Cafeteria Marvels

 by Mario Rizzo  

The market is a “marvel.” What does that mean? According to Marcus Tullius Cicero, the Roman orator and senator, a marvel is something contrary to or surpassing common understanding.  

In that sense, the market is a true marvel – so much so that it even surpasses the understanding of many economists.  

Richard Thaler (a University of Chicago Business School professor) and Cass Sunstein (a Harvard law professor and Obamian regulatory czar) have illustrated the benign qualities of paternalism with a curious example of cafeteria food placement. (An interesting and important exchange between Glen Whitman and Richard Thaler – among others – is now taking place at Cato Unbound.)

Here is a portion of an article, by Richard Thaler and Cass Sunstein, that appeared in the American Economic Association’s Papers and Proceeding, May 2003. They provocatively called it “Libertarian Paternalism.”

“The first misconception is that there are viable alternatives to paternalism. In many situations, some organization or agent must make a choice that will affect the choices of some other people…  

Consider the problem facing the director of a company cafeteria who discovers that the order in which food is arranged influences the choices people make.  To simplify, consider three alternative strategies: (1) she could make choices she thinks would make the customers better off; (2) she could make choices at random; or (3) she could maliciously choose those items that would make customers as obese as possible. Option 1 appears to be paternalistic, which it is, but why would anyone advocate options 2 or 3?  

The second misconception is that paternalism always involves coercion. As the cafeteria example illustrates, the choice of which order to present food items does not coerce anyone to do anything, yet one might prefer some orders to others on paternalistic grounds.”  

This is a very amusing story. The people who go to this cafeteria are apparently automatons who, time after time, are moved in the direction of the fruit or cake even though they know the other is coming up next. I’d like to meet these people and sell them bridges and such.  

As a “poor” graduate student at the University of Chicago many years ago I used to eat at a couple of cafeterias. One, on 53rd Street, in Hyde Park called “Valois” (Valoise, please) did not have fruit at all. So there was no issue. The other, the cafeteria at the University of Chicago Hospital, had the cake before the fruit in those days. It may be the other way around now. Anyway, after a couple of visits I knew the order and could easily ignore the “nudge.” I wanted the fruit. But maybe that is why I was able to get a Ph.D. in rational-choice economics! 

A few days ago I was at Union Station in Washington, D.C.  Here I saw a marvel worth more than a dozen articles on paternalism. There is a cafeteria set-up at a national restaurant chain there. So in my guise of empirical economist I looked to see the order of the cake and the fruit. Mirabile visu! There were two placements.  

At the one end of the cafeteria setup, the food display started with cake. At the other end, it stated with fruit. In other words depending on the direction from which you entered the line, you got fruit or cake first. Furthermore, there was no preferred side of entry – each was open to a large part of the station.  

So if I enter from the fruit side, I can easily ignore the cake – nothing more comes after it. If I enter from the cake side I can ignore the fruit. Obviously some people will enter randomly unaware of the placement. The manager is obviously not nudging one way or the other.  

It is as if the firm were providing two kinds of restaurants:  one that enables people to ignore the cake; the other that enables people to ignore the fruit. This is a microcosm of the market-provision of choice architecture. Different restrictions of the choice set are available for different people. This is similar to the effect competition has in providing all sorts of options – vegetarian, red meat, seafood, Kosher, etc. restaurants.  

I described all of this, quite naturally, without referring to it as paternalism. Why? First, in a market there is mutuality of gain – the firm manager is not providing a benefit to the customers without corresponding gain to himself as a paternalist usually does. Second, the restaurant is catering to expressed preferences, not hypothetical “true preferences.” A good entrepreneur tries to predict what people might like under specific conditions. Is market entrepreneurship paternalism because it tries to uncover latent demands? It is paternalism because it tries, in exchange for custom, to help people with their self-control problems?  

One big problem with using the word paternalism in this kind of situation is that it deflects us from the role of profit and mutual gain in solving the problems some consumers may have. A paternalist, on the other hand, is a kind of charity worker or possibly a Good Samaritan.  

The idea of paternalism adds nothing to our understanding of cafeteria economics. It is simply a facile rhetorical gimmick to slide us to real legal paternalism. We are not fooled.

26 thoughts on “Cafeteria Marvels

  1. So can you get the data from that cafeteria whether there is a difference in fruit vs cake buying between those entering from the left or right? Are there two cashiers? Or we could set up some easy video monitoring to mark lefties from righties and ignore repeat customers (who may be endogenously choosing their entry). I’d fund that study.

  2. I should have added that I dont know how much difference placement actually makes. What I am saying is if it makes the difference Thaler and Sunstein claim, here is market provision of something probably even better than they could have imagined.

  3. They define paternalism very broadly. The placement at most is a suggestion. That is hardly paternalism. As Mario indicates, if entrepreneurs make bad suggestions, they lose out to those who better know consumer preferences.

    Alchian’s concept of search is relevant. Much of what entrepreneurs do is an attempt to discover (search for) consumer preferences. They are not “given.”

  4. One marvelous aspect of markets is “Economic Darwinism”. That is to say that because a business optimizes profits, and must do so to survive, one can draw a general conclusion that a behavior of a business that survives is a profit maximizing behavior.

    Retail stores spend large amounts of money determining the positioning of goods in their stores. Therefore I presume they do it to maximize profits. Therefore placement influences purchasing. Does that means “The people who go to this store are automatons” — to a certain extent, yes.

    You might not like the conclusion, but another marvelous aspect of markets is that businesses don’t get to choose to do what they like. Businesses don’t have a choice, they must do what maximizes profits, whether they like it or not.

  5. Incidentally, George Will (that well known leftie commentator) wrote a article in favor of paternalism, suggesting that setting the default to the ‘beneficial’ value without limiting any choices, was acceptable. He suggested, for example, that participation in 401-Ks should be ‘opt-out’ not ‘opt-in’. The same might be said of organ-donation.

    The question posed by the “Obamian” is if you have a decision that could be made in a beneficial, random, or malevolent way, why would it be right to choose “random” or “malevolent”?

  6. I think that the big reason that paternalism is becoming so prominent is class.

    I live in Ireland. Here, and in the UK, there is a much bigger visible difference between classes than there was in the recent past. The lower class have taken on a certain culture of their own rather than borrowing from mainstream culture. We have “chavs” (and “knackers” in Ireland) in the US you have “douchebags”, “hipsters” and “butterballs”.

    I think that many people feel threatened by this and are supporting paternalistic methods of getting rid of them. In Britain for example the Labour party are proposing a law against under 18s using sunbeds! They’ve already specifically increased the tax on cider (alcoholic cider that is). If those aren’t anti-chav laws I don’t know what is.

    This nudge stuff is just a smokescreen to hide the distaste the middle-class and upper-class have for the lower-classes.

  7. Dave Pullin seems to be enagaged in a non sequitor. If firms department stores are maximizing profits by their placements, then they are placing items where their customers want them. How does that make customers automatons?

    A department is not a metahpor for the state. The state’s “suggestions” are inherently coercive. It is the difference between contract and compulsion.

  8. “Therefore placement influences purchasing.” -Dave Pullin

    Or placement is influenced by personal preferences.

  9. @Jerry O’Driscoll No. Your point contracts the facts for retail store placement. Often goods that customer want most are placed not in the most convenient place, but in a place that requires the customer to go past lots of displays that encourage impulse buying. There is a very famous, and very well documented, example of placement of six-packs of beer next to diapers in convenience stores.

    You will, no doubt, argue that whatever the customer does is what the customer wants to do, but the professionals know that what they want can be influenced. If a business can *make* a customer want something, isn’t that coercion?

    “The state’s “suggestions” are inherently coercive.” So it you can choose between X and not X – for example whether to be an organ donor or not – and the state says the default is “you are”, then that’s coercive. And if they say the default is “you are not” that’s coercive too. If they say “you must go to the DMV and declare whatever you are or are not an organ donor”, that is clearly coercive. … so your point is that everything can interpreted as being coercive.

  10. @Jerry O’Driscoll Some things governments do are coercive, like taxes. Therefore everything governments do is coercive?? Some coercive things are bad, therefore all coercive things are bad??

    I am very glad that governments coerce you (and me) in to demonstrating some minimal ability to drive safely before we can (legally) drive on the road. I am glad governments coerce enforcement of laws against murder, theft, etc. … aren’t you?

    Whether a department store can coerce you depends on the availability of competition. If they are the only supplier of food, then they are coercive. A monopoly is coercion. (of course you don’t have to eat food, just like you don’t have to pay taxes, you can starve and you can avoid having income, respectively).

    The cafeteria placement example does indeed reflect customer preference if there is perfect competition … enough cafeterias to choose from such the customer can find the placement that he prefers. But such perfect competition never occurs (just as perfect monopolies never occur.)
    Perfect competition in the cafeteria placement example requires an unreasonable investment in cafeterias and customers who make an unreasonable effort to choose among them based on food placement.

  11. Dave Pullin,

    I and almost everyone I know was driving safely, if illegally according to the state, long before we took the exam required by government. Every year upwards of 40,000 people die in auto accidents in the U.S. despite the fact that they were all tested by the DMV. Whether a specific person is or isn’t a safe driver has nothing to do with government issued licenses.

    Unless you happen to live in the boondocks, your argument about the ability of a department store to coerce you into shopping there is a non starter. In today’s world the internet makes your argument empty of any meaning. Even in the far reaches of Alaska I enjoyed two grocery stores and yes they had to compete. Believe me, if there were a need for a competitor in that business, some entrepreneur, out of self-interest, would fill the void.

    Yes, everything governments do is coercive, that’s the nature of the beast. Coercion of people who are going about their business while not encroaching on others’ rights is always negative.

    In your argument concerning government’s role in policing criminal behavior you seem to misunderstand the difference between punishment and coercion. No one needs to coerce me into not stealing, killing, etc.

  12. Dave Pullin:

    I’m not glad the State coerces anyone to demonstrate a minimal driving ability. In a free market, drivers would be certified by private certification firms, not by the State’s Department of Motor Vehicles.
    Law, courts, and cops would be supplied by private law, court, and defense firms.

    There’s no such thing as a monopoly on the free market. The USPS has a state-granted monopoly of mail delivery. Patentees have state-granted monopolies on their inventions. Writers, musicians, and artists have state-granted monopolies (copyrights) on copies of their creative works. All these monopolies-crookopolies depend on the State’s use of force.

    Government is defined as a legalized monopoly of coercive force over an arbitrarily circumscribed geographical area, and which gains its resources by theft. Weber’s famous definition left out the last part.

  13. I’d like to go back to the interesting point Jerry O’Driscoll made. Armen Alchian did make an important contribution by emphasizing that firms search for prices and demands. Israel Kirzner’s theory of entrepreneurship takes this idea even further.

    So now think of this real-word example: People believe that they are eating too much junk food and that they lack self-control. Producers of junk food realize this. So they start to produce calorie-contolled portions of junk food (100 calories, etc.) In this way the phenomenon of eating the whole bag once it is opened is channeled in a better direction. (Let us avoid using the now-loaded word “nudge.”)

    The firm probably did some research to discover the demand for junk food under certain ancillary conditions. This demand, before the opportunities were presented to the public, was not an explicit or expressed demand. But it was, in a sense, there.

    Now is this paternalism or simply entrepreneurship? I guess if people like Thaler and Sunstein *approve* of the new product, they are tempted to call it paternalism.

    On the other hand,if they do *not approve*, perhaps it is just entrepreneurship. Consider: producers of junk food discover that the demand for junk food would increase if it were sold in packages with super-heroes
    on it. The demand for this is not immediately explicit — but it is there, in a sense. But since children’s behavior is moved in an unapproved direction it is not called paternalism.

    Now I ask: Is all of this talk about the difference between observed, expressed preferences and “true” or “informed” preferences really the issue. Clearing away all the excess: Isn’t it really a matter of approved demands versus unapproved demands?

  14. Bill Stepp: Interesting points. I’m glad you recognize that there are monopolies in the actual (US) market. We agree that they are no free markets – in fact far from it. And those monopolies (patents, copyright) only exist as a result of state coercion.

    I believe I have a good understanding of how a perfect market works (in theory) and that how competition forces an alignment of self-interest (or greed if you prefer) with wealth creation. I can see how much of the US economy does not even approximate to a free market.

    What I don’t understand is how you would expect a “completely free laisez faire market” to work. Companies buy their competitors so they don’t have to compete – so they end up with monopolies.

    Yes a private firm could certify drivers, but why would drivers bother to be certified, without coercion?

    We seem to have a highly dysfunctional system now – the worst of all worlds – markets that aren’t free, governments that don’t govern, but I don’t see “completely free laisez faire market” as any better.

    Can you help?

  15. Dave Pulllin,

    Can you provide one example, sans government involvement, when a company came to so dominate it’s particular market that it could have been considered a monopoly? Every one I can come up with existed only with the protection provided by government.

  16. Companies buy their competitors so they don’t have to compete – so they end up with monopolies.

    This isn’t the main reason why firms buy competitors. And I can’t think of a real world example where a company bought a competitor and thereby gained a monopoly.
    I think you misunderstand my point, although maybe I didn’t make it clear.
    What I’m saying is that in the real world, the only monopolies are government-granted monopolies, including regulated monopolies (e.g., utilities, although even these are eroding under some forms of regulated “competition”). You can’t point to a monopoly that eliminated competition by buying it out, or by any other contractual, market-based means. If you can, provide a company name that gained a monopoly by, for example, buying its competitor.

    As for drivers’ certifications, some people might not get certified to drive, just as not all accountants are CPAs. So what?
    Assuming the rule of law exists, they will drive by adhering to the rules of the road, or they will suffer appropriate legal consequences.
    Here in the People’s Republic of New York, a DWI conviction earns more or less a slap on the wrist, I assume because of the liquor lobby. A little rent seeking can have dangerous consequences. Get rid of the State, and watch rent seeking disappear and DWI become a smaller problem.

  17. “Companies buy their competitors so they don’t have to compete – so they end up with monopolies.”

    I suggest you buy some books on economics from the Mises institute and read them.

  18. Crawdad (re can I provide an example …):
    Yes – with a couple of caveats to ensure we are having a useful decision.

    1. There are no perfect monopolies, just like there is no perfect competition. Anti-trust defense lawyers always argue that there is a bigger market in which the putative monopolist plays in which they do not have a monopoly. Monopolies are better identified by the presence of the maladies of monopolies and the absence of the benefits of free competition, than they are by arguing whether 80%, 90% or 99.9999% dominance marks a monopoly.

    2. All companies exist to some extent by virtue of some government protection. The first requirement of economic development is the definition and enforcement of property rights. Trading efficiency is greater improved by the definition and regulation of symbols for wealth (aka money). Investment is greatly encouraged by shareholder limited liability protection. All these protections are usually provisioned and enforced by governments and are inherently coercive.

    I assume you are referring to government protection that is specific to the monopolist and not available to would be competitors.

    The example I give is Microsoft.

    Microsoft obtained a monopoly (in DOS) by something between “dumb luck” (their own lawyers words) and Bill Gate’s foresight or IBM’s lack thereof.

    From then on, they maintained their monopoly and leveraged that monopoly into new monopolies (Windows, Office, IE) using abusive practices that are well documented, and which had little to do with competition on the basis of the best product. AFIK there was no special government protection afforded Microsoft; indeed no special government action at all up to the anti-trust suit which, ironically, had the net effect of better securing, rather than destroying, their monopoly.

    IBM is another example.

    You may point out that both companies depend not just on property rights, but specifically Intellectual Property Rights (Patents and copyrights) which are arbitrarily defined by government edict, without any form of market valuation, and that is arguably very harmful. I would agree, but I don’t know that anyone has a better way to do it. Do you?

  19. Bill Stepp: I’m sure you will argue whether Oracle has obtained a monopoly in database software, but it’s acquisition of MySQL put it in control of a major competitor.

    Generally Oracle’s major competitor is seen to be SAP. The practical reality is that both Oracle and SAP have monopoly-like control over separate sets of customers, as evidenced by their ability to charge record high maintenance fees that earn gross profit margins above 90% (by not actually providing any maintenance). Oracle’s threat was not SAP, but customers migrating to open source databases – i.e. MySQL. But now Oracle owns MySQL so their customers cant escape. That’s not a free market; that’s monopolization.

    BTW I am not saying that acquisitions are always anti-competitive, nor that the only way to create a monopoly is buying your competitors. I’m saying it happens.

    What’s more, if that CEO of a company, had the opportunity to buy his competitor, such that he would obtain a monopoly and charge monopoly rent prices that would yield a net profit to his shareholders in excess of the cost of the acquisition, then he has fiduciary responsibility to his shareholders to do the acquisition. He should be fired and could be sued by his shareholders if he does not do it.

    Companies have no responsibility for maintaining the existence of competition, any more than they have for maintaining the existence of the spotted owl, or protecting the environment, or protecting the existence of the human race. They have only one responsibility which is to maximize profits.

    If maintaining competition, the spotted owl, or the human race, are seen as desirable objectives they have to be achieved by some mechanism other than (or in addition to) profit-maximizing companies. Isn’t that what governments are supposed to be for? (which is not the same question as whether they achieve it).

  20. Microsoft obtained a monopoly (in DOS) by something between “dumb luck” (their own lawyers words) and Bill Gate’s foresight or IBM’s lack thereof.

    From then on, they maintained their monopoly and leveraged that monopoly into new monopolies (Windows, Office, IE) using abusive practices that are well documented, and which had little to do with competition on the basis of the best product. AFIK there was no special government protection afforded Microsoft; indeed no special government action at all up to the anti-trust suit which, ironically, had the net effect of better securing, rather than destroying, their monopoly.

    MSFT never had a monopoly in an OS or in anything else. A programmer friend told me there are at least a dozen non-MSFT OSs available. The DOJ’s definition of a monopoly is 70% market share, which is absurd.
    How come MSFT’s stock price has gone nowhere since 1999?

  21. As for IBM, it’s long been the patent monopoly champion. The solution is to abolish patents and let competition prevail.
    BTW, MSFT also has some monopolies, as do other software firms.

  22. “70% market share, which is absurd.”

    Perfect free competition delivers specific benefits: it gives maximal genuine choice to individuals while efficiently allocating resources. It channels self-interest into alignment with genuine wealth creation. It rewards valued innovation.

    Perfect monopolies delivers specific maladies: it individual no choice. Deploys resources inefficiently, and self-interest becomes wealth taking, and stifles innovation.

    Your are right the transformation between one and the other does not occur between 69% and 71% market share. Nor between 99.9999 and 100% share. It does not occur when the putative monopolist sudden matches someone’s definition of “monopoly”.

    Rather the benefits decrease and the maladies increase along the spectrum between perfect free competition and perfect monopoly.

    Your programmer friend is technical right. There are many alternative OS — but that’s like saying the Ford would not have a monopoly if only Ford cars worked on roads. You could buy cars from any manufacturer as long as you didn’t want to drive on a road.

    MSFT had a sufficient monopoly to control the market without regard to the quality or price of the market. It made 90%+ profit margins on the product that was “in competition” with the dozens of alternatives your programmer friend told you about, which made no profit at all. It stifled innovation. For 15 years no VC would invest in an idea whose success was predicated on beating Microsoft.

    “How come MSFT’s stock price has gone nowhere since 1999?” Because by 1999 MSFT’s ability to extract future monopoly rent was built into the stock price. And because a monopolies ability to ignore customers requirements eventually catches up on them. The most obvious and clear evidence is the XP/Vista petition. What part of a free market requires customers to petition a supplier not to force them to take the “new” product and plead with a producer to provide the “old” one.

    During the life of MSFT, PC related products (both hardware and software) dropped in price/performance by several orders of magnitude — up to 10,000 times improvement. Meanwhile the MSFT products increased in price by 10 to 100 fold. So I can increase my price by 100 times while my competitor is dropping it by a similar or greater factor, but I’m not a monopolist?

  23. “The solution is to abolish patents and let competition prevail.”

    Yes but the contrary argument was merit too. Inventors create wealth. Without patents or copyright, they cannot appropriate any of the wealth they create. If they can’t get paid for their work, why would they do it?

    So you can explain how a market for inventions would work?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s