultural Economics, Zhen, et al. on Predicting the Effect of Sugar-Sweetened Beverage Taxes on Food and Beverage Demand in a Large Demand System. One issue we discussed was whether diet and sugar sodas were complements or substitutes. As a reminder, substitutes means that as one price goes up, people buy more of the other product. A complement means that that as one price goes up, people buy less of both goods.
Are they substitutes or complements? I would suggest that it’s an empirical question. Zhen, et al. mention research that finds that the two types of sodas are complements (Dharmasena and Capps, 2012, for those playing at home).
If someone is buying diet soda, I would not expect the prices to affect their decision within the ranges of price movements we usually see. Price may lead them to switch to other low-calorie drinks (water, tea) or other diet foods, but the significant calories in a sugar soda probably represent large dis-utility to the diet-conscious consumer. On the other hand, consumers of sugar drinks have already rejected the diet drinks and probably switch to other good-tasting drinks (energy drinks, flavoured milk).
Me, I can’t stand diet drinks and price won’t make me switch. I will drink water instead.
We could also think about the purchase scenario. When consumers buy soft drinks, perhaps they are buying them for a group: a family or a party. They buy both diet and sugar drinks to cater to the different tastes in the group. Higher and lower prices change the total amounts but have little effect on the ratio.
There are two methodological issues that can affect these sorts of estimates. First, there is significant correlation in the diet and sugar soft drink prices. Companies tend to price both types the same. That correlation would make it difficult (or perhaps impossible) to estimate cross-price elasticities (XPE). Secondly, a common assumption is that XPEs are symmetric. That is, the diet–> sugar XPE is the same as the sugar–> diet XPE. That may not be the case. A diet drinker may be reluctant to change, but a sugar drinker may be quite price sensitive. If we assume symmetry (which is common) then we won’t estimate the different XPEs.
11/06/2013 § 3 Comments
I mentioned my Turkish colleague before. Today, he is sputtering about the new law restricting alcohol. Like many, he makes the link between the legal rules around alcohol sale and consumption, and the push for religious restrictions. This new law indicates that Turkey’s government is moving closer to the religious powers, rather than focusing on keeping the government secular.
I can’t vouch for this post on the law change, but I draw your attention to this statement:
Since alcohol abuse is not really an issue in Turkey (a government study found that 83 percent of adults never even touch a drink and only one percent have a drink every day), at its heart the new alcohol law is a political move.
That ‘government study’ link takes you to a news article with this information:
You may think Turkey has a serious alcohol problem, but according to the Turkish Statistical Institute’s Household Budget Surveys, only around 6 percent of Turkish households consume alcohol. These results are consistent with the Institute’s 2011 Family Structure Survey, which found that 83 percent of Turkish adults never use alcohol. Most of the rest are casual drinkers; less than 1 percent said they drank every day.
I haven’t gone looking for the household surveys, on the assumption that it’ll be too hard to find the right statistic, so I’m accepting this as true (or at least true-ish): alcohol consumption isn’t particularly high in Turkey. As a result, the social costs of alcohol — health, employment, and criminal externalities — wouldn’t be particularly high. And so, we in the West can look at this law change and recognise that it is motivated by religious preferences.
Why can’t we do the same for New Zealand?
I’ve generally stayed away from the alcohol debate on this blog, because it crosses over into my paid work a bit too much. Not that I have to say much — Eric Crampton has done a bang-up job of dealing with the issues in a theoretically consistent way. Once you start working through the various economic analyses and the associated social and medical literature, a few things become clear:
- people drink booze because they like to, or because it makes things less bad. Either way, it makes them happier than the baseline
- alcohol is a ‘problem’ for a minority of people and/or on occasion. A few people are dipsos; the average adult occasionally gets blotto. Making the problem out to be more than that is disingenuous
- price is a stupid way to deal with the externalities. First, by definition addicts are insensitive to price. Secondly, the behavioural response of binge drinkers to increased prices is to cut out the moderate drinking sessions, not the harmful ones
- the research that shows otherwise is flawed. Crampton’s done the heavy lifting, so go see his work, but I’ll back him up. Poor assumptions, begging the questions, faulty parameters — embarrassing, really.
Just because the alcohol activists in New Zealand aren’t banner-waving members of the CWTU doesn’t make their desire to impose their preferences on the rest of the population any better. We can see it clearly on the other side of the world. Let’s be clear about it at home, too.
06/09/2012 § 4 Comments
Price-based policies are all the rage. A core idea is that we need to get prices ‘right’, and then market forces and consumer rationality will get us to maximum satisfaction. But, as Eric Crampton likes to remind us, slopes are often slippery.
Rather than worry about the coefficient of friction, however, I want to lay out how price-based policies actually work and what that means for policy and welfare.
First, we start with how consumers respond to prices. The technical term for this is ‘elasticity’, but let’s just go with the simpler ‘response’. There are three responses when the price of some item goes up:
- the item itself — when the price goes up, we usually buy less
- income response — because spending on the item itself has changed, the money we can spend on everything else also changes
- other items — we will buy more of some things and less of other things.
One of the key things to understand is whether total spending on the item itself goes up or down. Take petrol, for example. When petrol prices go up, we buy less (in litres) but not a lot less. As a result, spending on petrol goes up. On the other hand, when prices for some things go up (cherries, mangoes, merino-possum jerseys,…) our total spending on them goes down.
That leads to the income response. This is somewhat simplified, but basically we can have more or less money to spend on other things. We change how much we spend on everything else — heating, healthcare, holidays, etc.
Finally, we also need to take into account how the item itself fits into our total consumption. Take petrol again. When petrol goes up, cars and holidays become more expensive, so we buy less of them. Phone calls and internet connections can substitute for driving somewhere, so we buy more of them.
Now, let’s think about a tax on something like alcohol, ignoring for a moment where the tax revenues go. The tax makes the alcohol more expensive. That pushes up total spending on alcohol. We then have less money to spend on everything else, good and bad. Finally, our spending on everything else shifts around, depending on whether those things go with alcohol or substitute for it.
Are we happier? No, we are not, by assumption. The economic model assumes that we spend our money on the things that provide the most satisfaction. Otherwise, we could change our spending patterns and be happier. When prices go up (without compensating somehow), we are poorer and less able to get satisfaction.
Are we healthier? Well, that’s complicated. Let’s assume that our alcohol consumption (or sugar or fat consumption) is in the ‘harmful’ range. Assume, further, that by consuming less of it (although spending more on it) we are doing ourselves less harm. We have to set against that the income response and the spending on other items. We are poorer, so we buy less healthcare and less home heating. Our spending patterns shift, so we might buy less tobacco (if we smoke when we drink) but we also might buy less tomato juice, celery, and horseradish (because we like Bloody Marys). We may end up healthier, but we may not.
Are we ‘better off’ — has our welfare increased? In one sense, no, it hasn’t, by assumption (see, ‘happier’, above). If, however, we assume that people lack information or we assume that they are too myopic (that is, they know what’s good for themselves but don’t place enough weight on the future), then such policies might increase experienced welfare even as they decrease happiness in the present.
Note, however, the chains of ifs and buts, the assumptions on assumptions. If the lack of information or myopia is sufficient, and if the buying patterns shift towards ‘good’ products on balance, and if the net health impacts are positive, then price-based policies can make us better off.
I’ve left out a lot, a discussion of what happens with the tax revenue being the obvious gap. Even in the simplified example, though, the complexity is obvious. To quote the Dread Pirate Roberts, ‘Anyone who says otherwise is selling something.’