Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Edward Prescott, Idiot

16 views
Skip to first unread message

Robert Vienneau

unread,
Oct 27, 2004, 5:24:42 AM10/27/04
to

"It's easy to get over $200,000 in income with two wage earners in a
household."
-- Edward Prescott
<http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott1
9.html>

--
Mostly economics: <http://www.dreamscape.com/rvien/#PublicationsForFun>
r c
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

sinister

unread,
Oct 27, 2004, 7:07:47 AM10/27/04
to

"Robert Vienneau" <rv...@see.sig.com> wrote in message
news:rvien-C00412....@news.dreamscape.com...

>
> "It's easy to get over $200,000 in income with two wage earners in a
> household."
> -- Edward Prescott
> <http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott1
> 9.html>

Did you see the thread I started, "Fine work by our new Nobelist"?

Igor

unread,
Oct 27, 2004, 2:09:25 PM10/27/04
to
sinister wrote:

> "Robert Vienneau" <rv...@see.sig.com> wrote in message
> news:rvien-C00412....@news.dreamscape.com...
>
>> "It's easy to get over $200,000 in income with two wage earners in a
>> household."
>> -- Edward Prescott
>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott1
>>9.html>
>
>
> Did you see the thread I started, "Fine work by our new Nobelist"?
>

I did and it was it obvious you never read the paper you were attacking.
You were asking question clearly answered in the paper.

Igor

unread,
Oct 27, 2004, 2:10:15 PM10/27/04
to
Robert Vienneau wrote:

> "It's easy to get over $200,000 in income with two wage earners in a
> household."
> -- Edward Prescott
> <http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott1
> 9.html>
>

Interesting the page can not be found. Did he really say this?

ro...@telus.net

unread,
Oct 27, 2004, 2:19:38 PM10/27/04
to
On Wed, 27 Oct 2004 18:09:25 GMT, Igor <jjwea...@houston.rr.com>
wrote:

Yes, and I showed, in our exchange, that the answers were idiotic.

-- Roy L

MS

unread,
Oct 27, 2004, 2:33:57 PM10/27/04
to

Igor

unread,
Oct 27, 2004, 4:33:56 PM10/27/04
to
ro...@telus.net wrote:

Only if you do not understand the concept and really believe that labor
and land are the only resources. If you believe that mud pies have the
same cost as apple pies because labor is paid is the same and they are
made on the same land then anything seems idiotic. If you understand
there are other inputs then you have the beginnings of understanding the
concepts.

sinister

unread,
Oct 27, 2004, 9:14:52 PM10/27/04
to

"Igor" <jjwea...@houston.rr.com> wrote in message
news:pvRfd.3090$EI6....@fe2.texas.rr.com...

Roy showed in his argument with you---in which you were utterly
destroyed---that the paper is incoherent.

And why should we take any of your posts seriously, given that you once
wrote, "There is a limit interest compounds exponentially while GDP grows in
a linear fashion." ?? (Link at
http://groups.google.com/groups?q=g:thl1955457383d&dq=&hl=en&lr=&c2coff=1&selm=4xGcd.16172%24Rf4.14720%40fe2.texas.rr.com&rnum=2
)


sinister

unread,
Oct 27, 2004, 9:14:52 PM10/27/04
to

"Igor" <jjwea...@houston.rr.com> wrote in message
news:bwRfd.3091$EI6...@fe2.texas.rr.com...

Apparently you're not any better with using the web than with understanding
economics. The page is there, and according to the newspaper he did say it.
Extended excerpt:
---------------------------------------------
Bush's campaign on Monday released a letter signed by Prescott and five
other Nobel laureates critical of Kerry's proposal to roll back tax
reductions for families earning $200,000 or more.

In The Republic interview, he said such a policy would discourage people
from working.

"It's easy to get over $200,000 in income with two wage earners in a

household," Prescott said. "We want those highly educated, talented people
to work."


Robert Vienneau

unread,
Oct 28, 2004, 12:06:19 AM10/28/04
to
In article <7kLfd.6791$LT1.5725@trnddc09>, "sinister"
<sini...@nospam.invalid> wrote:

> "Robert Vienneau" <rv...@see.sig.com> wrote in message
> news:rvien-C00412....@news.dreamscape.com...
> >
> > "It's easy to get over $200,000 in income with two wage earners in a
> > household."
> > -- Edward Prescott
> > <http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott
> > 1
> > 9.html>

> Did you see the thread I started, "Fine work by our new Nobelist"?

Yes. It was about another dubious statement by Prescott.

Igor

unread,
Oct 28, 2004, 1:48:54 AM10/28/04
to
sinister wrote:

> "Igor" <jjwea...@houston.rr.com> wrote in message
> news:bwRfd.3091$EI6...@fe2.texas.rr.com...
>
>>Robert Vienneau wrote:
>>
>>
>>> "It's easy to get over $200,000 in income with two wage earners in a
>>> household."
>>> -- Edward Prescott
>>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott1
>>>9.html>
>>>
>>Interesting the page can not be found. Did he really say this?
>
>
> Apparently you're not any better with using the web than with understanding
> economics.

Oh part of the link was not placed in the html code. Something with
Mozilla cut some of the code. Now look at what he says "We want those
highly educated, talented people
to work.".

He means it is easy for highly educated talent people to make over
200000 in a two earner household. The truth is that it is easy given
what he said. He did not say it was easy for every two income family to
make 20000+.

The point is that Labor economist have estimated married women to have
the highest income elasticities. The second earner is much more
sensitive to wages when deciding to work. This is especially true when
the primary earner has a high income level. Presscot is simply talking
about the disincentive for these people to work. You raise their taxes
and they will work less.

There is nothing wrong or stupid about this statement. Rob cut the
qualification on the quotation to make Presscot sound as though he was
speaking of the general population. He was being deceptive and dishonest.

Igor

unread,
Oct 28, 2004, 1:52:51 AM10/28/04
to
sinister wrote:

> "Igor" <jjwea...@houston.rr.com> wrote in message
> news:pvRfd.3090$EI6....@fe2.texas.rr.com...
>
>>sinister wrote:
>>
>>
>>>"Robert Vienneau" <rv...@see.sig.com> wrote in message
>>>news:rvien-C00412....@news.dreamscape.com...
>>>
>>>
>>>>"It's easy to get over $200,000 in income with two wage earners in a
>>>>household."
>>>> -- Edward Prescott
>>>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott1
>>>>9.html>
>>>
>>>
>>>Did you see the thread I started, "Fine work by our new Nobelist"?
>>>
>>
>>I did and it was it obvious you never read the paper you were attacking.
>>You were asking question clearly answered in the paper.
>
>
> Roy showed in his argument with you---in which you were utterly
> destroyed---that the paper is incoherent.
>

It is incoherent to Roy because he thinks resources are only land and
labor. To any trained economist the paper is very coherent. If you
really think Roy made sense then you really seriously are biased against
any mainstream economic theory, like Rob, or you just do not understand
what is being said.


> And why should we take any of your posts seriously, given that you once
> wrote, "There is a limit interest compounds exponentially while GDP grows in
> a linear fashion." ?? (Link at
> http://groups.google.com/groups?q=g:thl1955457383d&dq=&hl=en&lr=&c2coff=1&selm=4xGcd.16172%24Rf4.14720%40fe2.texas.rr.com&rnum=2
> )
>
>

Because this is true. GDP growths are mean reverting meaning they are
close to be constant in the long run. The misunderstanding is that
economist would never do time series analysis without taking logs. That
is why it is trend stationary. I forget sometimes that people do not
know that. The statement is true. I even showed the formulas that back
up the point that there are limits to deficits.

Igor

unread,
Oct 28, 2004, 1:55:20 AM10/28/04
to
Robert Vienneau wrote:

> In article <7kLfd.6791$LT1.5725@trnddc09>, "sinister"
> <sini...@nospam.invalid> wrote:
>
>
>>"Robert Vienneau" <rv...@see.sig.com> wrote in message
>>news:rvien-C00412....@news.dreamscape.com...
>>
>>> "It's easy to get over $200,000 in income with two wage earners in a
>>> household."
>>> -- Edward Prescott
>>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott
>>>1
>>>9.html>
>
>
>
>>Did you see the thread I started, "Fine work by our new Nobelist"?
>
>
> Yes. It was about another dubious statement by Prescott.
>

A "dubious" statement supported by data. A "dubious" statement that was
robust to different measurements. The statement surprised me but the
data supported it when Fisher did his analysis and Prescott's analysis
supported the statement as well.

If you want to still call statements backed by data dubious then I can
understand why you often can not see the obvious.

Robert Vienneau

unread,
Oct 28, 2004, 3:17:26 AM10/28/04
to
In article <gKXfd.4574$jD4.1943@trnddc06>, "sinister"
<sini...@nospam.invalid> wrote:

> "Igor" <jjwea...@houston.rr.com> wrote in message
> news:pvRfd.3090$EI6....@fe2.texas.rr.com...

> And why should we take any of your posts seriously, given that you once

> wrote, "There is a limit interest compounds exponentially while GDP grows
> in
> a linear fashion." ?? (Link at
> http://groups.google.com/groups?q=g:thl1955457383d&dq=&hl=en&lr=&c2coff=1&
> selm=4xGcd.16172%24Rf4.14720%40fe2.texas.rr.com&rnum=2
> )

I've added a page to my "Immortal Fumbles" list. See:

<http://www.dreamscape.com/rvien/Fumbles/GDPGrowthIsLinear.html>

Igor

unread,
Oct 28, 2004, 3:40:23 AM10/28/04
to
Robert Vienneau wrote:

> In article <gKXfd.4574$jD4.1943@trnddc06>, "sinister"
> <sini...@nospam.invalid> wrote:
>
>
>>"Igor" <jjwea...@houston.rr.com> wrote in message
>>news:pvRfd.3090$EI6....@fe2.texas.rr.com...
>
>
>>And why should we take any of your posts seriously, given that you once
>>wrote, "There is a limit interest compounds exponentially while GDP grows
>>in
>>a linear fashion." ?? (Link at
>>http://groups.google.com/groups?q=g:thl1955457383d&dq=&hl=en&lr=&c2coff=1&
>>selm=4xGcd.16172%24Rf4.14720%40fe2.texas.rr.com&rnum=2
>> )
>
>
> I've added a page to my "Immortal Fumbles" list. See:
>
> <http://www.dreamscape.com/rvien/Fumbles/GDPGrowthIsLinear.html>
>

Oh no I have to hang my head in shame now. The statement is correct.
REAL GPD Growth per capita is mean reverting. It averages 1% - 2% a
year. REAL GDP GROWTH IS MEAN REVERTING. Most time series methods used
shows a linear trend. I say most because some older methods can not
reject the unit root hypothesis. It is well known that GDP for the US is
trend stationary with a break. THE CONFUSION IS THAT TIME SERIES
ANALYSIS ALWAYS ANALYZES GDP IN LOGS. THAT MEANS THE TREND IS LINEAR.
The statement is correct.

Again you are being deceptive and dishonest because you know this is the
case. Any economist reading that statement knows it too.

Robert Vienneau

unread,
Oct 28, 2004, 5:49:25 AM10/28/04
to
In article <Hn1gd.6962$EI6...@fe2.texas.rr.com>, Igor
<jjwea...@houston.rr.com> wrote:

> Robert Vienneau wrote:

> > In article <gKXfd.4574$jD4.1943@trnddc06>, "sinister"
> > <sini...@nospam.invalid> wrote:

> >>And why should we take any of your posts seriously, given that you once
> >>wrote, "There is a limit interest compounds exponentially while GDP
> >>grows
> >>in
> >>a linear fashion." ??

> > I've added a page to my "Immortal Fumbles" list. See:


> >
> > <http://www.dreamscape.com/rvien/Fumbles/GDPGrowthIsLinear.html>

> Oh no I have to hang my head in shame now. The statement is correct.
> REAL GPD Growth per capita is mean reverting. It averages 1% - 2% a
> year. REAL GDP GROWTH IS MEAN REVERTING. Most time series methods used
> shows a linear trend. I say most because some older methods can not
> reject the unit root hypothesis. It is well known that GDP for the US is
> trend stationary with a break. THE CONFUSION IS THAT TIME SERIES
> ANALYSIS ALWAYS ANALYZES GDP IN LOGS. THAT MEANS THE TREND IS LINEAR.
> The statement is correct.

Anybody want to explain to this idiot with a PhD - or perhaps ABD -
some simple high school mathematics?



> Again you are being deceptive and dishonest because you know this is the
> case. Any economist reading that statement knows it too.

I know Mr. Weatherby is ignorant, illiterate, and innumerate.

Igor

unread,
Oct 28, 2004, 6:09:51 AM10/28/04
to
Robert Vienneau wrote:

> In article <Hn1gd.6962$EI6...@fe2.texas.rr.com>, Igor
> <jjwea...@houston.rr.com> wrote:
>
>
>>Robert Vienneau wrote:
>
>
>
>>>In article <gKXfd.4574$jD4.1943@trnddc06>, "sinister"
>>><sini...@nospam.invalid> wrote:
>
>
>>>>And why should we take any of your posts seriously, given that you once
>>>>wrote, "There is a limit interest compounds exponentially while GDP
>>>>grows
>>>>in
>>>>a linear fashion." ??
>
>
>>>I've added a page to my "Immortal Fumbles" list. See:
>>>
>>> <http://www.dreamscape.com/rvien/Fumbles/GDPGrowthIsLinear.html>
>
>
>
>>Oh no I have to hang my head in shame now. The statement is correct.
>>REAL GPD Growth per capita is mean reverting. It averages 1% - 2% a
>>year. REAL GDP GROWTH IS MEAN REVERTING. Most time series methods used
>>shows a linear trend. I say most because some older methods can not
>>reject the unit root hypothesis. It is well known that GDP for the US is
>>trend stationary with a break. THE CONFUSION IS THAT TIME SERIES
>>ANALYSIS ALWAYS ANALYZES GDP IN LOGS. THAT MEANS THE TREND IS LINEAR.
>>The statement is correct.
>
>
> Anybody want to explain to this idiot with a PhD - or perhaps ABD -
> some simple high school mathematics?
>

It is not a misunderstanding of mathematics. It is loose language.
Whenever GDP is plotted by an economist or analyzed as a time series the
logarithm is used. It is a simple miscommunication that occurs between
people who actually work with data and people who do not work with data.

>
>>Again you are being deceptive and dishonest because you know this is the
>>case. Any economist reading that statement knows it too.
>
>
> I know Mr. Weatherby is ignorant, illiterate, and innumerate.
>

Typical Rob fashion when he knows a criticism is true he reverts to
insults. Oh come now Rob, I did not insult you when I showed that your
understanding of minimum wage analysis by mainstream economist was
wrong. I did state the fact that you were ignorant of the long run
prediction that being exactly what you argued. I did not insult you though.

How can I be illiterate? I was able to read your message. I can not be
fully illiterate. Oh yes that is the charge given when you have been
shown to be exageratting, dishonest, deceptive or just completely wrong.

sinister

unread,
Oct 28, 2004, 6:25:45 AM10/28/04
to

"Igor" <jjwea...@houston.rr.com> wrote in message
news:TO%fd.10534$186....@fe1.texas.rr.com...

> sinister wrote:
>
>> "Igor" <jjwea...@houston.rr.com> wrote in message
>> news:pvRfd.3090$EI6....@fe2.texas.rr.com...
>>
>>>sinister wrote:
>>>
>>>
>>>>"Robert Vienneau" <rv...@see.sig.com> wrote in message
>>>>news:rvien-C00412....@news.dreamscape.com...
>>>>
>>>>
>>>>>"It's easy to get over $200,000 in income with two wage earners in a
>>>>>household."
>>>>> -- Edward Prescott
>>>>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott1
>>>>>9.html>
>>>>
>>>>
>>>>Did you see the thread I started, "Fine work by our new Nobelist"?
>>>>
>>>
>>>I did and it was it obvious you never read the paper you were attacking.
>>>You were asking question clearly answered in the paper.
>>
>>
>> Roy showed in his argument with you---in which you were utterly
>> destroyed---that the paper is incoherent.
>>
>
> It is incoherent to Roy because he thinks resources are only land and
> labor. To any trained economist the paper is very coherent. If you

No, like anyone familiar with the classical liberals, Roy knows there are
three factor of production: land, labor, and capital.

> really think Roy made sense then you really seriously are biased against
> any mainstream economic theory, like Rob, or you just do not understand
> what is being said.

Nope. It's just that Roy---and I---are biased against the idea that the
proper way to value a firm is to add up its resource value or whatever name
Prescott has given to this concept.

>
>
>> And why should we take any of your posts seriously, given that you once
>> wrote, "There is a limit interest compounds exponentially while GDP grows
>> in a linear fashion." ?? (Link at
>> http://groups.google.com/groups?q=g:thl1955457383d&dq=&hl=en&lr=&c2coff=1&selm=4xGcd.16172%24Rf4.14720%40fe2.texas.rr.com&rnum=2
>> )
>>
>>
>
> Because this is true. GDP growths are mean reverting meaning they are
> close to be constant in the long run. The misunderstanding is that
> economist would never do time series analysis without taking logs. That is
> why it is trend stationary. I forget sometimes that people do not know
> that. The statement is true. I even showed the formulas that back up the
> point that there are limits to deficits.

Either
(a) you're an idiot when it comes to economics, as GDP growth is exponential
(i.e., geometric), or
(b) you're an idiot when it comes to language.


Igor

unread,
Oct 28, 2004, 6:45:07 AM10/28/04
to
sinister wrote:


>>
>>It is incoherent to Roy because he thinks resources are only land and
>>labor. To any trained economist the paper is very coherent. If you
>
>
> No, like anyone familiar with the classical liberals, Roy knows there are
> three factor of production: land, labor, and capital.
>
>

This is a simplification for models. In any empirical work you have to
add raw materials and energy. You can not ignore these resources or
their cost. Under a competitive market price is equal to marginal cost.
That means P= the extra cost of producing the last unit. That cost
includes raw materials and energy. This means the resource cost is the
price for paid for the good.

So empirically you add up all the costs of the good (which equals price
paid) and account for deprecation to find the stock. It does not make
sense to you because you do not understand what resource means. Once you
understand that you can understand the model.

All of this would be painfully clear if you actually read the paper
instead of criticizing the abstract. It is clearly explained before any
equation is written that these resource cost represent replacement cost.
That means how much it cost to replace the asset. Roy did not even
understand that the equations represent a return nor do you.

The equations used in the robustness check for Fisher's result are
simple and clear. It says your profits equal your after tax returns on
your investment. That is why taxes come into play. Plain and simple.

Roy still does not understand that the equation iK + (i-g)C does not
have to be negative if g > i. That is just simple algebra. That is where
his big misconception of the "stupidity" of the model came from.

> Nope. It's just that Roy---and I---are biased against the idea that the
> proper way to value a firm is to add up its resource value or whatever name
> Prescott has given to this concept.
>

The value is the discounted sum of returns. This means you have to value
the assets to find the returns. You can't see this simple concept
because you do not believe that a price in competitive market is equal
to the marginal cost of the last good made and that cost includes raw
materials and energy not just labor, land, and capital.

>
>>
>>>And why should we take any of your posts seriously, given that you once
>>>wrote, "There is a limit interest compounds exponentially while GDP grows
>>>in a linear fashion." ?? (Link at
>>>http://groups.google.com/groups?q=g:thl1955457383d&dq=&hl=en&lr=&c2coff=1&selm=4xGcd.16172%24Rf4.14720%40fe2.texas.rr.com&rnum=2
>>> )
>>>
>>>
>>
>>Because this is true. GDP growths are mean reverting meaning they are
>>close to be constant in the long run. The misunderstanding is that
>>economist would never do time series analysis without taking logs. That is
>>why it is trend stationary. I forget sometimes that people do not know
>>that. The statement is true. I even showed the formulas that back up the
>>point that there are limits to deficits.
>
>
> Either
> (a) you're an idiot when it comes to economics, as GDP growth is exponential
> (i.e., geometric), or
> (b) you're an idiot when it comes to language.
>

I would not go as far to say I am an idiot with language but perhaps
loose with language. Economist always use the logs of GDP when doing
analysis such as time series. It is so common that I forget not everyone
does this. You plot the log of GDP versus time and the trend is linear.
It is trend stationary with a break. The statement is correct but the
language was loose.

>

Andy F

unread,
Oct 28, 2004, 6:58:22 AM10/28/04
to
Igor wrote:

>Oh no I have to hang my head in shame now. The statement is correct.
>REAL GPD Growth per capita is mean reverting. It averages 1% - 2% a
>year. REAL GDP GROWTH IS MEAN REVERTING. Most time series methods used
>shows a linear trend. I say most because some older methods can not
>reject the unit root hypothesis. It is well known that GDP for the US is
>trend stationary with a break. THE CONFUSION IS THAT TIME SERIES
>ANALYSIS ALWAYS ANALYZES GDP IN LOGS. THAT MEANS THE TREND IS LINEAR.
>The statement is correct.
>
>Again you are being deceptive and dishonest because you know this is the
>case. Any economist reading that statement knows it too.
>

So when you said

"There is a limit interest compounds exponentially while GDP grows in a linear
fashion."

what you really meant was

"interest compounds exponentially while GDP grows in a linear fashion on a
logarithmic scale"

If you put the compounding interest on a log scale, wouldn't that also be
linear?


sinister

unread,
Oct 28, 2004, 9:13:42 AM10/28/04
to

"Igor" <jjwea...@houston.rr.com> wrote in message
news:Pz3gd.11901$186....@fe1.texas.rr.com...

> Robert Vienneau wrote:
>
>> In article <Hn1gd.6962$EI6...@fe2.texas.rr.com>, Igor
>> <jjwea...@houston.rr.com> wrote:
>>
>>
>>>Robert Vienneau wrote:
>>
>>
>>>>In article <gKXfd.4574$jD4.1943@trnddc06>, "sinister"
>>>><sini...@nospam.invalid> wrote:
>>
>>
>>>>>And why should we take any of your posts seriously, given that you once
>>>>>wrote, "There is a limit interest compounds exponentially while GDP
>>>>>grows in a linear fashion." ??
>>
>>
>>>>I've added a page to my "Immortal Fumbles" list. See:
>>>>
>>>> <http://www.dreamscape.com/rvien/Fumbles/GDPGrowthIsLinear.html>
>>
>>
>>>Oh no I have to hang my head in shame now. The statement is correct. REAL
>>>GPD Growth per capita is mean reverting. It averages 1% - 2% a year.
>>>REAL GDP GROWTH IS MEAN REVERTING. Most time series methods used shows a
>>>linear trend. I say most because some older methods can not reject the
>>>unit root hypothesis. It is well known that GDP for the US is trend
>>>stationary with a break. THE CONFUSION IS THAT TIME SERIES ANALYSIS
>>>ALWAYS ANALYZES GDP IN LOGS. THAT MEANS THE TREND IS LINEAR. The
>>>statement is correct.
>>
>>
>> Anybody want to explain to this idiot with a PhD - or perhaps ABD -
>> some simple high school mathematics?
>>
>
> It is not a misunderstanding of mathematics. It is loose language.

No, moron.

You originally wrote, "There is a limit interest compounds exponentially


while GDP grows in a linear fashion."

Clearly you're saying GDP grows linearly *in contradistinction* to another
quantity that grows exponentially.

Robert Vienneau

unread,
Oct 28, 2004, 2:33:59 PM10/28/04
to
In article <Pz3gd.11901$186....@fe1.texas.rr.com>, Igor
<jjwea...@houston.rr.com> wrote:

> Robert Vienneau wrote:
>
> > In article <Hn1gd.6962$EI6...@fe2.texas.rr.com>, Igor
> > <jjwea...@houston.rr.com> wrote:

> >>Again you are being deceptive and dishonest because you know this is
> >>the
> >>case. Any economist reading that statement knows it too.

> ...Oh come now Rob, I did not insult you...
> [ Rehashing of Mr. Weatherby's nonsense from a previous thread - ]
> [ deleted. ]

Robert Vienneau

unread,
Oct 28, 2004, 2:33:53 PM10/28/04
to
In article <aL%fd.6521$EI6....@fe2.texas.rr.com>, Igor
<jjwea...@houston.rr.com> wrote:

> sinister wrote:
>
> > "Igor" <jjwea...@houston.rr.com> wrote in message
> > news:bwRfd.3091$EI6...@fe2.texas.rr.com...

> >>Robert Vienneau wrote:

> >>> "It's easy to get over $200,000 in income with two wage earners in a
> >>> household."

"We want those highly educated, talented people to work."

> >>> -- Edward Prescott
> >>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott19.html>

> [A poor carpenter blaming his tools - deleted. ]


> Now look at what he says "We want those
> highly educated, talented people
> to work.".

> He means it is easy for highly educated talent people to make over
> 200000 in a two earner household. The truth is that it is easy given
> what he said.

The above statement, like Prescott's, is that of an idiot.

> He did not say it was easy for every two income family to
> make 20000+.

> There is nothing wrong or stupid about this statement. Rob cut the

> qualification on the quotation to make Presscot sound as though he was
> speaking of the general population. He was being deceptive and dishonest.

It is not my fault that the fake Nobel prize was awarded this year
for mistaken work by idiot savants.

Igor

unread,
Oct 28, 2004, 4:57:16 PM10/28/04
to
Robert Vienneau wrote:

> In article <Pz3gd.11901$186....@fe1.texas.rr.com>, Igor
> <jjwea...@houston.rr.com> wrote:
>
>
>>Robert Vienneau wrote:
>>
>>
>>>In article <Hn1gd.6962$EI6...@fe2.texas.rr.com>, Igor
>>><jjwea...@houston.rr.com> wrote:
>
>
>>>>Again you are being deceptive and dishonest because you know this is
>>>>the
>>>>case. Any economist reading that statement knows it too.
>
>
>>...Oh come now Rob, I did not insult you...
>>[ Rehashing of Mr. Weatherby's nonsense from a previous thread - ]
>>[ deleted. ]
>
>

Did you think this deleted the post from the server? Obviously you have
nothing to say about the fact that your model is long run AND PREDICTS
THE SAME THING GE DOES IN THE LONG RUN. You still want to hang on to the
thread of a notion that you are proving nobel laurates and educated
people wrong when you are not. Nice analysis to call someone who
disagrees with the statement an idiot. Many highly educated two income
households make over 200,000. These are the people with the most elastic
labor supply and will work less when taxes are raised. Why? Because
unlike the guys making 20,000 a year they can.

ro...@telus.net

unread,
Oct 28, 2004, 5:29:37 PM10/28/04
to
On Wed, 27 Oct 2004 20:33:56 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>ro...@telus.net wrote:
>
>> On Wed, 27 Oct 2004 18:09:25 GMT, Igor <jjwea...@houston.rr.com>
>> wrote:
>>
>>>sinister wrote:
>>>
>>>>Did you see the thread I started, "Fine work by our new Nobelist"?
>>>
>>>I did and it was it obvious you never read the paper you were attacking.
>>>You were asking question clearly answered in the paper.
>>
>> Yes, and I showed, in our exchange, that the answers were idiotic.
>

>Only if you do not understand the concept and really believe that labor
>and land are the only resources.

I have said no such thing, and that is in any case irrelevant to
Prescott's fundamental conceptual error of identifying cost with
value. It is you who does not understand that concept.

>If you believe that mud pies have the
>same cost as apple pies because labor is paid is the same and they are
>made on the same land then anything seems idiotic.

<sigh> I can spend $1K to make apple pies, or mud pies. That makes
their resource costs identical. Their values are not identical. You
(and Prescott) stand refuted.

>If you understand
>there are other inputs then you have the beginnings of understanding the
>concepts.

The other inputs are irrelevant to Prescott's error, as demonstrated
above.

-- Roy L

ro...@telus.net

unread,
Oct 28, 2004, 5:37:01 PM10/28/04
to
On Thu, 28 Oct 2004 05:52:51 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>sinister wrote:
>
>> "Igor" <jjwea...@houston.rr.com> wrote in message
>> news:pvRfd.3090$EI6....@fe2.texas.rr.com...
>>
>>>sinister wrote:
>>>
>>>>"Robert Vienneau" <rv...@see.sig.com> wrote in message
>>>>news:rvien-C00412....@news.dreamscape.com...
>>>>
>>>>>"It's easy to get over $200,000 in income with two wage earners in a
>>>>>household."
>>>>> -- Edward Prescott
>>>>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott1
>>>>>9.html>
>>>>
>>>>Did you see the thread I started, "Fine work by our new Nobelist"?
>>>
>>>I did and it was it obvious you never read the paper you were attacking.
>>>You were asking question clearly answered in the paper.
>>
>> Roy showed in his argument with you---in which you were utterly
>> destroyed---that the paper is incoherent.
>
>It is incoherent to Roy because he thinks resources are only land and
>labor.

That is a lie; and even if it were true, it is irrelevant to
Prescott's errors.

>To any trained economist the paper is very coherent.

Garbage. Any trained economist should know that the Labor Theory of
Value fails for precisely the same reason as Prescott's paper: cost is
not value.

>If you
>really think Roy made sense then you really seriously are biased against
>any mainstream economic theory, like Rob, or you just do not understand
>what is being said.

More garbage. Identifying cost with value is not mainstream economic
theory in any place I have ever heard of.

>> And why should we take any of your posts seriously, given that you once
>> wrote, "There is a limit interest compounds exponentially while GDP grows in
>> a linear fashion." ?? (Link at
>> http://groups.google.com/groups?q=g:thl1955457383d&dq=&hl=en&lr=&c2coff=1&selm=4xGcd.16172%24Rf4.14720%40fe2.texas.rr.com&rnum=2
>

>Because this is true. GDP growths are mean reverting meaning they are
>close to be constant in the long run. The misunderstanding is that
>economist would never do time series analysis without taking logs. That
>is why it is trend stationary. I forget sometimes that people do not
>know that.

No, you forgot that like GDP, compound interest is also linear on a
log graph.

>The statement is true. I even showed the formulas that back
>up the point that there are limits to deficits.

Please, Igor. Robert only has so many gigabytes of space in his
fumbles file.

-- Roy L

ro...@telus.net

unread,
Oct 28, 2004, 5:39:24 PM10/28/04
to
On Thu, 28 Oct 2004 05:55:20 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>Robert Vienneau wrote:
>
>> In article <7kLfd.6791$LT1.5725@trnddc09>, "sinister"
>> <sini...@nospam.invalid> wrote:
>>
>>>"Robert Vienneau" <rv...@see.sig.com> wrote in message
>>>news:rvien-C00412....@news.dreamscape.com...
>>>
>>>> "It's easy to get over $200,000 in income with two wage earners in a
>>>> household."
>>>> -- Edward Prescott
>>>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott
>>>>1
>>>>9.html>
>>
>>>Did you see the thread I started, "Fine work by our new Nobelist"?
>>
>> Yes. It was about another dubious statement by Prescott.
>
>A "dubious" statement supported by data.

No, it wasn't. The value data were simply fabricated.

>A "dubious" statement that was
>robust to different measurements.

Except the facts of history.

>The statement surprised me but the
>data supported it when Fisher did his analysis and Prescott's analysis
>supported the statement as well.

Except that history proved them both laughably wrong.

>If you want to still call statements backed by data dubious then I can
>understand why you often can not see the obvious.

As they say in Japan, "It's mirror time!"

-- Roy L

sinister

unread,
Oct 28, 2004, 9:23:42 PM10/28/04
to

"Andy F" <aft...@aol.com> wrote in message
news:20041028065822...@mb-m18.aol.com...

> Igor wrote:
>
>>Oh no I have to hang my head in shame now. The statement is correct.
>>REAL GPD Growth per capita is mean reverting. It averages 1% - 2% a
>>year. REAL GDP GROWTH IS MEAN REVERTING. Most time series methods used
>>shows a linear trend. I say most because some older methods can not
>>reject the unit root hypothesis. It is well known that GDP for the US is
>>trend stationary with a break. THE CONFUSION IS THAT TIME SERIES
>>ANALYSIS ALWAYS ANALYZES GDP IN LOGS. THAT MEANS THE TREND IS LINEAR.
>>The statement is correct.
>>
>>Again you are being deceptive and dishonest because you know this is the
>>case. Any economist reading that statement knows it too.
>>
> So when you said
>
> "There is a limit interest compounds exponentially while GDP grows in a
> linear
> fashion."
>
> what you really meant was

If you engage Igor, you're find yourself thinking, "well, if I'm charitable,
what he really meant was..." repeatedly.

>
> "interest compounds exponentially while GDP grows in a linear fashion on a
> logarithmic scale"
>
> If you put the compounding interest on a log scale, wouldn't that also be
> linear?

Shhh! You're going to confuse him!

>
>


Igor

unread,
Oct 28, 2004, 9:39:29 PM10/28/04
to
ro...@telus.net wrote:

> On Wed, 27 Oct 2004 20:33:56 GMT, Igor <jjwea...@houston.rr.com>
> wrote:
>
>
>>ro...@telus.net wrote:
>>
>>
>>>On Wed, 27 Oct 2004 18:09:25 GMT, Igor <jjwea...@houston.rr.com>
>>>wrote:
>>>
>>>
>>>>sinister wrote:
>>>>
>>>>
>>>>>Did you see the thread I started, "Fine work by our new Nobelist"?
>>>>
>>>>I did and it was it obvious you never read the paper you were attacking.
>>>>You were asking question clearly answered in the paper.
>>>
>>>Yes, and I showed, in our exchange, that the answers were idiotic.
>>
>>Only if you do not understand the concept and really believe that labor
>>and land are the only resources.
>
>
> I have said no such thing, and that is in any case irrelevant to
> Prescott's fundamental conceptual error of identifying cost with
> value. It is you who does not understand that concept.
>
>
>>If you believe that mud pies have the
>>same cost as apple pies because labor is paid is the same and they are
>>made on the same land then anything seems idiotic.
>
>
> <sigh> I can spend $1K to make apple pies, or mud pies. That makes
> their resource costs identical. Their values are not identical. You
> (and Prescott) stand refuted.
>

No it does not make them the same. How do you think that capital should
be empirically measured. It is measured by real replacement cost. That
means you adjust for inflation to get a measure in real units. How do
you suggest to do that other than add up investment( what was paid for
the capital) and subtract depreciation. I hate to break it to you but
accountants value assets in the same way!! So how do you propose it
should be measured?

Igor

unread,
Oct 28, 2004, 9:45:33 PM10/28/04
to
ro...@telus.net wrote:

A Labor theory of value does not say that cost is value. It says that
value is given because labor is added to it. It does not say that it is
reflected by the wages. In fact, Smith argues that if takes twice as
much labor to kill a deer than a rabbit then deer meat cost twice as
much not that the value is twice as much. This is where you
misunderstand what a labor theory is.

Objective prices in a competitive market says that PRICE = MC. That
means price is the COST OF THE LAST UNIT PRODUCED. This is not a labor
theory. It is quite the contrary. So first you have to understand what a
theory of value is and how a labor theory differs from objective value
to understand what is being argued.

>
>>If you
>>really think Roy made sense then you really seriously are biased against
>>any mainstream economic theory, like Rob, or you just do not understand
>>what is being said.
>
>
> More garbage. Identifying cost with value is not mainstream economic
> theory in any place I have ever heard of.
>

P = MC in a competitive market. IF YOU HAVE NOT HEARD OF THAT YOU HAVE
NEVER TAKEN A SINGLE COLLEGE COURSE IN MAINSTREAM ECONOMICS.


>>The statement is true. I even showed the formulas that back
>>up the point that there are limits to deficits.
>
>
> Please, Igor. Robert only has so many gigabytes of space in his
> fumbles file.
>

The statement is true. There are limits on deficits. Interest rates grow
faster than GDP growth this is almost always the case. That means if you
run primary deficits and continue to finance the interest then the DEBT
IS GROWING FASTER THAN GDP at some point interest payments will exceed
GDP. This is another simple concept you can not grasp.

> -- Roy L

Igor

unread,
Oct 28, 2004, 9:49:18 PM10/28/04
to
ro...@telus.net wrote:

> On Thu, 28 Oct 2004 05:55:20 GMT, Igor <jjwea...@houston.rr.com>
> wrote:
>
>
>>Robert Vienneau wrote:
>>
>>
>>>In article <7kLfd.6791$LT1.5725@trnddc09>, "sinister"
>>><sini...@nospam.invalid> wrote:
>>>
>>>
>>>>"Robert Vienneau" <rv...@see.sig.com> wrote in message
>>>>news:rvien-C00412....@news.dreamscape.com...
>>>>
>>>>
>>>>>"It's easy to get over $200,000 in income with two wage earners in a
>>>>>household."
>>>>> -- Edward Prescott
>>>>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott
>>>>>1
>>>>>9.html>
>>>
>>>>Did you see the thread I started, "Fine work by our new Nobelist"?
>>>
>>>Yes. It was about another dubious statement by Prescott.
>>
>>A "dubious" statement supported by data.
>
>
> No, it wasn't. The value data were simply fabricated.
>

Sorry this should be listed on alt.conspiracy now. The data come from
credible stockmarket data nothing is fabricated. In the first section
the equations you do not understand and think are stupid ARE NOT USED.
RTFM.


>
>>A "dubious" statement that was
>>robust to different measurements.
>
>
> Except the facts of history.
>

What facts are those that the stock market crashed due to excessive
margin calls not bad news about profits?

>
>>The statement surprised me but the
>>data supported it when Fisher did his analysis and Prescott's analysis
>>supported the statement as well.
>
>
> Except that history proved them both laughably wrong.
>

How? Why do think that after looking at the data. Just because the
market crashed due to margin calls not being met does not mean that the
stocks analyzed were not undervalued. Profits were high at that period
of time. On the day the market crashed conditions were good. Labor
disputes had been settled. It was subsequent events that caused the
economy to tank. You think it is silly because you have the silly notion
that the economy and not the stock market crashed on particular day.


ro...@telus.net

unread,
Oct 29, 2004, 5:57:05 PM10/29/04
to
On Thu, 28 Oct 2004 10:45:07 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>sinister wrote:
>
>>>It is incoherent to Roy because he thinks resources are only land and
>>>labor. To any trained economist the paper is very coherent. If you
>>
>> No, like anyone familiar with the classical liberals, Roy knows there are
>> three factor of production: land, labor, and capital.
>
>This is a simplification for models.

No, it is not.

>In any empirical work you have to
>add raw materials and energy.

Which are in turn products of land, labor and capital. You seem to
have no understanding of economics whatever.

>You can not ignore these resources or
>their cost.

Nobody is.

>Under a competitive market price is equal to marginal cost.

Only under certain ideal conditions which are very remote from the
conditions under which corporate assets are typically acquired and
owned.

>That means P= the extra cost of producing the last unit. That cost
>includes raw materials and energy. This means the resource cost is the
>price for paid for the good.

All of which has _zero_ to do with valuation of corporate assets.

>So empirically you add up all the costs of the good (which equals price
>paid) and account for deprecation to find the stock.

That is not an empirical measure of value. It is a priori.

>It does not make
>sense to you because you do not understand what resource means. Once you
>understand that you can understand the model.

Garbage. I understand the model fine. It is just nonsensical.

>All of this would be painfully clear if you actually read the paper
>instead of criticizing the abstract. It is clearly explained before any
>equation is written that these resource cost represent replacement cost.

Replacement cost is not value. Take a new car. Pay someone $10 to
hit it with a sledgehammer for an hour. How much is it worth? How
much would it cost to replace it?

_Get_it_??



>That means how much it cost to replace the asset. Roy did not even
>understand that the equations represent a return nor do you.

ROTFL!! The equations were vastly at variance with the returns
actually being made. That's what _you_ don't seem able to understand.

>The equations used in the robustness check for Fisher's result are
>simple and clear. It says your profits equal your after tax returns on
>your investment. That is why taxes come into play. Plain and simple.

You cannot calculate profits by means of equations that equate them
with expenses. You can only calculate profits by _subtracting_
expenses from revenues. Prescott has it completely backwards.
Companies don't make profits by increasing costs, but by reducing them
below revenues.

>Roy still does not understand that the equation iK + (i-g)C does not
>have to be negative if g > i.

<sigh> For the third time, I said no such thing. I said it can
_only_ be negative if g > i (assuming all variables are positive).

>That is just simple algebra.

It is simple logic, which you obviously cannot understand.

>That is where
>his big misconception of the "stupidity" of the model came from.

The model is stupid even without that nonsensical element.

>> Nope. It's just that Roy---and I---are biased against the idea that the
>> proper way to value a firm is to add up its resource value or whatever name
>> Prescott has given to this concept.
>
>The value is the discounted sum of returns.

In fact, Prescott ignores the returns _actually_being_made_.

>This means you have to value
>the assets to find the returns.

ROTFL!! You value assets by the _estimated_ returns, not vice versa.
That is the whole point.

>You can't see this simple concept
>because you do not believe that a price in competitive market is equal
>to the marginal cost of the last good made

Corporate assets are not "the last goods made."

>and that cost includes raw
>materials and energy not just labor, land, and capital.

Incredible.



>> Either
>> (a) you're an idiot when it comes to economics, as GDP growth is exponential
>> (i.e., geometric), or
>> (b) you're an idiot when it comes to language.
>
>I would not go as far to say I am an idiot with language but perhaps
>loose with language.

Don't be so modest. You can be an idiot with both language and
economics. In fact, you are.

>Economist always use the logs of GDP when doing
>analysis such as time series. It is so common that I forget not everyone
>does this. You plot the log of GDP versus time and the trend is linear.

As is the sum of compound interest.

>It is trend stationary with a break. The statement is correct but the
>language was loose.

The statement is false.

-- Roy L

ro...@telus.net

unread,
Oct 29, 2004, 5:58:50 PM10/29/04
to

Don't look at the man behind the curtain!

-- Roy L

ro...@telus.net

unread,
Oct 30, 2004, 12:43:16 AM10/30/04
to
On Fri, 29 Oct 2004 01:39:29 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>ro...@telus.net wrote:
>
>> On Wed, 27 Oct 2004 20:33:56 GMT, Igor <jjwea...@houston.rr.com>
>> wrote:
>>
>>>If you believe that mud pies have the
>>>same cost as apple pies because labor is paid is the same and they are
>>>made on the same land then anything seems idiotic.
>>
>> <sigh> I can spend $1K to make apple pies, or mud pies. That makes
>> their resource costs identical. Their values are not identical. You
>> (and Prescott) stand refuted.
>
>No it does not make them the same.

Yes, it does. $1K = $1K. I am not sure how you prevent yourself from
knowing such facts, but it is clear that you manage to do it somehow.

>How do you think that capital should
>be empirically measured.

Depends on the purpose. In any case, book value is not the same as
ability to generate profits.

>It is measured by real replacement cost.

Sounds OK for insurance purposes.

>That
>means you adjust for inflation to get a measure in real units. How do
>you suggest to do that other than add up investment( what was paid for
>the capital) and subtract depreciation.

Market value would be better. Having paid $1K for something does not
make it worth $1K, and most certainly does not mean it will generate
equivalent profits. Market value at least indicates a collective
judgment of the profit stream an asset can be expected to yield.

>I hate to break it to you but
>accountants value assets in the same way!!

Yes, well, accountants aren't economists. Accountants told us that
Enron was a brilliant success.

>So how do you propose it
>should be measured?

The problem with trying to measure capital for economic purposes is
that it is not homogenous, and different kinds of capital are not
commensurable except by converting them to money. So if you want to
estimate the value of capital, market liquidation value is probably
the closest you can get.

-- Roy L

ro...@telus.net

unread,
Oct 30, 2004, 1:14:52 AM10/30/04
to
On Fri, 29 Oct 2004 01:45:33 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>ro...@telus.net wrote:
>
>> On Thu, 28 Oct 2004 05:52:51 GMT, Igor <jjwea...@houston.rr.com>
>> wrote:
>>
>>>To any trained economist the paper is very coherent.
>>
>> Garbage. Any trained economist should know that the Labor Theory of
>> Value fails for precisely the same reason as Prescott's paper: cost is
>> not value.
>
>A Labor theory of value does not say that cost is value. It says that
>value is given because labor is added to it.

Same mistake, different words.

>It does not say that it is
>reflected by the wages. In fact, Smith argues that if takes twice as
>much labor to kill a deer than a rabbit then deer meat cost twice as
>much not that the value is twice as much. This is where you
>misunderstand what a labor theory is.

Labor is a cost; it doesn't matter if the wage is high or low, labor
expended is a real cost. But it may or may not add value.

>Objective prices in a competitive market says that PRICE = MC.

That is merely a theoretical limit. It is not reality.

>That means price is the COST OF THE LAST UNIT PRODUCED.

But in reality, it isn't.

>This is not a labor
>theory. It is quite the contrary.

It is just a more complicated version of the same fallacy.

>So first you have to understand what a
>theory of value is and how a labor theory differs from objective value
>to understand what is being argued.

Objective value is _market_price_, and _not_ the result of some
equation.



>>>If you
>>>really think Roy made sense then you really seriously are biased against
>>>any mainstream economic theory, like Rob, or you just do not understand
>>>what is being said.
>>
>> More garbage. Identifying cost with value is not mainstream economic
>> theory in any place I have ever heard of.
>
>P = MC in a competitive market.

That equation is a falsehood, and corporate assets do not trade in a
competitive market in any case. The purpose of a corporation is to
construct a profit-making _system_ that yields greater returns than
just selling off its assets.

>IF YOU HAVE NOT HEARD OF THAT YOU HAVE
>NEVER TAKEN A SINGLE COLLEGE COURSE IN MAINSTREAM ECONOMICS.

I guess it takes more than one course to unlearn that equation...

>>>The statement is true. I even showed the formulas that back
>>>up the point that there are limits to deficits.
>>
>> Please, Igor. Robert only has so many gigabytes of space in his
>> fumbles file.
>
>The statement is true.

No, it is not.

>There are limits on deficits.

Yes, but political ones, not the ones described by your formulas.

>Interest rates grow
>faster than GDP growth this is almost always the case.

True. Average growth through history has been much lower than average
interest rates. But growth doesn't go broke and reset to zero every
so often, which might help explain the fact that average interest
rates are so much higher than average growth.

>That means if you
>run primary deficits and continue to finance the interest then the DEBT
>IS GROWING FASTER THAN GDP at some point interest payments will exceed
>GDP.

The average interest rate on US government debt has been less than
average US GDP growth. If interest ever becomes too politically
burdensome, the Treasury can just print more money, resetting the
nominal GDP growth rate to more than the interest rate on the debt.

-- Roy L

ro...@telus.net

unread,
Oct 30, 2004, 1:34:59 AM10/30/04
to
On Fri, 29 Oct 2004 01:49:18 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>ro...@telus.net wrote:
>
>> On Thu, 28 Oct 2004 05:55:20 GMT, Igor <jjwea...@houston.rr.com>
>> wrote:
>>
>>>Robert Vienneau wrote:
>>>
>>>>In article <7kLfd.6791$LT1.5725@trnddc09>, "sinister"
>>>><sini...@nospam.invalid> wrote:
>>>>
>>>>>"Robert Vienneau" <rv...@see.sig.com> wrote in message
>>>>>news:rvien-C00412....@news.dreamscape.com...
>>>>>
>>>>>>"It's easy to get over $200,000 in income with two wage earners in a
>>>>>>household."
>>>>>> -- Edward Prescott
>>>>>><http://www.azcentral.com/arizonarepublic/business/articles/1019Prescott
>>>>>>1
>>>>>>9.html>
>>>>
>>>>>Did you see the thread I started, "Fine work by our new Nobelist"?
>>>>
>>>>Yes. It was about another dubious statement by Prescott.
>>>
>>>A "dubious" statement supported by data.
>>
>> No, it wasn't. The value data were simply fabricated.
>
>Sorry this should be listed on alt.conspiracy now. The data come from
>credible stockmarket data nothing is fabricated.

The valuation formulas you gave were just a way of generating numbers,
with little relation to actual market value.

>In the first section
>the equations you do not understand and think are stupid ARE NOT USED.

Then why did you say they were?



>>>A "dubious" statement that was
>>>robust to different measurements.
>>
>> Except the facts of history.
>
>What facts are those that the stock market crashed due to excessive
>margin calls not bad news about profits?

It crashed because the profits weren't going to be there to support
the prices, because the money the corporations had spent on their
capital (resource cost) was not all spent wisely, given changing
market conditions. Margin calls were just the trigger, not the cause.



>>>The statement surprised me but the
>>>data supported it when Fisher did his analysis and Prescott's analysis
>>>supported the statement as well.
>>
>> Except that history proved them both laughably wrong.
>
>How?

By seeing stock prices collapse.

>Why do think that after looking at the data. Just because the
>market crashed due to margin calls not being met does not mean that the
>stocks analyzed were not undervalued.

Yes, it does. You say P = MC in a competitive market, but then
proceed to claim the real-life price verdict of a real-life
competitive market was somehow not valid. Like Prescott, you are
throwing out the facts that disprove your theory on its own terms.

>Profits were high at that period of time.

They weren't going to stay high.

>On the day the market crashed conditions were good. Labor
>disputes had been settled. It was subsequent events that caused the
>economy to tank.

Nope. Money supply growth was already being reversed.

>You think it is silly because you have the silly notion
>that the economy and not the stock market crashed on particular day.

The particular day is irrelevant. Profits based on rapid money supply
growth were not going continue without it.

-- Roy L

Igor

unread,
Oct 30, 2004, 3:06:15 AM10/30/04
to
ro...@telus.net wrote:

> On Fri, 29 Oct 2004 01:39:29 GMT, Igor <jjwea...@houston.rr.com>
> wrote:
>
>
>>ro...@telus.net wrote:
>>
>>
>>>On Wed, 27 Oct 2004 20:33:56 GMT, Igor <jjwea...@houston.rr.com>
>>>wrote:
>>>
>>>
>>>>If you believe that mud pies have the
>>>>same cost as apple pies because labor is paid is the same and they are
>>>>made on the same land then anything seems idiotic.
>>>
>>><sigh> I can spend $1K to make apple pies, or mud pies. That makes
>>>their resource costs identical. Their values are not identical. You
>>>(and Prescott) stand refuted.
>>
>>No it does not make them the same.
>
>
> Yes, it does. $1K = $1K. I am not sure how you prevent yourself from
> knowing such facts, but it is clear that you manage to do it somehow.
>

1K of machines is not equal to one 1K of mud. You get a lot fewer
machines and they are more productive. That is what real measures are
used to isolate the effects of inflation. This puts the analysis is
units not dollars. You forget that the price deflators on these goods
will be different. So 1K of spending to buy a computer a year will not
have the same real valuation of 1k spent on a truck last year.

>
>>How do you think that capital should
>>be empirically measured.
>
>
> Depends on the purpose. In any case, book value is not the same as
> ability to generate profits.
>

Nice way to dodge the question. So you have no alternative to how the
measure the stock we wish to get a return on. In other words you just
want to complain not offer a solution.

>>That
>>means you adjust for inflation to get a measure in real units. How do
>>you suggest to do that other than add up investment( what was paid for
>>the capital) and subtract depreciation.
>
>
> Market value would be better.

How do you determine that if the goods are not being sold. A year old
machine with wear and tear is not going to sell for the same price as a
brand new machine.

> Having paid $1K for something does not
> make it worth $1K,

Yes it does if you are using an objective value. If the good sales for
1K it is worth 1K. Unless you are coming for a labor theory of value or
some other archaic obfusicated theory of value which says sale price is
not equal to the value.

> and most certainly does not mean it will generate
> equivalent profits.

Why not? The willingness to pay for the good is based on the
producitivity the capital brings. Willingnes to sell is based on cost.
The market will agree on a price that says Cost = Value or Price =
expected discounted return on the good. Price determines the amount
paid. Therefore resource cost, a fancy way of saying what was paid for
it, will reflect the value to buyer and the cost to the seller.

> Market value at least indicates a collective
> judgment of the profit stream an asset can be expected to yield.
>

How do you calculate this. If you spent 1K buying the goods on the
market then how can say that was not market value. Are you seriously
saying that would try to estimate the resell price on every single piece
of equiptment a firm given age, wear and tear, etc to estimate a stock.
I hate to tell but that DATA DOES NOT EXIST.


>>So how do you propose it
>>should be measured?
>
>
> The problem with trying to measure capital for economic purposes is
> that it is not homogenous, and different kinds of capital are not
> commensurable except by converting them to money. So if you want to
> estimate the value of capital, market liquidation value is probably
> the closest you can get.
>

How do you obtain that data when no firm keeps it? Book value may not be
the best way. In fact it is rarely used. The way economist construct
capital stocks is a little complicated. You have to use IO tables a
series of deflators for investment and it takes a lot of time and
effort. Saying that you add the prices paid for the good and subtract
deprication is the basic idea but actual methods are more complicated.
To get a sense of how it is done you can look up the NBER productivity
panel. The paper that documents the data has some descriptions of how
the measurements are made in practice. The potential problems and how
they have solved for some them.

If you are arguing the measure used is not perfect I have to agree.
However, there is not much better unless firms start keeping books much
differently than they do now.

Igor

unread,
Oct 30, 2004, 3:13:53 AM10/30/04
to
ro...@telus.net wrote:


>
> True. Average growth through history has been much lower than average
> interest rates. But growth doesn't go broke and reset to zero every
> so often, which might help explain the fact that average interest
> rates are so much higher than average growth.
>

What are you arguing. It is simple take two equations.

1. GDP(t+1) = (1+x)GDP(t)
2. Debt(t+1) = (1+i) debt(t)

It does not take rocket science to figure out that if i>x then debt is
growing faster than GDP. That means at some point it will converge to
GDP and become higher than GDP. Interest payment would eventaully be
higher than GDP. At this point you can not tax or borrow enough to pay
interest.

The analysis does assume the deficit occurs before interest payments are
made. Therefore, all interest is financed.


> The average interest rate on US government debt has been less than
> average US GDP growth. If interest ever becomes too politically
> burdensome, the Treasury can just print more money, resetting the
> nominal GDP growth rate to more than the interest rate on the debt.
>

No. The FED can print more money if it wishes. Congress can not force
the Fed to do anything. They would have to legislate the Fed out of
existance to do this. THE TREASURY CAN NOT PRINT MONEY. That is the
Fed's responsibility to control the money supply.

Interest rates are typically higher than growth rates of GDP.

Igor

unread,
Oct 30, 2004, 3:32:29 AM10/30/04
to
ro...@telus.net wrote:

I did not say there were used in the first cut of estimation. There are
two approaches to estimation used and reported. Again this is explained
IN THE PAPER. You have to read past the abstract. The equations are used
in the second estimation. Presscot uses data similar to Fisher for the
first section. I may be wrong but I think he uses values generated from
stock market data in the first section. I do not remember how exactly he
did it but he followed Fisher's method.

The second section takes into intangibles. The equations are way to find
out how much the intangibles are worth. That is the reason for equating
profits to returns. Capital stocks and there return can be measured. So
if you subtract capital returns from profits what is left is returns
from intangibles. In that section, the model is used as a way to
estimate the value of the unobserved intangible capital assets.

Again I do not have the paper in front of me but I believe the second
method was a robustness check to see if the Fisher results could be
obtained using a different method.


>>What facts are those that the stock market crashed due to excessive
>>margin calls not bad news about profits?
>
>
> It crashed because the profits weren't going to be there to support
> the prices, because the money the corporations had spent on their
> capital (resource cost) was not all spent wisely, given changing
> market conditions. Margin calls were just the trigger, not the cause.
>

Wrong. There was no reason to believe that. Fisher's point was there no
indications and no breaking news that caused expectations to drop. If
you take a freeze frame at that point in time there is absolutely no
indication the recession is about to begin. The dip in profits and the
recession that FED and Roosevelt would turn into the great depression
had not occurred yet. At the moment of the crash stocks were
undervalued. Where they undervalued 2 days later maybe not. Where they
undervalued a year later? Most likely not. The captial stocks may have
dropped or by the formulation intangibles may have dropped.

You arguing that because you know from hindsight a recession was coming
that at THAT moment stocks were undervalued. This is in correct. They
were not Over valued until later.


>
>>>>The statement surprised me but the
>>>>data supported it when Fisher did his analysis and Prescott's analysis
>>>>supported the statement as well.
>>>
>>>Except that history proved them both laughably wrong.
>>
>>How?
>
>
> By seeing stock prices collapse.
>

Stock prices can collaspe while the market is undervalued. Look at 1986.
Undervalued does not mean that the market will immediately jump or that
they are permenantly undervalued. The standard terms means a lower P/E
ratio. Prices are low compared to profits (earnings). At the time of the
crash data shows prices were still low related to the value of the firm.

It does not say one month or one year after the crash that stocks were
still under valued.

>
>>Why do think that after looking at the data. Just because the
>>market crashed due to margin calls not being met does not mean that the
>>stocks analyzed were not undervalued.
>
>
> Yes, it does. You say P = MC in a competitive market, but then
> proceed to claim the real-life price verdict of a real-life
> competitive market was somehow not valid.

That is because stocks do not have marginal cost. They are not a
produced good. Expected profits of the firm drives prices. If expected
profits are high then demand is rising and supply is decreasing. People
want to buy and few want to sell. Prices are high. This is not the same
concept as marginal cost. Stocks are not produced goods and what
determines supply and demand are different. P=MC does not make sense for
stocks because there are no marginal cost. They are an investment that
expect to gain returns.

Capital goods are produced products so if the market is competitive
supply represents MC. Demand represents value to the buyer. P wil equal MC.

>>Profits were high at that period of time.
>
>
> They weren't going to stay high.
>

This does not mean at that one point in time that stocks were
undervalued. Even after the crash profits were high for some period.

>
>>On the day the market crashed conditions were good. Labor
>>disputes had been settled. It was subsequent events that caused the
>>economy to tank.
>
>
> Nope. Money supply growth was already being reversed.
>
>

Yes but it had not taken effect yet. A reversal in policy does not fully
take effect over night. It takes time for the policy to "move through"
the system and have full effect. Freidman estimated at least 9 months
before monetary policy affects aggregate variables. In the 1920's it
actually took a little longer.


ro...@telus.net

unread,
Oct 30, 2004, 4:57:40 AM10/30/04
to
On Sat, 30 Oct 2004 07:13:53 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>ro...@telus.net wrote:
>
>> True. Average growth through history has been much lower than average
>> interest rates. But growth doesn't go broke and reset to zero every
>> so often, which might help explain the fact that average interest
>> rates are so much higher than average growth.
>
>What are you arguing. It is simple take two equations.
>
>1. GDP(t+1) = (1+x)GDP(t)
>2. Debt(t+1) = (1+i) debt(t)
>
>It does not take rocket science to figure out that if i>x then debt is
>growing faster than GDP. That means at some point it will converge to
>GDP and become higher than GDP. Interest payment would eventaully be
>higher than GDP. At this point you can not tax or borrow enough to pay
>interest.

You seem intent on ignoring the fact that in reality, that just can't
happen.

>> The average interest rate on US government debt has been less than
>> average US GDP growth. If interest ever becomes too politically
>> burdensome, the Treasury can just print more money, resetting the
>> nominal GDP growth rate to more than the interest rate on the debt.
>
>No.

Yes.

>The FED can print more money if it wishes. Congress can not force
>the Fed to do anything.

<sigh> I suspect Hummel can correct your ignorance on that score.
Congress can easily force the Fed to print more money, by just
spending without taxing.

>They would have to legislate the Fed out of
>existance to do this.

They legislated it into existence, didn't they?

>THE TREASURY CAN NOT PRINT MONEY. That is the
>Fed's responsibility to control the money supply.

And Congress can easily make it impossible for the Fed to refuse to
print money.

>Interest rates are typically higher than growth rates of GDP.

Not average interest rates on US government debt over the country's
whole history.

-- Roy L

Robert Vienneau

unread,
Oct 30, 2004, 5:35:57 AM10/30/04
to
In article <RaHgd.33367$186....@fe1.texas.rr.com>, Igor
<jjwea...@houston.rr.com> wrote:

> What are you arguing. It is simple take two equations.
>
> 1. GDP(t+1) = (1+x)GDP(t)
> 2. Debt(t+1) = (1+i) debt(t)

i represents the rate of interest. The second formula is wrong.
For example, it is inapplicable to the U.S. Federal debt right now.

Robert Vienneau

unread,
Oct 30, 2004, 7:19:46 AM10/30/04
to
In article <hsHgd.13294$lM1....@fe2.texas.rr.com>, Igor
<jjwea...@houston.rr.com> wrote:

"The 1929 Stock Market: Irving Fisher Was Right," Ellen R. McGrattan
and Edward C.Prescott
<http://research.mpls.frb.fed.us/research/sr/sr294.pdf>

> There are
> two approaches to estimation used and reported. Again this is explained
> IN THE PAPER. You have to read past the abstract. The equations are used
> in the second estimation. Presscot uses data similar to Fisher for the
> first section. I may be wrong but I think he uses values generated from
> stock market data in the first section. I do not remember how exactly he
> did it but he followed Fisher's method.
>
> The second section takes into intangibles. The equations are way to find
> out how much the intangibles are worth. That is the reason for equating
> profits to returns. Capital stocks and there return can be measured. So
> if you subtract capital returns from profits what is left is returns
> from intangibles. In that section, the model is used as a way to
> estimate the value of the unobserved intangible capital assets.
>
> Again I do not have the paper in front of me but I believe the second
> method was a robustness check to see if the Fisher results could be
> obtained using a different method.

As usual, Mr. Weatherby is incorrect about the whole line of argument.
Since Mr. Weatherby is an illiterate and innumerate idiot, it does
not do him any good to run his eyes over text.

McGrattan and Prescott argue that stocks were undervalued in 1929. To
make this argument, they need to compare two calculations of the
value of corporations - the market value and the "fundamental" value.

In Section 2, McGrattan and Prescott calculate market value. In
Section 3, McGrattan and Prescott estimate the sum of corporate
tangible and intangible capital - the fundamental value of
corporations - by applying a steady-state growth theory to data.

As far as I can see, McGrattan and Prescott nowhere claim to be
duplicating Fisher's calculations of fundamental value.

McGrattan and Prescott do not show that Fisher was correct. They
reach an absurd conclusion. What might somebody who was not an
idiot savant conclude from their paper? One of the following:

o What's seems initially absurd is actually correct.

o McGrattan and Prescott incorrectly apply the theory.

o The theory they are applying is false in these circumstances

o There's something wrong with the data.

By "data", I am not referring to their final estimates, but the
national accounting measures that go into their calculations.

A non-idiot that knows something about discussions of the stock
market knows that some have questioned whether any such thing
as "fundamental value" exists. An examination of certain properties
of steady state growth might help make this point. These properties
are highlighted in:

Robinson, J. (1953-1954). "The Production Function and the Theory
of Capital", Review of Economic Studies. V. 21, p. 81-106.

Along a steady-state growth path, the following three values are
equal:

o The sum of past costs in creating the current stock of capital
goods, costed up at the prevailing rate of interest

o The market value of capital goods (which is the same as the
replacement cost)

o The present value of the sum of future earnings from those
capital goods.

One might call the third value the fundamental value. But those
future earnings do not exist at a specific moment of time. They
depend on current and future actions of agents. So perhaps
fundamental value does not exist.

The U.S. economy was not in a steady state in 1929, and the second
value overstated the third value. This is easily seen by observing
the overcapacity existing in the next few years, the re-evaluation
of bad debts, etc. Since Prescott's theory is that the Great
Depression was a matter of workers deciding to take long vacations,
it is no wonder he does not see this.

In some sense, any stock is overvalued if it can be bought three
months later for a third less.

Where do McGrattan and Prescott go wrong in their estimations? Key
to their argument is that the estimate of the real interest be
conservative (on the high end). They state "0.0473 is our estimate
of the real interest rate based on the noncorporate sector".
McGratton and Prescott observe modern management techniques were
beginning to come on-line in the 1920s. This is all part of how
the corporate sector and non-corporate sector were beginning be
distinguished. Perhaps, the return to financial capital is
systematically higher in the corporate sector because of
barriers to entry and other non-competitive properties of that
sector. If so, McGratton and Prescott's estimation of the
real interest rate is low, not high.

What is needed is a growth theory that emphasizes this sort of
dualism in the modern economy. So one can read McGrattan and Prescott
as empirical evidence for Post Keynesian economics, if one wants.

At one point, the topic of this thread was Prescott (and Weatherby's)
idiotic ignorance of the distribution of income in the United States.
Mr. Weatherby has yet to put forth any data on topic.

sinister

unread,
Oct 30, 2004, 11:53:39 AM10/30/04
to

"Igor" <jjwea...@houston.rr.com> wrote in message
news:hsHgd.13294$lM1....@fe2.texas.rr.com...

In real time, what were aggregate P/E and P/D ratios? What was total market
cap to GDP?

Igor

unread,
Oct 30, 2004, 4:28:43 PM10/30/04
to
Robert Vienneau wrote:

> In article <hsHgd.13294$lM1....@fe2.texas.rr.com>, Igor
> <jjwea...@houston.rr.com> wrote:
>
> "The 1929 Stock Market: Irving Fisher Was Right," Ellen R. McGrattan
> and Edward C.Prescott
> <http://research.mpls.frb.fed.us/research/sr/sr294.pdf>
>
>>There are
>>two approaches to estimation used and reported. Again this is explained
>>IN THE PAPER. You have to read past the abstract. The equations are used
>>in the second estimation. Presscot uses data similar to Fisher for the
>>first section. I may be wrong but I think he uses values generated from
>>stock market data in the first section. I do not remember how exactly he
>>did it but he followed Fisher's method.
>>
>>The second section takes into intangibles. The equations are way to find
>>out how much the intangibles are worth. That is the reason for equating
>>profits to returns. Capital stocks and there return can be measured. So
>>if you subtract capital returns from profits what is left is returns
>>from intangibles. In that section, the model is used as a way to
>>estimate the value of the unobserved intangible capital assets.
>>
>>Again I do not have the paper in front of me but I believe the second
>>method was a robustness check to see if the Fisher results could be
>>obtained using a different method.
>
>

> McGrattan and Prescott argue that stocks were undervalued in 1929. To


> make this argument, they need to compare two calculations of the
> value of corporations - the market value and the "fundamental" value.
>

Yes two different measures. Roy did not know this because he did not
read the paper.


> In Section 2, McGrattan and Prescott calculate market value. In
> Section 3, McGrattan and Prescott estimate the sum of corporate
> tangible and intangible capital - the fundamental value of
> corporations - by applying a steady-state growth theory to data.
>
> As far as I can see, McGrattan and Prescott nowhere claim to be
> duplicating Fisher's calculations of fundamental value.
>

The whole point was to see if the results could be duplicated. They
constantly compared the method and data to Fisher. Read the paper. They
have a discussion about how they have some data Fisher did not have.


> A non-idiot that knows something about discussions of the stock
> market knows that some have questioned whether any such thing
> as "fundamental value" exists. An examination of certain properties
> of steady state growth might help make this point. These properties
> are highlighted in:
>
> Robinson, J. (1953-1954). "The Production Function and the Theory
> of Capital", Review of Economic Studies. V. 21, p. 81-106.
>
> Along a steady-state growth path, the following three values are
> equal:
>
> o The sum of past costs in creating the current stock of capital
> goods, costed up at the prevailing rate of interest
>
> o The market value of capital goods (which is the same as the
> replacement cost)
>
> o The present value of the sum of future earnings from those
> capital goods.
>
> One might call the third value the fundamental value. But those
> future earnings do not exist at a specific moment of time. They
> depend on current and future actions of agents. So perhaps
> fundamental value does not exist.
>

Yes and no. It does not necessicarily have to be future earnings from
capital goods. For instance in two chapters of Grossman and Helpman the
concept is applied to a model with no capital at all. Both chapters have
models where consumption goods change due to R&D. Chapter 3 talks about
expanding variety. Chapter 4 shows quality change. In Grossman and
Helpman the fundemental value is the present value of future profits but
NO CAPITAL IS INVOLVED. Theoritically this is fundemental value.
Presscot is basing these formulations for Grossman and Helpman and other
papers on new growth theory that stress imperfect competition.

Now obviously this fundemental can not be measured as theory
states.Therefore Presscot has use a second best solution which is the
second ideal. The market value is the cost of the goods deflated and
depreciated.

> The U.S. economy was not in a steady state in 1929, and the second
> value overstated the third value.

To say this is misleading. The US was in a long run steady state. A
deviation was about to happen. The truth is without steady state
assumptions it is difficult to get measurements. No one argues this is
perfect. Yet, you have to make some assumptions in able to construct
measurements.

>This is easily seen by observing
> the overcapacity existing in the next few years,

The next few years not at that point. There is no doubt a year later the
recession had begun. It is more doubtful that the recession had begun
when the stock market crashed.

> Since Prescott's theory is that the Great
> Depression was a matter of workers deciding to take long vacations,
> it is no wonder he does not see this.
>

Absolutely not. This paper assumes nothing about Business cycle models
prevailing. In fact time inconsistency problems deal with RATIONAL
EXPECTATIONS which can be seen as an extension of Keynes and not REAL
BUSINESS CYCLE MODELS. Under RBC things like consistency do not matter.
Recessions are caused by things other than the Fed.


>
> Where do McGrattan and Prescott go wrong in their estimations? Key
> to their argument is that the estimate of the real interest be
> conservative (on the high end). They state "0.0473 is our estimate
> of the real interest rate based on the noncorporate sector".
> McGratton and Prescott observe modern management techniques were
> beginning to come on-line in the 1920s. This is all part of how
> the corporate sector and non-corporate sector were beginning be
> distinguished. Perhaps, the return to financial capital is
> systematically higher in the corporate sector because of
> barriers to entry and other non-competitive properties of that
> sector. If so, McGratton and Prescott's estimation of the
> real interest rate is low, not high.
>
> What is needed is a growth theory that emphasizes this sort of
> dualism in the modern economy. So one can read McGrattan and Prescott
> as empirical evidence for Post Keynesian economics, if one wants.
>

One can also read it as evidence for the growth theory it is based upon.
That is a theory based on imperfect competition and in very recent work
the importance of firm differences. These models show that firm profit
maximizing decision decide on levels of R&D and therefore drive economic
growth. Barriers to entry are the basis of these models. That is why I
can not understand your constant resistance to any of this theory. It is
stating in a sense that Robinson was right. Imperfect competition is
imporant and necessary. I think this is where you misunderstand what is
mainstream.


Presscot bases this paper on Grossman and Helpman 1991. He is very
unclear about the growth theory he says he bases this on. This is a very
weak portion of the paper. Knowing the subject quite intimately I can
place his estimation of fundemental value solely on Grossman and Helpman
1991 Chapter 3. This is the nuts and bolts approach to valuing a firm
applied in Grossman and Helpman. The approach is necessary to find a no
arbritage condition in which models such as Peretto 1998 used to be able
to account for entry.

Igor

unread,
Oct 30, 2004, 4:35:06 PM10/30/04
to
sinister wrote:

I do not remember and I have lost the link to the paper. Presscot
reports the P/E ratios in the paper. I am sorry I will have to refer you
to the paper for the actual numbers. If you would have read the paper
before commenting you would realize that. You posted the link I assume
you can still find the paper. So why don't you read it to answer your
questions. I can only provide a summary of what Presscot said and
comment on the validity of methods. So I can not say more than the
paper. In some cases I can answer questions about methods but I
understand them but that is all the additional insight I have. I do not
have the data nor have I tried to replicate his results. Honestly, I do
not have enough interest in the question to do either. In fact, I am
surprised I have rewritten this much on a topic that really does not
interest me. The motivation is more to show that the Noble is not a fake
prize and that Presscot does know something about what is doing than to
offer another study confirming him.

ro...@telus.net

unread,
Oct 30, 2004, 8:53:09 PM10/30/04
to
On Sat, 30 Oct 2004 07:32:29 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>ro...@telus.net wrote:
>
>> On Fri, 29 Oct 2004 01:49:18 GMT, Igor <jjwea...@houston.rr.com>
>> wrote:
>>
>>>In the first section
>>>the equations you do not understand and think are stupid ARE NOT USED.
>>
>> Then why did you say they were?
>
>I did not say there were used in the first cut of estimation. There are
>two approaches to estimation used and reported. Again this is explained
>IN THE PAPER. You have to read past the abstract. The equations are used
>in the second estimation. Presscot uses data similar to Fisher for the
>first section. I may be wrong but I think he uses values generated from
>stock market data in the first section. I do not remember how exactly he
>did it but he followed Fisher's method.
>
>The second section takes into intangibles. The equations are way to find
>out how much the intangibles are worth.

No, they aren't. They are a way to find out the value the intangibles
were expected to have in 1929, at the time they were bought.

>That is the reason for equating
>profits to returns. Capital stocks and there return can be measured.

Not a priori, they can't.

>So if you subtract capital returns from profits what is left is returns
>from intangibles. In that section, the model is used as a way to
>estimate the value of the unobserved intangible capital assets.

Which don't remotely follow P = MC.

>>>What facts are those that the stock market crashed due to excessive
>>>margin calls not bad news about profits?
>>
>> It crashed because the profits weren't going to be there to support
>> the prices, because the money the corporations had spent on their
>> capital (resource cost) was not all spent wisely, given changing
>> market conditions. Margin calls were just the trigger, not the cause.
>
>Wrong. There was no reason to believe that.

Wrong. The Fed had already slammed on the brakes.

>Fisher's point was there no
>indications and no breaking news that caused expectations to drop.

Fisher pretended the Fed didn't exist.

>If you take a freeze frame at that point in time there is absolutely no
>indication the recession is about to begin.

Stock speculators are smarter than movie cameras, let alone still
cameras.

>The dip in profits and the
>recession that FED and Roosevelt would turn into the great depression
>had not occurred yet.

It was in the cards.

>At the moment of the crash stocks were
>undervalued.

Nope. They were _about_ to become _over_valued. That is very much
the point. Stock value is based on _future_expectations_of_profit_,
not past expenditures on asset acquisition.

>Where they undervalued 2 days later maybe not. Where they
>undervalued a year later? Most likely not. The captial stocks may have
>dropped or by the formulation intangibles may have dropped.

Stocks are value on expected returns, not returns hoped for when
assets were acquired.

>You arguing that because you know from hindsight a recession was coming
>that at THAT moment stocks were undervalued. This is in correct. They
>were not Over valued until later.

Wrong. They were overvalued the moment Fed policy became more
restrictive.



>>>>>The statement surprised me but the
>>>>>data supported it when Fisher did his analysis and Prescott's analysis
>>>>>supported the statement as well.
>>>>
>>>>Except that history proved them both laughably wrong.
>>>
>>>How?
>>
>> By seeing stock prices collapse.
>
>Stock prices can collaspe while the market is undervalued. Look at 1986.
>Undervalued does not mean that the market will immediately jump or that
>they are permenantly undervalued. The standard terms means a lower P/E
>ratio. Prices are low compared to profits (earnings).

But stock prices are not based only on current earnings, but mainly on
expected earnings.

>At the time of the
>crash data shows prices were still low related to the value of the firm.

Data shows no such thing. It shows only that investments had been
made in expectation of returns that the Fed was then in the process of
evaporating.

>>>Why do think that after looking at the data. Just because the
>>>market crashed due to margin calls not being met does not mean that the
>>>stocks analyzed were not undervalued.
>>
>> Yes, it does. You say P = MC in a competitive market, but then
>> proceed to claim the real-life price verdict of a real-life
>> competitive market was somehow not valid.
>
>That is because stocks do not have marginal cost.

Neither do rent collection privileges, some of which you call
"capital" and some "intangibles."

>They are not a
>produced good. Expected profits of the firm drives prices.

The prices of its assets are likewise driven by what the market thinks
they will return, not by what they cost to acquire.

>P=MC does not make sense for
>stocks because there are no marginal cost. They are an investment that
>expect to gain returns.

Like corporate assets. Right. So contrary to Prescott, _neither_ of
them can be valued based on cost.

>Capital goods are produced products so if the market is competitive
>supply represents MC.

Only to acquire them new. And not all the "capital" Prescott makes up
values for is capital goods. Land is not a product, and thus cannot
be valued by marginal cost.

>Demand represents value to the buyer. P wil equal MC.

No, it won't, especially for intangibles.

>>>Profits were high at that period of time.
>>
>> They weren't going to stay high.
>
>This does not mean at that one point in time that stocks were
>undervalued. Even after the crash profits were high for some period.

Irrelevant. So were Enron's.



>>>On the day the market crashed conditions were good. Labor
>>>disputes had been settled. It was subsequent events that caused the
>>>economy to tank.
>>
>> Nope. Money supply growth was already being reversed.
>
>Yes but it had not taken effect yet.

ROTFL!! What do you think caused the crash??

>A reversal in policy does not fully
>take effect over night.

Stock speculators are not as stupid as you and Prescott.

>It takes time for the policy to "move through"
>the system and have full effect.

And meanwhile, everyone just pretends they don't see it coming?
ROTFL!!

>Freidman estimated at least 9 months
>before monetary policy affects aggregate variables. In the 1920's it
>actually took a little longer.

It very obviously didn't take longer to hit the stock market.

-- Roy L

Igor

unread,
Oct 30, 2004, 9:43:44 PM10/30/04
to
ro...@telus.net wrote:
> On Sat, 30 Oct 2004 07:32:29 GMT, Igor <jjwea...@houston.rr.com>
> wrote:
>
>
>>ro...@telus.net wrote:
>>
>>
>>>On Fri, 29 Oct 2004 01:49:18 GMT, Igor <jjwea...@houston.rr.com>
>>>wrote:
>>>
>>>
>>>>In the first section
>>>>the equations you do not understand and think are stupid ARE NOT USED.
>>>
>>>Then why did you say they were?
>>
>>I did not say there were used in the first cut of estimation. There are
>>two approaches to estimation used and reported. Again this is explained
>>IN THE PAPER. You have to read past the abstract. The equations are used
>>in the second estimation. Presscot uses data similar to Fisher for the
>>first section. I may be wrong but I think he uses values generated from
>>stock market data in the first section. I do not remember how exactly he
>>did it but he followed Fisher's method.
>>
>>The second section takes into intangibles. The equations are way to find
>>out how much the intangibles are worth.
>
>
> No, they aren't. They are a way to find out the value the intangibles
> were expected to have in 1929, at the time they were bought.
>

running in circles again. I have explained this at least twice now I
will not explain it again. It is simple profit = return on tangible
capital plus return on intangibles. If you know profit and you know the
return on real capital then return on intangibles is profit - return on
tangible capital. It is a simple statement. What you argue is if the
psychical capital was MEASURED correctly. So far you have presented
absolutely no tractable alternative to measuring capital. I am not going
to run around in circles with you.

>>So if you subtract capital returns from profits what is left is returns
>
>>from intangibles. In that section, the model is used as a way to
>
>>estimate the value of the unobserved intangible capital assets.
>
>
> Which don't remotely follow P = MC.
>

Of course it does. Returns are equal to the interest rate * K. You use
the interest rate because in competitive markets returns on assets will
be equal therefore no one has incentive to switch assets unless
conditions change.

Now how do you measure capital at market value? The Price of one unit is
EQUAL TO ITS MARGINAL COST if the market is competitive. So the market
value is a measure of the cost of the good. You deflate these prices to
remove pure inflationary increases. Yes there are problems with
deflators and sometimes they take out quality increases instead of pure
inflation. After you do that you add us the values to get the value of
the stock. The problem is some of the stock has devalued due to wear and
tear. So you deprecate the capital at a fixed rate to take care of this.


In other words Change in K = Investment(deflated) - (deprecation
rate)K(t-1) . That is how you build a stock using the perpetual
inventory method. This is what Presscot shows in equation 3.

The stock is resource cost - minus deprecation because the price goods
were bought for reflect the marginal cost of the good.

>
>>>>What facts are those that the stock market crashed due to excessive
>>>>margin calls not bad news about profits?
>>>
>>>It crashed because the profits weren't going to be there to support
>>>the prices, because the money the corporations had spent on their
>>>capital (resource cost) was not all spent wisely, given changing
>>>market conditions. Margin calls were just the trigger, not the cause.
>>
>>Wrong. There was no reason to believe that.
>
>
> Wrong. The Fed had already slammed on the brakes.
>

I do not think that was until after the market crash. I think a few days
after. I would have to look up a time line to make sure but I do not
think the Fed. actions were before the crash.


>>At the moment of the crash stocks were
>>undervalued.
>
>
> Nope. They were _about_ to become _over_valued. That is very much
> the point. Stock value is based on _future_expectations_of_profit_,
> not past expenditures on asset acquisition.
>


Exactly. WHAT IS THE DEFINITION OF UNDERVALUED. The definition says that
expectations of future profits ARE BELOW CURRENT PROFITS. A low P/E
ratio means that stocks are undervalued because certain expectations
said this corporation was expected to make less than it is making now.

To say a stock is undervalued DOES NOT MEAN THAT THE PRICE WILL RISE. It
says people's expectations about future profits are less than current
profits. This can be a good sign for buying a stock but not always.
Somethings expectations about future profits are correct and investing
in an "undervalued" stock will cause you to lose money because future
profits really do fall. Some times they are wrong and you gain money as
the market adjusts it is expectations and send prices up. To Say a stock
is undervalued is not to say that PRICE WILL RISE. It is no guarantee of
a good buy. It says current expectations are that the corporation will
make less money than it is making now. It could because investors are
correct. It could also be because investors were surprised and the
corporation MADE MUCH MORE THAN EXPECTED.

To say stocks were undervalued in this case makes a strong case for
rational expectations. It says that people set expectations and they set
them correctly. The crash was an immediate adjustment of expectations.
If you want to say it was because of Fed policy then you have proved the
point was true. People adjusted expectations quickly causing low P/E
ratios. Undervalued IS NOT a 100% prediction of stocks rising. It says
that if people are wrong there is a potential to make money once
expectations adjust upward. An overvalued stock does not mean that price
will rise. It says that it has a low potential to make money because
people are already pricing the stock at expectations of higher profits.
So if profits rise meeting that expectation there is little gain to be
made. It is much riskier to invest because if expectations are wrong you
lose big time. You bought a stock at high and when profits are shown low
expectations will cause prices to drop.

Undervalued means a POTENTIAL to make big money. It means expectations
say low profits. While current profits are high profits. If you invest
and expectations are WRONG they you make a lot of money because price
will rise if expectations were wrong. If expectations were right you
don't lose much because you bought the stock at sometime close to the
future price.

The misunderstanding is what undervalued means. Undervalued means
expectations say that profits will be lower in the future. It DOES NOT
say that prices will rise. A stock is not necessiarly overvalued if the
price drops in three months. It is overvalued if PEOPLE EXPECTED PROFITS
to be much higher than they actually were.

>
>>Where they undervalued 2 days later maybe not. Where they
>>undervalued a year later? Most likely not. The captial stocks may have
>>dropped or by the formulation intangibles may have dropped.
>
>
> Stocks are value on expected returns, not returns hoped for when
> assets were acquired.
>

That is the price. Under and Over valued are determined by comparing
STOCK PRICES TO CURRENT PROFITS. It says does the market expect this
firm to have lower or higher profits.

>
>>You arguing that because you know from hindsight a recession was coming
>>that at THAT moment stocks were undervalued. This is in correct. They
>>were not Over valued until later.
>
>
> Wrong. They were overvalued the moment Fed policy became more
> restrictive.
>

No because PROFITS WERE STILL HIGH and PEOPLE EXPECTED PROFITS TO BE
LOWER LATER. That means it was UNDERVALUED. The P/E ratio was low.

Prescott calculates these P/E ratios. I do not remember what exactly
they were.


>
>>>>>>The statement surprised me but the
>>>>>>data supported it when Fisher did his analysis and Prescott's analysis
>>>>>>supported the statement as well.
>>>>>
>>>>>Except that history proved them both laughably wrong.
>>>>
>>>>How?
>>>
>>>By seeing stock prices collapse.
>>
>>Stock prices can collaspe while the market is undervalued. Look at 1986.
>>Undervalued does not mean that the market will immediately jump or that
>>they are permenantly undervalued. The standard terms means a lower P/E
>>ratio. Prices are low compared to profits (earnings).
>
>
> But stock prices are not based only on current earnings, but mainly on
> expected earnings.
>

Yes but UNDER AND OVER VALUED JUDGMENTS ARE BASED ON EXPECTED EARNINGS
COMPARED TO ACTUAL CURRENT EARNINGS. If the economy is headed for
recession and people know it then stocks will be undervalued. You will
have a low price while profits are still high. This is EXACTLY THE
SITUATION AT THE TIME OF THE 1929 CRASH.


>
>>At the time of the
>>crash data shows prices were still low related to the value of the firm.
>
>
> Data shows no such thing. It shows only that investments had been
> made in expectation of returns that the Fed was then in the process of
> evaporating.
>

Current profits were still high. Prices had already begun to drop or the
market had not expected the profits to be as high at that point. The
data show this.

>>They are not a
>>produced good. Expected profits of the firm drives prices.
>
>
> The prices of its assets are likewise driven by what the market thinks
> they will return, not by what they cost to acquire.
>

Exactly and when expectations of future profits lead to lower prices and
at that moment firms are making high profits what do you call that?
UNDERVALUED STOCKS.


>
>>P=MC does not make sense for
>>stocks because there are no marginal cost. They are an investment that
>>expect to gain returns.
>
>
> Like corporate assets. Right. So contrary to Prescott, _neither_ of
> them can be valued based on cost.
>

Corporate assets are produced and bought on the market. A machine has a
market value that reflects cost. A building has a market value that
reflects cost of production, and so on.

>
>>Capital goods are produced products so if the market is competitive
>>supply represents MC.
>
>
> Only to acquire them new.

Right WHICH IS WHY DEPRECIATION IS TAKEN OUT OF THE VALUE>

>And not all the "capital" Prescott makes up
> values for is capital goods. Land is not a product, and thus cannot
> be valued by marginal cost.
>

Land is not capital Roy. I hate to break it to you but is considered an
entire different resource. Resources are land, labor, capital, raw
materials, and energy. LAND IS NOT CAPITAL.


>
>>Demand represents value to the buyer. P wil equal MC.
>
>
> No, it won't, especially for intangibles.
>

That is why intangibles are a residual. We can not observe them. They
often have 0 MC but have a price above zero. This is why you have to
subtract the returns on tangible capital from Profits to estimate the
amount of return on intangibles. YOU CAN NOT OBSERVE THEM THAT IS WHY
THEY ARE INTANGIBLES. No cost measure is used to construct intangibles
THEY ARE A RESIDUAL.


>
>>>>Profits were high at that period of time.
>>>
>>>
>>>They weren't going to stay high.
>>
>>This does not mean at that one point in time that stocks were
>>undervalued. Even after the crash profits were high for some period.
>
>
> Irrelevant. So were Enron's.
>

It is releveant because UNDERVALUED MEANS EXPECTED PROFITS ARE BELOW
ACTUAL PROFITS. So at that moment the stock is undervalue. As I
explained that does not preclude future expectations are wrong and the
PROFITS COULD DROP causing the stock to be overvalued at a later time.

>
>>>>On the day the market crashed conditions were good. Labor
>>>>disputes had been settled. It was subsequent events that caused the
>>>>economy to tank.
>>>
>>>Nope. Money supply growth was already being reversed.
>>
>>Yes but it had not taken effect yet.
>
>
> ROTFL!! What do you think caused the crash??
>

Expectations of future profits could have caused this. You even admitted
this that is expectations of FUTURE PROFITS that drive prices not
CURRENT PROFITS. Over or under value refers to if profits are higher or
lower then people expect them to be.

A good portion of the crash had nothing to do with though. It was the
fact that people borrowed way too much to buy stocks. As EXPECTED
profits began to decrease they could not meet the margin calls. The
market open with tons of stocks needing to be dumped to meet margin. At
that point expectations did not matter. Supply was increasing not
because expectations of future profits were dropping but because people
had to liquidate to meet margins. As the sellers flooded in prices
dropped causing more stocks to be sold to meet the margin. It started a
deadly spiral.


>
>>A reversal in policy does not fully
>>take effect over night.
>
>
> Stock speculators are not as stupid as you and Prescott.
>

Exactly that is why expected profits were dropping causing the stock
prices to be UNDERVALUED. I do not think they are stupid nor does
Presscot. The misconception is that you do not understand what
UNDERVALUED means.

>
>>It takes time for the policy to "move through"
>>the system and have full effect.
>
>
> And meanwhile, everyone just pretends they don't see it coming?
> ROTFL!!
>

No as expectations drop stock prices drop but PROFITS ARE STILL HIGH
BECAUSE THE POLICY HAS NOT EFFECTED REAL VARIABLES YET. Therefore Stocks
are undervalued. Prices are very low while profits are very high. As I
said before this is NO guarantee stock prices will rise or the market is
wrong. When you invest in an undervalued stock you are betting the
market's expectations are off.

>
>>Freidman estimated at least 9 months
>>before monetary policy affects aggregate variables. In the 1920's it
>>actually took a little longer.
>
>
> It very obviously didn't take longer to hit the stock market.
>

No because those are expectations but CURRENT profits would not be
affected for some time. That sets the stock in an undervalued state.

AGAIN RTFM. This is explained. The P/E ratios are calculated. I am not
sure Presscot gives the definition of undervalued but the definition I
give is what he is referring to.

Robert Vienneau

unread,
Oct 31, 2004, 1:35:59 AM10/31/04
to
In article <krXgd.29451$EI6....@fe2.texas.rr.com>, Igor
<jjwea...@houston.rr.com> wrote:

"The 1929 Stock Market: Irving Fisher Was Right," Ellen R. McGrattan
and Edward C.Prescott
<http://research.mpls.frb.fed.us/research/sr/sr294.pdf>

> Returns are equal to the interest rate * K. You use

> the interest rate because in competitive markets returns on assets will
> be equal therefore no one has incentive to switch assets unless
> conditions change.
>
> Now how do you measure capital at market value? The Price of one unit is
> EQUAL TO ITS MARGINAL COST if the market is competitive.

That statement is close to the claim that the interest rate
is equal to the marginal product of capital. So I think it
mistaken, too.

> Change in K = Investment(deflated) - (deprecation
> rate)K(t-1) . That is how you build a stock using the perpetual
> inventory method.

Technically, the above equation is incorrect. The theoretically
exact way to treat fixed capital is as a kind of joint production.
Accounting has a lot of convention built into it.

> WHAT IS THE DEFINITION OF UNDERVALUED. The definition says that
> expectations of future profits ARE BELOW CURRENT PROFITS.

Nope. And that is not the definition McGrattan and Prescott use.

> Resources are land, labor, capital, raw
> materials, and energy.

"These ambiguities, redundancies, and deficiencies recall those
attributed by Dr. Franz Kuhn to a certain Chinese encyclopedia
entitled Celestial Emporium of Benevolent Knowledge. On those remote
pages it is written that animals are divided into (a) those that
belong to the Emperor, (b) embalmed ones, (c) those that are trained,
(d) suckling pigs, (e) mermaids, (f) fabulous ones, (g) stray dogs,
(h) those that are included in this classification, (i) those that
tremble as if they were mad, (j) innumerable ones, (k) those drawn
with a very fine camel's hair brush, (l) others, (m) those that have
just broken a flower vase, (n) those that resemble flies from a
distance."
-- Jorge Luis Borges

Robert Vienneau

unread,
Oct 31, 2004, 5:10:19 AM10/31/04
to
In article <7OOgd.179$cA4.53@trnddc01>, "sinister"
<sini...@nospam.invalid> wrote:

> "Igor" <jjwea...@houston.rr.com> wrote in message
> news:hsHgd.13294$lM1....@fe2.texas.rr.com...

> In real time, what were aggregate P/E and P/D ratios? What was total
> market
> cap to GDP?

"we find that conservative estimate for the fundamental value of
U.S.corporations in 1929 was ... 21.6 times 1929 after-tax
corporate earnings."

They do not comment on whether this is above historical trends.

sinister

unread,
Oct 31, 2004, 5:40:56 AM10/31/04
to

"Robert Vienneau" <rv...@see.sig.com> wrote in message
news:rvien-AC2C1A....@news.dreamscape.com...

> In article <7OOgd.179$cA4.53@trnddc01>, "sinister"
> <sini...@nospam.invalid> wrote:
>
>> "Igor" <jjwea...@houston.rr.com> wrote in message
>> news:hsHgd.13294$lM1....@fe2.texas.rr.com...
>
>> In real time, what were aggregate P/E and P/D ratios? What was total
>> market
>> cap to GDP?
>
> "we find that conservative estimate for the fundamental value of
> U.S.corporations in 1929 was ... 21.6 times 1929 after-tax
> corporate earnings."

Checking with a google search on
P/E US stocks 1929
the first two pages with an actual number claimed that P/E before the crash
was about 60.

Given the games one can place with "E," and given Prescott's quality of work
and reasoning, I'd wager they're using a very liberal version of "E".

ro...@telus.net

unread,
Oct 31, 2004, 1:20:24 PM10/31/04
to
On Sat, 30 Oct 2004 07:06:15 GMT, Igor <jjwea...@houston.rr.com>
wrote:

>ro...@telus.net wrote:
>
>> On Fri, 29 Oct 2004 01:39:29 GMT, Igor <jjwea...@houston.rr.com>
>> wrote:
>>
>>>ro...@telus.net wrote:
>>>
>>>>On Wed, 27 Oct 2004 20:33:56 GMT, Igor <jjwea...@houston.rr.com>
>>>>wrote:
>>>>
>>>>>If you believe that mud pies have the
>>>>>same cost as apple pies because labor is paid is the same and they are
>>>>>made on the same land then anything seems idiotic.
>>>>
>>>><sigh> I can spend $1K to make apple pies, or mud pies. That makes
>>>>their resource costs identical. Their values are not identical. You
>>>>(and Prescott) stand refuted.
>>>
>>>No it does not make them the same.
>>
>> Yes, it does. $1K = $1K. I am not sure how you prevent yourself from
>> knowing such facts, but it is clear that you manage to do it somehow.
>
>1K of machines is not equal to one 1K of mud.

The resource cost is the same.

>You get a lot fewer
>machines and they are more productive.

The formula Prescott used says _nothing_ about productivity. It says
value comes from cost, whether there is any production or not.
Period.

>That is what real measures are
>used to isolate the effects of inflation. This puts the analysis is
>units not dollars. You forget that the price deflators on these goods
>will be different. So 1K of spending to buy a computer a year will not
>have the same real valuation of 1k spent on a truck last year.

Gibberish.



>>>How do you think that capital should
>>>be empirically measured.
>>
>> Depends on the purpose. In any case, book value is not the same as
>> ability to generate profits.
>
>Nice way to dodge the question. So you have no alternative to how the
>measure the stock we wish to get a return on.

Oh, _that's_ easy enough to measure: whatever was spent. But wishes
do not create value.

>>>That
>>>means you adjust for inflation to get a measure in real units. How do
>>>you suggest to do that other than add up investment( what was paid for
>>>the capital) and subtract depreciation.
>>
>> Market value would be better.
>
>How do you determine that if the goods are not being sold.

Honestly.

>A year old
>machine with wear and tear is not going to sell for the same price as a
>brand new machine.

Bingo. And pretending it will (even depreciated by the interest rate)
is just stupid.

>> Having paid $1K for something does not
>> make it worth $1K,
>
>Yes it does if you are using an objective value.

No, it doesn't if you are using an objective value, as the mud pies
_prove_.

>If the good sales for
>1K it is worth 1K.

Wrong. It _was_ worth $1K.

>Unless you are coming for a labor theory of value or
>some other archaic obfusicated theory of value which says sale price is
>not equal to the value.

_Current_ price is equal to the value. Not deflated purchase price.

>> and most certainly does not mean it will generate
>> equivalent profits.
>
>Why not?

Because people are fallible.

>The willingness to pay for the good is based on the
>producitivity the capital brings.

Nope. It's based on the _hoped_for_ return. Not the _actual_ return.

>Willingnes to sell is based on cost.

Nope. Expected return.

>The market will agree on a price that says Cost = Value or Price =
>expected discounted return on the good. Price determines the amount
>paid. Therefore resource cost, a fancy way of saying what was paid for
>it, will reflect the value to buyer and the cost to the seller.

Only at the time of the exchange.

>> Market value at least indicates a collective
>> judgment of the profit stream an asset can be expected to yield.
>
>How do you calculate this.

Not a priori, that's for sure.

>If you spent 1K buying the goods on the
>market then how can say that was not market value.

It _was_. That doesn't mean it _is_.

>Are you seriously
>saying that would try to estimate the resell price on every single piece
>of equiptment a firm given age, wear and tear, etc to estimate a stock.
>I hate to tell but that DATA DOES NOT EXIST.

Right. So Prescott just made some up.

>>>So how do you propose it
>>>should be measured?
>>
>> The problem with trying to measure capital for economic purposes is
>> that it is not homogenous, and different kinds of capital are not
>> commensurable except by converting them to money. So if you want to
>> estimate the value of capital, market liquidation value is probably
>> the closest you can get.
>
>How do you obtain that data when no firm keeps it?

By comparing actual liquidation proceeds of bankrupt firms with the
kind of notional accounting-entry asset values Prescott used.

>Book value may not be
>the best way. In fact it is rarely used.

I agree book value also has its problems.

-- Roy L

Igor

unread,
Oct 31, 2004, 2:25:03 PM10/31/04
to
ro...@telus.net wrote:

> On Sat, 30 Oct 2004 07:06:15 GMT, Igor <jjwea...@houston.rr.com>
> wrote:
>
>
>>ro...@telus.net wrote:
>>
>>
>>>On Fri, 29 Oct 2004 01:39:29 GMT, Igor <jjwea...@houston.rr.com>
>>>wrote:
>>>
>>>
>>>>ro...@telus.net wrote:
>
>>A year old
>>machine with wear and tear is not going to sell for the same price as a
>>brand new machine.
>
>
> Bingo. And pretending it will (even depreciated by the interest rate)
> is just stupid.
>

Roy RTFM. I am tired of trying to have a discussion and you enter
complete misconceptions like depreciation = the interest rate. RTM. The
interest rate is not the rate of decprication.

You can not discuss something if you do not know what is says.

>>If the good sales for
>>1K it is worth 1K.
>
>
> Wrong. It _was_ worth $1K.
>

Which is why a measure of depreciation has to be subtracted from
purchase prices.


>
>>Unless you are coming for a labor theory of value or
>>some other archaic obfusicated theory of value which says sale price is
>>not equal to the value.
>
>
> _Current_ price is equal to the value. Not deflated purchase price.
>

Deflation puts todays dollars in terms of yesterdays PURCHASING POWER.
It adjust for inflation and gives a constant dollar measure. Not
deflating means you are comparing prices at points where the value of
the dollar is different. ie. $1 in 1900 does not = equal $1 in 2004. You
could buy MUCH MUCH MORE FOR $1 in 1900.

Again the problem is you have no absolutely no clue to the concepts
behind this or even what the terms mean.

Igor

unread,
Oct 31, 2004, 2:38:40 PM10/31/04
to
Robert Vienneau wrote:

> In article <krXgd.29451$EI6....@fe2.texas.rr.com>, Igor
> <jjwea...@houston.rr.com> wrote:
>
> "The 1929 Stock Market: Irving Fisher Was Right," Ellen R. McGrattan
> and Edward C.Prescott
> <http://research.mpls.frb.fed.us/research/sr/sr294.pdf>
>
>>Returns are equal to the interest rate * K. You use
>>the interest rate because in competitive markets returns on assets will
>>be equal therefore no one has incentive to switch assets unless
>>conditions change.
>>
>>Now how do you measure capital at market value? The Price of one unit is
>>EQUAL TO ITS MARGINAL COST if the market is competitive.
>
>
> That statement is close to the claim that the interest rate
> is equal to the marginal product of capital. So I think it
> mistaken, too.
>

It is a result for General equilibrium analysis in perfect capital
markets. It is not mistaken. What you mean is that the assumption may
not hold in real markets because they are not perfect. I agree but these
are simplifying assumptions so that you can actually measure something.

>
>>Change in K = Investment(deflated) - (deprecation
>>rate)K(t-1) . That is how you build a stock using the perpetual
>>inventory method.
>
>
> Technically, the above equation is incorrect. The theoretically
> exact way to treat fixed capital is as a kind of joint production.
> Accounting has a lot of convention built into it.
>

We are not talking about theoritically exact ways. We are talking about
what is done in econometric practice. Often therotically exact ways
leave with methods that are untractable or you can not find the complete
data on.

>
>>WHAT IS THE DEFINITION OF UNDERVALUED. The definition says that
>>expectations of future profits ARE BELOW CURRENT PROFITS.
>
>
> Nope. And that is not the definition McGrattan and Prescott use.
>

Completely wrong. All you have to do is look at the abstract to find
this out. I haven't found the link to the full paper again but this is
from the abstract.

" In this paper, we use growth theory to estimate the fundamental value
of corporate equity and compare it to actual stock valuations. "

The measure price compared to actual returns. When actual fundemental
value is high and actual stock market valuations are low then it is
undervalued. In other words when expectations of value are lower than
actual value the stock is undervalued. THIS DOES NOT SAY PRICE WILL
RISE. IT SAYS PEOPLE ARE EXPECTING FUTURE PROFITS TO BE LOWER THAN
CURRENT PROFITS. The result gives credence to the Rational expectations
hypothesis. It says that people forsaw the crash before it happened and
adjust expectations BEFORE the event occured.

You think the finding is absurd because you believe that undervalued
says that the price will be higher in 3 months. No one says this.
Undervalued says that expected profits are much lower than actual
profits. This means that the market believes profits will be lower or
they were surprised the firm did much better. NO stock broker will say
an undervalued stock is a guaranteed winner. They will buy undervalued
stocks if they believe the market expectations are off. If they believe
the market expectations are correct they will not buy an undervalued stock.

Robert Vienneau

unread,
Oct 31, 2004, 4:04:59 PM10/31/04
to
In article <4bbhd.25296$lM1....@fe2.texas.rr.com>, Igor
<jjwea...@houston.rr.com> wrote:

> Robert Vienneau wrote:

> > "The 1929 Stock Market: Irving Fisher Was Right," Ellen R. McGrattan
> > and Edward C.Prescott
> > <http://research.mpls.frb.fed.us/research/sr/sr294.pdf>

> ...I haven't found the link to the full paper again...

Igor

unread,
Oct 31, 2004, 7:33:56 PM10/31/04
to
Robert Vienneau wrote:

> In article <4bbhd.25296$lM1....@fe2.texas.rr.com>, Igor
> <jjwea...@houston.rr.com> wrote:
>
>
>>Robert Vienneau wrote:
>
>
>>> "The 1929 Stock Market: Irving Fisher Was Right," Ellen R. McGrattan
>>> and Edward C.Prescott
>>> <http://research.mpls.frb.fed.us/research/sr/sr294.pdf>
>
>
>>...I haven't found the link to the full paper again...
>
>

Nice job of cutting the quotation from the abstract that showed that you
had no idea what the definition of undervalued is. This is one of your
problems Rob you can read mathematics but do not understand terms and
concepts.

sinister

unread,
Nov 1, 2004, 5:59:10 AM11/1/04
to

"Igor" <jjwea...@houston.rr.com> wrote in message
news:Uvfhd.27772$lM1....@fe2.texas.rr.com...

Pot...kettle...black...


0 new messages