Frank Redux

Thirteen years ago, in 1999, when I wanted to illustrate the astonishing march of progress, my Exhibit A was a new $250 stereo system that held 60 CD’s and could play tracks in random order.

My new Exhibit A is the fact that it’s been only thirteen years since this was a great example.

That was in an essay focused mostly on Robert Frank’s hypothesis that people care largely about relative position as opposed to absolute wealth. I was reminded of that essay following yesterday’s post, and I managed to dig out a copy, which I’ve posted here. Despite the dated examples, I still think it’s pretty good.

Click here to comment or read others’ comments.

Print Friendly, PDF & Email
Share

24 Responses to “Frank Redux”


  1. 1 1 Rabault

    Interesting. I work in finance, where wages are well above average (including my own). When I see my colleagues picth for higher bonuses, so as to be able to change their Porsche, which is a car that makes sense only if you intend to breach the speed limits, I feel sympathy
    for Robert Frank. What flow of services do they derive from their Porsche ? Isn’t it that they feel good being seen in such a car ? What about taking a bit of that purchasing power to make sure poor people have adequate treatment for cancer ? What’s wrong with the argument that having a higher tax on luxury goods to redistribute to the poor is making the world a better place ? Would this tax really change the incentive to work ?

  2. 2 2 David R. Henderson

    Steve,
    A truly excellent book review. Nicely done! Here are my two attempts to untangle Frank’s views:
    http://econlog.econlib.org/archives/2011/12/my_review_of_ro.html
    http://www.cato.org/pubs/policy_report/v29n6/cpr29n6-1.html

  3. 3 3 Al V.

    Planet Money did a podcast some time ago where they interviewed a college professor, who likes to ask his students whether they would rather make $70,000. per year today, or make $70,000 per year in 1910. Obviously, if you earned $70K in 1910, you were rich – very rich. You were in the 1%, perhaps in the 0.1%. Today it’s a good income, but nothing special.

    The professor reported that when he asks the question, about 2/3 of respondants choose to live today, instead of in 1910.

  4. 4 4 nobody.really

    [R]esearchers have made the effort to tell precise, coherent stories about jockeying for relative position and to enumerate their consequences. My favorite example is a paper by Hal Cole, George Mailath, and Andrew Postlewaite (“Social Norms, Savings Behavior and Growth,” Journal of Political Economy 100 [1992]:1092–1125). Cole, Mailath, and Postlewaite (henceforth CMP) hypothesize that people care about their relative positions in the wealth distribution because a high relative position allows you to attract a better mate. Then, rather than just assert various consequences of their hypothesis à la Frank, they actually work out in detail what life would be like in such a world. And one thing they find is that the competition for mates drives most people to save too much.
    * * *
    I love that observation because it’s a clear, convincing, logical, and surprising consequence of a carefully stated assumption. It’s also, I think, self-evidently wrong.

    Ok, the second paragraph was not written about the first paragraph – but it should have been. It strikes me as self-evident that young men acquire expensive cars, rent a “flat that could flatten the Taj Mahal,” and engage in conspicuous acts of generosity in order to attract mates. Cole, Mailath, and Postlewaite will have to put forth a damn good explanation to overcome this widely-acknowledged phenomenon.

    And the explanation may be this: Competition for mates causes people to save more than they would otherwise. But how would potential mates know about this savings? They don’t. So mates look for indicia of wealth a/k/a conspicuous consumption. Young men then learn that they can get the benefits of savings without the cost of savings by simply acquiring the indicia of wealth.

    (A study suggests that higher social class predicts increased unethical behavior. While I find parts of the study sound, I was not impressed with the correlation between expensive cars and aggressive driving. I suspect the explanatory variable here is not social class, but testosterone.)

    People who enjoy conspicuous consumption in the present are likely to enjoy conspicuous consumption in the future, and they know that future consumption requires current saving.

    It is not clear to me that young men with the expensive cars and apartments do not eventually marry, settle down, and trade in the car for a minivan. In short, conspicuous consumption may be more important at some stages of life than others.

  5. 5 5 nobody.really

    Okay, let’s sort out this argument. In a society where people are jockeying for relative position, there are two possible justifications for a consumption tax. First, a consumption tax can discourage overwork and thereby make everyone happier. Second, insofar as people care only about relative position, and insofar as the tax leaves everyone’s relative position unchanged, it’s a means of generating tax revenue at no cost to the taxpayers. Those are separate arguments, and Frank doesn’t do a very good job of delineating them, so let me try to separate the issues.

    I’m missing the second point. Insofar as people care only about relative position, and insofar as the tax leaves everyone’s relative position unchanged, wouldn’t taxation generate revenue at no cost to the taxpayers? I don’t see where Landsburg refutes this proposition.

    More to the point, even if we relax the assumptions, would taxation of positional goods be a more efficient way of generating revenues than other forms of public finance?

  6. 6 6 Ken B

    nobody.really: “would taxation of positional goods be a more efficient way of generating revenues than other forms of public finance?”

    Yes, and yet I suspect you in fact support subsidizing Olympic athletes rather than taxing them.

  7. 7 7 Harold

    AlV. Perhaps the respondants knew about something happening in 1914 that made them choose today.

  8. 8 8 Harold

    I don’t follow your objections to the job part, which you say is self-evidently wrong.
    I am assuming we are talking about a gropup of employees on the same wage. The most productive workers see their less productive colleagues as a blessing and vice versa. This bit seems true. My partner was working on a student job placement, and was understandably keen. She was asked by other employees to do less because it was making them all look bad, so clearly less productive employees can see their more productive colleagues as a threat.

    So far so good. But then: “Therefore the least productive are overpaid”. Is this not obvious? They are getting more money for less production. I do not see where the status bit comes in. Who has more status – the productive worker who is liked by the boss but suspected by colleagues, or the slacker who is not liked by the boss, but who’s colleagues regard as a blessing?

    I would like to know which part you find “clear, convincing, logical, and surprising consequence” and why it is wrong.

  9. 9 9 Steve Landsburg

    Harold:

    I would like to know which part you find “clear, convincing, logical, and surprising consequence” and why it is wrong.

    I find the conclusion that the least productive are overpaid a clear, convincing, logical, and surprising consequence of Frank’s assumptions. I also think it is self-evidentlly wrong because, in my experience, the least productive are in fact the people everyone wants to get rid of. Your own experience, of course, may differ.

  10. 10 10 nobody.really

    I suspect you in fact support subsidizing Olympic athletes rather than taxing them.

    ??

    You would be mistaken. Indeed, I’ve never heard anyone advocate exempting Olympic athletes from taxation.

  11. 11 11 Ken B

    nobody.really has missed the.point. If it is more efficient to tax positional goods then we should apply extra taxes to Olympic competitors, not subsidies. We do the opposite, and I expect nobody.really supports the subsidies not the new efficient taxes.

  12. 12 12 Dave

    “By advertising that he’s held everyone else back, the politician can only make himself more attractive to you. The fact
    that no politician campaigns that way is evidence against Frank’s story”

    Wouldn’t such a politician only manage to gain 2 votes (from himself and the one person he hasn’t held back)? Seems like that would be a horrible campaign strategy even if Frank’s views are correct, so this is not evidence that he is wrong.

  13. 13 13 Steve Landsburg

    Dave Decker:

    Wouldn’t such a politician only manage to gain 2 votes (from himself and the one person he hasn’t held back)?

    Taking as given the actual effects of the policies, we each already know how we ourselves are doing. So any news that people in general are doing badly is good news for each of us.

  14. 14 14 Ken B

    @Dave: To amplify Steve’s last. I know my meagre pittance per week. Assume it has not changed. I learn that the average worker’s pay has decreased 2%. “Aha!” I reason, “I have not lost anything but the rest have, so I have gained in relative terms.” If I care very much about relative positions I must be pleased and approve. Not many other than Will A’s peasants do though.

  15. 15 15 nobody.really

    nobody.really has missed the.point.

    Cute! But perhaps not.accurate.

    If it is more efficient to tax positional goods then we should apply extra taxes to Olympic competitors, not subsidies. We do the opposite, and I expect nobody.really supports the subsidies not the new efficient taxes.

    And again, you’d be mistaken. Although I must confess, I don’t know a lot about how the Olympics are financed.

    But even if I did support subsidies for Olympic athletes, I don’t see how that would be inconsistent with a general preference for taxing positional goods/services. Olympic athletes seem like a pretty easily distinguished group.

    For better or worse, my neighbors seem to identify with athletes from our country and derive satisfaction from seeing those athletes succeed. To further that outcome, they may favor subsidies for those athletes. And one form of subsidy may be withholding taxes that might otherwise apply. In contrast, my neighbors don’t seem to identify with, or derive much satisfaction from, my Lamborghini. (I don’t let them into it, which might be part of the problem.) So I wouldn’t be surprised to learn that they would favor different tax/subsidy treatments for Olympic athletes than for my Lamborghini.

    In short, I see nothing odd about people raising public revenues in the most efficient way they can devise. And I see nothing odd about people subsidizing things they want to support. One phenomenon might offset the other, but I find nothing incongruous in favoring both. Or not.

  16. 16 16 Andrew

    Being an Olympic athlete IS a positional good. If you support taxing positional goods, then you should be taxing Olympic athletes instead of subsidizing them.

  17. 17 17 Harold

    I don’t think the job thing is too important – I would have to read Frank’s book to find out where he is coming from. I can only think that he assumes a lot of status attaches to work performance. I would guess that the more important the “status effect” e.g. employees with little power, the less status attaches to work performance. However, I don’t believe it is self-evident that it is wrong.

    People care about relative and absolute wealth. Frank may over emphasise the relative part, but it does play a significant role.

  18. 18 18 Mike H

    Now I’m wondering where I can get hold of some old shopping catalogues. If only I’d known they weren’t just junk mail! Hmm… maybe I should reconsider how I treat my spam folder?

  19. 19 19 Doc Merlin

    If thats true then why isn’t Obama trumpeting the fact that the top 1% took a massive hit to income in the recession and taking credit for the fact? Right there in the recession as usual income inequality dropped substantially.

  20. 20 20 Harold

    Take a market with only apples and oranges, with Alice and Bob representing two homogeneous, but distinct populations. The market ensures that Alice and Bob are as happy as they could be with the orange / apple situation, and we say their utility is maximised. Say for the sake of argument that this position is that each has equal numbers of apples and oranges.

    If oranges become status goods, then Alice and Bob will no longer have maximum utility with equal numbers, because oranges have become more desirable – not for their own sake, but for what they say about you. Bob buys more oranges, until his reduced utility of consuming more oranges (and fewer apples) than he prefers is balanced by his increase in status. However, Alice also buys more oranges, which reduces Bobs relative status. He now has all those extra oranges, and nothing to show for it. The number of oranges and apples produced and consumed must be different from the efficient result we had before, thereby reducing utility, but because the status is relative, it is impossible for the increase in utility from status to fully compensate.

    From this I conclude that if there is an element of positional motivation, the outcome must be inefficient. Is this correct? If not, where is the error?

  21. 21 21 iceman

    @Harold:

    My guess is the main thing missing in your example is that oranges, unlike oceanfront houses, are not inherently fixed in supply (but not sure why fancy cars are, or why (as others have pointed out) slots in good schools are not). If Bob and Alice want more oranges for whatever reason, growers can accommodate (price may not even change unless marginal cost does – but might it even go down?). As you describe, Bob enjoys at best a temporary increase in status, until Alice catches up and now they all wish they had more apples; but at that point what prevents apples from becoming the new status good? Moving beyond a two-good world, it seems to me the objects of our positional motivation (involving non-fixed resources) are constantly shifting; could one say fashion trends are all about creating opportunities for people continually to signal their astuteness through early adoption? You and I might tell ourselves that constant search for status is wasteful or stupid, but I’m not sure anything in utility theory would allow us to conclude it’s inefficient.

  22. 22 22 Harold

    iceman: Yes, my simple analysis ignores the fixed supply issue. It assumes some external source of “status”, that makes oranges more desirable only for this external reason. If the reason for the status is only limited supply, then my argument fails.

  23. 23 23 Advo

    I think Krugman wrote about the phenomenon of relative wealth at length a few years ago on his blog. If I remember correctly, he argued that income inequality actually had the biggest negative effect on the happiness of the top few percent, because that is where the income curve accelerates so much. So if you earn 400,000 USD, in your peer group there will be a lot of people making, twice, three times, ten times that amount, which is not the case for people making 40,000 USD.

  24. 24 24 TMS71

    If thats true then why isn’t Obama trumpeting the fact that the top 1% took a massive hit to income in the recession and taking credit for the fact? Right there in the recession as usual income inequality dropped substantially.

    Because people don’t like to admit that they care about relative wealth. It makes them seem petty and competitive. It is to admit that you care not so much about what you have but whether you have more than others. The fact that the top 1% took a massive hit is very comforting to most if they know about it but its very tricky to disseminate the information. You have to hide it as an ancillary fact when discussing other things. But I fail to see how Obama could benefit on balance by trumpeting this fact. He needs $$ from the 1% and if he’s celebrating and bragging about how he caused the 1% to lose a lot of money it seems that would be wildly counterproductive. None of this refutes that idea that it is relative wealth that is most important to people.

Leave a Reply