---
schema_version: "library-reader-v1"
source_type: "original_book"
title: "Digital Capitalism"
author: null
edition: null
publisher: null
year: null
isbn: "2021022153"
library_document_id: "50b7f922-b940-4a05-8eb0-c4dc0d8ec9ed"
source_sha256: "746bfc1e2253e11737253daad663d7176077b6b54343c2c35f6f132990452fbc"
pdf_physical_page_count: 343
pdf_physical_page_index_base: 0
range_start: 290
range_count: 10
html_url: "https://workshop.closeclaw.top/api/ai/read/50b7f922-b940-4a05-8eb0-c4dc0d8ec9ed?start=290&limit=10"
markdown_url: "https://workshop.closeclaw.top/api/ai/read/50b7f922-b940-4a05-8eb0-c4dc0d8ec9ed.md?start=290&limit=10"
previous_url: "https://workshop.closeclaw.top/api/ai/read/50b7f922-b940-4a05-8eb0-c4dc0d8ec9ed.md?start=280&limit=10"
next_url: "https://workshop.closeclaw.top/api/ai/read/50b7f922-b940-4a05-8eb0-c4dc0d8ec9ed.md?start=300&limit=10"
---

# Digital Capitalism

> 页码规则：PDF_PHYSICAL_PAGE_INDEX 是从 0 开始的权威文件物理页索引；PRINTED_PAGE 是页面中识别出的印刷页码，可能为空，二者不得混用。

## PDF_PHYSICAL_PAGE_INDEX=290 | PRINTED_PAGE=276

There are certainly informational occupations that do not result in the production of an information commodity, but rather in different commodity types. They are part of a collective workforce outside the information sector. The distinction between the occupational and the industry definition of industries certainly is feasible and can be combined. The question however is which one fits best for a Marxist understanding of the economy.

To avoid an elitist definition of creativity, one can also resort to Marx (1867, 284), who argued that at the “end of every labour process, a result emerges which had already been conceived by the worker at the beginning, hence already existed ideally”. All human labour requires mental planning and anticipation of the result, albeit with different qualities. All occupations have an information dimensions and are creative in the general understanding of the word that Marx hints at, namely that they create usevalues that satisfy needs.

This brief non-exhaustive discussion shows why this chapter uses a framework, in which the information industry is defined as the realm of the production, circulation, and consumption of information products, both information goods and services. Many information products take on the form of commodities. There is however a growing sector of the information economy that transcends the commodity form and in which peer produced information takes on the form of an informational commons. Given this definition of the information economy, the share of the information industry in total national value added was in 2010 15.6% in Finland, 14.1% in France, 12.4% in Germany, 12.0% in Italy, and 10.6% in Norway (data source: OECD STAN).<sup>3</sup> The total employment share of this sector was in the same year 16.3% in Finland, 13.8% in France, 13.2% in Germany, 11.7% in Italy, and 14.2% in Norway (ibid.). One may now say these are significant, but not overwhelming shares. Let’s, however, compare these data to the manufacturing sector:<sup>4</sup> it accounted in 2010 for 14.8% of value added in Finland, 9.7% in France, 14.8% in Finland, 20.0% in Germany, 17.1% in Italy, and 7.3% in Norway. Manufacturing comprised in the same year 13.4% of total employment in Finland, 10.2% in France, 15.9% in Germany, 15.1% in Italy, and 8.8% in Norway. Six out of ten country-specific variables in these example calculations have higher values in the information than in the manufacturing sector.

276

Themes

3 Comparable results can be obtained for other countries. OECD STAN at the moment of writing (January 2014) only provides data for 15 countries because ISIC Revision 5 is still a relatively new metric introduced in 2008 that has not been applied to older data and is not used for statistical purposes in all countries. No data are available for the United Kingdom. The data for the USA are not sufficiently disaggregated for use. See the appendix for a methodological discussion.

4 I calculated value added of manufacturing based on ISCI Revision 5 aggregated category D10T33 minus D18 and D26 that according I consider to be part of the information economy.

## PDF_PHYSICAL_PAGE_INDEX=291 | PRINTED_PAGE=277

In this chapter, section 12.2 studies an example case – the German information economy – with the help of the labour theory of value in order to formulate some assumptions. Section 12.3 tests based on the same case, the assumption that the labour theory of value cannot be applied to the information sector.

12.2 The Information Industry and Marx’s Labour Theory of Value

Some representatives of Autonomist Marxist theory argue for example that the law of value does not apply to the information economy because it is networked and based on co-operation. Antonio Negri (1991, 172) writes for example that Marx in the Grundrisse argues for the death of the law of value. Paolo Virno (2004, 100) says that the development of capitalism refutes the law of value. Hardt and Negri (2004, 145) claim that the law of value in contemporary capitalism “makes no sense”. Vercellone (2010, 90) speaks of capitalism’s “crisis of measurement”. But Marx (1857 [1858], 705), in the specific passage that these authors refer to, does not talk about the law of value in capitalism, but rather about the death of the law of value in a communist society (Rosdolsky 1977, 428). The law of value is a foundation of all forms of capitalism (Postone 2008, 126).

A comparable argument has been made by Nicholas Garnahm, who assumes that the labour theory of value is invalid in the information industries. Nicholas Garnham is one of these scholars. Garnham (1998, 103) says that the labour theory of value works if “labour time is largely a matter of energy and expended” and productivity rises due to saving of human energy expenditure based on technology. He questions whether this works in a “non-entropic economy of bits”. The labour theory of value for Garnham works well “in an economy based […] on the measurable increase in the productivity of human work-time and especially on the saving of human energy inputs through the substitution of natural energy sources” (Garnham 2000, 146). “In an economy dominated by information production any stable relation between labour and value breaks down” (Garnham 2000, 146). Garnham (2011, 47–48) writes that the economic concepts of productivity and efficiency cannot be applied to culture because of an “uncertain relationship between labour inputs and outputs” (Garnham 2011, 47). Information labour could not easily be automated and mechanised.

Chapter Twelve | The Information Economy

277

The Information Industry and Marx’s Labour Theory of Value

## PDF_PHYSICAL_PAGE_INDEX=292 | PRINTED_PAGE=278

In a newer version of his argument, Garnham argues that the labour theory of value

served a useful analytical function and within the relatively simple material economies of the day bore some relevant relationship to relative prices and the division of the surplus and thus to class relations. […] In an economy based upon a complex division of labour, on a growing ratio of dead labour to living labour and on the application of knowledge, it made, and makes, less and less sense. (Garnham 2016, 295–296; for a response see Fuchs 2016a)

Garnham argues that Piero Sraffa (1960) in his book Production of Commodities by Means of Commodities has shown that “the inequalities of capitalism can be generated within the process of commodity production and exchange without recourse to an exploitation theory of labour” (Garnham 2016, 297; for a critique see Fuchs 2016a).

One should bear in mind that Cantabrigian economics (Pierro Sraffa, Joan Robins, John Eatwell, and others) has resulted in Marxist critique. In this approach, “people as human beings – and, more importantly, as historical social classes – are given no role in the process of production” (Roosevelt 1975, 7). The Cantabrigians separate “relations between things from relations between people” (Roosevelt 1975, 19), which results in a commodity-fetishistic model of capitalism, in which “things produce things” (p. 19). “However ingenious the Cantabrigians are in analyzing price phenomena, they never connect such phenomena with social relations in the way that Marx did in Capital” (p. 20). Sraffa and Robinson’s approaches constitute a “[p]rice theory without value theory” (p. 19). Robinson (1966, 22) revises Marx theory and argues that “no point of substance in Marx’s argument depends upon the labour theory of value”. For Marx, distribution and production are intertwined social processes, for the Cantabrigians they are separate, production being technical in nature and distribution social. For Sraffa (1960, 9), surplus is not surplus-value, but a physical “surplus production” of commodities, “a surplus of some commodities” (p. 26). In contrast, for Marx surplus is surplus-value that stems from capital’s exploitation of labour in class relations.

For Marx, there is a social production process, in which humans enter social relations of production and these social relations use means of production for creating new goods that are distributed in relations of distribution (Fuchs 2016b). Production and distribution are interrelated social processes grounded in human activity. For Marx, human creativity and activity sustain any economy. Capitalism is a mode of production that is based on the general production of commodities that reflects socially necessary labour-time, that is an average quanta of human labour that is exploited by capital. Production is a social process, in which humans work together in order to produce a good or service. In capitalism, this involves a division of labour. In addition, in complex economies there are multiple organisational units of production creating similar types of goods and services. In capitalism, this involves a social process of competition for productivity levels, price advantages, markets, customers, and profits. Marx’s labour theory of value is a theory of time in capitalism (Fuchs 2016b; Fuchs 2015, chapter 4). It distinguishes between necessary labour-time and surplus labour-time. The first constitutes the value of labour-power and the value of the necessary means of production. The second is surplus-value that is unremunerated labour-time. Parts of the working day are not paid, which results in monetary profit and a surplus product owned by the capitalist class. Capitalism means a struggle focused on paid and unpaid labour-time. The working class has the interest that all of its time is remunerated, whereas the capitalist class has the interest that no labour-time is paid at all. The class struggles oscillates depending on the power of the involved forces somewhere between the poles of maximum wage and maximum profit that corresponds to the two extremes of the payment of all labour-time on the one end and the payment of no labour-time at all on the other end.

278

Themes

## PDF_PHYSICAL_PAGE_INDEX=293 | PRINTED_PAGE=279

To argue that increasing importance of the information economy invalidates Marx’s labour theory of value means that time and labour-time do no longer play any role in the economy. Consequently, the class conflict must have come to an end and the production of information must have become autonomous from humans. But none of the two can be observed today. There are several reasons why the labour theory of values continues to apply in information capitalism and digital capitalism (see Fuchs 2014a, 2015, 2016b, 2017 for details of these arguments):

• First, commercial software and other information goods are not just produced once and then copied, but there are often new versions, constant updates, and forms of support labour.

• Second, one has to see that large parts of the Internet’s political economy are based on targeted advertising. Google and Facebook are not communications corporations. They are the world’s largest advertising companies. Advertising is not just based on the labour-time of marketing professionals, but also on the attention time of audiences and on commercial Internet usage time that is (unpaid) labour time.

• Third, there is an international division of digital labour, in which various forms of labour are organised. It ranges from the exploitation of enslaved miners in the

Chapter Twelve | The Information Economy

279

The Information Industry and Marx’s Labour Theory of Value

## PDF_PHYSICAL_PAGE_INDEX=294 | PRINTED_PAGE=280

Congo, Tayloristic information and communication technology (ICT) assemblers at Foxconn in China, or software engineers in India or the Silicon Valley to various forms of unpaid online labour (Fuchs 2014a, 2014b, 2015). The production of information technology is highly exploitative and time-consuming.

• Fourth, there are various forms of irregular, unpaid, precarious, outsourced, crowdsourced, and click-worked digital labours. Examples include the usage of Facebook, Google, YouTube, Weibo, LinkedIn, Pinterest, and Instagram; online customer reviews on Amazon or Yelp; work via freelancer platforms such as Upwork, PeoplePerHour, Amazon Mechanical Turk, and ClickWorker; the participation in customer surveys, installing software updates, deleting spam, unsubscribing from spam lists, the time spent on online daring platforms such as match.com or Tinder, answering professional e-mails via the mobile or tablet out of regular working hours, working on the train, tube, or in cafés; online travel booking, etc.

• Consumer and prosumer labour is shadow work because it does not in an obvious way feel like work, but creates value for corporations. It takes time. And it takes time away that could be used outside the commodity culture. It substitutes paid labour by precarious and unpaid labour and by reducing corporations’ wage-sum helps increasing their profits. Consumers and users have become part of the working class.

The digital law of value has created new forms of exploitation as well as contradictions that allow the creation of new spheres of non-commercial, alternative, cooperative production and a solidarity, commons-based, and peer production economy outside the realm of capitalism that undermine the law of value. The political aim of destroying the law of value is not an automatism that flows from information and information technology. It can rather only be achieved in conscious political struggles for the decommodification of information, the economy, and the world. It requires the dialectical political unity of the social movement “crowd” and the party (Dean 2016).

In order to show that the labour theory of value applies to the information economy, I employ in this chapter the Temporal Single-System Interpretation (TSSI) of Marxian value theory for the analysis of the German information economy. One could certainly discuss if the TSSI is the right methodology or if another one should be used, which is a more scholastic discussion that I leave to others. One could simply in an equal manner apply other interpretations and check the feasibility of the results. The space of this chapter allows however to just apply one method of analysis and the TSSI seems to be an interesting approach.

280

Themes

## PDF_PHYSICAL_PAGE_INDEX=295 | PRINTED_PAGE=281

The TSSI is a version of Marxian value theory. It assumes a dynamic character of accumulation. It has been created by a group of economists around Andrew Kliman and Alan Freeman. Some of the basic insights of this interpretation are:

• Marx’s theory is not erroneous or inconsistent. It does not require a correction or revision with external assumptions, but is in itself consistent.

• Claims that Marx’s theory is inconsistent or erroneous often serve the purpose to suggested that it should not be used today and that its political implications are wrong. The consequential assumption of such approaches is that capitalism and class should not be abolished.

• Critics of Marx’s theory often either argue that his theory of the tendency of the rate of profit to fall, the labour theory of value, or his solution to the transformation problem (the transformation of labour values into prices) are false and therefore invalidate his theory.

• The TSSI shows that all three aspects are internally consistent on the grounds of Marx’s theory.

• The TSSI assumes the existence of a single-system of labour values and prices. Values and prices are not seen as being independent, but standing in a relationship to each other. The monetary expression of labour-time (MELT) is a variable that can be calculated for an economic system and that allows the transformation of labour values (calculated in hours) into prices (calculated in monetary units) and vice-versa.

• The TSSI is temporal because it assumes a dynamic character of the capitalist economy. This means that the outputs of one time period (typically one full year) act as inputs into and influence the outputs of the next time period. This means that the total hours worked during one period of time create means of production that are typically used during the next period of time. One therefore does not have to assume that the values and prices of the economic inputs have to equal those of the outputs during a certain period of time.

Alan Freeman (2010, 592) argues that the price of a commodity “at time t1 must be equal to the sale price at time t0” (Freeman 2010, 592), that is there is a connection between output and input prices of one production period and the next. “In each period, new prices are established in circulation on the basis of the values arising thus from the immediately preceding phase of production” (Freeman 1998, 11). The TSSI also assumes, as Marx did, that commodities do not necessarily sell at their value, but can be sold above or below their value:

Chapter Twelve | The Information Economy

281

The Information Industry and Marx’s Labour Theory of Value

## PDF_PHYSICAL_PAGE_INDEX=296 | PRINTED_PAGE=282

the production price of a commodity is not at all identical with its value. […] It has been shown that the production price of a commodity may stand above or below its value and coincides with it only in exceptional cases. (Marx 1894, 892)

In the TSSI, the

value is quite distinct from the price, and the difference is a quantitative one. In exchange the values that have arisen in the manner we have just described will be exchanged against money and this money, in exactly the same way but with new proportions, will represent value, measurable in hours. Some will realize more hours than their value, and some will realise less. Conversely, since every amount of value may equally well be represented in money, we may assign a money magnitude to their value and we will find that their value, measured in money, is in general systematically different from their price. (Freeman 1998, 11)

Two main assumptions of the TSSI are that:

1) “valuation is temporal, so input and output process can differ”;

2) “values and prices, though quite distinct, are determined interdependently” (Kliman 2007, 2).

As indicated in the introduction, there is an ongoing theory debate in Marxist political economy about value-creation in the information economy. This discourse has theoretical importance, but has remained on a purely philosophical and theory level devoid of applications to empirical economic analysis, especially mathematical and statistical political economy. This chapter takes a different approach and is especially interested in empirical analysis of available statistical data. The TSSI is an approach that is grounded in the labour theory of value and mathematical and statistical analysis. The TSSI is the most thorough and rigorous mathematical formulation of the relationship between labour and time, for which it uses the category of the MELT, a category introduced first by Ramos (1998 [1999]). Furthermore the TSSI allows conceiving the information economy as value-creating, which stands in contrast to some other contemporary approaches such as certain versions of Autonomist Marxism.

The TSSI implies that for calculating the value of a commodity (measured in hours or units of money) and how it differs from the commodity price (also measured in hours or units of money), the wage (variable capital) needs to be transformed into “the value of the money used to purchase that worker’s labour power” (Freeman 2010, 595), which is done based on a ratio called the MELT; the MELT is obtained by calculation ratio of the whole economy’s total money to total labour-time (measured in hours) during a specific period of time, such as one year (Freeman 1998, 13). It indicates the degree of value measured in money that workers create on average in one hour of labour. The MELT’s unit of measurement is £/hour (or another currency). When Marx speaks about value, he mostly means units of labour-time, but sometimes also monetary values. The MELT is a way of connecting these two measures of economic value.

282

Themes

## PDF_PHYSICAL_PAGE_INDEX=297 | PRINTED_PAGE=283

The TSSI implies that value is produced in the production process and not when the commodity is sold (Kliman 2007, 37). The latter interpretation has been given recently by Michael Heinrich (1999, 2012) in what can be characterised not as a labour theory of value, but a money theory of value that implies that the worker is not exploited if the commodity s/he produces cannot be sold on the market (for a critique, see Kliman et al. 2013).

I will outline how the labour theory of value can be mathematically applied to information industries with the help of an example. I have chosen data for specific industries in Germany for the years 2001–2011 that were obtained from the OECD’s Database for Structural Analysis (STAN). I have chosen Germany because the required data are relatively complete for this country and the time period 2001–2011 because data are available for it based on the International Standard Industrial Classification of All Economic Activities (ISIC), Revision 4, that in contrast to ISIC Revision 3 provides more detailed data for the information industries. The appendix of this chapter discusses differences between the ISIC Revisions 3 and 4 in relation to the information economy and the classification codes used for defining the information sector in this chapter.

The empirical model used in this chapter follows the assumption widely spread in Marxist theory that finance, insurance, and real estate constitute unproductive realms of the capitalist economy. Marx argues that productive labour creates use-values that satisfy human needs and in capitalism take on the form of exchange-value and commodities. Real estate is an unproductive sector because rent can be generated without any labour. Once housing or a piece of land exists, it can be rented out. Financial capital’s production process takes on the form M–M’, money results in more money through financial instruments such as loans, mortgages, financial derivatives, shares, etc. There is no commodity separate from money. This becomes evident when you think of a loan: you receive a sum of money that a bank lends to you. Over a specific period of time, you pay back the whole sum plus an interest that you pay over time from your salary. Given that there is no commodity separate from money involved and productive labour is in Marxian theory tied to the production of a new commodity, the realm of finance is often considered as being unproductive. It is therefore often excluded from calculations of productive labour (see Shaikh and Tonak 1994). The theoretical assumption that financial capital is not the result of productive labour can also be observed practical in financial crises: they emerge from financial bubbles that build up through processes that constitute a difference between the monetary profits attained from the sale of commodities produced by labour and market values the same companies selling these commodities achieve on financial markets. Large divergences between financial capital values and profits constitute a source of financial crisis. Marx (1894) speaks of financial capital as fictitious capital because it takes on artificial values that are decoupled from labour-values. From “time to time […] flows of fictitious capital got out of hand to form speculative financial and commercial bubbles” (Harvey 2014, 239). The 2008 US housing market crisis that triggered a larger international economic crisis was the result of “fictitious capital run wild” (Harvey 2014, 33) because mortgages had been bundled together into derivatives that were traded as high-risk financial instruments.

Chapter Twelve | The Information Economy

283

The Information Industry and Marx’s Labour Theory of Value

## PDF_PHYSICAL_PAGE_INDEX=298 | PRINTED_PAGE=284

The method applied to this data was inspired by Freeman (1998) and Kliman (2007, 25–26) and included the following steps:

• The following data were obtained from OECD STAN and taken to represent as suggested by Shaikh and Tonak (1994) the following Marxian variables:

INTI = intermediate inputs: corresponds to constant capital c,

LABR = compensation of employees: corresponds to variable capital v,

NOS = net operating surplus: corresponds to profit p,

HRSE = hours worked by employees h

• All data were available in aggregated form at the level of industries as defined by ISIC Revision 4.

• Not all labour is productive. Therefore, there are industries that add no value to the economy in a Marxian framework. They are either unproductive or consume value. This question is contested in Marxist value theory and the empirical results one obtains depend on the theoretical choices one makes. I am in favour of a relatively broad concept of productive labour (Fuchs 2014a, 2014b, 2015), but assume that the financial and insurance industry as well as real estate do not create value. Data for these industries are however included in STAN (categories D64–D66 Financial and insurance activities, D68 Real estate activities). I therefore transformed the obtained data in such a way that I excluded data for these categories and recalculated the totals for all variables. Finance produces interest and real estate creates rent. Interest and rent are paid for by wages and profits, they consume and do not create surplus and value.

284

Themes

## PDF_PHYSICAL_PAGE_INDEX=299 | PRINTED_PAGE=285

• At a specific time t, the MELT can be calculated as (Freeman 1998):

MELT $\mathrm { ( t ) = ( c _ { p } ( t ) + v _ { p } ( t ) + p _ { p } ( t ) }$ [in prices]) $/ \left( \mathfrak { c } _ { \mathfrak { h } } ( \mathfrak { t } - 1 ) + \mathfrak { v } _ { \mathfrak { h } } ( \mathfrak { t } - 1 ) + \mathfrak { p } _ { \mathfrak { h } } ( \mathfrak { t } - 1 ) \right)$ [in years or hours])

where $\mathtt { C } _ { \mathtt { p } } ( \mathtt { t } )$ stands for total constant capital at the price level at point of time t (measured in monetary units);

$\mathsf { V } _ { \mathsf { p } } ( \mathsf { t } )$ is total variable capital at the price level at point of time t (measured in monetary units);

$\mathsf { p } _ { \mathsf { p } } ( \mathsf { t } )$ is total profit at the price level at point of time t (measured in monetary units);

$\complement _ { \mathfrak { h } } ( { \sf t } - 1 )$ is total constant capital at the level of labour-time at the point of time t-1 (measured in hours or years);

$v _ { \mathrm { h } } ( t - 1 )$ is total variable capital at the level of labour-time at the point of time t-1 (measured in hours or years) (= necessary labour-time at the level of wages); $\mathsf { p } _ { \mathsf { h } } ( \mathsf { t } - 1 )$ is total profit at the level of labour-time at the point of time t-1 (measured in hours or years) (= surplus labour-time).

• I calculated the economy-wide MELT for the German economy for the years 2000–2011 (t=0, 1, 2….10). Table 12.1 gives an overview. The initial MELT (0) in the years 2000 was calculated as:

$$
{ \sf M E L T } \left( 0 \right) = \left( \mathrm { c } _ { \mathrm { p } } ( 0 ) + \mathrm { v } _ { \mathrm { p } } ( 0 ) + \mathrm { p } _ { \mathrm { p } } ( 0 ) \left[ \mathrm { i n \ p r i c e s } \right] \right) / \left( \mathrm { c } _ { \mathrm { h } } ( 0 ) + \mathrm { v } _ { \mathrm { h } } ( 0 ) + \mathrm { p } _ { \mathrm { h } } ( 0 ) \left[ \mathrm { i n \ h o u r s } \right] \right)
$$

The total hours worked during a year (= h) in the economy represent the hours during which living labour creates the value of its labour power (wages, necessary labour-time) and new value (surplus-value, surplus labour-time):

$$
\mathsf { h } ( \mathrm { t } ) = \mathsf { v } _ { \mathsf { h } } ( \mathrm { t } ) + \mathsf { p } _ { \mathsf { h } } ( \mathrm { t } )
$$

where h(t) is total hours worked in year t;

Chapter Twelve | The Information Economy

285

The Information Industry and Marx’s Labour Theory of Value

下一段：[继续读取](https://workshop.closeclaw.top/api/ai/read/50b7f922-b940-4a05-8eb0-c4dc0d8ec9ed.md?start=300&limit=10)