Open threads at the Open Thread tab every Sunday and Wednesday

Inefficient Hot Dogs

I know the “why do hot dogs come in packs of ten and buns in packs of eight” question is beloved of boring observational comedians everywhere, and has become a cliched example of people who don’t have anything more interesting to talk about.

But from an economic perspective, why do hot dogs come in packs of ten and buns in packs of eight? Last weekend I spent about ten minutes laboriously looking through all the hot dog buns trying to find one in a pack of ten so it would match my hot dogs. I finally gave up and bought one of the twenty or so different packs-of-eight. If there had been a ten-pack, I would have bought it instantly.

This sort of equilibrium shouldn’t be able to last ten seconds in a free market. It should literally have taken ten seconds from the moment the hot dog bun company executive noticed all the other companies sold their buns in an inconvenient number to the time she decided to make her company’s product stand out by offering what consumers are looking for.

I know it seems trivial. But then we posit theories like “Companies will react to consumers’ desire for safe goods by monitoring their own safety standards” or “Businesses can’t remain sexist over the long term, because refusing to hire competent women who are otherwise neglected would be like leaving a hundred dollar bill on the ground”.

And both of those arguments sound like they should work, unless of course for some mysterious reason entire industries can just completely fail to respond to incentives for indefinite periods of time based on some stupid reason like tradition. Right now the hot dog situation seems like an especially good example of evidence in favor of this possibility.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

54 Responses to Inefficient Hot Dogs

  1. Mestroyer says:

    If you buy 10 hotdogs and 8 buns, then you end up with two hotdogs left over, and then you want to buy more buns. The only way you consume all that you have (unless you eat one without the other) is to (a) cooperate with other consumers (overhead is way more costly than a couple of hotdogs) or (b) eat 40.

    This sounds like the same kind of thinking as video game companies that have their own currency, which they force you to buy with real money before buying their games with it, and selling the currency in discrete packets that don’t align with the prices of games. Exploit sunk cost fallacy.

    Could hot-dog and bun manufacturers be cooperating? As long each of them sticks to the inconvenient package sizes, they both benefit. An individual company could gain a temporary benefit by defecting and siding with the customer, but pretty soon the others would follow suit, and their defector’s advantage would be gone, and so would the gains they got from cooperating. I think it’s plausible that the expected benefit of future gains from cooperating to an individual is higher than the benefit of a short burst of defector’s payoff.

    • Scott Alexander says:

      I did consider that, but I reject your cooperation scenario for pretty much the same reason I reject most group selection. Even real smoke-filled-room collusion couldn’t work that well for that long. Or if it could, that level of success would be a way bigger threat to capitalism than a few people stubbornly ignoring incentives.

      • Mestroyer says:

        This is different from group solection because hot-dog and bun manufacturers have foresight, and evolution does not.

        How deadly would it be to capitalism if certain markets could occasionally pull this off? It’s not like through cooperation hot-dog and bun manufacturers could raise their prices to whatever they wanted, because they’re still competing with everyone else who sells food.

        I find it very implausible that whoever decides packaging sizes is stupid enough to not notice this or notice and stick with their existing package size out of tradition or something. I think it’s basically guaranteed someone with the power to change this has heard the question if boring observational comedians are using it (Friends/family would have relayed it to them).

        How many distinct companies making buns and hot-dogs are there?

      • FWIW, I was impressed by this post, and told an MBA friend of mine about it. She immediately had the same thought as Mestroyer. According to her, this kind of implicit collusion between companies does in fact happen in the real world, e.g. the point of “we’ll match any competitor’s price” offers isn’t so much about bringing in customers, but letting competitors know not get into a price war.

      • Max says:

        I’m pretty sure it wouldn’t take a lot of searching to find ways in which industries routinely operate on tacit collusion between companies. The businesses all recognize that it’s not in their best interests to instigate a race to the bottom, so they don’t defect, because they understand that the temporary advantage from doing so would be outweighed by the long term loss of profits as their competitors followed suit.

        Any product whose retail price is conventionally marked up orders of magnitude from its cost of production is likely to be an example. Companies also could defect, but don’t, in equilibria built around planned obsolescence, like razor blades which are designed to need frequent replacement.

        • Anonymous says:

          Seems like an effective altruist should identify those products and start companies that offer them at the lowest sustainable price level.

      • Jack V says:

        I think tacit cooperation definitely exists, but this case just seems too complicated for it to be worth it.

        Although come to think of it, it might be something similar. Say, there’s some other reason (tradition?) hot dogs come in tens. And
        bun makers sell in eights or twenties so you have to buy twenty?

        But I’m not sure — how many people always have an exact one to one ratio? I don’t expect that to happen in real life, I don’t know if it does.

      • Deiseach says:

        Which came first, the buns or the hot dogs? I mean, if hot dog rolls are a type of bread roll, and they were being made before mass-market hot dogs came on the market, it could just be “the way it is” – they were always sold in packs of eight.

        • I think there’s some of that, some explanation in terms of different traditional standards in the meat industry vs. the bread industry. But implicit collusion can still explain why it’s lasted as long as it has.

    • Yes, but this argument relies on all bun and hot-dog manufacturers cooperating over a very long term. The first one to defect would reap enormous profits as everyone said “Oh look, a ten-pack of buns” (or “eight-pack of hotdogs”, as the case might be) and bought only that brand. You can’t tell me that so large an industry as hot-dog manufacturing doesn’t contain a couple of selfish bastards who’ll say “screw the long-term benefits of maintaining the sunk-cost fallacy for the whole industry – I’m going to get a nice short-term profit and retire on my stock options in three years”.

    • suntzuanime says:

      So then does this mean companies could implicitly collude to produce unsafe products, or by not hiring competent women?

      That latter seems plausible – if you’re a bigshot businessman good-ol’-boy, and your friends are bigshot businessment good-ol’-boys, then on the one hand you could impress them by becoming super profitable, but on the other hand if you had to do something unmanly like hiring and promoting a bunch of women to do it, the shame and emasculation might outweigh it and act as a social pressure to conform.

      We can test this hypothesis if seeing where firms where the people who have power are relatively unlinked to the larger social networks of people who have power are relatively less gender-biased.

  2. randallsquared says:

    Maybe the problem is that the *other* person who would buy a pack of hot dogs because the count matches the count of bun packages doesn’t shop for hot dogs that often. The amount of money they’re leaving on the ground is probably very low. Some evidence is that Oscar Meyer apparently gets only 10-15 questions about this a year: http://www.straightdope.com/columns/read/560/why-do-hot-dogs-come-10-to-a-pack-while-buns-are-8-to-a-pack

    • Max says:

      But how many people wonder and care compared to the number who actually try to contact the company? Most people I know have never written in to a company to ask about *anything.*

  3. Sniffnoy says:

    One thing worth noting in the Straight Dope column is that Continental Baking *does* make 10-packs of hot dog buns, but only sells them in places where there’s high demand for them. The question then becomes, if you’re doing 10-packs, why not switch to only 10-packs? (Or switch to 10-packs by default and 8-packs where there’s high demand for them.) It does seem like there’s something going on here, some reason they must have for preferring 8 packs.

    The old “now you have to get 40 of each” reason brought up by Mestroyer is the obvious one but whether it’s right I obviously have no idea.

    • Leonard says:

      It does seem like there’s something going on here, some reason they must have for preferring 8 packs.

      Second. They can already do what you want, and they even do it — in limited circumstances. That they do not in all circumstances suggests that you don’t understand the problem correctly.

  4. anon1 says:

    Hot dogs without buns are pretty tasty (maybe better than with the bun), and buns without hot dogs are no good at all. Slightly fewer buns than hot dogs seems efficient to me, and if all bun packages must contain the same number then eight is a pretty reasonable number.

    Of course it’s still silly that there aren’t alternative packages of 10 for people who really like buns with their hot dogs.

    • Anonymous says:

      I disagree that buns alone are useless. You can put any other regular sandwich stuff into them for a little sub sandwich. Do it all the time.

  5. jimrandomh says:

    As others have pointed out, as long as everyone’s selling mismatched packages, this benefits the sellers by incentivizing people to buy buns for their leftover hot dogs and vise versa. This should be outweighed by the gains from defeating competition and capturing the whole market, and it would be, if economic competition were like a prisoner’s dilemma. But it isn’t; economic competition is like an *iterated* prisoner’s dilemma. If sellers are mostly using tit-for-tat and competing in small groups, then defecting (by lowering prices or fixing package sizes) is unprofitable.

  6. Maybe the bun makers try to fix the problem, but the retailers (whose incentives are less clear) won’t cooperate.

    Note that Smart Dogs come in packages of eight.

  7. MugaSofer says:

    In my experience, there’s usually someone hungrier than the other who snatches up “spare” hotdogs. But then, it’s usually me, so there’s a slight selection bias…

  8. misha says:

    larger hot dogs often come in smaller number packages, as well as brats and other sausages. It’s possible 8 buns is the optimal middle ground that gets the package grabbed by people buying the most kinds of processed meat.

    it’s also possible that 10 buns per package would be unwieldy. 2×5 rows might just bend and break buns a lot more than 2×4.

  9. ciil says:

    Shouldn’t the real question be: Why don’t hot dogs come in packs of 17 and buns in packs of 13?

  10. mareofnight says:

    If I remember correctly, hot dog buns usually come sort of… connected to each other? Could it have to do with the numbers their machines are built to produce, so that switching would be expensive? Or perhaps the size and shape of shelf space that stores designate for those products?

  11. For a more fraught and less well-defined example: clothing for fat women is well short of the demand. I’ve been hearing about this problem since the seventies.

    30-odd years later, there’s much more available online (online didn’t exist at all back then) and as relatively expensive clothing. Off-the-rack, and especially cheap off-the-rack clothing, tends to be hard to find, badly made, and unfashionable.

    I have no doubt this is partly a matter of prejudice. I would expect that even in a perfectly unprejudiced world, large clothing (and especially very large clothing) would be somewhat more expensive and less varied, but the difference would not be nearly as extreme.

    There’s also the matter of the lack of standardized sizing for women’s clothing, but that’s a hard coordination problem.

    However, I think there are some other factors as well. One is simply that there are always more plausible ways of improving a business than there’s resources to implement them. There’s also the plausible premise that people in business aren’t necessarily looking to make as much money as possible– it seems more likely that they’re trying to make money without thinking too hard or taking too many risks.

    A depiction of Lane Bryant being a solution for a while….

    http://www.xojane.com/it-happened-to-me/it-happened-to-me-working-at-a-plus-size-clothing-retailer-taught-me-to-love-my-body

    • Scott Alexander says:

      The first thought that comes to my mind here is that if a clothing company or clothing store does a good job making/distributing clothing for fat woman, it will become known as “the clothing company for fat women”, and that association will kill its business among thin women who want to avoid signaling that they are fat.

      This still doesn’t explain why all-new fat-only clothing companies don’t spring up, but it at least raises the transaction costs a little.

      • There are some businesses that offer clothes over a wide range of sizes– Land’s End has a plus-sized catalog.

        There was a company which offered clothes for teenagers that actually took a survey about whether their existing customers would mind if the stores offered larger clothing, and the customers said they didn’t mind. I’ll see if I can find out which company it was and how it worked out.

        So far as stigma and products is concerned, In Search of Excellence mentioned that Cadillac was the first car company which was willing to sell a high end car directly to black people. They’d noticed that well-off black people were using white agents to buy Cadillacs.

        Fast forward some decades, and Cadillacs are known as pimpmobiles. I don’t know whether this affected sales.

        It’s going to be legal relatively soon for crowd-funding to be available for stock in new companies– I’m hoping that this will make it easier for large-sized clothing to be available.

        • Creutzer says:

          There was a company which offered clothes for teenagers that actually took a survey about whether their existing customers would mind if the stores offered larger clothing, and the customers said they didn’t mind.

          Surely they all had an accurate self-model and were being honest…

        • mareofnight says:

          Abercrombie seems to believe that will happen, but I don’t think that’s very good evidence here. Marketers can be wrong. Do we know of any instances of stores that had sales of small clothing drop when they introduced plus-size clothing?

          I’d expect carrying larger sizes to have different effects (or at least differently sized effects) on different stores as well. For example, a store whose brand is selling clothes for cool teens, or for athletes, would have a harder time selling a wide range of sizes than a store whose brand focuses on comfort or quirkiness would.

        • Paul, your link mentions that H&M, which offers large sizes, is doing well. All that the link proves is that Abercrombie and Fitch *worries* that large sizes may hurt its image.

    • Damien says:

      “standardized sizing”

      Why not just use actual measurements, like men’s pants? If I wanted to go start a business I’d totally consider re-selling women’s clothing with measurements.

      • naath says:

        Mostly I think it’s a matter of there being too many measurements needed (different shops have different notions about waist:hip rations, thigh sizes, trouser length, etc.) as well as differences of opinion in how closely garments should fit.

        Alas the cost of producing clothing bespoke to the client’s own measurements and tastes is rather greater than most people are willing to pay for garments.

      • BlindMind says:

        The measurements on men’s pants are not really actual measurements:

        http://www.esquire.com/blogs/mens-fashion/pants-size-chart-090710

    • lmm says:

      Interestingly shops always seem to sell clothes at the same price whichever size they’re in. Might it be that larger versions of the same clothes would be more expensive to produce/stock/etc., but it would be politically unacceptable to charge more for them?

  12. Erik says:

    Cultural tangent: In my local experience in this part of Norway, hot dog buns are sold in packs of eight, but hot dog buns are generally not what people eat hot dogs with. Hot dogs are eaten with ‘lompe’ (a sort of pancake-like soft thin flatbread you wrap around the dog), which is sold in packs of ten. The only times I’ve seen hot dog buns in use were at large parties, camping events, picnics etc. where hot dogs, buns and lompe were all bought in large quantities and leftovers were inevitable no matter the exact quantity.

  13. Rob Miles says:

    Perhaps they’re assuming that in a given hot dog eating group, 20% will be Discordians, who are of course religiously obligated to Partake of No Hotdog Buns.

  14. Rob says:

    What a lazy article. Took me 2 seconds to find out that Ball Park Franks is owned by Sara Lee which makes…. BUNS. Betting most all the dog makers are in turn owned by companies that also makes buns.

    So this drivel about sexism & consumer mechanics – while interesting & worthwhile to discuss – is completely irrelevant to the issue of dog:bun ratios. The unwillingness of this author to do the most cursory of investigations into this topic is frustrating & plays into the overall dumbing-down of journalism.

    C’mon buddy! Put a little effort into your work.

    • Tyrrell McAllister says:

      What a lazy comment. Took me 2 seconds to find out that Sara Lee acquired Ball Park Franks in 1989… TWO YEARS after the article was written. Betting most all journalists do not reference facts that only because true after the time of their writing.

      So this drivel about the dumbing-down of journalism – while interesting & worthwhile to discuss – is completely irrelevant to the issue of whether bun-makers make hotdog. The unwillingness of this commenter to do the most cursory of investigations into this topic is frustrating & plays into the overall dumbing-down of blog-reader commentary.

      C’mon buddy! Put a little effort into your work.

      :). Just kidding (mostly). You still raise a relevant point. Even while this fact was not yet true at the time the article was written, the eventual truth of the fact does represent a failed prediction of the theory offered by the journalist.

      • Damien says:

        But who did Sara Lee acquire Ballpark Franks from? Was the latter independent and had it been so for a long time?

        • Douglas Knight says:

          Sara Lee acquired Hygrade Foods from Hanson, as you could have learned from wikipedia. Hygrade Foods appears to have sold only meat. Hanson owned other food divisions, but probably didn’t coordinate them much (eg, it acquired some in 1986 with Imperial Tobacco and immediately sold them off). It held Hygrade for 13 years.

  15. When I first came to the US sixteen years ago, ARCO gas stations were much cheaper than all other chains in California. In a free market, you would expect either other chains to drop their prices, or ARCO to take 100% of the market. But in 16 years, nothing happened: the percentage of ARCOs among California gas stations seems roughly the same.

    • Tab Atkins says:

      A personal anecdote relevant to this: When I moved from Texas to California, I had to roadtrip there, as I had pets that couldn’t stand a flight. At one point on day 3 or so I needed gas, and stopped at an ARCO. They didn’t seem to have a card-reader at the pump, so I went inside, only to discover that they simply *don’t take credit cards*. Cash only. We didn’t have any cash, so we had to go to the *next town over* to find another station, and hope we didn’t run out in the meantime.

      I’m still not entirely certain if this is a company-wide thing or just a quirk of that particular station, but since then I’ve sworn off of ARCO entirely in protest. Most places in California offer a cash discount, but to refuse CCs *entirely* is just bullshit. If it *is* a company-wide thing, that might help explain why they haven’t gotten too popular.

  16. Phil Goetz says:

    This is similar to what I said in my LessWrong post “The cup-holder paradox“. I have another example: Mint toothpaste. I’ve always hated mint toothpaste because it’s much too strong. I supposed I was just odd, but recently I read in one of A.J. Jacobs’ books about his attempt to optimize toothbrushing by finding a non-nasty toothpaste. The only one he found was Tom’s of Maine. (I’ve gone through that same process, and Tom’s of Maine is also the only popular toothpaste not approved by the American Dental Association. My dentist warned me against it, saying it’s abrasive and could damage my gums.) AJ mentioned this to someone else in the book, who also said he hated mint toothpaste.

    So I’ve started bringing this up with other people, and so far, everyone I’ve talked to hates mint toothpaste. I asked my hygenist for unflavored toothpaste at my most-recent tooth cleaning, and she said they can’t find it anymore, and that she herself never has her teeth cleaned because the mint flavor is too revolting.

    Yet in an American drugstore, you can find thirty varieties of mint-flavored toothpaste, some non-mint, high-priced Tom’s of Maine, and nothing else.

    • Douglas Knight says:

      I’m reposting because I think the spam filter ate my previous comment. But without the link.

      Are you really complaining about the mint being too strong, or about it being too sweet?

      Here are two hypotheses: (1) you and everyone you talked to like sweet things less than the general public; (2) children like sweet things (or maybe minty things) more than adults, but people get fixed very early, either on a brand, or on their sense of what a toothpaste should taste like.

    • nydwracu says:

      They used to have cinnamon toothpaste; is that not around anymore? (I always thought it was worse than mint, but I don’t mind mint toothpaste… and Wikipedia says “case reports of plasma cell gingivitis have been reported with the use of herbal toothpaste containing cinnamon” so that may explain that. Is ‘herbal toothpaste’ relevantly different from regular?)

      My question for toothpaste is: why doesn’t everyone buy Aim? Is there some difference in ADA-approved toothpaste brands that makes it worth spending three to four times as much on a better-known brand, a difference that just about everyone but me knows about? Or is it just that people tend to be suspicious of products priced at a quarter of the conventional cost?

    • I like mint toothpaste, and I spent part of my childhood looking for stronger and stronger mints. Vick’s had a light blue cough drop which was the best. I’m not sure whether it’s still made. Altoid’s “curiously strong” mints aren’t.

      Anyway, if you’re handy to Philadelphia, Terralyn at the Reading Terminal Market makes a toothpaste that has little or no mint in it.

      I’m not sure whether she does mail order.
      https://www.facebook.com/pages/Terralyn-Naturally/178834865489904
      (215) 922-3170

  17. Cadmium says:

    Correct me if I’m wrong, but the/an argument Scott seems to be making resembles the following:

    As someone who is partial to GMU-economist type arguments, the counter-argument initially struck me as compelling. On further reflection and discussion, though, I think there a couple of points to be made.

    As has been pointed out, we don’t know what kind of costs that a large-scale baker, such as the makers of Wonder bread, would have to incur in order to gain the obvious first-mover advantage. It might involve a complete re-jigging of ovens or shipping containers or something.

    Partly because we don’t know this, we also don’t know the decision matrix emplyed by the board and CEO of Wonder when considering whether to implement the change across the board. The change might take four years to turn a profit, and the CEO’s compensation scheme might motivate her to take a shorter-term view.

    On the other hand, I don’t need to know the specifics of anyone’s decision-making process in order to profit from what I believe is a clear bias in the market. All I have to do is convince someone that actually knows something about finance to craft me an investment vehicle that is weighted toward companies that hire more women and weighted even stronger for corporations with women on the board or in senior management. I can use this heuristic to profit from a broad-based strategy.

    I don’t do this because I’m confident that most ongoing gender discrepancies in wages and hiring are, in fact, not due to bias (again, in the Bayesian sense, which could include some rational decisions that could be labelled “sexist”).

    Failures to respond to incentives occur for different reasons. While the failure to respond to the hot dog discrepancy may be “mysterious” and seems like a market failure because no obvious explanation presents itself, this does not mean that “market failure” should become the go-to option when we think we see a bias, particularly one this politically charged.

    Obviously, the deductive argument does strike a blow against the infallibility of the free market as a problem-solver, but since most of us are at least willing to acknowledge the concept of “market failure”, I think it is safe to say that we would likely not posit such absolute infallibility in the first place.

    • Cadmium says:

      Sorry for my HTML incompetence – here is how the comment should look.

      Correct me if I’m wrong, but the/an argument Scott seems to be making resembles the following:

      “Conservatives and reactionaries claim that lingering gender discrepancies between wages and hiring are not primarily the result of irrational sexism (i.e. sexism that constitutes a bias in the Bayesian sense).

      Libertarians and GMU-economist types provide them with a supporting argument: corporations are profit motivated. Any overt bias should be corrected in short order by the ruthless profit-maximizers. It would be as easy as picking a hundred-dollar bill off the ground. Because corporations have not taken action to hire women or pay them more and large investors have not demanded corporations take such steps, there must be no bias to be corrected for.

      This argument is essentially deductive. Since we know that one sound counterexample negates such an argument, the conservative position can be negated by proving the existence an analogous bias that remains unaccounted for.

      The first actor to take action to market hot dog buns in 10-packs would reap huge sales benefits. Nonetheless, there has been no overt action to market hot dog buns in this way. The potential for profit here is so obvious that failure to act on it can only be the result of bias on the part of the profit-motivated individuals. Because the deductive argument fails here, it cannot be used to justify ongoing gender-based wage and hiring disparities.”

      As someone who is partial to GMU-economist type arguments, the counter-argument initially struck me as compelling. On further reflection and discussion, though, I think there a couple of points to be made.

      As has been pointed out, we don’t know what kind of costs that a large-scale baker, such as the makers of Wonder bread, would have to incur in order to gain the obvious first-mover advantage. It might involve a complete re-jigging of ovens or shipping containers or something.

      Partly because we don’t know this, we also don’t know the decision matrix emplyed by the board and CEO of Wonder when considering whether to implement the change across the board. The change might take four years to turn a profit, and the CEO’s compensation scheme might motivate her to take a shorter-term view.

      On the other hand, I don’t need to know the specifics of anyone’s decision-making process in order to profit from what I believe is a clear bias in the market. All I have to do is convince someone that actually knows something about finance to craft me an investment vehicle that is weighted toward companies that hire more women and weighted even stronger for corporations with women on the board or in senior management. I can use this heuristic to profit from a broad-based strategy.

      I don’t do this because I’m confident that most ongoing gender discrepancies in wages and hiring are, in fact, not due to bias (again, in the Bayesian sense, which could include some rational decisions that could be labelled “sexist”).

      Failures to respond to incentives occur for different reasons. While the failure to respond to the hot dog discrepancy may be “mysterious” and seems like a market failure because no obvious explanation presents itself, this does not mean that “market failure” should become the go-to option when we think we see a bias, particularly one this politically charged.

      Obviously, the deductive argument does strike a blow against the infallibility of the free market as a problem-solver, but since most of us are at least willing to acknowledge the concept of “market failure”, I think it is safe to say that we would likely not posit such absolute infallibility in the first place.

  18. Douglas Knight says:

    Who is more ruthless than CEOs?

    Economists !

    In the 50s and 60s, Greenspan found that women economists were underpriced and hired a lot of them. I’m not sure about the 70s and 80s. The earliest source I have for this is this 1983 profile. I am somewhat surprised that this isn’t the standard example in econ textbooks.

    I always valued men and women equally, and I found that because others did not, good women economists were cheaper than men. Hiring women does two things: It gives us better quality work for less money, and it raises the market value of women.