Economists Do It With Models

Warning: “graphic” content…

Bookmark and Share
Reader Question: Is (Behavioral) Economics Behaving Badly?

July 20th, 2010 · 24 Comments
Behavioral Econ · Policy · Reader Questions

A number of readers emailed, tweeted, sent carrier pigeons, etc. to ask what I had to say regarding a recent NYT op-ed piece by George Loewenstein and Peter Ubel that discusses the limitations of using behavioral economics to drive policy. The authors summarize their central thesis as follows:

But the field has its limits. As policymakers use it to devise programs, it’s becoming clear that behavioral economics is being asked to solve problems it wasn’t meant to address. Indeed, it seems in some cases that behavioral economics is being used as a political expedient, allowing policymakers to avoid painful but more effective solutions rooted in traditional economics.

Loewenstein and Ubel then go on to use the mandatory calorie labeling plans in places like New York City as an example of a case where “fashionable” behavioral nudges are being written into policy at the expense of more substantive and (theoretically) effective legislation. (Seriously, it’s just easier to read the article and then come back here.) There are two things that are important to keep in mind when mulling over the article’s logic.

First, the divide between behavioral economics and traditional microeconomics is not nearly as wide or as clear as the wording in the above quote implies. Put simply, the difference between traditional and behavioral economics is that traditional economics assumes that individuals are always able to maximize their long-term utility and aren’t swayed by silly things like discrepancies between short-term and long-term happiness, the framing, wording or context of their choices, self-control problems, cognitive limitations and the like, and behavioral economics…well, doesn’t. (When economists refer to “irrational” behavior, you can roughly interpret it as “behavior in conflict with one’s objective long-run utility or well-being,” though personally I tend to envision a hysterical 1960’s Betty Draper-type female.) In the calorie labeling example, the rationale is that people are making “bad” choices because they are either unaware of the characteristics of the products they are choosing from (rational but ignorant) or they are willfully ignoring those characteristics and seeking immediate gratification at the expense of long-term happiness (irrational). Any Economics 101 course will tell you that a required condition for markets to be efficient (read, value-maximizing for society) is that consumers have full information about the products they are considering consuming. In this way, the calorie-labeling legislation is helping to push the fast food market in the direction of efficiency as much as anything else. What’s so behavioral-y about that?

This is why I get annoyed when the calorie-labeling laws are reported as failures. A failure at what exactly? The law was supposedly put in place in order to give people easy access to information so that they can make better choices for themselves, and it certainly succeeds in this regard. It’s not really the law’s fault that people still want their Big Macs and super-size fries even when they have the calories staring at them in the face. (It’s also kind of funny to see the upper middle-class mostly white lawmakers struggle with the notion that not everyone is obsessed with being as thin and wrinkle-free as possible. Just wait until this sort of issue reaches the House of Representatives and Nancy Pelosi gets her hands on it.) Maybe those people still eating the Big Macs are being irrational and are going to regret it later, or maybe- shocker- the Big Macs are utility-maximizing despite the squishiness that they induce. If policy-makers are attempting to enact legislation under the guise of “saving people from themselves,” they should at least consider the possibility that people actually are doing what is best for them and don’t need or want to be saved. It’s interesting to note, however, that studies have reported that the calorie-labeling laws have caused people to make healthier choices for their children, which (weakly) suggests that people may still be having trouble making the best long-term choices for themselves but is far from being conclusive evidence.

The second, and more important, point is that the framing of the difference between the two lines of policy thought as “behavioral” versus “traditional” masks the inherent functional differences between the policy ideologies. When a government uses information and choice architecture to encourage what it feels is good behavior, it is subscribing to a philosophy of libertarian paternalism (yes, that’s actually a thing) whereby the regulator is nudging without restricting consumers’ options in any way. For example, the consumer has exactly the same set of choices available to her regardless of whether calorie counts are on the menus or not. Because of this feature, it’s hard to argue that this sort of legislation is significantly bad for anyone- here, the worst-case scenario is that some people keep eating unhealthy food but are no longer blissfully ignorant and instead feel guilty. This is different from, say, a tax on fast food, which would raise the price of food and thus limit the choices available to consumers. Given this difference, why shouldn’t lawmakers try to achieve their goals via legislation that doesn’t impose monetary costs and doesn’t force people’s hands via the almighty dollar before resorting to the blunt kick in the pants that taxes usually end up being?

The idea of saving people from themselves is nice and all, but, in reality, the main motivation and justification for intervening in the fast-food market is that eating fast food imparts costs on the health-care system that end up being paid in part by people who didn’t specifically choose to live on Double Quarter Pounders with Cheese. (negative externatlities, in economic terms) Again, Economics 101 textbooks will tell you that when negative externalities are present in a market, putting a tax in place can actually increase overall societal welfare. The tax is theoretically helpful because it causes people to internalize the costs that their actions have on others, and these people produce and consume less of the thing with the negative side effects as a result. However, improving overall societal welfare doesn’t necessarily correspond with making everyone in the society better off- for example, requiring that a person be sacrificed so that his organs can save the lives of multiple individuals waiting for transplants would also increase this textbook definition of societal welfare, but we’d certainly think twice before implementing this sort of plan.

Are there reasons that a government should think twice about implementing a fast-food tax as well? The fact of the matter is that, by putting a fast-food tax in place, the government is raising the price of food. (This is true even if people end up switching away from the fast food.) The people who are going to feel this price increase the most are those who consume more fast food, and these people are generally of the lower-income variety, except perhaps for my friend Brian who takes his Ferrari through the Taco Bell drive-thru on a regular basis. I’m all for encouraging healthy eating and whatnot, but something doesn’t quite sit right when it comes to artificially raising the price of food for people who are likely struggling to feed their families in the first place, especially when the justification is that the rest of us don’t have to look at their fat as…er, so society doesn’t have to foot the bill on health care costs. (Yes, I also get that the people will live longer and healthier to some degree, but that’s sort of like saying “hey, you know how you’re struggling to get by and all? Well, guess what? We’re going to make life harder for you now so that you can keep struggling for longer. You’re welcome!!!”) It’s not surprising that the discussion usually takes a philosophical turn pretty quickly- one could argue that it’s not fair for poor people to pay $1 more for their food so that Bill Gates can put $1 less into the health care system, but it’s also not really fair that Bill Gates has to pay extra money into the health care system because other people chose to eat a ton of bacon cheeseburgers.

The other issue with bluntly using a tax to correct for the externalities of unhealthy food is that fast food is not like pollution- it’s not like I am adding some small amount to the costs of the health care system every time I have my chicken McNuggets. Therefore, taxing me each time I eat said McNuggets is not the right approach, and it is not necessarily the case that taxing in this way will even lead to an improved outcome for society. (For the wonks in the audience: The reason that a tax can be a good thing is that it decreases the cost of the externality by more than the deadweight loss that it creates. However, if the people that change their behavior due to the tax are in large part those who weren’t creating the externality in the first place, society will see the downside of the tax without a whole lot of the upside.) Maybe this is just me being bitter about having to pay because other people can’t handle their McNuggets, but I don’t really want to hear about most of these “sin taxes” until someone figures out a reasonable way to specifically tax excess consumption.

The bottom line is that designing good legislation that limits people’s choices and hits them where it hurts is harder than simply nudging people in the right direction. Done right, it might also be more effective, but from the above discussion it should be clear that “doing it right” is a tall order. Make no mistake, however, that the distinction is between easy and hard much more so than it is between “behavioral” and “traditional.”

To read more about nudges and libertarian paternalism, I recommend Nudge: Improving Decisions About Health, Wealth, and Happiness by Richard Thaler and Cass Sunstein. You can also read a response to the NYT article on the Nudge blog here.

Tags: Behavioral Econ · Policy · Reader Questions

24 responses so far ↓

  • 1 Joshua // Jul 20, 2010 at 3:21 pm

    Another great post, and one I may link to at my blog, since it has to do with the ethics of eating, and therefore, the environment.

    We’ve really tied ourselves up pretzel-like in trying to simultaneously heavily subsidize bad food (monocrops, processing, etc.), oppose universal health care, and allow pollution externalities (e.g., carbon) to go un-priced and unregulated.

  • 2 Paul // Jul 20, 2010 at 3:24 pm

    But is unhealthy eating actually a negative externality? This is only because we chose to make it one with a public health care system?

    It seems to me, that (in your example) government is trying to fix a problem it created in the first place. Allowing individuals to put the costs of their unhealthy lifestyle on other people.

  • 3 Chuck Dolci // Jul 20, 2010 at 3:59 pm

    I agree completely with Paul.
    Once you accept the premise that society (read – “your more affluent neighbor”) is responsible for your HEALTH CARE (i.e. taking care of you when you get sick) then it is just logical to assume that society (read – “your more affluent neighbor”) is also responsible for your HEALTH CARE (making sure you don’t get sick in the first place).
    So if you want society to be responsible for your health care then you have to be prepared for society dictating how you live your life.

    And “society” always means “your more afffluent neighbor” – because if you could afford it in the first place (meaning you give your health care greater priority than season tickets to the Celtics) you would not have to rely on “society”.

    One more thing – I strenuosuly avoid all healthy foods – for two (make that three) reasons.
    1 – too many of my friends and colleagues who were health food junkies died at an early age;
    2 – I don’t particularly want to live into my 90s (eating healthy foods may add years to your life – but remember – those added years come at the end. It is not adding years to your 30s and 40s. It is not more weekends partying in Cancun – it is more weekends in the old folks home, wasting away); and
    3 they generally taste lousy.

    Thanks, but I don’t need Aunt Nanny State telling me what is good for me.

  • 4 econgirl // Jul 20, 2010 at 4:24 pm

    @ Paul and Chuck: You’re mostly right, but it’s important to remember that these sorts of cross-subsidy and moral hazard issues don’t arise solely in situations where health care is publicly funded. Even if there were no Medicare/Medicaid/whatever, the non-cheeseburger people would end up subsidizing the cheeseburger people whenever the two groups crossed paths on an insurance plan. So yeah, if everyone strictly paid their own way it would be more of a non-issue, but it’s not specifically the public nature of the systems that causes the problem. Also, there’s nothing to say that society could even refer to your less-affluent neighbors, as would be the case if Bill Gates had health insurance and was the cheeseburger eater.

    That said, even if you take health care out of the picture, there are side effects of obesity and bad health on disability payments and such, as well as on general productivity, and it’s harder to argue that these shouldn’t be publicly-funded programs, so it’s unclear how we could have a complete “every man for himself” approach in this situation. In principle, however, I do support a “I take care of my own business, so don’t tell me what I should be doing” mentality.

  • 5 Scott Ritchie // Jul 20, 2010 at 6:02 pm

    Penn and Teller did an episode on Fast Food recently, and their point was that it’s far more logical to repeal the agricultural subsidies than tax fast food. Fast food is artificially cheap because its main inputs are the largest recipient of agg subsidies (corn, wheat, and potatoes), while healthier produce is often unsubsidized.

  • 6 Marc // Jul 21, 2010 at 3:35 am

    That is absolutely bizarre to use nutritional labeling of products as an example of behavioral economics. Transparency like that is really one of the most basic free market concepts out there. A tax on the other hand, would be the quintessential product of behavioral economics.

    Where the real problem lies with behavioral economics is in the statement “The bottom line is that designing good legislation that limits people’s choices and hits them where it hurts is harder than simply nudging people in the right direction.” It’s more of an issue of not being possible, especially over the long-term perspective. The mistake is the assumption the human behavior can be modeled in scientific matter.

    Human interactions and social behavior are emergent and unpredictable. An issue modern economics has chosen to leave out of it’s approach to modeling.

    There’s a lot of work being done on this subject, yet it seems to intentionally ignore by the economics profession?

  • 7 Michael // Jul 21, 2010 at 4:00 am

    @Marc Human behavior is not as mysterious and ineffable as you seem to imply. Think of a bell curve; there are certainly edge cases but well-written legislation will push the majority of people in the proper direction (though with the caveat of being in the short-term, as you noted).

    One of the primary assumptions of economics is “people [economic actors] respond to incentives” (see While the reaction is not always perfectly mappable and second-order effects can spawn unintended consequences, you have to work within the parameters you’ve been given.

    That is, perhaps in the long run many of the issues we rush to solve would simply work themselves out. It is because we are impatient that we push for solutions quickly, sometimes becoming part of the problem. But what is the alternative, to do nothing? Certainly economics, as imprecise a science as it is (behavioral or otherwise), has better solutions for us than that.

  • 8 Marc // Jul 21, 2010 at 5:38 am

    Michael, you make a point. The simple individual aspects of behavior can be addressed this way in an academic sense. Given many things simplified out the equation and my aspects of cause and effect taken on assumption.

    And it’s indeed true that people in a general sense respond to incentives. Modeling a simple behaviors given many assumption about initial conditions and other factors can be vary informative and can allow us to better under stand the possible cause of observed outcomes.

    The disconnect comes when trying to apply what we learn from these academic models to the creation of forward looking models that we defend as accurate enough to base policy on.

    Where modeling of human behavior in the real world (forward looking), completely breaks down can be addressed most basically by examining initial conditions and incentives. In the real world, the initial condition of the model are very complex and not easily accounted for, in addition these variable adapt to each other over time further complicating the possibilities of modeling a situation accurately. At the same time an individual’s reaction to one incentive adapts based on the movements of all other incentives and conditions and thus are adaptive as well.

    So that’s a very basic explanation why human behavior can not be modeled well. Do we create models that seem to be accurate? Of course. Is it because of there hard scientific validity? No.

    Last, to address your comment. Should we do nothing? For the most part, Yes. I suggest you do some reading on the informational nature of markets.

  • 9 Daniel // Jul 21, 2010 at 10:28 am

    Great Post.

    Quick question: In your extensive, unparalleled experience and opinion, what has been the least-succesful (to be read: most disastrous, terrible, worst, etc.) policy that was actually put in to effect, and failed miserably?

    Love the blog.

  • 10 econgirl // Jul 21, 2010 at 1:02 pm

    @ Marc (re your 1st comment): Exactly. I think that the calorie labeling is being categorized as a behavioral economics thing because Richard Thaler wrote about it in Nudge and he is a behavioral economist. (I’m leaving Sunstein out of this discussion because I’m not entirely sure what he is.) This logic, however, would imply that I could write about puppies and that would count as economics because I’m an economist.

    The most perplexing part about the original article’s premise is that arguing that people behave irrationally and could be made better off by an outside organization “saving them from themselves” lends itself to arguments for stronger legislative controls, whereas the authors are lamenting that behavioral economics is offering weak solutions to problems.

  • 11 Matt H. // Jul 21, 2010 at 1:46 pm

    Marc, well put. At first I thought you were arguing that modeling needed to attempt to predict emergent and adaptive behavior, but then realized you were advocating inertia in favor of inducing unforseen, negative incentives. If regulators thought more about that in the lead up to the housing crisis they may have patrolled home lending more vigorously, restricted hedge fund coaxing of CDO managers, or required ratings agencies to not fall asleep on the job.
    In light of how heavy handed legislation can lead to negative and unpredictable results, I think I would prefer the small nudge on as many issues as applicable.

  • 12 JJ // Jul 21, 2010 at 4:03 pm

    I largely agree with your points, but I would add that providing information is costly. So requiring a firm to provide more information is akin to a tax.

  • 13 Dan L // Jul 21, 2010 at 9:41 pm

    I agree that this has nothing to do with behavioral vs non-behavioral economics. It has to do with nudges.

    Here’s the real story: The libertarian zealots say that the government shouldn’t do anything, whether it be forcing people to do things, or paying people to do things. Enter Richard Thaler and others with Nudge etc, who say, “Well, can’t we at least nudge them to do things? You can achieve a lot with nudging, and you’re not forcing anyone to do anything, nor does it cost much.” People in power see this and say, “Ooh, cheap policy that accomplishes stuff and is non-objectionable. Yes!” Not surprisingly, nudge policies don’t change anyone’s lives. Not surprising because they were never meant to. The author’s valid point is this: If you want to solve real big problems, you’re going to have to spend real money and do real stuff. But I think he’s attacking a straw man because no one ever claimed that nudges could save the world. People aren’t failing at the big problems because they’re obsessed with nudges. They’re failing at the big problems just for the usual reasons–politics as usual, right-wing crazies, the complicated nature of the problems, legitimate differences of opinion, etc.

    An aside on Nudge: The sad thing about this book is that it’s a very long appeal to libertarian zealots that they should accept no-brainer policies.

  • 14 Marc // Jul 22, 2010 at 5:04 am

    Jodi, one excellent paper on the fundamental flaws of Behavioral Economics is The Knowledge Problem of New Paternalism, by Mario Rizzo.

  • 15 The Demand For Ignorance «  Modeled Behavior // Jul 22, 2010 at 6:05 pm

    […] Modeled Behavior, obesity | by Adam Ozimek Jodi Beggs, aka Economists Do It With Models, argues that paternalism need not be justified by assuming irrational agents, but can simply be an […]

  • 16 Dave // Jul 22, 2010 at 7:22 pm

    I did not see the details on this proposed labeling law for restaurants. Would it require all restaurants to provide the same type of nutrition information as is currently found on packaged food? If so, another economic effect would be to give large chain restaurants an advantage over local “mom and pops”.

    For a large chain, determining nutrition information for their menu is a relatively small expense. However, for a “mom and pop” restaurant, which may have a large, variable menu and only one or two restaurants, the expense will be relatively greater. This would be especially significant if the law requires a proper laboratory analysis (as opposed to simply totaling up the nutrition value of the ingredients) — the restaurant would have to cook up everything on the menu and send it off to a qualified laboratory for analysis.

    This sort of thing seems to be one of the things that legislators often overlook when they create regulations: Regulations often create overhead expenses and barriers to entry that give larger/established/static businesses a relative advantage over smaller/new/dynamic businesses. (For example, consider how expensive the cost of meeting regulations would be if you were starting an automobile company. What is a middling expense for the present-day Ford Motor Company is a barrier to entry for an entrepreneur running a shop the size of Henry Ford’s original shop.)

  • 17 Dan L // Jul 22, 2010 at 8:12 pm

    Dave, I think that even our politicians are not mean/stupid enough to screw over mom and pops that badly. In New York at least, only restaurants with a certain minimum number of locations are required to post calorie information.

  • 18 econgirl // Jul 22, 2010 at 9:07 pm

    If I remember correctly, the threshold in New York is either 11 or 14 locations. It should not shock you, especially if you’ve been reading here for a while, that restaurants are employing creative strategies to stay under this threshold. I think the Houston’s chain goes by about 4 different names by this point, for example.

  • 19 Marc // Jul 23, 2010 at 5:40 am

    Dave, you make an important point that regulation does have to general effect of stifling economic growth and hindering innovation, regardless if the intents are “good” or “evil”

    Which brings up the argument, is it best to approach regulation from the side of caution? Only regulating were regulation becomes absolutely necessary. As opposed to the currently popular view of regulating for the intent to create change.

  • 20 Joao Pedro Afonso // Jul 23, 2010 at 8:45 am

    What perplexes me in the referred article from the NYTimes is that, while it appears a reminder of the limits of behavior economics and a critic to too much reliance on that discipline to produce solutions instead of more orthodox economic solutions, some of the critic only makes sense in the light of the findings of the same “behavior economics” discipline criticized. As Marc said, “nutritional labeling” is orthodox economics to the fullest in a regulated market, meaning the necessary transparency for a rational costumer to do their choice and be the desired “economic man”. Since what we want to prevent is high probable life threatening situations connected to obesity and no one “rational” can put a price on his life, is something to admire that such available information will not automatically translate in saner life styles. So there is a curious inversion in all this because “nutritional labeling” appears to me to be an economical 101 solution (working by revealing the true costs of a decision in the long term), while if that fails, I’ll trust behavior economics more to offer the best explanation for that.

    To be fair, the opinion voiced is not that hostile to behavior economics, but the way it starts establishes the mood.

    And then, there is the question of the “true” cost and how to calculate it. In the nytimes article I think we can speak of two distinct kinds of goals while discussing the tax increases they proposed as prospective more effective solutions. Continuing with the obesity example, one goal is to have less obesity,… how much less in the goal will determine how much we need to tax the food to create the desirable effect. On another, there is an implicit goal set when they say “An increase in the gas tax that made the price of gas reflect its true costs”… what is implied here is that the situation is what it is because there is a price dumping promoting the wrong consume, and an higher price reflecting the true cost will lower that obesity incidence. This is a kind of different goal to me because it is implicit: it is accepted as long as it fulfill the seeming fairer approach of equalizing the price to the real cost of the thing… considering all its impact in the society. But how it is calculated?

    It is not just the implication here that the public sector is already incurring costs with this situation, or the pertinence of having taxes collected for something which might well be individually solved by the private sector without those. We could discuss if that means some obligation from the public sector to solve the problem, but nothing forbids a national program funded with that, where all the solution actors may drink… inclusive supplementing the otherwise more costly insurances plans referred by Econgirl.

    My problem is other: with so many interdependence of policies and multiple goals for each, how may we calculate the “true” cost of a commodity?… because once the concept is out, we will demand rigor in it. Take the car’s fuel price for instance. When the article uttered about the true costs of fuel, he was probably thinking on correlated costs of the usual suspects like the carbon footprint or pollution, and even some subsides to the oil industry. But a “true” cost approach probably would have demanded more than that. It can be argued that a large part of the USA’s military complex, some operations (like Iraq’s war) and some costly support to nations over-seas are linked to the pretension of securing stable oil sources around the world. True costs would have to account for that too, and being the USA, one of the countries which expend more on those areas, I would have expected some serious fuel inflated prices by them. The good side of the coin would have be less income taxes or public debt, and likely less success in demagogic political gestures (imagine a politician having to announce, “citizens, we arm ourselves to go to war to pay less for the oil, so lets pay more for each gallon now”… so delicious as “we make war to make peace”), reasons enough to deserve my sympathy. Still, there is the question of knowing the “true” cost. And while I could argue that the fuel true costs must cover these, others might argue that the same parcel are in reality to defend the nation or freedom, to help friends, or simply to train troops and help them to be ready. In other words, I’m not sure that a fair true cost assessment can be done, and that will inevitably poison the discussion of how much taxes to implement. Because the electorate might swallow painful measures with the argument of the true costs but only if they are indeed the “true” ones.

    (Anyway, I’m only a layman on this, so everything I said might be wrong. Sorry for the comment extension)

  • 21 Ezequiel Colanero // Jan 30, 2012 at 2:25 pm

    Wow, fantastic internet site! I actually liked the content! Make sure you keep on writing about this content material, I will likely be subscribing next!

  • 22 racing pigeons, pigeons,buy racing pigeons,porumbei, porumbei voiajori,porumbei de vanzare,pigeons for sale // Feb 24, 2014 at 6:23 am

    If you want to buy pigeons you have come into the right place. This is your chance to own the best racing pigeons.

  • 23 Behavioral Economics, Correspondence, Bohr, Hayek, and Keynes | The Incidental Economist // Feb 26, 2014 at 10:31 am

    […] the many excellent points made by Econgirl in her post this week discussing a recent NYT op-ed piece by George Loewenstein and Peter Ubel criticizing the […]

  • 24 // Jun 23, 2014 at 1:23 pm

    The latest increases are because insurers are inflating costs for repairs and
    hire vehicles. 5 live Investigates took up Ashley Bennett case with Emotor
    fleet insurance, the company will be charging
    as of July 1, 2013 and will offer VW VOWG_p. 7m UK drivers dodged motor fleet insurance last year.

Leave a Comment