Paradocs
John Raymond Tyler
Isn't there a contradiction between the concept of equilibrium and the fact
of
growth?
How do we reconcile time preference with ordinal utility and rational
expectations?
John Tyler
------------------------------------------------------------
Get your FREE web-based e-mail and newsgroup access at:
http://MailAndNews.com
Create a new mailbox, or access your existing IMAP4 or
POP3 mailbox from anywhere with just a web browser.
------------------------------------------------------------
ro...@telus.net
<JRT...@MailAndNews.com> wrote:
>Isn't there a contradiction between the concept of equilibrium and the fact
>of growth?
Sure. Just like the contradiction between the concept of "ecological
balance" (or protection of endangered species, for that matter) and
the fact of evolution. In both cases, the former is merely a
theoretical construct that sometimes aids understanding, but just as
often hinders it.
-- Roy L
John Raymond Tyler
I would say that the concept of equilibrum serves no useful purpose. By
definition, it is a state to which the system tends to return. That seems
to
describe a stationary state, but no one would ever say that the economy is
in
a stationary state. That is something quite different from general market
clearing, which seems to be the usual meaning of general equilibirum.
John T.
Mark Patrick Witte
John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>===== Original Message From ro...@telus.net =====
>>On Tue, 10 Jul 2001 12:34:45 -0400, John Raymond Tyler
>><JRT...@MailAndNews.com> wrote:
>>
>>>Isn't there a contradiction between the concept of equilibrium and the fact
>>>of growth?
No, there is a well established concept of equilibrium growth
models where capital/labor ratios and the like preserve over growth of
population and/or technology.
Robert Vienneau
(Mark Patrick Witte) wrote:
> In article <3B82...@MailAndNews.com>,
> John Raymond Tyler <JRT...@MailAndNews.com> wrote:
> >>===== Original Message From ro...@telus.net =====
> >>On Tue, 10 Jul 2001 12:34:45 -0400, John Raymond Tyler
> >><JRT...@MailAndNews.com> wrote:
> >>>Isn't there a contradiction between the concept of equilibrium and the
> >>>fact
> >>>of growth?
> No, there is a well established concept of equilibrium growth
> models where capital/labor ratios and the like preserve over growth of
> population and/or technology.
My game of Bukharin can help one explore quantity flows in two-sector
generalizations of such models. (It is well-established that
one-sector growth models are deeply questionable as useful and
empirically validated tools for help in providing insight into
actual economies. See two most recent issues of Journal of Post
Keynesian Economics.)
Perhaps Mr. Tyler's question can be stated more perspicaciously.
Some have distinguished between economic growth and development. Isn't
there a contradiction between the concept of equilibrium and economic
development?
Structural dynamics, which one can also explore in my game, go
some way in exploring this question.
--
Try http://csf.colorado.edu/pkt/pktauthors/Vienneau.Robert/Bukharin.html
r c .../Keynes.html
v s a Whether strength of body or of mind, or wisdom, or
i m p virtue, are found in proportion to the power or wealth
e a e of a man is a question fit perhaps to be discussed by
n e . slaves in the hearing of their masters, but highly
@ r c m unbecoming to reasonable and free men in search of
d o the truth. -- Rousseau
Mark Patrick Witte
Robert Vienneau <rv...@see.sig.com> wrote:
>In article <9ij6oj$dot$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>(Mark Patrick Witte) wrote:
>
>> >>>Isn't there a contradiction between the concept of equilibrium and the
>> >>>fact
>> >>>of growth?
>
>> No, there is a well established concept of equilibrium growth
>> models where capital/labor ratios and the like preserve over growth of
>> population and/or technology.
>
>My game of Bukharin can help one explore quantity flows in two-sector
>generalizations of such models. (It is well-established that
>one-sector growth models are deeply questionable as useful and
>empirically validated tools for help in providing insight into
>actual economies. See two most recent issues of Journal of Post
>Keynesian Economics.)
It seems that while this is "well established" to certain
autodictats who seem to have an emotional attachment to an obscure
journal, it remains a useful research tool to many empirically trained
professional economists who are familiar with the wide range of work that is
done in the field.
>Perhaps Mr. Tyler's question can be stated more perspicaciously.
>Some have distinguished between economic growth and development. Isn't
>there a contradiction between the concept of equilibrium and economic
>development?
>
>Structural dynamics, which one can also explore in my game, go
>some way in exploring this question.
Or if one is actually interested in this subject, one could read a
book like "Development Economics" by Debraj Ray or "Development, Geography,
and Economic Theory" by Paul Krugman for an excellent introduction.
Robert Vienneau
(Mark Patrick Witte) wrote:
> In article <rvien-C012CC....@news.dreamscape.com>,
> Robert Vienneau <rv...@see.sig.com> wrote:
> >(It is well-established that
> >one-sector growth models are deeply questionable as useful and
> >empirically validated tools for help in providing insight into
> >actual economies. See two most recent issues of Journal of Post
> >Keynesian Economics.)
>
> [ Emotional outburst undirected to cognitive values - deleted. ]
Here's an account of economists filling up journal articles with
discussion of Solovian growth theory without actually empirically
testing it. It's in a book apparently intended for a popular
audience:
"The response of economists to the Solow model in the first two
decades of its existence was a case of, as Sherlock Holmes put it,
the dog that did not bark. In natural sciences, confronted by a new
and elegant theory from a distinguished member of the profession,
the first instinct of many would be to try to test the theory
empirically. In other words, to confront the properties and
implications of the model with actual data from the economy, and
to see if the two were in broad conformity.
The response in economics was completely different. Except in the
most rudimentary sense, the model was not tested. Rather, it was
assumed to be a reasonable description of reality, and was used
to try to account for the sources of actual growth. A vast and
increasingly sophisticated elaaboration of Solow's original model
filled the pages of academic economic journals. But empirical
testing remained noticeable by its absence."
I don't think the next paragraph is about the lack of empirical
support for the Solow growth model. But I include it anyways.
"An esoteric debate did take place, led by economists at Cambridge
University, in which the Solow model was attacked, not on
empirical grounds, but on a mixture of ideological and
theoretical points. The exact details of this argument, to which
I was subjected at inordinate length as a student at Cambridge
in the early 1970s, need not concern us here. In comparison,
the medieval theological wrangles about how many angels could
stand on the end of a pin appear positively illuminating. It was,
however, held that the Solow model attempted to provide an
intellectual justification for the existence of profits. Some of
the finest minds in economics diverted their energies into
trying to show that profits could not be justified in this way.
One figure above all was revered. Piero Sraffa, an old
Stalinist Fellow of Trinity College, had published virtually
nothing for over thirty years, until his masterpiece appeared,
the enigmatically entitled _Production of Commodities by Means
of Commodities_. In one of Evelyn Waugh's novels, the hero is
plagued by voices, which he imagines are being broadcast by
fellow passengers on his ship. In a futile effort to distract
them, he takes from the ship's library and reads aloud the most
boring book he can imagine, Charles Kingsley's _Westward Ho!_
Had Sraffa's book been available, he would certainly have
succeeeded in his task."
-- Paul Ormerod (1998)
There's no accounting for taste.
> >Perhaps Mr. Tyler's question can be stated more perspicaciously.
> >Some have distinguished between economic growth and development. Isn't
> >there a contradiction between the concept of equilibrium and economic
> >development?
> >
> >Structural dynamics, which one can also explore in my game, go
> >some way in exploring this question.
> Or if one is actually interested in this subject, one could read a
> book like "Development Economics" by Debraj Ray or "Development,
> Geography,
> and Economic Theory" by Paul Krugman for an excellent introduction.
I like Luigi Pasinetti on structural economic dynamics.
Mark Patrick Witte
Robert Vienneau <rv...@see.sig.com> wrote:
>In article <9il43l$m63$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>(Mark Patrick Witte) wrote:
>
>> In article <rvien-C012CC....@news.dreamscape.com>,
>> Robert Vienneau <rv...@see.sig.com> wrote:
>
>> >(It is well-established that
>> >one-sector growth models are deeply questionable as useful and
>> >empirically validated tools for help in providing insight into
>> >actual economies. See two most recent issues of Journal of Post
>> >Keynesian Economics.)
>>
>> [ Emotional outburst undirected to cognitive values - deleted. ]
Ah, it is a good feeling to so kick someone's ass that he feels the
need to completely delete the remarkes. My "emotional outburst" was just
that readers should be skeptical about accepting the views of someone with
little training in economics who holds a facination for a minor journal.
Oh dear, it seems I've gotten all emotional again; it must be hormones!
>Here's an account of economists filling up journal articles with
>discussion of Solovian growth theory without actually empirically
>testing it. It's in a book apparently intended for a popular
>audience:
Ah yes, such books are the height of scholarship, and this one is
classic in the level of uninformed claims it makes. Consider below.
> "The response of economists to the Solow model in the first two
> decades of its existence was a case of, as Sherlock Holmes put it,
> the dog that did not bark. In natural sciences, confronted by a new
> and elegant theory from a distinguished member of the profession,
> the first instinct of many would be to try to test the theory
> empirically. In other words, to confront the properties and
> implications of the model with actual data from the economy, and
> to see if the two were in broad conformity.
Which, since Solow's paper was empirical...as where so many of those
that followed up upon it...was essentially what was done. It could be that
Olmerod is complaining that the Solow approach in some sense is a sort of
accounting identity that breaks growth of output into contributions from
growth of labor, growth of capital, and growth of "productivity", with the
essential problem with testing identities. However, structures built upon
the Solow context generated testable hypotheses with lead to a huge
empirical literature. With the extension of Solow's work by Brock and
Mirman, and the empirical implementation of this by Kydland and Prescott,
the RBC literature was launched.
> The response in economics was completely different. Except in the
> most rudimentary sense, the model was not tested. Rather, it was
> assumed to be a reasonable description of reality, and was used
> to try to account for the sources of actual growth. A vast and
> increasingly sophisticated elaaboration of Solow's original model
> filled the pages of academic economic journals. But empirical
> testing remained noticeable by its absence."
Again, a simply amazing claim.
>I don't think the next paragraph is about the lack of empirical
>support for the Solow growth model. But I include it anyways.
>
> "An esoteric debate did take place, led by economists at Cambridge
> University, in which the Solow model was attacked, not on
> empirical grounds, but on a mixture of ideological and
So the Cambridge folks didn't challenge the empirical content of
Solow's work, but instead it's "ideology?" Ah, now that's good science!
> theoretical points. The exact details of this argument, to which
> I was subjected at inordinate length as a student at Cambridge
> in the early 1970s, need not concern us here. In comparison,
> the medieval theological wrangles about how many angels could
> stand on the end of a pin appear positively illuminating. It was,
> however, held that the Solow model attempted to provide an
> intellectual justification for the existence of profits. Some of
> the finest minds in economics diverted their energies into
> trying to show that profits could not be justified in this way.
Huh? Solow's work attempts to justify profits? For tractability,
he assumes perfectly competitive firms facing competitive factor markets
where labor and capital earn their marginal products. As such, profit
really doesn't come into it and it is quite a few years before someone is
able to extend a Solow setting to include non-zero profits (Hall; Rotemberg
and Woodford?).
susupply
"Mark Patrick Witte" <mwi...@merle.acns.nwu.edu> wrote in message
news:9ir797$lom$1...@news.acns.nwu.edu...
> >> [ Emotional outburst undirected to cognitive values - deleted. ]
>
> Ah, it is a good feeling to so kick someone's ass that he feels the
> need to completely delete the remarkes.
Ah, c'mon. He didn't claim a talent for provoking you to rant yet. And
when he feels the need to recycle several years of ad hominems directed at
you that you KNOW your boot has connected like Barry Bonds bat on a batting
practice fastball.
Patrick
susupply
"Robert Vienneau" <rv...@see.sig.com> wrote in message
news:rvien-68CAAA....@news.dreamscape.com...
> In one of Evelyn Waugh's novels, the hero is
> plagued by voices, which he imagines are being broadcast by
> fellow passengers on his ship. In a futile effort to distract
> them, he takes from the ship's library and reads aloud the most
> boring book he can imagine, Charles Kingsley's _Westward Ho!_
> Had Sraffa's book been available, he would certainly have
> succeeeded in his task."
> -- Paul Ormerod (1998)
Sounds like Vienneau on sci.econ to me.
Patrick
David Lloyd-Jones
"John Raymond Tyler" <JRT...@MailAndNews.com> asks:
>
> Isn't there a contradiction between the concept of equilibrium and the
fact
> of growth?
There might be a contradiction if equilibrium were some sort of fixed point,
right there in an assigned position in some sort of Newtonian mechanical
universe. This is not, however, how the word equilibrium is used in
economics. It is, as you say above, a concept. Think dynamic equilibrium --
like when you ride a bicycle. Then you'd be a good bit closer to what it
means in economics.
> How do we reconcile time preference with ordinal utility and rational
> expectations?
Hunh? To need reconciliation there has to be some sort of problem. What is
it you think is inconsistent here?
-dlj.
Robert Vienneau
(Mark Patrick Witte) wrote:
> In article <rvien-68CAAA....@news.dreamscape.com>,
> Robert Vienneau <rv...@see.sig.com> wrote:
> >In article <9il43l$m63$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
> >(Mark Patrick Witte) wrote:
> >> In article <rvien-C012CC....@news.dreamscape.com>,
> >> Robert Vienneau <rv...@see.sig.com> wrote:
> >> >(It is well-established that
> >> >one-sector growth models are deeply questionable as useful and
> >> >empirically validated tools for help in providing insight into
> >> >actual economies. See two most recent issues of Journal of Post
> >> >Keynesian Economics.)
> >>
> >> [ Emotional outburst undirected to cognitive values - deleted. ]
> [ Whining about deletions of portions of posts directed to ]
> [ non-cognitive values - deleted. ]
Mayhaps Mr. Witte can provide a post that does not suggest he's in
Oakland - a place well known for being someplace where's no there
there.
> >Here's an account of economists filling up journal articles with
> >discussion of Solovian growth theory without actually empirically
> >testing it. It's in a book apparently intended for a popular
> >audience:
> Ah yes, such books are the height of scholarship, and this one is
> classic in the level of uninformed claims it makes. Consider below.
Mr. Witte doesn't like "obscure" scholarship for some obscure reason -
probably not a reason that would withstand critical scruntity in
an academic culture without perverse norms. So I thought that I
would pick a source directed towards a popular audience.
I do find Ormerod's book amazing in one regard. He argues that typical
RBC business cycle models fail basic empirical tests - their spectrums
and their correlation functions don't match the data. He also produces
a business cycle model of his own that he says works better. This
model has a disaggregated accelerator effect combined with some
chaos math.
All well and good. But he cannot go into detail about these tests
because he's assuming an audience that doesn't necessarily fully
understand Fourier analysis. Furthermore, I cannot see any sign
that he submitted a write-up of his model to peer-reviewed journals;
this is its first presentation.
He says neoclassical economics is not wrong. It just needs to be
generalized. What does it say for his assessment of the culture of
the profession, though, that this is how he first chooses to present
his model?
> > "The response of economists to the Solow model in the first two
> > decades of its existence was a case of, as Sherlock Holmes put it,
> > the dog that did not bark. In natural sciences, confronted by a new
> > and elegant theory from a distinguished member of the profession,
> > the first instinct of many would be to try to test the theory
> > empirically. In other words, to confront the properties and
> > implications of the model with actual data from the economy, and
> > to see if the two were in broad conformity.
> Which, since Solow's paper was empirical...as where so many of those
> that followed up upon it...was essentially what was done.
Nope. Lots of this literature was not CONFRONTING the model with
data, but using the model to interpret data, assuming the correctness
of the model. As Mr. Witte realizes:
> It could be that
> Olmerod is complaining that the Solow approach in some sense is a sort of
> accounting identity that breaks growth of output into contributions from
> growth of labor, growth of capital, and growth of "productivity", with
> the
> essential problem with testing identities. However, structures built
> upon
> the Solow context generated testable hypotheses with lead to a huge
> empirical literature. With the extension of Solow's work by Brock and
> Mirman, and the empirical implementation of this by Kydland and Prescott,
> the RBC literature was launched.
Consider Fisher and Shaikh's results. Was the Solow model tested? Or
were many supposedly empirical results merely a test of the stability
of income distribution? According to an article in the Journal of Post
Keynesian Economics of two issues ago, Solow was incorrect in his
response to Shaikh where he said if the data were like so, "we would
not now be having this discussion." Apparently the data was more or
less like that.
> > The response in economics was completely different. Except in the
> > most rudimentary sense, the model was not tested. Rather, it was
> > assumed to be a reasonable description of reality, and was used
> > to try to account for the sources of actual growth. A vast and
> > increasingly sophisticated elaaboration of Solow's original model
> > filled the pages of academic economic journals. But empirical
> > testing remained noticeable by its absence."
> Again, a simply amazing claim.
Not to those who understand it.
> >I don't think the next paragraph is about the lack of empirical
> >support for the Solow growth model. But I include it anyways.
> > "An esoteric debate did take place, led by economists at Cambridge
> > University, in which the Solow model was attacked, not on
> > empirical grounds, but on a mixture of ideological and
[>> theoretical points. ]
> So the Cambridge folks didn't challenge the empirical content of
> Solow's work, but instead it's "ideology?" Ah, now that's good science!
Notice Mr. Witte's truncation to impose a false dichotomy. Ah,
what transparent rhetoric!
I did not endorse Ormerod's opinion. I thought if I did not include
this quote, somebody that might actually look up its source would
criticize me for twisting Ormerod's words to support my previously
expressed opinions.
Good scholars tend to note where other scholars have disagreed with
them. Sort of like Mr. Witte on the communities of scholars who think
Solovian growth theory lacks empirical and theoretical support. Oh,
wait...
> > theoretical points. The exact details of this argument, to which
> > I was subjected at inordinate length as a student at Cambridge
> > in the early 1970s, need not concern us here. In comparison,
> > the medieval theological wrangles about how many angels could
> > stand on the end of a pin appear positively illuminating. It was,
> > however, held that the Solow model attempted to provide an
> > intellectual justification for the existence of profits. Some of
> > the finest minds in economics diverted their energies into
> > trying to show that profits could not be justified in this way.
> Huh? Solow's work attempts to justify profits?
More reading comprehension problems from Mr. Witte. Ormerod suggests
that it is a misunderstanding of Solow's work to take it as an attempt
to justify profits. (J. B. Clark's work, on the other hand...)
> For tractability,
> he assumes perfectly competitive firms facing competitive factor markets
> where labor and capital earn their marginal products.
Strangely enough, "capital" does not earn its marginal product in
equilibrium in many multigood models with perfectly competitive firms
facing competitive factor markets. The same even holds in models
with many identical firms producing (net) one identical commodity.
This inequality results from price Wicksell effects. But Mr. Witte,
of course, is free to reject good economic theory.
> As such, profit
> really doesn't come into it and it is quite a few years before someone is
> able to extend a Solow setting to include non-zero profits (Hall;
> Rotemberg and Woodford?).
More reading comprehension problems from Mr. Witte and lack of
acquaintance with vast tracts of the literature. The above was not
about pure economic profits. Bohm-Bawerk started a trend among
neoclassical economists in calling "interest" what many others call
"profits" - returns to capital, including equilibrium returns. If
one cannot put this convention aside, one is unlikely to understand
what is being said in lots of literature.
Let's see what an author of a good survey text of capital theory
has to say:
"The issue was settled in favour of Cambridge University when
Samuelson wrote (1976) that wherever 'informed economic theory
is taught', the 'paradoxes' are accepted, and their consequences
for the concept of capital known. It is another matter that, on
the basis of this criterion, many seats of learning in North
America, as perhaps also elsewhere, do not teach informed
economic theory. For instance, the 'new classicals', claiming
otherwise to be meticulous theorists, keep on blithely using
the Clarkian concept of capital in their production functions.
The case for the implicit excuse that the absence of any easy
alternative justifies such high-handedness has not so far
been made in the literature."
-- Syed Ahmad (1998)
(One might think a very recent contribution by Burmeister makes
some such claim in the literature.)
Mr. Witte, of course, is free to continue to ignore vast chunks of
the literature and the implications of rigorous theory.
Mark Patrick Witte
Robert Vienneau <rv...@see.sig.com> wrote:
>In article <9ir797$lom$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>(Mark Patrick Witte) wrote:
>
>> In article <rvien-68CAAA....@news.dreamscape.com>,
>> Robert Vienneau <rv...@see.sig.com> wrote:
>
>> >In article <9il43l$m63$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>> >(Mark Patrick Witte) wrote:
>
>> >> In article <rvien-C012CC....@news.dreamscape.com>,
>> >> Robert Vienneau <rv...@see.sig.com> wrote:
>
>> >> >(It is well-established that
>> >> >one-sector growth models are deeply questionable as useful and
>> >> >empirically validated tools for help in providing insight into
>> >> >actual economies. See two most recent issues of Journal of Post
>> >> >Keynesian Economics.)
>> >>
>> >> [ Emotional outburst undirected to cognitive values - deleted. ]
>
>> [ Whining about deletions of portions of posts directed to ]
>> [ non-cognitive values - deleted. ]
>
>Mayhaps Mr. Witte can provide a post that does not suggest he's in
>Oakland - a place well known for being someplace where's no there
>there.
Wow, two complete deletes in two posts. I'll have to remember to
wash my kicking foot extra carefully tonight. (And what's this "Mr." crap?
I went to evil econ school for years to lose that one!)
>> >Here's an account of economists filling up journal articles with
>> >discussion of Solovian growth theory without actually empirically
>> >testing it. It's in a book apparently intended for a popular
>> >audience:
>
>> Ah yes, such books are the height of scholarship, and this one is
>> classic in the level of uninformed claims it makes. Consider below.
>
>Mr. Witte doesn't like "obscure" scholarship for some obscure reason -
>probably not a reason that would withstand critical scruntity in
It is unfortunate that I my ass-kicking is so painful to some
poster that he feels the need to delete my remarks yet reply to them. My
point was a response to a broad accusation against a large literature by
someone enraptured by a very minor journal.
>an academic culture without perverse norms. So I thought that I
>would pick a source directed towards a popular audience.
A refereed source.
>I do find Ormerod's book amazing in one regard. He argues that typical
>RBC business cycle models fail basic empirical tests - their spectrums
>and their correlation functions don't match the data.
RBCs do poorly in spectral and correlations relative to what
alternative published in a refereed journal?
>He also produces
>a business cycle model of his own that he says works better. This
>model has a disaggregated accelerator effect combined with some
>chaos math.
Which survived the scrutiny of the book editor at Wiley & Sons?
Fine scholars they!
>All well and good. But he cannot go into detail about these tests
>because he's assuming an audience that doesn't necessarily fully
>understand Fourier analysis. Furthermore, I cannot see any sign
>that he submitted a write-up of his model to peer-reviewed journals;
>this is its first presentation.
Nor has such appeared in the four years since the book was written.
>He says neoclassical economics is not wrong. It just needs to be
>generalized. What does it say for his assessment of the culture of
>the profession, though, that this is how he first chooses to present
>his model?
That he wants to sell books to unsophisticated readers?
>> > "The response of economists to the Solow model in the first two
>> > decades of its existence was a case of, as Sherlock Holmes put it,
>> > the dog that did not bark. In natural sciences, confronted by a new
>> > and elegant theory from a distinguished member of the profession,
>> > the first instinct of many would be to try to test the theory
>> > empirically. In other words, to confront the properties and
>> > implications of the model with actual data from the economy, and
>> > to see if the two were in broad conformity.
>
>> Which, since Solow's paper was empirical...as where so many of those
>> that followed up upon it...was essentially what was done.
>
>Nope. Lots of this literature was not CONFRONTING the model with
>data, but using the model to interpret data, assuming the correctness
>of the model. As Mr. Witte realizes:
Some people evidently think "Lots" equals "all." But even using the
model to interpret data implies tests, as with the work it inspired by Romer
et al.
>> It could be that
>> Olmerod is complaining that the Solow approach in some sense is a sort of
>> accounting identity that breaks growth of output into contributions from
>> growth of labor, growth of capital, and growth of "productivity", with
>> the
>> essential problem with testing identities. However, structures built
>> upon
>> the Solow context generated testable hypotheses with lead to a huge
>> empirical literature. With the extension of Solow's work by Brock and
>> Mirman, and the empirical implementation of this by Kydland and Prescott,
>> the RBC literature was launched.
>
>Consider Fisher and Shaikh's results. Was the Solow model tested? Or
>were many supposedly empirical results merely a test of the stability
>of income distribution? According to an article in the Journal of Post
>Keynesian Economics of two issues ago, Solow was incorrect in his
>response to Shaikh where he said if the data were like so, "we would
>not now be having this discussion." Apparently the data was more or
>less like that.
"Apparently the data was more or less like that." What's the point
in responding so such nonsense?
>> > The response in economics was completely different. Except in the
>> > most rudimentary sense, the model was not tested. Rather, it was
>> > assumed to be a reasonable description of reality, and was used
>> > to try to account for the sources of actual growth. A vast and
>> > increasingly sophisticated elaaboration of Solow's original model
>> > filled the pages of academic economic journals. But empirical
>> > testing remained noticeable by its absence."
>
>> Again, a simply amazing claim.
>
>Not to those who understand it.
None of whom evidently have ever taken a class in graduate
macroeconomics. Or who have read Denison (1967, 1985) or Bailey and Gordon
(1988) or a host of others.
>> >I don't think the next paragraph is about the lack of empirical
>> >support for the Solow growth model. But I include it anyways.
>
>> > "An esoteric debate did take place, led by economists at Cambridge
>> > University, in which the Solow model was attacked, not on
>> > empirical grounds, but on a mixture of ideological and
>[>> theoretical points. ]
>
>> So the Cambridge folks didn't challenge the empirical content of
>> Solow's work, but instead it's "ideology?" Ah, now that's good science!
>
>Notice Mr. Witte's truncation to impose a false dichotomy. Ah,
>what transparent rhetoric!
This is pure idiocy. Critiquing Solow on theoretical points is good
science, critiquing his "ideological" points is not. Duh.
>I did not endorse Ormerod's opinion. I thought if I did not include
>this quote, somebody that might actually look up its source would
>criticize me for twisting Ormerod's words to support my previously
>expressed opinions.
>
>Good scholars tend to note where other scholars have disagreed with
>them. Sort of like Mr. Witte on the communities of scholars who think
>Solovian growth theory lacks empirical and theoretical support. Oh,
>wait...
One can disagree with the quality of the work that has been done;
one cannot disagee that considerable empirical testing of Solow's growth
model has taken place.
>> > theoretical points. The exact details of this argument, to which
>> > I was subjected at inordinate length as a student at Cambridge
>> > in the early 1970s, need not concern us here. In comparison,
>> > the medieval theological wrangles about how many angels could
>> > stand on the end of a pin appear positively illuminating. It was,
>> > however, held that the Solow model attempted to provide an
>> > intellectual justification for the existence of profits. Some of
>> > the finest minds in economics diverted their energies into
>> > trying to show that profits could not be justified in this way.
>
>> Huh? Solow's work attempts to justify profits?
>
>More reading comprehension problems from Mr. Witte. Ormerod suggests
>that it is a misunderstanding of Solow's work to take it as an attempt
>to justify profits. (J. B. Clark's work, on the other hand...)
At no point to ascribe this idiocy to Ormerod. Reading
comprehension indeed.
>Mr. Witte, of course, is free to continue to ignore vast chunks of
>the literature and the implications of rigorous theory.
Pot...kettle....
Robert Vienneau
(Mark Patrick Witte) wrote:
> In article <rvien-7B5250....@news.dreamscape.com>,
> Robert Vienneau <rv...@see.sig.com> wrote:
> >In article <9ir797$lom$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
> >(Mark Patrick Witte) wrote:
> >> In article <rvien-68CAAA....@news.dreamscape.com>,
> >> Robert Vienneau <rv...@see.sig.com> wrote:
> >> >In article <9il43l$m63$1...@news.acns.nwu.edu>,
> >> >mwi...@merle.acns.nwu.edu
> >> >(Mark Patrick Witte) wrote:
> >> >> In article <rvien-C012CC....@news.dreamscape.com>,
> >> >> Robert Vienneau <rv...@see.sig.com> wrote:
> >> >> >(It is well-established that
> >> >> >one-sector growth models are deeply questionable as useful and
> >> >> >empirically validated tools for help in providing insight into
> >> >> >actual economies. See two most recent issues of Journal of Post
> >> >> >Keynesian Economics.)
> >> >> [ Emotional outburst undirected to cognitive values - deleted. ]
> >> [ Whining about deletions of portions of posts directed to ]
> >> [ non-cognitive values - deleted. ]
> >Mayhaps Mr. Witte can provide a post that does not suggest he's in
> >Oakland - a place well known for being someplace where's no there
> >there.
> [ More posting from Oakland - deleted ]
It remains the case that it is well-known among economists that
some have raised good reasons to think Solovian growth theory is
without good theoretical and empirical foundation.
"It is important, for the record, to recognize that key
participants in the debate openly admitted their mistakes.
Samuelson's seventh edition of _Economics_ was purged of
errors. Levhari and Samuelson published a paper which
began, 'We wish to make it clear for the record that the
nonswitching theorem associated with us is definitely
false. We are grateful to Dr. Pasinetti...' Leland Yeager
and I jointly published a not acknowledging his
earlier error and attempting to resolve the conflict
between our theoretical perspectives.
However, the damage had been done, and Cambridge, UK,
'declared victory': Levhari was wrong, Samuelson was wrong,
Solow was wrong, MIT was wrong and therefore neoclassic
economics was wrong. As a result there are some groups of
economists who have abandoned neoclassical economics for
their own refinements of classical economics. In the
United States, on the other hand, mainstream economics
goes on as if the controversy had never occurred.
Macroeconomics textbooks discuss 'capital' as if it were
a well-defined concept - which it is not, except in a very
special one-capital-good world (or under other
unrealistically restrictive conditions). The problems of
heterogeneous capital goods have been ignored in the
'rational expectations revolution' and in virtually all
econometric work"
-- Edwin Burmeister (2000)
John Raymond Tyler
>"John Raymond Tyler" <JRT...@MailAndNews.com> asks:
>>
>> Isn't there a contradiction between the concept of equilibrium and the
>fact
>> of growth?
>
>There might be a contradiction if equilibrium were some sort of fixed point,
>right there in an assigned position in some sort of Newtonian mechanical
>universe. This is not, however, how the word equilibrium is used in
>economics. It is, as you say above, a concept. Think dynamic equilibrium --
>like when you ride a bicycle. Then you'd be a good bit closer to what it
>means in economics.
>
I am afraid that dynamic equilibrium is a meaningless concept, although it
is
widely used in economics. We might speak of an optimal path, but that is
something else.
>> How do we reconcile time preference with ordinal utility and rational
>> expectations?
>
>Hunh? To need reconciliation there has to be some sort of problem. What is
>it you think is inconsistent here?
>
> -dlj.
>
Time preference requires that we compare utilities in different time
periods.
This clearly requires a cardinal measure of utility. And if our
expectations
were rational, we should we have time preference; it is not rational to
prefer
present over future utility.
John Tyler
John Raymond Tyler
=====
>In article <3B82...@MailAndNews.com>,
>John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>>===== Original Message From ro...@telus.net =====
>>>On Tue, 10 Jul 2001 12:34:45 -0400, John Raymond Tyler
>>><JRT...@MailAndNews.com> wrote:
>>>
>>>of growth?
>
>models where capital/labor ratios and the like preserve over growth of
>population and/or technology.
>
>
original position,which is what equilibrium means. In equilibrium, there is
no tendency to leave the original position, the entire position, not just
some
ratio.
It has been some time since I read Solow, which cropped up in this
discussion.
As I recall, he makes two elementary mistakes in calculus in deriving the
equation to be tested. It is true that they tend to cancel each other, but
his work is fatally flawed.
John Tyler----------------------------------
ro...@telus.net
<JRT...@MailAndNews.com> wrote:
>>===== Original Message From "David Lloyd-Jones" <dav...@sympatico.ca> =====
>
>I am afraid that dynamic equilibrium is a meaningless concept, although it
>is
>widely used in economics. We might speak of an optimal path, but that is
>something else.
Imagine that the economy is a marble slowly rolling around on an
enormous plate. It is always close to equilibrium, and getting
closer. Now imagine that the plate is borne on the backs of four
colossal elephants called Technology, Popular Taste, Government Policy
and Business Sentiment. Economic equilibrium is like that.
>>> How do we reconcile time preference with ordinal utility and rational
>>> expectations?
>>
>>Hunh? To need reconciliation there has to be some sort of problem. What is
>>it you think is inconsistent here?
>>
>Time preference requires that we compare utilities in different time
>periods.
>This clearly requires a cardinal measure of utility. And if our
>expectations
>were rational, we should we have time preference; it is not rational to
>prefer
>present over future utility.
Sure it is. The longer the delay, the greater the risk that you won't
get to enjoy the future utility.
"The pig might die. The king might die. I might die. And maybe, the
pig will learn to sing."
-- Roy L
Mark Patrick Witte
John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>"John Raymond Tyler" <JRT...@MailAndNews.com> asks:
>>>
>>> Isn't there a contradiction between the concept of equilibrium and the
>>fact
>>> of growth?
>>
>>There might be a contradiction if equilibrium were some sort of fixed point,
>>right there in an assigned position in some sort of Newtonian mechanical
>>universe. This is not, however, how the word equilibrium is used in
>>economics. It is, as you say above, a concept. Think dynamic equilibrium --
>>like when you ride a bicycle. Then you'd be a good bit closer to what it
>>means in economics.
>>
>
>I am afraid that dynamic equilibrium is a meaningless concept, although it
>is
>widely used in economics. We might speak of an optimal path, but that is
>something else.
No, dynamic equilibria need not be optimal, as in the capital-over
accumulation case. However, a dynamic equilibrium would just be one where
ratios are at their long run values, or which can be normalized to a
stationary state.
>
>
>>> How do we reconcile time preference with ordinal utility and rational
>>> expectations?
>>
>>Hunh? To need reconciliation there has to be some sort of problem. What is
>>it you think is inconsistent here?
>>
>> -dlj.
>>
>Time preference requires that we compare utilities in different time
>periods.
>This clearly requires a cardinal measure of utility. And if our
>expectations
>were rational, we should we have time preference; it is not rational to
>prefer
>present over future utility.
So...you'll lend me $5 today in exchange for $5 a year from now?
Mark Patrick Witte
As so often happens in these sorts of things, we find
ourselves in a weird situation. John Tyler began this by
asking whether there was "a contradiction between the
concept of equilibrium and the fact of growth." It was
explained to him that equilibrium steady-state growth
models displayed both growth and what economists
would call equilibrium, in the sense that there is no internal
force to change various ratios, that the system can be
normalized to a steady state.
Then a poster injects an irrelevancy that "It is well-
established that one-sector growth models are deeply
questionable as useful and empirically validated tools for
help in providing insight into actual economies. See two
most recent issues of Journal of Post Keynesian
economics." When it is pointed out that "well-established"
is far more than an overstatement, this poster replies with a
quote from a popular press book by Paul Ormerod that
claims that the Solow growth model was not empirically
tested for over twenty years. Ormerod claims that at least
at Cambridge, the model was not attacked empirically but
rather on the basis of theoretical and ideological points.
The poster also supplies a quote from Edwin
Burmeister that the poster claims criticizes the empirical
foundations of Solow's work. In fact, as is usual from this
poster on matters relating to data and econometrics, the
Burmeister quote makes no empirical comment whatsoever
and instead essentially complains about assumptions of the
models.
The poster also enquires on several occasions about
whether I am in Oakland or not. This indicates a strange
psychology and evidently a deep attraction to the clichis of
Gertrude Stein. More on this point below.
Invective aside, was Solow's growth theory work
empirically tested? The answer, as I've stated repeatedly, is
yes. Beyond work by Solow himself, there was certainly
work by Denison, Madisson, Gordon, DeLong,
Lucas, and many others including a great debate lately
between Chang-Tai Hsieh and Alwyn Young.
But Ormerod's claim should raise a skeptical flag
itself. Why wouldn't important work be empirically tested?
(Was it some conspiracy among all the participants in the
field? Perhaps a big meeting was held where it was
decreed, "Lay off Solow!") The answers are many, but two
good ones present themselves. One is a lack of data at a
given time that allows separating hypotheses, and this did
pose some limits on testing the usefulness of Solow's
framework. A second is the lack of ways to construct
tests, and this is often a problem. How does one test a
residual? However, Denison's work on the US was a first
try at what Gordon calls "top down versus bottom up"
comparison's of answers. Did Solow's model give crazy
results, as compared with competing methodologies? The
answer, generally, was yes. Still, this goes to the notion
of scientific progress. Solow's contribution was not that
he correctly measured total factor productivity growth, but
rather that he built a simple framework that gave limiting
estimates of what it might be, estimates that could be
compared with other methodologies (Denison, Hsieh, etc.)
and a framework that could be built upon to capture
important aspects of the growth process (Barro, Hall, Romer,
etc.). As such, all scientific contributions are
valuable if they provide a fertile building place for more
work, and Solow's certainly did. However, I have been too
unkind to Solow here. His work does make predictions
about convergence that have been borne out well in papers
Madisson and Christiano concerning post-war capital stock
accumulation. Furthermore, with the addition of
uncertainty through Brock and Mirman led to the hugely
empirical RBC literature with its host of competing
transition and propagation mechanisms that generate
testable temporal and intertemporal moments and
correlations.
In a Lakatosian sense, there are limits to how Solow's
work can be tested, but it certainly generates a testable
empirical literature. In this way, it is wonderfully useful
unlike the work of say...Sraffa, where the empirical
predictions and implementations are vanishingly few.
As to whether I am in Oakland, I would hope that my
e-mail name would give some idea of my affiliation
although for all although for part of this exchange, I've
actually been in Corsica. This made things tough because
the keyboards there are definitively non-QWERTY and
require some odd combinations like having to hit the shift
key to type a period. I don't know if this is a matter of path
dependence or just some strange characteristic of the
French. (If anyone following this thread has interests or
questions about QWERTY and path dependence, I know
that there is someone who follows and occasionally posts
on this group who has done research in this area and who
has recently been invited to present his work in a coming
forum. I'm sure he'd be glad to help if any interested party
who would post on this subject or e-mail him at
susu...@mindspring.com.)
In article <rvien-ABBFDC....@news.dreamscape.com>,
Mark Patrick Witte
John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>===== Original Message From mwi...@merle.acns.nwu.edu (Mark Patrick Witte)
>=====
>>In article <3B82...@MailAndNews.com>,
>>John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>>>===== Original Message From ro...@telus.net =====
>>>>On Tue, 10 Jul 2001 12:34:45 -0400, John Raymond Tyler
>>>><JRT...@MailAndNews.com> wrote:
>>>>
>>>>Isn't there a contradiction between the concept of equilibrium and the fact
>>>>of growth?
>>
>> No, there is a well established concept of equilibrium growth
>>models where capital/labor ratios and the like preserve over growth of
>>population and/or technology.
>>
>>
>Preserving capital/labor ratios is not the same thing as a return to the
>original position,which is what equilibrium means.
Where does this definition come from? What if the original ratios
have an overly low or high capital stock?
> In equilibrium, there is
>no tendency to leave the original position, the entire position, not just
>some ratio.
>
>It has been some time since I read Solow, which cropped up in this
>discussion.
> As I recall, he makes two elementary mistakes in calculus in deriving the
>equation to be tested. It is true that they tend to cancel each other, but
>his work is fatally flawed.
Yes, yes, and the editors at the QJE are drooling idiots, almost as
dumb as Frank Ramsey.
Robert Vienneau
(Mark Patrick Witte) wrote:
> Then a poster injects an irrelevancy that "It is well-
^^^^^^^^^^^
> established that one-sector growth models are deeply
> questionable as useful and empirically validated tools for
> help in providing insight into actual economies. See two
> most recent issues of Journal of Post Keynesian
> economics."
Mr. Witte misspells "parenthetical comment".
> When it is pointed out that "well-established"
> is far more than an overstatement, ...
Mr. Witte's lack of understanding why, say, there was a
conference organized in Rome a few years back asking whether
one could ethically make policy recommendations based on
one good models does not point out anything more than
his lack of understanding.
> The poster also supplies a quote from Edwin
> Burmeister that the poster claims criticizes the empirical
> foundations of Solow's work.
As usual, Mr. Witte's comments are unrelated to the text
he is pretending to summarize.
> In fact, as is usual from this
> poster on matters relating to data and econometrics, the
> Burmeister quote makes no empirical comment whatsoever
"The problems of heterogeneous capital goods have been
ignored in the 'rational expectations revolution' and in
virtually all econometric work"
> and instead essentially complains about assumptions of the
> models.
Mr. Witte seems not to understand the theory that Burmeister
mentions. Anybody who cares could always look at, for instance,
Neri Salvadori and Ian Steedman, "No Reswitching? No
Switching!" _Cambridge Journal of Economics_, 12 (1988):
481-486. Reprinted in _Understanding 'Classical'
Economics: Studies in Long-period Theory_ (edited by
H. D. Kurz and N. Salvadori), Routledge, 1998.
"There is no consistent long-period neoclassical theory
other than in exceptionally special cases that are of
no economic interest."
-- Kurz and Salvadori, (1998:15)
"Suppose that a number of economies, satisfying the growth,
savings, and technology assumptions made above, but having
various tax rates, were observed by a naive, and therefore
imaginary, econometrician. The data collected would all
satisfy (3) and thus a completely spurious 'production
function' could be observed. The value of capital per man
and the output-capital ratio would be observed to be quite
independent of the net rate of profit, while their relation
to the gross rate of profit would provide most intricate
and complex curves. If [the rate of growth of the labor
force] and [the fraction of disposable income saved by
capitalists] were to differ between the economies, the
confusion would be worse confounded and the single technique
technology might even generate observations claimed by
our naive econometrician to show the effects of changes in
technique. But this is just fantasy, since no real-world
econometrician would allow mere price-effects to influence
his cross section estimates of real technical relationships."
-- Ian Steedman (1973)
Does Mr. Witte summarize econometric work subject to this
criticism? One cannot tell from his comments, for Mr. Witte
does not seem to understand price Wicksell effects.
I don't think the widespread use of Sraffa's theory in empirical
work in the form of Leontief Input-Output analysis is on
point to this thread. T. K. Rymes' work (which I don't know
much about), on the other hand...
> [ More stuff, that judging by content, is posted from ]
> [ Oakland - deleted. ]
Robert Vienneau
(Mark Patrick Witte) wrote:
> In article <3B59...@MailAndNews.com>,
> John Raymond Tyler <JRT...@MailAndNews.com> wrote:
> >>===== Original Message From mwi...@merle.acns.nwu.edu (Mark Patrick
> >>Witte)
> >=====
> >>In article <3B82...@MailAndNews.com>,
> >>John Raymond Tyler <JRT...@MailAndNews.com> wrote:
> >It has been some time since I read Solow, which cropped up in this
> >discussion...
> >his work is fatally flawed.
> Yes, yes, and the editors at the QJE are drooling idiots, almost as
> dumb as Frank Ramsey.
"Those who have followed the literature of the past 20 years know
that...many properties of the Clarkian paradigm do NOT all hold
generally. In every Cambridge, and wherever informed economic
theory is taught, there is no...disagreement on this."
-- Paul A. Samuelson, (QJE 1976)
I realize Samuelson grounds are different than Mr. Tyler's.
"We have a macroeconomics squarely based on perfect foresight,
infinite time optimization, and universal perfect competition.
What Ramsey took to be a normative model, useful for working
out what an idealized omniscient planner should do, has been
transformed into a model for interpreting last year's and
next year's national accounts.
Of course that is the economics of Dr. Pangloss, and it bears
little relation to the world. In a decade that has seen vast
progress in our study of asymmetric information, 'missing
markets,' contracts, strategic interaction, and much else
precisely because those aspects are regarded as real phenomena
that require analysis, macroeconomics has ignored them all.
The consequence is this: no account has been given of how
and why a decentralized economy could behave as if guided by
a Ramsey maximizer. It is true that an Arrow-Debreu equilibrium
is an allocation that maximizes a special social welfare
function, but that is not the case, for instance, when some
insurance markets are abscent, or indeed when any even mildly
realistic phenomena are included."
-- Frank Hahn and Robert Solow (1995)
Robert N. Newshutz
>
> On Mon, 16 Jul 2001 11:35:21 -0400, John Raymond Tyler
> <JRT...@MailAndNews.com> wrote:
>
> >>===== Original Message From "David Lloyd-Jones" <dav...@sympatico.ca> =====
> >
> >I am afraid that dynamic equilibrium is a meaningless concept, although it
> >is
> >widely used in economics. We might speak of an optimal path, but that is
> >something else.
>
> Imagine that the economy is a marble slowly rolling around on an
> enormous plate. It is always close to equilibrium, and getting
> closer. Now imagine that the plate is borne on the backs of four
> colossal elephants called Technology, Popular Taste, Government Policy
> and Business Sentiment. Economic equilibrium is like that.
>
What does the turtle the elephants stand on represent?
--
Robert N. Newshutz
Mark Patrick Witte
growth and equilibrium are compatible has, as is often the
case on newsgroups, become something else, generally due
to misunderstandings of basic issues of research
methodology. As has been explained so some posters
repeatedly, even posters who have claimed to have read
some work by Thomas Kuhn and believe themselves to
have understood aspects of it, the progress of understanding
advances by building models to try to explain phenomena,
testing these models against data, and then comparing the
empirical results with those of other models. The world
itself is too complicated to implement empirically (or even
theoretically) so all models involve making use of
simplifying assumptions. Whining that some assumptions
are "unrealistic" or "cases of no economic interest" is not
doing good work. Good work is implementing a model and
showing that by some metric it does a better job with the
data than do competing models.
In discussion pieces, researchers will commonly make
"wish lists" of what they personally find unsatisfying about
current research. This is not particularly useful work. The
other frequent poster to this thread has in the past expressed
preferences for a role of "market power" and "class
struggle" in economic models, just as some of the quoted
economists below would like to see explicit heterogeneous
capital goods and reswitching. There are a lot of things all
researchers would like to see put into models, but good
intentions are not the path to heaven. Instead, good work
means actually putting these features into implementable
models and showing that the data supports that this is an
improvement over alternatives; not whining that others
have not done so. This is why I have a lot of respect for
Maryanne Baxter, Patricia Reynolds, and Valarie Ramey
who have all done empirical work with heterogeneous
capital goods. When last I checked, there was vanishingly
small empirical support for the importance of reswitching.
Attempts to give Sraffa's some empirical value by
attempting to steal from the work of Wassily Leontief
are pathetic. (Further, it is the height of either hypocrisy or
ignorance for someone who complains about the difficulty
of testing Solow's framework to then trumpet I-O matrices
as good empirical work.) And alas, I can only wish that
Leontief's input-output work was more widespread these
days since that would make a project I'm working on much
easier.
So, in conclusion, is it "well-established that one-
sector growth models are deeply questionable as useful and
empirically validated tools for help in providing insight into
actual economies"? No, unless one means that by "well-
established" such models are only used in the widest
circulated journals and by macroeconomic policy makers in
at all levels of government and international organizations.
This is not to say that other approaches aren't used in some
cases as well, or that such models represent ultimate truth.
From experience, I'm sure while some folks had a great
time complaining about the limits of current theoretical and
empirical work during their junket in Rome, their time
would have likely been more productively (but less
pleasurably) spent actually building models that better
match the data than do popular competing approaches.
Given that policy must be made, what is "ethical" is using
the models that have the best empirical support. Given that
such models are popular enough to generate a circle of
whiners in Rome, they must be doing something right.
In article <rvien-E385CE....@news.dreamscape.com>,
Christopher Auld
>> >It has been some time since I read Solow, which cropped up in this
>> >discussion...
>> >his work is fatally flawed.
>
>> Yes, yes, and the editors at the QJE are drooling idiots, almost as
>> dumb as Frank Ramsey.
>
> "Those who have followed the literature of the past 20 years know
> that...many properties of the Clarkian paradigm do NOT all hold
> generally. In every Cambridge, and wherever informed economic
> theory is taught, there is no...disagreement on this."
> -- Paul A. Samuelson, (QJE 1976)
Rob's ability to interpret almost anything as ringing the death bell
for neoclassical economics is truly staggering. Recently, he surely
caused giggling the world over by claiming Deidre McCloskey and
Andreu Mas-Colell as intellectual fellow travellers, which would
surely come as shocking news to these researchers. Rob follows up here
by implying that Paul Samueulson thinks Bob Solow's work is "fatally
flawed." Which would surely come as shocking news to Bob Solow, who
may have read, for instance,
Samuelson, P. (1989) "Robert Solow: an affectionate portrait,"
JEP 3(3) 91-97.
Which includes, for instance, the comment:
"The Nobel authorities have properly singled out Solow's 1956 growth
model and his 1957 econometric measurements of that model."
Like almost everyone else, Samuelson recognizes that Solow's model is
a very important methodological contribution. Mark Witte has already
pointed out the vast literatures which stem from this these seminal
pieces. Attacking models following Solow on the grounds that aggregate
production functions don't exist is an attack on the realism of
assumptions, and is therefore only interesting if one can present
alternate analytical techniques which both evade the criticism and
perform better in some well-defined sense. Attacking aggregation
assumptions needed for tractability is a particularly dull and vacuous
line, as such assumptions are extremely common, and often problematic,
in all areas of scientific inquiry. In this context, this sort of
attack usually smacks of ideology rather than genuine attempts to
further undertanding, a notion Rob inadvertently confirms by quoting
Ormerod, of all people, mocking such ideological arguments from Rob's
"favorite" group of economists at Cambridge UK in the early 70s.
Incidentally, Samuelson published one paper in the QJE in 1976, and
it does not include Rob's quote above.
Since this thread has, of course, turned into yet another opportunity
for Rob to discuss the CCC (any excuse will do), the fact he dredges
up this quote is interesting:
> Of course that is the economics of Dr. Pangloss, and it bears
> little relation to the world. In a decade that has seen vast
> progress in our study of asymmetric information, 'missing
> markets,' contracts, strategic interaction, and much else
> precisely because those aspects are regarded as real phenomena
> that require analysis,
I've never seen any indication from Rob that he acknowledges that
these issues, and their innumerable applications in a wide variety of
contexts, are important in modern economic thought. One would think
from reading Rob's posts (and some radical meta-analysis, which seems
to be all the economics Rob reads) that mainstream economics is
Marshall, end of story. Above, Rob's quoting criticism of one type of
modelling in macroeconomics, but mysteriously Rob applies this
criticism to all of mainstream economics.
Of course, Rob has never read a paper using an RBC model, he's just
quoting, Jay Hanson-like, meta-analysis of a literature he's entirely
unfamiliar with. Personally, and like many RBC theorists themselves,
I too am disappointed with the relatively small contributions RBC
models have made in helping us understand the effects of macroeconomic
policy changes. I don't think, however, that that failure has anything
to do with the CCC or reswitching. On the other hand, the statistical
methods developed by RBC researchers are revolutionary and have proved
themselves useful in a wide variety of micro and macroeconometric -- and
non-economic for that matter -- contexts.
--
Chris Auld
Department of Economics
University of Calgary
au...@ucalgary.ca
ro...@telus.net
Historical context. So it's "turtles, all the way down"...
;^)
-- Roy L
John Raymond Tyler
><JRT...@MailAndNews.com> wrote:
>
>>>===== Original Message From "David Lloyd-Jones" <dav...@sympatico.ca>
=====
>>
>>I am afraid that dynamic equilibrium is a meaningless concept, although it
>>is
>>widely used in economics. We might speak of an optimal path, but that is
>>something else.
>
>Imagine that the economy is a marble slowly rolling around on an
>enormous plate. It is always close to equilibrium, and getting
>closer. Now imagine that the plate is borne on the backs of four
>colossal elephants called Technology, Popular Taste, Government Policy
>and Business Sentiment. Economic equilibrium is like that.
>
perpetual pursuit of equilibrumm, but that is definitely not what is done.
I am afraid that the concept of equilibrium has such powerful normative
connotations that economists cannot bring themselves to recognize that it is
never in full equilibrum.
>>> How do we reconcile time preference with ordinal utility and rational
>>> expectations?
>>>
>>>Hunh? To need reconciliation there has to be some sort of problem. What is
>>>it you think is inconsistent here?
>>>
>>Time preference requires that we compare utilities in different time
>>periods.
>>This clearly requires a cardinal measure of utility. And if our
>>expectations
>>were rational, we should we have time preference; it is not rational to
>>prefer
>>present over future utility.
>
>Sure it is. The longer the delay, the greater the risk that you won't
>get to enjoy the future utility.
>
>"The pig might die. The king might die. I might die. And maybe, the
>pig will learn to sing."
>
>-- Roy L
Time preference has nothing to do with uncertainty; that is another concept
entirely. As a matter of fact, it might be useful to explicitly introduce a
discount for risk, but that is not done.
Frank Ramsey is one economist who argued that a rational consumer would not
have time preference.
John Tyler
John Raymond Tyler
>John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>>===== Original Message From "David Lloyd-Jones" <dav...@sympatico.ca>
=====
>>>
>>>fact
>>> of growth?
>>>
point,
>>>right there in an assigned position in some sort of Newtonian mechanical
>>>universe. This is not, however, how the word equilibrium is used in
>>>economics. It is, as you say above, a concept. Think dynamic equilibrium --
>>>like when you ride a bicycle. Then you'd be a good bit closer to what it
>>>means in economics.
>>>
>>
>>is
>>widely used in economics. We might speak of an optimal path, but that is
>>something else.
>
>accumulation case. However, a dynamic equilibrium would just be one where
>stationary state.
>
>>
I am afraid that no physicist would describe a moving bicycle as being in a
state of equilibrium.
General equilibrum is obviously just that; everything must be maintained,
not
just a few ratios.
>>
>>> How do we reconcile time preference with ordinal utility and rational
>>> expectations?
>>>
>>>Hunh? To need reconciliation there has to be some sort of problem. What is
>>>it you think is inconsistent here?
>>>
>>>
>>Time preference requires that we compare utilities in different time
>>periods.
>>This clearly requires a cardinal measure of utility. And if our
>>expectations
>>were rational, we should we have time preference; it is not rational to
>>prefer
>>present over future utility.
>
>
>>
who will offer more than $5 in the future for $5 now.
John Raymond Tyler
=====
>In article <3B59...@MailAndNews.com>,
>John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>>===== Original Message From mwi...@merle.acns.nwu.edu (Mark Patrick Witte)
>>=====
>>>John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>>><JRT...@MailAndNews.com> wrote:
>>>>
>>>>Isn't there a contradiction between the concept of equilibrium and the
fact
>>>>of growth?
>>>
>>>models where capital/labor ratios and the like preserve over growth of
>>>population and/or technology.
>>>
>>>
>>Preserving capital/labor ratios is not the same thing as a return to the
>>original position,which is what equilibrium means.
>
> Where does this definition come from? What if the original ratios
>have an overly low or high capital stock?
>
>> In equilibrium, there is
>>no tendency to leave the original position, the entire position, not just
>>some ratio.
>>
>> As I recall, he makes two elementary mistakes in calculus in deriving the
>>equation to be tested. It is true that they tend to cancel each other, but
>
> Yes, yes, and the editors at the QJE are drooling idiots, almost as
>dumb as Frank Ramsey.
>
>>
>>
have no idea how Frank Ramsey got into this.
In my first class in my graduate career at the University of Chicago, I
questioned Milton Friedman's methodology. Milton quite literally shouted me
down. Of course, this really made my reputation among the other students,
but
I am afraid that this not the way to carry on scholarly discourse.
I another part of this thread you stress the need for simplification in
builing models that can predict in the real world. There are two problems
with this. the real purpose of science is not prediction, it is
explanation.
Prediction without explanation is magic, not science. Quite meaningless
predictions can be made on the basis of some models. Many models assume
that
there is no depreciation, and Solow's is one of them. On the basis of these
models, we must then predict that no one ever has to buy a new car. This is
obvioius nonsense.
John Tyler
David Lloyd-Jones
incorrectly attributed to me.
-dlj.
"John Raymond Tyler" <JRT...@MailAndNews.com> wrote in message
news:3B61...@MailAndNews.com...
Robert Vienneau
(Christopher Auld) wrote:
> Robert Vienneau <rv...@see.sig.com> wrote:
> > "Those who have followed the literature of the past 20 years know
> > that...many properties of the Clarkian paradigm do NOT all hold
> > generally. In every Cambridge, and wherever informed economic
> > theory is taught, there is no...disagreement on this."
> > -- Paul A. Samuelson, (QJE 1976)
> Incidentally, Samuelson published one paper in the QJE in 1976, and
> it does not include Rob's quote above.
I thank Chris for alerting me to this error. I have confused
Samuelson's February 1975 article in the QJE with a later 1976
paper. The quote is from the 1976 paper not in the QJE.
As for the rest of Chris' post - poor Chris Auld.
Robert Vienneau
(Mark Patrick Witte) wrote:
> [ No there there. ]
I remain skeptical of claims that substantial theoretical and
empirical questions can be resolved by appeals to methodology, with
no discussion of the substance of the questions. (That's
consistent with contemporary trends in methodology, including
the trend away from prescriptive methodology - a trend given
a large shove by that obscure writer, Thomas Kuhn.)
Mr. Witte simply does not address the economic theory that leads
some economists to conclude that it is of questionable ethics
to use Solovian growth theory for policy recommendations.
Perhaps he does not understand the theory. His refusual to recognize
a connection between Sraffa and Leontief certainly suggests as
much.
One of my favorite titles for a journal article:
"'This Age of Leontief ... and Who?' An Interpretation"
It remains the case that it is well-known among economists that
some have raised good reasons to think Solovian growth theory is
without theoretical and empirical foundation.
Some interesting articles on the (lack of) empirical foundations:
J. S. L. McCombie, "What does the aggregate production function
show? Further thoughts on Solow's 'Second thoughts on growth theory'".
JPKE, Summer 20001, 23:4, 589-615
J. S. L. McCombie, "The Solow Residual, Technical Change, and
Aggregate Production Functions". JPKE, Winter 2000-2001, 23:2,
267-297.
If Mr. Witte is going to respond, while completely ignoring
all points of substance, I'm sure many would rather read him
on his surroundings, than some boring misdirected rant on
methodology.
con...@email.rahul.net
My favorite quote from P. Samuelson is from 1952:
"The search of the great minds of recorded history for the perfect
democracy, it turns out, is the search for chimera, for logical
self-contradiction. Now scholars all over the world-in
mathematics, politics, philosophy, and economics-are trying to
salvage from [Kenneth Arrow's 1951 so-called Impossibility
Theorem; the voters paradox], devastating discovery that is to
mathematical politics what Kurt Godel's 1931
impossibility-of-proving-consistency theorem is to mathematical
logic."
His most significant contribution, IMHO, in 1965, was the formal proof
of Bachelier's empirical theories on the characteristics of
speculative markets in, "Proof that Properly Anticipated Prices
Fluctuate Randomly," which became the underlying foundations of
Black-Scholes methodology, (where the standard deviation of the
marginal increments in a stock's price is a metric of investment risk,
which establishes the isomorphism between the value of financial
instruments and fractals.)
His contributions are often overlooked, IMHO.
John
Robert Vienneau writes:
> In article <9j221e$5l...@acs4.acs.ucalgary.ca>, au...@acs.ucalgary.ca
> (Christopher Auld) wrote:
>
> > Robert Vienneau <rv...@see.sig.com> wrote:
>
> > > "Those who have followed the literature of the past 20 years know
> > > that...many properties of the Clarkian paradigm do NOT all hold
> > > generally. In every Cambridge, and wherever informed economic
> > > theory is taught, there is no...disagreement on this."
> > > -- Paul A. Samuelson, (QJE 1976)
>
> > Incidentally, Samuelson published one paper in the QJE in 1976, and
> > it does not include Rob's quote above.
>
> I thank Chris for alerting me to this error. I have confused
> Samuelson's February 1975 article in the QJE with a later 1976
> paper. The quote is from the 1976 paper not in the QJE.
>
> As for the rest of Chris' post - poor Chris Auld.
>
--
John Conover Tel. 408.370.2688 con...@email.rahul.net
631 Lamont Ct. Fax. 408.379.9602 http://www.johncon.com/
Campbell, CA 95008 Cel. 408.772.7733
Mark Patrick Witte
John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>>
>>>I am afraid that dynamic equilibrium is a meaningless concept, although it
>>>is
>>>widely used in economics. We might speak of an optimal path, but that is
>>>something else.
>>
>> No, dynamic equilibria need not be optimal, as in the capital-over
>>accumulation case. However, a dynamic equilibrium would just be one where
>>ratios are at their long run values, or which can be normalized to a
>>stationary state.
>>
>>>
>I am afraid that no physicist would describe a moving bicycle as being in a
>state of equilibrium.
>General equilibrum is obviously just that; everything must be maintained,
>not
>just a few ratios.
I'm not so sure there. I recall working problems with
(frictionless) gyroscopes where there was certainly a concept of
equilibrium. However, I'm badly out of my area here. As for dynamic growth
equilibria versus optimal paths, note my remark above about normalization.
The best discussion I can recall of this is the series of papers by
King-Plosser-Rebelo in the JMonetaryE.
>>>Time preference requires that we compare utilities in different time
>>>periods.
>>>This clearly requires a cardinal measure of utility. And if our
>>>expectations
>>>were rational, we should we have time preference; it is not rational to
>>>prefer
>>>present over future utility.
>>
>> So...you'll lend me $5 today in exchange for $5 a year from now?
>>
>Of course not. There are plenty of consumers who obviously aren't rational
>who will offer more than $5 in the future for $5 now.
Well, perhaps we can agree that there is someone who's irrational
Grinch
Witte) wrote:
>In article <rvien-7B5250....@news.dreamscape.com>,
>Robert Vienneau <rv...@see.sig.com> wrote:
>>In article <9ir797$lom$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>>(Mark Patrick Witte) wrote:
>>
>>> In article <rvien-68CAAA....@news.dreamscape.com>,
>>> Robert Vienneau <rv...@see.sig.com> wrote:
>>
>>> >In article <9il43l$m63$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>>> >(Mark Patrick Witte) wrote:
>>
>>> >> In article <rvien-C012CC....@news.dreamscape.com>,
>>> >> Robert Vienneau <rv...@see.sig.com> wrote:
>>
>>> >> >(It is well-established that
>>> >> >one-sector growth models are deeply questionable as useful and
>>> >> >empirically validated tools for help in providing insight into
>>> >> >actual economies. See two most recent issues of Journal of Post
>>> >> >Keynesian Economics.)
>>> >>
>>> >> [ Emotional outburst undirected to cognitive values - deleted. ]
>>
>>> [ Whining about deletions of portions of posts directed to ]
>>> [ non-cognitive values - deleted. ]
>>
>>Mayhaps Mr. Witte can provide a post that does not suggest he's in
>>Oakland - a place well known for being someplace where's no there
>>there.
> Wow, two complete deletes ...
> It is unfortunate that my ass-kicking is so painful to some
>poster that he feels the need to delete my remarks yet reply to them.
But this is a remarkable posting technique...
~~
>>>> My comments repeated...
>>>[All of yours deleted]
>>My reply repeated ...
>[Yours deleted]
I reply again ... and conclude from the forgoing that I'm right!
~~
This is such a natural for usenet it's hard to imagine how we haven't
seen it before. If it sweeps through the groups now, remember where
you saw it first!
Mark Patrick Witte
John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>>
>>>It has been some time since I read Solow, which cropped up in this
>>>discussion.
>>> As I recall, he makes two elementary mistakes in calculus in deriving the
>>>equation to be tested. It is true that they tend to cancel each other, but
>>>his work is fatally flawed.
>>
>> Yes, yes, and the editors at the QJE are drooling idiots, almost as
>>dumb as Frank Ramsey.
>>
>I am afraid that an argument from authority doesn't carry much weight, and I
>have no idea how Frank Ramsey got into this.
The Ramsey comment recalls a post from some time back with a similar
claim. It's hard to do anything but argue from authority against vague
claims about how a professor of statistics made to elementary mistakes in
calculus yet got them by the editors of the QJE.
>In my first class in my graduate career at the University of Chicago, I
>questioned Milton Friedman's methodology. Milton quite literally shouted me
>down. Of course, this really made my reputation among the other students,
>but I am afraid that this not the way to carry on scholarly discourse.
Friedman was never one to suffer fools gladly.
>I another part of this thread you stress the need for simplification in
>builing models that can predict in the real world. There are two problems
>with this. the real purpose of science is not prediction, it is
>explanation.
>Prediction without explanation is magic, not science. Quite meaningless
>predictions can be made on the basis of some models.
No. This is the difference between "prediction" and
"forecasting", although both come from attempts at explaining some
phenomenon. For instance, Greg Mankiw's first durable goods paper predicted
an ARMA(1,1) in consumer spending on such goods, but such a result was not
found. Similarly, some of Einstein's work had the famous prediction of
bending of light, a result that found rapid confirmation. These two results
were fun because they really were "prediction", rather than pure attempts at
explanation of observed behavior, but even there it is common to say that a
model "predicts" certain empirical results, which then may or may not be
found.
>Many models assume that
>there is no depreciation, and Solow's is one of them. On the basis of these
>models, we must then predict that no one ever has to buy a new car. This is
>obvioius nonsense.
In many such models, such as Solow's, depreciation is implicit in
the net rate of return and is just suppressed for notational simplicity.
Mark Patrick Witte
assumptions of a model without offering a comparison to an alternative model
that contains a "preferred" assumption set that offers improved empirical
support (by some metric)? It's called "wanking."
If someone likes a given set of assumptions, let him find a way to
construct an implementable model that includes them and then let the data be
a judge. Or go wank in private.
In article <rvien-B7BBCA....@news.dreamscape.com>,
Robert Vienneau <rv...@see.sig.com> wrote:
>In article <9j1tl5$sub$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>(Mark Patrick Witte) wrote:
>
> [ No there there. ]
All these Oakland refernces...is it about baseball? Perhaps Robert
Vienneau is an Athletics supporter?
>I remain skeptical of claims that substantial theoretical and
>empirical questions can be resolved by appeals to methodology, with
>no discussion of the substance of the questions. (That's
>consistent with contemporary trends in methodology, including
>the trend away from prescriptive methodology - a trend given
>a large shove by that obscure writer, Thomas Kuhn.)
>
>Mr. Witte simply does not address the economic theory that leads
>some economists to conclude that it is of questionable ethics
Hmm...from "well established" to "some economists"....
>to use Solovian growth theory for policy recommendations.
>Perhaps he does not understand the theory. His refusual to recognize
>a connection between Sraffa and Leontief certainly suggests as
>much.
On no, the connection is well known. Sraffa is to Leontief
as...footnote is to magnum opus.
Christopher Auld
Robert Vienneau <rv...@see.sig.com> wrote:
>(Mark Patrick Witte) wrote:
Well, after Rob's devastating response to my post, I suppose I have been
put in my place. "Poor Chris Auld." It is indeed humbling to have one's
arguments put to rest with a mere three words; we are truly in the
presence of greatness. Perhaps I will be forgiven the affront of
commenting on Rob's post to Mark Witte nonetheless.
Rob says:
>It remains the case that it is well-known among economists that
>some have raised good reasons to think Solovian growth theory is
>without theoretical and empirical foundation.
But earler contended:
>It is well-established that
>one-sector growth models are deeply questionable as useful and
>empirically validated tools for help in providing insight into
>actual economies.
Now, to mere mortals such as myself, these are not equivalent statements.
But likely I have just misunderstood Rob's wisdom, so let's consider the
most recent statement. Is it in fact "well known among [sic] economists
that some have raised good reasons...?" Rob cites two articles, both by
one JSL McCombie, and both in the Journal of Post Keynesian Economics, in
support of his claim. I read the latest one:
> J. S. L. McCombie, "What does the aggregate production function
> show? Further thoughts on Solow's 'Second thoughts on growth theory'".
> JPKE, Summer 20001, 23:4, 589-615
Not being a specialist in macroeconomics, I don't have strong priors on
this literature. I'm also quite open to claims along the lines that some
macro subliteratures are barking up the wrong tree, so it actually
wouldn't surprise me much if McCombie had some truly stinging critiques.
Does he, in my opinion? Not even close. Does he even establish Rob's
much weaker (second) claim that it's "well-known" that Solovian growth
theory is "deeply questionable?" Nope, aside from the obvious and much
more broadly applicable aggregation problems inherent to much of
economics.
First, if we limit ourselves to papers published since 1964, this
literature comprises a handful of papers published in outlets ranging from
minor mainstream journals (e.g., Applied Economics) to obscure radical
journals (e.g., JPKE and JRPE). Since none of these journals are
particularly widely circulated or influential, this branch of criticism
isn't really "well known" at all, much less constituting such widespread
and damning criticism as to justify Rob's initial histrionics.
Second, is McCombie right? The argument revolves around regressions like
lnQ = a + b(lnL) + c(lnK) + noise.
McCombie presents some simple simulations which could have just as
easily been expressed analytically. He generates data off just such a
process, but in a pointlessly obtuse fashion which masks the fact that the
errors in his simulated data have almost zero variation, an unknown
distribution, and will be correlated by construction with the regressors
(there's a *lot* of highly questionable econometrics in this piece).
He then runs the regression above and finds that \hat b and \hat c very
precisely estimate the true underlying process. Vindicating Solow? No,
according to McCombie, who proceeds to add a new regressor of the form:
e(lnL) + f(lnK)
to the specification. Obviously, when put this way (the paper is
somewhat obtusely written, a problem aggravated by and a number of
mathematical typos), the regression recovers (b+e) and (c+f) rather
than b and c. McCombie triumphantly announces than b and c are no
longer being accurately estimated, so so much for Solow.
All of this could have been expressed analytically without any
simulations, and the results of the simulations are presented as if they
are saying something about econometric results when in fact they say
nothing, if for no other reason than the fit is so close to perfect (all
the R^2's are reported as 1.000) that the specification and hypothesis
tests reported are meaningless. Beneath the surface, there's nothing more
here than: omitted variable bias and other specification errors can lead
to misleading regression results. Since this observation applies with
equal force everywhere, it's not an interesting critique of Solow's
econometric method.
For the rest of the paper, McCombie discusses aggregation problems.
Aggregation problems are, of course, rampant in macroeconomics. McCombie
quotes Solow and Fisher (1977), who conclude "aggregate production
functions are simply too useful to pass up, especially when they can work,
as our experiments show [even when the aggregation conditions are
violated]." McCombie disagrees, but he doesn't disagree on compelling
grounds.
So, in short, McCombie adds to a niche literature, and he adds only some
badly done simulations and a rehash of age-old aggregation problems. His
article certainly doesn't do anything to prop up Rob's initial contention,
and doesn't even salvage the much weaker revisionist history Rob presented
yesterday. Of course, none of this has any bearing on the putative point
of the thread, someone's strange contention that "dynamic equilibrium" is
a nonsense concept, followed by a misunderstanding of the old argument
that the social discount rate ought to be zero as an argument that
individuals shouldn't discount. Even Rob ought to admit those particular
critiques of economics are silly, but, alas, Rob has never met a critique
of economics he hasn't liked, and any post with anything related to
economics in it looks though Rob-colored glasses like an invitation to
discuss the CCC.
Now, maybe I should write some of this up as a note for some obscure
journal, so my morning won't be entirely wasted from a research point of
view.
Mark Patrick Witte
Robert Vienneau <rv...@see.sig.com> wrote:
>In article <9j1tl5$sub$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>(Mark Patrick Witte) wrote:
>
>
>Some interesting articles on the (lack of) empirical foundations:
>
> J. S. L. McCombie, "What does the aggregate production function
> show? Further thoughts on Solow's 'Second thoughts on growth theory'".
> JPKE, Summer 20001, 23:4, 589-615
>
> J. S. L. McCombie, "The Solow Residual, Technical Change, and
> Aggregate Production Functions". JPKE, Winter 2000-2001, 23:2,
> 267-297.
>
>If Mr. Witte is going to respond, while completely ignoring
>all points of substance, I'm sure many would rather read him
>on his surroundings, than some boring misdirected rant on
>methodology.
For those who are interested, I'm in New Mexico, in a place where
people seem to really like using the word "chaos." Getting bored, I managed
to score two recent copies of the JPKE and gave the cited articles a read.
Chris Auld's comments are more thought through than mine but McCombie struck
me as just summarizing results and adding nothing that anyone who's taken a
class in time series would not already have seen.
Of course, he only presents a critique of direct attempts to
estimate aggregate production functions; he does not present any alternative
empirical model that contains elements acceptible to CCC addicts.
Is anyone who has followed such threads surprised by this?
con...@email.rahul.net
had no mean?
John
BTW, or even worse, if it was, as mentioned, chaotic, (as in NLDS,)
and the mean diverged?
John Conover Tel. 408.370.2688 con...@email.rahul.net
ro...@telus.net
>Wouldn't it be interesting if aggregate production was fractal, and
>had no mean?
>
>BTW, or even worse, if it was, as mentioned, chaotic, (as in NLDS,)
>and the mean diverged?
Sounds all too plausible.
-- Roy L
Robert Vienneau
(Mark Patrick Witte) wrote:
> Just a reminder...
> [ And more silliness. ]
> >Perhaps he does not understand the theory. His refusual to recognize
> >a connection between Sraffa and Leontief certainly suggests as
> >much.
> On no, the connection is well known. Sraffa is to Leontief
> as...footnote is to magnum opus.
So Mr. Witte concedes he was mistaken in denying that Leontief
input output analysis provides an empirical implementation of
Sraffa's work by Mr. Witte's standards. Or maybe Mr. Witte is
trying to resolve questions by personalities and authority.
Mr. Witte wants a model. Nothing is easier to those who understand
what's going on.
Postulate a 1-good model with an aggregate production function of
the Leontief fixed coefficients form. The coefficients change
over time as a consequence of technical progress. Let's
postulate the Kahn-Kaldor-Pasinetti-Robinson theory of
distribution.
Is this model supported by the data? I suggest all the empirical
work Mr. Witte cited in support of Solovian growth theory also
support this theory equally well. (I'm rely mostly on secondary
literature for this claim.)
To investigate this claim, one would have to understand what the
data can and cannot show with the techniques used to date. Perhaps
there is no empirical support for either theory in that the
techniques fail to distinguish these theories from
accounting tautologies. Serious people interested in this
question would be interested in the Phelps-Brown/Simon/Fisher/
Shaikh argument cumulating, for now, in the work of McCombie that
I previously cited.
The above questioning of the existence of empirical support
for Solovian growth theory is different from questioning
aggregation conditions or the meaning of capital. The latter
are two separate questions. Some have asked how could the world
be, in the multi-commodity case, such that Solovian growth theory
(or other neoclassical theories) apply. They think they have
found a reductio ad absurdum in the theory of production. That
is neoclassical theory is logically absurd.
Both classical and marginalist economists provided accounts of
the long-period (uniform rate of profit) theory of value and
distribution, but whereas a classical economist could take the
real wage as a datum for the purpose of such analysis (whatever
the implicit 'background' theory of wages might be), the
marginalist economist had to 'close the system' in some other
manner. In effect, since 'resource supplies' were often taken as
given, this meant that 'the supply of capital' had to be taken
as given, IN ONE WAY OR ANOTHER. Just how the given supply of
capital was to be represented was an issue that led to
considerable heterogenity amongst even those marginalist
economists who shared the long-period method of analysis with
the classical economists and with each other. That heterogenity
cannot be entered into here (see Kurz and Salvadori, 1995:
427-43) but it is now widely recognized that each version of
such traditional long-period marginalist theory of value and
distribution encountered insoluble problems (ibid.: 443-48).
-- Ian Steedman (1998)
Harcourt: ...John Green did a good paper on aggregation, but it was
a bit peripheral to the problems, especially if you think the critique
is not about aggregation but about the meaning of capital, which as
a Marxist I'm sure you'd accept as a fairer characterization. Just
because it's hard to aggregate labour is neither here nor there,
because in real life you don't need to aggregate labour. If you're
going to have a comprehensive system you can have as many classes of
workers as you want, and all you're arguing is that the wage for any
one class in a competitive system will be the same, there'll be a
law of one price. But for capital you have to explain the return in
the economy as a whole, and therefore you have to have a theory of
the rate of profits in the economy as a whole. If you're going to
use a supply-and-demand analysis you have to know, before the analysis
starts, what a quantity of capital is, because it's one of the
exogeneous or determining variables. It's as simple as that. It's
got nothing to do with mutual determination or general equilibrium.
It's to do with what's inside and what's outside, and the trouble
with capital in a supply-and-demand approach is that it's
simultaneously exogeneous and endogeneous. That is, you're arguing
in a circle.
-- J. E. King, _Converstions with Post Keynesians_, St. Martin's
Press, 1995
Note the above cannot be summarized by a strawperson such as
"complain[t]s about the assumptions of a model."
It remains the case that it is well-known among economists that
some have raised good reasons to think Solovian growth theory is
without theoretical and empirical foundation.
I think John Tyler's objections were elsewhere. I originally took
them to be about the well-known difficulties of getting into
equilibrium.
con...@email.rahul.net
http://www.johncon.com/john/correspondence/981014233807.19309.html
Monthly, quarterly, and yearly data on the dynamics seems to be
related by the square root function. If it is NLDS, it seems that
stochastic calculus is a good expedient "fit", (depending on who is
telling the story, of course.)
John
BTW, the 60 some programs at
http://www.johncon.com/ndustrix/utilities.html have had the entire US
DOC database, (its available on CD, Re: http://www.stat-usa.gov/,) run
through them. The only exception was MSELINFS which is a derived
index, (it seemed to be anti-persistent,) which was traced to the way
the data was presented.
John Conover Tel. 408.370.2688 con...@email.rahul.net
David Lloyd-Jones
news:9j8k7j$13o$1...@samba.rahul.net...
> Wouldn't it be interesting if aggregate production was fractal, and
> had no mean?
John,
I think this leads to one of the most convincing lines of support for the
Bachelier/Mandelbrot view of economic timeseries.
Critics of Mandelbrot say that if his view of economics were correct,
everything would fly to pieces every now and then.
Well?
-dlj.
Mark Patrick Witte
Robert Vienneau <rv...@see.sig.com> wrote:
>In article <9j5o8m$jbu$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
>(Mark Patrick Witte) wrote:
>
>> Just a reminder...
>> [ And more silliness. ]
>
>> >Perhaps he does not understand the theory. His refusual to recognize
>> >a connection between Sraffa and Leontief certainly suggests as
>> >much.
>
>> On no, the connection is well known. Sraffa is to Leontief
>> as...footnote is to magnum opus.
>
>So Mr. Witte concedes he was mistaken in denying that Leontief
>input output analysis provides an empirical implementation of
>Sraffa's work by Mr. Witte's standards.
I deny that the magnum opus in question is an empirical
implementation of the footnote in question.
Specifically, what are the empirical implications of Sraffa's work?
Christopher Auld
>So Mr. Witte concedes he was mistaken in denying that Leontief
>input output analysis provides an empirical implementation of
>Sraffa's work by Mr. Witte's standards.
Leontief I/O analysis is not only rarely used in academic research,
it is not an "empirical implementation of Sraffa's work."
>Mr. Witte wants a model. Nothing is easier to those who understand
>what's going on.
This is rich.
>Postulate a 1-good model with an aggregate production function of
>the Leontief fixed coefficients form.
Rob apparently understands absolutely nothing about the obscure
debate he's fixated upon if he thinks doing nothing but making
the aggregate production function model dramatically *more*
restrictive somehow evades any problems with that approach.
>accounting tautologies. Serious people interested in this
>question would be interested in the Phelps-Brown/Simon/Fisher/
>Shaikh argument cumulating, for now, in the work of McCombie that
>I previously cited.
>The above questioning of the existence of empirical support
>for Solovian growth theory is different from questioning
>aggregation conditions or the meaning of capital.
Again amusing. If Rob actually ever read any of this literature
rather than abstracts and dictionary articles about it, he'd
find that it is indeed mostly about "aggregation conditions and
the meaning of capital." The first line of the "McCombie piece
he previously cited" (Rob has apparently never read the article)
is
"The traditional defense of the use of the aggregate production
function, in the face of serious problems posed by aggregation
issues and the Cambridge Capital Theory Controversies, is...."
> The latter
>are two separate questions. Some have asked how could the world
>be, in the multi-commodity case, such that Solovian growth theory
>(or other neoclassical theories) apply.
This is an "aggregation problem," Rob. Get your dictionary
out, look it up.
>Note the above cannot be summarized by a strawperson such as
>"complain[t]s about the assumptions of a model."
Yes, Rob, but they're talking about different issues that
empirical implementations of Solovian growth theory. The
"debate" you want to talk about, but haven't read, is exactly
about "the assumptions of a model."
>It remains the case that it is well-known among economists that
>some have raised good reasons to think Solovian growth theory is
>without theoretical and empirical foundation.
That's right, Rob: just follow Jay Hanson's lead, dig your
heels in, cover your ears, and chant the same thing over
and over again until the bad people pointing out the problems
with it get bored and go away.
To repeat: the handful of papers Rob refers to are not well-known,
and the issues they raise are not considered important "among
economists." If the critics were to propose alternate methods
which evade the theoretical problems with aggregate production
functions and also provide useful empirical frameworks, the
story might change.
>I think John Tyler's objections were elsewhere. I originally took
>them to be about the well-known difficulties of getting into
>equilibrium.
Anyone unfortunate enough to have read the thread knows very
well Rob completely ignored what Tyler actually said and launched
into an irrelevant diatribe on the CCC.
con...@email.rahul.net
assign epistemological meaning to our interpretations of mathematical
abstractions, but there are several schools of political thought that
claim social things have a very high fractal dimension.
Its an attempt to explain why society's paradigms remain static for
long periods, and then make a jump, in some kind of a
revolution/rebellion.
A Cauchy bell shaped distribution is often evangelized for such
things-it has no mean, (the mean diverges,) and one is forced to work
with the median. Also, root mean square math can't be used, and one is
forced to work with the interquartile range-the variance has no
meaning.
A time series with a Cauchy distribution can be manufactured:
x = tan [pi * (0.5 - u)]
where u is a uniform variable in the interval [0,1].
John
BTW, if the tangent operator is replaced with a slight exponentiation,
it produces a log normal distribution, which is probably ubiquitous in
economic data, since aggregates are a sum of things; its distribution
is a bell shaped curve, too-but with fatter tails than a normal
distribution.
A more precise formulation for economic data is to include
non-linearity; the random operator is multiplication instead of a
sum-which is straight Bachelier. The variables can be extrapolated
over several orders of magnitude for a lot of economic data sets.
But its not clear what the word "aggregate" means in such things.
Robert Vienneau
(Christopher Auld) wrote:
> Robert Vienneau <rv...@see.sig.com> wrote:
> >It remains the case that it is well-known among economists that
> >some have raised good reasons to think Solovian growth theory is
> >without theoretical and empirical foundation.
> ...Is it in fact "well known among [sic] economists
> that some have raised good reasons...?" Rob cites two articles, both by
> one JSL McCombie, and both in the Journal of Post Keynesian Economics, in
> support of his claim. I read the latest one:
> > J. S. L. McCombie, "What does the aggregate production function
> > show? Further thoughts on Solow's 'Second thoughts on growth theory'".
> > JPKE, Summer 20001, 23:4, 589-615
> Not being a specialist in macroeconomics, I don't have strong priors on
> this literature. I'm also quite open to claims along the lines that some
> macro subliteratures are barking up the wrong tree, so it actually
> wouldn't surprise me much if McCombie had some truly stinging critiques.
Chris fails to mention that he's already taken a position on this
argument in his ranting about Scott Moss last year. If he were
to accept McCombie's argument, he would have to admit much of what
he said then was misdirected or mistaken. So the comment about
lacking "strong priors" is either misleading or a reflection of
a lack of understanding of what he's talking about.
> Does he, in my opinion? Not even close. Does he even establish Rob's
> much weaker (second) claim that it's "well-known" that Solovian growth
> theory is "deeply questionable?" Nope, aside from the obvious and much
> more broadly applicable aggregation problems inherent to much of
> economics.
> First, if we limit ourselves to papers published since 1964, this
> literature comprises a handful of papers published in outlets ranging
> from
> minor mainstream journals (e.g., Applied Economics) to obscure radical
> journals (e.g., JPKE and JRPE).
1964 is totally arbitrary. McCombie says in his first paragraph that
his argument is a continuation of the argument in these papers:
E. H. Phelps Brown, "The Meaning of the Fitted Cobb-Douglas
Function", _Quarterly Journal of Economics_, 1957.
H. A. Simon and F. K. Levy, "A Note on the Cobb-Douglas Function",
_Review of Economics and Statistics_, 1963.
A. Shaikh, "Laws of Production and Laws of Algebra: The Humbug
Production Function", _Review of Economics and Statistics, 1974.
H. A. Simon, "On Parsimonious Explanations of Production
Relations", _Scandinavian Journal of Economics_, 1979.
A. Shaikh, "Laws of Production and Laws of Algebra: Humbug II"
in some book, 1980.
A. Shaikh, "Humbug Production Function", in _New Palgrave_, 1987.
J. S. L. McCombie, "Does the Aggregate Production Function Imply
Anything About the Laws of Production? A Note on the Simon and
Shaikh Critiques", _Applied Economics_, 1987.
J. S. L. McCombie and R. Dixon, "Estimating Technical Change in
Aggregate Production Functions: A Critique," _International
Review of Applied Economics_, 1991.
J. S. L. McCombie, "Are There Laws of Production? An Assessment of
the Early Criticisms of the Cobb-Douglas Production Function"
_Review of Political Economy_, 1998.
J. S. L. McCombie and A. P. Thirlwall, _Economic Growth and the
Balance-of-Payments Constraint_, Macmillan, 1994.
J. Felipe, "On the Myth and Mystery of Singapore's 'Zero TFP'",
_Asian Economic Journal_, 2000.
J. S. L. McCombie, "Regional Production Functions and the Accounting
Identity: A Problem of Interpretation", _Australian Journal of
Regional Studies_, 2000.
J. S. L. McCombie, "The Solow Residual, Technical Change, and
Aggregate Production Functions", JPKE, Winter 2000-2001.
And a couple of forthcoming articles. According to McCombie (2000-2001),
Herb Simon mentioned this argument in his "Nobel" prize acceptance
speech.
Also important to the argument in the paper are:
F. M. Fisher, "Aggregate Production Functions and the Explanations
of Wages. A Simulation Experiment," _Review of Economics and
Statistics_, 1971.
F. Fisher, R. M. Solow, and J. M. Kearl, "Aggregate Production
Functions: Some CES Experiments", _Review of Economic Studies_,
1977.
R. M. Solow, "Laws of Production and Laws of Algebra: The Humbug
Production Function: Comment". _Review of Economics and Statistics_,
1974.
R. M. Solow, "Second Thoughts on Growth Theory" in some book, 1987.
The Solow articles, including the co-authored one, argue against
the line of thought in the first set of articles. That is, they
argue that the failure of Solow's growth accounting approach to
test aggregate neoclassical theory is unimportant.
> Since none of these journals are
> particularly widely circulated or influential, this branch of criticism
> isn't really "well known" at all, much less constituting such widespread
> and damning criticism as to justify Rob's initial histrionics.
Right. Nobody pays attention to arguments among "Nobel" laureates,
the QJE, and RES. It's quite clear why Chris would have a problem with
the very existence of ROPE. And Thirlwall is a nonentity, not to
mention Franklin Fisher or Phelps Brown.
In my opinion, anybody unfamiliar with the issues raised in this
literature is hard to take seriously in discussions of whether
Solovian growth theory is empirically well-founded (just as anybody
unfamiliar with the CCC has trouble arguing neoclassical economics is
theoretically valid). Suppose I were to accept Chris' claim that
this literature is not well-known. What conclusion should I draw?
On the other hand, what conclusion should I draw from this
literature being well-known, but ignored.
> Second, is McCombie right?
Since Chris doesn't come close to summarizing the point of this paper,
I'll have to do it.
Solovian growth theory has been used extensively in applied work,
particularly in assigning the causes of growth to technical
change and increases in capital per worker. One can find regressions
of growth rates and such like in this literature. For example,
Solow (1957) reports five regressions of different forms of
production functions, with correlation coefficients ranging
from 99.64% to 99.96%. The question addressed here is to what
extent this literature can be said to have tested aggregate
neoclassical theory, the aggregate marginal productivity theory
of distribution, and so on.
The contention of this paper, and the previous literature on
which it draws, is that much of these empirical results are
not tests and give no reason to suppose aggregate neoclassical
theory is empirically valid. Rather, these supposed empirical
results are the consequence of an accounting identity and a
few stylized facts:
o There is no long run trend in the rate of profits (although
it varies procyclically)
o The shares of wages and profits in income is roughly
constant
o Wages grow at a roughly constant rate.
o The share of wages is roughly 75% of national income.
These facts held better in the United States during the postwar
"golden age". As I show in an appendix, these facts imply a Cobb-Douglas
production function will fit the data, as assessed by some simple
regressions of rates of growth. Factor shares will be equal to
estimated output elasticities. Total factor productivity growth
(i.e., the growth in output per worker not due to increases in inputs
of capital per worker) must be about three-quarters of the growth in
output per worker.
Thus, this fit of a Cobb-Douglas prodution function cannot be
used to support aggregate neoclassical parables. The fit would
result even if some other non-marginal productivity theory of
distribution generated the data. McCombie mentions a markup
theory of prices, for example. (McCombie does not provide
a tutorial on alternate theories of value and distribution;
he can assume that regular readers of the JPKE have been
exposed to a variety of "models".) Furthermore, estimates of
output elasticities and total factor productivity growth may
not tell us anything about the underlying technology, marginal
productivities, etc. McCombie says these conclusions extend
even to cases where the above stylized facts are not exact:
"If the factor shares are not exactly constant, it may be possible
to find more flexible functional forms, such as the CES... or
translog, that could improve the statistical fit. Thus, it should
be emphasized that this critique is not confined just to the
Cobb-Douglas, although for expositional ease attention will be
confined to this particular form."
The particular purpose of this paper is to address Solow (1987).
Solow argues that the above argument applies to physical (engineering)
production functions, as well as production functions with data (Y, K)
given in price terms. Since one can estimate engineering production
functions, the argument cannot be correct. Solow also identifies
where he thinks Shaikh and others go wrong.
McCombie argues output elasticities can be misestimated as a wide
range of values from quantity data if technical change is proxied
badly. So there is a difference in the implications of the argument
for aggregate production functions (estimated with price data) and
for engineering production functions (estimated with quantity data).
The point of the simulation exercises is to illustrate these
claims. The point is the simulator, such as Franklin Fisher,
can control whether marginal productivity theory applies and
the actual values of income shares, etc.
Before turning to more detail about simulation, I'd like to
mention that it is not McCombie's contention that there are no
possible tests of aggregate neoclassical theory. In his JPKE
2000-2001 paper, he considers Solow's initial response to Shaikh's
humbug example. Using Shaikh's made up data, Solow regressed:
log y = a + b t + c log k
where y is income per head, k is capital per head, and t is time.
He found the independent variables do not explain the variation
in the dependent variables and the point estimate for c to be
negative.
"If this were the typical outcome with real data
we would not now be having this discussion."
(Solow 1974)
McCombie (2000-2001) regresses Solow's data (with Hogan's
correction). He does not impose the condition that the
exponents in the Cobb-Douglas must sum to unity.
He finds the point estimate of c is negative, but
statistically insignificant. So Solow's comment seems to have
been mistaken or misleading.
Now for simulation. McCombie first does some simulations
with engineering production functions. As I understand
it, the regressions here are intended to demonstrate that one
can tell something about the underlying process from the data
and straightforward analyses.
McCombie then turns to simulations generating value data.
For value data to matter, he introduces two industries. Chris
would have it:
> For the rest of the paper, McCombie discusses aggregation problems.
McCombie writes:
"There is now the problem of how to value the stock
of (different) machines in each industry, which is
necessary to sum them. There has, of course, been an
extensive discussion on the theoretical difficulties
of this, which we will simply bypass here."
I read that as putting aside the CCC and aggregation problems.
> Aggregation problems are, of course, rampant in macroeconomics. McCombie
> quotes Solow and Fisher (1977), who conclude "aggregate production
> functions are simply too useful to pass up, especially when they can
> work,
> as our experiments show [even when the aggregation conditions are
> violated]." McCombie disagrees, but he doesn't disagree on compelling
> grounds.
McCombie quotes Solow and Fisher further:
"Our parting advice is to handle [aggregate production functions]
the way the old garbage man tells the young garbage man to handle
garbage wrapped in plastic bags of unknown provenance: 'Gingerly,
Hector, gingerly.'"
McCombie has a footnote about here:
"It is difficult to know what to make of this advice. Does one,
for example, use the aggregate production function to calculate
(as in the growth accounting approach), or to estimate (by
regression analysis), the rate of technical change (growth of
total factor productivity) and then add the rider that these
estimates may well be totally spurious as the conditions for
aggregation are (except under some highly implausible
assumptions) violated? Needless to say, this caveat concerning
the aggregate production function is normally never stressed
by those who actually use the concept (Solow being the
notable exception).
Anyways, part of the point is that Fisher came to the opposite
conclusion in his own solo papers. Fisher discovered through
simulation what Shaikh later showed analytically: good fits
of aggregate production functions using standard techniques
are the consequence of stable income shares. Economists
ought to work on developing better techniques and ought to
be aware that good reason has been given to think Solovian
growth theory is not empirically supported.
> So, in short, McCombie adds to a niche literature, and he adds only some
> badly done simulations and a rehash of age-old aggregation problems. His
> article certainly doesn't do anything to prop up Rob's initial
> contention,
> and doesn't even salvage the much weaker revisionist history Rob
> presented
> yesterday.
> Of course, none of this has any bearing on the putative point
> of the thread, someone's strange contention that "dynamic equilibrium" is
> a nonsense concept, followed by a misunderstanding of the old argument
> that the social discount rate ought to be zero as an argument that
> individuals shouldn't discount.
Chris is being misleading. One of the points of that thread was
whether growth models made any sense. I'm quite happy to agree that
my contribution to that thread may have been in a different
direction than John Tyler's original intent. The non-authoritarian
will recognize that that's the nature of Usenet and discussion in
general. It's not as if I was trying to refute John Tyler by
discussing strawpersons and ignoring all substance in his post.
> Even Rob ought to admit those particular
> critiques of economics are silly, but, alas, Rob has never met a critique
> of economics he hasn't liked,...
I have made no comment on John Tyler's ideas about intertemporal
utility maximization. Chris seems to be so "autistic" as to think
economics = mainstream economics = neoclassical economics. My
discussions of critiques of neoclassical economics I find devastating
give no support for Chris' silliness.
Poor Chris Auld.
APPENDIX
Consider the accounting identity:
Y = w L + r K (A-1)
where Y is net national income, w is the wage, L is the labor force,
r is the rate of profits (interest), and K is the value of the
capital stock. Differentiate:
dY/dt = w dL/dt + L dw/dt + r dK/dt + K dr/dt (A-2)
Multiply by unity in some complicated form:
dY/dt = w L Lhat + w L what + r K Khat + r K rhat (A-3)
where xhat is the rate of growth of x:
xhat = (1/x) dx/dt (A-4)
From A-3:
yhat = (w L/Y) what + (r K/Y) rhat + (w L/Y) Lhat + (r K/Y) Khat (A-5)
Or
yhat = a(t) what + (1 - a(t)) rhat + a(t) Lhat + (1 - a(t)) Khat (A-6)
where:
a(t) = w L/Y (A-7)
and the notation emphasizes income shares is a function of time.
Assume that the rate of profits shows no trend (rhat = 0). Thus,
yhat = a(t) what + a(t) Lhat + (1 - a(t)) Khat (A-8)
Consider an arbitrary production function:
Y = F( L, K, t ) (A-9)
Differentiate:
dY/dt = dF/dt + dF/dL dL/dt + dF/dK dK/dt (A-10)
Or:
dY/dt = dF/dt + L dF/dL Lhat + K dF/dK Khat (A-11)
Or:
yhat = (1/F) dF/dt + alpha Lhat + (1 - alpha) Khat (A-12)
where
alpha = ( L dF/dL )/F (A-13)
Compare A-6 or A-8 and A-12. Conclude factor shares equal
estimated output elasticities by definition.
Assume income shares are constant. From A-8:
yhat = a what + a Lhat + (1 - a) Khat (A-14)
Assume the rate of growth of wages is a constant:
yhat = lambda + a Lhat + (1 - a) Khat (A-15)
Integrate:
ln Y = A + lambda t + a ln L + (1 - a) ln K (A-16)
Or:
Y = A exp( lambda t ) L^a K^(1-a) (A-17)
So the stylized facts explain how a fit of the Cobb-Douglas
production function follows from an accounting identity.
Wages are about 75% of total national income. Since lambda is
then 0.75 what, total factor productivity growth (i.e., the
growth in output per worker not due to increases in inputs of
capital per worker) must be about three-quarters of the growth
in output per worker.
Robert Vienneau
(Mark Patrick Witte) wrote:
> In article <rvien-EE9F0B....@news.dreamscape.com>,
> Robert Vienneau <rv...@see.sig.com> wrote:
> >In article <9j5o8m$jbu$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
> >(Mark Patrick Witte) wrote:
> >> Just a reminder...
> >> [ And more silliness. ]
> >> >Perhaps he does not understand the theory. His refusual to recognize
> >> >a connection between Sraffa and Leontief certainly suggests as
> >> >much.
> >> On no, the connection is well known. Sraffa is to Leontief
> >> as...footnote is to magnum opus.
> >So Mr. Witte concedes he was mistaken in denying that Leontief
> >input output analysis provides an empirical implementation of
> >Sraffa's work by Mr. Witte's standards.
> I deny that the magnum opus in question is an empirical
> implementation of the footnote in question.
> Specifically, what are the empirical implications of Sraffa's work?
If Mr. Witte wanted to have or was capable of having an intellectually
serious conversation - all the evidence on this newsgroup is against
it - I suggest the burden would be on him to clarify how he thinks
input output analysis relates to the models in Sraffa's book.
Consider the subtitle of Sraffa's book. It suggests that Mr.
Witte's question is not the right metric to evaluate that book.
"But self-absorption and consistent policy error are just two of
the endemic problems of the leading American economists, and not
even the most serious among them. The deeper problem is the nearly
complete collapse of the prevailing economic theory--of the
structure of thought that supports their policy ideas. It is a
collapse so complete, so pervasive, that the profession can only
deny it by refusing to discuss theoretical questions in the first
place."
-- James Galbraith
http://www.prospect.org/print/V11/7/galbraith-j.html
Robert Vienneau
(Mark Patrick Witte) wrote:
--
Robert Vienneau
(Mark Patrick Witte) wrote:
> In article <3B61...@MailAndNews.com>,
> John Raymond Tyler <JRT...@MailAndNews.com> wrote:
> >>>
> >>>It has been some time since I read Solow, which cropped up in this
> >>>discussion.
> >>> As I recall, he makes two elementary mistakes in calculus in deriving
> >>> the
> >>>equation to be tested. It is true that they tend to cancel each
> >>>other, but
> >>>his work is fatally flawed.
> >> Yes, yes, and the editors at the QJE are drooling idiots, almost as
> >>dumb as Frank Ramsey.
> >I am afraid that an argument from authority doesn't carry much weight,
> >and I
> >have no idea how Frank Ramsey got into this.
> The Ramsey comment recalls a post from some time back with a similar
> claim. It's hard to do anything but argue from authority against vague
> claims about how a professor of statistics made to elementary mistakes in
> calculus yet got them by the editors of the QJE.
Yes, it's hard to imagine a response like:
"What mistakes are you talking about?"
Or
"Are you refering to the problems that follow from
Solow's use of a discrete approximation to the rate
of growth?"
Anyway, Mr. Witte seems to have an idealized picture of the
referreeing process. Here's some empirical results to contrast:
<http://www.econ.duke.edu/~erw/Preprints/JHETFinal.html>
Mark Patrick Witte
calculus that eluded an MIT stats professor, the editors of the QJE, and
myself, yet were caught by someone who does not understand how imputed
depreciation is usually handled in such models?"
And this poster replies with nothing.
In article <rvien-F47CCF....@news.dreamscape.com>,
Mark Patrick Witte
but at least didn't use any text to do it. Very efficient!
In article <rvien-38831A....@news.dreamscape.com>,
Mark Patrick Witte
Sraffa's work?"
And this poster replies with nothing, other than a quote of someone
whining about assumptions.
In article <rvien-9BA6FF....@news.dreamscape.com>,
Christopher Auld
>(Christopher Auld) wrote:
> Chris fails to mention that he's already taken a position on this
>argument in his ranting about Scott Moss last year. If he were
>to accept McCombie's argument, he would have to admit much of what
>he said then was misdirected or mistaken. So the comment about
>lacking "strong priors" is either misleading or a reflection of
>a lack of understanding of what he's talking about.
Scott Moss' arguments only somewhat overlap the methodological issues in
growth econometrics raised by McCombie. Which is beside the point, since
my comments are still applicable to my reading of the relevant literature
last year. (I wonder if Rob is ever going to give up comments like he
"lacks understanding of what he's talking about." As usual, I am more
than happy to let readers decide which one of us knows of what he speaks.)
>> First, if we limit ourselves to papers published since 1964, this
>> literature comprises a handful of papers published in outlets ranging
>> from
>> minor mainstream journals (e.g., Applied Economics) to obscure radical
>> journals (e.g., JPKE and JRPE).
>
> 1964 is totally arbitrary. McCombie says in his first paragraph that
>his argument is a continuation of the argument in these papers:
[ Snip: list of a couple of papers from the 50s in major journals
and "a handful of papers published in outlets ranging from minor
mainstream journals...."]
I'd like to thank Rob for taking the trouble to type in all that stuff
verifying my comment. I'd also like to point out that anyone can find
numerous examples of empirical pieces drawing on the aggregate production
framework simply by randomly opening major general journals or macro field
journals and randomly picking papers.
>> Since none of these journals are
>> particularly widely circulated or influential, this branch of criticism
>> isn't really "well known" at all, much less constituting such widespread
>> and damning criticism as to justify Rob's initial histrionics.
>
> Right. Nobody pays attention to arguments among "Nobel" laureates,
>the QJE, and RES.
... in the late fifties and early sixties. Since then, Solow's method has
sparked several major and ongoing research paradigms, and earned him a
Nobel prize.
> It's quite clear why Chris would have a problem with
>the very existence of ROPE.
?
> In my opinion, anybody unfamiliar with the issues raised in this
>literature is hard to take seriously in discussions of whether
>Solovian growth theory is empirically well-founded (just as anybody
Rob is, of course, free to hold and express whatever opinion he likes.
However, it certainly isn't clear why anyone should take his opinion on
issues he doesn't understand seriously. I notice he correctly hasn't
tried to defend his "just make the aggregate production function
Leontief" fix (even though that was the obvious solution if one deeply
understands the issues). One might think such a profound error would
cause Rob to adjust his beliefs on whether he actually has a clue what
this literature is all about, or maybe crack open an intermediate
theory text, but no.
> Suppose I were to accept Chris' claim that
>this literature is not well-known. What conclusion should I draw?
>On the other hand, what conclusion should I draw from this
>literature being well-known, but ignored.
If Rob were actually interesting in trying to figure out how the
world works, or at least reading about others attempts at that
vexing goal, he'd probably actually read some of the thousands of
pieces of empirical and theoretical economics that have been
written drawing on Solow's seminal contributions.
In another thread, Rob blasts Mark Witte for not asking Tyler why
he thinks Solow made "elementary mathematical errors," which would
seem a fair point until one observes that Mr. Tyler himself operates
under elementary delusions and it is unlikely, although possible,
he's right in this case. Notice it never occurs to Rob to ask
"Why, Auld and Witte, and indeed thousands of professional economists
across the globe, do you not agree with this line of criticism?"
Notice it never occurs to Rob to read any of the actual applied
work drawing on Solow's contribution. Instead, he leaps to a
conclusion, then proceeds to lecture the economists on, oddly, what
economists know.
>> Second, is McCombie right?
>
> Since Chris doesn't come close to summarizing the point of this paper,
>I'll have to do it.
Oh no.
> Solovian growth theory has been used extensively in applied work,
>particularly in assigning the causes of growth to technical
>change and increases in capital per worker. One can find regressions
>of growth rates and such like in this literature. For example,
>Solow (1957) reports five regressions of different forms of
>production functions, with correlation coefficients ranging
>from 99.64% to 99.96%. The question addressed here is to what
>extent this literature can be said to have tested aggregate
>neoclassical theory, the aggregate marginal productivity theory
>of distribution, and so on.
1. The econometric methodologies used in the relevant literatures
have advanced a little since 1957. Incidentally, much of
McCombie's critique doesn't apply, for that reason, to modern
studies.
2. The question is NOT to "what extent this literature can be
said to have TESTED aggregate...." This is a elementary
error explicitly addressed by Solow himself nearly 45 years
ago.
3. As Witte commented, many of these issues are not unique to
this issue but rather just issues in time series econometrics
generally. This includes the 99+% R^2s, which are not signs
of a good model, quite the opposite.
[ snip: paraphrasing of McCombie's paper. ]
>Anyways, part of the point is that Fisher came to the opposite
>conclusion in his own solo papers. Fisher discovered through
>simulation what Shaikh later showed analytically: good fits
>of aggregate production functions using standard techniques
>are the consequence of stable income shares. Economists
>ought to work on developing better techniques and ought to
>be aware that good reason has been given to think Solovian
>growth theory is not empirically supported.
Yes, given the stylized facts describing the U.S. post-War macroeconomy,
the simple regression of lnQ on lnL and lnK will yield a good fit, etc.
This is a prediction of the model, in fact, it's one reason *why* the
Cobb-Douglas form is popular. Unfortunately, the "standard techniques"
Rob refers to haven't been "standard" in some time. Again, modern
implementations typically use some variant of Generalized Method of
Moments estimation and more flexible forms in order overcome these
problems.
Overarching it all is the simple observation that assuming the existence
of an aggregate production function is simultaneously a grave error and a
powerful technique. Every economist knows that it isn't "true" (which is
one reason McCombie et al's comments on aggregation issues are just dull),
it's rather a tractable simplification of an intractable world. Everyone
knows that under, to borrow Rob's phrase, "rigorous economic theory," this
stuff is technically invalid. As in every *model*, the question isn't
whether it's "true," it's whether it's useful. Writing papers that do
nothing but challenge assumptions without offering anything useful is, in
Witte's evocative phrasing, wanking. This observatio has particular
bite when everyone already knows why the assumptions are wrong.
> The non-authoritarian
Considering all Rob has done in this thread, as usual, is act in an
authoritarian fashion, but without enough grasp of the literature to
realize his authorities are, well, not, this would seem a fine place to
exit.
susupply
"Robert Vienneau" <rv...@see.sig.com>
apparently having relocated to a city where the baseball team is 19 games
out of 1st place,
wrote in message news:rvien-9BA6FF....@news.dreamscape.com...
> In article <9j9dam$q81$1...@news.acns.nwu.edu>, mwi...@merle.acns.nwu.edu
> (Mark Patrick Witte) wrote:
> > Specifically, what are the empirical implications of Sraffa's work?
>
> If Mr. Witte wanted to have or was capable of having an intellectually
> serious conversation - all the evidence on this newsgroup is against
> it - I suggest the burden would be on him to clarify how he thinks
> input output analysis relates to the models in Sraffa's book.
>
> Consider the subtitle of Sraffa's book. It suggests that Mr.
> Witte's question is not the right metric to evaluate that book.
In other words, there's no there in that there book?
Patrick
Mark Patrick Witte
so often comes in these sorts of exchanges with the likes of Bob
Veinneau or Jay Hanson.
Recall that RV claimed that: "It is well-established that
one-sector growth models are deeply questionable as useful and
empirically validated tools for help in providing insight into
actual economies. See two most recent issues of Journal of Post
Keynesian Economics.)
Chris Auld and I point out that actually the opposite is
true. While there are certainly well known technical and empirical
problems to doing macroeconomic research, generally the Solovian
approach to growth accounting has been found to be remarkably
useful by the authors who publish in the most influential journals.
Chris points out that far from it being "well-established", to find
much challenge to this approach, we must "limit ourselves to
papers published since 1964, this literature comprises a handful
of papers published in outlets ranging from minor mainstream
journals (e.g., Applied Economics) to obscure radical journals
(e.g., JPKE and JRPE)." Veinneau evidently agrees but puts the
date at which this debate dies in the general journals as recently
as 1974. (In this post, I am not commenting on the validity of
this debate, just upon how it was incorrectly characterized as
being widely accepted important.)
Now in an otherwise vacuous post today, Veinneau
gives a link to an article by Roy Weintraub and Ted Gayler
about how the work on GE by Arrow and Debreu gained
acceptance. The authors write:
"How did it come to pass that a particular paper,
in a journal at that time read by very few economists, came
to be accepted as having established a foundational truth
about market economics?"
Exactly how obscure was this journal, Econometrica?
Its president in 1944-1945 was a back-bencher named John
Maynard Keynes, its executive committee in 1956 had a
membership that included Samuelson, Koopmans, and Stone.
That year it published work by Mitchell, Bellman, Allais,
Tinbergen, Cowles, Dorfman, Alchian, Solow, Clower, Machlup,
Hahn, Morgenstern, Frish, Muth, Neyman, Roy, Baumol, and
many others.
My, how standards vary!
(I have two observations about the Journal of
Post-Keynesian Economics. I see that a former frequent
sci.econ poster, Warren Mosler, is on the board. I also wonder
what is up with their webpage and that of their publisher. Did
they forget to pay their domain bill?
http://www.mesharpe.com/pke_2303.htm)
In article <rvien-F47CCF....@news.dreamscape.com>,
Edward Flaherty
Mark Patrick Witte wrote:
> (I have two observations about the Journal of
> Post-Keynesian Economics. I see that a former frequent
> sci.econ poster, Warren Mosler, is on the board.
In half-hearted defense of the JPKE, having Mosler on their board
doesn't necessarily mean they know nothing about
economics. Jaques Jaikaran, author of "Debt Virus"
and tax evasion convict, was on the board of a Texas
bank. Having an idiot on its board didn't mean the
bank didn't know banking, nor did it mean the idiot
knew banking.
--
Edward Flaherty
On-line since 1993
E-mail: flahe...@earthlink.net
Web site: http://members.home.net/flaherty15/index.htm
Mark Patrick Witte
that I thought would be of interest to sci.econ people. Warren Mosler is
not an academic but is certainly a very intelligent and creative fellow with
a lot of experience in financial markets.
In article <3B5C7D07...@earthlink.net>,
William F Hummel
Patrick Witte) wrote:
> No, no! This was not an attack on the JPKE, just an observation
>that I thought would be of interest to sci.econ people. Warren Mosler is
>not an academic but is certainly a very intelligent and creative fellow with
>a lot of experience in financial markets.
Yes, and he probably has a net worth over $100 million, the
result of his expertise.
Mark Patrick Witte
Well...does this mean that I should take international trade advice
from Ross Perot?
William F Hummel
Patrick Witte) wrote:
>In article <djhpltkbrfc0feptm...@4ax.com>,
>William F Hummel <wfhu...@mediaone.net> wrote:
>>On 23 Jul 2001 23:41:47 GMT, mwi...@merle.acns.nwu.edu (Mark
>>Patrick Witte) wrote:
>>
>>> No, no! This was not an attack on the JPKE, just an observation
>>>that I thought would be of interest to sci.econ people. Warren Mosler is
>>>not an academic but is certainly a very intelligent and creative fellow with
>>>a lot of experience in financial markets.
>>
>>Yes, and he probably has a net worth over $100 million, the
>>result of his expertise.
>
> Well...does this mean that I should take international trade advice
>from Ross Perot?
Who is Ross Perot?
John Raymond Tyler
comment
about the relevance of prediction? A clever astrologer might well come up
with a formula that predicts eclipses very well. It may call for the
sacrifice of a virgin at the right time, but what the hell, it predicts
beautifully. Prediction can never be the sole criteria for the acceptance
of
a theory.
John Tyler
------------------------------------------------------------
Get your FREE web-based e-mail and newsgroup access at:
http://MailAndNews.com
Create a new mailbox, or access your existing IMAP4 or
POP3 mailbox from anywhere with just a web browser.
------------------------------------------------------------
Mark Patrick Witte
John Raymond Tyler <JRT...@MailAndNews.com> wrote:
>>On 24 Jul 2001 02:30:52 GMT, mwi...@merle.acns.nwu.edu (Mark
>>Patrick Witte) wrote:
>>
Recall that "prediction" in the sense of a model's ability to match
moments of the data is not the same as "forecasting" the future. But, pray
tell, what is the "sole criteria (sic) for the acceptance of a theory?"
Robert Vienneau
(Christopher Auld) wrote:
> Robert Vienneau <rv...@see.sig.com> wrote:
> >(Christopher Auld) wrote:
> > Chris fails to mention that he's already taken a position on this
> >argument in his ranting about Scott Moss last year. If he were
> >to accept McCombie's argument, he would have to admit much of what
> >he said then was misdirected or mistaken. So the comment about
> >lacking "strong priors" is either misleading or a reflection of
> >a lack of understanding of what he's talking about.
> Scott Moss' arguments only somewhat overlap the methodological issues in
> growth econometrics raised by McCombie. Which is beside the point, since
> my comments are still applicable to my reading of the relevant literature
> last year.
Actually, this literature is one of Scott Moss' points. Chris has
"strong priors".
> >> First, if we limit ourselves to papers published since 1964, this
> >> literature comprises a handful of papers published in outlets ranging
> >> from
> >> minor mainstream journals (e.g., Applied Economics) to obscure radical
> >> journals (e.g., JPKE and JRPE).
> > 1964 is totally arbitrary. McCombie says in his first paragraph that
> >his argument is a continuation of the argument in these papers:
> [ Snip: list of a couple of papers from the 50s in major journals
> and "a handful of papers published in outlets ranging from minor
> mainstream journals...."]
Perhaps Chris can list two articles in my list with publication dates
in the 50s.
It's bizarre to describe a certain former oil company executive
as "radical". And of course the label has nothing to do with the
validity of arguments advanced in a journal which he edits.
> I'd like to thank Rob for taking the trouble to type in all that stuff
> verifying my comment. I'd also like to point out that anyone can find
> numerous examples of empirical pieces drawing on the aggregate production
> framework simply by randomly opening major general journals or macro
> field
> journals and randomly picking papers.
Simply noting somebody using Solow's growth accounting approach
in "empirical" work does not vitiate the argument in these papers.
Chris would go on and on if I suggested any natural inference
from Chris' curious notions of what's relevant to "point out".
> >> Since none of these journals are
> >> particularly widely circulated or influential, this branch of
> >> criticism
> >> isn't really "well known" at all, much less constituting such
> >> widespread
> >> and damning criticism as to justify Rob's initial histrionics.
> > Right. Nobody pays attention to arguments among "Nobel" laureates,
> >the QJE, and RES.
> ... in the late fifties and early sixties.
Chris says economists are encouraged to be ignorant of the literature.
> Since then, Solow's method has
> sparked several major and ongoing research paradigms, and earned him a
> Nobel prize.
No, you don't say!
[ >> H. A. Simon, "On Parsimonious Explanations of Production ]
[ >> Relations", _Scandinavian Journal of Economics_, 1979. ]
[ >> According to McCombie (2000-2001), ]
[ >> Herb Simon mentioned this argument in his "Nobel" prize acceptance ]
[ >> speech. ]
Chris has problems with the whole notion of linear time. And he
seems to think the mere passage of time makes an argument invalid.
Or, perhaps, his refutation of the argument in this literature seems
to be, as usual, a mere matter of counting noses - a rather obvious
fallacy.
> > In my opinion, anybody unfamiliar with the issues raised in this
> >literature is hard to take seriously in discussions of whether
> >Solovian growth theory is empirically well-founded (just as anybody
> Rob is, of course, free to hold and express whatever opinion he likes.
> However, it certainly isn't clear why anyone should take his opinion on
> issues he doesn't understand seriously. I notice he correctly hasn't
> tried to defend his "just make the aggregate production function
> Leontief" fix (even though that was the obvious solution if one deeply
> understands the issues). One might think such a profound error would
> cause Rob to adjust his beliefs on whether he actually has a clue what
> this literature is all about, or maybe crack open an intermediate
> theory text, but no.
I think I'll leave the above silliness here.
There was no such error there. Chris failed to understand my point
and misrepresented my text. (I wonder if Chris is ever going to give
up comments like "Rob [should] adjust his beliefs on whether he
actually has a clue what this literature is all about, or maybe
crack open an intermediate theory text, but no." As usual, I am more
than happy to let readers decide which one of us knows of what he
speaks.)
> > Suppose I were to accept Chris' claim that
> >this literature is not well-known. What conclusion should I draw?
> >On the other hand, what conclusion should I draw from this
> >literature being well-known, but ignored.
> [ Silliness deleted ]
> In another thread, Rob blasts Mark Witte for not asking Tyler why
> he thinks Solow made "elementary mathematical errors,"
> [ ...more silliness - deleted. ]
I never "blast"ed Mark Witte on that point.
> >> Second, is McCombie right?
> > Since Chris doesn't come close to summarizing the point of this paper,
> >I'll have to do it.
> Oh no.
Notice Chris chooses not to say anything about simulation in this
post.
> > Solovian growth theory has been used extensively in applied work,
> >particularly in assigning the causes of growth to technical
> >change and increases in capital per worker. One can find regressions
> >of growth rates and such like in this literature. For example,
> >Solow (1957) reports five regressions of different forms of
> >production functions, with correlation coefficients ranging
> >from 99.64% to 99.96%. The question addressed here is to what
> >extent this literature can be said to have tested aggregate
> >neoclassical theory, the aggregate marginal productivity theory
> >of distribution, and so on.
> 1. The econometric methodologies used in the relevant literatures
> have advanced a little since 1957. Incidentally, much of
> McCombie's critique doesn't apply, for that reason, to modern
> studies.
Notice Chris does not say how these advancements have surmounted
McCombie's critique. (McCombie is aware that Solow's particular
method of estimating technical growth, A(t), is rarely used.)
There is some hint of the ghost of an argument below in Chris'
comments about Generalized Methods of Moments.
> 2. The question is NOT to "what extent this literature can be
> said to have TESTED aggregate...." This is a elementary
> error explicitly addressed by Solow himself nearly 45 years
> ago.
The questioned addressed by McCombie's paper is to what extent
this literature can be said to have tested aggregate neoclassical
theory. Solow's intentions are irrelevant to whether that is the
question which McCombie is addressing.
> 3. As Witte commented, many of these issues are not unique to
> this issue but rather just issues in time series econometrics
> generally. This includes the 99+% R^2s, which are not signs
> of a good model, quite the opposite.
Actually, according to McCombie, they are issues that arise
when one uses value data to estimate underlying "engineering"
relations. McCombie mentions some other cases than Solovian
growth accounting where these issues arise. Strangely enough,
I fail to see how the arising of similar issues in other
contexts makes these issues unimportant.
I guess Chris thinks it is well-known that there are
theoretical problems in interpreting what empirical
results there are supporting Solovian growth theory.
About those correlation coefficients over 99% - it would
seem Chris is saying that Solow's growth accounting paper
contained a technical error. At least, I read Solow's text
as saying these correlation coefficients are signs of
a good model.
> >Anyways, part of the point is that Fisher came to the opposite
> >conclusion in his own solo papers. Fisher discovered through
> >simulation what Shaikh later showed analytically: good fits
> >of aggregate production functions using standard techniques
> >are the consequence of stable income shares. Economists
> >ought to work on developing better techniques and ought to
> >be aware that good reason has been given to think Solovian
> >growth theory is not empirically supported.
> Yes, given the stylized facts describing the U.S. post-War macroeconomy,
> the simple regression of lnQ on lnL and lnK will yield a good fit, etc.
> This is a prediction of the model, in fact, it's one reason *why* the
> Cobb-Douglas form is popular.
And there we have it. My understanding of the mathematics here
leads me to conclude some economists have perverse norms if they
accept that as a reason for using the Cobb-Douglas.
Chris could always look up a dictionary article on humbug to see what
the prediction of a model with Leontief fixed coefficient production
functions under neutral technological change would be under any
theory of distribution yielding stable income distribution.
> Unfortunately, the "standard techniques"
> Rob refers to haven't been "standard" in some time. Again, modern
> implementations typically use some variant of Generalized Method of
> Moments estimation and more flexible forms in order overcome these
> problems.
Of course, McCombie gave references to how his article applies to
more flexible functional forms.
> Overarching it all is the simple observation that assuming the existence
> of an aggregate production function is simultaneously a grave error and a
> powerful technique. Every economist knows that it isn't "true" (which is
> one reason McCombie et al's comments on aggregation issues are just
> dull),
> it's rather a tractable simplification of an intractable world. Everyone
> knows that under, to borrow Rob's phrase, "rigorous economic theory,"
> this
> stuff is technically invalid. As in every *model*, the question isn't
> whether it's "true," it's whether it's useful.
I guess Chris agrees that it is well-known that Solovian growth
theory is without theoretical foundation.
> Writing papers that do
> nothing but challenge assumptions...
> [ Further silliness deleted ]
Poor Chris Auld.
Christopher Auld
>(Christopher Auld) wrote:
>Actually, this literature is one of Scott Moss' points.
Scott Moss claimed (claims?) that Solow "agreed in substance"
with Shaikh's critique in building an argument that economists
are "intellectually dishonest" for still using Solow's method
after even Solow had disowned it. Anyone, regardless of
whether they know anything about economics, can glance at Solow's
rejoinder to Shaikh in the RES and see quite plainly that Solow
did not "agree in substance" with the critique. That argument
was over what Solow wrote not about which one of Shaikh or Solow
was right per se.
>It's bizarre to describe a certain former oil company executive
>as "radical". And of course the label has nothing to do with the
>validity of arguments advanced in a journal which he edits.
Of course not. Once again, Rob's claim was that this literature
is "well known." Pointing out that this literature is, in fact,
less than two dozen papers (half of which are by the same couple
of authors) over a forty year period, most of which are in obscure
journals and rarely cited, is a perfectly valid objection.
>> I notice he correctly hasn't
>> tried to defend his "just make the aggregate production function
>> Leontief" fix
>There was no such error there. Chris failed to understand my point
>and misrepresented my text.
Well, the floor's all yours, Rob. Here's your opportunity
to sketch your methodology, one that's "obvious" to anyone who
understands the issues yet which has never been employed, one
which evades all of the problems brought up by McCombie et
al.
>Solow's intentions are irrelevant to whether that is
>the question which McCombie is addressing.
All right, so McCombie is still addressing a misunderstanding
cleared up when Eisenhower was President?
>I guess Chris
>thinks it is well-known that there are theoretical problems in
>interpreting what empirical results there are supporting Solovian
>growth.
Of course there are. There are well-known theoretical problems
in all of macroeconomics, for that matter. This is all taught
in undergrad theory courses.
>About those correlation coefficients over 99% - it would seem
>Chris is saying that Solow's growth accounting paper contained a
>technical error. At least, I read Solow's text as saying these
>correlation coefficients are signs of a good model.
I am saying that time series econometrics has advanced since 1957.
>> Yes, given the stylized facts describing the U.S. post-War
>> macroeconomy, the simple regression of lnQ
>> on lnL and lnK will yield a good fit, etc. This is a prediction
>> of the model, in fact, it's one reason *why* the
>> Cobb-Douglas form is popular.
>And there we have it. My understanding of the mathematics here
>leads me to conclude some economists have perverse norms if they
>accept that as a reason for using the Cobb-Douglas.
?
I think Rob's "understanding of the mathematics here" may be a
tad flawed.
[ in Rob's phrasing, "Silliness deleted." ]
Why not humour us, Rob, and actually read a paper drawing on
Solow's method to tackle some real-world problem? If nothing
else, it would give you further insight into the evil and
incompetent world of academic economics, and amuse you while
you work on your paper outlining the "obvious" method you've
developed which doesn't succumb to McCombie et al's critique.
How about Grossman and Helpman on innovation or Hansen's
seminal paper on GMM?