Good heavens, what charming people we find on newsgroups!  So, to
recap the history of this thread, I am requested, by name, to review a
piece written by this poster and offer suggestions.  I do so, but then
my suggestions for making the piece more convincing are met with the
response that my past efforts to help (by attempting to explain the
rudiments of empirical testing of theory) are characterized as
"misrepresented."  Ah well, charming people.

Returning to the point at hand, the originator of this thread wrote a
piece that promotes the contributions of Sraffa to modern economics. 
Further, the reader asks whether Sraffa's work is important beyond
capital aggregation issues.  I'll admit that I'm open minded and that
after forty years am not yet willing to give up hope that some
empirical value to Sraffa's work can be found.  However, my Baysian
prior expectation for seeing such a contribution at this point is not
strong.

My main suggestion for where to go with this poster's work was to make
the usual remark that it is important to stress the way the paper
improves upon the current literature in explaining the observed world.
 This need not call for formal empirical testing, although that is
generally a strong plus.  I ask in what way would the central aspects
of the paper change or improve upon some major published work in
economics in the last ten years.  What is the response?

> Some time ago I pointed out that, due to price Wicksell effects, the first numbered
> equation in Romer (1990) is mistaken. I'm not the only one to have noticed:
> 
>   "Vintage models have attracted the attention of neoclassical
>   economists for another reason. In these models, using severely
>   restrictive assumptions, it is possible to assign a value to
>   a 'heterogeneous' capital stock that is independent from
>   distribution (Fisher 1965, and the contribution by Petri in
>   this volume). Note that in these models, 'capital' is
>   heterogeneous in the limited sense of different efficiency of
>   vintages of a unique type of capital good (i.e., corn-seeds
>   of different efficiency). Recently, Romer (1990) adopts this
>   approach in an EG model without any discussion of its severe
>   limitations."
>    -- Sergio Cesaratto, "New and Old Neoclassical Growth Theory:
>       A Critical Assessment", in _Value, Distribution, and Capital:
>       Essays in Honour of Pierangelo Garegnani_. Routledge, 1999.

What is the poster claiming?  That Romer is wrong and that the editors
at the JPE in 1990 made a mistake?  If so, submit a correction and see
how the editors respond.  In support of his position that Romer's work
is in error, Mr. Vienneau quotes Casaratto here, but there is no such
support in the quote, rather just a complaint about a lack of
"discussion of its severe limitations."  (Further, the Cesaratto quote
does not come from an article that has undergone formal referee
review.)

Yes, models have assumptions that come with limitations; such is the
way research works.  A contribution is not to point out this well
known truth but rather to build alternative models and show how their
results contrast.  Would Romer's results be overturned with a
different specification?  Work it out and see, then send it in to the
editors for review and get their feedback, and maybe even get the work
published and thus into the intellectual discourse of the field.  This
is how knowledge is formed and human understanding advanced, not by
repeatedly spamming newsgroups.

>>>>>    "Reswitching poses a serious problem for capital aggregation
>>>>>    and measurement, and that's about it."
>>>>>     -- Poor Chris Auld, 9 April 1999
> 
>>>>> Perhaps Mr. Witte can tell us if he agrees that Chris Auld's 
>>>>> statement
>>>>> is incorrect.
>  
>>> Notice how Mr. Witte refuses to answer the question.
>
>> In the end, the only answer is the question, "Does this matter?"    
> 
> Mr. Witte seems not to know what an answer is. For example, an
> answer would start with "Yes" or "No".
> 
> Perhaps Mr. Witte can tell us if he agrees that Chris Auld's
> statement is incorrect.
>
>> I'll repeat the question I asked, "Most readers would likely want to 
>> know how would applying the approaches or results in this paper
>> produce differences from the
>> findings in some well cited papers in the literature."
> 
> I would hope many readers might wonder also why textbooks are full of
> outdated, exploded, inconsistent, and contradictory theories.
> 
>   M. I. Naples and N. Aslanbeigui, "What DOES Determine the Profit
>   Rate? The Neoclassical Theories Presented in Introductory
>   Textbooks", Cambridge Journal of Economics, V. 20, pp. 53-71, 1996.
>
>> Perhaps the
>> next time this paper surfaces, the value of Sraffa's work for its
>> ability to improve upon standard methods in modeling the observed
>> world will be explained.
>> 
>>    Without an answer to this basic line of questions, this paper will
>> continue to be greeted with indifference.  I believe that the author
>> of the paper would say that the results described have all been around
>> since the work of Sraffa in the 1960s.
> 
> No. My paper contains at least one original speculation, namely in
> the last paragraph of 2.4.
> 
>   "What counts though are the big insights a genius brings to an
>   important problem of science. By this test Piero Sraffa is not
>   fully recognized in the mainstream literature."
>     -- Paul A. Samuelson, "The Special Thing I Learned from
>        Sraffa", in _Value, Distribution, and Capital:
>        Essays in Honour of Pierangelo Garegnani_. Routledge, 1999.
>
>> With forty years to try to 
>> find some relevance to describing the world, the importance of this
>> work seems slight.
> 
> Personally, I'm of the opinion that one's ability to understand the
> world is improved by throwing out internally inconsistent theories.
> 
> Mr. Witte's misrepresentation has been clarified many times here. The
> previous post cited Leontief's work. That PDF paper, which apparently Mr.
> Witte did not read, cites Albin (1975), Rosser and Prince (1985), and
> Ozanne (1996). (A number of years ago, Mr. Witte has misrepresented
> the first two of these three references.) Some time ago I pointed
> out that, due to price Wicksell effects, the first numbered
> equation in Romer (1990) is mistaken. I'm not the only one to
> have noticed:
> 
>   "Vintage models have attracted the attention of neoclassical
>   economists for another reason. In these models, using severely
>   restrictive assumptions, it is possible to assign a value to
>   a 'heterogeneous' capital stock that is independent from
>   distribution (Fisher 1965, and the contribution by Petri in
>   this volume). Note that in these models, 'capital' is
>   heterogeneous in the limited sense of different efficiency of
>   vintages of a unique type of capital good (i.e., corn-seeds
>   of different efficiency). Recently, Romer (1990) adopts this
>   approach in an EG model without any discussion of its severe
>   limitations."
>    -- Sergio Cesaratto, "New and Old Neoclassical Growth Theory:
>       A Critical Assessment", in _Value, Distribution, and Capital:
>       Essays in Honour of Pierangelo Garegnani_. Routledge, 1999.
> 
> Also, consider
> 
>   "David Laidler suggests that Sraffa was ahead of his time in
>   pointing out that neoclassical models based on an aggregate
>   production function lack proper microfoundations, because this
>   is a far more telling criticism of modern real business cycle
>   theory, whose exponents make strong claims about such matters,
>   than it was of 1950s vintage growth models."
>     -- Avi J. Cohen and G. C. Harcourt, JEP, Winter 2003.
> 
> (Cohen is writing a book subtitled "From Bohm-Bawerk to Bliss"
> and Cohen, Harcourt, and Bliss are putting together a collection
> of the most important articles on capital theory.)
> 
> And there are others bringing out certain implications.
> 
>>> Note follow-ups.
>> [I deleted the sci.math cross-post out of common decency.] 
> 
> What might substantial comments on the following look like?
> 
>  http://csf.colorado.edu/pkt/pktauthors/Vienneau.Robert/Sraffa3.pdf
> 
> One might state whether the numerical examples there are correct
> or not. And if one claims that they are incorrect, one might
> indicate why and what a correct solution looks like.
> 
> One might indicate whether one thinks technology can be such
> that the implications indicated in this paper could occur. If
> one thinks technology could not be like that, one might indicate
> sufficient special case assumptions on technology one is inclined
> to impose such that the principle of substitution follows. Whether
> one can or cannot outline such special case assumptions, one
> might provide some argument for why technology could not be
> like that, if one thought it cannot.
> 
> On the other hand, if one agrees with the literature
> that the neoclassical theory of value and distribution is
> faulty, one might indicate that one would like economists
> to abandon it in their teaching and practice.