Virtual special issue editorial essay: ‘The shitty rent business’: What’s the point of land rent theory?

Introduction

In a letter to Engels recounting the misery of his family’s impoverishment, Marx found an unlikely reason for optimism in land rent theory; pronouncing that he had ‘at last been able to sort out … the shitty rent business’. One would not begrudge a desperate man even the most unlikely source of succour, yet the subsequent century of economic thought has shown land rent to be anything but sorted out. Confusion and conflict has reigned throughout and today the topic is largely neglected in the social sciences except as a set of inherited assumptions in the models of urban economists or a heuristic for the idiosyncratic analyses of some Marxists.

That is not to say that the issues surrounding land rent have disappeared. They are central to many key contemporary topics in urban studies such as the financialisation of land, the dynamics of house prices, the governance of urban infrastructure, land grabs, expulsions, gentrification and redlining; to name but a few. Yet even as interest in these issues has intensified, their political economic kernel – land rent – has been black-boxed as many researchers have either turned away from political economy approaches in general or simply found little applicability from the imposing, esoteric debates in the rent theory literature. The aim of this virtual special issue is to clarify the raison d’être of a theory of rent and encourage debate around its uses.

First, a definition of terms. Land rent is ‘a payment made to landlords for the right to use land and its appurtenances’ (Harvey, [1982] 2006: 331). It is the total rent paid. Ground rent is the rent paid for the use of the land, minus that paid for the fixed capital on the land (buildings and other appurtenances). The distinction itself is not crucial to our discussion and we use the terms interchangeably but, properly speaking, the discussions on rent theory pertain to ground rent because the part afforded to buildings and their appurtenances is usually considered a straightforward return on capital invested. Similarly, land values and the land market fall under the purview of rent theory because the value of land is held to be the result of its estimated rental value over the future (typically a period of two or three decades). Ground rent, then, is seen as the major determinant of both the contracted rent paid by tenants and the land’s purchase price.

Much of the focus of neoclassical theories of rent, and to some extent approaches in geography since the ‘spatial turn’, has been on spatial differences in land price. However, the overarching problem requiring a theory of rent is not this, but that of explaining the existence of ground rent at all: why does land command large values, the largest portion of which cannot be attributed to labour or interest on capital investment but seemingly appears for nothing? The attempt to account for this payment generates a corresponding problematic, of which spatial differentiation is a part. In her insuperable review of the topic, Haila asserts that a theory of rent broaches one or more of three questions:

In our view these questions are, of themselves, important enough to justify a pursuit of rent theory. Yet in the quarter-century since Haila offered this outline and called for a modern theory of rent, the trend has been one of neglect punctuated by isolated calls for a revival amongst critical scholars (Anderson, 2014; Haila, 2015; Harvey, 2010: 140–183; Jäger, 2003; Park, 2014). What remains to be demonstrated, perhaps, is not only that the questions themselves are significant but also the efficacy of extant rent theory in answering them and the explanatory power of these answers in understanding the geographies of capitalism.

In this introduction to a virtual special issue on land rent, we sketch out the history of rent theory, encompassing: classical political economy; Marx’s political economy; the marginalist turn and foundations of urban economics; and the Marxist revival of rent theory during geography’s spatial turn. We then overview some of the strands of literature that have been prominent since the breakdown of this consensus in political economy approaches centred on capital switching, institutionalism of various stripes and the rent gap theory.

We offer a critical urban political economy perspective on the topic, highlighting three problems hampering a full political economic theorisation of rent. First is a tendency in recent critical literature to ignore Haila’s first question, of the emergence of rent, and focus only on Haila’s second question, the nature and operation of the actors of rent. By thus taking the question of why we pay rent at all as unproblematic and ignoring the particular dynamics of rent that the idiosyncrasies of land imbue it with, the contemporary critical literature risks reproducing the conflation of land and capital that underpins many of the contradictions of capitalist urbanism (Harvey, [1982] 2006). This needs to be addressed with a robust theorisation of the categories of rent and of the features of land as a primary factor of production. Second, the critical literature on rent has eschewed a theorisation of the bid-rent function but in doing so loses the conceptual grounding with which to build a non-functionalist theory of land markets and their role in the capitalist coordination of space. Third, ‘absolute rent’ has been rejected in the literature but should be the basis of a critical theory of monopolies. Indeed, as the form of rent that arises only through the violence of asserting property rights or class position, this category should not only be rehabilitated but requires extension beyond land to an increasingly extractive financialised capitalism rife with distributional conflicts.

The roots of a problem: Classical rent theory

For the French physiocrats and Adam Smith ([1776] 1982), land rent is the price paid for the value contributed by nature itself. However, to maintain that land is a source of value is incompatible with the labour theories of value which prevailed in classical political economy. A labour theory of value holds that the economic value of a commodity depends on how much labour must be spent in order to produce it. It follows from this theory that land cannot command a value in itself as it is permanent and does not require labour to produce. Ricardo’s ([1817] 2004) solution was to bring rent theory into accordance with the labour theory of value using the notion of ‘differential rent’.

Differential rent was first formulated by James Anderson ([1777] 1984; see Clark, 1988: 21) in his assertion that rent derives from differences in fertility of the soil which, therefore, determine the profitability of a farmer using a particular plot. The landlord’s rent is her claim on the increased profitability that results from using her plot of land over others. Thus differential rent entails only a redistribution of the profits (rent is in ‘an ingenious contrivance for equalising the profits to be drawn from fields of different degrees of fertility, and of local circumstances’ (Anderson, quoted in Clark, 1988: 22). As such, if no rent were charged the price of the commodity would be unaffected, the tenant farmer would simply not have to share her profits.

For Ricardo, labour is the only source of value so all rent must be differential rent. Fertility is a feature of nature but is economically valuable only as a factor affecting how much labour must be applied in order to produce the commodity. With differential rents being derived from the advantage a particular plot of land holds over inferior plots, Ricardo held rents to be determined by the fertility of marginal land in cultivation – that is, the least profitable land that is in use (the amount of land in use being determined by demand). Such marginal land commands no rent itself but forms a sort of baseline: any fertility above this level is a productivity gain due to the land and the resulting extra yield is taken by the landlord as a condition of the farmer’s access to it. Ricardo’s theory can be regarded the definitive classical political economy statement on rent and is the departure point for neoclassical economics and Marxist theory, the two dominant approaches to the subject.

Marx’s theory of rent

Marx reformulated the labour theory of value and introduced two important innovations to Ricardo’s theory of rent. In the theory of value, he argued that it is not the labour put into creating a specific commodity that determines its value, but the socially necessary labour time required to produce that or similar commodities across society as whole, that is, the average labour time required to produce something under current technological and social conditions determines its value.

This reconstituted labour theory of value led to the first of Marx’s innovations to Ricardo’s rent theory: it is no longer necessary to posit marginal land commanding no rent as a baseline for differential rent. Instead, differential rent is understood as charged on the basis of enhancements over a socially determined acceptable level of profitability of land in use (see Ball, 1977; Fine, 1979). Differential rent in Marx’s theory, then, is not purely technical or ahistorical but depends on the specificities of prevailing socio-economic relations (Haila, 1990: 283).

The second innovation was to incorporate theories of monopoly rent. Ricardo had rejected Smith’s proposition that rent is a determinant of price, arguing instead that all rent is differential and so only a transferral of profit. Marx, however, reincorporated a theory of monopoly pricing into his rent theory in (a) allowing for the existence of ‘natural’ monopolies (Ramirez, 2009) where the unavoidable scarcity of something means that its price is limited only by effective demand; and (b) arguing for the existence of ‘absolute rent’, where the barriers imposed by the existence of a rentier class in itself is the source of rent.

The general logic of the argument is to explain how the existence of rent is consistent with a labour theory of value and to deduce the economic conditions and social relations that must be in place for this to be so. On this basis the two categories of rent are identified: differential rent, in which the landlord claims the excess profit from the competitive advantages of using their land and so is a rent based on the distribution of profits that would exist anyway and does not affect the price of the final commodity produced; and monopoly rents, based on the impairment of competition which, as such, does enter into the costs of production and affects the price of the commodity produced.

These rents are further subdivided into two categories of differential rent (DR): DR1, also known as ‘extensive rent’, being due to increased productivity attributable to an existing feature of the land; and DR2, also known as ‘intensive rent’, being due to increased productivity attributable to investment upon that land. And two categories of monopoly rents, distinguished based on ‘whether the rent flows from a monopoly price, because a monopoly price of the product or of the soil exists independently of it, or whether the products are sold at a monopoly price, because a rent exists’ (Marx, [1894] 1981: 910). This first is monopoly rent in which the impairment of competition is due to some natural feature of the land of which there is a limited supply; and the second absolute monopoly rent, in which the impairment is attributable to the existence of the class of rentiers themselves (see Table 1). The different forms of rent, it must be made clear, may be at work simultaneously and are empirically indistinguishable as the actual rent is only paid in lump sum at a price determined by the tenancy contract negotiations (in the case of annual rents) and the bid-rent process (in the case of land purchases).

OPEN IN VIEWER

Table 1.

Differential, monopoly and absolute rent.

Differential forms	Description	Examples
Differential rent 1	Rent arising from increases in productivity due to some feature of the land	Classic example: Fertility of the soil Modern example: Distance from workplace/market, as per Alonso-Muth model
Differential rent 2	Rent arising from increases in productivity as a result of investment on the land	Classic example: Investment in improving the fertility of the soil Modern example: A shopping mall which invests in facilities and services to ensure that its tenants receive greater custom (see Lamarche, 1976)
Monopoly forms
Monopoly rent	A rent arising from some unique, non-substitutable feature of the commodity which is, as such, limited only by effective demand	Classic example: Fine wine from a particular vineyard Modern example: A toll road that is the only viable route, or the sale of a Picasso painting (see Harvey, 2012)
Absolute rent	A rent arising because of the existence of a class of landlords acting as a barrier to entry for capital or consumers. Can take the form of: 1) a reservation price which keeps land out of supply; 2) concerted, cartel-like action amongst landowners in order to circumscribe competition and/or exploit consumers	Classic example: Class of landlords preventing the entry of capital into the agricultural industry and so preventing the equalisation of the profit rate, maintaining higher rents as a result Modern example: 1) housing which the landlord keeps vacant rather than rent out at a loss (see Walker, 1974) 2) protection/creation of a monopoly through litigation despite substitutability otherwise, i.e. brand protection of wine from the Champagne region of France (see discussion in Harvey, 2012: 89–112)

Further, the basis of rent is the monopolisation of particular portions of the globe by a certain class demanding a payment for its use, so in this sense every rent is an absolute rent: it is only the application of private property to land and the existence of a class of landlords demanding a certain rate of profit that allows the existence of rent in the first place – an aspect Evans (1999, this special issue) explores in his attempt to translate it into mainstream economic theory through the concept of ‘minimum rent’. This is the notion of absolute rent as a ‘reservation price’. However, while this class-monopoly is the necessary precondition for rent, it is not sufficient to explain as to how the minimum rent is met or exceeded – monopoly does not, in itself, create value. This is what the categories of rent describe: a set of conditions (and implicit corresponding social relations) in which rent above a minimum tribute is possible in a capitalist economy.

Much of Marx’s work on this topic centred on building a theory of rent commensurate with his value theory. It is not surprising, then, that the transformation of classical political economy into economics – centring around the shift from labour to marginalist theories of value – corresponded with a long period of quietude on rent theory. In addition, the agricultural question was no longer as salient in a century characterised by rapid and mass (sub-)urbanisation, and it was not until the 1950s that any serious effort was made to adapt agricultural theories of rent to the urban context. It is to the marginalist revolution in value theory and subsequent attempts to apply the tools of economics to understand urban land use which we now turn.

The marginalist turn

For the classical theorists, the labour theory of value was held to be central because competition pushed the value of commodities down towards the costs of production, so over the long run and across the economy as a whole (except, importantly, those situations where competition is hampered and so monopoly rents arise) the determining factor of commodity prices is the value of the labour imbued in them (the price here is not understood as 1:1 with value but varies around it, averaging roughly the same over the long-term). This suited the agenda of classical political economy which took production and the process of capital accumulation as the starting point of analysis, with a focus on the social character of economic activity (Mandel, 1962).

Economists of the marginal revolution, by contrast, posited some exogenously determined given supply of productive factors and demand as an independent factor, so that ‘the economic problem was to search for the conditions under which given productive services were allocated with optimal results among competing uses’ (Blaug, 1985: 295). These results were to be optimal in the sense of creating maximal satisfaction (utility), and value was held to be determined not by the intensity or absolute utility provided (whereby bread would be more valuable than diamonds) but that provided by the last unit needed to be completely satisfied; hence the term ‘marginal utility’ (Mandel, 1962). On the basis of this principle, utility curves can be constructed demonstrating the point of equilibrium, being that at which supply and demand is balanced and utility is maximised in terms of resource allocation.

With the emergence of macroeconomic and institutional approaches in the 20th century, the neoclassical paradigm of pure microeconomics would not maintain a complete hegemony over economics but its marginalist reconception of value shifted the perspective of value theory from ‘objective’ (in the sense of being determined by costs of production) to subjective, and from the long-term perspective of the wealth of nations to the abstract a-temporality of mathematical modelling. As such, ‘[f]or the first time, economics truly became the science that studies the relationship between given ends and given scarce means that have alternative uses’ (Blaug, 1985: 295).

Insofar as the marginalist theory of value became dominant, the problem animating much of classical rent theory – to explain the apparent existence of values paid to landowners that do not correspond with any labour imbued in a product – disappeared. Yet this does not eliminate the need for a theory of ground rent. Despite all appearances to the contrary, the basic principles determining ground rent are relatively simple, as Foldvary summarises: ‘the supply of land of a particular quality, relative to marginal land, sets the rent, utility being equivalent to the productivity’ (2008: 11). We can take issue with the Ricardian assumption of marginal land as the yardstick in Foldvary’s summary, as well as whether one wants to use the concept of utility and productivity as equivalents, but the basic principle applies for any treatment of land rent: it is determined by the supply of land of a particular sort on the one hand, with shortages in supply of that sort creating monopoly rents; and the productivity and/or utility increase that that particular plot of land provides on the other, so creating differential rents. Of course, in that the market value is determined by the supply of the product in relation to the utility its purchase provides, land is no different than any other commodity. The crucial question is whether land is like other commodities within this valuation process or if it has some unique feature as a factor of production that sets it apart and requires a distinct theorisation.

Within classical political economy, land was considered a free gift of nature and, as such, was seen as a primary factor of production requiring a separate theory. Marginalists began to question this and the issue of how far to erase the classical economists’ distinction between capital and land was a major debate which never saw a satisfactory conclusion (see Blaug, 1985: 79–83; Clark, 1988: 32–52; and Foldvary, 2008, for extensive reviews). Ultimately, at least within the economists’ paradigm, the problem is reducible to one of the elasticity of supply: the assumption made in mainstream economics is that the supply of land will be responsive to market demand through the augmentation of available land via infrastructure extensions, the depth of land (digging down or building up) or simply changes in use (Blaug, 1985). On the other hand, those, such as Marshall, who felt land should be separated as a factor of production, argued that land is unique (so it is difficult to find an adequate supply of a given quality), difficult to adapt to other uses because of the irreversibility of changes and other path-dependencies, and impossible to augment the supply of in some contexts (Clark, 1988: 32–52; Marshall, 1893, 1961: 430–432). In this virtual special issue, Neutze (1987) outlines some of the established problems in changing the supply of land but also raises the problem of uncertainty over future land use (and hence value), something the prevailing static models in urban economics struggle to account for.

If these latter arguments are accepted then land cannot be said to be a normal form of capital responsive to supply/demand, is not subject to many of the assumptions of the marginalists’ models, and so requires a distinct understanding of the emergence and dynamics of its market. Within mainstream economics the treatment of land as a form of capital is almost ubiquitous, but this appears to have been mostly due to mathematical convenience in being able to take only two factors of production (labour and capital) into account as opposed to being due to any persuasive argumentation. It seems clear to us that land is distinct from capital and that this underpins many of the contradictions of the capitalist production of space. Conveniently, however, making the assumption that land is perfectly responsive to pressures of supply and demand also allowed economists to construct models of perfect competition and subsequent equilibrium in land markets, as in the Alonso-Muth model outlined in the next section.

Von Thünen and the Alonso-Muth model

Ricardo’s differential rent theory only considered agricultural land and so did not take into account competing uses of the land, while rent differentials were based on non-marginal benefits to productivity – two fundamental incompatibilities with marginal theory. Yet the core logic of the theory ‘is formally identical with the marginal productivity theory’ (Blaug, 1985: 79). As such, the development of location theory in mainstream economics proceeded as an adaptation of Ricardian differential rent but with its basic assumptions replaced by marginalist ones (of competing uses determined by marginal utility curves translated into the price mechanism and resulting in equilibrium). In fact, location theory in economics took its departure point not directly from Ricardo but from a 19th-century German landowner, Von Thünen, who modified fundamental Ricardian assumptions by making rent differentials not dependent on increased productivity over the worst land, but a function of distance and transportation costs to market.

Von Thünen (1926) developed an agrarian land use model in order to price his own land by assuming that there is a single price at the market for each agrarian product. Further, production costs and the required income for the farmer are assumed to be the same for each land unit. The market price per hectare minus the farmer’s required income and the production costs then equal the bid-rent plus the transportation costs. Therefore, land that is closer to the market has lower transportation costs and commands higher bid-rent. There are competing uses of the land in terms of what agricultural commodity the farmer produces and some products need shorter transportation times to the market (e.g. dairy products) than others (e.g. grain). It follows that there are concentric rings of land use, commonly termed ‘Von Thünen rings’, as formalised by Launhardt (1885; Blaug, 1985: 618; Shieh, 2003), who produced Figure 1 demonstrating the rent gradient in such a model.

OPEN IN VIEWER

Figure 1.

Von Thünen rings.

Although the earliest examples of a marginal mode of analysis were thus developed in relation to agricultural land, land and space are neglected topics in mainstream economics largely left to the specialist subdiscipline of urban economics which emerged in the 1960s with the work of Alonso. Alonso’s (1964) innovation was to translate Von Thünen’s agrarian land use model to one of urban land uses and within a neoclassical framework based on utility and spatial equilibrium. Like Von Thünen, Alonso assumed a central node and rising transportation costs as one moves away from the centre. However, Alonso also introduced the possibility of substitution and variations in taste. So some users are willing and able to pay more for central land than others, which explains the location of the typical North-American Central Business District (CBD), but the poor may live more centrally in more intensively used land, or the rich may choose to live further out and pay more towards transport costs so as to own a larger plot of land. This became enshrined as the Alonso-Muth model, after Muth (1969) complemented Alonso’s focus on urban land with one on urban housing and a simplified conception of utility assuming equilibrium of locational costs/benefits.

In this virtual special issue, Spivey (2008) is representative of a number of recent urban economics articles (see also Lauridsen et al., 2013; Sexton et al., 2012; Verhetsel et al., 2010) empirically applying the Alonso-Muth model and finding that, despite its simplifying stylistic assumptions, it captures an approximate mechanism shaping the morphology of cities. This approach is essentially static – the models capture a point in time and do not account for change – so Abelson’s (1997) article is remarkable in incorporating an historical account of the development of an urban area. Finally, Phe and Wakely (2000) offer an interesting attempt to improve the realism of the Alonso-Muth model by adapting its assumptions to include the meaning and perception of place, thus attempting to incorporate culturally determined relational factors affecting the value of location.

In summary, for Von Thünen, Alonso and neoclassical economists in their wake, differential rent/marginal utility is key while monopolies, if recognised at all, are seen as aberrations and are not incorporated in their models. For Von Thünen, rent is paid for the relative advantage of a place compared with the most marginal, e.g. as a result of lower transportation costs, higher fertility (DR1, in Ricardian/Marxist terms), or because of differences in the costs of preparing the land for agrarian use such as irrigation (DR2). The productivity of the land, and therefore the land price, can be increased if the costs of creating higher productivity are lower than the potential increase in the rent. For Alonso, not only transportation costs (DR1), but also the intensity of land use and differences in the costs of preparing land for urban land use (DR2) are sources of rent. Differential rent/the marginal utility of different land, therefore, is an expression of the relative advantage of a place and helps to explain a crucial geographical aspect of land rent theory: why are there spatial differences in land price? As an approximate mechanism, this is useful beyond the assumptions of urban economics. If we reject the assumption of spatial equilibrium, for instance, one can see this as an important driver of uneven development (see Smith, 1984: 175–205). The subsequent critical consensus on rent theory, however, was predicated on the rejection of this mechanism.

The Marxist revival

The short-lived revival and subsequent decline of heterodox rent theory has received extensive review, definitively so by Haila (1990), but see also, among others, Ball et al. (1985), Clark (1988), Park (2014). Given this, we restrict ourselves to a summary overview here, seeking to offer an interpretation of the direction in which this literature has moved in the 25 years since Haila’s review. Following her periodisation, heterodox rent theory is understood as enjoying a consensus during the 1970s under the influence of Harvey’s (1973) critique of urban economics, then a period of transition as this consensus broke down (notably marked by Harvey’s reformulation of rent theory in The Limits to Capital, [1982] 2006), and a mid- to late-1980s ‘rupture’ as any common framework for a theory of rent fractured and researchers began to question its foundations and function.

The magic roundabout

As Haila (1990) styles it, two main camps developed in this period of dissensus: a ‘nomothetic’ one led by Harvey, which seeks to derive generalisable laws; and an ‘idiographic’ one led by Ball, which advocated describing specific social relations of property development as opposed to relying on a general theory of rent. Following his exhortation to look at ‘detailed historical situations rather than to make gestures towards some grand general theory’ (Ball, 1985a: 86), Ball advocated a ‘structures of provision’ approach which would focus on describing the established sets of agents within a given context and the patterns of their interactions (see Ball, 1998, this virtual special issue). However, as he continues immediately after rejecting ‘gestures towards a grand general theory’:

Even though the effects of rent depend on historical circumstances the conditions that structure the operations of landed property at those points in time still need to be theorised; analysing rent mechanisms and evaluating their consequences are part of that theorisation. (Ball, 1985a: 86)

Effectively, Ball is emphasising that rent is only one aspect of understanding land markets and property development. Such an emphasis on variegation in property markets and the importance of institutions was undoubtedly necessary, yet it is not clear how Ball jumps from this to pronouncing the death of urban rent theory (Ball, 1985b: 504). As Haila (1990: 285) points out, ‘it is self-evident that relations in reality involve much more than an abstract rent relation’. Further, we would add, while the dynamics of rent is only one aspect deciding a given socio-spatial outcome, it is the only one that we know we can find across any capitalist context (capitalism being itself a historically specific set of relations) and the only one that amounts to a set of necessities conditioning the nature and existence of land markets. This does not mean that they are mechanistic or deterministic (indeed, they offer little predictive power as to specific outcomes) but it certainly means they are a crucial component of any analysis.

Ball’s critique of rent theory depended on refuting it as a theory that aspired to be able to explain everything in every context. This, Haila (1990: 287) points out, amounts to a critique of rent theory as a universal theory, not a general one in the sense of seeking generalisable laws applicable to many instances. For Haila, meanwhile, the problem with rent theory was that it appeared to take its generality as a given whereas, she asserted, any such generalities must be substantiated through empirically observed mechanisms. She identified the tendency for landowners to increasingly treat their land as a financial asset as just such a mechanism, for insofar as land is mobilised as a commodity then landlords become subject to the general laws of accumulation and rent comes to have a coordinative function over space. Harvey ([1982] 2006) initially posited this tendency but Haila departs from him in arguing that it cannot be theoretically deduced from posited tendencies internal to the logic of capital, but instead must be empirically investigated with an account of landlords’ behaviour. So, in contrast with the ‘old’ theory of rent which ‘explains rent within the system of production’, a ‘new’ theory of rent hinges on empirical exploration of ‘the existence, scope, and meaning of the tendency of land to become a pure financial asset’ (Haila, 1990: 270, 292).

For Kerr (1996), however, both Ball and Haila offer circular theories because they focus on the contingencies of real estate market dynamics and the actors therein to explain rent but, at the same time, allow that rent and rent-seeking are important aspects of that dynamic. Rejecting the ‘crossroads’ Haila posits between old ‘ossified theory’ and the new theory of rent she outlines (1996: 294), Kerr mischievously ¹ offered a counter-characterisation of rent as a ‘magic roundabout’ because both Ball and Haila’s theories of rent ‘start and end with the activities of landowners, rather than with capital accumulation and the capitalist users’ of landed property’ (Kerr, 1996: 80). The sole concern with the nature of the agents of rent led to a focus on the influence that landowners/property developers have on land prices without connecting that analysis either to the dynamics of capital production and circulation, nor land use. In contrast:

this tautology can only be transcended … if the theory of rent recognises the real estate sector’s dynamics does not explain rent but rather presupposes its existence and the changing ability of users to pay such rent. (Kerr, 1996: 82)

This accords with our reading of rent literature over the last 25 years in which there has been a convergence upon institutional approaches describing a diversity of actors of rent, their immediate motivations and social relations without any connecting analysis of rent as a political economic category itself. The effect has been to implicitly reproduce the economists’ denial of any fundamental difference between land and capital. Rent revenues from the land do, in practice, become treated as pure financial assets indistinguishable from capital; but it takes a complex set of institutional, regulatory, socio-cultural, calculative and political practices to make it so. We have become very good at documenting these in the literature on calculative practices but in doing so have tended to forget the caveat that rent is fundamentally different to capital proper; arising, as it does, in a very different way and with a peculiar set of characteristics. The conflation of rent and capital in actual practice is a fundamental contradiction of capitalism exactly for this reason and ends in disastrous rounds of market ‘rationalisation’ being applied to socio-spatial configurations (Harvey, [1982] 2006). To reflect this conflation in analysis is to reproduce a contradiction of practice into one of theory also.

In the rest of this review, we will look at prominent contemporary approaches which can be said to have spun off from this ‘magic roundabout’: ‘capital-switching’ approaches within the urban political economy tradition focusing on the entry of capital into the built environment institutional approaches developed from a wide range of theoretical perspectives but which all in some way explicitly focus on a theorisation of organisations and actors of rent as their guiding frame, and the literature on the ‘rent gap’ which remains the most consistent contemporary application of rent theory.

Capital-switching approaches

Scott’s article in this virtual special issue is to be located within the transitional stage of the development of the rent literature, explicitly placing itself within the urban political economy problematic which ‘seeks at the outset to conceptualise the urban process in relation to the structure and dynamics of commodity production’ (1982: 112). While Scott’s piece is somewhat atypical of this literature in drawing upon a Sraffian rather than Marxist approach, it is characteristic of what Haila deemed ‘old’ rent theory in that it embeds a theory of rent and location within the system of production. It was a growing rejection of this productionist focus that underpinned the move towards what she christened a new theory of rent (Haila, 1990: 290) focusing on investment flows into the built environment.

This rejection was intertwined with metatheoretical changes in geography, with the focus on production perceived as a feature of structuralism: Gottdeiner’s (1985) application of structuration theory to the development of the built environment was a frequent touchstone in the literature’s growing assertion that real estate has its own internal dynamics linked to those of finance as opposed to being subservient to that of manufacturing (see Aalbers, 2007; Beauregard, 1994; Feagin, 1987; Gotham, 2002). Surprisingly, however, the insight that real estate has autonomous dynamics to those of manufacturing did not provoke any exploration of the economic category of the rent on the land as quite distinct from the category of profit on capital. The result has been the proliferation of studies emphasising contingent practices of property development which, nevertheless, blackbox the one thing at the heart of the whole process in the appropriation of rent, so also undermining the basis for generalising their insights outside of the particular context under investigation vis-à-vis the motivations and logics of rent.

Here we have selected two papers engaged with this tradition, the first from Bryson (1997), bucks these trends and offers a substantiation of the sort of rent theory Kerr (1996, see above) had called for. Bryson points out that a series of intermediaries determine whether and how supply reacts to demand and argues with respect to the power of capital markets that investors’ criteria are a crucial determinant in the production of the built environment (Bryson, 1997: 1440, 1442). Yet he does not claim they are independent of the dynamics of rent. Rather, ‘[w]hat is built and where it is built is determined by current rental levels and yields as well as by the actions, perceptions and motivations of a variety of property development and investment interests’ (Bryson, 1997: 1445), and his empirical analysis of development in a marginal property market depends on exactly this: on the combination of investor requirements and the manipulation of rent mechanisms by property developers in order to ensure revenue from the specific properties in question. In this approach combined with an analysis of the sort recently offered by Smet (2015), which attempted to draw connections between housing prices and the geographically bound production and circulation of economic revenues, one could imagine how theories connecting the dynamics of surplus production and the circulation of rents might be constructed without being deterministic or productionist.

The other article from this strand of literature selected in this issue, by Guironnet et al. (2016), encapsulates why it is problematic to replace an analysis of rent with an account of the links between finance and real estate. Rejecting the focus on rent maximisation in the literature on the mobilisation of urban land as a financial asset (Charnock et al., 2014; Harvey, [1982] 2006; Kaika and Ruggiero, 2016; Moulaert et al., 2003), they assert that ‘in adopting a conception of financialisation as a general process affecting all landowners irrespective of their characteristics this approach paradoxically fails to fully engage with the growing importance of financial markets and investors’ (Guironnet et al., 2016: 1443). The problem, we suggest, is precisely the opposite. To assert that the literature claims financialisation is an even process affecting all landowners irrespective of their characteristics omits the body of work reviewed above emphasising exactly the historical contingencies of landowner characteristics in accounting for the tendency to treat land as a financial asset and the associated switching of financial capital into the built environment. The problem at the core of much of this literature is, as Bryson (1997: 1456) put it, ‘a confusion between the actions of landowners and the role of rent as a mechanism to control the operation of the urban land market’.

Guironnet et al. not only reproduce this confusion but compound it further by obscuring the dynamics of the land market as the subjectivities of investors. Thus, the developer makes particular demands over the surface area of the development for that which they have ‘deemed profitable in the light of … market circumstances’, it demands particular allowances on the basis of ‘[c]laiming an intimate understanding of the market’ and certain features of the wider built environment in the locality are sought by investors because ‘this is believed to influence both resale and rental liquidity’ (2015: 15–16); and all of this is proffered as proof that investors’ expectations shape urban development. Rather than entertaining the notion that these demands might correspond to strategies to maximise rent on that particular plot of land and in that particular land market, the analysis is halted at the fact that international investors and their local intermediaries form expectations about the market and act upon them.

As a result, their analysis begins to look very much like the ‘radical idealism’ of which Smith (1996, see below) accuses Bourassa (1993). A more charitable interpretation may be that their approach amounts to a form of what Ball (1998, this issue) termed ‘conflict institutionalism’ and the authors do gesture towards this in calling for the development of a financialised ‘structures of provision’ approach. However, they neither define the concept nor deploy it in analysis. This is one demonstration of how political economy approaches concerned with the entry of capital into the built environment, shorn of the substantive political economic analysis of rent theory, have begun to converge upon a rather ad hoc institutionalism.

Institutional approaches

For Marx ([1894] 1981) the institution of the state was crucial in creating the possibility of rent as it is only through property rights (defined and maintained by and through states) that land is monopolisable. The corrective to this provided by institutional approaches are important in their theoretical formalisation and expansion of the role of institutions beyond merely enforcing property titles, as well as their insistence that property regimes and their implementation are variegated. However, institutionalism itself is a wide tent with oft vaguely defined concepts and little by way of shared epistemologies or method between different approaches, as Ball’s (1998, this virtual special issue) review of institutions in British property research demonstrates.

Perhaps the most notable thing about institutionalist approaches is their lack of a shared definition over what an institution is and what status a theory of institutions should hold. Ball ascertains two main definitions: a ‘formal’ one based on the framework of property rights (distinguishing between organisations as the players and institutions as the rules) and ‘casual’ one in which agencies involved in property development are understood to be institutions (1998: 1502). Ball’s adoption of the casual definition on the basis that it appeals to the common sense meaning of the term appears to sit uneasily with his criticism of ‘ad hoc’ institutionalism (and what he terms ‘conflict institutionalism’ as ad hoc) on the basis that ‘[t]here is no clear theory of institutions and how to study them, rather elements are drawn together in ad hoc explanations’ (Ball, 1998: 1506).

As, indeed, does his avocation of a structures of provision approach as ‘not a complete theory in itself … [but rather] a series of statements about how to examine institutions and their roles’ (Ball, 1998: 1514). However, this is not a contradiction for Ball as he appears to be content with a theory of institutions as a ‘bolt on’ for other theories as and when including institutions in the analysis provides greater explanatory power. This is what the SOP is designed as: theoretical guidelines about institutions and their role in mediating supply/demand which can be appended to other theories (see also Ball, 2002). There is a lack of studies using this framework but Ball himself (2003) offers a study of factors affecting housing supply, while in this virtual special issue Wu (1998) deploys the framework in the context of Chinese urbanism.

For others, such as Needham et al. (2011), in this virtual special issue, a ‘bolt on’ approach to institutionalism runs the risk of institutions becoming a deus ex machina, deployed to explain away empirical results that run counter to the core theory. Embedded in new institutional economics which reduces institutions to transaction costs, they aim to complement this by looking to the more casually defined ‘old institutional approaches’ to build a theory which makes institutions internal to the theory of land markets. However, they maintain the methodological commitment to deductive, predictive model-making of mainstream economics and within this paradigm find that they cannot construct a general theory of markets which take into account institutions, instead arguing for partial theories tailored to explain the context of interest. That their attempts at a general theory fail is hardly surprising, for they attempt to integrate an ‘old’ institutionalist approach acknowledging that man-made institutions can affect preferences (Needham et al., 2011: 166) within a neoclassical methodology that is predicated on taking preferences as given.

Offering an institutionalism more rooted in political economy, Healey and Barrett’s framework attempts to ‘combine the insights from the traditions of institutional analysis … with the neoclassical analyses of the operation of the urban land markets and Marxist approaches to the way capital flows through the built environment’ (1990: 90, this virtual special issue). However, their treatment of rent is indicative of the obfuscation of rent prominent in the political economy literature and outlined in the previous section. In short, they reject the applicability of theories of rent and argue that to understand the way capital flows through the built environment is to understand the financial agents investing into it (1990: 92–94, this virtual special issue), an assertion which leaves them subject to the critique offered of Guironnet et al. (2016), above. Nonetheless, their framework demonstrates the potential of a more synthetic approach to land rent and urban development.

Indeed, while some, such as Guy and Henneberry (2000), are sceptical of economic approaches to land markets in favour of agent-focused institutionalism, there is no inherent mutual exclusivity. Institutionalism can be ‘bolted on’ to mainstream economics (see Ball, 2002; Guy and Henneberry, 2002; although this is a superficial solution in our view, as per Needham et al., 2011) and from a political economy perspective the institutional approaches to land and rent surveyed could be said to be variations on the Polanyian ([1944] 2002: 187) theme that ‘land is an element of nature inextricably woven with man’s institutions’. Polanyi deemed the commodity of land itself ‘fictitious’, meaning it is a commodity only through social construction as opposed to being the result of a production process. This being so, institutional factors fundamentally shape the market in general and give rise to a relatively unique position for landowners in that their engagement in relevant institutions directly shapes the form, content and profitability of their own commodity. There is no fundamental logical contradiction between land being institutionally constituted as a commodity, and that commodity being a concrete one from the point of view of the market and so subject to general laws of accumulation.

Rent gap

The rent gap literature provides a synthetic conceptual tool which has been a consistent application of rent theory at the urban level but has remained curiously isolated from wider theorisations of rent. Neil Smith (1979) developed the rent gap as an explanation of where and why gentrification takes place. Emphasising that the ground rent and the house value are separate components making up the house price, he pointed out that as houses age – and if they are undermaintained – the house price, the house value and the capitalised ground rent all go down but that the potential ground rent remains stable or even goes up (following the assumption that more central places have higher ground rents and that these go up if the metropolitan area extends). Smith labelled this difference between the potential and capitalised ground rent the ‘rent gap’. Over time, the rent gap widens until the point at which it becomes profitable enough to attract investment in redeveloping and/or revalorising the land, with the gap then closed through the actions of property-based capital. In this supply-side explanation, gentrification thus represents ‘a back to the city movement by capital, not people’ (Smith, 1979).

Rent gap theory offers a powerful understanding of the way in which the dynamics of rent determine the geographies and temporality of investment into the built environment. Smith’s explanation of gentrification came to dominate the literature on the subject throughout the 1980s at the expense of demand-side explanations, although it attracted some criticism (among others, Hamnett and Randolph, 1984; Ley, 1987). Bourassa (1993, this virtual special issue) argued that Smith’s distinction between two forms of ground rent (capitalised and potential) does not contribute to the explanation of either the location or timing of changes in land use. For Bourassa, rent, by definition, is based only on the current use of land, making it conceptually impossible to speak of potential rent. In his neoclassical account there can only be a difference between ‘current and potential, feasible land uses [b]ecause land rent and value change as soon as perceptions about the future change and do not wait for land use to change’ (1993: 1741, emphasis in original). That is to say, any future potential rent is the current rent because the capitalised rent is adjusted to reflect the best use of the land regardless of the actual use.

This critique emanates from a neoclassical methodology that cannot account for change: it simply assumes that the best price will be reached immediately and automatically, regardless of the actual use of the land or any informational and/or power asymmetries. This, Smith argues, amounts to ‘a radical idealism centred on the desquamation of taste’ (1996: 1201, this virtual special issue). Clark (1995, this issue), meanwhile, points out a number of technical points Bourassa misunderstood regarding rent gap theory and offers an adjusted representation of the rent gap as compared with Smith’s (1979) original, allowing that speculation drives up the land rents prior to a change in land use (Figure 2).

OPEN IN VIEWER

Figure 2.

The rent gap.

Further, as both Clark and Smith argue, rent gap theory is not intended to be predictive but an explanatory tool to understand the geography of gentrification ‘in particular places at particular times’ (Smith, 1996: 1202). Bourassa incorrectly separates rent gap theory from the larger theoretical framework in which it is embedded, that of ‘a political economic theory of uneven development on the urban scale [which] as such cannot be divorced from the societal relations and power struggles involved in the creation and capture of values in the built environment’ (Clark, 1995: 1489).

It is in this particularity, the adoption of the viewpoint of a particular neighbourhood, that we find the limits of rent gap theory. Hammel’s (1999) article in this issue points to this in arguing for greater attention to be paid to scale in rent gap theory on the basis that ‘potential land rent is determined at the metropolitan scale and capitalised land rent at the neighbourhood scale’ (1289). Indeed, within the reduction of ‘potential rent’ is a whole world of demand-side factors and the wider dynamics of rent. Regarding this latter, Smith’s supervisor David Harvey – himself mired in the categories of rent theory in writing The Limits to Capital at the time – was famously dismissive of his graduate student’s efforts (Slater, 2015), and this can be said to have rather anticipated a sympathetic but definite distance between rent gap and land rent theory. Although the link to a wider analysis of rent is explicitly made in rent gap theorists’ emphasis on uneven development (Clark, 1988; Smith, 1984), their particular scalar focus has led to a lack of integration between the two and Bryson (1997, this virtual special issue) is rather exceptional in deploying both rent gap theory and an analysis of the rent mechanisms at work in a particular development.

Challenges for future research

To summarise, land has unique features as a factor of production that sets it apart from capital in general and requires a theory of rent. In this, rent is determined by the supply of land of a particular sort on the one hand, with shortages in supply of that sort creating monopoly rents; and the productivity and/or utility increase that that particular plot of land provides on the other, so creating differential rents. Aspects of these differential rents as an approximate mechanism in the urban context are captured well by neoclassical models, and aspects of the monopolistic, socially constructed nature of land ownership are captured well by institutional analyses; however, with the convergence towards institutional approaches in the critical literature, the emergence of rent has been neglected and so an understanding of how capital flows through land made untenable. In doing so the critical literature reproduces the conflation between land and capital of both mainstream economic analysis and the extant practices of investors treating land as a financial asset, thus losing sight of a crucial contradiction which should be central to critique.

Heterodox rent theory has fallen into a state of dilapidation due – we suggested – to three major problems that emerged following its ‘rupture’ (Haila, 1990). Here we outline them again alongside associated suggestions for rehabilitating rent back into the centre of analysis.

First, within the context of urban land rent, a mixture of confusion over the applications of rent theory alongside rejection of structuralist explanations led to the proliferation of approaches emphasising contingent mediations and agentic factors rather than connecting analyses to a general theory of rent. This was an important corrective to a literature which often paid insufficient attention to mediating factors. However, in the course of this redressal theorists have failed to distinguish rent as an economic category distinct from its constitutive institutions (in the case of Ball, 1985b), typology of its actors (in the case of Haila, 1990), or fictitious capital in general (in the case of Guironnet et al., 2016). In attempting to combine an emphasis on the importance of all of these with a consideration of the dynamics of rent mechanisms, Bryson (1997) offers an example of an alternative to this ‘magic roundabout’.

Second, reservations regarding theories centring on the phenomenal form of price (as opposed to underlying value dynamics) meant that when Marxists conceded differential rent a central place in rent theory, little attention was paid to the bid-rent mechanism. This rendered their account of landlords as a class functionalist and disconnected it from an understanding of the wider land market (e.g. Harvey, [1982] 2006: 330–372). So instead of connecting their research to macro-level analyses and theories of the land market, researchers in this tradition have tended to adopt – as Ball (1985a) suggests they should – the rent categories as a political heuristic to expose extractive power relations within an institutional analysis (e.g. Baxter, 2014; Charnock et al., 2014), or obfuscated the issue of rent altogether. The bid-rent mechanism is crucial to understanding the land market in a way that avoids slipping into functionalism and should be integrated with considerations of power, capital accumulation and associated uneven development.

Third, a confusion over the status of absolute rent has led to disarray. This has meant that even where absolute rent has been unavoidable for analysis, as in those using the concept of class-monopoly rent to understand urban property markets, there has often been a lack of integration with wider understandings of the dynamics of rent. Most frustratingly, absolute rent should be the basis of a general Marxist theory of monopoly and its neglect has foreclosed potentially fertile ground to extend the theory beyond land to other situations where the existence of a class of rentiers itself creates rents – for instance, in the case of immaterial commodities where profit is reliant on the imposition of intellectual property rights, and the process of financialisation across the economy generally. Indeed, insofar as we accept that much new ‘production’ is effectively enclosure of various commons (Zeller, 2008), then this form of rent should be the central category for understanding capitalism today.

Amongst current talk of an ‘Anthropocene’ and ‘planetary urbanisation’ it is surprising, to say the least, that the economic fulcrum of the capitalist remaking of geography has fallen so completely off the agenda. One would reasonably expect the interplay of capitalist and spatial dynamics and their metabolism through the rent relation to be at the very core of geography and urban studies. A theory of ground rent is required not only for analyses of the politics of rural land (see Lefebvre, [1956] 2016) – especially its contemporary issues of large-scale land grabbing – but is a crucial link between urban political economy and the burgeoning field of political ecology more generally (Andreucci et al., forthcoming). Further, if the challenge of the last century was to apply land rent theory to the urban context, the challenge of this looks to be to take the categories of rent beyond land in the analysis of a capitalism increasingly reliant on flows of rentier income through financial instruments (recently theorised in the context of real estate by Haila, 2015, as ‘derivative rents’), immaterial commodities enforced by property rights such as in the case of carbon trading (Felli, 2014) and so-called ‘sharing economies’ on digital platforms; while, correspondingly, contemporary social struggles increasingly centre upon the existence and distribution of these new and old forms of rent. The challenges for future research are manifold, but the conceptual foundations exist. It is time to get serious and bring rent back into the analysis of land, cities and capitalism.