The Diffusion of Activities

One of the striking features of the recent period has been the diffusion of manufacturing and service activities from the countries of the core to the periphery.  The logic of competitive striving for the export market among the many “labour reserve” economies in the periphery leads to the accumulation of ever-growing reserves and a constraint on domestic absorption.  To believe that the contradictions that emerge will disappear if only the economies with current surpluses appreciate their exchange rates is a fantasy.  Such an appreciation, if it is not to reduce the level of activity in the appreciating economies, must be accompanied by an enlargement of the fiscal deficit in these economies, which means both an abandonment of the tenet of “sound finance” and increased domestic absorption entailing a retreat from the strategy of export-led growth.

1  Introduction

The diffusion of activities from the “metropolitan core” to the under- developed economies of the “periphery” can occur under a variety of circumstances.  The import-substituting development strategy that many economies of the “periphery” had followed in the post-war period had led to a diffusion of activities from the “core” to these economies, but that instance of diffusion constituted what one may call “forced diffusion”, as distinct from the “spontaneous diffusion” that is occurring more recently within a regime of “economic liberalisation”.1  In what follows I shall use the term “diffusion” to refer exclusively to spontaneous diffusion.

One of the striking features of the recent period has been the diffusion, in this sense, of manufacturing activities from the metropolitan capitalist economies, notably the United States (US), to China, and of a range of IT-related services to India.  But these instances of diffusion are only the “leading species of a large genus”.  Such diffusion has been going on for sometime, initially to the east Asian countries, subsequently to the south-east Asian countries, and more recently to China and India.

It may be instructive to look at this process of diffusion in “Lewisian” terms, i e, as an application to the world economy of Arthur Lewis’ (1954) perception of economic development “with unlimited supplies of labour”.  The third world to which the Asian economies belong is the repository of substantial labour reserves in the world economy.  Just as, in the Lewisian perception, labour was drawn from the existing reserves in an underdeveloped economy to the “modern” sector of that economy, likewise in the world economy the diffusion of activities that is taking place is akin to a spatial expansion of the modern sector from the core, by making use of the existing labour reserves, at least those that happen to be located in the Asian economies.

The main difference between the current reality and the Lewisian perception consists in the fact that this “spread” of the modern sector does not actually lead to a using up of, or a reduction in the magnitude of, world labour reserves, whether in absolute terms or even relative to the world population.  In fact, the labour reserves relative to the workforce are not even declining in the economies to which such diffusion is currently taking place.  This is because the rate of technological change in the diffusion-receiving economies, entailing both the shift in frontier technologies at the core which, in turn, are transplanted to such economies, and the “catching up” with the frontier technologies in a number of spheres within such economies, is extremely rapid.  The rate of growth of labour productivity therefore is also extremely rapid.  It severely truncates the labour-absorbing capacity of the growth process which emerges in the diffusion-receiving economies on account of this diffusion, and hence prevents the labour reserves from diminishing.

2  Tendency towards Underconsumption

One implication of this fact, in Lewisian terms, is that the wage rate in the modern sector in the diffusion-receiving economies continues to remain at the “subsistence” level.  Of course, the modern sector in these economies employs skilled labour which is paid much higher wages than what the unskilled labour surrounding it gets.  But what the statement, about the wage rate being tied to the subsistence level despite the occurrence of the diffusion of activities to these economies, means is the following.

Skills are a produced good.  Skilled labour may be seen as the outcome of a production process where unskilled labour is an input.  Its price of production, or what comes to the same thing, its wages, may therefore be seen as a multiple, independent of demand, of the subsistence wage of “simple labour”.  This multiple, and the level of the “subsistence wage”, are assumed to be more or less the same in all diffusion-receiving economies because of competition among them.  And this “subsistence” level remains unchanged as long as the labour reserves remain undiminished.  (The assumption that simple labour used in the modern sector obtains some “subsistence wage”, and more skilled labour, or what may be called “complex labour”, obtains some independently-determined multiple of it, was employed by Marx in his value theory.  Keynes, though he was talking of money wages in a single period, made a similar assumption of skilled labour getting a wage rate that was some independently-determined multiple of the wage rate of unskilled labour, in putting forward his concept of the “wage unit”.)

The second implication is that there is a rise in the share of surplus in the “modern” sector, in the diffusion-receiving economies, and hence by implication in the entire modern sector of the world, i e, taking both the core and the diffusion-receiving economies together.  This is because as labour productivity rises, the wage rate remains stubbornly stuck at the subsistence level.  It follows that an increase in income inequalities is built into the dynamics of the growth process in diffusion-receiving economies.

It may be thought that if the labour reserves of a particular country get used up, because of the growth occurring on account of the diffusion of activities to this economy, then at last its real wage rate will start rising.  But this is not true.  Even if a diffusion-receiving economy exhausts its own labour reserves, its wage rate will still not rise above the subsistence level until labour reserves get exhausted everywhere else, for otherwise a rise in its wages will entail a shift of activities away from this economy to some other economy, i e, the location of diffusion will shift.  Hence, as long as the labour reserves of the world are not exhausted, diffusion-generated growth in the diffusion-receiving economies, must necessarily continue to be ac- companied by an increase in income inequalities.  The third implication, discussed below, follows from this.

This third implication is a tendency towards “underconsumption” in the world economy as a whole.  Since the share of the surplus in the modern sector’s output increases over time, taking the world economy as a whole, this would create an ex ante tendency towards underconsumption.  While this ex ante tendency towards underconsumption originates in the diffusion-receiving economies, since they are the ones where the wages remain at the “subsistence” level even as labour productivity keeps increasing, the effect of this ex ante tendency is to produce an ex post realisation crisis, if at all, not in these economies but at the core itself.  This is because the product wage at the core, relative to labour productivity, is higher than in these economies, which after all is why the diffusion of activities to these economies occurs at all, and this reduces the core’s ability to compete with these economies in the event of any deficiency of aggregate demand.

In other words, when ex ante savings at the base level of capacity utilisation in the world economy exceed ex ante investment, which we assume is the same as ex post investment, the reduction in output that occurs as a result of this deficiency of demand occurs at the core itself rather than in economies which are the recipients of diffusion, and where the tendency towards underconsumption originates in the first place.  The reduction in output at the core reduces the ex post savings at the core, which, given the fact that investment remains unchanged there, creates a current account deficit.  Corresponding to this deficit there would be a current account surplus in the economies to which diffusion is taking place.  Of course, the latter economies may in any case be having a current account surplus.  The effect of the tendency towards underconsumption originating in these economies is therefore to increase their current account surplus and, correspondingly, to increase the current account deficit at the core, in situations where the tendency towards underconsumption expresses itself as a realisation crisis.

On the other hand, in situations where the tendency towards underconsumption does not express itself as a realisation crisis, the reason for its not doing so must lie, apart from specific conjunctural factors, in the deliberate countervailing action on the part of the State at the core, through, for instance, a fiscal deficit.  Given the fact that an ex ante tendency towards underconsumption, if it gets translated to a realisation crisis, is likely to affect the core rather than the diffusion-receiving economy, the State in the latter economy has little reason to adopt any countervailing measures against the prospects of such a crisis.  Whatever has to be done to ward off such a crisis will have to be done by the State at the core.  But in such a case too, there will be an increased current account deficit at the core and a correspondingly increased current account surplus in the diffusion-receiving economy.  Thus, in either case, whether the ex ante tendency towards underconsumption is thwarted or realised, it causes an increase in the current account surplus of the diffusion-receiving economy and a corresponding increase in the current account deficit at the core.

Current Account Movements and ‘Savings Glut’

The view that a tendency towards underconsumption arising in the diffusion-receiving economies manifests itself as an increase in their current account surplus and hence a corresponding increase in the current account deficit at the core, appears in a theoretically distorted form in the “savings glut” argument, put forward by US Fed Chairman Ben Bernanke (2005) among others.  Bernanke not only misstates the argument theoretically, since his concept of a savings glut does not distinguish ex ante from ex post, but argues that this is what has actually been occurring in the world economy.2  In claiming this, he is not just saying that the ex ante tendency towards underconsumption in “emerging market economies”, which he calls a savings glut, has actually manifested itself as an increased current account deficit at the core.  Saying this after all is quite unexceptionable.  But he is saying something more, namely that underlying the observed increase in the current account deficit at the core is nothing else but such an ex ante tendency towards underconsumption.

This last-stated position, though theoretically tenable as we have seen earlier, is empirically unsustainable.  This is because an ex ante tendency towards underconsumption can never give rise to an increase in the level of activity at the core.  If the tendency gets actually realised, then there will be a reduction in the level of activity at the core.  If it does not get realised, because of countervailing action on the part of the State at the core, in the form, say, of a fiscal deficit, then the level of activity at the core would remain exactly what it otherwise would have been in the absence of such a tendency.  In neither case would there be an increase in the level of activity at the core.  Hence, if we find an increase in the current account deficit that is accompanied by a rise in the level of activity at the core, by boom conditions that is, then this increase cannot be attributed exclusively to the tendency towards underconsumption.  Its proximate explanation must lie in the fact of the boom itself, which, as is only usual, sucks in goods and services from the rest of the world through an enlarged current deficit.  Since the period about which Bernanke is writing is one characterised by a really pronounced boom in the US economy (and by extension in the core as a whole), whose end has precipitated the current recession in the world economy, to attribute the increase in the current deficit during this boom to the ex ante tendency towards underconsumption originating in the emerging market economies is clearly untenable.

3  Displacement

This underconsumption element has to be distinguished from another element which, in the absence of a better word, I shall call the “displacement” element.  To understand the latter, let us abstract from the underconsumption element altogether by assuming that there is no technical progress at all and that the level of labour productivity everywhere is given.  (To avoid any possible confusion we make the extreme assumption that labour reserves exist in the form of pure unemployment, i e, there are no other activities in the diffusion-receiving economies apart from the modern sector activities.  This, of course, raises the question: how do the unemployed manage to survive in such an economy?  But since our assumption is only for expositional purposes, we need not go into this question here.)  In such a world the very fact of “diffusion of activities” from the core would bring about a secular reduction in the level of activity at the core and hence a secular increase in the ratio of current deficit at the core to its GDP in the following manner.

Because of the lower ratio of product wage to labour productivity in the diffusion-receiving economies, they would always be more competitive and hence be at full capacity (except for transitory effects of disturbances).  During a boom in the world economy, the demand for their goods will increase, which ceteris paribus would suck in more goods from their economies.  One possible way this could occur is through a “profit inflation”, i e, a rise in the price level relative to the money wage rate.  But if the real wage rate was already at the subsistence level, then their governments will wish to avoid a further push below this level.  Besides, a labour-reserve economy which prevents such a profit inflation, by reducing domestic absorption via a cutback, say, in the fiscal deficit, will obtain a larger share of the export market than another “labour-reserve” economy that does allow such a profit inflation, since the latter’s real effective exchange rate would have appreciated relative to that of the former.  There would be a tendency among all labour-reserve economies in a period of world boom therefore to reduce domestic absorption and make more goods and services available to the core economies that are sucking in such goods and services, while keeping the domestic real wage rate at the subsistence level.

But the reverse, namely an increase in domestic absorption in such economies, does not happen during a period of world slump.  Since the impact of reduced aggregate demand is felt on the output at the core, the labour reserve economies are under no compulsion to enlarge their fiscal deficits.  Besides, any such economy that does increase domestic absorption is likely to lose export markets to another such economy that does not expand domestic absorption.  Labour-reserve economies therefore do not expand domestic absorption in periods of crisis.  They keep real wages at the subsistence level, so that workers’ consumption does not become a source of increased domestic absorption; and they do not enlarge the fiscal deficit as there is no need for it.  (They may do so only transitorily as China has done in the context of the present world recession, but even in the case of China the very recent increase in its exports despite the persistence of world recession suggests that as its fiscal deficit gets reduced, the current account surplus will increase to take its place, confirming the hypothesis about the resilience of such economies against reductions in world aggregate demand.)

There is, therefore, a “ratchet effect” at work whereby the current surpluses of the diffusion-receiving economies keep increasing secularly relative to their GDP, with a corresponding secular curtailment in the proportion of domestic absorption.  This also means ceteris paribus a secular rise in the ratio of current account deficit to GDP at the core.

The observed effect of this factor is exactly the same as that of underconsumption and indeed the two effects are additive.  Any observed secular increase in the current deficit relative to GDP at the core therefore can be attributed to the combined effects of the tendency towards displacement and the tendency towards underconsumption.

The tendency towards the displacement of core production can also be subsumed under the savings glut argument, in the sense that since increased domestic absorption in the diffusion-receiving economies would prevent a secularly growing current account surplus, any actual secular growth in their current account surplus can ipso facto be attributed to a lack of adequate domestic absorption, and hence to a so-called savings glut.  But apart from the fact that the reasons for the inadequate domestic absorption in diffusion-receiving economies are never examined (more on this later), no distinction is drawn in the savings glut hypothesis between the tendency towards underconsumption and the tendency towards displacement of core production.

4  Competition among Diffusion-Receiving Economies

The real question that arises is: why do the diffusion-receiving economies not increase domestic absorption?  Their doing so, instead of simply piling up foreign exchange reserves, will be beneficial for the world economy in at least three distinct ways.  First, the living standards of their domestic population will improve, which after all is the objective of development.  Even if the real wages remain at the subsistence level, an increased domestic absorption in the form, say, of a fiscal deficit that finances transfers to the workers and peasants, or is used for providing collective consumption goods, or is used for raising the social wage, will improve the condition of the domestic population.  Second, an increase in their domestic absorption will certainly get rid of the world imbalances, since it would reduce their current account surpluses and hence the current deficits of the core economies, especially the US.  And third, it is likely to raise world employment, especially in the US, by raising, ceteris paribus, the level of world aggregate demand.  This presumes, of course, that the world economy is demand-constrained, which it almost invariably is, and especially at the present juncture.  But even if perchance it was not, and no increase in world employment occurred as a consequence of increased domestic absorption by the diffusion-receiving economies, even then a substitution of larger domestic absorption by them for a larger US fiscal deficit (or a larger US private sector deficit) as the stimulus for world demand, is preferable both on the grounds of greater stability of the world economic system and on the grounds of greater equity in consumption levels across the economies of the world.

The answer to this puzzle, about why countries appear to prefer current account surpluses to larger consumption by the domestic population, is usually given in terms of neo-mercantilist attitudes characterising such countries.  Such attitudes, of course, do exist and do constitute some part of the explanation.  And they are even understandable up to a point in societies that had been subject, directly or indirectly, to colonial exploitation.  The desire for national economic power as a counterweight to metropolitan hegemony is understandable, even though we know that “defensive” nationalism can quickly give way to big power chauvinism.  Likewise, the desire for national economic security of the sort that the holding of large foreign exchange reserves brings is understandable, especially in the wake of the east Asian crisis, even though we may feel that the quest for security is being pushed too far.

But overriding these “national” considerations of power and security, there is a more objective factor that explains the reluctance to increase domestic absorption.  And this consists in the fact that any increase in domestic absorption in any one of the diffusion-receiving economies will entail a loss of its export market to other diffusion-receiving economies that do not expand domestic absorption.  In other words, it is competition between the actual and potential diffusion-receiving economies (or what I have called earlier the “labour-reserve” economies) that ensures not only that their wage rates remain at the subsistence level but also that their domestic absorption is kept down (and declines over time through the “ratchet effect” mentioned earlier).

Diffusion occurs within an international regime of neoliberalism.  This ties the growth rate of economies to their capacity for export, their ability to capture a share of the world market and to keep increasing this share as much as possible.  A reduction in this share is quite capable of setting up a vicious cycle leading to a cumulative downward slide.  Hence each such economy, willy-nilly, is coerced into stinting on domestic absorption, lest the growth strategy being successfully pursued gets jeopardised.  The kind of Darwinian struggle that Marx had detected as characterising the different capitals within a national market also characterises the different national economies in the global market, in a regime of what the World Bank benignly christens as “outward orientation”.

In short, it is the objective coercion of the export-led growth strategy within a neoliberal paradigm, rather than the subjective desire for power or prestige, that largely explains the reluctance to expand domestic absorption.  What Bernanke considers a vice on the part of the EMEs, or others, even progressive American economists like Paul Krugman, consider the manifestation of a “beggar-my-neighbour” policy on the part of EMEs, is, in fact, inherent to the logic of a neoliberal regime.

5  Implications

The foregoing has two extremely significant implications.  First, under neoliberalism, poverty and unemployment will never get eliminated.  All that will happen is a perpetuation of dualism.  It may be argued that with rising economic surplus relative to GDP the investment ratio will rise in the diffusion-receiving economies and this would keep raising the growth rate until labour reserves dwindle and the wage rate starts rising, as in the original Lewis model.  But since the rate of growth of labour productivity rises with the output growth rate, the difference between the two, which is the rate of growth of employment, will rise, if at all, at an extremely minuscule rate as the growth rate rises.  If its magnitude still remains less than or equal to the rate of growth of the labour force, then the relative size of the reserve army will never fall, the wage rate will remain tied to the subsistence level, and the poverty ratio will never decline.  Even if the rate of growth of employment marginally noses ahead of the rate of growth of the labour force, the decline in the relative size of the reserve army will occur at such a tardy pace that for all practical purposes the state of poverty and unemployment will continue.

This conclusion is reinforced by the fact that the possibility of state intervention to ameliorate poverty also gets limited by the constraints on domestic absorption mentioned earlier.  The logic of competitive striving for the export market among the many labour reserve economies comes in the way of any effort to raise domestic absorption through state policy.  Indeed, if anything, domestic absorption is curtailed to push out more exports which only results in the accumulation of even larger reserves.

It follows from the above that the exhaustion of labour reserves constitutes a necessary condition for the eradication of poverty, which requires a significant difference, whose absolute magnitude must exceed the rate of growth of labour force, between the rates of growth of output and of labour productivity.  This can be ensured only with some restrictions on the pace of technological change (to which the structural change in the economy is a contributory factor), and such restriction can never be imposed in a neoliberal economy.  Neoliberalism, therefore, is incompatible with the removal, or even an amelioration, of poverty, of dualism, and of unemployment and underemployment, all of which constitute characteristic features of the state of underdevelopment.

This argument against economic liberalisation is completely different from the old argument that used to be advanced in the 1950s and 1960s.  The argument then was that the pursuit of economic liberalisation, notably trade liberalisation, did not lead to any significant “spontaneous” diffusion of activities from the core to the third world.  Hence the strategy of export-led growth had limited prospects; and trade liberalisation only succeeded in getting the country mired in stagnation, as in the colonial times.  This proposition advanced by many, from P C Mahalanobis (1963) to Surendra Patel (1959) in India, was critiqued by a host of authors like Little, Scitovsky and Scott (1970); Bhagwati and Desai (1970); Manmohan Singh (1964); and Benjamin Cohen (1964).  The present argument, by contrast, presumes that diffusion of activities does take place, and that too to a degree where, for all practical purposes, the problem of effective demand disappears in the labour reserve economies (exactly as neoclassical economists assume); and that there is even a rising current account surplus relative to GDP (so that there is no question of any foreign exchange constraint).  But even under these conditions, poverty and underdevelopment continue to hold sway in the labour reserve economies pursuing the trajectory of export-led growth within a neoliberal regime.

The second major implication is that in a world economy characterised by such diffusion, “imbalances” are bound to grow.  In short, the existence and aggravation of imbalances is endemic to the situation, since both the tendency towards the displacement of core activities, and the tendency towards underconsumption, are its inherent features.  In short, a neoliberal economic policy regime exacerbates the contradictions in the world economy.

To believe that these contradictions will disappear if only the economies with current surpluses appreciate their exchange rates is a fantasy.  Such an appreciation, if it is not to reduce the level of activity in the appreciating economies, must be accompanied by an enlargement of the fiscal deficit in these economies, which means both an abandonment of the tenet of “sound finance” so dear to finance capital, and increased domestic absorption entailing a retreat from the strategy of export-led growth.  It follows that the resolution of the contradictions plaguing the world economy, lies, whichever way we look at it, in the abandonment of the logic of neoliberalism.

But neoliberalism is above all the preferred regime of international finance capital which wants all restrictions on the movement of goods and services, and of finance above all, removed, so that it can move around the world with complete freedom and ease.  The fact that this preferred regime of finance capital exacerbates the contradictions of the world economy, only underscores yet again the point that the hegemony of international finance capital constitutes the biggest barrier to the economic progress of mankind at the present juncture.



1  For a detailed discussion of the different kinds of diffusion see Patnaik (1997).

2  For a theoretical critique of the Bernanke argument see Patnaik (2010).



Bernanke, Ben (2005): “The Global Savings Glut and the US Current Account Deficit”, Sandridge Lecture delivered at the Virginia Association of Economics, Richmond, Virginia, 10 March 2005; available at

Bhagwati, J N and P Desai (1970): India: Planning for Industrialisation (Oxford: OUP).

Cohen, B I (1964): “The Stagnation of Indian Exports, 1951-61”, Quarterly Journal of Economics, November.

Lewis, W Arthur (1954): “Economic Development with Unlimited Supplies of Labour”, The Manchester School, May.

Little, I M D, T Scitovsky, and M Fg Scott (1970): Industry and Trade in Some Developing Countries (Oxford: Clarendon Press).

Mahalanobis, P C (1963): The Approach of Operational Research for Planning in India, Statistical Publishing Society, Calcutta.

Patel, S J (1959): “Export Prospects and Economic Growth: India”, The Economic Journal, Vol LXIX.

Patnaik, P (1997): Accumulation and Stability under Capitalism (Oxford: Clarendon Press).

— (2010): “The Theory of the Global Savings Glut”,

Singh, Manmohan (1964): India’s Export Trends and the Prospects for Self-Sustained Growth (Oxford: Clarendon Press).

Prabhat Patnaik is at the Centre for Economic Studies and Planning, Jawaharlal Nehru University, New Delhi.  This article is the text of a paper presented at an international conference organized by IDEAs on “Recovery or Bubble?  The Global Economy Today” in New Delhi on 29-30 January 2010.  It was published in Economic & Political Weekly on 6 March 2010; it is reproduced here for non-profit educational purposes.

| Print