I'm not understanding what you're stuck on -- context-dependent goods are always going to end up more expensive due to the competition factor. While everyone can have a, say, flat screen TV once the price goes down far enough, only half of everyone can have a house bigger than the median, regardless of how much money they spend on housing.
I checked out some reviews on Amazon, and haven't found any analysis similar to yours. People seem to think that Frank is dividing purchases between those that are in some sense tied to necessity, and purchases which are driven more by competition.
The puzzle arose out of reading the book. In ch. 11, "understanding conspicuous consumption," Frank argues that his way of understanding individual/group conflict provides an explanation of conspicuous consumption spending patterns. This requires the premise that I alluded to in the post. E.g., p. 162:
In brief, we expect an imbalance in the choice between two activities if the individual rewards from one are more context sensitive than the individual rewards from the other.
But there's not really an argument for this; what precedes it is the obvious military-spending example, which is sort of misleading, because the results of losing that arms race are very different from the results of losing the 'bigger houses, shinier cars' arms race. The puzzle is that, in the case of many context-sensitive but inessential goods, we might expect rational agents to see that achieving the benefits requires ever-increasing expenditures (since the payoff depends on relative position and others are after the same goods). In response, they would shift their resource allocation from the context-dependent goods to the independent goods, which provide higher SWB/dollar.
I have not read the book, but I don't think much of his argument if I'm understanding it properly. I'd think that it would be very hard to extricate conspicuous consumption from our drive for social ties, for example. You buy the big house to be in a neighborhood, and develop friendships with, a certain class of people. You can't directly spend money on those friendships - it would vitiate the myth of friendship - so you spend it indirectly. If that's right, there's a painful punishment for not keeping up with the Joneses - your relationship with them changes, and not for the better.
In response, they would shift their resource allocation from the context-dependent goods
This assumes that context-independent goods can be substituted for context-dependent goods and produce the same benefit. If the benefits you get from the context-dependent goods aren't fungible, then you're stuck in the arms race.
If the benefits you get from the context-dependent goods aren't fungible, then you're stuck in the arms race.
I assume everything's being reconciled as SWB, LB. So a large house isn't shelter, it's a given amount of SWB. If SWB is not fungible, then I would think that the whole project breaks down.
So the claim is that given the choice between buying a context-sensitive good and a context-independent good, a rational agent will purchase the context-sensitive good instead.
I haven't read the book, but it seems that here is one possibility. The context-sensitive good is valued primarily due to its context (thought it may have some other instrumental or intrinsic value). If that context is allowed to decay, then the value of the context-sensitive goods also diminishes. The resources go to keep up the context in order to keep up the value of goods already purchased; like SCMT says keeping up with the Joneses means you keep the Joneses.
Simplified, probably wrong, example:
So I buy a big house and a large SUV. My neighbors buy bigger houses, bigger SUVs, and iPods. Now I can buy an iPod or some other non-status symbol thingy. If I buy the iPod, I have another context-sensitive good that helps maintain the social context I participate in. If I don't buy the iPod, I'm starting to slip out of the social context that values the big house and big SUV.
If I were to slip out of the game entirely, I'd still have a house and an SUV, but I wouldn't be participating in the context that adds the extra context-sensitive value to them. So I keep going in this arms race of consumption to maintain the value of the goods I already have.
Basically I'm suggesting that context-sensitivity is a sort of recursive value function.
we might expect rational agents to see that achieving the benefits requires ever-increasing expenditures
Is Frank using 'rational' in the way economists use it--the rationality of homo economicus? (Homoeconomic rationality, in the delightful and possibly accidental coinage of one of my students last year?) Because under that definition I believe we wouldn't expect rational agents to see this; rationality is defined as going for the best payoff for you, if we hold everyone else's actions fixed assuming that your action doesn't causally impact theirs. Just that, in general, homoeconomically rational agents always defect in a prisoner's dilemma (including repeated PDs with fixed endpoints), while what I and maybe you would call rational agents will cooperate at least some of the time.
Back to the main argument--it does seem like there's a missing premise. One thing could be that agents are especially sensitive to losses, so that if an agent has a smaller house, which decreases SWB, they'll be particularly likely to rectify it. Another one--maybe this is more likely--is that in most areas gain in SWB per additional dollar spent decreases as you gain more and more of the good, so rational agents will direct more money to areas in which they're currently getting less SWB (instead of piling more money into areas in which they're already doing pretty well). But in position-sensitive areas everyone's always pretty close to zero, so it's always rational to spend money there. And the more money your neighbors spend, the farther you are into zero/negative territory, so it makes even more sense for you yourself to put more money into the position-senstive good; pushing your neighbors farther into zero/negative territory, making it rational for them to allocate more money to this good; etc.
I haven't read the book, but here's how I'd transpose this setting to a prisoner's dilemma:
Economists would argue that the optimal allocation of consumption would be the point where the marginal increase in utility for the last dollar spent in each category is the same.
If there are two agents choosing consumption of context sensitive and insensitive consumption, this will suggest a natural equilibrium. However, each of the agents will anticipate that if they shift consumption towards context sensitive goods then they will have a bigger house than the other anticipating an additional boost to satisfaction. This shifting of consumption is like finking in the p.d. It's not socially optimal, but it dominates because no matter what the other agent does, your better off shifting money towards context sensitive goods. But this could leave the society with much lower overall utility because the agents are shifting so much consumption away from the other category to keep up with one another in context sensitive goods.
I possess no context-dependent goods in that sense, nor do I know anyone who does. I'm sure there are such people, but, for an explanation that proposes to tackle such a large social problem, I have a feeling his perspective is misguided.
I possess no context-dependent goods in that sense, nor do I know anyone who does.
Does the following qualify as a 'context dependent good'?
In 1970, you would have been pleased as punch to watch football on a 15" TV. Today, that would seem small and poor. The difference isn't that it's any harder to see what's happening on a 15" TV, it's that other people have much bigger TVs now, and a little screen appears frustratingly inadequate in that context.
18. I think we have bigger TVs now because they're cheaper, yes? I mean, I wasn't alive back then, but wouldn't the price of a 15" color TV be comparable to the price of, I don't know, maybe a 30" color TV today? Or maybe even less?
But if people in the 70s were happy with their tiny TVs, wouldn't it make sense for you to be just as happy with a tiny TV, and spend your money one something else? Instead, regardless of the price, you need a bigger TV to be as happy as your father was with a little TV.
It could easily be that people in the 70s valued TV size in exactly the same degree as we do. As prices dropped, the intersection point for TV utility maximization shifted, such that everyone now buys larger TVs. If they cost the same amount as they did in the 70s, we would have the same sizes as in the 70s. The demand is the same, but demand is only one axis, cost the other.
Economists' explanations of the rationality of behavior aren't useful for anyone making a decision who wants their decision to be rational. They just take statistically prevalent decisions as given, and then tweak rationality until they're able to show that the prevalent decisions are rational. Because if they aren't, economics is less powerful as a justification of our economy.
Franks' book won't help you make better decisions, though it might make you feel better about certain decisions you have already made which you feel vaguely ashamed of.
A lot of the rationality in everyday arguments is pretty fake. It's claimed that some choice maximizes or optimizes some kind of unmeasurable intangible, for example. That's quite a different thing than optimizing something that actually can be described and measured (market share, profit, wealth, or minimizing risk).
Or hypothetical arguments are made based on calculations of odds and assignments of values that are purely imaginary.
"Would you drop a nuclear weapon on Mecca if there was a 10% chance that if you didn't, an innocent 12 year old schoolgirl would be horribly killed? If not, what if there were 100 innocent schoolgirls and a 40% chance? Or a 1% chance of saving 10,000 schoolgirls, of whom half were innocent and the other half not really very innocent....."
Is it just me, or does that last hypo look suspiciously like an excuse for dwelling on the existence of not-particularly-innocent schoolgirls? Mr. Emerson, for shame!
(But other than that, I agree completely. "Economic rationality," as used in argument, is often completely tautological.)
But why, Michael? Surely if anyone should get upset it's wolfson? Yours is an honest profession, but his tipping habits are atrocious and shameful(besides which they reinforce certain unfortunate stereotypes).
I'm not understanding what you're stuck on -- context-dependent goods are always going to end up more expensive due to the competition factor. While everyone can have a, say, flat screen TV once the price goes down far enough, only half of everyone can have a house bigger than the median, regardless of how much money they spend on housing.
What's the confusing bit?
Posted by LizardBreath | Link to this comment | 09-11-05 11:03 AM
Have you read the book? I bet it's in the book.
I checked out some reviews on Amazon, and haven't found any analysis similar to yours. People seem to think that Frank is dividing purchases between those that are in some sense tied to necessity, and purchases which are driven more by competition.
Posted by Michael | Link to this comment | 09-11-05 11:42 AM
The puzzle arose out of reading the book. In ch. 11, "understanding conspicuous consumption," Frank argues that his way of understanding individual/group conflict provides an explanation of conspicuous consumption spending patterns. This requires the premise that I alluded to in the post. E.g., p. 162:
But there's not really an argument for this; what precedes it is the obvious military-spending example, which is sort of misleading, because the results of losing that arms race are very different from the results of losing the 'bigger houses, shinier cars' arms race. The puzzle is that, in the case of many context-sensitive but inessential goods, we might expect rational agents to see that achieving the benefits requires ever-increasing expenditures (since the payoff depends on relative position and others are after the same goods). In response, they would shift their resource allocation from the context-dependent goods to the independent goods, which provide higher SWB/dollar.
Posted by FL | Link to this comment | 09-11-05 12:17 PM
I have not read the book, but I don't think much of his argument if I'm understanding it properly. I'd think that it would be very hard to extricate conspicuous consumption from our drive for social ties, for example. You buy the big house to be in a neighborhood, and develop friendships with, a certain class of people. You can't directly spend money on those friendships - it would vitiate the myth of friendship - so you spend it indirectly. If that's right, there's a painful punishment for not keeping up with the Joneses - your relationship with them changes, and not for the better.
Posted by SomeCallMeTim | Link to this comment | 09-11-05 12:29 PM
In response, they would shift their resource allocation from the context-dependent goods
This assumes that context-independent goods can be substituted for context-dependent goods and produce the same benefit. If the benefits you get from the context-dependent goods aren't fungible, then you're stuck in the arms race.
Posted by LizardBreath | Link to this comment | 09-11-05 12:38 PM
If the benefits you get from the context-dependent goods aren't fungible, then you're stuck in the arms race.
I assume everything's being reconciled as SWB, LB. So a large house isn't shelter, it's a given amount of SWB. If SWB is not fungible, then I would think that the whole project breaks down.
Posted by SomeCallMeTim | Link to this comment | 09-11-05 12:49 PM
So the claim is that given the choice between buying a context-sensitive good and a context-independent good, a rational agent will purchase the context-sensitive good instead.
I haven't read the book, but it seems that here is one possibility. The context-sensitive good is valued primarily due to its context (thought it may have some other instrumental or intrinsic value). If that context is allowed to decay, then the value of the context-sensitive goods also diminishes. The resources go to keep up the context in order to keep up the value of goods already purchased; like SCMT says keeping up with the Joneses means you keep the Joneses.
Simplified, probably wrong, example:
So I buy a big house and a large SUV. My neighbors buy bigger houses, bigger SUVs, and iPods. Now I can buy an iPod or some other non-status symbol thingy. If I buy the iPod, I have another context-sensitive good that helps maintain the social context I participate in. If I don't buy the iPod, I'm starting to slip out of the social context that values the big house and big SUV.
If I were to slip out of the game entirely, I'd still have a house and an SUV, but I wouldn't be participating in the context that adds the extra context-sensitive value to them. So I keep going in this arms race of consumption to maintain the value of the goods I already have.
Basically I'm suggesting that context-sensitivity is a sort of recursive value function.
I have no idea if this is even plausible.
Posted by Cala | Link to this comment | 09-11-05 12:58 PM
we might expect rational agents to see that achieving the benefits requires ever-increasing expenditures
Is Frank using 'rational' in the way economists use it--the rationality of homo economicus? (Homoeconomic rationality, in the delightful and possibly accidental coinage of one of my students last year?) Because under that definition I believe we wouldn't expect rational agents to see this; rationality is defined as going for the best payoff for you, if we hold everyone else's actions fixed assuming that your action doesn't causally impact theirs. Just that, in general, homoeconomically rational agents always defect in a prisoner's dilemma (including repeated PDs with fixed endpoints), while what I and maybe you would call rational agents will cooperate at least some of the time.
Back to the main argument--it does seem like there's a missing premise. One thing could be that agents are especially sensitive to losses, so that if an agent has a smaller house, which decreases SWB, they'll be particularly likely to rectify it. Another one--maybe this is more likely--is that in most areas gain in SWB per additional dollar spent decreases as you gain more and more of the good, so rational agents will direct more money to areas in which they're currently getting less SWB (instead of piling more money into areas in which they're already doing pretty well). But in position-sensitive areas everyone's always pretty close to zero, so it's always rational to spend money there. And the more money your neighbors spend, the farther you are into zero/negative territory, so it makes even more sense for you yourself to put more money into the position-senstive good; pushing your neighbors farther into zero/negative territory, making it rational for them to allocate more money to this good; etc.
Posted by Matt Weiner | Link to this comment | 09-11-05 1:01 PM
I should have asked this initially, but what's SWB?
Sense of Well Being is what I'm coming up with, but I haven't seen the term defined.
Posted by LizardBreath | Link to this comment | 09-11-05 1:19 PM
Subjective Well Being (right?)
Posted by Matt Weiner | Link to this comment | 09-11-05 1:22 PM
Sisters With Boyces
Posted by Matt Weiner | Link to this comment | 09-11-05 1:23 PM
I haven't read the book, but here's how I'd transpose this setting to a prisoner's dilemma:
Economists would argue that the optimal allocation of consumption would be the point where the marginal increase in utility for the last dollar spent in each category is the same.
If there are two agents choosing consumption of context sensitive and insensitive consumption, this will suggest a natural equilibrium. However, each of the agents will anticipate that if they shift consumption towards context sensitive goods then they will have a bigger house than the other anticipating an additional boost to satisfaction. This shifting of consumption is like finking in the p.d. It's not socially optimal, but it dominates because no matter what the other agent does, your better off shifting money towards context sensitive goods. But this could leave the society with much lower overall utility because the agents are shifting so much consumption away from the other category to keep up with one another in context sensitive goods.
Posted by cw | Link to this comment | 09-11-05 1:44 PM
Don't mind comment 2 too much. I awoke feeling like a little bitch this morning.
Posted by Michael | Link to this comment | 09-11-05 1:45 PM
A context dependent good is a good for which the value *to you* is dependent on others who you are in competition with?
Posted by Michael | Link to this comment | 09-11-05 2:09 PM
That's my understanding.
Posted by LizardBreath | Link to this comment | 09-11-05 2:53 PM
I awoke feeling like a little bitch this morning.
Why, did b-dub not tip you enough?
Posted by Mitch Mills | Link to this comment | 09-11-05 3:25 PM
I possess no context-dependent goods in that sense, nor do I know anyone who does. I'm sure there are such people, but, for an explanation that proposes to tackle such a large social problem, I have a feeling his perspective is misguided.
Mitch, I'm gonna get you.
Posted by Michael | Link to this comment | 09-11-05 3:59 PM
I possess no context-dependent goods in that sense, nor do I know anyone who does.
Does the following qualify as a 'context dependent good'?
In 1970, you would have been pleased as punch to watch football on a 15" TV. Today, that would seem small and poor. The difference isn't that it's any harder to see what's happening on a 15" TV, it's that other people have much bigger TVs now, and a little screen appears frustratingly inadequate in that context.
Posted by LizardBreath | Link to this comment | 09-11-05 5:02 PM
You can read the passage from the book starting here.
Posted by ogged | Link to this comment | 09-11-05 5:28 PM
Well, that was not convincing at any level. But thanks anyway, ogged.
Posted by SomeCallMeTim | Link to this comment | 09-11-05 5:44 PM
18. I think we have bigger TVs now because they're cheaper, yes? I mean, I wasn't alive back then, but wouldn't the price of a 15" color TV be comparable to the price of, I don't know, maybe a 30" color TV today? Or maybe even less?
Posted by Michael | Link to this comment | 09-11-05 6:36 PM
But if people in the 70s were happy with their tiny TVs, wouldn't it make sense for you to be just as happy with a tiny TV, and spend your money one something else? Instead, regardless of the price, you need a bigger TV to be as happy as your father was with a little TV.
Posted by LizardBreath | Link to this comment | 09-11-05 6:58 PM
It could easily be that people in the 70s valued TV size in exactly the same degree as we do. As prices dropped, the intersection point for TV utility maximization shifted, such that everyone now buys larger TVs. If they cost the same amount as they did in the 70s, we would have the same sizes as in the 70s. The demand is the same, but demand is only one axis, cost the other.
Posted by text | Link to this comment | 09-11-05 7:05 PM
I see that TV size could be "context dependant" but it isn't necessarily so, solely based on the fact that the sizes have increased.
Posted by text | Link to this comment | 09-11-05 7:06 PM
when I said cost I of course meant supply. I'm an ass.
Posted by text | Link to this comment | 09-11-05 7:14 PM
Economists' explanations of the rationality of behavior aren't useful for anyone making a decision who wants their decision to be rational. They just take statistically prevalent decisions as given, and then tweak rationality until they're able to show that the prevalent decisions are rational. Because if they aren't, economics is less powerful as a justification of our economy.
Franks' book won't help you make better decisions, though it might make you feel better about certain decisions you have already made which you feel vaguely ashamed of.
A lot of the rationality in everyday arguments is pretty fake. It's claimed that some choice maximizes or optimizes some kind of unmeasurable intangible, for example. That's quite a different thing than optimizing something that actually can be described and measured (market share, profit, wealth, or minimizing risk).
Or hypothetical arguments are made based on calculations of odds and assignments of values that are purely imaginary.
"Would you drop a nuclear weapon on Mecca if there was a 10% chance that if you didn't, an innocent 12 year old schoolgirl would be horribly killed? If not, what if there were 100 innocent schoolgirls and a 40% chance? Or a 1% chance of saving 10,000 schoolgirls, of whom half were innocent and the other half not really very innocent....."
Posted by John Emerson | Link to this comment | 09-11-05 7:35 PM
Is it just me, or does that last hypo look suspiciously like an excuse for dwelling on the existence of not-particularly-innocent schoolgirls? Mr. Emerson, for shame!
(But other than that, I agree completely. "Economic rationality," as used in argument, is often completely tautological.)
Posted by LizardBreath | Link to this comment | 09-11-05 8:31 PM
The schoolgirls were 16 years old and fully clothed, though not innocent.
Posted by John Emerson | Link to this comment | 09-11-05 8:53 PM
Also, I have no interest whatsoever in Olympic gymnastics or figure skating.
Posted by John Emerson | Link to this comment | 09-11-05 8:54 PM
pear thiefs, the lot of them!
Posted by Michael | Link to this comment | 09-11-05 11:08 PM
thieves, before any of you punks corrects me.
Posted by Michael | Link to this comment | 09-11-05 11:09 PM
we allocate resources unevenly between spending categories because of an interesting asymmetry
I take it this is supposed to read "interesting assymetry"?
Posted by ogmb | Link to this comment | 09-12-05 12:13 AM
Mitch, I'm gonna get you.
But why, Michael? Surely if anyone should get upset it's wolfson? Yours is an honest profession, but his tipping habits are atrocious and shameful(besides which they reinforce certain unfortunate stereotypes).
Posted by Mitch Mills | Link to this comment | 09-12-05 4:10 AM
I am not the victim of gay whore stereotypes!!!
You can make up for it by giving me your job.
Posted by Michael | Link to this comment | 09-13-05 3:47 PM