Wages and Productivity: They Meet at the Margin
rv...@dreamscape.com
>d...@inforamp.net (David Lloyd-Jones) writes:
>>The tradeoff between consumption and investment is an arithmetical
>>tautology which exists only in statics.
>
>The tradeoff between consumption today and capital goods -- and hence
>consumption tomorrow -- is neither tautological nor does it "exist only in
>statics." If more resources are devoted to producing capital goods, less
>resources have to be devoted to producing something else.
Chris has cleverly changed the question. See my essay
ftp://csf.colorado.edu/econ/authors/Vienneau.Robert/sraffa2.html
for a clarification, especially Section 7.5, "Interest as A Reward for
Waiting." Chris' assertion may be true, but it does not easily translate
into a statement about "investment." Even its truth value can be argued.
Since Keynes, many may have doubted that interest is paid because "capital"
(finance) is scarce. When unemployment exists and industry operates below
capacity (i.e. almost always), more capital goods can be created without
giving up current consumption goods.
>Lloyd-Jones better quickly write up his novel idea that no such tradeoff
>exists, as all the mainstream journals are stubbornly sticking to the idea
>that savings rates may affect investment. One can only wonder where the
>resources required for investment come from if not, ultimately, foregone
>consumption?
See, here Chris changes back to the original question. Is he confusing
a tautological accounting identity with an assertion about the world?
I suspect he is wrong about what *all* mainstream articles assert. The
pre-Keynesian idea that the proper method of increasing investment
is to increase savings rates may still be questioned here and there
(assuming I'm reading Chris correctly).
Wherever informed economic theory is taught (Samuelson), economists know
there is a difference between statements about savings and investment and
statements about physical quantity flows. Pre-Keynesian ideas still have
logical difficulties.
>My
>claim that, and let's get it right, in the long run, increases in
>productivity will be result in increases in wages is theoretically sound
>and has very strong empirical support...
>
>To make this crystal clear: I assert that if the value of output were to
>rise 1%, in the long-run, all else equal, we would observe the mean wage
>rise roughly 1%. On the other hand, if wages were to rise 1% for reasons
>unrelated to productivity, I am hard pressed to come up with a mechanism
>that would push the value of output up. That is, at an aggregate level,
>the causation runs only one way.
Productivity in the US may have slowed in growth since about 1970, but
it has still grown as Chris is measuring it. Empirically, this supposed
tie between productivity changed.
Furthermore, Chris has yet to even outline a mechanism in theory to
connect an increase in productivity to increase wages. At the highest
level of abstraction, a shift outward in the wages-rate of profit frontier
("factor price frontier") need not be associated with a transverse to any
particular long run equilibrium point in which either wages, the rate of
profits, or both have increased. Who gains that increase cannot be
determined from production conditions alone.
So Chris, what is your mechanism?
Robert Vienneau Whether strength of body or of mind, or
rv...@future.dreamscape.com wisdom, or virtue, are always found...in
proportion to the power or wealth of a man
[is] a question fit perhaps to be discussed
by slaves in the hearing of their masters,
but highly unbecoming to reasonable and
free men in search of the truth.
-- Rousseau
Mark Witte
<rv...@dreamscape.com> wrote:
>Chris Auld writes:
>
>>d...@inforamp.net (David Lloyd-Jones) writes:
>>>The tradeoff between consumption and investment is an arithmetical
>>>tautology which exists only in statics.
>>
>>The tradeoff between consumption today and capital goods -- and hence
>>consumption tomorrow -- is neither tautological nor does it "exist only in
>>statics." If more resources are devoted to producing capital goods, less
>>resources have to be devoted to producing something else.
>
>Chris has cleverly changed the question. See my essay
>
> ftp://csf.colorado.edu/econ/authors/Vienneau.Robert/sraffa2.html
>
>for a clarification, especially Section 7.5, "Interest as A Reward for
>Waiting." Chris' assertion may be true, but it does not easily translate
>into a statement about "investment." Even its truth value can be argued.
>Since Keynes, many may have doubted that interest is paid because "capital"
>(finance) is scarce. When unemployment exists and industry operates below
>capacity (i.e. almost always), more capital goods can be created without
>giving up current consumption goods.
But does this make new capital costless? No. Further, as long
as capital has a positive marginal product, it will be scarce.
rv...@dreamscape.com
Chris Auld writes:
>(Aside to Vienneau: "investment" is generally a term used to refer to
>production of capital goods; my argument in no way depends on Keynesian
>accounting identities.)
Whatever argument that is. Does Chris remember that I have asked him
several times now how productivity drives wages? Anyways...
Case 1. Investment refers to physical quantities such as the number of
blast furnances of a given quality, tons of steel, etc. Then one cannot
evaluate how much consumption today is traded off for consumption tomorrow.
Capital goods are not measured in the same units as consumption goods.
Case 2. Investment is measured in the same units as consumption goods.
Then one is implicitly assuming a certain price structure, including
wages. It seems one would have in mind an accounting identity here,
not a physical relationship.
I have worked out an example illustrating Pasinetti's strictures on
these topics at:
ftp://csf.colorado.edu/econ/authors/Vienneau.Robert/sraffa2.html
Especially 7.5. The logic explained there also implies why measuring
productivity as the ratio of the value of output to labor inputs
requires knowledge of prices, including wages (Figure 7-4). This was
Ricardo's problem. Mark Witte may be interested in knowing that an
extension of this sort of model shows that different qualities of "land"
cannot generally be ordered be ordered from high rent to low rent based on
some physical measure of "fertility." The order of rentability may vary
with distribution and the resulting prices. See (Pasinetti 1980) - full
reference in Sraffa3.ps at the above site. Likewise, with heterogeneous
labor, I doubt one can generally map from physical measures of
"productivity" to wages.
I think Chris Auld correct (but for the wrong reason) in suggesting
any competently refereed journal would not publish (without additional
work) David Lloyd-Jones' idea that a trade-off between consumption and
investment applies at best to statics, not dynamics. I'm not at all
sure David has exhibited the necessary originality, and I suspect he
would agree.
As an aside to an aside, I have trouble locating the logic of Mark
Witte's interjection:
RV: When unemployment exists and industry operates below
RV: capacity (i.e. almost always), more capital goods can be created without
RV: giving up current consumption goods.
MW: But does this make new capital costless? No. Further, as long
MW: as capital has a positive marginal product, it will be scarce.
In the neoclassical theory of competitive equilibrium, the equilibrium price
of a factor in excess supply is zero. Perhaps Mark is displaying some
common sense here in violation of his logic*. Which goods are non-free
("scarce") is indeed a property of the competitive solution, not a physical
given. But "capital," as opposed to individual capital goods, is not a
factor in the disaggregated theory and has no marginal product.
* An allusion to a passage in Keynes.
While I'm on the subject of the history of thought, what economist of his
own era did Schumpeter characterize as exhibiting the "Ricardian vice," and
what was that vice? Chris Auld is encouraged to answer, for the answer might
help him understand some of David's comments.
Robert Vienneau Whether strength of body or of mind, or
rv...@future.dreamscape.com wisdom, or virtue, are always found...in
proportion to the power or wealth of a man
[is] a question fit perhaps to be discussed
by slaves in the hearing of their masters,
but highly unbecoming to reasonable and
free men in search of the truth.
-- Rousseau
David Lloyd-Jones
>Chris Auld writes:
>>(Aside to Vienneau: "investment" is generally a term used to refer to
>>production of capital goods; my argument in no way depends on Keynesian
>>accounting identities.)
>Whatever argument that is. Does Chris remember that I have asked him
>several times now how productivity drives wages? Anyways...
Robert,
It finally dawned on me why Chris's arguments are totally out to
lunch. His famous starting point which I felt uncomfortable about
from my very first post on this now very boring subject was
"Productivity determines..."
Well, there probably does not exist a scientific proposition of the
form "X determines..." and there _certainly_ does not exist a valid
economic proposition of the form "X determines Y."
"A, B and C determine, plural, C" is conceivable, though it's a little
on the cocky side; but single cause one-way causation is just very
very powerfully one of the things that economics was invented to
replace.
Cheers,
-dlj.
Mark Witte
<rv...@dreamscape.com> wrote:
>
>Chris Auld writes:
>
>>(Aside to Vienneau: "investment" is generally a term used to refer to
>>production of capital goods; my argument in no way depends on Keynesian
>>accounting identities.)
>
>Whatever argument that is. Does Chris remember that I have asked him
>several times now how productivity drives wages? Anyways...
>
>blast furnances of a given quality, tons of steel, etc. Then one cannot
>evaluate how much consumption today is traded off for consumption tomorrow.
>Capital goods are not measured in the same units as consumption goods.
>
>Case 2. Investment is measured in the same units as consumption goods.
>Then one is implicitly assuming a certain price structure, including
>wages. It seems one would have in mind an accounting identity here,
>not a physical relationship.
Hmm, if only there was some way to "add up" the value of society's
output in a given year....
>
>I have worked out an example illustrating Pasinetti's strictures on
>these topics at:
>
> ftp://csf.colorado.edu/econ/authors/Vienneau.Robert/sraffa2.html
>
>Especially 7.5. The logic explained there also implies why measuring
>productivity as the ratio of the value of output to labor inputs
>requires knowledge of prices, including wages (Figure 7-4).
How about this measure of productivity?
Value of output/labor hours.
>This was
>Ricardo's problem. Mark Witte may be interested in knowing that an
>extension of this sort of model shows that different qualities of "land"
>cannot generally be ordered be ordered from high rent to low rent based on
>some physical measure of "fertility."
Oddly, Mark does not find this terribly interesting.
>The order of rentability may vary
>with distribution and the resulting prices. See (Pasinetti 1980) - full
>reference in Sraffa3.ps at the above site. Likewise, with heterogeneous
>labor, I doubt one can generally map from physical measures of
>"productivity" to wages.
Hence the complete lack of piece work?
>
>I think Chris Auld correct (but for the wrong reason) in suggesting
>any competently refereed journal would not publish (without additional
>work) David Lloyd-Jones' idea that a trade-off between consumption and
>investment applies at best to statics, not dynamics. I'm not at all
>sure David has exhibited the necessary originality, and I suspect he
>would agree.
>
>As an aside to an aside, I have trouble locating the logic of Mark
>Witte's interjection:
>
>RV: When unemployment exists and industry operates below
>RV: capacity (i.e. almost always), more capital goods can be created without
>RV: giving up current consumption goods.
>
>MW: But does this make new capital costless? No. Further, as long
>MW: as capital has a positive marginal product, it will be scarce.
Ah, the importance of context. Too bad so much is missing.
>
>In the neoclassical theory of competitive equilibrium, the equilibrium price
>of a factor in excess supply is zero.
Given variable utilization rates and title value as present value
of future returns, this would qualify as a CE.
Chris Auld
>Case 1. Investment refers to physical quantities such as the number of
>blast furnances of a given quality, tons of steel, etc. Then one cannot
>evaluate how much consumption today is traded off for consumption tomorrow.
>Capital goods are not measured in the same units as consumption goods.
And consumption goods are not measured in the same units as consumption
goods. Hence, no tradeoffs exist in the economy. Or is that what Rob is
really trying to argue? Recall the context: my innocent, or so I thought,
notion that if we are to produce more capital goods, we must give
something up, all else equal. Does Rob disagree with this idea, or is he
just seeking an excuse to talk about Sraffa?
>I have worked out an example illustrating Pasinetti's strictures on
>these topics at:
>
> ftp://csf.colorado.edu/econ/authors/Vienneau.Robert/sraffa2.html
Ah.
--
Chris Auld Department of Economics
Internet: au...@qed.econ.queensu.ca Queen's University
Office: (613)545-2264 Kingston, ON K7L 3N6
rv...@dreamscape.com
> How about this measure of productivity?
> Value of output/labor hours.
It depends on wages. See Ricardo (1821), Chapter 1, Section IV.
I'd give a more contemporary reference, but somehow Chris Auld has a
nonrational objection to the one I find most obvious. He seems to want
to use an Irving Fisher-like approach to reify a trade-off between
consumption and investment, but I'm not supposed to refer to a famous
discussion of that approach since then. I suppose I'd also get
criticized for referring to Schumpeter (1911) (English translation 1934)
for a discussion of how "capital" might be said to be created out of
thin air, given accomodating financial institutions, especially as one
of his students at Harvard later built on an unmentionble Italian
economist to illustrate how Schumpeter's ideas might relate to the
mathematics of nonlinear dynamics.
Mark Witte
<rv...@dreamscape.com> wrote:
>Mark Witte writes:
>
>> How about this measure of productivity?
>> Value of output/labor hours.
>
>It depends on wages. See Ricardo (1821), Chapter 1, Section IV.
No it doesn't. See _Economic Report of the President_ (1996), data
tables.
(Ricardo? You're trotting out the labor theory of value?)
>
>I'd give a more contemporary reference, but somehow Chris Auld has a
>nonrational objection to the one I find most obvious. He seems to want
>to use an Irving Fisher-like approach to reify a trade-off between
>consumption and investment, but I'm not supposed to refer to a famous
>discussion of that approach since then. I suppose I'd also get
>criticized for referring to Schumpeter (1911) (English translation 1934)
>for a discussion of how "capital" might be said to be created out of
>thin air, given accomodating financial institutions, especially as one
>of his students at Harvard later built on an unmentionble Italian
>economist to illustrate how Schumpeter's ideas might relate to the
>mathematics of nonlinear dynamics.
If you're intimidated by getting stuffed by Chris Auld again,
then don't don't worry about posting a more contemporary reference.
rv...@dreamscape.com
:::
::: How about this measure of productivity?
::: Value of output/labor hours.
:>
:>It depends on wages. See Ricardo (1821), Chapter 1, Section IV.
:
: No it doesn't. See _Economic Report of the President_ (1996), data tables.
:
: (Ricardo? You're trotting out the labor theory of value?)
No. Apparently you want to uphold the labor theory of value as some sort
of empirical regularity though - a position not absurd.
Markku Stenborg
>Mark Witte writes:
>
>> How about this measure of productivity?
>> Value of output/labor hours.
>
>It depends on wages. See Ricardo (1821), Chapter 1, Section IV.
Eh? Value of output = price of output * amount sold; Labor hours = how
many hours all workers have spent on total at work-place (or something
similar). I can't see where wages could enter? You can't be suggesting
Marx's labor theory of value, can you?
>I'd give a more contemporary reference, but somehow Chris Auld has a
No, this has nothing to do w/ the age of references.
>nonrational objection to the one I find most obvious. He seems to want
>to use an Irving Fisher-like approach to reify a trade-off between
>consumption and investment, but I'm not supposed to refer to a famous
>discussion of that approach since then. I suppose I'd also get
>criticized for referring to Schumpeter (1911) (English translation 1934)
>for a discussion of how "capital" might be said to be created out of
>thin air, given accomodating financial institutions, especially as one
>of his students at Harvard later built on an unmentionble Italian
>economist to illustrate how Schumpeter's ideas might relate to the
>mathematics of nonlinear dynamics.
Markku Stenborg <mar...@utu.fi>
Take my advice, I have no use for it
Key fingerprint = 0C D5 B6 5D E8 9E 01 C0 4C 8F 7A 60 A9 A7 BA B1
Mark Witte
<rv...@dreamscape.com> wrote:
>In his usual haphazard fashion, Mark Witte writes:
>:::
>::: How about this measure of productivity?
>::: Value of output/labor hours.
>:>
>:>It depends on wages. See Ricardo (1821), Chapter 1, Section IV.
>: No it doesn't. See _Economic Report of the President_ (1996), data tables.
>:
>: (Ricardo? You're trotting out the labor theory of value?)
>
>No. Apparently you want to uphold the labor theory of value as some sort
>of empirical regularity though - a position not absurd.
Well, yes it is. And as for wages being necessary for measuring
productivity, I guess you're agreeing they're not?
rv...@dreamscape.com
: In article <v01530500ada4501e1d7e@[206.64.128.195]>,
: <rv...@dreamscape.com> wrote:
:>In his usual haphazard fashion, Mark Witte writes:
:>:::
:>::: How about this measure of productivity?
:>::: Value of output/labor hours.
:>:>
:>:>It depends on wages. See Ricardo (1821), Chapter 1, Section IV.
:>:
:>: No it doesn't. See _Economic Report of the President_ (1996), data tables.
:>:
:>: (Ricardo? You're trotting out the labor theory of value?)
:>
:>No. Apparently you want to uphold the labor theory of value as some sort
:>of empirical regularity though - a position not absurd.
:
:Well, yes it is.
Since you apparently have no idea of the contents of the labor theory of
value, I find your uniformed and unargued opinion of no consequence.
: And as for wages being necessary for measuring productivity, I guess
: you're agreeing they're not?
No. I still assert your definition of productivity has a theoretical
dependence on wages. Whether this effect shows up only as noise in
the stats, I do not say.
Even you should be able to follow another dependence than the above. What
happens to (value of output)/(labor hours), when aggregated over several
industries, if the composition of demand is different for goods bought
out of wages and goods bought of other sources of income? Assume some
limiting land-like factor exists.
Mark Witte
<rv...@dreamscape.com> wrote:
>Mark Witte writes:
>: In article <v01530500ada4501e1d7e@[206.64.128.195]>,
>: <rv...@dreamscape.com> wrote:
>:>In his usual haphazard fashion, Mark Witte writes:
I am something of an expert on being haphazard, having read
Sraffa3.ps.
>:>:::
>:>::: How about this measure of productivity?
>:>::: Value of output/labor hours.
>:>:>
>:>:>It depends on wages. See Ricardo (1821), Chapter 1, Section IV.
>:>:
>:>: No it doesn't. See _Economic Report of the President_ (1996), data tables.
>:>:
>:>: (Ricardo? You're trotting out the labor theory of value?)
>:>
>:>No. Apparently you want to uphold the labor theory of value as some sort
>:>of empirical regularity though - a position not absurd.
>:
>:Well, yes it is.
>
>Since you apparently have no idea of the contents of the labor theory of
>value, I find your uniformed and unargued opinion of no consequence.
Well, here's a chance for you to display your empirical acumen.
Since you have begun to come around to the viewpoint that matching the
observed data is important, how about if you demonstrate the empirical
regularity that underlies the labor theory of value.
>
>: And as for wages being necessary for measuring productivity, I guess
>: you're agreeing they're not?
>
>No. I still assert your definition of productivity has a theoretical
>dependence on wages. Whether this effect shows up only as noise in
>the stats, I do not say.
>
>Even you should be able to follow another dependence than the above. What
Even me? Well, that is high praise. You know, I honed my
reading skills trying to find a charitable way of interpretting the
aforementioned Sraffa3.ps in a way that might allow it to be reconciled
with reasonable economic theory.
>happens to (value of output)/(labor hours), when aggregated over several
>industries, if the composition of demand is different for goods bought
>out of wages and goods bought of other sources of income? Assume some
>limiting land-like factor exists.
What a foolish comment. There are many ways of measuring
labor productivity and I mentioned one that does not depend upon wages.
It is a method well suited to comparing productivities between nations,
sectors, or firms. Your attempts to bring class struggle into this
are inane (and of dubious empirical relevance.)