Price-Value Deviations and the Labour Theory of Value: Evidence from 42 Countries, 2000–2017

ABSTRACT

The relationship between prices and labour values has been the source of fruitful controversy since the earliest Classical Political Economists. The alleged refutation of the labour theory of value was an integral part of the marginalist attack against Classical and Marxist analysis. However, statistical analysis of price-value relationships made possible by the data available since the later 20th century suggest considerable empirical strength of the labour theory of value. We trace the intellectual history of the price-value relationship and its inseparable link to capitalist competition through Smith, Ricardo, Marx and Sraffa. Following Shaikh and Ochoa, we present an empirical model of testing their hypotheses that (1) labour values regulate prices of production and (2) serve as gravitational centres for market prices. The analysis of a large dataset of 42 countries and 15 years reveal only small and stable deviations and thus lend support to the Classical Political Economic analysis. With a sample of over 36,000 price vectors, we provide the most comprehensive empirical application of its class and generalize the results that have been established in the relevant literature.

KEYWORDS:

• Price
• value
• input–output
• classical political economics
• labour theory of value

JEL CLASSIFICATION:

• B12
• B51
• C67
• O57
• P16

Acknowledgements

We thank the anonymous reviewers for their careful reading of our paper and their many insightful comments. We also thank Anwar Shaikh, Persefoni Tsaliki and Lefteris Tsouldfis for their instruction and advice as well as William Paul Cockshott for very helpful comments on an earlier version of this paper.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 We use this term to mainly refer to Smith, Ricardo, and Marx.

2 Where use tables U and make tables V are available rather than flow tables Z, we construct the diagonal commodity output matrix $\hat{Z}$ = diag(z₁, z₂, … , z_n) and calculate the coefficients matrix $A=V{\hat{Z}}^{-1}U{\hat{X}}^{-1}$ (BEA 2017).

3 We include additional IOTs only if they recorded 100 or more industries, as compared to the harmonized 55 industries structure in the WIOD (Timmer et al. 2015).

4 The criteria for omissions are listed in Online Appendix A, which is available as attachment to the online article. Similarly, Online Appendix B (and Tables B1 and B2) are only available in the online version of the paper.

5 It should be noted that the cross-sectional plots summarize the relationships for all countries and years, which have quite different economic dynamics regulating the individual outcomes. The plots are meant to illustrate the intuition rather than provide a solid ground for interpretation of the results.

6 In the WIOD sample, we approximate energy industries’ output as the output sum of the following industries: ‘B' for mining and quarrying, ‘C19' for manufacture of coke and refined petroleum products, and ‘D35' for electricity, gas, steam and air conditioning supply.

7 Both results are significantly different from zero with Pearson product-moment tests rejecting insignificance at a p-value below 0.01.

8 Both results are significantly different from zero at a 0.01 p-value according to Pearson product-moment test.