Sunday, September 25, 2011

Mark Bittman's war on junk food

Mart Bittman had another complaint about junk food following last month's long article on why we should tax it. Today, he explained that junk food was actually more expensive than "better food" and so that the difference in prices is not the reason why junk food is so popular.

I have a couple of problems with his argument. First, he argues that the main problem is that people do not want to cook. But this can be explained as a price problem: once you account for the cost of cooking(and not only the concrete costs of utensils and all, but also time), is it clear that fast food is more expensive than, say, organic food?

Second, Bittman links junk food and tobacco in an interesting way. First, he explains that junk food is addictive:
A 2009 study by the Scripps Research Institute indicates that overconsumption of fast food “triggers addiction-like neuroaddictive responses” in the brain, making it harder to trigger the release of dopamine. 
This leads him to argue that "real cultural changes are needed to turn [the problem] around". For him, the question is "How do you change a culture?" and he explains that this is what the tobacco market has done since the late nineties "Smoking had to be converted from a cool habit into one practiced by pariahs". The weirdest part, for me, is when he argues that we should "give [our children] the pleasures of nourishing one another and enjoying that nourishment together".

First, I have a problem with the "cultural" argument, because it seems to be quite vague, and also intractable. Indeed, how do you change a culture? More problematically, what is culture? Who decides what "culture" is the target culture?

Second, although the addictive component of junk food is a nice way of to see that junk food is similar to, say, tobacco, I don't understand the follow-up. It is not clear to me that the objective, and the effect, of the anti-cigarette campaigns were to transform people into pariahs. It seems to me that there was a campaign of information and an increase in costs so that people would be better informed of the consequences(e.g. pictures on cigarette packs), and so that the externality cost would be internalized by consumers(increase taxes on packs). I am not sure there was a culture war against cigarette.

This being said, even if one accepts the cultural argument, I think Bittman has his policy prescription wrong. It is not clear to me that you need a big cultural shift. Malcolm Gladwell has a nice discussion of the cigarette problem in The Tipping Point. First, smoking and depression are closely related, and anti-depressants have a significant impact on the rate of smoking quits. Second, addiction is not instantaneous and it affects people differently(e.g. here).  You can wait years before being addicted, and some people can smoke regularly and never get addicted. This means that the share of nicotine content in cigarettes can be lowered slightly and have huge impacts. More importantly for the argument here, Gladwell says that
We've been obsessed with changing attitudes toward tobacco on a mass scale, but we haven't managed to reach the groups whose attitude needs to change the most. 
For cigarettes, informing and increasing prices are ways to make the "market" better. Targeting nicotine levels and using anti-depressants are ways to solve the addiction problem. The former thing was what makes me more sympathetic to Bittman's first argument in favor of taxing food, and also argues in favor of the new health care law requirement that calories be posted(there is evidence, however, that that does not work. But it is not clear!). On the second point, well,  apparently nicotine and junk food both act on the level of released dopamine, so hopefully we can find ways to target addiction.

I want to conclude on another link, slightly related. As I mentioned above, calorie posting on menus does not seem to work especially for low-income people, which is, I believe, the population that Bittman wants to target. Interestingly, in their new book, Abhijit Banerjee and Esther Duflo explain that when people are budget constrained and cannot afford the desirable number of calories, giving them extra money will not lead them to spend all of this money on food, and the food they'll purchase will be less "efficient" in terms of calories("television is more important than food!"). They refer to a recent paper by Jensen and Miller who ran a randomized-controlled trial showing that "households that received subsidies for rice or wheat consumed less of those two items and ate more shrimp and meat(...) Remarkably, overall, the caloric intake of those who received the subsidy did not increase". Duflo and Banerjee seem to argue in favor of paternalistic policies, such as giving way fortified foods to parents and more generally, providing more directly better-quality food to poor families. It is not clear, however, that this is what they would advocate in a country with better institutions. Interestingly, they mention the information problem:
They[The poor] often behave as if they think that any change that is significant enough to be worth sacrificing for will simply take too long. This could explain why they focus on the here and now(...)
This is clearly not specific to any category of people. The instantaneous pleasure of junk food or of cigarette, and the general short-termism, affects everybody except George Clooney. This means that we do not compute the real costs of short-term behavior(and Banerjee and Duflo show evidence that the poor, for instance, would like to constrain their future behavior on health the same way anybody would like to have a commitment device for new year's resolutions). So improving information and, more controversially, increasing costs(also because of the externalities we're talking about), is still fine with me.

Thursday, September 15, 2011

You're beautiful, no matter what they say

Is being beautiful good or bad? The question is important, because it will help us know whether Greg Mankiw will write an editorial on whether we should tax them(I am a fervent advocate of his proposed tax on height)

I don't have an answer to the question, but there are several things to think about. First, Hamermesh and Abrevaya had a working paper at NBER last month arguing that beautiful paper are happier. Should we accuse them of Captain Obviousness? It's not clear. As they mentioned, there are two ways beauty could affect happiness: indirectly through "market-related" outcomes, and directly, because beauty is awesome. They find that indirect channels are more important:

Among both men and women at least half of the increase in satisfaction/happiness generated by beauty is indirect, resulting because better-looking people achieve more desirable outcomes in the labor market (higher earnings) and the marriage market (higher-income spouses). That relatively more of the impact among women is direct, not mediated through the effects of beauty on market outcomes, might help to explain gender differences in people’s attitudes about their own looks.
Interstingly, there are a lot of discussion on whether beauty has a positive impact on "market-related outcomes.". PsyBlog lists 5 benefits and 5 problems with being hot. For instance, the Halo effect means that attractive people are seen as better in various unrelated characteristics. Beautiful people are paid more. Hot athletes are paid more(" a “good-looking” quarterback like Kerry Collins or Charlie Frye earned approximately $300,000 more per year than his stats and other pay factors would predict.") Apparently, through other characteristics, attractive people are more persuasive. Men are more ready to forgive attractive women after an apology. Hamermesh(again) finds that beauty improves earnings in China but that investing in beauty is extremely costly(not only because it has decreasing returns to scale), and so advocates for protection of less beautiful people

However, being too pretty has negative impacts. Mostly, it is related to a couple of things:

  • People don't like you because you're too hot: attractive people have problems in job interviews when the interviewer is of the same sex(apparently mostly a woman problem). Also, you scare people. A friend of mine told me of a cameraman on a movie set asking Natalie Portman out, and she said yes because nobody dared asking her out. Unconsciously, hot people are considered less suitable for long-term relationships. Women are less ready to forgive attractive women after an apology
  • People think you are less competent: basically, if you see two identical persons only differing in how attractive they are, you discount the quality of the pretty one on the criteria that you(consciously or unconsciouly, opposing the halo effect) think are positively affected by beauty(or this is what Bayes tells you). Attractive women are discriminated against in masculine jobs, again, consciously or unconsciously

Finally, an interesting dynamic was underlined in a recent OkTrends blog post, which might underline a last problem with being objectively hot. It is summed up in this picture:


Girls get more private messages on OkCupid when there's debate on whether they're attractive, one possible reason being that guys think they have more chances(competition is less fierce).

Wednesday, September 14, 2011

Why Spain?

A friend of mine asked me why Spain, which has a low debt-to-GDP ratio, was attacked by the market in the same way Italy is.  For the debt/GDP ratio,  here is the evolution over the last decade, via Eurostat(Spain is in pink):
Now, as I mentioned here before, Paul Krugman, Martin Wolf and Paul De Grauwe convincingly argue that Spain and Britain have a similar trend in debt to GDP ratio(though as I mentioned earlier, the maturity profiles of the two countries might also be a factor) but Spain pays a far higher interest rate on its 10-year bonds because they do not have the option to devalue their currency. Now, the comparison with Italy is trickier. First, a graph of 10-year bond yields since the beginning of the year: 
Now, here is what I quickly answered(with some probable mistakes along the road, sorry)
  • I think there are several things that matters. First, without talking about the papers  I am currently reading, there are various weird things about Spain. Unemployment is extremely high. It fell because of the construction bubble before, but Spain usually had an unemployment rate of more than 15%. Now it's around 20, I think, with a 40% unemployment among young people. Structurally, they have problems. Second, their banks are indebted. Spain had the majority of failed banks in the two stress tests the EU has conducted. Third, Spain provinces are quite independent from the central government and there's no cooperation. The central government has less power than other countries to control public debt. Try this article from today with Google Translate.
  • Now, actually, debt to GDP is not a big factor in bond spreads, from what I gather from articles, so it is not clear we should look at those numbers. A recent  NBER paper has that debt/taxbase has a significant impact, though.  Growth prospects are important, and you see that Italy, Greece, Spain have structural pb(I think it's less the case with Ireland). We don't know what are their comparative advantage(although tourism is probably one of them). Trade is important, and Spain had a HUGE trade deficit in the second part of the decade. Exports gets you reserves and helps you pay your debt, especially if it's dollar denominated. Being in the Eurozone also means that the terms of trade(how much imports you can pay for a given amount of exports) have been quite bad for a while. This is the trade balance, via Eurostat again: 


Update(09/15/11): I forgot to list a couple of things. First, via Brad Plumer, the list of countries with the highest share of small businesses is quite amazing




Still, it does not say anything about Spain vs. Italy, which both rank at the top. A report by the IMF has some nice numbers and figures. For instance, in 2007, Spain had the second highest current account deficit, after the US(numbers are not adjusted by GDP)
(Source: IMF, World Economic Outlook database, as of August 21, 2007.)

If you go look at table 3 of the report, you'll see that Spain has one of the lowest reserves to GDP ratio.

Finally, on the structural problems in Spain and the fact that restoring competitiveness appears to be hard, the graph of startups in the last 5 years is telling: