### Rational Riddles

Many years ago, when soft pretzels were available on every street corner in downtown Philadelphia at the going price of

Ten Cents Apiece/Three for a Quarter
there was one vendor who occupied a prime location in the City Hall courtyard, and was therefore able to command a premium price. His sign read
Ten Cents Apiece/Two for a Quarter

I always thought we could explain that one away as a case of poor math skills. But now our frequent (and frequently brilliant!) commenter Thomas Bayes sends along this photo of a sign that he recently spotted at a gas station convenience store, and which I’m finding a little harder to get my head around:

Thomas reports:

I asked the person behind the counter if she could sell me one pack for \$1. She said no. I asked if she would throw one of the packs away for me if I bought two. She seemed genuinely puzzled. I drive an SUV, so I wish they’d apply this scheme to the gas they sell.

Here’s your chance to get creative. Give me an explanation consistent with rational behavior and orthodox economic theory.

#### 55 Responses to “Rational Riddles”

1. 1 1 Phil

Maybe they mean buy 2 for \$1.00 EACH.

2. 2 2 Jonatan

Lets assume the two comes in packages of two. This could include one type of m&ms that customers never buy, but that the shopkeeper thinks they could develop a taste for if they just tried it.

3. 3 3 dave

the normal (retail profit margin markup) price of the m&m’s is 50 cents. this sale is just a device to encourage you to think about buying m&m’s.

4. 4 4 Xorph

I see two possible and not mutually exclusive explanations
- “\$1.09″ is a complicated and weird price
- “2 for \$1″ looks like a bargain compared to the first weird & expensive offer. Wow, a bargain ! Can’t miss this !

5. 5 5 Jonatan

@Dave and Xorph: These suggestions assume that the customer is irrational, don’t they? I don’t think you are wrong in that assumption, but the riddle stated that we should assume rational behavior.

6. 6 6 prior probability

Dude, maybe people are just plain stupid

7. 7 7 Scott H.

@4 Xorph…

Yes. It’s not just a bargain; it’s an irrationally good bargain! Who among us can resist the irrationally good bargain?

8. 8 8 Andy

Maybe they are conducting an experiment on rationallity

9. 9 9 Ron

I think Phil @1 has it. Management meant the sign to be
\$1 EACH, but the counter people take it literally.
Good thing they didn’t have a sign, “All M&Ms \$1″.

Second most likely: The candy is nearing its expiration
date. Instead of simply pricing them at 50 cents each,
they price per the sign to encourage people to buy two
at once and help turn over the stock.

10. 10 10 RPLong

I think Dave @3 has it right (also Ron’s second explanation @9). They list the M&Ms at regular price and then say, “But if you buy TWO, I’ll sell it to you for a dollar!” They just want to sell a bunch of M&Ms at a discount. They’re trying to get them off the shelves.

My time spent working at convenience stores makes me think that the manager accidentally ordered too many M&Ms and now they’re taking up valuable shelf space. There isn’t a lot of room for storage of inventory in convenience stores, so if you end up with too much of a slow-moving product it becomes “opportunity-costly” to store it. Gotta get rid of it so that it can be replaced with a better-selling item.

Having said that, the cashier was definitely not behaving rationally. I think the cashier just missed the point of the special.

11. 11 11 FC

M&Ms are selling at 40c each at the competitor’s. And are fresher. This guy wants to clear inferior product quickly at a premium price.

(FWIW: I had a company that looked for fake products on the internet e.g. xerox cartridges. The most reliable indicator for a fake was that it was listed for a higher price than the real one, as the rational expectation would be quality+brand commands a premium).

12. 12 12 Smylers

RPLong @10, you’re presuming that \$1.09 is the “regular” price. The only thing we know is that it’s currently the price for one bag; it could be that before this special offer the price for a single bag was lower, and has been increased to make the offer look better and encourage people to buy more bags.

The UK consumer magazine ‘Which?’ highlighted that Asda (owned by Wal-Mart) sold Müller yoghurts at 30p each, then increased the price to 61p each for the duration of an “offer” where you could buy 10 yoghurts for £4:
http://press.which.co.uk/whichpressreleases/supermarket-special-offers-not-so-special/

13. 13 13 Tony N

The convenience store is engaging in a form of confectionary stimulus: it’s attempting to increase M&M2

14. 14 14 Dave Smith

Ron #9 has it. Inventory reduction. It’s loosely akin to those buy one get one 50% off sales….why not just put everything at 25% off?

15. 15 15 ThomasBayes

Thanks to everyone for the comments and insights. According to the person at the counter, they were trying to clear inventory from a previous M&M promotion that didn’t go as well as they had planned.

My best guess about this is that they wanted to move the M&Ms by selling two packs for \$1.00. Had they simply put that on the sign, though, they would have had to deal with people wanting to buy one pack for \$0.50. To avoid that “confusion”, they posted the price for one pack. I probably wouldn’t remember the sign today if they had posted a “buy one for \$1, get one free” offer, or if the price for one pack was less than or equal to the price for two packs. (Assuming, of course, that the price for one pack wasn’t less than half the price for two packs.) But I still struggle with a rational explanation for offering two packs for less than the price of one. I could see this for some goods or services, but with M&Ms you could simply throw one pack away on your way out the door if you wanted one instead of two. Or, if you don’t like to waste, you could leave one pack on the counter.

On a related note, I recall seeing an advertisement from a car dealer announcing that they were having a half-price sale on cars. Sure enough, the prices they listed below each of the advertised cars was half what you would have expected the prices to be. In fine print on the bottom of the page, there was a note that read “Half-price sale means that the price you see is half the price you pay.”

16. 16 16 Pat T

Price discrimination?

Really fat people are poorer than thin people. People who want two bags tend to be poorer and fatter. Giving them a cheap price allows you to compete better with BJ’s/Sam’s Club bulk drums of chocolate awesomeness

Thinner wealthier people pay a premium to not have the temptation of a second bag screaming C’MON EAT ME ALREADY YOU KNOW YOU WANT TO.

This is a half-baked wild guess.

17. 17 17 Steve Reilly

Didn’t Lewis Carroll already solve this one?

‘I should like to buy an egg, please,’ she said timidly. ‘How do you sell them?’
‘Fivepence farthing for one—Twopence for two,’ the Sheep replied.
‘Then two are cheaper than one?’ Alice said in a surprised tone, taking out her purse.
‘Only you MUST eat them both, if you buy two,’ said the Sheep.
‘Then I’ll have ONE, please,’ said Alice, as she put the money down on the counter. For she thought to herself, ‘They mightn’t be at all nice, you know.’

18. 18 18 Rflood

It’s a fun conversation starter – not a real price anomaly.

19. 19 19 RJ

I thought inventory reduction would be the reason, but I too still struggled with why they wouldn’t sell 2 for \$1.09 and then 1 for \$1.

I think now it has to do with the fact that they want to clear inventory so badly that they’re trying to discourage people from just buying one. Assuming M&M’s cost at least \$0.50 to make and deliver, \$1.00 is breaking even and they don’t lose any money. This way, they could care less if you throw one pack away because they still get their dollar.

In response to Bayes question “What if they just leave it on the counter?” Well, then that’s great! They can sell that one at a profit now if someone partakes in the 2 for \$1 dollar deal since they’re using the same bag someone left behind.

20. 20 20 RJ

By the way, this reminds me of that Beavis and Butthead video from back in the day.

Choc full of economic and accounting fallacies :)

21. 21 21 Dave

“But now our frequent (and frequently brilliant!) commenter Thomas Bayes…”

I’m pretty sure “Thomas Bayes” doesn’t want to be considered “frequently” anything…

22. 22 22 Roger

At grocery stores where I live, a sign says “2 for \$1″ does mean that the store will sell the items for 50 cents apiece. Unless the sign also has a price for single item.

I have never seen the price of 2 being less than the price of 1, but it would not surprise me greatly if it occasionally happened that the price of 2 with a coupon (which might even be picked up free in the front of the story) is less the the price of 1 without a coupon.

23. 23 23 Scott F

They have an extremely burdonsome inventory of M&Ms through an irreversible clerical error from the Mars distributor. They don’t want to come out and tell the world this verbatim for fear of being robbed by ravenous M&M consumers.
They assume that when consumers see a good deal, even if idiotic, they will take advantage of it. Thus, they are trying to diminish their overabbundance of M&M’s as quickly as possible while minimizing the risk of being robbed. All under the guise of complete stupidity. Based on Thomas’ comment, the ruse is working.

24. 24 24 BC

The store’s main inventory system was designed to handle only even numbers of packs of M&M’s. (If the number of packs of M&M’s is known to be an even number, then the computer memory required is less by 1 bit, leading to a cheaper system.) Whenever the store has an odd number of M&M packs, they need to employ additional inventory processes, the cost of which they estimate to be between 0.01 and 0.09. The store is also inefficient at disposing unopened M&M packs, which also costs them between 0.01 and 0.09 to dispose.

That’s why the clerk refused to sell you 1 pack for 1.00 or to just throw away 1 pack if you bought two for 1.00. Either option would cost the store 0.01-0.09, either in additional inventory accounting costs or in disposal costs. Thus, they will only do it if you are willing to pay an additional 0.09.

If you only want 1 pack, you can buy 2 for 1.00 and dispose of the unused pack yourself rather than having the store do it, thus taking advantage of your lower (zero) cost of M&M disposal. We see here an example of market efficiency where those with lower disposal costs than the store are the ones that end up disposing of unwanted M&M’s.

25. 25 25 Scott F

@BC
This would also lead to the, quite testable, hypothesis that they would not have garbage cans outside the station, or that they would fill them with disgusting and hazardous waste to prevent scavenging of disposed bags.

26. 26 26 Jimbino

Hell, salesmanship is everything. When I sold Fuller Brushes door-to-door, I could sell anything labeled as a “special.” I sold 29 cent toothbrushes in a two-dozen-pack “on special” for \$7.19 and got zillions of takers. I combined spray cans with prices varying from 89 cents to \$1.89 to sell “on special” at 2 for \$3.79 or 3 for \$5.19.

There’s many a sucker who can’t add and who won’t admit it. Those are the ones who no doubt support all the ripoffs like Obamacare.

The important thing was that success at door-to-door sales amounted to spending some time identifying the suckers. Then you don’t have to ever again go door-to-door; you only need to return to soak the same suckers. Even that process could be accelerated by knocking only at the houses that had a “NO SOLICITORS” sign posted, as that instantly identified a housewife mark whose husband knew he’d had married a sucker.

27. 27 27 Jonathan Finegold

The first one you mention, my guess is that the person was basing their sign on other peoples’ and just marked down the “three for 25 cents” to “two for 25 cents,” without realizing that he was giving people a worse deal.

The only thing I can think of for the second one is that 1 hotdog is not really worth \$1.09, but if they put that and offer 2 for \$1 then the latter looks like a really good deal.

28. 28 28 Jonathan Finegold

*Err, M&Ms, not hot dogs.

29. 29 29 Robert

After two m&m’s you are more likely to then go and purchase a salty treat from the same vendor.

30. 30 30 ken p

This is commonplace in the corporate world. What happened was a consultant was hired to find efficiencies. Part of the resulting report showed the total expenses of the store, divided by the number of transactions as transaction expense.
Some brilliant new manager has decided to reduce the M&M transactions by selling two bags at a time, thus reducing expenses!
Logic leapt, ingenuity established, promotion assured.

31. 31 31 Gordon Landwirth

An answer to the question requires rationality and consistency with orthodox economic behavior on the part of both buyer and seller. So, while offering a lower total price for higher quantity may make sense for a couple of reasons given, I haven’t seen the buyer’s perspective accounted for in comments yet (unless I missed it as I scanned through quickly).

First, from the seller’s perspective, other commenters have already mentioned upcoming expiration of a perishable item, and desire to clear inventory to free up space for other items. More generally, the seller may face storage and (alternatively) disposal costs that are both so high that the seller makes more profit selling 2 for less revenue vs. selling 1 for more revenue.

From the buyer’s perspective, one would normally expect that every buyer would prefer more for less, making the price for 1 unit useless and thus irrational, but: The buyer could also face high storage and disposal costs, and additionally, I can tell you, as someone who counts calories to maintain my fitness level, a consumer may not want the temptation of a second unit of M&Ms even if he were, in effect, paid to take that second unit — or stated more generally, there may be negative marginal utility as the quantity is increased above a particular quantity (in this case, 1 unit).

32. 32 32 Harold

We have another Ken! You can never have too many. I don’t think anyone has nailed a convincing rational argument yet.

33. 33 33 Harold

The explanations for the consumers’ rationality seem to be based on his realisation of his own irrationality, including that of Gordon Landwirth. Sort of “I know I am not rational enough to throw away a bag of M&M’s even if I don’t want them, so I had better pay extra and only get one bag”. This does not make it rational.

Many explanations for seemingly irrational economic behaviour seem to fall into this catagory.

It is not sufficient to say that whatever choice is made maximises utility, although this may be a sort of “get out of jail” card. It is based on a circular argument.

34. 34 34 prior probability

After reading read William Poundstone’s book “Priceless: the myth of fair value” over the weekend, I need to take back my previous comment about the irrationality of this pricing scheme. The price scheme is a rational way to sell more candy by taking advantage of people’s sense of “loss aversion”. Even if I just want one bag of candy, I will purchase both bags to avoid the negative emotion of the loss that occurs if I just buy one bag

35. 35 35 Harold

#34 pertains to my comment. Is loss aversion rational under the Rosencrantz and Gildenstern analysis? (Or is that von Neuman Morgenstern?). With loss aversion, the same amount of money (or other good) is valued differently depending on where it starts. I think under vN/M the final position only should be considered. Is this correct?

36. 36 36 Gordon Landwirth

@ Harold #34,

Re: The explanations for the consumers’ rationality seem to be based on his realisation of his own irrationality, including that of Gordon Landwirth. Sort of “I know I am not rational enough to throw away a bag of M&M’s even if I don’t want them, so I had better pay extra and only get one bag”. This does not make it rational.

I’m not sure if you’re disagreeing with me or not re: my point about negative marginal utility of the second bag. I realize it would arguably be irrational to eat the second bag of M&Ms, but that doesn’t mean it is irrational to, in effect, lose money (9 cents) to avoid creating a situation conducive to that net-harmful (and arguably irrational) decision I may make at some later point. Also, even if I stored the second bag until such time as it is net-beneficial to eat it (let’s say I consider it optimal to give myself such a treat once each week, so I’d be storing the second bag for one week), I may anticipate that I will feel stress in the interim period simply due to the temptation. It could be rational for me to sacrifice 9 cents (plus the cost of purchasing something like that one week from now) to avoid that stress.

Moreover, as I said, it could be rational to sacrifice 9 cents and avoid taking purchasing — and thus taking possession of — the second bag if there were storage and/or disposal costs (as well as selling costs, just to cover that option, and/or transport or other logistical costs) that were high enough to make it rational to sacrifice 9 cents to avoid the costs resulting from taking possession of that second bag.

Granted, in reality, it’s hard to imagine such high costs for a bag of M&Ms, but before we say something must be irrational, we need to know that no one could possibly have any costs of such magnitudes, and given that we don’t know scenarios for everyone, we may as well be talking about unspecified widgets and hypothetical scenarios, and I’d ask you to consider a scenario in which they are mattresses rather than M&Ms — i.e., something associated with considerable costs of taking possession (transport, storage (if one doesn’t have much extra space), perhaps disposal, and opportunity cost of selling).

37. 37 37 Mike Rulle

Well, it would never make sense to buy one bag for \$1.09. One can only assume that a rational buyer would purchase the two bags. They can chose to do what ever they want to do with the second bag. As one writer noted, the seller could have said, “buy one and get one free”. Perhaps the seller wanted to make it extra clear he was not going to sell one bag for 50 cents, so he created a 59 cent penalty to buy just one bag, his presumed marginal cost plus profit to be forced to keep candy on the shelf.

The bottom line is he wanted to sell the candy for 1.00 for 2 bags. The 1.09 is irrelevant. Since I do not have a baseline for the normal price, it is not completely obvious why it is 2 for a dollar. He wanted to unload supply by selling two bags per purchase. Candy has a long shelf life—but maybe he needed extra shelf space fast. I can hypothesize many psychological reasons why a buyer might consider it a hassle to be responsible for an extra bag of candy he does not want that is worth 9 cents to pay for—(because he does not want the second bag it is 9 cents not 59 cents). But this seems like a dopey path of thinking to go down.

38. 38 38 Harold

Gordon Landwirth #36. This is an interesting point that I feel I never quite got to the bottom of. Whilst in general terms it may be rational not to buy an extra item, such as a matress as you suggest, we are clearly dealing with M&M’s here. In this specific case, there is almost no storage or disposal costs for the consumer.

In the Alice case mentioned earlier, Alice is rational in choosing only one egg, because she has been told she must eat both if she buys two. She does not know the quality of the eggs – they might be horrible. This is also not the case here as M\$M’s are pretty uniform, and there is no compulsion to eat both or even either packet.

The consumer buying only one packet for more money is therefore only saving himself either the emotional cost of throwing or giving away the packet, or the negative emotional cost of temptation, or of giving in to temptation.

If we are to say that this is rational, then anything becomes rational if I merely want it to be so.

Whilst in one sense it may be considered rational, I don’t think it is in the vN/M sense, which I think is what we are talking about here.

Reference to the locavore is illustrative. In an economic sense, I think the locavore is irrational. However, we can make him rational by simply invocing his distaste at buying food that has travelled a long way. I have not heard an argument that makes the M&M buyer rational but leaves the locavore irrational. If we are to cast then both rational, then anything goes.

39. 39 39 Gordon Landwirth

@Harold #38,

The consumer buying only one packet for more money is therefore only saving himself either the emotional cost of throwing or giving away the packet, or the negative emotional cost of temptation, or of giving in to temptation.

If we are to say that this is rational, then anything becomes rational if I merely want it to be so.

Choosing A over B because B would bring emotional cost is not (or at least not necessarily) irrational. Obviously it can be subjective, but a cost is a cost, so I’m not sure what you mean by it being rational only if “anything becomes rational if I merely want it to be so.” Someone may not want or choose to incur that emotional cost. It may occur due to values, feelings, perspectives, perceptions, etc., etc., that are sufficiently ingrained that he/she may not be able to remove the emotional cost or it may not be rational to even work on doing so simply for the benefit of a second bag of M&Ms plus 9 cents. (It’s also possible one wouldn’t want to change one’s values, etc.)

I’d also add that there could be yet more scenarios involving negative marginal utility to the buyer of a second bag, and not all involving just the behavior and emotions of the buyer in isolation. For example, as for simply throwing it away, in addition to possible actual cost (at least theoretically), and emotional cost to the individual, a parent might not want his/her child to see him/her being wasteful (bad lesson re: environment or denying someone else a chance to purchase a bag later if/when that vendor’s supply runs out), but also may not want the kids to either bug the heck out of him/her for it before they are allowed to eat it or perhaps eat it without permission.

40. 40 40 Peter Tennenbaum

Any trick to promote sugar and, hence, tooth decay; M&Ms (etc.) are subsidized by The Dental Lobby. Huge scandal. Follow the money and you’ll find the real sugar daddies who are promoting this insidious addiction.

Type II diabetes is also exploding. Medical catastrophes everywhere. Think of THE CHILDREN. People before profits!

41. 41 41 iceman

Harold – is Weight Watchers irrational?

42. 42 42 Harold

#41. Interesting question, and perhaps a good example that will tease out this problem. I will need to think it through a bit more, but my initial reaction is that Weight Watchers is rational, but only because of the earlier irrationality of eating more than you would really want to.

The rational economic actor knows what they want, and knows which choice will get what they want and is consistent. The rational economic actor would not eat more than they really wanted, then pay extra to encourage themselves to eat less.

Given that the real person acts irrationally sometimes by over-indulging, then the rational person will take steps to obtain the result they really want. This may take the form of joining weight watchers which will make it easier to avoid succumbing to their irrational indulgencies.

43. 43 43 iceman

I know this is a recurring theme for you (kinda like nobody.really never misses a chance to claim rights aren’t real). Again I highly recommend the 1st chapter of David Friedman’s “Hidden Order” for a fascinating essay on what the rationality assumption really means. To your basic point, he says that in order to get very far economists must assume people’s objectives are reasonably simple and they tend to choose the correct means of achieving them; otherwise utility devolves into “why did I stand on my head with a burning \$1,000 bill between my toes? Because I wanted to.” However, re: the m&ms example he also tells a story about not buying a campus parking permit so he’d be forced to walk more. A colleague challenged that as a believer in rationality he should be able to make the correct choice without rigging the game. He responded that rationality is an assumption he makes about *other* people (as the best guide to their behavior); he knows enough about himself to make allowances for his own irrationality.

Personally I find taking preemptive action based on ones self-knowledge of human frailty – like paying someone to help you eat less – to seem pretty reasonable / rational.

44. 44 44 Harold

“he knows enough about himself to make allowances for his own irrationality.” Pretty much in agreement there. We can reduce the impact of our irrationality sometimes if we recognise it. This fails to provide an explanation that is based on rationality.

45. 45 45 Harold

Looking back at Friedman’s comment, rationality is a good starting point when considering other people. This allows us to develop theories and models etc. We must however not forget that this is an assumption. Allowing for variations from rationality is not as simple as allowing for wind resistance perhaps, but we must not forget that we have made a simplifying assumption.

46. 46 46 iceman

I’m not sure economists forget they’ve made an assumption, but there are really only two other possibilities – irrationalities are basically random / unpredictable, or as the behaviorists argue there may be some systematic tendencies that warrant adjusting models and theories. (Where competition is involved there is also a survivorship bias in rationality.)

Whatever the case I still think the m&ms is a weak example – you’re looking at someone recognizing a temptation and paying (a small price) to avoid it and calling that irrational; others might call it an admirable level of self-awareness. As so often happens in these posts this seems maybe even to devolve into semantics.

47. 47 47 Harold

Although it sometimes does devolve into semantics, I don’t think it does in this case. It seems to me a strong case – No one has come up with an explanation where the retailer is rational that does not involve the consumer being irrational. The consumer mitigates his irrationality, but that does not remove it.

48. 48 48 iceman

To me irrationality typically refers to people doing things that are against their own best interests, as opposed to our hypothetical single-pack m&m buyer who’s consciously paying to avoid a greater loss. You say “but if you were rational you just wouldn’t overeat!” But given that we’re human (a fact I don’t think is lost on too many economists) and thus possess varying degrees of willpower against natural urges, perhaps accounting for that by developing appropriate coping strategies is about the best we could hope to observe in the real world. One could argue this is like me saying “I really wanted to be an NBA player” and you responding “well if you were rational you would’ve been taller then!”

I think what’s really happening here is either they meant \$1 each (maybe you call that mistake seller irrationality), or they’re using an overly high reference price to try to make a ‘special offer’ look better. Of course we haven’t established that anyone even takes the single-pack offer seriously, but the art of retail is filled with techniques for appealing to *time-challenged* consumers who use quick mental processing rules to make decisions, and arguably this is categorically different from irrationality (of which there are plenty of purer examples too). At least Prof. Richard McKenzie argues this in an interesting book on pricing puzzles.

49. 49 49 Harold

“consciously paying to avoid a greater loss” But only a loss that he is in control of, so does not need to pay extra for – unless he is irrational.

“time challenged” consumers are un-informed consumers.

50. 50 50 iceman

The point is it *can* be fully rational to not spend the time and effort to be fully informed. (This is true in spades in the political arena.)

One last attempt on this – I bet we would agree there is a difference, certainly in degree and to me in kind, between someone eating m&ms because he believes they will help him lose weight, vs. paying someone to help him not eat them because he knows they will cause him to gain weight.

51. 51 51 Harold

#50 – I agree- it is often (even always) rational to remain less than fully informed. However, it is not possible to be both fully rational and fully informed. This may be obvious, but it is significant.

There is a difference in the m&m scenarios you suggest. In the first the consumer is ill-informed. We assume he wants to lose weight, believes eating m&m’s will help, so rationally (but erroneously) eats m&ms. I think we can agree that this outcome is far from efficient since he spends all the money on m&m’s to negative effect. In the second, he is properly informed that eating m&ms will cause him to gain weight. The rational thing to do would be to not eat m&m’s. He recognises that he cannot be rational, so he instead pays someone to help him do what he would do if he were rational. This is also inefficient, because the money he spends keeping himself rational would be better spent elsewhere.

52. 52 52 iceman

Actually it’s not obvious to me what you mean by “not possible to be fully rational and fully informed”; I thought this is precisely what behaviorists criticize economic orthodoxy for assuming.
But I think it’s safe to say we’re working from different definitions.

Rationality – I would say a guy who eats m&ms to lose weight is irrational if he has *no reasonable basis for believing* this outcome to be consistent with his actions. Certainly if he just made up this theory, and even if based on bad information this gets into source credibility.
I would say the person who pays (an amount he’s willing to pay) because he recognizes he can’t otherwise resist the urge to eat m&ms is acting as rationally as possible under the circumstances – that is, taking an action (albeit preventive) that is consistent with his goals & beliefs.
Are nicotine patches irrational?

Efficiency – I agree the first example is inefficient. Not sure about the second, I guess it depends on the extent to which we think the guy ‘ought to be’ able to control his urge. Again one could view this in the same sense that given my wish to be an NBA player, my failure to make myself taller is inefficient.
[And of course for someone who doesn’t mind the consequences of being overweight and derived great pleasure from eating, we may have to say this is perfectly rational and efficient.]

Q – Suppose someone is (legitimately) afraid of getting mugged and can reduce this risk/fear by an equal amount by either carrying a gun, or hiring a bodyguard to follow him around at greater cost. If he has an aversion to handling a firearm so opts for the bodyguard, is this inefficient? Irrational?

Enjoyed the exchange, have a good weekend.

53. 53 53 Harold

” I would say a guy who eats m&ms to lose weight is irrational if he has *no reasonable basis for believing* this outcome to be consistent with his actions”. That could be true if he could not rationally hold that belief.

” is acting as rationally as possible under the circumstances”. I agree, but the circumstances include an inevitable irrationality. Rationally as possible is not the same as rationally.

Nicotine patches – are addictions rational? I had thought that addictions could cause one to act irrationally. Therefore, given a cause of irrationality, it is rational to ameliorate this by nicotine patches. Overall the behaviour is not based on rationality.

Fear of mugging and distaste of firearm. This would depend on why he has an aversion to handling firearms. Say he believes handling a firearm will give him a brain tumour – that would be irrational. It is probably irrational, because I cannot think of a rational reason why he should have this aversion. This does illustrate the problem with this discussion- if you push any choice far enough it is difficult to arrive at a rational reason for anything.

54. 54 54 iceman

Nothing else going on here so what the hay…

“if you push any choice far enough it is difficult to arrive at a rational reason for anything”

I think economists make this more tractable – if we observe someone eating less to lose weight, their actions are consistent with a goal that (we assume) provides net utility. If in the process they incur greater cost than we deem ‘necessary’, we might call that inefficient. But like you said even there we have an issue of ‘control’, i.e. cognitive deficiencies vs. physical realities. E.g. if an addiction is a medical condition it would seem “as rationally as possible” becomes simply “rational”. (Surely addictions *can* cause people to act irrationally, but here we’re talking about someone acting correctly to overcome the addiction.) A biological urge is perhaps less deterministic, but still requires some level of effort to resist which represents a real cost, so paying something less to eliminate the conflict (and thereby increase net utility) seems quite rational. On the firearm, if he has moral objections I don’t think we’d call that irrational (though paying someone else to do the same thing would seem hypocritical). If he just finds the idea of directly handling them “yucky”, to me this would just fall under accounting for tastes.

55. 55 55 Harold

Iceman. I have drafted a few responses, but they have all ended up rambling. One thought that occurs to me is when does a preference become a prejudice? Taking the gun case as an example, he might not want to actually shoot someone himself, but he is perhaps quite happy to be the cause of someone being shot in the saem way as if he had pulled the trigger. You point out the hypocrisy, but is it irrational? If I don’t like carrots because I don’t like the taste, then fine, but if I don’t like them because I have been told they are inferior, is that rational? What if I don’t like the taste because I was told they were inferior as a child and have developed an aversion?