Unproductive Activities and the Rate of Surplus Value at the Industry Level in Korea, 1995–2015

ABSTRACT

In order to explain how the Korean economy underwent structural change through the two crises of 1997– 1998 and 2008 within the context of globalisation, this article focuses on class analysis and inter-sectoral value transfer by estimating the sectoral rates of exploitation along with the sectoral monetary expressions of labour time. Our data indicate the possibility that the expansion of unproductive activities, accompanied by intensification of exploitation within the unproductive sectors, might not have overtaken capital accumulation in Korea during 1995–2015. It can also be concluded that the condition for manufacturing’s capital accumulation steadily improved since the 1997–1998 crisis but started to deteriorate after 2011. Our value-theoretic analysis provides a foundation for understanding the context of the regime change, which may plausibly characterise the Korean economy in the last couple of decades.

KEY WORDS:

• Sectoral rates of surplus value
• unproductive labour
• monetary expression of labour time
• Korean economy

Acknowledgments

The authors are grateful to Sangyong Joo and Su Min Jeon for generously providing the data on the labour income share of Korea. This study would have been impossible without their support. We also thank Jonathan F. Cogliano, Jeong Joo Kim, Xiao Jiang, Fred Moseley, Naoki Yoshihara, Jayoung Yoon, two anonymous referees, and participants at the 2018 Winter Conference of the Korean Association for Political Economy in Seoul and the 2018 Annual Meeting of the Eastern Economic Association in Boston for useful comments and suggestions. The usual caveat applies.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1. This is one of the advantages of the NI over the other interpretations of Marxian value theory and the main reason why we adopt it as a framework for the empirical analysis in this article. We recognise that there is a critique of the NI that it explains values from prices rather than the other way around and thus that “the whole relation between surplus value and profit is turned on its head” (Shaikh and Tonak 1994, 179). However, this view is misdirected since in the NI the labour value categories are not determined by, but recovered from, the price categories.

2. Jeong and Jeong (2020), which appears in this special issue, differs from ours in two ways. First, their methodology in distinguishing between productive and unproductive labour is more sophisticated, relying upon the input–output table. Second, in their article, an estimation of the rate of profit and the rate of surplus value is based on a more traditional approach; for instance, the value of labour power is still interpreted in connection to the value of means of consumption. The result is that their empirical results are somewhat different from ours.

3. Sources of the data used in the article are explained in the appendix and details of estimations are found in the note of each figure.

4. Two points should be observed. First, we computed the capital share of national income by following Joo and Jeon’s (2014) method, that is, to decompose the operating surplus of self-employed, unincorporated sector – which is officially categorised as capital income – into capital income and labour income according to the profit–wage ratio of the corporate sector. Second, the measure of the profit rate in Figure 2 assumes away the distinction between productive and unproductive labour. When the distinction is taken into consideration, however, the wage of unproductive industries will need to be added to the numerator of the profit rate as part of total surplus value. We will start to incorporate the productive–unproductive labour distinction into our analysis from the next section.

5. In Hong (2013), the manufacturing industry’s output–capital ratio, which corresponds to our capital productivity, was in an upward trend from the 1997–1998 crisis to 2009. This difference can possibly be explained by the fact that Hong measures the ratio in real terms while we use nominal terms and that Hong uses value added while we use net national income.

6. The productive industries include agriculture, forestry and fishery, mining, manufacturing, utilities, construction, accommodation and food service, transportation and warehousing, information services, education service, arts and entertainment.

7. The underlying assumption is that the labour share in income is identical between the incorporated and unincorporated sectors.

8. See Mohun (2006) for a summary of the theoretical framework of the NI in the context of an empirical analysis.

9. While Ⓑ could be a useful approach in separating out the supervisory labour from the productive labour as shown in Sung (2007) and Jeong (2015), it is difficult to apply without some restrictive assumptions since the data published by the Ministry of Employment and Labour (2003–2017) is based on a sample inspection instead of a full and complete one.

10. In relation to the discussion of Piketty above, we can examine the behaviour of the value composition of capital (VCC) using K/W_P as a proxy. From the estimation, the result of which is omitted in this article, we have observed that for both the whole economy and manufacturing the VCC increased during the two crisis episodes and then stagnated after 2010. In particular, the rising trend has been weaker in manufacturing than in the whole economy.

11. Although value and surplus value are not produced in the unproductive industries, exploitation may take place therein, as Foley (1986, 122) states that exploitation “through the wage labor relation occurs when a worker expends more labor hours than he or she receives an equivalent for in wages.”

12. It should be noted, however, that this relation holds only under certain conditions.

13. Paitaridis and Tsoulfidis (2012) document the rise in the share of the unproductive sectors in terms of employment and wages whereas Basu and Foley (2013) focus on employment and value added. It should be noted, however, that Basu and Foley adopted the distinction between service and non-service sectors which is less controversial than the productive versus unproductive industry distinction.

14. For more discussions on the determinants of the sectoral MELTs, see Rieu, Lee, and Ahn (2014).

15. To our knowledge, Vaona (2011) is the only paper that precedes ours in using panel unit root tests to examine the hypothesis of equalisation of either the rate of profit or the rate of exploitation. As Vaona studies three European countries with a larger data set than ours, the Choi (2001) test, which is another available panel unit root test, is used in addition to the two tests we used; our data set is too small to employ the Choi (2001) test. The results reported in Vaona more strongly support the hypothesis that the sectoral rates of surplus value are not equalised.

16. The assumption of an equalised value-creating capacity of each industry’s aggregate labour is necessary in measuring the sectoral rates of exploitation from the real-world data. Duménil, Foley, and Lévy (2009) state that in “order to recover a measure of this value productivity from real-world price and wage data by sector, some additional assumption about relative rates of exploitation (which Marx often explicitly assumes to be equal) is required.” Hahn and Rieu (2017) present simulations that examine the consequence of the assumption that the value-creating capacity of the aggregate labour in each industry is not the same.