Well what's more resource-efficient, having thousands of customers keep a spare cheap phone at home as a standby replacement, or having the store keep a few to give out as loaners?
Econ 101 also teaches that that conclusion is only valid when certain pre-conditions are met. If those pre-conditions are not met then there's no reason to assume the market solution is optimal. In particular, if users don't know about a particular advantage or disadvantage of a product at purchase time, then the market solution won't be optimal. Take Comcast throttling of its' users access to BitTorrent; do you think that's what the users wanted? In the case of phone insurance, most users have no idea when walk out of the store with a new phone, whether the store would give them a loaner phone if they made an insurance claim to get their main phone replaced. So the precondition for market optimality is not met.
Well obviously I assume they'd pass the costs on in the insurance premium (which means you won't be subsidizing my dumb ass if you forgo the insurance). My argument is that the benefit to consumers of not having to go without a phone for two days, is great enough, that most of them would come out ahead, even after the increased costs get passed on.
My real goal was not to gain sympathy (on Slashdot?). The point I'm making in many of these articles is that we should not assume "the market" will lead to optimal solutions. People have accepted or argued for many terrible situations by blindly assuming that "the market" leads to what's best for us -- if that were really true, there would be no Net Neutrality issue, for example, because if a company were blocking or slowing access to a website, customers would just leave that company. It doesn't work that way in real life, which is why Net Neutrality is an issue.
I happen to think every example of market failure helps to get the point across. I don't much care about the phone.
Well I'd need a smartphone to keep using it the way I'd use my normal phone, and the backup phone would have to be purchased without a contract, meaning a few hundred dollars at least. I doubt every '"sane person" does that:) If you mean a backup dumbphone, well maybe, but it's much harder to adjust to trying to find people and places when you're on the go, when you've suddenly lost Internet.
Well first of all you're double-counting if you're adding the $6500 in inventory cost PLUS the $3500 from depreciation:) If the depreciation of the phones is a loss, it's a depreciation on an asset that you still have, so you can't also count the up-front cost of the phone as a loss at the same time.
Also, it's unrealistic to think you'd need 10 of EVERY phone. It's not as if 10 people are all going to make replacement claims on the same model in the same 2 days (it takes at most 2 days for the replacements to arrive from the insurance provider).
More generally, it would probably benefit enough customers enough, if there were just a requirement to provide loaner phones while replacing the insurance phones, even if the loaner phones wouldn't have to be exactly the same model. I was probably overreacing on the idea of having loaner phones in every model since I really, really prefer having a slide-out keyboard, but even I could live with a virtual keyboard for one day.
This can't be that much of an imposition, because it's something that some cell phone stores actually do. (But not all of them, and that's where customers get screwed because they have no working phone for two days.)
That would indeed minimize the cost. But the goal should not be to minimize cost but to maximize benefit-minus-cost. (Otherwise, you could just sell empty boxes without phones in them, to "minimize cost".) Having a loaner phone that is guaranteed available to consumers to borrow while their replacement phone is being mailed to them, would benefit consumers a lot for only a little cost.
My goal was not really to get a regulation like this passed. My goal was to get people thinking about how often the market leads to non-optimal solutions, because there is a lot of dogma claiming that the market cannot make mistakes like this, and it needs to be counteracted.
The specific circumstances that apply to cell phones are (1) they're really small, so it would cost the store less to carry spares, than, say, for an auto dealership to carry spare cars; and (2) when people's phones break, there's a huge benefit to them of getting a new one right away instead of waiting several days.
After all, as some commenters pointed out, some stores provide free loaner phones voluntarily, so it can't be that hard.
My problem is with the assumption that the free market will take care of these things by itself, and so whatever "the market" has given us must already be resource-optimal. That's only true for attributes of a product that the customer is fully aware of up-front when they're comparing options. If customers don't know in advance what the experience will be like to get a phone replaced, there's no reason to expect people to make the most informed choice when buying the phone and the insurance in the first place.
I would not be in favor of this regulation for a small business, but there aren't a lot of mom-and-pop cell phone carriers.
If there were some carriers small enough that this would be truly burdensome, it could always be written into the regulation that the requirement only applied to companies with more than X number of stores.
My point is that the benefit to the consumer would be a lot less than the cost to the company, which means customers would come out ahead even if the companies passed the costs along.
Yeah I was thinking that it would probably be good enough for most customers making insurance claims, if the store were just required to give out a loaner phone, not necessarily the same model. I would have strongly preferred the same model since I'm hooked on my slideout keyboard, but a virtual keyboard phone is better than nothing.
Well (1) we had been talking to them from beside their boat for a while before; (2) I couldn't believe it either; (3a) we were not in the lock, we were in the canal waiting area waiting to go into the lock; (3b) this lock in question was being filled up, not drained, reducing the risk of being sucked under; and (3c) in any case it's moot because I didn't get into the water, I got into their boat, at emergency speed, and the phone fell out of my pocket into the water at the bottom of their boat, not into the canal.
Now, (4) what if we mitigated the cost to the store by (a) only requiring them to give out a loaner phone, not necessarily the same model that you have (as I wrote elsewhere, I hate anything that doesn't have a slide-out keyboard, but I'd live) and (b) only requiring the loaner phones to be available from some store in the area, not necessarily the one where you bought your phone?
Thanks, this is a lot more thoughtful than a lot of the ranting comments that got posted in response. So what if you minimized the cost by (1) only requiring the loaner phones to be available from one store in the area, that the customer could drive to (instead of in every retail location) and (2) only requiring the store to give out a loaner phone, not necessarily the same model. Now you've driven the cost down from $2,500 per store to $50, per, say, every 10 stores, or $5 per store? Presumably that's a much less burdensome regulation.
Oh OK. Well actually I considered that, but the problem is how to weigh the opinions of people who have only ever used virtual keyboards because they've never known anything else, possibly because the store didn't even offer anything else as an option. I think it would be absurd to count those all as votes for "virtual keyboards". Maybe some of them are just sure that they don't want a slide-out keyboard, but based on the evidence from the stores, it looks as if far more of them just didn't have that option, or didn't know that they did. In the end I decided just to count the opinions of people who had tried them both.
copied from a comment I just wrote elsewhere:
it was just a quick and dirty survey to show that the proportion of people who want slideout keyboard phones is not zero, like the stores are pretending that it is
What I wrote was: "Obviously that's too small of a sample to be very precise about the percentage of users that prefer slide-out keyboards (apart from the fact that Mechanical Turk users are unrepresentative of the general population in several ways), but it does mean that the near-extinction of slideout-keyboard phones in retail stores is probably not in proportion to what people actually want."
i.e., it was just a quick and dirty survey to show that the proportion of people who want slideout keyboard phones is not zero, like the stores are pretending that it is.
What about the Stratosphere, which I wrote about previously? That didn't have a slide-out keyboard and so presumably wasn't serving a niche market. That was the one where the calendar app highlighted the wrong date as "today", because it (apparently) computed "today" based on GMT rather than the phone's current time zone.
If you're taking it as a given that the article statements were correct, my final statement was that we could improve existing phones by methodically testing them for idiotic problems (the Stupid Shit Index), so that consumers know how to find phones that have the least stupid shit wrong with them, the makers of those phones are rewarded, and the next iteration of phones has incrementally less stupid shit as a result. Since this is a big reward for relatively little effort, isn't it worth doing?
Take the auto-correct for example. The auto-correct application on the high-end phone presumably doesn't change "you're" to "you"re". So in what plausible scenario would they end up stripping down that high-end version to a low-end version that does?
If the high-end autocorrect occupies too much memory (too large a database of words, for example) for it to work correctly on the low-end phone, wouldn't it be easier to strip down the dictionary, rather than starting from scratch with a new dictionary that contains incorrect entries?
The UI bugs that I was talking about are in the kinds of features that you'd think they could just develop once, and re-use for both their low-end and high-end phones. e.g. they could have developed an auto-correct for the high-end phone that doesn't change you're to you"re, and then just re-used it on the low-end phone. Wouldn't it actually be more work to develop a separate buggier auto-correct for the cheap phone?
I think that's part of the explanation, but it doesn't account for why the phone doesn't suck in other ways -- if manufacturers are lazy and cheap, why has the camera resolution, for example, evolved to the point where it's really pretty good? And I think the answer is that camera resolution is quantifiable in a standard way, so it puts more pressure on the manufacturers to compete, whereas usability and bugginess are not.
Slashdot Mirror
Well what's more resource-efficient, having thousands of customers keep a spare cheap phone at home as a standby replacement, or having the store keep a few to give out as loaners?
Econ 101 also teaches that that conclusion is only valid when certain pre-conditions are met. If those pre-conditions are not met then there's no reason to assume the market solution is optimal. In particular, if users don't know about a particular advantage or disadvantage of a product at purchase time, then the market solution won't be optimal. Take Comcast throttling of its' users access to BitTorrent; do you think that's what the users wanted? In the case of phone insurance, most users have no idea when walk out of the store with a new phone, whether the store would give them a loaner phone if they made an insurance claim to get their main phone replaced. So the precondition for market optimality is not met.
Well obviously I assume they'd pass the costs on in the insurance premium (which means you won't be subsidizing my dumb ass if you forgo the insurance). My argument is that the benefit to consumers of not having to go without a phone for two days, is great enough, that most of them would come out ahead, even after the increased costs get passed on.
My real goal was not to gain sympathy (on Slashdot?). The point I'm making in many of these articles is that we should not assume "the market" will lead to optimal solutions. People have accepted or argued for many terrible situations by blindly assuming that "the market" leads to what's best for us -- if that were really true, there would be no Net Neutrality issue, for example, because if a company were blocking or slowing access to a website, customers would just leave that company. It doesn't work that way in real life, which is why Net Neutrality is an issue.
I happen to think every example of market failure helps to get the point across. I don't much care about the phone.
Well I'd need a smartphone to keep using it the way I'd use my normal phone, and the backup phone would have to be purchased without a contract, meaning a few hundred dollars at least. I doubt every '"sane person" does that :) If you mean a backup dumbphone, well maybe, but it's much harder to adjust to trying to find people and places when you're on the go, when you've suddenly lost Internet.
Well first of all you're double-counting if you're adding the $6500 in inventory cost PLUS the $3500 from depreciation :) If the depreciation of the phones is a loss, it's a depreciation on an asset that you still have, so you can't also count the up-front cost of the phone as a loss at the same time.
Also, it's unrealistic to think you'd need 10 of EVERY phone. It's not as if 10 people are all going to make replacement claims on the same model in the same 2 days (it takes at most 2 days for the replacements to arrive from the insurance provider).
More generally, it would probably benefit enough customers enough, if there were just a requirement to provide loaner phones while replacing the insurance phones, even if the loaner phones wouldn't have to be exactly the same model. I was probably overreacing on the idea of having loaner phones in every model since I really, really prefer having a slide-out keyboard, but even I could live with a virtual keyboard for one day.
This can't be that much of an imposition, because it's something that some cell phone stores actually do. (But not all of them, and that's where customers get screwed because they have no working phone for two days.)
says the guy commenting five levels deep :-P
That would indeed minimize the cost. But the goal should not be to minimize cost but to maximize benefit-minus-cost. (Otherwise, you could just sell empty boxes without phones in them, to "minimize cost".) Having a loaner phone that is guaranteed available to consumers to borrow while their replacement phone is being mailed to them, would benefit consumers a lot for only a little cost.
My goal was not really to get a regulation like this passed. My goal was to get people thinking about how often the market leads to non-optimal solutions, because there is a lot of dogma claiming that the market cannot make mistakes like this, and it needs to be counteracted.
The specific circumstances that apply to cell phones are (1) they're really small, so it would cost the store less to carry spares, than, say, for an auto dealership to carry spare cars; and (2) when people's phones break, there's a huge benefit to them of getting a new one right away instead of waiting several days.
After all, as some commenters pointed out, some stores provide free loaner phones voluntarily, so it can't be that hard.
My problem is with the assumption that the free market will take care of these things by itself, and so whatever "the market" has given us must already be resource-optimal. That's only true for attributes of a product that the customer is fully aware of up-front when they're comparing options. If customers don't know in advance what the experience will be like to get a phone replaced, there's no reason to expect people to make the most informed choice when buying the phone and the insurance in the first place.
I would not be in favor of this regulation for a small business, but there aren't a lot of mom-and-pop cell phone carriers.
If there were some carriers small enough that this would be truly burdensome, it could always be written into the regulation that the requirement only applied to companies with more than X number of stores.
My point is that the benefit to the consumer would be a lot less than the cost to the company, which means customers would come out ahead even if the companies passed the costs along.
Yeah I was thinking that it would probably be good enough for most customers making insurance claims, if the store were just required to give out a loaner phone, not necessarily the same model. I would have strongly preferred the same model since I'm hooked on my slideout keyboard, but a virtual keyboard phone is better than nothing.
Well (1) we had been talking to them from beside their boat for a while before; (2) I couldn't believe it either; (3a) we were not in the lock, we were in the canal waiting area waiting to go into the lock; (3b) this lock in question was being filled up, not drained, reducing the risk of being sucked under; and (3c) in any case it's moot because I didn't get into the water, I got into their boat, at emergency speed, and the phone fell out of my pocket into the water at the bottom of their boat, not into the canal.
Now, (4) what if we mitigated the cost to the store by (a) only requiring them to give out a loaner phone, not necessarily the same model that you have (as I wrote elsewhere, I hate anything that doesn't have a slide-out keyboard, but I'd live) and (b) only requiring the loaner phones to be available from some store in the area, not necessarily the one where you bought your phone?
Thanks, this is a lot more thoughtful than a lot of the ranting comments that got posted in response. So what if you minimized the cost by (1) only requiring the loaner phones to be available from one store in the area, that the customer could drive to (instead of in every retail location) and (2) only requiring the store to give out a loaner phone, not necessarily the same model. Now you've driven the cost down from $2,500 per store to $50, per, say, every 10 stores, or $5 per store? Presumably that's a much less burdensome regulation.
stop posting spoilers without a warning
Oh OK. Well actually I considered that, but the problem is how to weigh the opinions of people who have only ever used virtual keyboards because they've never known anything else, possibly because the store didn't even offer anything else as an option. I think it would be absurd to count those all as votes for "virtual keyboards". Maybe some of them are just sure that they don't want a slide-out keyboard, but based on the evidence from the stores, it looks as if far more of them just didn't have that option, or didn't know that they did. In the end I decided just to count the opinions of people who had tried them both.
What specifically do you think was the "wrong question" and what do you think would be a "right question"?
copied from a comment I just wrote elsewhere: it was just a quick and dirty survey to show that the proportion of people who want slideout keyboard phones is not zero, like the stores are pretending that it is
What I wrote was: "Obviously that's too small of a sample to be very precise about the percentage of users that prefer slide-out keyboards (apart from the fact that Mechanical Turk users are unrepresentative of the general population in several ways), but it does mean that the near-extinction of slideout-keyboard phones in retail stores is probably not in proportion to what people actually want."
i.e., it was just a quick and dirty survey to show that the proportion of people who want slideout keyboard phones is not zero, like the stores are pretending that it is.
hig fve!
If you don't think the statements are correct, then it's incumbent on you to say what you think is incorrect.
If you do think they're correct but you "don't see why you should comment on them", what on Earth are you doing in the comments section?
What about the Stratosphere, which I wrote about previously? That didn't have a slide-out keyboard and so presumably wasn't serving a niche market. That was the one where the calendar app highlighted the wrong date as "today", because it (apparently) computed "today" based on GMT rather than the phone's current time zone.
If you're taking it as a given that the article statements were correct, my final statement was that we could improve existing phones by methodically testing them for idiotic problems (the Stupid Shit Index), so that consumers know how to find phones that have the least stupid shit wrong with them, the makers of those phones are rewarded, and the next iteration of phones has incrementally less stupid shit as a result. Since this is a big reward for relatively little effort, isn't it worth doing?
Take the auto-correct for example. The auto-correct application on the high-end phone presumably doesn't change "you're" to "you"re". So in what plausible scenario would they end up stripping down that high-end version to a low-end version that does?
If the high-end autocorrect occupies too much memory (too large a database of words, for example) for it to work correctly on the low-end phone, wouldn't it be easier to strip down the dictionary, rather than starting from scratch with a new dictionary that contains incorrect entries?
The UI bugs that I was talking about are in the kinds of features that you'd think they could just develop once, and re-use for both their low-end and high-end phones. e.g. they could have developed an auto-correct for the high-end phone that doesn't change you're to you"re, and then just re-used it on the low-end phone. Wouldn't it actually be more work to develop a separate buggier auto-correct for the cheap phone?
You're very articulate. Your English teacher would be proud.
Now, on behalf of your math teacher, do you actually have a point? i.e. is there a statement in the article you think is incorrect.
I think that's part of the explanation, but it doesn't account for why the phone doesn't suck in other ways -- if manufacturers are lazy and cheap, why has the camera resolution, for example, evolved to the point where it's really pretty good? And I think the answer is that camera resolution is quantifiable in a standard way, so it puts more pressure on the manufacturers to compete, whereas usability and bugginess are not.