Surplus Value and Transfer Value

Gernot  Köhler

School of Computing and Information Management
Sheridan College
Oakville, Ontario, Canada L6H 2L1

November 1999


This study attempts to clarify the conceptual relationship between "transfer value" (unequal exchange, superexploitation)  and "surplus value".


The global income gap is widening; the First World is dominating the world economy and shaping it so as to serve its own interests. "Unequal exchange" is taking place between periphery and center of the world, so that, in effect, the poor countries subsidize the rich and the labourers in the periphery countries are the most exploited in the world-system. Amin calls this the "superexploitation" of the labourers of the Third World. The process of "unequal exchange", "center-periphery exploitation", or "superexploitation" works through such channels as biased terms of trade and biased exchange rates, or, generally speaking, through the biased (global) "structure of prices". Thus, Amin writes:   " follows that there is superexploitation of labor in the periphery which totals as much as $300 billion and which is largely hidden in the structure of prices" (Amin 1980: 19) The figure of US $300 billion is Amin's estimate of the annual outflow of value from the world's periphery to the center (around 1980, in current prices), due to "unequal exchange". Amin calls this value alternatively "unequal exchange" or "transfer value" and states: "It is this superexploitation that prevents the development of productive forces on the periphery." (Amin 1980: 25)  I have argued separately that the same world-level mechanism is also responsible for massive unemployment in the center (First World, OECD). An interesting problem arises, namely: What is the conceptual relationship between "transfer value" and "surplus value"?


The concept of "surplus value"  is central to  Marxist literature. Similar concepts appears under different names and symbols also in post-Keynesian and  OECD  literature. Conceptually, "surplus value" is the amount of value which the worker helps to produce but does not get. In the aggregate, the "mass of surplus value" in the economy is the difference between aggregate income and total wages. Thus, if a total of 10 billion dollars worth has been produced and earned by society and if the total amount of wages paid is 6 billion dollars, then the "mass of surplus value" (i.e., aggregate surplus value) is 4 billion dollars. Conceptually, this is clear.  The heterodox "national accounting identity" uses a similar distinction, namely, Q  =  w L + r K   where wL is all wage income (namely, wage rate w  times number of workers L) and rK is all other income (namely, rate of return on capital r times volume of capital K) (see, for example, Shaikh 1987). In Keynes's General Theory we find the expression "wage bill (W)" which corresponds to wL in the heterodox national income accounting identity (Keynes 1964). Finally, one finds a category of "operational surplus" in the statistical accounts of OECD (e.g., OECD 1995: 34.-- This "operational surplus" comes out to be about 25% of GDP, which is only about one half of "surplus value", but the concepts have some similarity.) At the measurement level (operational level) there are differences between authors. Some of these differences have to do with the fact that Marx wrote at a time long before modern GDP-type national accounting had been developed. A study which uses authentic Marx-based concepts and measurements of  surplus value is Moseley (1991). Other Marxist authors calculate "surplus value" in a GDP-type context (see, the discussion in Moseley 1991). (As a result, measurements of surplus value differ between various authors.)


The intellectual history of the concept of  global transfer value starts, perhaps, with the Latin American school and Prebisch.
Prebisch criticized the terms of trade between center and periphery countries and argued that the terms of trade were deteriorating for the Third World. The notion of unfair terms of trade was then further developed by Arghiri Emmanuel who developed the theory, and coined the term, of "unequal exchange" (Emmanuel 1962, 1972). Emmanuel proposed a conceptual definition of unequal exchange, but did not have any operational definition nor quantitative estimates of the magnitude of unequal exchange. Emmanuel's theoretical definition of "unequal exchange" is, as follows:

"...unequal exchange is the proportion between equilibrium prices that is established through the equalization of profits between regions in which the rate of surplus value is ‘institutionally’ different..." (Emmanuel 1972: 64).

This is a theoretical definition and not an operational definition. How unequal is the exchange? How much value is unfairly transferred from the periphery to the center? Emmanuel did not give any precise figures. However, Emmanuel gave a sense of magnitudes involved when he wrote that: the "loss [sc. resulting from unequal exchange] ... is enormous in relation to the poverty of the underdeveloped countries while being far from negligible in relation to the wealth of the advanced countries." (Emmanuel 1972: 265) Some authors provided estimates for individual countries. Gibson (1980) estimated a loss for Peru in 1969 in its trade with USA as 38% of its exports to the USA (see, Raffer 1987: 94). Webber and Foot (1984) estimated for the Philippines in 1961 that Philippine exports would have amounted to 5.269 billion pesos, if valued at Canadian wages and prices, instead of the actual 1.129 billion pesos (see, Raffer 1987: 95). Worldwide estimates of unequal exchange are available from Amin and Köhler. Amin mentions global figures for unequal exchange in two separate publications -- an estimate of $22 billion for 1966 (Amin 1976: 143-144) and an estimate of $300 billion for around 1980 (Amin 1980: 19). Köhler presented estimates of global unequal exchange for 1965 and 1995  (Köhler 1998a, 1998b).

In the intellectual development from a purely theoretical definition of unequal exchange (Emmanuel) to some quantification of the concept (Amin) and more quantification (Köhler), the concept of a "transfer value" crept into the discussion (e.g., Amin 1980). The situation can, perhaps, be seen in such a way that the expressions of  "unequal exchange" and "transfer value" are two sides of the same thing -- namely, "unequal exchange" describes the process or mechanism, whereas "transfer value" describes the quantity of value which moves from A to B in the process of unequal exchange. Some of the literature uses the term "unequal exchange" interchangeably for both the process and the quantity.


In addition to the expressions "transfer value" and "unequal exchange", Amin also uses the term "superexploitation".
While this expression can be understood in the sense of "superlative exploitation" or "extreme exploitation", there is more to it than that, namely, "superexploitation" has also the meaning of "exploitation superimposed on exploitation" or "double exploitation". In Amin's usage, superexploitation does not happen to labour in the center, no matter how exploited they may be. Superexploitation (double-exploitation) happens to labour in the periphery--namely, because the labourers in the periphery generate both surplus value (to the business owners, capitalist class) and global transfer value (to the rich countries, center countries). Who benefits from the "superexploitation" of the laborers of the periphery of the world-system? Here Amin lists three categories of beneficiaries, namely: "(1) the local classes exploiting them (landowners and local capitalists); (2) the capital that dominates the whole system -- that of the monopolies; and (3) the workers of the imperialist centers (or at least some of them)." (Amin 1980: 19) This can be interpreted in such a way that laborers anywhere (both in the periphery and the center) generate surplus value; but, in addition, the laborers of the periphery also generate global transfer value, part of which is benefiting labour in the center countries via the mechanism of unequal exchange


From an orthodox Marxist point of view, "unequal exchange" is not "exploitation", even though it is exploitation in the common sense of the word. That is probably one of the reasons why Emmanuel chose a  name other than "exploitation" (namely, "unequal exchange") and why Amin coined yet another expression, namely "superexploitation", and why others use yet other special expressions like "world technological rent" (Çakmak 1998) and why the term "transfer value" comes in handy for purposes of communication. The exact relationship between "surplus value" and "transfer value" contains the following problematique: Whereas both concepts refer to value which is extracted from somebody by somebody, the "somebodies" on the giving and receiving end are different ontological entities and are overlapping (not mutually exclusive). When we speak of surplus value, the two entities are worker and capitalist (at the micro-level) or labour and capital (at the macro-level) -- i.e., functionally defined roles or classes. When we speak of  transfer value, the two entities, which are referred to, are periphery country and center country, or, periphery and center -- i.e., territorially defined entities or entity classes. The categories "labour and capital" and "center and periphery" are overlapping. As a result, the value flows in the two types of relationships (surplus value, transfer value) are also overlapping categories. That requires some clarification.


The problem can be discussed around two formulae. We can define aggregate surplus value, as follows:

SV =  Y - W

in words, aggregate surplus value (SV) is the difference between aggregate income of a community (Y) less aggregate wages of a community (W). [Authors disagree about the measurement of Y and W, but let us ignore that for the time being.]

We can define "transfer value", as follows:

T  =  d*X  - X

in words, aggregate transfer value (T) is the difference between the true value of exports from periphery to center (d*X) less the actually paid value for the same exports (X). [d is a distortion factor, measuring undervaluation. This is the formula which I proposed in Köhler 1998a. Other components of transfer may have to be added in order to capture all value that is being transferred. But this formula is sufficient for the present exercise.]


Let's assume that there is a world with only two countries -- namely, country N (for "North", rich) and country S (for "South", poor). Each country has 10 workers and one capitalist entrepreneur.

(a)  the GDP of rich country N is 1000 francs
(b) the GDP of poor country S is 1000 yuan
(c) the wage bill W of rich country N is 600 francs (shared by 10 workers)
(d) the wage bill W of poor country S is 600 yuan (shared by 10 workers)

For this example we can calculate aggregate surplus values, as follows:
SV = Y - W, therefore:

(e) SV of country N = 1000 - 600 (francs) = 400 francs
(f) SV of country S  = 1000 - 600 (yuan)  = 400 yuan

What is the transfer value?
In order to answer that question, we must have additional information. Let's make the further assumptions that  there is international trade between N and S;
(g) the exchange rate is 5 : 1, namely, 5 yuan = 1 franc
(h) country S exports to N in the amount of 250 yuan
(i) country N exports to S in the amount of 50 francs
(i.e., the trade is balanced, since 250 yuan = 50 francs)

In order to calculate the transfer value, we must have still more information, namely, is this a fair exchange? If not, how unfair is it? Let us assume that it is an unequal exchange (unfair trade) and that it is unfair because the exchange rate is unfair. Let us assume that:
(j) the fair exchange rate would be 1 : 1, namely, 1 yuan = 1 franc

We can now calculate the aggregate transfer value, as follows:
T = d*X - X , therefore:

(k) the distortion factor d is equal to the actual exchange rate divided by the fair exchange rate, or, (5/1) : (1/1) = 5.0
(l) T (in terms of yuan) = 5*250 - 250 = 1000 yuan  [expressed in undervalued yuan]
(m) T (in terms of francs) = 5*50 - 50 =   200 francs

The example shows that surplus value and transfer value are as different as apples and oranges.  The surplus values (SV) for this situation are 400 francs and 400 yuan, while the transfer value (T) is either 1000 yuan or 200 francs. [See, lines (e), (f) and (l), (m)]


I stated above that surplus value (SV) refers to a value flow between functionally defined entities/classes (labour, capital), whereas transfer value refers to a value flow between territorially defined entities/classes (periphery, center). When we use the notation "c" for "capital" and (letter) "l"  for "labour", then surplus value flows, as follows:

l ---[SV]---> c                   [flow of surplus value, from labour to capital]

When we use the notation "P" for "periphery" and "C" for "center", then transfer value is a flow, as follows:

P ---[T]----> C                 [flow of transfer value, from periphery to center]

The categories "capital/labour" and "center/periphery" overlap. We may want to use the following notations:
lP    labour of periphery
cP   capital of periphery
lC    labour of center
cC   capital of center

Given these notations, it can be stated that transfer value T flows, as follows:

from {lp and cP} ----[T]------> to {lC and cC}

in words, transfer value is an aggregate value which is "sent" (lost) jointly by the inhabitants of territory P (i.e., peripheral labour and peripheral capital) and is "received" (gained) jointly by the inhabitants of territory C (i.e., central labour and central capital). Amin is talking about the fact that the global "labour aristocracy" is benefiting from Third World labour -- in my notation, that is the portion of T value which goes to lC. Amin points out that laborers of the Third World are "superexploited" -- in my notation, that is the idea that lP loses surplus value to cP and, in addition,  loses a portion of T to the center countries.

The above formal notation brings out an interesting aspect, namely, that there are, in the global pyramid, classes with dual status. In my simplified formal scheme, there are two classes with no contradictory status and two classes with contradictory status (dual status), namely:

(1) capital of center (cC) -- receives surplus value SV and a portion of transfer value T -- homogenous status as exploiter
(2) labour of periphery (lP)--loses surplus value SV and a portion of transfer value T -- homogenous status as exploited
(3) capital of periphery (cP) -- receives surplus value SV and loses a portion of transfer value T -- dual status
(4) labour of center (lC) -- loses surplus value SV and receives a portion of transfer value T -- dual status

It follows from these notions that there is an important difference between surplus value and transfer value, as follows. Surplus value flows, in theory, from a single class (labour) to a single class (capital). In contrast, transfer value flows from two classes (lP and cP; labour and capital of the periphery) to two classes (lC and cC; labour and capital of the center). We can thus decompose the transfer value T in terms of either origin or destination, namely:

by origin:
T =  TlP + TcP

by destination:
T = TlC  + TcC

in words, the aggregate transfer value has two components in terms of origin, namely, TlP (the portion of transfer value lost by peripheral labour) and  TcP (the portion of transfer value lost by peripheral capital); and there are two components in terms of destination, namely, TlC  (the portion of transfer value gained by central labour) and TcC (the portion of transfer value gained by central capital).

The combined gains and losses for each of the four classes, factoring in both surplus value and transfer value, can now be stated as follows ( SVC = surplus value of center and SVP = surplus value of periphery):

Table 1
      CLASS                         PAYOFFS (gets+/loses-)

(1) capital of center (cC)         + SVC +  TcC
(2) labour of center (lC)          - SVC  +  TlC
(3) capital of periphery (cP)    +SVP   -   TcP
(4) labour of periphery (lP)      -SVP   -   TlP


The measurement of the various combinations of gains and losses due to SV and T (surplus value, transfer value) is difficult because the T value is based on the assumption that there is an actual (unfair) value and a postulated (fair) value, so that one must think in terms of two valuation systems, the unfair and the fair one. The example which I used above can be further developed to illustrate the various gains and losses, as follows.

Let's express the distribution in terms of the fair values (purchasing power parity values, PPP values) and work with the figures given in the example above. Note that we assumed that  the (fair) PPP value of 1 yuan = 1 franc; the (unfair) exchange rate is 5 yuan = 1 franc. In PPP valuation, we have yuan=franc, therefore.

Let's call the two countries N and S together "the world-system" and use the following notations:
Z = world income = Y south + Y north
Wg = globally aggregated wages
SVg = globally aggregated surplus values

(n) global wages Wg            =   600 +   600 = 1200 PPP francs (yuan)      [from col. c,d]
(o) global surplus value SVg =   400 +   400 =   800 PPP francs (yuan)      [from col. e,f]
(p) world income Z              = 1000 + 1000 =  2000 PPP francs (yuan)     [from col. a,b]

(q) global transfer value T    =                             200 PPP francs (yuan)      [from col. m]

We do not have any firm information about the various components of the T value (namely,  TlP , TcP , TlC , and TcC). Let us make a further assumption, namely, that these components are apportioned in the same way as wages and surplus values (in my example, a 60 : 40 split in both countries and the world). That leads us to the following situation:

Table 2

                             COUNTRY         NET INCOME (after transfer value is subtracted/added)
                                                         Labour            Capital             Both Classes

                             South                     480                 320                  800
                                                          (600-120)       (400-80)          (1000-200)

                             North                     720                 480                1200
                                                           (600+120)      (400+80)        (1000+200)

                             WORLD             1200                  800                2000

Another breakdown of the same figures for each class is:

Table 3

                                                           GETS            GIVES                        GIVE/GET RATIO
                                                                                 TO OTHERS

                            Labour South          +480                -520                         520/480  =     108% (superexploited)
                            Capital South          +320                -  80                            80/320 =       25%
                            Labour North          +720               - 400                         400/720  =      56%
                            Capital North          +480                      0                             0/480  =       0%

                            TOTAL                + 2000              -1000
                                                                                    (800 SVg + 200 T)


The above example uses hypothetical values. I selected a 60 : 40 split for wages and surplus value in order to have some simple and recognizably different numbers to work with. Empirical values are to be determined separately. The same goes for transfer value. That notwithstanding, the example shows in an interesting way that, at the world-system level, surplus value (SVg, global surplus value) is an  inadequate measure of total global exploitation (in the common sense of the word). In the example, the total value which moves upward in the global pyramid is 1000 (composed of 800 for global surplus value and 200 for global transfer value), which is more than global surplus value (800 in the example).


Amin, S. (1976) Unequal Development. New York, USA: Monthly Review Press [translated from the 1973 French original]

Amin, Samir (1980)  "The Class Structure of the Contemporary Imperialist System", Monthly Review (USA), 31 (1980), 8: 9-26

Çakmak, A. (1998), "living peoples", several memoranda to the internet forum wsn, September 1998, at:

Emmanuel, A. (1972) Unequal Exchange: A Study of the Imperialism of Trade. New York, USA: Monthly Review Press [translated from the 1969  French original] [his first article on "unequal exchange" in 1962, French]

Keynes, J.M. (1964) The General Theory of Employment, Interest, And Money. New York, USA: Harcourt Brace Jovanovich [reprint of the 1936 original]

Köhler, Gernot (1998a), "The Structure of Global Money and World Tables of Unequal Exchange." Journal of World-Systems Research 4: 145-168

Köhler, Gernot (1998b), "Unequal Exchange 1965 - 1995: World Trend and World Tables", World-Systems Archive, Working Papers, /archive/papers/kohlertoc.htm

Moseley, Fred (1991) The Falling Rate of Profit in the Postwar United States Economy. New York, USA: St. Martin's Press

OECD (1995)  National Accounts 1960-93. Vol. 1. Main Aggregates.

Raffer, K. (1987) Unequal Exchange and the Evolution of the World System. London, England: MacMillan Press

Shaikh, Anwar (1987)  "humbug production function", The New Palgrave (London, UK: MacMillan), Vol. 2, p. 690-692


Köhler, Gernot (1999)  "A Simulation of Global Exploitation", World-Systems Archive, Working Papers,  /archive/papers/kohlertoc.htm

Köhler, Gernot (1999)  "Global Keynesianism and Beyond", Journal of World-Systems Research,  5:225-241

* end