University of East London Institutional Repository: http://roar.uel.ac.uk
This paper is made available online in accordance with publisher policies. Please
scroll down to view the document itself. Please refer to the repository record for this
item and our policy information available from the repository home page for further
To see the final version of this paper please visit the publisher’s website.
Access to the published version may require a subscription.
De Angelis, Massimo.
Globalization no questions! Labour commanded and Foreign Direct
Year of publication:
De Angelis, M. (2008) ‘Globalization no questions! Labour commanded
and Foreign Direct Investment.’ Review of Radical Political Economics, 40, (4) 429-
Link to published version:
Information on how to cite items within roar@uel:
Globalisation? No Question:
Foreign Direct Investment
and Labour Commanded
Massimo De Angelis
Economics-ELBS, University of East London, Longbridge Road, Dagenham,
Essex RM8 2AS; M.DeAngelis@uel.ac.uk
School of Management, Ken Edwards Building, University of Leicester,
Leicester LE1 7RH; email@example.com
A key argument of the globalisation thesis’s sceptics, such as Linda Weiss and Hirst
and Thompson, is that most Third World countries remain marginal to the international
economy in terms of both investment and trade. The sceptics’ argument is supported by
empirical evidence on foreign direct investment (FDI) and trade flows, which are
presented in terms of US dollars. In this paper we re-examine the empirical evidence on
international investment drawing on the concept of labour commanded, central to
Classical Political Economy. Using data on exchange rates and wage rates (or labour
costs), combined with that on dollar values of FDI, we remap the patterns of global
capital flows in terms of the quantities of labour which such investment can mobilise.
On this basis we draw a very different conclusion from the sceptics. In a nutshell, our
conclusion is the following: developing countries are far more integrated into the global
economy than the FDI data suggests, as a result of the amount of labour that can be
commanded with the absolute levels of FDI, itself due to low wages.
JEL Classification: D460; F020; O100; P160
Keywords: labour commanded; FDI; globalisation; capitalism
Globalisation? No Question:
Foreign Direct Investment
and Labour Commanded
Over the last few decades, the number of different definitions of ‘globalisation’ has
mushroomed, together with their associated explanations and rationalisations. This vast
literature is divided across academic/intellectual disciplines and the partialities inherent
in each disciplinary framework have led to different conceptualisations of
‘globalisation’, regarding it as principally economic, social, political or cultural, for
example. Held, et al. (1999: 2–10) propose a useful classification of approaches towards
the study of the phenomenon, distinguishing sceptical, transformationalist and
hyperglobalist theses. Following this classification, Hoogvelt (2001: 120) suggests that
‘these approaches correspond [respectively] to whether one views globalization as
primarily an economic, a social or a political phenomenon.’
The sceptics, whose argument we will critically discuss in this paper, adopt a primarily
economic perspective. They question the relevance of notions such as globalisation to
describe global trends in foreign direct investment (FDI) and trade in the last quarter-
century. In contrast, the transformationalists regard the process of globalisation as
‘primarily a social phenomenon that has brought qualitative changes in all cross-border
transactions’ (Hoogvelt 2001: 120). The phenomenon in question is what David Harvey
(1989) has called ‘time-space compression’ and its emergence can be seen in the fusion
between information and telecommunication technology, as well as in the reduction in
transport costs (Dicken 2002). These two factors have combined to bring the
‘annihilation of space through time’. They have thus created a ‘new economy’ based on
networks, and a consequent transformation of cross-border activities, which is then
called globalisation (Castells 1996).
Finally, the hyperglobalists tend to emphasise power and politics. Their focus is the
nation state, the relevance of which is problematised in the context of global trends.
Here the thesis advanced (see, for example, Strange, 1996) is the declinist view of the
state. Comparing the power of business and transnational production networks, on the
one hand, with that of nation-states, on the other, these authors conclude that the former
is growing relative to the latter. A common illustration of this approach is the ranking of
TNC and government powers, as measured by their net revenue. Such a ranking
positions companies such Ford, Texaco and GM above Brazil and other poorer states
(see Sklair, 2002). The declinist thesis is that nation states have lost power over their
own economies and instead are simple ‘transmitters of global market discipline to the
domestic market’ (Hoogvelt 2001: 120).
Our modest aim in this paper is to problematise the sceptics’ thesis from a critical
political economy perspective. By doing this we open up the ‘economic’ point of view
to “contamination” with issues of power and qualitative change, which are of relevance
to other discussions of globalisation. It must be clear that our purpose here is not the
discussion of this “contamination”, but rather the proposal of an entry point to this
This entry point is the discoursive problematisation of the empirical
evidence on foreign direct investment (FDI) that the sceptics provide in support of their
argument. FDI is defined in terms of some monetary unit, such as US dollars. Following
an old tradition of radical political economy, we argue that this monetary definition
conceals underlying social relations; in fact, money is a social relation, yet it appears to
us as a thing. Here we propose an alternative measure, one that can make help to make
more visible these social relations and can therefore open to account for their critical
problematisation. Drawing on classical political economy’s category of labour
commanded, we thus define the new variable Labour Commanded FDI.
knowledge, neither the large empirical current literature on FDI nor the current
theoretical-historical literature on labour commanded has opened to the question of
power in the way we propose.
The structure of the paper is as follows. In section 2, we first outline the sceptics’
criticisms of the globalisation thesis, before suggesting some problems with this
economic approach. In section 3, we discuss the theoretical foundations of the concept
of labour commanded, which are to be found in the work of Adam Smith, David
Ricardo and Karl Marx. The paper’s quantitative heart is section 4. Here we explain
how we transform the empirical evidence on FDI, cited by the sceptics in support of
their arguments, into a measure of labour commanded and present results. These results
show that, far from being marginal to the global economy, in terms of quantity of labour
commanded, developing countries are, in fact, highly integrated into it. We conclude, in
section 5, by countering anticipated criticisms of our methodology and suggesting
directions for future research.
2. Globalisation and its sceptics
According to sceptics of the globalisation thesis, such as Hirst and Thompson (1999),
Linda Weiss (1997, 1998) and David Gordon (1998), the extent of globalisation, and in
particular its novelty, have been grossly overstated. Hirst and Thompson claim even that
they are ‘convinced that globalization, as conceived by the more extreme globalizers, is
largely a myth’ (1999: 2). To make their argument, the sceptics have charted
quantitative historical comparisons of foreign trade and capital movements and have
concluded that globalisation, as a worldwide integration of national economies, is
nothing new. In fact, taking proxy measures of integration, such as share of foreign
direct investments over production or incidence of trade in national economies, they
suggest the world is less integrated now than it was in the early part of the nineteenth
Thus, for example, Glyn and Sutcliffe write:
The system has ... become more integrated or globalized in many respects. ... Nevertheless what has
resulted is still very far from a globally integrated economy. ... In short, the world economy is
considerably more globalized than 50 years ago; but much less so than is theoretically possible. In
many ways it is less globalized than 100 years ago. The widespread view that the present degree of
globalization is in some way new and unprecedented is, therefore, false. (Glyn and Sutcliffe 1992: 91,
cited in Dicken 2003: 11)
Hirst and Thompson reach a similar conclusion. Examining post-war investment and
trade flows, they find that ‘between 54 per cent and 70 per cent of the world’s
population was in receipt of only 16 per cent of global FDI flows in the first half of the
1990s. In other words, between a half and two-thirds of the world was still virtually
written off the map as far as any benefit from this form of investment was concerned’
(Hirst and Thompson 1999: 74). Kleinknecht and ter Wengel, focusing on the EU, find
that ‘to the extent that trade [and FDI] exceeds the frontiers of the European Union, the
lion’s share of transaction still takes place among the rich OECD countries, notably
with the US. Looking at long-run trade figures, one can also question the proposition
that we are currently experiencing an historically unique stage of internationalisation’
In the sceptics’ approach, then, globalisation as global integration is put under question
or even treated as a ‘myth’ because the bulk of FDI and trade are concentrated in the
‘triad’ of North America, Europe and Japan, the dominant economic blocs. However,
there are several broad problems with this solely economic approach to globalisation.
Here we focus on one them, that of measure.
The problematic of measure permeates almost every issue of interest to (political)
economists. Regarding globalisation, if this phenomenon is understood as one of the
integration of people and livelihoods across the globe, then to what extent do patterns of
FDI (and trade) flows measure it? To what extent does a knowledge of trade and
investment quantities give insights into the mutual relations between, a mother’s work
of reproduction in Indonesia, say, and a steel worker’s work of production in Indiana,
USA or a call-centre worker’s service labour in India? It is not just that there is no
monetary measurement for mothers feeding children, there is no market value attributed
to this work at all.
But despite the lack of measurement of such work, patterns of
capital investments cannot be theorised independently of it, that is, independently of
differentials in the conditions of reproduction, much of it unwaged, of labour-power in
Hence, and perhaps paradoxically, capital movement does in a sense
measure the relative conditions of the reproduction of labour power and for the
accumulation of capital more generally. That is, capital flows provide an index of an
amalgam of wage rates differentials, degree of revolts and insubordinations, degrees of
normalisation to markets, extent of state public spending on entitlements and public
services and so on. For, if we assume that capital flows to those locations where it can
find workers who are healthy, sufficiently willing and hardworking, appropriately
skilled and where, moreover, wage rates (and labour costs in general) are sufficiently
low, then the fact that capital does (or does not) flow into a particular location indicate
that these conditions do (or do not) exist
What is the implication of all this for our critique of the economic view of
globalisation? The implication is that monetary measures for us matter more as a
moment in a process (indeed, a contradictory process based on conflict and on the
articulation between monetised production and non monetised reproduction) than a
static picture of a ‘structure’. For this reason, to argue, as the sceptics do, that trade and
FDI are concentrated in the triad, does not in fact question globalisation as process of
capitalist integration. On the contrary, this empirical evidence perhaps reveals the
capitalist character of this process of integration, one based on the command over
labour and its differentiation along a continual reconfiguring international division of
labour. Given the miserable wages of the global South in relation to those paid in the
Northern developed countries, and the overall lower value of labour power in these
countries, the fact that only 15 or 20 per cent of world FDI flows into the South may
demonstrate, not that global investment is unfairly distributed, but rather that it is fairly
distributed, according its capacity to command labour within the process of capitalist
For example, in the United States, $20 will employ one worker for one hour, that is, it
will command just a single hour of labour time. But, in China or Thailand, $20 can put
four people to work each for ten hours, whilst in India that $20 is sufficient to put ten
people to work, each for ten hours. When the difference that $20 makes is between
commanding one hour of labour time, on the one hand, and commanding 40 hours or
100 hours, on the other, it matters much less that less FDI goes to the South. This is the
problematic introduced by what classical political economy calls labour commanded.
3. Theoretical foundations of labour commanded (Smith, Ricardo, Marx)
As is well-known, Adam Smith introduced the notion of labour commanded in one of
two theories of value. In his first, ‘labour-embodied’ theory, a commodity’s value is
determined by the labour time materialised, or embodied, in it, that is, the quantity of
labour necessary to produce it. In the second, ‘labour-commanded’ theory, the
commodity’s value depends on the labour it can itself command. Now these two
definitions of value are in contradiction as the former (embodied or materialised labour) is
independent from the value of labour (wages), while the latter depends on the value of
labour. According to Smith, materialised labour was true in
the early and rude state of society which preceded both the accumulation of stock and the
appropriation of land, the proportion between the quantities of labour necessary for acquiring different
objects seems to be the only circumstance which can afford any rule for exchanging them for one
another. (Smith 1970: 150)
This condition is altered ‘as soon as stock has accumulated in the hands of particular
persons’ (Smith 1970: 151). That is, as soon as private property is introduced, ‘something
must be given for the profits of the undertaker of the work who hazards his stock in this
adventure’ (ibid: 151). Further, as soon as land becomes private property, ‘the landlords,
like all other men, love to reap where they never sowed, and demand a rent even for its
natural produce’ (ibid: 152).
Thus, in the ‘civilised’ state of society, the value of a commodity resolves into labour
The value of any commodity … to the person who possesses it, and who means not to use or consume
it himself, but to exchange it for other commodities, is equal to the quantity of labour which it enables
him to purchase or command. Labour, therefore, is the real measure of the exchangeable value of all
commodities. (Smith 1970: 133)
Ricardo advances two interlinked objections to Smith’s theory of labour commanded.
First, Smith’s notion of labour commanded depends on the value of ‘labour’. But the value
of ‘labour’ depends in turn on the value of those commodities constituting workers’
subsistence. Thus, we go round in circles. Second, suppose the labour required to produce
a given quantity of food doubles. ‘[Y]et, the labourer’s reward may possibly be very little
diminished.’(Ricardo 1951: 15) This is because Ricardo assumes subsistence-level wages.
Thus, if we measure the value of that quantity of food in terms of labour embodied, value
has doubled. But if we measure that same quantity of food in terms of the labour for which
it will exchange, than value has remained constant (the same amount of food sets in motion
the same amount of labour). For this reason, Ricardo proposes his own version of the
labour embodied theory of value:
The value of a commodity, or the quantity of any other commodity for which it will exchange,
depends on the relative quantity of labour which is necessary for its production, and not on the greater
or less compensation which is paid for that labour (Ricardo 1951: 11).
Marx also rejects Smith’s theory of labour commanded, as a theory of value. Instead, he
founds his value theory, not on the quantity of actual (concrete) labour embodied in an
individual commodity, but rather on the quantity of (abstract) labour socially necessary to
produce it. Marx’s critique recognises that labour time does not stop being the immanent
measure of value ‘from the moment when the conditions of labour confront the wage-
labourer in the form of landed property and capital’ (Marx 1969: 73). Rather, it is the
‘expressions “quantity of labour” and “value of labour” [that] are not identical’ and
therefore the value of commodity ‘although determined by the labour-time contained in
them, is not determined by the value of labour’ (ibid: 73)
If labour commanded is not for Marx the immanent measure of value, it gives us another
important indication, and he qualifies Smith’s view in its role:
When [Smith] comes to the exchange between materialized labour and living labour, between
capitalist and workers, and then stresses that the value of the commodity is now no longer determined
by the quantity of labour it itself contains, but by the quantity … of living labour of others which it can
command … he is not in fact saying by this that commodities themselves no longer exchange in
proportion to the labour-time they contain; but that the increase of wealth, the increase of the value
contained in the commodity, and the extent of this increase, depends upon the greater or less quantity
of living labour which the materialised labour sets in motion. And put in this way it is correct (Marx
Surprisingly, this acknowledgement of ‘something deeper’ (ibid: 71) in Smith’s
argument has generally been overlook by the extensive exegetic literature of Marx’s
theory of value. If the value of labour-power is not an indication of the value of
commodities, it is certainly an important factor in determining the amount of living
labour that can be put to work by a given quantity of capital. It can therefore provide us
with an idea of the ‘increase in wealth’ (in value terms) that a certain quantity of capital
(still in value terms) can potentially generate, through setting living labour in motion.
This meaning, in which Marx refers to labour-commanded as that quantity of living
labour which is set in motion by a given amount of capital, is also evident in other
contexts of his writing (see, for example, Marx 1981: 323). There is however another
sense in which we can gain insight by the term labour commanded. This is the potential
living labour that can be put in motion by a certain money-value of capital. This
understanding, in fact, relates back to Hobbes’ insight that wealth is power and to
Smith, who also links labour commanded with power. This power consists precisely of
that ‘command over all the labour’ (Smith 1970: 134). Marx, in turn, argues that,
The power which each individual exercises over the activity of others or over social wealth exists in
him as the owner of exchange value, of money. The individual carries his social power … in his
pocket. (Marx 1973: 156–57)
This conception of labour-commanded stresses the power of money to control others’
time, to put people to work, to command labour, whether or not this power is actually
exercised. Indeed, the command over labour and the exercise of this command, refer to
two different concepts within Marx’s theory of value and surplus-value, which is based
on the distinction between labour and labour power. The former is not a commodity, but a
life-activity creating value. The latter is a commodity to be exchanged on the market and
has a price like any other commodity. Labour commanded therefore is not yet a measure of
labour expended, although it gives us an indication of the amount of labour that can be
expended, that can potentially be set in motion.
Changes in quantities of labour commanded therefore, as reflected in changes in monetary
FDI patterns translated into labour commanded, for example, do not give us an indication
of labour actually expended or embodied; rather, they point to changes in the quantity of
waged labour that can potentially be set in motion within the accumulation process.
However, this quantity is also dependent upon the level of wages, which in turn depends
upon general conditions of labour-power reproduction. Thus, in this context, the notion of
labour commanded opens up the problematisation of a variety of factors, including
relations between classes and between waged and unwaged sections of the working class,
which the simple monetary measures of FDI disguise. This problematisation is of course
beyond the scope of this paper.
Our effort here is . that a similar concept to the problematicemerging from a discussion of
more recent use of the comparing of It is perhaps important to point out at that the
emphasis on the potential character of labour set in motion by quantity of money is a
different problematic than the one of finding a measure to
(Amado 2003) Adri ana More i ra Amado
To summarise this section, there are two ways in which we can conceive of labour-
commanded: first, as a measure of value (for Smith) or increase in value, that is, of
surplus value (for Marx); second, as a measure of the (possibly potential) quantity of
living labour which can be set in motion by a quantity of money as capital. It is the
second meaning which is of interest to us in this paper. By converting statistics on
foreign direct investment and trade from money terms into terms of labour commanded,
we can gain insights of the increase in control over wealth, understood in terms of
labour time that can potentially be set in motion by a given quantum of money, which
results from these indicators of economic globalisation. In short, the question of the
extent to which global capital is inserting itself into people’s lives cannot be answered
by considering only absolute quantities of money. Instead, we must examine also the
potential labour (life) time that these quantities can set in motion in different contexts.
We turn to this task in the next section.
4. Foreign direct investment and labour commanded
Given monetary flows of foreign direct investment, valued in US dollars, we obtain
figures for annual labour commanded by dividing these FDI inflow figures by US dollar
values of hourly wage rates in manufacturing. That is, the number of hours of labour
commanded in country i in year t by foreign direct investment inflows, lc
, is given by,
is annual FDI inflow (in US dollars) into country
exchange rate against US dollars and
is the hourly labour cost in local currency.
We then sum across developed and developing countries to obtain aggregate annual
We follow the United Nations’ conventions regarding definitions of ‘developed’ and
We then select the top 20 developed-country and top 20
developing-country recipients of FDI over the period 1970–2002. These countries are,
in descending order:
: United States, United Kingdom, Belgium-Luxembourg,
Republic of Germany, Netherlands, Canada, Spain, Australia, Italy, Ireland, Denmark,
Japan, Switzerland, Austria, Finland, Norway, Portugal, New Zealand.
: China, Hong Kong, Brazil, Mexico, Singapore, Argentina, Malaysia, Chile,
Thailand, India, Colombia, Taiwan, Peru, South Africa, Philippines, Indonesia,
Pakistan, Sri Lanka, Bangladesh, Democratic Republic of Korea.
Aggregate FDI inflow figures for each of these two groups of countries are plotted in
Also plotted in figure 1 are total FDI inflows into developed countries, into
developing countries and globally. It is clear from this figure that the 20 developed
countries selected receive the lion’s share of all FDI inflows into developed countries
(more than 90 percent in every year and 96 percent on average). The selected
developing countries account for at least 50 percent of all FDI inflows into developing
countries in all but two years (1975 and 1982) and 78 percent on average.
FIGURE 1 ABOUT HERE
We can observe from figure 1, the empirical basis for sceptics’ critique of the
Over the three decades, FDI inflows into developed countries
have dwarfed those into developing countries, averaging, respectively, 73 percent and
27 percent of total flows.
The data is presented slightly differently in figure 2, in which we aggregate FDI flows
over sub-periods. The story is the same however: FDI has grown exponentially over the
three decades, but inflows into developed countries dominate those into developing.
In figure 3 we present results for our new labour commanded FDI variable for the
It is clear here that the sceptics’ interpretation of globalisation is
completely reversed. In terms of labour commanded FDI, the lion’s share now ‘belongs’
to developing countries. This is even clearer in figure 4. When we measure capitalist
investment in terms of its potential to mobilise labour, i.e., in terms of the social power
of money, there would seem to be no doubt: capital’s pervasive globalisation across the
globe can also be made intelligible quantitatively. As far as capital in concerned
therefore, there is no need for greater investment in the South in relation to the North: it
is already able to command masses of living labour there, and this it is able to so by
paying pitiful wages.
FIGURES 3 AND 4 ABOUT HERE
We should stress here that none of the figures presented here should be treated as exact.
First, we have drawn on data for labour costs or wages in manufacturing only, since this
is far more readily available than economy-wide figures. Second, for many countries
complete data on
labour cost is not available; for these countries hourly figures
were estimated by also utilising working-hours series, which themselves are frequently
themselves sometimes had to be extrapolated or estimated
from earnings or wage series.
Finally, the figures published by the various bodies (the
US BLS, the ILO and the UN) are themselves likely to be subject to errors and not
always directly comparable, given that sources, coverage, sample sizes and so on, vary
from country to country. However, these figures do present a broad-brush overview,
which illustrates general
in FDI-labour-commanded and comparisons between
developed and developing economies.
We can also note that the figures are likely to
-estimate quantities of labour
commanded in developing countries vis-a-vis developed countries for two reasons.
, as discussed in section 4, above, the 20 selected developed countries received on
average 96 per cent of all FDI inflows (in dollar terms) into developed countries, whilst
the corresponding figure for the 20 developing countries is 78 per cent. To obtain a
more accurate reflection of FDI labour commanded in developing countries as a whole
then, the figures presented here should perhaps be inflated by a factor of perhaps 25 to
30 per cent.
In contrast, the figures for developed countries need only be inflated by 5
, as noted in the Appendix, below, for many countries and years, labour
costs are estimated from figures for earnings or wages. But, because of higher rates of
business taxation, more stringent laws regulating workplace health and safety, as well as
working condition more generally being more favourable to workers, the ratio of labour
costs to earnings will tend to be higher for developed countries than it is for developing.
The estimation algorithm does not take this into account, however; thus, developing-
country labour-costs estimates are likely to be biased upwards and labour-commanded
estimates will be biased downwards.
We should also emphasise that the concept of labour commanded refers to the
labour (life) time that can be put to work. It is of secondary importance whether a
particular quantum of money capital is actually advanced to employ people, rather than
invested in fixed capital, or used to transfer ownership of existing productive assets,
say. Thus, the objections of, for example, Weiss (1997, 1998) that a high proportion of
FDI is either directed towards ‘non-productive’ assets or is concentrated on merger and
acquisition (M&A) activity do not invalidate our argument.
In this paper we have begun to explore the classical idea of labour commanded in its
application to modern processes of globalisation. In this preliminary work, we have
suggested that monetary measures of global FDI trends, when translated into terms of
labour commanded, can reveal results which are quite the opposite of those cited by
economic critics of the globalisation thesis. In fact, according to our estimates of labour
commanded, the populations of the global South (developing countries) are ‘benefiting’
from this form of investment far more even than populations in the North, and are
certainly far from being ‘virtually written of the map’, as Hirst and Thompson suggest.
The approach we have adopted here allows us to problematise the notion, which is held
dear by conventional economic wisdom and embedded in economic discourse, that
investment is uniquely associated with a ‘benefit’ to the recipient local population. In
fact, a large quantity of labour-commanded FDI could well be associated with poorly
performing social and environmental indicators, which results in a high level of labour
commanded per dollar. As one example of the double-edged nature of investment, one
could reflect on the investment programme to build a series of dams along the Narmada
river and its tributaries in central India. This investment can certainly be seen to
‘benefit’ local unemployed labourers and engineers, but hardly those thousands of
families who have to be displaced to make room for the development. High
displacement rates and, in general, the high vulnerability of the local population would
be reflected in prevailing wage rates through something akin to Marxian theory of the
reserve army of labour (Marx, 1976).
The monetary figures of FDI are not able to
capture the social costs associated with investment programmes. In contrast, labour
commanded FDI figures, through their emphasis on power and their link to conditions
of reproduction captured by the prevailing wage rate, are better able to reflect such
This methodology thus provides thus a framework within which other questions on
power can be posed and we conclude by suggesting a few such possible studies. In the
first place, the same approach can be applied to trade figures, which we anticipate could
be revealing of patterns of global integration along the lines we have defined in this
paper. If relatively small amounts of FDI in developing countries become relatively
large amounts of labour commanded FDI, so the same would apply for the relatively
small monetary figures of global trade when measured in hours of labour commanded.
Second, for accounting convenience, our analysis has aggregated figures such that the
globe has been divided into simply ‘developed’ and ‘developing’ countries. But, by
considering more disaggregated regions –- for example: ‘old’ (Western) Europe; eastern
Europe (the ‘transition economies’ of the former Soviet bloc countries); the United
States and Canada; Japan; Asian NICs; China and other Asia; Latin America; Africa —
this static analysis can be extended to explore the dynamics and patterns of capital’s
flows within and between blocs, in such a way as to compare FDI and trade estimates in
terms of both dollars and labour commanded.
Third and finally, we can gain further insights by more directly investigating the
determinants of labour commanded FDI. As suggested above, these include the general
conditions of reproduction of labour-power such as literacy, health, education, as well
the existence or likelihood of social conflict, which can be proxied by rates of
unionisation, figures on industrial disputes and so on.
The aim here would be to model
patterns of labour commanded FDI flows in terms of general conditions of the
reproduction of labour power and of social conflict or harmony.
Appendix: Methodology, Data Sources and Estimation
Foreign direct investment
All figures are drawn from the United Nations Conference on Trade and Development
(UNCTAD) Foreign Direct Investment database (available on-line at
http://www.unctad.org), which reports annual FDI inflows in US dollars.
Hourly labour costs
We draw on three data sources: the United States Bureau of Labor Statistics (USBLS),
the International Labour Organization (ILO) and the United Nations (UN).
The US Bureau of Labor Statistics (BLS) publishes data on hourly labour compensation
costs and hourly direct pay for production workers in manufacturing in 30 selected
countries, compiled as part of its Foreign Labor Statistics programme
(http://www.bls.gov/fls/home.htm). Data is available for the period 1975–2002 and is
published both in local currencies and the US dollar equivalent, which we employ. The
30 countries include all 20 developed countries, plus Brazil, Hong Kong, Korea,
Mexico, Singapore, Sri Lanka and Taiwan of our developing countries. The BLS’s
compensation measures include all items of labour compensation, including: employer
social insurance expenditures and other labour taxes; overtime pay, shift differentials,
other premiums and bonuses, and cost-of-living adjustments; holiday pay; the cost of
benefits in kind; employer legally-required expenditures on retirement and disability
pensions, health and other insurance schemes, and family allowances. The BLS argues
their figures ‘are appropriate measures for comparing levels of employer labor costs’.
These figures are thus suitable for our purposes, too.
The International Labour Organization (ILO) also compiles data on manufacturing
labour cost/employee compensation and wages, published on its on-line Laborsta
database (http://laborsta.ilo.org). Although the technical definitions of labour cost and
employee compensation differ slightly, the two measures are closely related and we do
not distinguish between them. Both concepts also share many common elements with
the USBLS’s labour-compensation measure. The principal difference between the ILO
and BLS measures is that the former includes costs of recruitment, employee training
and plant facilities and services, such as cafeterias and medical services. According to
the BLS, these ‘account for no more than 4 percent of total labor costs in any country
for which the data are available’. Substantially complete (over the period 1970–2002)
ILO labour cost series are available for seven developed and six developing countries,
whilst partial series are available for a further three developed and two developing
The ILO wage or earnings measure includes employee remuneration in cash and in
kind, both for time worked and for time not worked such as annual vacations or other
paid leave or holidays. The measure also includes bonus payments and family and
housing allowances; it does not include employer contributions so social security and
pension schemes or the benefits received by employees under such schemes.
Substantially complete ILO wage series are available for all but seven of the 40
Both ILO measures — labour cost/employee compensation and wages/earnings — vary
in their reporting unit: per hour, per day, per month or per year. Where the measure is
not reported on an hourly basis, we draw on ILO hours of work data in order to compute
Finally, the United Nations publishes wages/earnings data, adopting similar definitions
to the ILO (United Nations,
, New York: United Nations, various
years). Substantially complete wages series are available for 22 of our 40 countries.
Again where these figures are not provided on an hourly basis, we adjust them using
both ILO and UN data on working hours.
The two labour cost series — compiled by the USBLS and the ILO — are clearly most
appropriate for our purposes, but are incomplete. We employ missing variable analysis
in order to complete these series.
Finally, we take the two series’ mean.
In order to investigate the reliability of earnings series in predicting labour cost, we
calculate various ratios of the latter to the former. In three out of four possible ratios we
find that it is significantly higher for developed countries than for developing. This is to
be expected, since we would expect developed countries to have more generous systems
of social security, more stringent health and safety legislation and so on. But since the
labour costs series are more complete for developed countries, our estimates of these
figures for developing countries are likely to be biased upwards. As a consequence we
-estimate labour-commanded for this group of countries
s. Data are drawn from the International Monetary Fund’s International
Financial Statistics database (http://ifs.apdi.net/imf/).
An earlier version of this paper was presented at the Association for Heterodox
Economics’ Annual Conference, University of Leeds, 16–18 July 2004, and we are
grateful for other participants’ comments. We are also grateful for Paul Dunne’s useful
suggestions. The usual caveat applies.
We are of course aware that many economists, mostly working outside the
mainstream, do refuse to consider the ‘economic’, the ‘social’, and the ‘political’ as
independent spheres and instead regard them as interrelated. Yet, in our approach we
are prone to reject this distinction altogether. For us the understanding of money as
labour commanded that we develop in the paper, implies that the three “spheres” are
simply three analytical determinations of one immanent social relation. We are trying to
attract attention to this social relation.
Although the category of labour commanded as discussed within the classical political
economy tradition can offer interesting insights when compared to Keynes use of wage-
units in his
, this comparison is beyond the scope of this section. For an
interesting case of such a comparative analysis see Amado (2003).
Some of the recent discussion of the category of labour commanded include for
example Naldi (2003), Screpanti (2003), and Glyn (2006).
The value of women’s unwaged work was recently estimated to be US$11 trillion per
annum (United Nations: 1995).
The tendency to ignore questions of reproduction and its relationship with production
is one of the other problems with purely economic approaches to globalisation. If
globalisation is viewed solely as a question of integration of different
, that is,
set of human activities which produce commodities, then we ignore the
crucial set of questions concerning the integration of reproductive activities, which
include large chunks of unwaged labour. This problem is thus closely linked to the
problem of measure.
Of course, whether such labour is actually set in motion is also an interesting question.
The answer will depend both upon the ‘quality’ of the labour-power (its skill levels,
degree of subordination and so on), upon market conditions, and ultimately, upon power
For a general discussion of the link between waged production and unwaged reproduction within the
context of current global dynamics of production, see De Angelis (2007).
Data are obtained from the United Nations, the International Labour Organisation, the
United States Bureau of Labor Statistics and the International Monetary Fund. For more
details of these sources and the methodology used to estimate labour costs, see the
Clearly, such definitions are historically contingent. For example, Mexico was
admitted to the Organisation for Economic Cooperation and Development in 1994,
whilst in international trade statistics the Southern African Customs Union is treated as
a developed region. For our purposes, we define both Mexico and South Africa as
Prior to 2002 the UNCTAD reports only aggregate figures for Belgium and
Luxembourg. We therefore treat Belgium-Luxembourg as a single country.
Note that the vertical-axis scale is logarithmic.
Our figure 1 resembles very closely Hirst and Thompson’s figure 3.2, though their
series end in 1995 (Hirst and Thompson 1999: 71).
Again, the scale of the vertical-axis is logarithmic.
See the Appendix for more details.
The appropriate factor is not constant over the period. In fact, for some years, it
would be near to 100 per cent.
This is not to say that this question is unimportant. It makes a great deal of difference
to the citizens of a host country whether foreign capital invests in a labour-intensive
garment factory in an export-processing zone, say, or a fleet of high-tech trawlers
employing relatively few fishermen. Similarly, it matters whether this capital is used to
set-up new facility or simply assumes ownership of existing plant. But our argument
concerns the metric used to assess the degree to which states are integrated into the
global capitalist economy, rather than the specifics of how capital exploits workers in a
The Sardar Sarovar Project, the largest single dam in the Narmada Valley
Development Project, was only able to start through a World Bank loan of $450 million.
Following international pressure and an independent review, however, the Bank was
forced to withdraw its support of the project. See Caufield (1998: chapter 1).
Adopting a slightly different, though complementary, theoretical framework, we
might use labour costs figures, in conjunction with estimates of the organic composition
of capital (the capital-labour ratio), to obtain estimates of labour-embodied or labour
(1983) ‘social model’, which uses such variables to better
explain post-war U.S. productivity growth.
We use the package in SPSS, choosing the expectation-maximisation method. The
algorithm uses information on the relationships between the respective variables — the
five earnings or labour cost series, plus the year — where observations are available, in
order to compute the most likely values for years or countries where they are not.
Amado, A.M. (2003) Smith e Keynes sobre as unidades de salario. Nova-economia.
January-June; 13(1): 167-184
Castells, M. (1996)
The Rise of the Network Society
, Cambridge, Mass and Oxford UK:
Caufield, C. (1998)
Masters of Illusion: The World Bank and the Poverty of Nations
De Angelis, M. (2007)
The Beginning of History. Value Stuggles and Global Capital.
London: Pluto Press.
Dicken, P. (2003)
Global Shift. Reshaping the Global Economic Map in the 21st
, 4th edition, London: Sage.
Glyn, A. and R. Sutcliffe (1992) ‘Global but leaderless? The new capitalist order’, in R.
Miliband and L. Panich (eds.)
New World Order: The Socialist Register
Merlin Press: 76–95.
Glyn, A. (2006) ‘The Corn Model, Gluts and Surplus Value’,
Cambridge Journal of
, 30(2): 307-312.
Gordon, D. (1998) ‘The Global Economy: New edifice or crumbling foundation?’,
Harvey, D. (1989)
The Condition of Postmodernity,
Held, D., A. McGrew, D. Goldblatt and J. Perraton (1999)
Politics, Economics and Culture,
Hirst, P. and G. Thompson (1999)
Globalization in Question: The International
Economy and the Possibilities of Governance
, Second Edition, Cambridge: Polity.
Hoogvelt, A.M.M. (2001)
Globalization and the postcolonial world : the new political
economy of development
, Baltimore, MD : Johns Hopkins University Press.
Kleinknecht, A. and J. ter Wengel (1998) ‘The Myth of Economic Globalisation,
Cambridge Journal of Economics
’, 22(5): 637–647.
Marx, K. (1969)
Theories of Surplus Value
, Part 1, London: Lawrence and Wishart.
, Harmondsworth: Penguin.
, Volume 1, Harmondsworth: Penguin.
, Volume 3, Harmondsworth: Penguin.
Naldi, N. (2003) ‘Labour Employed in Production and Labour Commanded: a Ricardian
Cambridge Journal of Economics
, 27(4): 547-562
Perraton, J. (2001) ‘The global economy—myths and realities’,
Cambridge Journal of
, 25 (5): 669–684.
Ricardo, D. (1951)
On the Principles of Political Economy and Taxation
(Volume I of
The Works and Correspondence of David Ricardo
, edited by P. Sraffa),
Cambridge: Cambridge University Press.
Screpanti, E. (2003) ‘Value and Exploitation: a Counterfactual Approach’,
Sklair, L. (2002)
Globalization: capitalism and its alternatives
. Oxford: Oxford
Smith, A. (1970)
The Wealth of Nations
, Books I—III, Harmondsworth: Penguin.
Strange, S. (1996)
The Retreat of the State. The Diffusion of Power in the World
, Cambridge: Cambridge University Press.
United Nations (1995)
Human Development Report
, New York: United Nations.
Weiss, L. (1997) ‘Globalization and the Myth of the Powerless State’,
New Left Review
Weiss, L. (1998)
The Myth of the Powerless State
, New York: Cornell University Press.
Weisskopf, T.E., S. Bowles and D.M. Gordon (1983) ‘Hearts and Minds: A Social
Model of U.S. Productivity Growth’,
Brookings Papers on Economic Activity
Figure 1. Annual FDI inflows.
1970 1974 1978 1982 1986 1990 1994 1998 2002
US$, billions (log scale)
FDI, selec ted developed
FDI, selec ted developing
Figure 2. Total FDI (in constant prices) inflows to 20 developed and 20 developing
1970--78 1979--84 1985--90 1991--96 1997--2002
FDI inflows , billions of cons tant 1995 US$
Figure 3. Annual FDI inflows and labour commanded for 20 developed and 20
: various, see text.
1970 1974 1 978 1982 1986 1990 1994 1 998 2002
FDI inflows , $US billion (log sc ale)
Labour commanded, billion hours (log scale)
FDI, selec ted developed
FDI, selec ted developing
Lab c om, developed
Lab c om, developing
Figure 4. Total FDI labour commanded inflows to 20 developed and 20 developing
: various, see text
1970--78 1979--84 1985--90 1991--96 1997--2002
Labour commanded, billion hours