Drewk, thanks for the references and the useful critique of C&C, Ochoa _et al_. I'll have to look at your paper. As mentioned, for me the law of value involves not only some correspondence between relative values and relative prices -- expecially on the aggregate level -- but also deviations (so that people inside the system don't see how capitalism is actually working). So if it turned out that there was perfect correlation between values and prices, it would be a strike against Marx's LOV.
As for your macro-correlation between "multifactor productivity" and the CPI, I have some questions:
1. for Marx's LOV, shouldn't it be labor productivity that affects prices, not the multifactor productivity?
(In my opinion, MFP is a bogus concept. It's based on adding apples (labor-power hired) and oranges (means of production) using weights that assume that each factor is paid its marginal product.)
2. isn't it commonplace for economists to say that productivity increases lead to price falls, cet. par.?
3. how did you deal with the fact that most time series tend to be highly correlated with each other even when causation is absent?
------------------------
Jim Devine [EMAIL PROTECTED] & http://bellarmine.lmu.edu/~jdevine
> -----Original Message-----
> From: Drewk [mailto:[EMAIL PROTECTED]]
> Sent: Friday, June 13, 2003 1:24 PM
> To: [EMAIL PROTECTED]
> Subject: Re: [PEN-L] Falsifiability and the law of value
>
>
> Chris Burford wrote:
>
> "the marxian law of value, probably cannot be
> falsified, but may be "true".
>
> "The only empirical study I know about it is by Paul Cockshott and
> Allin Cottrell showing that it could fit with macroeconomic data,
> I
> think for Scotland if I recall correctly."
>
> There have been numerous studies like this (not only by Cockshott
> and Cottrell, but also by Ochoa, Shaikh, Petrovic, Guerrero,
> Maniatis & Tsoulfidis, and others). But their results are not
> meaningful. The strong cross-sectional correlations they find are
> SPURIOUS CORRELATIONS, correlations that stem only from their
> failure to control for variations in industry size. (It is not
> surprising, nor meaningful, that the total value of output and the
> total price of output are both large in large industries and small
> in small industries.)
>
> Once one DOES control for variations in industry size, the
> correlations between values and market prices simply disappear. I
> found this, using US data, in a study published in the Cambridge
> Journal of Economics last year. More recently, Rubén Osuna has
> found the same thing, using Spanish data, in a study that's part
> of his superb dissertation on the temporal single-system
> interpretation of Marx's value theory.
>
> The references are:
>
> Kliman, Andrew J. 2002. The law of value and laws of statistics:
> sectoral values and prices in the US economy, 1977–97, _Cambridge
> Journal of Economics_, vol. 26, 299-311
>
> Osuna Guerrero, Rubén. 2003. _Un Modelo Secuencial para el Cálculo
> de Precios. El Caso Español 1986-1994_. Ph.D. dissertation,
> Departamento de Análisis Económico I, Facultad de Ciencias
> Económicas y Empresariales. Universidad Nacional de Educación a
> Distancia (UNED), Spain.
>
>
> However, properly understood, the law of value does generate a
> falsifiable hypothesis that has not been falsified: productivity
> increases tend to reduce commodities' prices. My preliminary
> computation for the US during the past 1/2 century indicates that
> a 1 percentage point increase in multifactor productivity leads,
> ceteris paribus, to almost exactly a 1 percentage point decline in
> the CPI.
>
> Andrew Kliman
>
