Powers and HisPet Monkey

[From Bill Williams UMKC 7 November 2002 6:15 AM CST]

[From Bill Powers (2002.11.04.0817 MST)]
[From Rick Marken (2002.11.04.0950)

THere are many things in Bill Powers' comment, with which I do not agree.
Powers, however, does confirm my understanding that the paper contains an error
in regard to a claim it makes regarding the behavior of goods when the budget
is above the point at which the budget for meat and for bread more that
provides enough purchasing power to satisfy the reference levels. There are
other errors in the paper which are not mine, but this claim regarding what
happens when the budget exceeds the point of satiation is the most important
mistake of which I was not the source.

There are many things in Rick's reply which I do not find at all satisfactory.
The one which I will reply to here is Rick's suggestion that I might have been
"somewhat dishonest" when I published a paper when Powers had done the work.
Several kinds of "work" went into the paper. I wrote about the Giffen paradox
in a dissertation completed in 1972. But I didn't have a method which would
generate a solution to the paradox. It took me 13 years to find Powers work.
Then Powers and I collaborated in finding a solution. Since I didn't know
anything about programing at that time Powers wrote the code. I had, however
had the experience of designing the electronics for control units to be
installed on agricultural machinery. So, I did know something about control
theory. And, at one point in Bill Powers' effort to get a simulation running a
seemingly intractible problem developed. Looking at the code I saw that the
price of calories obtained by meat and bread had been set to the same value.
Given this specification, of course, the program couldn't run properly. So, I
did contribute not only the problem, but also a small portion of the work that
went into the initial giffen program. But, your complaint whatever the
distribution of effort within the colaboration and proper distribution of
credit is not with me. It was Powers who told me he didn't want his name to go
on the paper. (Now after Bill's recent disclosure, it can be inferred that he
did so as a result of a reaction to his experience with senior researchers
hogging the credit.) I would have been more than happy to have Bill Powers'
and my name appear on a paper as co-authors. I still would be. (But, I wouldn't
expect this is a likely prospect regarding a paper on "leakages.") So, my
answer would be no. If I made an ethical mistake it was in not arguing with
Powers and emphatically insisting that he sign onto the paper. In my upbring,
however youngsters were not encouraged to argue with the adults. So, If the
paper was going to be published, it was Powers' wish that his name not appear
on it.

You didn't respond to my charge that you "induced" me to republish the Giffen
paper, and then were heard complaining about my wanting to publish it. This
for me was equally as serious as your behavior as editor.

However, in contrast to my objection of much of what you have to say, I agree
with you when you say,

   "it's possible that I did not show you the final version that I sent to the
   printer."

OK. This has always been my understanding of what happened. I would be
satisfied if you would send me a letter stating that you introduced changes in
substance into the paper and that I did not approve of these changes. As a
"factual matter" I don't think there is any question that this happend. By
stating that this is what happened you would free me from responsiblity for a
paper, substansively significant parts of which which I did not write.

You argue at one point, that the printed version isn't all that bad. I
disagree. What it does is, ... well, it makes me sound like you. And, that is
just unbearable.

If you'll agree to release me from responsiblity for the paper, there will be
some issues to consider about the specific language. I don't intend to be
difficult this. But, there's a guy on campus who serves as an "ethicist"
resource, and I'll want to optain his advice concerning language the letter
should contain-- things like being signed in blue ink and other such strange
details.

Bill Williams

First I will reply to Rick's comment that "...is it not somewhat dishonest
to write a paper describing work carried out by someone else as though
it were done by you?" First the solution to the problem of the Giffen
paradox was achieved as a result of a colaboration between Powers and I.
I had writen about the paradox as a part of a dissertation completed in
1972. But, I didn't then have the methods required to solve it. It took me 13
years to find Powers' work. And even then I couldn't see a way to apply control
theory. I had experience developing control systems for agricultural machinery,
but that's quite different than models where control systems are in conflict.
For your information its not like I presented Powers with the problem and Bill
then solved it all by himself. It was quite a bit more difficult to think about
what was going on before I got the idea that the caloric and budget functions
could be depicted graphically. At one point the attempt to get the computer
model to work was stuck until I realized that the price of the meat and bread
had by accident be specified, so that the price per calorie was the same-- you
don't get the Giffen effect if this is the case. So, I was able to make
something of a contribution to the development of the model. Second, it was
Bill Powers who said not to put his name on the paper. I would have been more
than happy then to have had my name associated with POwers as co-author of a
paper. And, I told Powers this. ANd, I continue to be of this inclination. When
I eventually get the book completed, it will contain a generous acknowledgement
of Powers' contribution-- not only to the initial solution of the Giffen
effect, but as having supplied the method used throughout the volume. Third, if
this is so obvious to you now, why did you publish a paper which you now
describe as "somewhat dishonestd?"

You did not respond to my complaint that when you requested that I submit a
paper you demanded the Giffen paper rather than the one I wished to publish.
THen you c

···

Bill Williams UMKC 3 November 2002 3:00 PM CST --

>Rick Marken: The question of dishonesty.
>
>Rick Marken edited a set of papers by members of the Control Systems
>group which appeared in "The American Behavioral Scientist volume 34
>number 1 September/October 1990 under the caption "Purposeful Behavior
>The Control Theory approach" A paper that is attributed to me appears
>in the issue under the title "The Giffen Effect: A Note on Economic
>Purposes." The paper however is not the one I wrote for this issue. When
>I saw the published version it was immeadiately apparent that the paper
>had been cut without my permission to about half of the length of the
>paper I submitted. And, Rick in additon had made changes in substance.

I checked my recollection of this with Rick, We both think that if
substantive changes were made in the paper without your permission, that
was a serious mistake. Rick will address that point, and others, himself.

The cut in length was no more than the rest of us who contributed were
required to do. I shortened my paper; Wayne Hershberger had a large section
of his paper dropped. Ed Ford's paper was drastically edited by Rick in
consultation with Ed over the phone. I don't know about the others, but I'm
sure everyone had to cut the length, because the journal's editor said that
the issue was much too large and had to be cut to be published. I recall,
not very clearly, a general notice send out to all authors by Rick warning
that the journal issue was too large. I do not know the circumstances
regarding your particular paper, but it's hard for me to believe that Rick
would single you out as the one person whose paper would be reduced in size
and edited without notification or getting permission to make editorial
changes. Could it be that you were unavailable at the critical time?

I agree with you that the diagram as shown in Fig. 1 does not allow
concluding that if the budget is large enough, the "normal" demand curve
will be seen, with the consumption of bread decreasing when the price
rises. Rick was simply mistaken in adding that sentence on p. 108 to the
paper. On the other hand, the Figure by itself does not allow concluding
that consumption of bread will rise when the price increases, either. As
you may remember from the development of the Basic program, it is necessary
to add a preference for meat over bread (you suggested calling this
preference "prestige") in order for that effect to appear, and this
preference is not indicated in Fig. 1, though it is mentioned in the text
and it does appear in the Basic program cited just prior to that part of
your paper. According to Fig. 1, if the person prefers a (vegetarian) diet
consisting mostly or wholly of bread, the Giffen paradox will not occur.
The diagram does not show any reference levels, so these effects have to be
added, as it were, offstage.

I'm not sure -- you may want to investigate this -- but in the original
Basic program, a moderate preference for meat will, I think, lead to a diet
consisting partly of meat and partly of bread, in the absence of any
budgetary constraint. If the price of bread then increases, the calorie
requirement will demand that the sum of meat and bread calories remain the
same, so the relative gains of the meat, prestige, and bread control
systems will determine whether the consumption of bread will increase or
decrease.

Also, I point out that control theory _in general_, if not Figure 1 in your
paper, requires that as the obtained amount of a good increases, the effort
to obtain more of it (i.e., and amount of money that will be spent) will
decrease, as the discrepancy between the amount obtained and the amount
wanted decreases. So the downward-sloping demand curve is an inherent
prediction of control theory, and of the Basic model of the Giffen effect.
Technically, therefore, the second mention of this effect (p. 109) is
correct, though it may not have been part of the original paper. In terms
of the intent of the paper, of course, it is simply a repeat of the first
mistake.

Grievances seem to grow in proportion to the time they are nurtured. Twelve
years is a long time to wait before confronting the source of this
grievance. If I appeared indifferent at the time (to what?) it may be
because of your habit of letting your complaints leak out only in the most
indirect and allusive way, until the internal pressure makes you blow your
stack completely. It may not have struck me that having a second paper on
the Giffen effect published only a year after the first appearance of this
model in print was a big deal to you, so any minor changes in it, while
annoying, were certainly not a big problem for you -- I probably thought.
Hard to remember what I _really_ thought that long ago.

>For instance, on page 108 Rick says "Increase the budget and the _same_
>model behaves as classical economics predicts-- an increase in the price
>of bread leads to a decrease in the reference level (and hence the demand)
>for bread."

Not having your original paper, I can't be sure what was and was not
changed, but it strikes me that your "for instance" might not be just one
of many such additions by Rick as the above implies, but in fact the _only_
such change (repeated a page later in the summing-up paragraph). I hope you
realize, Bill, that in 1990 your writing was not up to your present
standards and it was often very hard to figure out what you meant. What you
see as changes could, in some cases, be nothing more than an editor's
attempt to clarify (erroneously) what you wrote. Like the commentator who
changed "Behavior: the control of perception" to "Perception: the control
of behavior," kindly correcting my slip of the pen.

Is it possible that, on this occasion and perhaps during a later argument,
Rick said some things that were a bit dishonest? Sure. Same goes for me and
you. Does this mean Rick is a Dishonest Person? To quote a local authority,
"Jesus wept." Or more plainly, hell no. Are you a Trait Psychologist?

Bill P.

______________________________________________________________________
Do you want a free e-mail for life ? Get it at http://www.email.ro/

[From Rick Marken (2002.11.07.0845)]

Bill Williams (UMKC 7 November 2002 6:15 AM CST) --

There are many things in Rick's reply which I do not find at all satisfactory.
The one which I will reply to here is Rick's suggestion that I might have been
"somewhat dishonest" when I published a paper when Powers had done the work.
...It was Powers who told me he didn't want his name to go
on the paper...If I made an ethical mistake it was in not arguing with
Powers and emphatically insisting that he sign onto the paper.

I don't think you made an ethical mistake, at least, not one of any significance.
I think you might have mentioned that Powers designed the architecture of the
model (two level model, as I recall). But I wasn't questioning your ethics. I
really don't think you did anything particularly wrong. I was just pointing out
that ethics aren't black and white. You made a judgment regarding the mention of
Powers contribution to the modeling and I made a judgment about editing your
paper. Both judgments were questionable, but not terribly wrong.

You didn't respond to my charge that you "induced" me to republish the Giffen
paper, and then were heard complaining about my wanting to publish it. This
for me was equally as serious as your behavior as editor.

As best as I can recall, I asked you (along with all the others whose articles
appear in the issue) to contribute a paper to the ABS issue. You sent me a paper
(on Veblen or something) that I judged (as editor) to be unacceptable (for many
reasons). So either you or I suggested replacing it with the Giffin paper. I
remember it as being me who asked _you_ to submit the Giffin paper. This seemed
like a better option (as I recall) than simply rejecting the Veblen paper and
ending it there, with no paper from you in ABS. Since I had decided to publish
papers from everyone who submitted, I think I must have asked you to contribute a
paper that I knew would be the closest to being acceptable to me (as editor).

I knew that you had already published a paper on Giffin in _Volitional Action_ but
I was willing to try to publish it in ABS because it did involve PCT modeling. I
don't remember complaining about you wanting to publish it. If I did complain,
perhaps I was complaining about the fact that the Giffin paper you sent was so
similar to the one in _Volitional Action_. I guess I'm sorry you heard me
complaining at all. But, since you did, I guess I might as well be totally honest
and admit that my evaluation of the quality of your work is about the same as your
evaluation of the quality of mine. I thought I was doing you a favor by publishing
your paper in ABS. I guess your enmity is my reward for trying to do you a favor.
I am definitely getting out of the "doing favors" business!

If I were acting as a real editor I would have rejected your papers and simply not
published them. But if I were acting as a real editor I would have had to reject
all but about four of the papers I received for that ABS issue. So there would
have been no PCT issue of ABS, which would probably have been the right thing to
do (not publish it) because that would have given a more honest indication of the
state of PCT science in 1990.

OK. This has always been my understanding of what happened. I would be
satisfied if you would send me a letter stating that you introduced changes in
substance into the paper and that I did not approve of these changes.

I will gladly send you an e-mail to that effect. But there seems to be something
wrong with your e-mail address. I got a "bounce" message from
w.d.williams@email.ro. But here's what I will say:

As Editor of the 1990 ABS issue on "Purposful Behavior" I introduced changes in
substance into the "Giffen Effect" paper written by William D. Williams. Dr.
Williams did not approve of these changes.

You argue at one point, that the printed version isn't all that bad. I
disagree. What it does is, ... well, it makes me sound like you. And, that is
just unbearable.

I think that's fair. As I said, in the Giffen paper you sounded like you, which
was obviously unbearable to me.

If you'll agree to release me from responsiblity for the paper, there will be
some issues to consider about the specific language. I don't intend to be
difficult this. But, there's a guy on campus who serves as an "ethicist"
resource, and I'll want to optain his advice concerning language the letter
should contain-- things like being signed in blue ink and other such strange
details.

Send me whatever you like. I'll be happy to sign it.

Best regards

Rick

···

--
Richard S. Marken, Ph.D.
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[From Rick Marken (2002.11.07.1100)] Happy Birthday Mary!

Bill Powers (2002.11.07,0926 MST) [Mary's birthday]

I think that accusations of dishonesty are unseemly and exaggerated to
begin with, but to reply as Rick did by saying "Well, you are too" is even
worse. First, while such a retort is an admission that the accusation is
correct, it is also a statement that if someone else is also dishonest,
then that makes it all right to have been dishonest oneself. That sort of
argument belongs in Second Grade, not in an adult discussion.

I hope you can see from my earlier post [Rick Marken (2002.11.07.0845)] that I was
not saying "Well you are too". I was trying to say (not well, perhaps) that these
are complex situations. I _could_ have found fault with Bill Williams' failure to
mention you as the architect of the control model of the Giffen effect but I
didn't because I suspected that there were other considerations involved. And
there were, as you mention in your post. In my case, there were also other
considerations involved, such as the desire to publish Bill Williams' paper no
matter what and my assumption that he (like all the other authors I edited)
accepted my edits in advance as the implicit "price" of having their articles
published (rather than rejected outright). I should also point out that at the
time I was editing the papers for the ABS issue, everyone (including Bill
Williams) respected my knowledge of PCT and (I think) liked me personally. Bill
Williams respect for and liking of me may have ended when he saw the published
version of his paper. But going into the editing process I think all those who
submitted papers 1) appreciated the fact that I had given them an opportunity to
be published 2) respected my judgment regarding PCT and 3) respected my
integrity. The tone of the whole editing process was very friendly and
respectful. I, of course, tried to make sure that all authors were happy with any
edits I made. But many of them (as I recall) had a "do what you think is right,
Rick" attitude. The ABS issue came out well before the "Rick is an asshole"
mythology started in the mid 1990s.

So, though I certainly should have gotten Bill Williams' approval before
publishing his paper (if I didn't), I think he may be remembering the whole thing
as being worse than it actually was because he is looking back at it through the
"Rick is an asshole" glasses that he didn't (to my knowledge) have on back then.

Best

Rick

···

--
Richard S. Marken, Ph.D.
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[From Bill Powers (2002.11.07,0926 MST)] [Mary's birthday]

Bill Williams UMKC 7 November 2002 6:15 AM CST --

>There are many things in Bill Powers' comment, with which I do not agree.

All right, let's get them straightened out.

>Powers, however, does confirm my understanding that the paper contains
an >error in regard to a claim it makes regarding the behavior of goods
when the >budget is above the point at which the budget for meat and for
bread more >than provides enough purchasing power to satisfy the reference
levels.

Yes, I still agree with that.

>There are many things in Rick's reply which I do not find at
all >satisfactory. The one which I will reply to here is Rick's suggestion
that I >might have been "somewhat dishonest" when I published a paper when
Powers had >done the work.

I think that accusations of dishonesty are unseemly and exaggerated to
begin with, but to reply as Rick did by saying "Well, you are too" is even
worse. First, while such a retort is an admission that the accusation is
correct, it is also a statement that if someone else is also dishonest,
then that makes it all right to have been dishonest oneself. That sort of
argument belongs in Second Grade, not in an adult discussion.

My second objection to Rick's remark is that while I did develop the
control-system model to reproduce the Giffen effect, the solution had
already been worked out by you, Bill Williams, in graphical form. My
contribution was to show how multiple control systems (at first two, then
three) operating simultaneously could produce the same net effect that Bill
W. was illustrating graphically, and was explaining as a sort of sequential
process.

Several kinds of "work" went into the paper. I wrote about the Giffen paradox
in a dissertation completed in 1972. But I didn't have a method which would
generate a solution to the paradox. It took me 13 years to find Powers work.
Then Powers and I collaborated in finding a solution.

This is correct, except that you already had the solution in that you
showed that the effect was explainable and not a paradox at all.

>Since I didn't know anything about programing at that time Powers wrote
the >code.

That's a little disingenuous, Bill -- I did more than "coding" a model that
you had already laid out While you had seen that the explanation lay in a
person's trying to reach several goals at the same time, you were not then
sufficiently acquainted with the idea of interactions among simultaneously
operating control systems with different goals to see what kind of model
was needed. The explanations you were offering were all cast in terms of
adjusting one thing, looking at the result, adjusting something else, and
so on -- a sequential model. My model was based on three independent
control systems operating simultaneously. It took you some time to
understand it (as you understand it now, I presume).

>I had, however had the experience of designing the electronics for
control >units to be installed on agricultural machinery. So, I did know
something >about control theory.

I can hardly comment on that without seeing the design you came up with. If
you ever showed it to me, I've forgotten what it was.

And, at one point in Bill Powers' effort to get a simulation running a
seemingly intractible problem developed. Looking at the code I saw that the
price of calories obtained by meat and bread had been set to the same value.
Given this specification, of course, the program couldn't run
properly. So, I did contribute not only the problem, but also a small
portion of the work that went into the initial giffen program.

The "intractible problem" was resolved when I realized that the system
needed some way to express the assumed preference for meat. It was you who
suggested that we add a "prestige" goal, while I worked out the details of
how that would affect meat versus bread purchases.

I think your recollection of the price per calorie having been set to the
same value is irrelevant, even if correct, because the whole point of the
Giffen demo (which I understood perfectly well) is to _vary_ the price of
bread relative to the price of meat (per calorie) and see the effect on
purchases of bread. The prices may have been set to the same value as a
starting point, but the program I wrote had provision for changing those
prices, as it had to in order to demonstrate the effect. You may have
misinterpreted seeing those prices set to the same value at the start of
the program, not realizing that they were variables to be adjusted. You did
contribute to the solution of the real problem, but I think you are
misremembering what it was.

But, your complaint whatever the distribution of effort within the
colaboration and proper distribution of credit is not with me. It was
Powers who told me he didn't want his name to go on the paper. (Now after
Bill's recent disclosure, it can be inferred that he did so as a result of
a reaction to his experience with senior researchers hogging the credit.)

Wrong inference: my name was first on the final paper on which Clark and
MacFarland collaborated. The issue was not credit but control of what is
now called intellectual property (yuck).

I have often passed on programs and ideas to other people to use as their
own, hoping they would benefit professionally from it, and that they would
develop their own expertise, as they often did. I have been in print often
enough to have got over the novelty of seeing my name published. And so
far, people have always given me credit in a general way, which is enough
for me.

I should also mention that, in addition to the above, I usually suggest
that my name not be used if there are substantial aspects of an article
with which I disagree, but which I would not want to prevent being
published just because of that. I do not generally give that as my reason,
however, even if that is a little bit dishonest.

>In my upbring, however youngsters were not encouraged to argue with
the >adults. So, If the paper was going to be published, it was Powers'
wish that >his name not appear on it.

I don't think of you as a youngster, Bill, whatever your illusions of youth
may be ;-).

You didn't respond to my charge that you "induced" me to republish the Giffen
paper, and then were heard complaining about my wanting to publish it. This
for me was equally as serious as your behavior as editor.

Yeah, that's a pretty bad lapse. But wasn't it you, Bill, who suggested
that we might observe a statute of limitations on such sins? I think a red
face on Rick is payment enough.

I would be satisfied if you would send me a letter stating that you
introduced changes in substance into the paper and that I did not approve
of these changes. As a "factual matter" I don't think there is any
question that this happend. By stating that this is what happened you
would free me from responsiblity for a paper, substantively significant
parts of which which I did not write.

How about this: "I, Rick Marken, did with malice aforethough and intent to
defraud and discredit Bill Williams arbitrarily and without reason alter
the text of Williams' article [title] without consultation with said
Williams and caused same to be published without his permission or
knowledge. I agree to resign from the Control Systems Group and never to
speak of anything relating to it again, as well as paying damages of $[to
be agreed] to Williams and writing letters to every person who may have
read the article explaining that any errors in it, concerning any subject,
were mine, not those of Bill Williams."

Let's see, you want that signed in blue ink, you say. Anything else we
could get this guy on? This is starting to sound like fun. Maybe we can get
this guy to auger in totally. Oh, it's so lovely to be RIGHT.

Best,

Bill P.

Best,

Bill P.

"Pay no attention to that man behind the curtain."
The Wizard Of Oz

"Convince a man against his will;
He's of the same opinion still."
Embroidery by Grammy Alice

[From Bill Williams UMKC 7 November 2002 7:50 PM CST]

[From Rick Marken (2002.11.07.0845)]

Send me whatever you like. I'll be happy to sign it.

Good. I am checking with the schools affirmative action officer who also takes
care of academic disputes. I think pretty much what you suggested should do,
but I'll wait to see what the Affirmative action office says. NOt really a
rush, but I don't want to delay closing this out.

It will be nice to have this behind me.

Bill Williams

···

______________________________________________________________________
Do you want a free e-mail for life ? Get it at http://www.email.ro/

[From Rick Marken (2002.11.08.0840)]

Bill Williams (UMKC 7 November 2002 7:50 PM CST) --

> [From Rick Marken (2002.11.07.0845)]
>
> Send me whatever you like. I'll be happy to sign it.

Good. I am checking with the schools affirmative action officer who also takes
care of academic disputes. I think pretty much what you suggested should do,

And that's pretty much all I'll sign. I don't think I'd sign the disclaimer that
Bill Powers suggested (hysterically funny, by the way, Bill. But who knows what
people will think of as funny these days).

It will be nice to have this behind me.

And I'm glad you got it off your chest, not only because it will make you feel
better but also because it has rejuvenated my interest in the Giffen phenomenon.
I wrote up a little model of the phenomenon last night (which turns out to be a
close approximation to Bill Powers' GEM model that you describe in your
_Volitional Action_ paper, though the fact that they are similar was just a
coincidence; I wasn't trying to reproduce Bill's model). One thing I've already
learned is that, although the Giffen effect (increase in consumption with increase
in price) does depend on the size of the budget, this dependence does not occur
for the reason I thought. This misunderstanding was the reason why I added the
offending sentence ("Increase the budget and the _same_ model behaves as classical
economics predicts-- an increase in the price of bread leads to a decrease in the
reference level (and hence the demand) for bread) to your paper.

What I have found is that, in the control model of the Giffen effect, the "upward
sloping demand curve" for bread results from the fact that consumption of bread
serves control of calories only. That is, in both Bill Powers and my model, the
system controlling for keeping expenditures within budget acts to control
expenditures by varying only the reference for consumption of meat, not the
reference for consumption of bread. The system controlling for caloric level
controls calories by varying _both_ the reference for consumption of meat and the
reference for consumption of bread.

The "upward sloping demand curve" goes away when budget is increased, but this is
because the system no longer needs to buy _any_ bread to maintain caloric intake.
I had thought that the effect of budget was on willingness to buy bread and that
when budget increased the effect of price on consumption of bread would be as
expected; decrease in consumption of bread with increasing price. But the fact is
that the price of bread is irrelevant to the model, except as a disturbance to the
expenditure control system. And this disturbance can be compensated for _only_ by
varying consumption of meat; variations in consumption of bread are _not_ used to
control expenditures. So there will be no "upward sloping demand curve" for bread
not matter what the budget. (This shows up in my experimental demo of the Giffen
effect, too, by the way: those interested can see the demo at
http://home.earthlink.net/~rmarken/ControlDemo/Economics.html

This control model has very interesting implications regarding observed "demand
curves" (curves relating price to consumption). The model suggests that demand
curves are, to some extent, a side effect resulting from control of other
variables besides expenditures (such as calories). People do have to control
expenditures and keep them in budget; but they also are controlling many other
perceptions besides expenditures via consumption. So we would expect that the
demand curves for some of these products (such as bread in the case of the Giffen
example) will not appear to be as "responsive" (or to "respond" as expected) to
price variations as others (such as meat in the Giffen example). I think this is
relevant to the fact that there are observed differences in the "elasticity" of
demand for different products. In other words, the Giffen effect is just a special
case of an "odd looking" demand curve but such curves are expected when people are
controlling many perceptual variables by consuming many different goods and
services, where the same variables are often controlled by consuming several
different goods and services and where variations in consumption of many of these
goods and services are not the basis for controlling expenditure.

Best regards

Rick

···

--
Richard S. Marken, Ph.D.
The RAND Corporation
PO Box 2138
1700 Main Street
Santa Monica, CA 90407-2138
Tel: 310-393-0411 x7971
Fax: 310-451-7018
E-mail: rmarken@rand.org

[From Bill Williams UMKC 8 November 2002 3:00 PM CST]

Rick,

I would agree that there is lots remaining in the Giffen Effect to explore.
There is a Giffen "Curve" which predicts that as the slopes of the budget
and caloric curves become more similiar as a result of a price change the
rate of reaction to a price change will increase. ( Easier to see if you
plot this graphically.) And, there is a Giffen Surface which results as a
combination of Giffen curves with a change in income. The Giffen Surface
predicts the survival behavior of an organism under the Giffen effect.
Sometime I will use mathematica to plot the surface-- I find it difficult
to draw freehand.

As long as "we" are into funny, how about:

   Free at last!
   Free at last!
   Thank God Almighty, Free at last!

If you wish to continue this, I suggest a new thread heading-- "Giffen
Effect."

You've been speculating, or rather making assumptions about my attitude
toward you. They're not much if at all better than my speculations about
what Fred was concerned about.

best

  Bill Williams