Lots of people (including Kevin Drum, Brad DeLong and Tyler Cowen have jumped in on this post by Will Wilkinson about this NBER study of habituation to changes in income and status. Wilkinson and most commentators focus on the findings regarding the subgroups on the right and left of the political spectrum, which I’ll come to, but it’s worth mentioning the general findings first.
Most people (in the German sample population) initially react more, as regards self-reported happiness, to a change in income than to a change in occupational status, but gradually get habituated to changes in income. This is consistent with the standard view of the happiness literature, that income changes don’t have a big effect on happiness, so that people in rich countries aren’t on average much happier than those in poor countries. Moreover, by looking at the same people over relatively short periods of time the analysis overcomes, to a significant extent, the objection I’ve made previously, that the scale on which happiness is measured is inherently relative to some notion of what is reasonable to expect.
The point picked up by Wilkinson and others is that, while those on the left of the political spectrum follow the general pattern (as do employees, taken as a group) those on the right (and the self-employed) habituate to changes in occupational status, but not to changes in money income. This seems to correlate pretty well with the fact that people on the left tend to worry about relative notions like inequality, authority relations and so forth, while people on the right dismiss these concerns and favour policies that create opportunities for making money. As quite a few commentators point out, the direction of causation is unclear here.
Rather than duplicate all the excellent discussion, I’ll offer the possibility that those on the right may not be so different from the rest of us as they seem to be at first sight. Money is valued because it provides access to goods and services, but it can also be used to keep score in a competition. It doesn’t seem likely, to take an Australian example, that the late Kerry Packer’s lifestyle would have been much different if he had been say, Australia’s tenth-richest man instead of the richest, but it’s hard to believe that this shift would have been a matter of indifference to him. On the other hand, in this kind of competition, the way in which the money is made (occupational status) may be of secondary importance.
{ 2 trackbacks }
{ 32 comments }
abb1 06.10.07 at 11:14 am
They get a pay raise and they’re happy about it for many years? Weird. Sounds like some anomaly or some important piece of data is missing.
abb1 06.10.07 at 11:25 am
Money is valued because it provides access to goods and services, but it can also be used to keep score in a competition.
Oh, OK. Yup, this sounds right on the nose. Heard it on AM talk radio a few times – some caller offering for no apparent reason that he makes ‘well over a hundred grand’.
Aidan Kehoe 06.10.07 at 11:49 am
Abb1, meh, that’s pretty much my attitude, and it’s hardly irrational.
alphie 06.10.07 at 12:26 pm
Looking at the current crop of Republican presidential hopefuls…I’d venture the theory that right-wing mistresses and ex-wives are more expensive than their left-wing counterparts to explain the data.
george w 06.10.07 at 2:23 pm
Fascinating implications. But it looks like they fail to reject the null on the effects of status (if I’m parsing that syntax correctly) so how can any conclusion be drawn? Did anybody read this paper beyond the abstract? It does cost five whole dollars after all.
Slocum 06.10.07 at 4:13 pm
I’ll offer the possibility that those on the right may not be so different from the rest of us as they seem to be at first sight.
But your explanation does not account for the finding that those on the right habituate to changes in status. Your thesis is that these folks care about money only as a means of score-keeping in a status competition (which is to say, they’re still status driven), but then you have to explain why they seem indifferent to status in other forms. Because it’s not as if one could not care about being the nth richest man AND also the nth coolest AND nth most famous …
And, of course, the group of people who are self-employed (as, full disclosure, I am) include countless millions who are nowhere near the top ten in any country (or state/province, or city) and who don’t have an accurate idea of how one’s income compares, so genuine score-keeping is near impossible.
I know this probably seems bizarre to academics, but not only is my job title meaningless, but my income is not a function of my rank nor does it rise steadily with seniority nor cost-of-living nor anything else. In fact, it varies pretty dramatically depending on business conditions (as much as a factor of 2 from lean year to fat year).
Which means operating with a buffer of savings many times greater than that of a tenured academic with a defined pension plan. Which, in turn, means spending during a given year is really not a function of income. So for a self-employed person like me, status in terms of titles/rank/promotion is generally meaningless and income is not of much use in score-keeping, since comparisons are difficult and the data is very noisy. But money IS extraordinarily useful in filling up the savings buffer — it has a much stronger effect on the sense of short and long-term security (which, of course, are complete non-issues for tenured academics) than it does on status.
I really don’t think we can reject the hypothesis that people in different life circumstances have different attitudes toward money & status (or that people who view money & status differently make different career choices).
abb1 06.10.07 at 4:25 pm
people in different life circumstances have different attitudes
But Slocum, you probably can find a job with higher status and less money; you prefer to be a contractor, it’s not your destiny.
Seth Finkelstein 06.10.07 at 4:34 pm
abb1 / #7 – I think the point slocum is making that while of course people trade-off income for status, the details of how much they do this trade-off can include a large component of risk and assocated variability that may not be accounted for in a simplistic model of “income”.
In particular, tenured academic is (relatively) low-income/high-status, but also low variability.
On the other hand, NON-tenured academic is low-income/high-status, but with a high variability too.
abb1 06.10.07 at 4:56 pm
I dunno, maybe it’s different now, but in the old days if you chose being a consultant, lean year or fat year – you’d still make a helluva lot more than a salaried employee of the save skill level. If that’s true, then the risk thing is just a red herring.
Seth Finkelstein 06.10.07 at 5:44 pm
That’s much less true now in the era of outsourcing/offshoring/”global economy”. Companies have adapted, and pushed a lot of risk onto the consultants. They’ve basically figured out that it’s a way they can avoid paying health benefits and unemployment insurance, and well as escape many labor laws. There have been some big lawsuits around this area, notable involving Microsoft.
Like many economic changes, the very top often make out like bandits, but everyone else can be worse off.
Slocum 06.10.07 at 7:14 pm
But Slocum, you probably can find a job with higher status and less money; you prefer to be a contractor, it’s not your destiny.
But isn’t this discussion about preferences? Yes, I could probably find a job as an W2 employee with greater status and less money, but I don’t prefer that — that’s the point. And nor is the premium for working as a independent a form of status competition by other means (e.g. score-keeping).
I dunno, maybe it’s different now, but in the old days if you chose being a consultant, lean year or fat year – you’d still make a helluva lot more than a salaried employee of the save skill level. If that’s true, then the risk thing is just a red herring.
On average over the last decade plus, I’ve made more than I would have as a salary employee (though that depends on what I assume about the hypothetical career track I didn’t follow), even after figuring paying for my own fringe benefits, vacation, and continuing education out of pocket. But in the few lean years I’ve made less. And “more money” was not really a deciding motivation in comparison to, say:
– greater autonomy
– getting away from big organization politics
– working from a home office to have time with my kids
– not wasting hours a week commuting
In other words — quality of life, rather than cash per se (and definitely rather than status — more than one casual acquaintance has come away with the impression that I might not have a real job at all, let alone a high-status one).
abb1 06.10.07 at 7:26 pm
I guess I don’t understand what your objection to JQ’s explanation is. Suppose my idea of status is my grade in the hierarchy and your idea of status is independence (which, incidentally, requires a lot of cash). Seems all right, what does it have to do with risk?
Tom T. 06.10.07 at 7:57 pm
The study abstract simply says “German” population. Given that the study stretches back to 1984, it seems relevant to ask whether “German” in this context means “West German”? If the study includes former East Germans whose notions of income and status were formed under a Communist regime, that might skew the results a bit.
Tom T. 06.10.07 at 8:03 pm
And maybe someone who’s read the study could expound upon what is meant by “one standard deviation in status”? Status seems like a quantity ill-suited for for such apparent precision. We don’t normally speak of “one standard deviation in beauty,” for instance. Moreover, changes in status seems hard to classify. As #8 points out, “tenured academic” seems high-status to a non-tenured academic, but suppose the tenured academic then takes a job in government or private industry — is that an increase in status or a decrease?
John Quiggin 06.10.07 at 8:32 pm
Slocum, my point is that if you measure your success by income, then that (rather than occupational prestige) is your main status measure.
Tracy W 06.10.07 at 9:09 pm
Most people (in the German sample population) initially react more, as regards self-reported happiness, to a change in income than to a change in occupational status, but gradually get habituated to changes in income.
This is interesting – it implies that money has more of a long-lasting impact on most people than their occupation, which perhaps I can madly generalise to income being more important to many people than occupational status.
Thomas 06.11.07 at 3:07 am
I think that, to the right, one’s “score” in life is very important. They value the status that goes along with both income and with occupational status as it gives them a perceived social edge, a superiority, over people who lack such status.
Simply, to someone on the right, fair is what allows them to win and to someone on the left fair is synonymous with equality. I’m not sure that either is correct.
Tom T. 06.11.07 at 3:39 am
#15 and #17, those assertions may be true, but insofar as they conflate income and status, they don’t really have anything to do with the NBER study, do they, since that study measured income and status separately?
Slocum 06.11.07 at 3:45 am
Slocum, my point is that if you measure your success by income, then that (rather than occupational prestige) is your main status measure.
Well, but the way this argument generally leads (from the left) is that absolute wealth is worthless from the point of view of subjective well-being because all people do with extra wealth is to spend it on ‘positional goods’ in a wasteful, zero-sum game. So (according to this view) we would all be better off with a lot more taxation and redistribution, since nobody really cares about wealth for it’s own sake, but only for score-keeping, and that could be done just as well at a lower level of absolute wealth as a higher one.
What I’m arguing is that it is quite likely that more money would, indeed, have a more lasting effect on me, personally, than increased status (in the form of a title or some kind of recognition) but NOT because it would increase my score in the zero-sum capitalist game of life where dollars=points, but rather because more money would buy a variety of things that *I* value, but that are not generally associated with status (and no, just because I value them does not make them a form of status. Status is inherently social. Self-esteem is not status).
Maybe my caring about quality of life factors such as work-family balance, security, and independence is highly idiosyncratic. Perhaps thoughts of earlier rather than later retirement are odd. Possibly very few others are interested in leaving their kids with college degrees unencumbered by student loans. But I doubt it. And these all depend on absolute, not relative wealth and are valuable in themselves, not because they get one ahead of a few more of the Joneses.
Consider the possibility that at least some of the strange people who habituate to status but not to extra income are like that not because money=points=status for them but because status really isn’t what they care about most. It is possible, isn’t it?
BillCinSD 06.11.07 at 5:10 am
Hasn’t Robert H. Frank been writing about this point for nigh unto 20 years now? Of course i haven’t yet checked how frank and the NBER paper correlate.
Matthias Wasser 06.11.07 at 5:15 am
Maybe my caring about quality of life factors such as work-family balance, security, and independence is highly idiosyncratic. Perhaps thoughts of earlier rather than later retirement are odd. Possibly very few others are interested in leaving their kids with college degrees unencumbered by student loans. But I doubt it. And these all depend on absolute, not relative wealth and are valuable in themselves, not because they get one ahead of a few more of the Joneses.
Certainly, but these are precisely the sorts of things that the redistribution is supposed to go towards.
You can of course make an argument that redistribution will hamper total growth, by reducing incentives and so on, and that would hardly be ridiculous. But for it to really validate your point you’d have to show (or show that it’s reasonable to expect) that it would reduce growth sufficiently that we couldn’t provide all of those above things to everyone. At this point I think it’s clear that there’s enough total wealth that the above can be provided at reasonable levels to everyone and that the redistribution that achieves it wouldn’t crash the economy below the point that we couldn’t provide enough for everyone. So inasmuch as the only barriers to these objective material goals are distributional…
It’s possible that your standards for these things are higher than I expected, but based on the happiness literature available and how absolute wealth for countries stops mattering around averages of $20K US or so, my guess is that you’re (in the aggregate) not.
Matthias Wasser 06.11.07 at 5:18 am
Shorter version of the above: the argument from the left isn’t that absolute wealth doesn’t affect happiness, but that it does so with decreasing marginal utility.
Sebastian Holsclaw 06.11.07 at 5:46 am
“Slocum, my point is that if you measure your success by income, then that (rather than occupational prestige) is your main status measure.”
This is a tricky area. Money is multi-functional. It CAN be used as merely another status measure, but it doesn’t have to. And even identifying money as someone’s ‘main status measure’ does not show that status measures are as important to that person as they are to others.
We all know people who seek desperately after status. We all know others that don’t care so much. There seems to be a continuum of status seeking. There isn’t any reason to believe that continuum is evenly distributed.
Donna 06.11.07 at 9:34 am
Does the article specifically define what they mean by happiness? If the group studied have universal health care, it would seem, that the results can only be applied to other countries who take care of their sick — meaning we can’t apply their results to the good ole USA.
I may be missing the point here, of course.
John Quiggin 06.11.07 at 10:34 am
“they don’t really have anything to do with the NBER study, do they, since that study measured income and status separately?”
The authors of the study measure occupational prestige and call it status.
Tom T. 06.11.07 at 12:25 pm
So to flip #1 around, they get a new title and years later they’re three times happier than they would have been with a raise? Weird. And what a boon to employers.
abb1 06.11.07 at 12:48 pm
#27 Weird? Maybe it is. But isn’t it clearly more typical?
Slocum 06.11.07 at 1:09 pm
At this point I think it’s clear that there’s enough total wealth that the above can be provided at reasonable levels to everyone and that the redistribution that achieves it wouldn’t crash the economy below the point that we couldn’t provide enough for everyone. So inasmuch as the only barriers to these objective material goals are distributional…
I don’t think it’s clear at all that there’s enough total wealth so that the kinds of things I listed could be provided to everyone via redistribution (leaving aside the likelihood that massive redistribution would shrink that wealth — which is a dynamic product not a fixed lump). For example, the long-term trend is toward more time out of the paid work force, both on the front end (more years of schooling) and back end (many more years of retirement). For that trend to continue, absolute levels of wealth will need to continue to grow.
And, again, this is the point of the discussion, but everyone doesn’t have the same preferences as I do. Some would rather have a sharp stick in the eye than work from a home office and supervise the kids. And some have absolutely no idea what to do with themselves when not gainfully employed, so early retirement would be a nightmare. And many, of course, do care very much about traditional forms of status, and derive satisfaction from being promoted to full-professor or becoming department chair or editor of a prestigious journal and dream of being elected a member of the NAS.
And then consider the recent rhapsodizing here over the iPhone. Is that all about a positional good that Kieran would wield to demonstrate his rank and humble his inferiors (and which opportunity for zero-sum one-upmanship a just society would deny him)? Or is it just possible that he would experience non-status derived aesthetic pleasure from a beautiful, functional object that might reasonably considered a work of art? And is it possible that a life with more of such gracefully designed objects is more satisfying than one without — regardless of what the Joneses have? Is Kieran’s attraction to the iPhone an indication that even CTers can’t escape the consumerist status competition? Or was he accidentally lending support to Virginia Postrel’s thesis?
Michael B Sullivan 06.11.07 at 6:37 pm
While we’re all offerring unsupportable hypotheses about How People Are, I have one of my own:
I think that people who don’t habituate to monetary differences may be using money to keep score, but using it to keep score for themselves, not “in relation to other people.” They’re playing an essentially solitary game, trying to exceed their previous personal best. That’s why they aren’t terribly affected by status changes — status is inherently social, and they aren’t interested in approval by others. Money at least appears to be an objective standard, and they can say, “If I’m making $100k now and was making $50k ten years ago, then that means I’m better at (whatever) now — I’ve progressed, go me.”
clew 06.12.07 at 5:47 pm
Perhaps thoughts of earlier rather than later retirement are odd. Possibly very few others are interested in leaving their kids with college degrees unencumbered by student loans. But I doubt it. And these all depend on absolute, not relative wealth and are valuable in themselves…
I agree that these are valuable in themselves, but the astounding increases in college and real-estate costs are partly because they are dependent on relative wealth.
clew 06.12.07 at 5:49 pm
Oh, and as a possible explanation for the non-habituation to job title changes; these do often reflect a change in the actual job done, with an increase in autonomy and interest, decrease in risk, etc.
Tom T. 06.13.07 at 12:08 am
#32: Perhaps, but the study purports to be measuring job prestige, not job responsibility. Certainly, these elements can be hard to separate, but they’re often hard to separate from income, too.
Comments on this entry are closed.