David Harvey’s theory of uneven geographical development: A Marxist critique

The geographical concentration thesis

Capitalism requires the purchase and sale of commodities. Commodities (including labour power) need to move in geographical space. ‘Time is money for capital’, and ‘Traversing space takes both time and money’, so ‘Economy of time and money is a key to profitability’ (Harvey 2014: 147). One way to reduce time and cost of movement is spatial concentration: ‘capitalists locate so as to minimize their costs of procuring means of production (including raw materials) and labour supplies and of getting to the market’ (p. 149). Enterprises (e.g. car parts, tyre factories and car assembly plants) cluster together, so they ‘can share facilities, access to labour skills, information and infrastructures’ (p. 149). And in search of employment, labour is also attracted to where enterprises cluster. All this ensures labour supplies to a cluster of firms; one firm can employ people trained by others. In other words, agglomeration economies (i.e. cost savings owing to geographical concentration) arise because many companies (and a large pool of labour) cluster together in space. ²

Urban agglomerations are spatial environments that capital specifically constructs in its own interests. They are characterized by a structured coherence, ³ produced by flows (i.e. inter-firm economic transactions) within them. Cultural processes (lifestyles, work habits, religious and political loyalties, etc.) are also involved in reinforcing these flows. Dominant classes and class alliances can lend some identity to particular agglomerations or regions. ⁴ Specialized economic regions (Leeds for wool, Manchester for cotton and Sheffield for steel) form a mosaic of UGD of capitalism. Within this mosaic ‘some regions tend to become richer while poor regions get poorer’ (p. 149). This happens, Harvey says, because of the cumulative causation mechanisms that Myrdal (1957) talked about. The regions that are already advanced draw new activity because they have bigger markets, and better physical and social infrastructures, which is supported by their better local tax base. New transportation lines may connect to these regions. All this makes the advanced regions even more attractive to capital and labour. Other regions do not receive much investment. They are allowed to decay. ‘The result is uneven regional concentrations of wealth, power and influence’ (Harvey 2014: 150).

The geographical dispersal thesis

Geographical concentration coexists with its opposite tendency: spatial dispersal. This occurs for two reasons. One is that unlike in the Myrdal’s thesis, there are limits to the continuous concentration, the limits which lie within the structure of market relations. A given pattern of regional concentration is not permanent. Why?

In a city/region where there is a lot of capital investment, costs of doing business gradually rise. Because of increased demand on land, the rentier class increases rent. The level of pollution and traffic-congestion rises. Local taxes may increase as well. ‘Rising local costs of living lead to wage demands’ (Harvey 2014: 150). And regional concentration of workers allows them to better fight for higher wages. All these factors limit profit opportunities in what were once developed regions, relative to the less developed regions. Capital from the advanced regions looks elsewhere. The less developed regions, with their low wages and high unemployment, attract capital from the more developed ones. In other words, locational change, just as technical change, is deployed to reduce costs, ⁵ producing the opposite of what can be called the ‘Myrdal effect’ (cumulative causation). ‘This is particularly so when new technological and production mixes are emerging and labour struggles are acute’ (Harvey 2014: 150). One set of places such as Bavaria, Tuscany and Silicon Valley displace another, more-established, set of places such as the Ruhr, Turin and Detroit, respectively. Internationally, newly industrializing States, or at least some regions within them, have displaced the more-established states (or certain regions within them).

Apart from rising costs in the developed regions, there is a second reason – an enabling condition – for spatial dispersal. Spatial dispersal is made possible by innovation in transportation and communication technology. Given that time is money, ‘A premium is … placed on innovations – technical, organizational and logistical – that reduce the costs and time of spatial movement’ (Harvey 2014: 147). Of course, the actual impact of new technologies depends on the type of production and the form that the commodity/capital takes. Money and images (e.g. telecast of sports events) as forms of capital or commodity can move faster than physical commodities (e.g. metal and food). But as a general rule, continuous investment in transport and communication technologies facilitates dispersal across increasingly larger spaces. ⁶ At a concrete level, there are several decentralization strategies. For example, divisions of labour within a firm can be decentralized to different locations such that parts of a complex commodity (e.g. car) can be made in different places. In addition, offshoring becomes possible: cars or toys can be made outside a country and imported back in. Harvey (2014) argues that

The near elimination of transport costs and times as a factor in location decisions permits capital to explore differential profit opportunities in widely disparate places … Regional specializations and divisions of labour become even more marked because small differences in costs (such as local taxes) translate into higher profits for capital. (p. 148, italics added)

Places are different not necessarily because one is more accessible than another (all places, increasingly, are more or less accessible) but because of their other properties which make some places more profitable than others. New geographical patterns of production arise because of sharpened spatial competition facilitated by cheaper and more efficient transport and communication technologies. For example, as costs of movements drop, steel can be produced in Korea and be imported, so steel industries in Sheffield close down. Detroit’s industrial base was destroyed due to competition from foreign countries and from low-wage regimes of Tennessee and Alabama. Cheap farm imports have had similar impacts on the United Kingdom. De-industrialization – the ‘other’ of geographical expansion – is going on.

What all this suggests is that capitalism is not only about competition, but that this competition happens in a geographical space. Because capital can go to a variety of places, ‘the monopolistic element in competition [spatial monopoly] is reduced’ (p. 148). Capital in a place cannot hide behind the veil of spatial monopolies (distinctive locational advantages) to the extent it was able to do earlier.

The spatial fix thesis

That does not mean that space relations are not important to capital. To annihilate space, capital has to produce space. However, the production of space – including new centres of accumulation – is a contradictory process. It signifies the contradiction between fixity and motion. In what is one of his most memorable statements (and there are many), Harvey says, ‘In order for capital to circulate freely in space and time, physical infrastructures and built environments must be created that are fixed in space’ (p. 155). This means that a part of the total capital has to be fixed for the remainder of it to be mobile. An important implication of this for UGD is that once a certain amount of capital is fixed in a place (in the form of factories, houses, airports, roads, etc.), it cannot move elsewhere (without risking devaluation), and thus, opportunities for capitalist investment in places where it is relatively absent become relatively limited.

Space is important not just for the spatial circulation of capital. It is also important from the standpoint of crisis: it becomes a spatial or spatio-temporal fix. Capitalism inevitably produces surpluses: of capital and of labour, relative to opportunities for their profitable employment, and this happens because of the imperative to innovate and accumulate. Such a state of overproduction of capital is called the ‘over-accumulation of capital’. The production of new landscapes, including physical infrastructures which are fixed in and on the land for a long period of time, absorbs surplus capital slowly. So, it is a fix (‘a solution’) to capitalism’s over-accumulation problem (pp. 151–152). ‘Urbanization and regional development become autonomous spheres of capitalist activity, requiring large investments (usually debt-financed) that take many years to mature’ (p. 151, italics added). Sometimes, these projects are set up in underdeveloped locations.

However, the spatial fix is a contradictory process. ‘Capital creates a geographical landscape that meets its needs at one point in time to have to destroy it at a latter point in time to facilitate capital’s further expansion’ (p. 155). Although the production of new dynamic regions, new resource complexes and new territorial divisions of labour can provide new opportunities for profits and can absorb surplus, such a process can threaten ‘the values already fixed in place elsewhere’ (p. 152) as in older cities such as Detroit. Furthermore, when capital moves out from a place to be invested in another place, it leaves behind a trail of devastation and devaluation, producing regional crises of employment and production (p. 151). But if capital stays put, it generates capital surpluses which cannot be deployed profitably. ⁷ There is another set of contradictions. When surplus capital from one place is sent to another place and put to profitable use, the new place absorbs capital and addresses the over-accumulation problem of the originating place; this is especially the case if the capital invested is in bulky projects (ports, railroads, etc.). However, over time, the new place will generate its own surplus (as China does now), which has to find new opportunities, including perhaps in the originating place, which had supplied capital to the new place. This can result in geopolitical tensions between old and new places, which compete to stave off crisis (and to have access to cheaper raw materials). But if the new place fails to absorb the surplus, then there is a problem for the place of origin of capital.

The UGD-as-ideological thesis

Thus, spatial mobility of capital keeps the whole system stable, but parts of the system can be in difficulties (deindustrialization here or some devaluation there; p. 153). To the extent that this is true, an aspect of it can be pointed to in order to stave off criticism against the system as a whole. While today this area and tomorrow that area is in trouble, it is also the case that there is always a successful neighbourhood, region or country where things are going right. Because some regions are always doing well, relative to others, people feel that all is well with the system, and all will be well if all regions do what the successful regions do (p. 154). And, when specific places have problems, it is said, they have to get it right (maybe hire an imaginative city planner or a Richard Florida). Capital itself gets off scot-free. Or, one blames the problem on greedy unions or political mismanagement (e.g. corruption). The rural and urban landscapes hide the fact that it is capital that is behind them, and hide the power of capital to create UGD. ‘The landscape of capitalism [both mansions and deprived areas] exists as a diversionary image of another world closer to some transcendental sense of human longing and desire’ (p. 160).

The UGD-and-state connection thesis

UGD must be seen in terms of the dialectics between the territorial logic and the capitalistic logic, Harvey (2003, 2014) says. The relative spatial fixity and distinctive logic of territorial power as manifested in the state interacts with the fluid dynamics of capital accumulation in space and time. The state stands for fixity, and capital, for motion (p. 155). The state’s practices are not monolithic, but it generally responds to business interests. The state also uses the forces of capital to ‘support its own powers of governmentality over potentially restive populations’, all the while enhancing its wealth, power and standing within a highly competitive interstate system. The state is involved in the co-production of uneven development in many ways. For one thing: ‘the state carries out urban and regional planning practices whose aim is to contain the otherwise chaotic consequences of unregulated market development’ (p. 157). The state also does this through its economic policies: for example, supplanting local banking system by a national banking system, allowing free flow of money across the national space and thus altering regional dynamics, nationalizing banks, and channelling resources to certain areas. The state is also involved in making geographical agglomeration possible: the necessity to produce and maintain collective goods (e.g. social and physical infrastructures) requires some state intervention. State-funded infrastructural projects are set in motion during crises to mop up surplus. In sum, the state creates collective conditions for capital. ⁸ And these can vary between places, regions and countries, including for reasons that are to do with differences in the tax base of the state and state-provided credit or state-backed credit needed for bulky investments. In recent times, mobile (financial) capital puts pressure on the state to create good business climate within countries/regions, including via state-enforced wage repression (and sub-national withdrawal of benefits). As mentioned earlier, small differences in costs between places can ‘translate into higher profits for capital’ (p. 148) and thus determine where capital will go. It is the state that can create these small differences, including through its policies of local taxes. ⁹ In addition, given the threat of devaluation, caused by the tendency towards over-accumulation, there is an intense competition between countries (and sub-national regions): competition for markets, or for scarce raw materials, and how this competition is resolved depends on coercive state power. Popular pressure plays a role in UGD, for ‘Capital has to be somewhat sensitive to the wants and needs of the populations it exploits’. So,

The geographical landscape of capitalism is … shaped by a multitude of interests as individuals and groups seek to define spaces and places for themselves against the background of macroeconomic processes of uneven geographical development that the rules of capitalist accumulation and state power jointly effect.

And even if capital were not sensitive to people’s wants and needs, ‘social and class struggles would surely force it to the table to compromise with critics’ (p. 159).

The political thesis

Capital is the main reason for uneven development (and many other problems). It must be resisted. Harvey says, it must be resisted by social movements in which workers will play a role but not necessarily a dominant role. Also, ‘It is first vital to recognize that capital is always a moving target for opposition’ (p. 162). Any anti-capitalist movement must cope with this fact. Anti-capitalist resistance is likely to be as ‘chaotic, volatile and geographically specific’ (p. 163) as molecular processes of capital accumulation are. Furthermore, ‘Anti-capitalist movements must abandon all thoughts of regional equality and convergence around some theory of socialist harmony. These are recipes for an unacceptable and unachievable global monotony’ (p. 162). So, anti-capitalist (social) movements have to liberate and coordinate their own dynamics of UGD, the production of emancipatory spaces of difference, in order to reinvent and explore creative regional alternatives to capital (p. 162). ‘Different social movements and resistances are emerging … from Stockholm and Cairo to Sao Paulo, Shanghai, Paris and London. These constitute a mosaic of different but loosely interconnected seedbeds for transformations of capitalism towards an anti-capitalist future’ (pp. 162–163), a mosaic which, so to speak, mimics the mosaic of regional economies capital creates.

Inadequate treatment of class relations

Harvey tends to think of capital more in terms of things, assets, and built environments and less in terms of class relations. ¹⁵ An implication of this is that he generally equates capitalism to the capitalism where productive forces are advanced and technological change is constantly happening. In other words, he conflates capitalism-in-general with capitalism at a specific stage of development (advanced capitalism). ¹⁶ This line of thinking implicitly endorses linearity: he (like Robert Brenner) assumes that the presence of the wage-labour relation itself (the differentia specificia of capitalism-in-general) will cause capitalism at a higher level of economic development, based on systematically occurring technological change, albeit one that is uneven. One must question this implied underlying linearity and the associated tendency to generalize on the basis of the experience of a few capitalist countries which currently have an advanced level of productive forces, rather than thinking about capitalism-in-general and thinking about capitalism internationally. The capitalist relation in itself in a region may not cause what Brenner (1986: 24) calls a systematic and continuous tendency towards a rise in labour productivity through technological change. This is because entrepreneurs can simply hire workers for long hours and with increased work intensity (Marx’s formal subsumption), often paying wages that do not even cover the cost of reproduction (i.e. payment of wages below the value of labour power) (Das 2012). In Capital volume 1, Marx says that capitalist class relation is historically differentiated into at least two forms/stages: the formal subsumption of labour under capital and the real subsumption of labour under capital. In the initial stage of the evolution of capitalist class relation, wage labour is only formally subsumed under capital: ‘at first, the subjection of labour to capital [is] … only a formal result of the fact, that the worker, instead of working for himself, works for and consequently under the capitalist’ (Marx 1977: 448). As Marx elaborates in the Appendix to Capital volume 1 (Marx 1977: 1025–1029), there are four main features of formal subsumption. First, surplus is pumped out via economic, and not extra-economic, coercion: that is, labourers, who do not own means of production, are generally free to choose their employers in a labour market. Second, the means of production and consumption are bought in the market (both by capital and labour). They confront the worker as capital. Third, no more labour time is used in production than is socially necessary: there is competition to reduce the cost of production of commodities for sale, suggesting that the law of value is operating. Fourth, an economic relation of supremacy exists at the point of production, as the worker is supervised by the capitalist (or their manager). One may add the payment of low wages (e.g. lower than the value of labour power) as an aspect of formal subsumption as well. Real subsumption has all these characteristics of formal subsumption. In addition, it involves the reduction of necessary labour time through the use of technology, reflected in an increase in the ratio of constant capital to variable capital; labour productivity increases as a result (Marx 1977: 762). Historically, when workers in an area resist formal subsumption, entrepreneurs tend to adopt productivity-raising technological change, which allows them to appropriate surplus value in its relative form. ¹⁷ Such resistance varies in space, producing spatial unevenness in whether and how the transition to real subsumption of labour associated with a developed capitalism based on systematic technological change happens. ¹⁸ Such spatial unevenness is a fundamental cause of uneven development (Das 2012). And in this approach, place-specificity of social processes matters.

UGD cannot be seen merely in terms of where capital investment is/occurs/moves to. What capital accumulation/investment does to labour must be taken into account in a much more rigorous way than Harvey’s approach allows. And especially whether accumulation is based on and leads to real subsumption of labour must be seen as a crucial element in the theory of UGD. Harvey’s inadequate approach to the distinction between formal and real subsumption is symptomatic of his inadequate treatment of the class character of capitalism itself (Das 2017). Harvey (2006 [1982]) has got the class character of capitalism wrong when he says that

Monetary relations have penetrated into every nook and cranny of the world and into almost every aspect of social, even private life. This formal subordination of human activity [or labour], exercised through the market, has been increasingly complemented by that real subordination which requires the conversion of labour into the commodity labour power through primitive accumulation. (p. 373)

One can see how Harvey’s conceptualization of formal and real subordination/subsumption is different from Marx’s which is briefly presented above. Little does Harvey realize that according to Marx, capital, in its most general form, is based on the formal subsumption of nominally free wage labour which is purchased by capital in the market, and that the capitalism of Harvey’s theory of UGD is merely the capitalism in its more economically developed form.

UGD must be treated at the levels of both capitalism in general and capitalism at a specific stage of development (and also at the level of class, a point clarified below). ¹⁹ Capitalism is fundamentally a class relationship which progresses through primitive accumulation and subsequently through formal and real subsumption of labour (Das 2017). What is uneven is not just the development of the physical properties of capital (e.g. built environments, etc.) but also the non-linear transition from one stage to another. Harvey’s under-emphasis on class relation between capital and labour is expressed in another form, i.e. in the way in which Harvey thinks about spatial unevenness in relation to economic crisis.

Inadequate treatment of the relation between crisis and UGD

Relevant in this context is Harvey’s deep scepticism towards Marx’s theory of falling rate of profit. This problem stems from the fact that in Harvey’s work, the most important relation, the class relation, that is, the relation of surplus appropriation from labour by capital, is not that significant and that merely competitive relations in an anarchic market can explain what we want to explain, including UGD. For Harvey, UGD is about cutting costs (including especially costs of geographical movement), which is associated with an increase in the ratio of immobile to mobile capital. But what about the ratio of constant capital (including fixed/immobilize capital) to variable capital? He has been, more or less, rejecting (the significance of) Marx’s theory of falling rate of profit; his views on this can occupy the space of an entire article. Let me first provide some brief evidence here.

In 1982, Harvey (2006 [1982]) had said, ‘Marx … is lured into an erroneous specification of [the theory]’, that ‘Marx diverts from the logic of his own argument to such a degree that what should have been a tangential proposition appears fundamental while the fundamental proposition gets interred in a mass of tangential argument’ (p. 181; italics added). He doubted the ‘epistemological status of the law’ of falling rate of profit (p. 181) on several grounds. One is that because Marx specifies the law in terms of value and not prices, ‘the law cannot be used to describe the “surface appearance” of capitalist dynamics; one also needs to be wary of “treating the law as a direct historical and empirical proposition”’ (pp. 180–181). This is partly because one cannot assemble data on corporate profits in a given country for a given period and prove and disprove the law for that period; Marx’s argument is also not ‘particularly well-honed or rigorously defined’ (p. 181). Two decades after he wrote these lines, here is what he says in the Introduction to a new edition of Limits: Marx ‘attached so many caveats, conditionalities and compensating circumstances’ to the theory of falling profit that ‘it is difficult to sustain it as a general theory of crisis’ and that like the profit squeeze and under-consumption, the falling profit rate is a surface manifestation of ‘something else’ (p. xxxiii). That something else is the tendency towards over-accumulation (surplus capital) – the fact that capitalists produce a lot more surplus value than what can be profitably absorbed, a process that is caused by the contradiction between productive forces and relations (Harvey 2006 [1982]). He thus explains contradiction of capitalism using an approach that applies to all class societies rather than deploying a broader, a more totalizing, approach which combines mechanisms that apply to all class societies with those that are associated with capitalism as such, and with advanced capitalism. ²⁰ He knows that what is at stake here is profitability. But ‘this has nothing directly to do with the supposed law of falling profits’ (Harvey 2006 [1982], italics added). Four years after he wrote his 2006 statement in Limits, this is what he says in Enigma (Harvey 2010): ‘It is hard to make Marx’s theory of the falling rate of profit work when innovation is as much capital or means of production saving (through, for example, more efficient energy use) as it is labour saving’ (p. 94).

For him, the rate of profit falling has little or nothing to do with rising organic composition of capital. One reason for this is that – and here he echoes a statement he made in 2006 quoted above – Marx’s list of countervailing forces is so ‘long that it renders the neat explanation for a solid “law” of falling profits as a mechanical response to labour-saving technological innovation more than a little moot’ (p. 94). Besides, our world is different from Marx’s:

the astonishing penchant for creating wholly new products lines … has placed the development of consumerism and a rising effective demand at the center of the sustainability of contemporary capitalism in ways that Marx, for one, would have found hard to recognise. (Harvey 2010: 95)

He adds, ‘Lack of aggregate effective demand in the market … creates a serious barrier to the continuity of capital accumulation. It leads to falling profits’ (Harvey 2014: 81; also pp. 191–192).

Harvey (2010) thinks that

(There are) multiple ways in which crisis can form in different historical and geographical situations … [W]hile the epicenter lies in the technologies and organisational forms of the credit system and the state-finance nexus, the underlying problem is excessive capitalist empowerment vis a vis labour and consequent wage repression, leading to problems of effective demand papered over by a credit-fuelled consumerism of excess in one part of the world and a too rapid expansion of production in new product lines in another. (pp. 118–119)

The disequilibrium happens unevenly; the over-accumulation crisis is localized. This is in part because of the geography of fixed capital. If for neo-classicals, market is the solution to ills, for Harvey market causes problems.

Harvey’s focus is clearly on the inadequate effective demand and over-accumulation. But must over-accumulation not be explained in terms of what is happening to profit opportunities: why are profit opportunities limited which makes accumulation over-accumulation? Crisis – the breakdown in the circuit of capital – is about the decrease in the ratio of total surplus value to total capital, in a society at large, or over a large part of global society over a long period of time, although crisis is not an unmediated product of such decrease. And this decrease is caused by the fact which Harvey knows but does not appreciate, that the only source of fresh surplus value – living labour – is increasingly replaced by machines, dead labour. A drop in the profit rate at some point in time will lead to an insufficient amount of capital. So, a decline in profit rate produces, from time to time but not necessarily always (thanks to countervailing forces), what appears to be a surplus of capital relative to profit opportunities. ²¹ The rate of profit sets limit to how much money is available at any time for investment. ²² The drop in the rate of profit is compounded by another tendency. With increases in the social productivity of labour, the minimum amount of capital required to set in motion a given quantity of labour power in a sector rises. This further limits profit opportunities. There are counter-tendencies, but they can work in the short term but cannot offset the main tendency for ever (Carchedi 2011: 94–95). ²³ The long list of counter-tendencies itself does not invalidate the existence of the central tendency. For Marx, it is the decrease in profitability that produces, although not in an unmediated fashion, a crisis of unemployment, a glut in the market, financial panics and so on. Machines remain unsold or labourers remain unemployed because firms, facing a drop in profitability, do not invest. Crises raise the rate of profit when firms are compelled to sell off means of production at below value, and when firms force wages down, increasing the rate of exploitation. What the (place-specific) devaluation of capital through deflation of commodity prices and by other means does is to increase the rate of profit in particular places.

Harvey must reconsider his scepticism of – and empiricist view towards – the falling rate of profit theory in the light of the massive amount of theoretical and empirical work produced in recent years, and especially since he had a chance to write about this in the early 1980s. In his work since Limits, there is (almost) no discussion of the recent scholarship on that theory (e.g. Beitel 2009; Carchedi 1991, 2011; Freeman & Carchedi 1996; Harman 2007; Kliman 2007, 2010, 2012; Laibman 2004; Mattick 2008; Shaikh 2010; Sinha 2014; Smith & Butovsky 2012). Some of these scholars (e.g. Smith, Basu and Kliman) have produced empirical evidence, the alleged absence of which is a reason for Harvey’s scepticism towards the theory.

Inadequate treatment of space and place

Harvey has correctly said that historical materialism, like all social theory, has under-stressed space. He is also right to emphasize how capitalism makes use of spatial mobility or immobility in its own interest. However, he incorrectly gives a lot more power to space than space can carry. He says that UGD ‘is a key means by which capital periodically reinvents itself’ and that without UGD (and associated competition between areas) ‘capital would surely have stagnated, succumbed to its sclerotic, monopolistic and autocratic tendencies and totally lost legitimacy as the dynamic engine of society …’ (Harvey 2014: 161). The spatial mobility of capital keeps the whole stable, but parts can be in difficulties. Inter-place competition is a means by which the old is replaced by the new, a process in which capital reproduces itself.

An internal relation can be symmetrical and asymmetrical (Sayer 1992: 90). ²⁴ Harvey is conflating the asymmetrical internal relation between space and society with the symmetrical internal relation between them. It is one thing to say that capital, which inevitably (necessarily) produces unevenness, makes use of unevenness, just as it makes use of inequalities between racial and gender groups, in its own interest. ²⁵ But it is another thing to say that if tomorrow the world economy or the United States or Germany becomes more geographically equal, and more socially equal (in terms of gender and race relations), capitalism will necessarily ossify/stagnate. What ossifies and stagnates capital is not all this, although, given that all things are interconnected to a variable degree, spatial and social equalities will create some difficulties for capital, by empowering labour and subordinate groups (in less developed areas) to fight against exploitation, and against social oppression through which mechanisms of exploitation often work at a concrete level. What really and necessarily ossifies and stagnates capital is the thing the significance of which Harvey definitely dilutes: capital’s inability to appropriate surplus value at a normal rate, the tendency for the rate of profit to fall, and capital’s ongoing fight to increase the rate of surplus value. If capital has survived, it is less because it has produced UGD and moves between places, although all that is surely not immaterial, and more because it has been able to engage in a dual war. It has engaged in a war against workers, semi-proletarians and small-scale (communal) producers within a nation (i.e. attack on their living standards). This war makes a region/country more profitable for capital, by making use of state power and the power of union bureaucrats (and even some ‘left’ intellectuals) who control the anger of the masses. ²⁶ And then there is an imperialist war, the main victims of which are, once again, ordinary working people, both at home and abroad.

It should also be pointed out that the idea about the see-saw movement of capital (that the rust belt of today can be the Sunbelt tomorrow) does not often hold: many areas (e.g. resource hinterlands in Canada, the Appalachia and numerous regions of the Third World) see little sign of capital investment for decades. This situation points to the existence of durable structure of uneven development. From another angle, there is little sign of capital moving out from certain places, because capital has other ways to cut costs than relocation. And even when capital moves, very little can be said about the actual pattern of spatial movement theoretically. ²⁷ Harvey says that reduction in costs of movement due to innovation in transportation and communication technologies allows capital to offshore. But it is not the cost of geographical movement which is the real issue. Consider in-shoring now, which Harvey ignores. With the support of union bureaucrats, the Obama government launched its crackdown on unions, thus driving down the wages, and this has made it possible for some industries to move back to the United States. This type of movement has less to do with constraining effect of distance and more to do with class relations and class struggle, including in the form that the state engages in on behalf of capital, all of which creates Third world–type conditions in advanced countries such as the United States or Canada.

Harvey makes a distinction between areas where there is an over-accumulation of capital and areas which receive over-accumulated capital, and between capital that is fixed/immobile and capital that is not. These distinctions are important. But is it not possible to think about geographical areas in terms of the organic composition of capital: areas where there is a higher than average ratio of organic composition capital (c/v) and areas where the ratio is lower? Such a difference, other things constant, may generate capital flows from the former areas to latter areas. As Grossman (1929) implies (Kuhn 2009, 1995), geography can be important from the standpoint of the falling rate of profit owing to the changes in the organic composition of capital. ²⁸ Because Harvey down-plays the class relation between capital and labour despite his Marxist instincts, and because he wants to show that spatial unevenness is directly caused by capital’s own competitive dynamics, he ends up giving more power to space than is due.

An alternative view of spatial unevenness would be this at a rather general level (which Harvey fails to consider). ²⁹ Capital operates under geographical conditions which it, at a given point in time, hardly chooses: it operates in a given system of places. Each place is, more or less, unique in terms of such features as natural resources, cultures of work, deployment of gendered and racial discourse to justify payment of low wages to some people, labour struggles, government policies, non-capitalist labour processes and linkages of a place to other places and to the world economy. Epistemologically, the place-specific features constitute the level of ‘the here and the now’ (Ollman 2003: 88). Capital’s more general tendencies/mechanisms operating at the level of capitalism-in-general and at the level of capitalism at a higher stage of development – these include competition, including in geographical space, exploitation of labour, tendency towards crisis via technological changes, imperialism, etc. which are, more or less, independent of time and place within the history of capitalism – interact with the time-and-place-specific features of capitalism. Such interaction produces geographical unevenness, which in turn may impact the ways in which the relatively general mechanisms work (including capital’s tendency to agglomerate or disperse). This is a view in which place (or place-time) specificity is given more importance than Harvey allows. What is needed is a more totalizing view of uneven development, which includes a consideration of more general mechanisms some of which Harvey discusses, and a place-based perspective, which stresses the place-specificity of social processes. ³⁰

Relations between places – of which uneven development is an aspect – are, to a large extent, relations between capital and labour. Harvey’s theory of UGD can be more or less read without having to know much about the relation between capital and labour as such, including what happens in the hidden abode of production. ³¹ Geographical variation in forms of surplus appropriation is an important cause of UGD (as mentioned earlier). While highlighting the distinction between two types of areas – an area that generates over-accumulation and an area that receives over-accumulated capital through a spatial fix of the first area, he consequently ignores the fact that a territory or a nation experiences not only uneven development (e.g. capitalist growth is more here and less there) but combined and uneven development. The latter, at the risk of simplifying, can be described thus: regions with a history of less-developed technology and its corresponding social relations (including non-capitalist processes) are combined with those with more advanced technology and relations, a process which simply indicates a drawing together, in space, of different historical stages of a country’s developmental journey, a combination of separate steps and an amalgam or a melding of more archaic with more contemporary forms (Trotsky 1932: 5; also Ashman 2009; Bond & Desai 2006; Lowy 2010). ³² This process of combination is expressed regionally, with some regions with their more developed social relations and higher level of economic development (expressed as higher labour productivity) interacting with regions with less-developed social relations and lower level of economic development.

Under-theorizing the relation between UGD and the state

Harvey (2009) says that ‘The theory of uneven geographical development needs further development’ (p. 71) and that his own theory of the state ‘needs some work’ (Harvey 2006 [1982]: 159). If we place these two statements in relation to one another, we are likely to arrive at this conclusion with some justification: the state has not been made an integral part of his uneven development theory, although it is not absent in his theory (see Harvey 2014: 155–116). Harvey is right to stress that even in neoliberalism, the state policies, no matter how incoherently, must create at least some of the general conditions for accumulation and help capital ride over periodic crises. The state plays this role by helping with the absorption of surplus capital through the production of built environment needing bulky investment. The state is not only involved in urban and regional planning practices. It is also involved in interstate relations. While these are all good points, there is little recognition of the fact that like capital, the state also works in a system of places, that is, in a geographical environment which it does not entirely choose at a point in time. In other words, at a given time, it works in a space that is already geographically differentiated by capital accumulation (and other processes). This means that a given national level policy (e.g. policy of neoliberal reforms) in support of capital must generally work differently in different regions, contributing to UGD.

As argued by Ollman, who provided the impetus to Harvey’s dialectical thinking in the 1970s: given the internal relation between the state and capital, state policies are a form of class struggle. Harvey abstracts from this view. He assumes that the state will be able to at least temporarily help capital absorb its surplus capital. ³³ For Harvey, the only limit to what the state can do for capital is that state-promoted built environment production will experience a glut. He fails to consider other limits. For one thing, the falling rate of profit sets limit on what the state can do on any permanent basis, although state policies do try to counteract the tendency. A most important way to increase the rate of profit is to increase the rate of exploitation, and the built environment production does relatively little in this regard directly. ³⁴ Furthermore, the state’s success in helping a regional economy in a country is always shaped/limited by the fact that it must ensure a given level of profitability in that country: if it taxes away a part of the super-profits that companies enjoy in an area to provide subsidies to companies in another area, the companies in the former area may resist (Mandel 1973). Over-accumulation is not the only limit to state’s capacity to help capital. And if capital has survived, it has survived less because the state has helped it absorb over-accumulated capital via fixed capital investment, which produces space, and more by the sheer class power of the state: attack on workers and small-scale producers at home and abroad to increase capitalists’ wealth (in value and non-value forms).

Reformist politics

Problems with Harvey’s (2014) intellectual analysis of the relation between capitalism and UGD are connected to his problematic political strategies. According to him, the capitalist class people should be made to ‘look to their responsibilities’ (p. 162). For this author, to insist that bankers merely ‘look to their responsibilities’ – and hear us – is to agree that they have a right to exist as bankers. Harvey wants to see a certain reversal of neoliberal (pro-market, pro-business) policies, a curtailment of the freedom of business. ³⁵ Harvey’s politics is mainly anti-neoliberalism (or Polanyian). It is not specifically Marxist anti-capitalism, that is, anti-capitalism from the standpoint of the international proletariat and the building of international socialism. His political stance – as that of many from the so-called Left – is potentially rooted in his conflation of levels of generality: the fight against capitalism in general becomes merely a fight against a specific form of capitalism (i.e. neoliberalism). ³⁶ He wants a new New Deal, which ‘is surely enough to fight for in the present conjuncture’ (Harvey 2003: 210–211; italics added).

Now, how will this new deal be implemented? He suggests a strategy which is no less reformist: social movements can put pressure on capital and on the state to move society in a progressive direction, to curb some of capitalism’s worst excesses. And, social movements have to work geographically. He supports the idea that radical change will happen if people can ‘occupy the parks, squares and streets of our cities until our opinions are heard and our needs attended to’ (Harvey 2012: 162). Besides, because capitalist development has been geographically uneven (in the sense that the capitalist space is a mosaic of regional economies), there must be a mosaic of anti-capitalist resistances (Harvey 2014: 163). There is no need for an internationally coordinated action both to fight capital and to construct socialism.

And who will be the agent – that force – of Harvey’s ‘anti-capitalist’ social movement? Let me first explain who, Harvey thinks, will not be the agent. In an interview, Harvey and Rivera (2010) said,

within Marxism we also have to … be very critical of [the] very conservative, rather dogmatic understandings of the world’ (italics added). One ‘can’t simply go back and cite Lenin as if somehow this is the solution. What a good Marxist does is to look at the conventional situation and do an analysis all over again given Marx’s method to try and understand the dynamics of the situation and therefore try to intervene in a way which is going to push society toward more democratic and more egalitarian solutions, and ultimately to solutions that are entirely non-capitalistic.

Harvey’s social ‘subaltern’ (from below) Left social-democratic belief is that small quantities of push will lead to a qualitative transformation. He continues,

the classic left-wing configurations … have a problem … Their notion of the factory worker as the vanguard proletarian figure that is going to make the revolution, I don’t think that works; I don’t think it ever really worked very well. ³⁷

He rejects what he calls a narrow view of the proletariat as ‘factory workers’. He insists on a definition of class based on concrete labour, saying that the term should include ‘all those who facilitate the reproduction of daily life: the care givers and teachers, the sewer and subway repair men’ and so on. He says, ‘You have to have a broader notion of an alliance of forces in which the conventional proletariat is an important element, but not necessarily an element that has a leadership role’ (Harvey & Rivera 2010).

So, who will play a vanguard role? Harvey demands a critical rethinking: ‘I think the left groups need to sit back and ask themselves who is likely to play a vanguard role in the current situation’ (Harvey & Rivera 2010). He offers a suggestion by connecting his ideas about the vanguard role to his theory of UGD:

those who are involved in the production of urbanization, the people who produce cities and the people producing city life. Right now, to the degree that the struggle is likely to be between public sector workers and the state apparatus, this is a very specific form of struggle, which is not based in the factories. It’s going to be the teachers unions and these (types of) groups that are likely to be pushed into a more vanguard role.

In that context, he says, ‘Thousands of delivery trucks [who] clog the streets of New York every day. Organised, those workers would have the power to strangle the metabolism of the city’ (Harvey 2012: 31).

To the extent that Harvey is saying that struggles around reproduction and ‘urban issues’ are important for the Left, this author accepts this view. But the point is that these struggles are (or should be) class struggles. And to view these struggles as class struggles requires one to have a proper conception of class, and such a concept is rooted in the ideas of Marx, Engels, Lenin, Luxemburg, Trotsky and so on (Das 2017). In such a conception, class is both a relationship and it refers to large groups of people; class is defined in terms of the relation of control over the means of production and relation of exploitation; and the property-less wage earners are the class that will fight capitalism as the tribune of the people, the class which will be at the head of all the oppressed masses. Harvey clearly does not agree with this view. Conceptually, for him, workers have no essential political role; sometimes, in some situations, some groups of workers may play a crucial role. ³⁸ The distance between Marxist theory of class relations and class struggle and Harvey’s could not be greater. ³⁹

Harvey’s suggested political strategy has several underlying theoretical problems. First, Harvey’s view of the proletariat is too narrow. In Marxism, one does not define the proletariat merely as factory workers: anyone is a proletarian if she/he, irrespective of the type of work (concrete labour performed) they perform, is separated from the means of production and depends mainly on wage work to survive. ⁴⁰ Second, Harvey (2014) says that capital is ‘always a moving target for opposition, and this is because of its uneven geographical movement’ (p. 162) and that ‘Oppositional movements in one space have often been defanged because capital moved to another’. This is a problematic view. Capital is a moving target only if one sees, like Harvey does, capital less in terms of class relations, and more in terms of things (e.g. factories, offices and currencies) which are more here and less there and which can be ‘moved’ around. If one sees capital mainly as a class relationship, as a relationship between who controls property and who does not, a relationship which is expressed in the form of things, then one can see that it is everywhere. Capital’s logic, its law of motion, is tendentially present everywhere. Capital is not a moving target. It is a ‘fixed’ target in that it is the target of opposition everywhere, although its concrete embodiment – whether factories, farms, banks, railways or airports – may change from one place to another. Third, advancing from the view that capital is a moving target because of its uneven development, Harvey then thinks – mistakenly – that there is a regional alternative to capital. But what does a regional alternative to what is a global social relation with a globally operating law of value mean? In what sense is there a regional alternative to the imperative of a workers’ government, at the head of a mass socialist movement, coming to power, everywhere? Such a government must take control of the most important parts of the economy and bring the large companies, whether in production or finances, in every regions of a country, under democratic control to establish a society where resources are used to satisfy human economic and cultural needs. Harvey takes away something from Marxism and gives it a little in return. With sympathy for the occupiers, he occupies a middle ground between reactionary capitalism and revolutionary socialism. According to Lenin, for whom Harvey, like many on the Left, has little/limited support, there is nothing significant between capitalist and socialist views. ⁴¹

Finally, for Harvey, there is no need for a centralizing democratic coordinating agency to mobilize proletarians and semi-proletarians. He forgets a whole history of Marxist thinking: only a democratically functioning vanguard ‘world party’ of revolution, which is mindful of inter-national and intra-national differences in the conditions for revolution, would carry out revolution and construct socialism, that is, democracy in every sphere of life and at multiple scales. It is a democracy which represents a qualitative break from the so-called liberal democracy (Bobbio 2006; Wood 1995). Harvey’s overall political perspective is not the overthrow of the present dictatorship of finance capital, but the development of a social movement (not a socialist movement) to pressure its representatives to just listen and throw some concessions (new New Deal). How will what are mere social movements challenge capitalism, backed up by the powerful state, which is the other arm of the body that is called class relations? Class struggle is what is needed at multiple geographical scales, especially in forms that Harvey is critical of: the ‘traditional’ forms of class struggle in workplaces and class struggle everywhere, although these have to be supported by other forms of struggle, ⁴² including against non-class oppression, the main victims of which are often workers and petty producers. The working class – conscious elements of the working class – is the vanguard of all sorts of exploited masses, including petty producers, and all those who are oppressed because of their skin colour or sexual orientation, even though it is not the only agent of revolution. Harvey does not share this view.