SHOULD WAGES RISE WITH PRODUCTIVITY
Bill Cooper
between productivity and wages. It has just never made any sense to
me.
If I run a machine and my employer goes out and buys a brand new
one that does twice as much, but requires me to do the same thing, why
should he pay me more to run it. After all, the number of people who
are qualified to run the machine isn't any less than it used to be
(the supply of labor hasn't decreased). In fact, if a lot of
employers who use the same type of machine all upgrade there may even
be an oversupply of labor which would drive wages (the price of labor)
down.
Is there a more sophisticated version of the wage/productivity
link that makes more sense. After all, countries with higher
productivity do pay higher wages, so the rule does work (more or less)
for comparing across economies. Perhaps, more productivity (i.e.
greater capital investment) gives greater scope to investments in
worker productivity (education) which in turn commands higher wages.
Maybe some of the Econ cogniscenti out there can set me straight.
Bill Cooper
FxX+FyY=pF(X Y)
what you need to know.
price of a good = number of labor required to produce a unit of good * wage
rate
this of course assumes perfect competition. if price is > number of labor
required* wage rate; there is profit to be earned and more firms will
enter the industry and reap those profits until price=number of labor*wage
rate.
so if price stays constant, and productivity increases; (number of labor
required to produce a unit of good goes down); wage will increase.
another explaination is the specific model
marginal productivity of labor* price = wage rate;
MPP.P = wage rate
again this does not assume that the hiring firm is a monopsony; and we
are living in a competitive economy.
when MPP increases; productivity increases, wage rate goes up when price is
constant.
this is the explantion on why productivity goes up with wages.
however, in non competitive case; it is true that a monopoly can afford
to give higher wages if productivity increases. but will it do so?
probably very very reluctantly. there will involve some bargaining.
for elegant treatments; try some game theory books on wage bargaining.
they are usually more interesting than the classical assumption of
'competitive economy'.
William F. Hummel
: I've read and heard, forever, it seems, that there is a link
---------------------------
In my opinion, there is no direct link. In the short run the business
owner (capitalist) will reap the reward of a better machine. In the long
run, wages will rise but not necessarily in proportion to the improved
productivity. How far depends on various factors: collective bargaining
(unions), social pressure, goodwill, and simply intelligent economics.
The more insightful capitalist will recognize that his own profits depend
upon a working class that has the money to buy his product.
William F. Hummel
Julien Maisonneuve
|> I've read and heard, forever, it seems, that there is a link
|> between productivity and wages. It has just never made any sense to
|> me.
However, for productivity gained through people, it is understandable that a
rasing wage is used as an incentive to increase personal productivity, or to
favour training.
Julien.
Markku Stenborg
writes:
> If I run a machine and my employer goes out and buys a brand new
>one that does twice as much, but requires me to do the same thing, why
>should he pay me more to run it. After all, the number of people who
>are qualified to run the machine isn't any less than it used to be
>(the supply of labor hasn't decreased). In fact, if a lot of
>employers who use the same type of machine all upgrade there may even
>be an oversupply of labor which would drive wages (the price of labor)
>down.
Yes, wages should rise w/ productivity, if by productivity you mean
*marginal productivity*. The rule that maximizes the profits of the firm
is to set wage (W) = marginal productivity of labor (MPL); more
specifically, in competitive environment where the W is a given parameter
to the employer, hire workers until W = MPL. (BTW, marginal productivity
== the increase in revenue generated by the last unit [eg, hour] of
labor.)
In "real worl", the wage depends on the relative bargainig power of the
employer and the worker; the wage is set between two bounds: W = MPL is
lower bound, and W = what ever a monopolist-worker would charge the
employer the upper bound. Both of these bounds go up w/ an increrase in
productivity.
> Is there a more sophisticated version of the wage/productivity
>link that makes more sense. After all, countries with higher
>productivity do pay higher wages, so the rule does work (more or less)
>for comparing across economies. Perhaps, more productivity (i.e.
>greater capital investment) gives greater scope to investments in
>worker productivity (education) which in turn commands higher wages.
Yes, there are a many links. Food for your thoughts: Suppose the profit
of the firm depends on the effort you put in the job (think of a CEO
deciding how much to work vs play golf), and your wage depends on the
firm's success. Then, the more you are paid the more you work, and the
higher the profit [up to some limit]. Or, in developing world, your
effort depends on your food intake which depends on your wage. In both
cases, higher wage leads to higher effort and bigger profits.
Markku Stenborg mar...@utu.fi
Take my advice, I have no use for it
Bill Cooper
>well, the ricardian model can be used to prove
>what you need to know.
Thanks very much for responding.
Well let me see if I understand what you're saying.
The new machine that I work on has a higher return on investment than
the old Machine (thst's why my employer invested in it). Other
capatilist's realize the same thing and also invest in the new
machines, which creates a greater demand for operators, which drives
up wages.
Obviously, the industry reaches a new equilibrium with a new price and
a new overall production level. How do you know that the new higher
productivity level will be large enough to make up for the
productivity gain (decrease in the labor that is needed for each
unit)? If my output went up from one widget an hour to two w/h, but
consumption only went up by 50%, then the number of operators needed
would decrease, leading to a decrease in wages. Of course, it's also
possible that consumption could go up by 150% which would result in
higher wages too.
Maybe the problem is that I.m concentrating to much on just one
position. Even though the wages for the operator go down maybe wages
for other people (manufactuers of the machine, persons involves in
distributuion) go up enough to compensate. I'm not sure why there is
a necessary connection.
Thanks again
Bill Cooper
Bill Cooper
>|> I've read and heard, forever, it seems, that there is a link
>|> between productivity and wages. It has just never made any sense to
>|> me.
>What you say about productivity gained with machines is probably true.
>However, for productivity gained through people, it is understandable that a
>rasing wage is used as an incentive to increase personal productivity, or to
>favour training.
Aren't most increases in procuctivity related to an improvement in
technology?
certainly an improved education, combined with new technology can
result in higher wages, but the education without a productive process
that makes use of the education isn't going to change much.
Thanks for responding.
Bill Cooper
Bill Cooper
>:Bill Cooper said
>: Is there a more sophisticated version of the wage/productivity
>: link that makes more sense. After all, countries with higher
>: productivity do pay higher wages, so the rule does work (more or less)
>: for comparing across economies. Perhaps, more productivity (i.e.
>: greater capital investment) gives greater scope to investments in
>: worker productivity (education) which in turn commands higher wages.
>: Maybe some of the Econ cogniscenti out there can set me straight.
>---------------------------
>In my opinion, there is no direct link. In the short run the business
>owner (capitalist) will reap the reward of a better machine. In the long
>run, wages will rise but not necessarily in proportion to the improved
>productivity. How far depends on various factors: collective bargaining
>(unions), social pressure, goodwill, and simply intelligent economics.
>The more insightful capitalist will recognize that his own profits depend
>upon a working class that has the money to buy his product.
>William F. Hummel
So you don't think that the free enterprise/market/capitalist system
will deliver higher wages on it's own? Yet you think that higher
wages are a good thing? Why are you in favor of capitalism at all?
I think of free market as an efficient algorythm (and no I don't have
a really clear understanding of what I mean by that). Society gives
it a set of inputs and the market delivers an optimal allocation of
capital to satisfy society's needs. If you don't like the results you
must have used the wrong inputs (by inputs I mean things like
antitrust or attaching costs to pollution or patent laws). So go back
and change them till you get the results you want. If you try to
dictate the outcome of the process you introduce tremendous
inefficiencies.
If you don't like markets what would you replace them with?
Bill Cooper
Bill Cooper
Thanks for responding to my post, Markku.
>Yes, wages should rise w/ productivity, if by productivity you mean
>*marginal productivity*. The rule that maximizes the profits of the firm
>is to set wage (W) = marginal productivity of labor (MPL); more
>specifically, in competitive environment where the W is a given parameter
>to the employer, hire workers until W = MPL. (BTW, marginal productivity
>== the increase in revenue generated by the last unit [eg, hour] of
>labor.)
I think that I understand this. But does my original example
necessariliy mean that the MPL will go up? I see measurements of
productivity (i.e. the U.S. Natrional productivity went up 3% last
year). These measurments are about widgets made per hour of labor,
not MPL (I think, tell me if I'm wrong).
>In "real worl", the wage depends on the relative bargainig power of the
>employer and the worker; the wage is set between two bounds: W = MPL is
>lower bound, and W = what ever a monopolist-worker would charge the
>employer the upper bound. Both of these bounds go up w/ an increrase in
>productivity.
I'm trying to connect you're comment about MPL with the idea of the
market determining the price of labor. If the price of labor (i.e.
wages) is the result of the intersection between the demand and the
supply of/for a particular set of skills and abilities. In my
original example I stipulated that the skills and abilities necessary
to run this new machine (technology or whatever) . Assume that when
the new technology is completely integrated into the market that the
demand for labor decreases (admittedly a big if). The same supply of
labor is chasing a smaller demand so wages should decrease.
In your MPL frame of reference, when the new machine is installed the
MPL goes up, then as other manufactuers replace their old machines
with the new, more efficient ones (nad other manufactuers just stop
produciing altogether) prices fall and the MPL goes down again (I
think)?
>> Is there a more sophisticated version of the wage/productivity
>>link that makes more sense. After all, countries with higher
>>productivity do pay higher wages, so the rule does work (more or less)
>>for comparing across economies. Perhaps, more productivity (i.e.
>>greater capital investment) gives greater scope to investments in
>>worker productivity (education) which in turn commands higher wages.
>Yes, there are a many links. Food for your thoughts: Suppose the profit
>of the firm depends on the effort you put in the job (think of a CEO
>deciding how much to work vs play golf), and your wage depends on the
>firm's success. Then, the more you are paid the more you work, and the
>higher the profit [up to some limit]. Or, in developing world, your
>effort depends on your food intake which depends on your wage. In both
>cases, higher wage leads to higher effort and bigger profits.
Sure, HIgher wages can result in greater producivity, but this effect
is pretty limited. People in advanced counties are many time more
productive than people in undeveloped countries This is due to
better technology, not better motivation.
>Markku Stenborg mar...@utu.fi
>Take my advice, I have no use for it
Thanks
Bill Cooper
FxX+FyY=pF(X Y)
for jobs elsewhere that offers them higher wages if you refuse to pay them
more.
but its hard to know who are those high productive workers so they
usually opt for higher education as a signal thta they are productive.
it's in one of those game theory papers, i can't recall which one.
i found it will studying adverse selection and moral hazard.
Edward Vytlacil
In a competitive equillibrium with complete information,
wages will equal their marginal product. And, of course,
the rents on the capital they use will equal the marginal
product of the capital.
If there is incomplete information in regards to the
quality of the worker then, depending on
how the model is formulared, a seperating or a pooling equillibrium
will result. In a seperating equillibrium, wages will still
equal marginal productivity. In a pooled equillibrium wages
will equal a weighted average of workers' marginal productivities.
Above you refer to a seperating equillibrium, and it should
be noted that in a seperating equillibrium wages will equal
marginal productivity despite workers being forced to engage
in costly signalling (education).
If there is incomplete information regarding a workers effort
but output is observed, a principal agent problem results. If effort
is observed with an impertect technology, an efficiency wage
model (a la Shapiro and Stiglitz) results. If effort is observed
but matching workers to firms is costly, a matching model (a la
Pissarides) results. In all these cases, wages do not
equal marginal product but are a function of the workers marginal product.
My point of all this is there is a multitude of alternative
models that can be study how wages are determined. In every model
that I have seen in mainstream economics, wages are either
equal to the worker's marginal productivity or directly related
to marginal productivity.
Ed Vytlacil
Darren Howard Lubotsky
capital rises - not the employee's labor productivity. That is, your
wage should not rise.
In another post someone argued to the affect that price equals amound of
labor required times the wage rate. Did you forget about demand? This
is a derivative of the labor theory of value - the basis of most
classical economics. Since at least Alfred Marshall in 1871 we have
realized that the price of goods reflects the interaction of both supply
and demand.
Darren Lubotsky
UC-Berkeley
Dept. of Economics
Bill Cooper (wfco...@tiac.com) wrote:
: I've read and heard, forever, it seems, that there is a link
: between productivity and wages. It has just never made any sense to
: me.
: If I run a machine and my employer goes out and buys a brand new
: one that does twice as much, but requires me to do the same thing, why
: should he pay me more to run it. After all, the number of people who
: are qualified to run the machine isn't any less than it used to be
: (the supply of labor hasn't decreased). In fact, if a lot of
: employers who use the same type of machine all upgrade there may even
: be an oversupply of labor which would drive wages (the price of labor)
: down.
: Is there a more sophisticated version of the wage/productivity
: link that makes more sense. After all, countries with higher
: productivity do pay higher wages, so the rule does work (more or less)
: for comparing across economies. Perhaps, more productivity (i.e.
: greater capital investment) gives greater scope to investments in
: worker productivity (education) which in turn commands higher wages.
: Maybe some of the Econ cogniscenti out there can set me straight.
: Bill Cooper
--
_____________________________
Darren Lubotsky
Department of Economics
University of California, Berkeley
lubo...@uclink2.berkeley.edu
William F. Hummel
: My point of all this is there is a multitude of alternative
: models that can be study how wages are determined. In every model
: that I have seen in mainstream economics, wages are either
: equal to the worker's marginal productivity or directly related
: to marginal productivity.
---------------
Continual reference to "mainstream economics" (which no one has yet
defined on sci.econ) reminds me of Phil Gramm's constantly telling us
what the American public wants or believes.
So all the models employ a direct relation between wages and worker's
marginal productivity. Does that mean such a direct relation in fact
exists? Or is it just easier to build (tractable) models of that sort?
One should be able to make a convincing argument in words alone the
existence of such a direct relation, or lack thereof.
William F. Hummel
Rob Vienneau
relationship between wages and productivity on the macroeconomic
level, a level not to be addressed by partial equilibrium models. In
the post-war U.S., at least, increases in productivity were reflected
in increased wages. Demand kept pace with output during this long
boom.
But in the last few decades, wages have stagnated, even when
productivity increased. Although I do not think anybody has yet
formulated a comprehensive answer to why things have come unhinged,
tentative suggestions can be offered.
I consider this very much a political result. We have seen a
counter-revolutionary response to the 60s, and an attack on working
people and democratic institutions. (E.g. consider Reagan's attack on
the air traffic controllers' union, the strikes that became struggles
over the issue of mere union survival, and the decision of the
"quality" press to no longer have reporters with a labor beat.) Union
membership has decreased dramatically, and wages stagnate. Cost of
Living Agreements (COLAs) cover a smaller part of the workforce, and
"mark-up" pricing prevails in an ever smaller part of the economy.
International trade is another important aspect of this change. In
less developed countries, agriculture often exhibits disguised
unemployment. A steady stream of formerly rural workers are available
to industry. It as if the tendency in industrialized countries to move
production off-shore has converted a Post Keynesian distribution
process to a Classical one in which wages are given, however
productivity increases.
Analytical models demonstrate that there's no hard and fast
determinate function from technology to distribution. Piero Sraffa has
shown how to construct a system of long run prices, given
technology. Distribution cannot be determined within the same
framework and either wages or the rate of profits/interest must be
given from outside the model. (Rents of no-longer-produced outdated
capital goods and nonexhaustible natural resources ("land") are
determined within the model.)
Luigi Pasinetti, with his work on vertical integration and structural
dynamics, has shown how to relate input-output models to productivity
changes. I like to think of productivity changes within the following
conceptualization. Collectively, the labor force works a certain
percentage of the year to produce the goods they consume. The
remainder goes to profit. If the the time needed to produce the
current wage basket decreases, that increase in productivity can
result in an increased wage basket, more leisure (a decrease in the
working day), or more profit. Which will happen is not determinate.
As Frank Hahn has shown in "The neo-Ricardians," one can reinterpret
this math in terms of marginal productivity theory. Not a single
formal proposition in Sraffa's book is inconsistent with marginal
productivity theory, according to Hahn. But rigorous theory does not
permit claims that wages are determined by [the value of] the marginal
product of labor. That proposition is confined to vulgar political
economy in which social relations are confused with natural ones.
Robert Vienneau Voice: +1-315-734-3671
Kaman Sciences Corporation FAX: +1-315-734-3699
258 Genesee Street Email: rvie...@utica.kaman.com
Utica, NY 13502 USA r...@utica.kaman.com
Mirjam Fritsch
Mirjam Fritsch
rises, either output alone rises, or (if both outout and input rise) it
rises more than input.
The input in the example consists of capital (= machine) and labour. The
output are widgets (say: 1/hour).
In the example Cooper presented rising input (a new and better machine).
So the input didn't rise on the labour side. He even adds, that the
machine requires him to do the same thing. So in the example the rise in
productivity depends only upon the machine, not upon labour at all (so it
is no use talking about an increased productivity of labour, education
etc.) The output increases (e.g. 2 w/h).
At first sight it would seem that the cost of the capital (= e.g.
interest, for the loan for purchasing the new machine) rises, but not the
cost of labour (wage).
Still I believe there can be a link between rising productivity and wage
rate.
A rise in productivity means, that the cost for one widget produced on
the new machine is lower than the cost for one widget produced on the old
one.
Now the relation between cost and price depends on the market you are on.
If the owner of the plant can keep the old price of one widget equal to
the new price, he can (after having paid everything related to the new
machine) reap the rewards of the improoved technology - and that's what
he's most likely to do, I don't believe in goodwill in this case.
So: no increase in wages.
But if others could join in the market and produce the same widgets on
the new machines because they ,smell" the high rewards, setting the price
just a little bit lower, the price on the whole market must fall to the
point, where no newcomers will join in the market anymore (price equals
marginal cost plus market entry cost).
In an oligopoly with blocked market entry, the price would probably fall
less, leaving higher rewards.
If the market is narrow (there is not much unsatisfied demand that would
buy the products even at a significantly lower price, at least not enough
to make profits out of the lower price) and the original price was
considered high already before introducing the new machine, the owner might
want/need to lower prices in order to be able to sell anything at all.
In this case, the price will lower to maintain the competitiveness against
other producers, or: to survive.
But: If the demand level in the beginning was low because of high prices
and if the prices fall, then the overall market demand will raise in
terms of number of widgets. Now if the productivity increase through the
new machines is smaller than the increase in demand given the new lower
price, and market entry is not deterred, then you could argue that new
plants will join in the market, requiring new labour, that is able to work
on the old or on the new machines (since it makes no difference). So the
demand for labour rises. If there are any unemployed in this field, they
will get employment at the current wage rate. If then there is still need
for more labour to satisfy the demand, only then wages will rise.
Best regards,
Mirjam Fritsch
William F. Hummel
: Analytical models demonstrate that there's no hard and fast
: determinate function from technology to distribution. Piero Sraffa has
: shown how to construct a system of long run prices, given
: technology. Distribution cannot be determined within the same
: framework and either wages or the rate of profits/interest must be
: given from outside the model. (Rents of no-longer-produced outdated
: capital goods and nonexhaustible natural resources ("land") are
: determined within the model.)
-----------------
When you say "analytic models demonstrate that...", it reminds me of the
oft used phrase "computer analysis shows that...". I'm sure you didn't
mean it that way, but such language betrays an unwarranted respect for
analytic models. The following quote says it better than I can:
"The discipline of [neo-classical] economics has so far successfully
resisted all efforts to alter its character as an exercise in how to
reason deductively from axiomatic principles. That is, it has
insisted on remaining the Euclidian geometry of the social sciences.
This Cartesian position has not been without its advantage to the
economist themselves. They have been known to remark, "We travel with a
light tool kit." By this, they mean that economic theorists have not
had to burden themselves much with factual detail. They have been
content to reason a priori -- and hence their preference for elegance
over relevance."
William F. Hummel
Henry Choy
: I've read and heard, forever, it seems, that there is a link
: between productivity and wages. It has just never made any sense to
: me.
: If I run a machine and my employer goes out and buys a brand new
: one that does twice as much, but requires me to do the same thing, why
: should he pay me more to run it.
If you just run the machine, you don't deserve any more, but if you
figure out how to improve output, then you should get a raise.
--
Henry Choy "Math class is hard" - Barbie
e-mail: ch...@cs.usask.ca "Stupid is as stupid does." - Mrs. Gump
Henry Choy
: well, the ricardian model can be used to prove
: what you need to know.
: price of a good = number of labor required to produce a unit of good * wage
: rate
Does "number of labor" include machines? Does "wage rate" include the
cost of running machinery?
: this of course assumes perfect competition. if price is > number of labor
: required* wage rate; there is profit to be earned and more firms will
: enter the industry and reap those profits until price=number of labor*wage
: rate.
: so if price stays constant, and productivity increases; (number of labor
: required to produce a unit of good goes down); wage will increase.
If the productivity increases, the employer can pay more to
machines. In other words, the employer invests in technology. That
way, people who don't do any more labor don't get paid more.
Henry Choy
: In my opinion, there is no direct link. In the short run the business
: run, wages will rise but not necessarily in proportion to the improved
: productivity. How far depends on various factors: collective bargaining
: (unions), social pressure, goodwill, and simply intelligent economics.
: The more insightful capitalist will recognize that his own profits depend
: upon a working class that has the money to buy his product.
Why can't a capitalist do best by paying as small a wage as possible
and charging as high a price as possible?
Consider a closed capitalist system where every capitalist does just
that. I'm talking about a free market where everyone is absolutely
greedy. If someone wants something, s/he has to pay for it
somehow. The money can be obtained in many ways (e.g., selling to
society, borrowing, receiving a gift or grant). In this system,
wouldn't the insightful capitalist recognize that his/her profits
depend on selling enough products to get enough money to buy what s/he
wants?
Henry Choy
: Well let me see if I understand what you're saying.
: the old Machine (thst's why my employer invested in it). Other
: capatilist's realize the same thing and also invest in the new
: machines, which creates a greater demand for operators, which drives
: up wages.
Why a greater demand for operators? Wasn't the original question based
on the ability to produce more per operator?
: Obviously, the industry reaches a new equilibrium with a new price and
: a new overall production level. How do you know that the new higher
: productivity level will be large enough to make up for the
: productivity gain (decrease in the labor that is needed for each
: unit)?
What do you mean "make up for"??
The hypothesis is a doubling of productivity.
: If my output went up from one widget an hour to two w/h, but
: consumption only went up by 50%, then the number of operators needed
: would decrease, leading to a decrease in wages. Of course, it's also
: possible that consumption could go up by 150% which would result in
: higher wages too.
How does the argument for this go?
We have no idea whether the cost of a widget changes. Suppose the cost
does not, initially. Demand is not entirely dependent on cost. Then we
have widgets that come out twice as fast, and quitting time is a lot
earlier. If demand doesn't rise much (say 50% at most), quitting time
is still early so wages fall. However, if demand grows more than the
speed of the machines, the workers have to work longer and need to be
paid more.
: Maybe the problem is that I.m concentrating to much on just one
: position. Even though the wages for the operator go down maybe wages
: for other people (manufactuers of the machine, persons involves in
: distributuion) go up enough to compensate. I'm not sure why there is
: a necessary connection.
The poor suckers who build the machines are going to want their money.
FxX+FyY=pF(X Y)
>When the new machine is purchased, the productivity of the employer's
>capital rises - not the employee's labor productivity. That is, your
>wage should not rise.
explain then:
L=labor K=capital
r= return to capital
w= wage
p=price of good (assumed constant)
P=profit
max profit = price * total output - cost of inputs
L
P = p*Q(K,L) - wL - rK
dP = p dQ - w
-- ---
dL dL
at stationary point, dp/dL=0
therefore w = p dQ(k,L)
-------
dL
using implicit function theorem,
w = - p.(dK/dL)(dQ/dK)
since dQ/dK is the productivity of capital, wage is a function of productivity
of capital.
actually, i used to just apply MRP=w until i was trying to refute
an earlier arguemtn and someone mentioned about giving the extra profit
to capital. so i decided to re derive the whole optimisation and came
to this conclusion. any mistakes or wrong interpretatoin,please notify me. i'll
be grateful.
Edward Vytlacil
William F. Hummel <wfhu...@netcom.com> wrote:
>Edward Vytlacil (vac...@cicero.spc.uchicago.edu) wrote:
>
>: My point of all this is there is a multitude of alternative
>: models that can be study how wages are determined. In every model
>: that I have seen in mainstream economics, wages are either
>: equal to the worker's marginal productivity or directly related
>: to marginal productivity.
>---------------
>Continual reference to "mainstream economics" (which no one has yet
>defined on sci.econ) reminds me of Phil Gramm's constantly telling us
>what the American public wants or believes.
>
>William F. Hummel
Well, "mainstream" seemed like the most neutral term I could think
of. What I really meant was economics as practiced by almost all
academic economists. As differentiated from, say, the various
alternative schools of thought which dominate sci.econ and which
almost no modern academic economist would recognize as being economics
or have every heard of.
Ed Vytlacil
FxX+FyY=pF(X Y)
>FxX+FyY=pF(X Y) (lbt) wrote:
>: well, the ricardian model can be used to prove
>: what you need to know.
>
>: price of a good = number of labor required to produce a unit of good * wage
>: rate
>
>Does "number of labor" include machines? Does "wage rate" include the
>cost of running machinery?
no. the ricardian model assumes that labor is the only cost and the value
of the labor determines the price, p=unit labor requiremnt*wage rate
the purpose of bringing up the ricardian model is not to support labor
theory of value or anything but to provide insight on why marginal productivity
of labor increases can lead to wage increase.
>
>
>: this of course assumes perfect competition. if price is > number of labor
>: required* wage rate; there is profit to be earned and more firms will
>: enter the industry and reap those profits until price=number of labor*wage
>: rate.
>
>: so if price stays constant, and productivity increases; (number of labor
>: required to produce a unit of good goes down); wage will increase.
>
>If the productivity increases, the employer can pay more to
>machines. In other words, the employer invests in technology. That
>way, people who don't do any more labor don't get paid more.
>
maybe you are not assuming perfect competition. in perfect competition
wages will be competed up to p.mpp=w.
net benefit from hiting labor = revenue from hiring L labor - cost of hiring L
labor
max,
L
net benefit = revenue - wL
0 = d revenue
------ -w
d L
w = marginal revenue of labor=p.mpp in PC
the same goes for capital
i think so.
Mark Uffen
In article <40dbe2$9...@sundog.tiac.net>, Bill Cooper (wfco...@tiac.com) writes:
> I've read and heard, forever, it seems, that there is a link
>between productivity and wages. It has just never made any sense to
>me.
> If I run a machine and my employer goes out and buys a brand new
>one that does twice as much, but requires me to do the same thing, why
>are qualified to run the machine isn't any less than it used to be
>(the supply of labor hasn't decreased). In fact, if a lot of
>employers who use the same type of machine all upgrade there may even
>be an oversupply of labor which would drive wages (the price of labor)
>down.
> Is there a more sophisticated version of the wage/productivity
>link that makes more sense. After all, countries with higher
>productivity do pay higher wages, so the rule does work (more or less)
>for comparing across economies. Perhaps, more productivity (i.e.
>greater capital investment) gives greater scope to investments in
>worker productivity (education) which in turn commands higher wages.
>
> Maybe some of the Econ cogniscenti out there can set me straight.
>
>Bill Cooper
>
>
firm. When a new machine with greater productivity is installed,
assuming that an improved competitive position results in higher
potential profitability, then the benefits can go to consumers
(through reduced price for the output), shareholders (through
increased dividends or enhanced share value through profit
retention), the IRS through a higher tax bill on increased profits,
and to those who work for the firm as salaries and wages.
With the exception of the IRS (and ignoring creative
accounting), the decision on the distribution of the benefits are
made by management in the firm. Their pricing decision will
probably result in higher profits (perhaps at a reduced price and/or
margin). They will wish to enhance their own earnings, but must
bear in mind that the shareholders can vote them out of office,
and they depend on the workforce to secure the enhanced output.
Even if they lay some workers off, they still depend on
those who remain. If the workforce does not have much bargaining
power (e.g. because of high unemployment, lack of unionisation,
limited public social pressure on management actions), then they can
choose not to increase wages and expect to get away with it.
Otherwise, it is in their own self interest to increase wages.
The surplus pool of labour arising from technical progress
may or may not apply. This is more of a macroeconomic question,
depending on whether enough extra jobs are created by the spending
of the benefits of the technical progress. Sometimes the bad
effects on workers can be magnified if old skills are no longer
required (e.g. auto assembly line workers such as welders replaced
by robots), or if a large number of people are thrown out of work
in a locality (the auto industry is also a good example of this).
William F. Hummel
: Well, "mainstream" seemed like the most neutral term I could think
: of. What I really meant was economics as practiced by almost all
: academic economists. As differentiated from, say, the various
: alternative schools of thought which dominate sci.econ and which
: almost no modern academic economist would recognize as being economics
: or have every heard of.
I think it will come as a great surprise to modern academic economists
to hear that most "practice" the same economics, whatever that means.
Are you implying that the neo-classical and neo-Keynesian schools are
birds of a feather? Would you say Barro and Blinder practice the same
economics?
William F. Hummel
FxX+FyY=pF(X Y)
w = - p.(dK/dL)(dQ/dK) -------------(1)
by symmetry,
r= -p (dL/dK)(dQ/dL)
therefore:
w/r = dK/dL
substituing back into (1) yields
r = -p (dQ/dK)
similarly,
w= -p (dQ/dL)
the w and r cancels out respectively!
therefore proof is completed that increase in productivity of capital
does not affect wage rate and increase in productivity of labor has
no effect on r.
Henry Choy
: ch...@skorpio3.usask.ca (Henry Choy) wrote:
: >FxX+FyY=pF(X Y) (lbt) wrote:
: >: well, the ricardian model can be used to prove
: >: what you need to know.
: >
: >: price of a good = number of labor required to produce a unit of good * wage
: >: rate
: >
: >Does "number of labor" include machines? Does "wage rate" include the
: >cost of running machinery?
: no. the ricardian model assumes that labor is the only cost and the value
: of the labor determines the price, p=unit labor requiremnt*wage rate
: the purpose of bringing up the ricardian model is not to support labor
: theory of value or anything but to provide insight on why marginal productivity
: of labor increases can lead to wage increase.
Is the model -->--> correct <--<-- if machines are not counted?
: >: this of course assumes perfect competition. if price is > number of labor
: >: required* wage rate; there is profit to be earned and more firms will
: >: enter the industry and reap those profits until price=number of labor*wage
: >: rate.
: >
: >: so if price stays constant, and productivity increases; (number of labor
: >: required to produce a unit of good goes down); wage will increase.
: >
: >If the productivity increases, the employer can pay more to
: >machines. In other words, the employer invests in technology. That
: >way, people who don't do any more labor don't get paid more.
: >
: maybe you are not assuming perfect competition. in perfect competition
: wages will be competed up to p.mpp=w.
How can the investment by one firm in more productive machinery be
observed by other firms? In the original problem statement there is no
indication that there are observable effects. Are there some unstated
assumptions that we are reasoning with?
If there are no observable effects, the other firms might not make any
changes. Then wages do not change.
Henry Choy
: upon further work, i found out why i reached that 'funny' conclusiong of:
: w = - p.(dK/dL)(dQ/dK) -------------(1)
: by symmetry,
: r= -p (dL/dK)(dQ/dL)
: therefore:
: w/r = dK/dL
: substituing back into (1) yields
: r = -p (dQ/dK)
: similarly,
: w= -p (dQ/dL)
: the w and r cancels out respectively!
: therefore proof is completed
There is one point that needs to be made clear:
B4I sqrt(u)qt(Pi)ru
-------------------
16
:that increase in productivity of capital
: does not affect wage rate and increase in productivity of labor has
: no effect on r.
Robert Vienneau
>>: [...] In every model
>>: that I have seen in mainstream economics, wages are either
>>: equal to the worker's marginal productivity or directly related
>>: to marginal productivity.
[...]
>Well, "mainstream" seemed like the most neutral term I could think
>of. What I really meant was economics as practiced by almost all
>academic economists. As differentiated from, say, the various
>alternative schools of thought which dominate sci.econ and which
>almost no modern academic economist would recognize as being economics
>or have every heard of.
Edward seemed to have been using the above proposition to argue for
a determinate relationship between changes in productivity and movements
in wages. If this is something that every "modern academic economist"
recognizes, so much the worst for academia. Although I have learned
economics from self-study, I can assure you that there are many first
rate economists in universities throughout the world who would reject
such a relationship, or at least any argument for it on the sole basis
of marginal productivity theory. (I don't necessarily think of myself as
a partisan of "alternative schools of thought which *dominate* sci.econ"
[emphasis added].)
"FxX+FyY=pF(X Y)" <lbt>'s mathematics can be ignored. His conclusions
are false in a multicommodity world, as I have explained at length before,
and his attempt to foist this nonsense off on Ricardo has no scholarly
foundation.
Robert Vienneau Voice: 1-315-734-3671
Kaman Sciences Corporation FAX: 1-315-734-3699
258 Genesee Street Email: rvie...@utica.kaman.com
Utica, NY 13502 USA r...@utica.kaman.com
"If I had to sum up the twentieth century, I would say that it raised
the greatest hopes ever conceived by humanity, and destroyed all
illusions and ideals." -- Yehudi Menuhin
FxX+FyY=pF(X Y)
>observed by other firms? In the original problem statement there is no
>indication that there are observable effects. Are there some unstated
>assumptions that we are reasoning with?
no. i assumed perfect competiion model where
number of firms--->infinity therefore perceived marginal revenue
=(1/n)MR + (1+1/n) P(X)
where MR is true marginal revenue
and P(X) is demand
therfoer when n--->infinity the perceeived mearginal revenue converges
to the demand curve.
also, perfect comp. assumes complete information.
everything i do, you know, and everything you do, i know.
it also assumes homogenous product. what you produce, is exactly the
same thing as what i produce.
and some more...
if you don't assume this, which makes it more like real world;
then you are right in saying that
FxX+FyY=pF(X Y)
rvie...@utica.kaman.com wrote
"FxX+FyY=pF(X Y)" <lbt>'s mathematics can be ignored. His conclusions
are false in a multicommodity world, as I have explained at length before,
and his attempt to foist this nonsense off on Ricardo has no scholarly
foundation.
may i add
it is false only if firms don't behave as:
max profit = price * total output - cost of inputs
L
and......
and it is definitely plausible in real world that
firms can't behave this way. eg. threats of strike by labor etc.
if the way i formulated the problem was correct, the answer must be correct
(except for the negative sign which i forgot to cross out at the end) unless
it was a mathematical error.
to model it more towards real world you may try L(P) where suply
of labor is a funciton of the profit or other similar cases (work
satisfaction L(Utilty from work) or L(prestige) etc..
in my model i assumed L as an independent variable so utility from
work or any other thing does not affect the supply of labor to the firm.
further note..
let's not make a big deal out of the ricardian model. everyone knows
that it has some flaws.
Mark Witte
>: Well, "mainstream" seemed like the most neutral term I could think
>: of. What I really meant was economics as practiced by almost all
>: academic economists. As differentiated from, say, the various
>: alternative schools of thought which dominate sci.econ and which
>: almost no modern academic economist would recognize as being economics
>: or have every heard of.
>I think it will come as a great surprise to modern academic economists
>to hear that most "practice" the same economics, whatever that means.
>Are you implying that the neo-classical and neo-Keynesian schools are
>birds of a feather? Would you say Barro and Blinder practice the same
>economics?
>
>William F. Hummel
I'd say that Barro and Blinder are much alike in that they both use
optimization based, marginal utility models and very similar econometric
tools.
FxX+FyY=pF(X Y)
Still I believe there can be a link between rising productivity and wage
rate.
you are quite right in saying that. i tried to remodel the optimisation
problem closer to real world by assuming wage and rate of return, w and r
as dependent variables , w(L) and r(K).
P=profit
p=price
Q=quantity of output
K=capital L=labor
w = wage
max by choosing K and L
P=p.Q(K,L) - w(L).L - r(K).K
s.t. no constraints
F.O.C=
dP/dL= p.dQ/dL - (dw/dL . L + w(L) ) = 0
w(L) = p.dQ/dL - dw/dL . L
by symmetry
r(K) = p.dQ/dL - dr/dK. K = 0
applying implicit function theorem:
w(L) = - p.dK/dL.dQ/dK - dw/dL.L ------(1)
unlike in the previous problem here, i can't divide the equations to
find the marginal rate of substitution of capital for labor.
equation (1) is the best i could simplify which implies
there is a possiblity that productivity of capital can affect
wage rate. unless of course dK/dL and dw/dL contains a w term
that will allow w to be cancelled out.
ah! which reminds me of euler
------------------
another explanation :
if however, firm is commited to a certain output Q, price p and capital employed K, and production
is a homogeneous function of degree 1, then
Q(K,L) = dQ/dK . K + dQ/dL . L
if Q is constant, when dQ/dK increases, and firm insists on keeping K
constant, tehn the term dQ/dL.L must go down.
if
w = p. dQ/dL
multiplying L on both sides:
wL = p.dQ/dL . L
therefore
dQ/dL.L = w.L/p must go down
if firm again insist on price staying at a certain level,
then the term w.L must go down.
therefore given the above complcated assumptions, any increase in productivity
of capital is a threat for workers since either wage w goes down, or
labor employed L goes down or a little of both.