Land, Abstraction, and Housing Provision

As Anne Haila (1990) noted, when the debates around land rent of the 1970s and 80s extinguished themselves, its participants split in two directions. Some took a “nomothetic” route out of it: they formulated general principles about the spatial dimensions of capital accumulation from the vantage point of abstract speculation. For Haila, David Harvey’s work was the foremost example. In The Limits to Capital (Harvey 1982), Harvey’s in-depth examination of Marx’s categories of rent culminated in his famous thesis that land has a tendency to assume the behaviour of a pure financial asset under capitalism. This assertion, which provides the conceptual basis of his famous “capital-switching” approach, was the product of a deductive exercise. The extent to which this was actually true he left for others to determine: “[h]ow far capitalist social formations have advanced down such a path is a matter for historical investigation” (Harvey 1982, 371). Others took an alternative route out of the rent debate: an “idiographic” path. This second group grew suspicious of grand, abstract theorizing and opted for bringing concrete phenomena into focus instead. Haila identified Michael Ball as the prime example. In a series of pieces in the 1980s, Ball criticized the infatuation of his then-fellow Marxists with Marx’s cryptic writings on rent, accusing them of losing sight of how the “structures of building provision” actually worked (Ball 1983, 1986). Of course, Ball’s interest in rent theory, as well as in historical materialism more broadly, fizzled out around the turn of the 1990s. But his focus on analysing concrete “structures of provision” remained, continuing to use this approach on a series of detailed studies of housing and urban institutions (e.g. Ball 1994, 1999).

Undergirding this bifurcation is a perennial dispute in the social sciences. Can social systems be represented as predictable entities, governed by law-like general principles and logics of action (i.e. the nomothetic approach)? Or does this impute natural properties onto fundamentally contingent human phenomena? Should then historical change only be understood at the level of dynamic interactions within specific institutional contexts (i.e. the idiographic approach)? Or does this make analysis slip into description, rendering any theoretical knowledge superfluous? Most researchers in the social sciences make their peace with this philosophical problem by hovering somewhere in between both positions.

Residential Accumulation was a conscious attempt to craft a theoretical framework from what I think is a necessary middle ground. Given how abstract my theoretical framework is, it may come as a surprise that my starting point was heavily skewed towards the idiographic side. Much of my research in this area comes out of an in-depth historicisation of housing provision in Spain.¹ For a historically minded researcher, deductive forms of structural reasoning like Harvey’s above are automatically suspicious. Theorising first and consulting history second has always struck me as putting the cart before the horse (see Knafo 2002). That said, I have also become acutely aware of the poverty of structurelessness. In spending long stretches focused on the minutia of a particular case study, I sometimes found myself oblivious to obvious comparative patterns, as well as to the common mechanisms which explained them. The embrace of “nomothetic” forms of reasoning that is patent in my article was a reluctant decision that came out of necessity. I wrote Residential Accumulation as an exercise in self-clarification, in an attempt to find a theoretical anchor that would keep me from drifting into empiricism. However, the aim of my framework was not to produce a hard theory with predictive power, or prefabricated theoretical structure meant to act as a substitute for historical research. Rather, the aim of my article was to produce a flexible and open-ended frame of reference, with a lot of room for contingency; one designed to help critical housing researchers situate the phenomena they encounter within a broader whole.

Questions of abstraction figured prominently in most of the responses, albeit for different reasons. The thrust of Manuel Aalbers’ and Reneé Tapp’s comments is that I have swung the pendulum too far towards the nomothetic side, whereas Callum Ward contends that I have not done so enough. Aretousa Bloom has her reservations, but overall seems content with the balance I have struck. She has even started running with my framework, making her own adjustments and modifications, putting it to use precisely as intended. In what follows, I try to address the points raised by these respondents in a systematic way. But first, I address Michael Ball’s comments separately. Not only is he a neoclassical outlier in this discussion, but he also misrepresents my argument greatly.

Land Changes Everything

Michael Ball’s reply can be summarized as a reassertion of the neoclassical theory of rent. Or rather, the lack thereof. After all, from a neoclassical perspective having a theory of land rent is superfluous, since, as Ball puts it, “landlord behaviour is essentially no different from any other profit-maximiser, capitalist or otherwise, and that their behaviour plays a limited analytical role in rent determination” (Ball 2024, this issue). The reason being that:

price is determined by the equilibrium of demand and supply, rather than by labour value or pure supply conditions alone. That recognition led to the late nineteenth-century “marginal revolution” in economics when the principle was applied to all goods and services and to the rejection by the mainstream of labour value theories. (Ball 2024, this issue)

If the concept of “rent” seeks to capture why certain assets, like land, resist the coercive force of price competition, then there is no real need for it, because the movement of land prices can be analysed through the market’s self-balancing competitive mechanisms, just like any other commodity. A rising demand for a particular use of land, residential or otherwise, “increases prices which encourages more land into that use, while simultaneously choking off some demand, until demand and supply are brought into balance”. The implication for our discussion is that any existing imbalance in housing supply – e.g. too few affordable homes – is bound to be temporary. Developers will register the high price of housing and respond by building more homes, bringing house prices down as they satisfy demand. All of this is provided, of course, that there are no blockages to the free flow of the market. Ball raises the example of NIMBY homeowners blocking planning permissions to defend the value of their properties, though he could have just as well blamed excessive environmental regulations or bureaucratic red tape. If such fetters are cleared, then any excess profits arising from stunted competition in the property market (“rents”) will necessarily wither away. Or at least that is what neoclassical theory says.

Sparring with a Strawman

Ball’s response misses the mark, not so much because his arguments are not internally coherent, but because he misrepresents my own, leaving him sparring with a strawman.

The bottom line of Ball’s position is that there is nothing inherently monopolistic about land and landed assets. For him, this renders moot all my talk about the rent-bearing qualities of housing. Apparently, to argue such a thing implies that landowners “have some improbable skill at combining together in order to keep supply short and thereby (rents) high” (Ball 2024, this issue). What I do not seem to understand, according to Ball, is that land (and house) prices are not just determined by “pure supply conditions alone” but by demand also. Ball alleges that my adherence to a labour theory of value makes me oblivious to the role of demand in the determination of land prices, hence why I am forced to hallucinate the existence of a rentier conspiracy to explain why house prices are so high. He admits, passingly, that “Moreno Zacarés discusses a mix of demand factors” – “but then, in contradiction, suggests that only supply conditions in the form of landlords” inflationary behaviours actually determine rents’ (Ball 2024, this issue, emphasis added).

It is hard to say for sure, since he never quotes me, but the culprit is likely the following line from my discussion of land banking: [i]n periods of high demand [emphasis added], large urban landowners can use their hoarding power to put a chokehold on residential provision and squeeze the maximum amount of land rent from developers, who in turn pass the cost onto the consumer (Moreno 2024, this issue). Indeed, land rentiers can behave in this way, and in specific historical circumstances they may even organize themselves into interest groups to defend the value of their assets to block attempts to expand the supply of urban space (just as the NIMBY homeowners that Ball admonishes). But the existence of this kind of behaviour is not fundamental to my argument. Land rent exists even in contexts where land ownership is dispersed, or where there are no landlord lobbies. To reiterate: at no point does my theory of rent necessitate the existence of a conspiracy of rentiers to explain the behaviour of land (and house) prices.

The ownership of a fixed factor of production (land) is sufficient to explain the existence of land rent. If the behaviour of landlords – or of any other rentier – is “inflationary”, it is thus not because they scheme in smoke-filled rooms, but rather because they own scarce assets, which allow them to extract rents from consumers when the location is in demand, prompting what I refer to as “rent waves”. In my article, there are three main determinants of a house-rent wave and demand plays a key role in all of them: (1) First, the tendency towards spatial agglomeration of economic activities. This concentrates demand for urban space in economically dynamic areas, pushing up the worth of their land – a fixed factor of production of which there is only a limited quantity. (2) Second, I note how the differential advantages of particular locations (e.g. transport connections, access to schools, particular social identities, the list is not exhaustive) “are bound to increase demand for a location and bolster the rents afforded by its buildings [emphasis added]”. (3) A third determinant, which only comes into play later in my argument, is finance, which extends the lifespan of a rent wave by stretching demand capacity. Note that, contrary to what Ball claims, I am not ascribing here any “irrational” motive to landlords (or to any other urban rentier for that matter, Ball’s response is just particularly defensive of landlords). Urban landlords extract surplus value out of tenants in the form of house rent, whereas capitalists extract surplus value at the point of production. The difference here has little to do with their rationality as profit maximizers. It simply reflects that landlords happen to own land, the supply of which tends to lag demand in capitalist societies, which are marked by mass, continuous urbanization.

So, given that I do take demand into consideration, Ball’s argument about the limits of “absolute rent” falls flat. “absolute rent” – the rent afforded by a landed asset on account of the power of landholders to block the free entry of investment – performs only a secondary role in my argument, as I have just explained. As I address in a hefty footnote in the article, my view is that “differential rent” – the rent afforded by a landed asset on account of its productivity differentials – does not apply to housing insofar it is not a space of production.² In my view, the type of rent that housing assets yield is primarily “monopoly rent”: rent afforded by a landed asset simply on account of the capacity and a willingness of a consumer to pay for it. In the paper, I quote Simon Clarke and Norman Ginsburg (1976), who long ago made the case that house rents are monopoly rents because they “cannot be determined without reference to the demand of the consumer … [they are] in essence, the premium which a consumer is compelled to pay for a house in a particular location”. In short, in mistaking my argument for a theory of absolute rent, Ball has engaged in an absolute misreading.³

Ricardian Scarcity Vs. Walrasian Equilibrium

To Ball’s surprise, then, we agree on the importance of demand in the formation of land prices. Where we have irreconcilable differences is on why the supply side remains under-responsive to demand. After dismissing my very un-Marxist position that land rent originates from landlords’ conspiratorial behaviour (a position I do not hold), he advises me to “simplify and fall back on the Ricardian principle of [land] scarcity” (I position I already hold). In any case, Ball himself disagrees with the Ricardian position:

Ricardo suggested that, as land is the fixed factor, the benefits of rising agricultural productivity would eventually accrue to landowners and, along with that, inequality would become more extreme. […] Ricardo’s prediction failed because the future turned out to be different; transport improvements globalised agriculture leading to a collapse of agricultural prices and the great landed estates of Europe in the late nineteenth and early twentieth centuries. (Ball 2024, this issue)

Transposing this argument onto urban housing, what Ball is saying here is that land is not actually all that scarce. Contrary to that old speculators’ saying attributed to Mark Twain, “buy land, they are not making it anymore”, Ball asserts that more land can, in fact, be produced. And up to an extent, this is true of course. In the movement of capital across space, there is a certain “equalising” tendency to balance out the effects of uneven capitalist development. For example, new transport routes can bring peripheral locations within orbit of overcrowded cities, opening them up for suburban development. It follows that since land is still abundant, rising house prices should prompt additional construction and bring prices down, or at least provided that any restrictions to the movement of capital across space are removed.⁴ If the housing market is not properly meeting demand it is simply because we are not letting it. Laissez faire!

Leaving aside the bleak urban morphology that would arise from cutting corners in land-use planning or environmental regulations, the fundamental problem with this neoclassical view is that it exaggerates the tendency towards “market equilibrium” in space (Christophers 2009). Capital moves across space in a dialectic between equalization and differentiation (Smith 2010), but this dance is skewed towards differentiation, making uneven geographical development the norm rather than the exception. Once fixed capital is installed in a particular location – e.g. machinery, buildings, transport infrastructure, etc. – a social division of space arises with strong path-dependent effects (Lipietz 1980). The more costly it is to replicate a substantial modification of space in another place, the more that capital will tend to agglomerate around already developed areas. The result is a concentration of demand that exerts an upward pressure on land prices. Recent developments – e.g. working from home – may well ease the pressure somewhat. But until teleporting is invented, thus annihilating the spatial dimensions of capital accumulation entirely, it seems unlikely that the land supply will ever be as elastic as Ball presumes it already is.

A comparative look at the historical record should raise doubts as to Ball’s conviction that the mechanism of competitive equilibrium applies in space in full. Katharina Knoll et al. (2017) have compiled a house-price index for 14 advanced economies between 1870 and 2012. Notwithstanding some important variations between countries, their data series reveals the existence of a global secular increase in real house prices; a trajectory was only seriously disrupted by the destruction of the World Wars and the outbreak of the Global Financial Crisis. From the second half of the twentieth century up until 2007–9, global house prices saw a steep incline, reflecting the spread of homeownership, the extension of mortgage credit, and the retreat of social housing. When the authors analyse the composition of these house-price increases, the inflationary role of land becomes strikingly clear: “84 percent of the house prices during 1950 to 2012 can be attributed to rising land prices” (Knoll et al., 2017, 348). Since the principle of uneven capitalist development (or Ricardian scarcity) can readily explain why this is the case, the burden of proof is on Ball to explain why his faith in the operation of Walrasian equilibrium is not borne out by the data. It could well be that full capitalism has not been tried yet; but this explanation seems rather improbable. Alternatively, Ball could just “simplify and fall back on the Ricardian principle of [land] scarcity”.

The Uses and Misuses of Abstraction

To return to the opening discussion, my article sought to provide a framework for analysing housing provision under capitalism from a middle level of abstraction. The aim was to strike a balance between the grand, abstract theorizing of Harvey’s “capital-switching” and the context-specific focus of Ball’s “structures of provision”. As noted in the introduction, my own theoretical journey has had me move from ideographic towards nomothetic forms of reasoning; that is, from focusing on contextual specificities to exploring general principles.

Part of what inspired this turn was the theoretical under-specification of the structures of provision approach, to which I have always been sympathetic. In his many deployments of this approach, Ball never really developed a framework to guide those of us who were receptive and wanted to emulate his analytical method. Despite regularly renewing his call for others to study the provision process, Ball never really made clear how to do a structures of provision analysis. It thus not all that surprising that the approach never really caught on, unlike Harvey’s. At the same time, the empiricist bent of Ball’s use of the structures of provision approach did not facilitate asking the broad, critical questions about political economy that I wanted answered: What is the historical specificity of housing provision under capitalism? What place does housing hold in the capitalist economy? What contradictions arise from the provision of housing in a capitalist society and in what ways are these managed? In many ways, Residential Accumulation was my attempt to carry forth a theoretical project that, to my mind at least, Ball left unfinished after his abandonment of rent theory and his departure from Marxism. The challenge was reattaching his institutionalist sensibilities onto a Marxist theoretical matrix capable of explaining how surplus value is produced and appropriated in capitalist societies. A combination of rent theory and Brenner’s concept of “social-property relations” provided me with the conceptual connective tissue that I had been missing all along. Put differently, the concept of “residential social-property relations” was, if you will, my own attempt to make Michael Ball Marxist again.

The theme of abstraction figures prominently in the other responses, albeit in different ways. Some think I have leaned too much on nomothetic forms of reasoning (Tapp, Aalbers), others not enough (Ward). Bloom, by contrast, seems content with my level of abstraction and has taken my framework in a more generative direction.

Ideal-Type Overreach?

Reneé Tapp finds my framework too ambitious and too abstract. It just tries to cover too much. Indeed, she does not hesitate to represent it as a kind of blob-monster. The framework “mushrooms, morphs, and sprawls, gobbling up the commercial residential and owner-occupied housing markets into one narrative. Rental and homeowner markets are fundamentally different housing sectors with different logics, incentives, and barriers” (Tapp 2024, this issue). At the same time, it seems too broad to be useful: “specificities are lacking in residential accumulation, making the framework hard to utilize”. A desire for clearer policy implications undergirds her reservations:

Should we not insist that political economy engage with more policy-relevant and empirical work? […] Should political economy undertake a more sustained engagement with policy relevancy, how would residential accumulation be operationalized? What exactly should political economists measure? How do you quantify the relations between “producers” and “exploiters”?. (Tapp 2024, this issue)

Here it is worth returning to the old distinction between “problem-solving” and “critical” theory (Cox 1981). The main aim of Residential Accumulation was to produce a critical-theoretical framework for analysing the contradictions and power dynamics of housing provision under capitalism, not to provide hands-on policy prescription nor easily quantifiable categories. Indeed, if the purpose is critique, then we should be weary of restricting our theorizing to the limits of policymakers’ language since their conceptual instruments reflect the biases of an existing order. To be sure, I have nothing against policy-focused or “problem-solving theory”, which should ideally arise from a process of theoretical translation – from critique to problem-solving – but this was not the purpose of my article. Working towards specific policy reforms is of course valuable, but asking bigger, abstract questions, even if they go beyond the limits of reform, is valuable too.

Manuel Aalbers seems more receptive to what I am trying to do, but expresses similar reservations regarding my level of abstraction:

can we construct a general theory applicable to housing provision under capitalism in general, that is, without it being about specific places or times? My own answer to this question goes in short: yes, this is possible but only limitedly so. There is only so much we can say about residential accumulation that applies to housing in general and to housing provision under capitalism in different places. (Aalbers 2024, this issue)

In trying to temper my idiographic instincts, I have gone too far the other way. He argues that I express a degree of nomothetical reasoning that sometimes reminds him of the parsimony of neoclassical economics. “What I am missing from the framework”, he notes, “is room for noise, for a bit more Bourdieu and perhaps also Weber, for variation and variegation, for an ‘and+and’ story … ” (Aalbers 2024, this issue). That said, he also requests less Weber. He argues that “the framework applies to some sort of ideal type of housing provision under capitalism” (Aalbers 2024, this issue):

Moreno seems to assume some sort of “pure” capitalist housing provision, that may exist in some places and in some times, but only under specific conditions. […] It also runs the risk of assuming there is only one kind of “real” capitalist condition and that all other types are not truly capitalist. If this were the case, I cannot imagine any full, real or pure capitalist conditions and it would render the framework irrelevant. (Aalbers 2024, this issue)

In other words, I have incurred in “ideal-type overreach”: trying to explain too much with an oversimplified model. The result is a framework that is too rigid, without enough room for “geographical and market specificity, for different expressions of residential accumulation” (Aalbers 2024, this issue).

Aalbers admits, however, that a Weberian method is something he is imputing to me: “Moreno does not use the phrase ‘ideal type’, but now I am making an assumption, and that assumption is that he is referring to an ideal type or something similar” (Aalbers 2024, this issue). I would dispute that my main method of abstraction is Weberian – it is, rather, of the Marxian kind.⁵ Both involve a similar process of abstraction, where the essential features of a social phenomenon are isolated from non-essential complexities, but they do so in fundamentally different ways. Crafting a Weberian ideal type involves collecting a number of cases and then rising to a high level of generality until the observed cases start blurring into one, revealing their common denominator. The “ideal type” is thus captured in the form of a zoomed-out, low-resolution image. By contrast, the Marxian process of abstraction moves in the opposite direction. It descends beneath surface appearances to grasp the workings of a core structure. It is an exploratory, dialectical process that involves repeated returns to the surface to map out the intervening mechanisms in between. It requires, as Marx put it, “rising from the abstract to the concrete”. As Søren Mau (2023, 13) explains:

Contrary to popular belief, this “rising” is not simply a matter of gradually approaching the empirically observable reality. It refers, rather, to a gradual increase in conceptual complexity as a result of the introduction of more and more concepts and the specification of their interrelations; by being situated withing a more and more elaborate theoretical structure, the methodological abstraction of earlier stages of theoretical progression is gradually sublated.

It may seem like a rhetorical trick, but there is an important difference here. Weberian ideal types provide a rough picture of surface appearances whereas Marxian abstraction is about charting the complex, subterranean structure beneath them. Contingency and divergence are treated differently in each case. In Marxian abstraction, the inner structure is connected to its surface appearances by a series of contingent processes that, depending on how they play out, will result in a variety of concrete manifestations. In Weberian ideal-types, contingency is registered as degrees of deviation from a standard of purity, from which a taxonomy of varieties follows.

Residential Accumulation is primarily an exercise in Marxian abstraction, with much more room for contingency and variation than Aalbers realizes. In the first part of the article, I map out what I think the basic inner structure of residential capitalism looks like. In the second part of the article, on residential social-property relations, I start tracing the connections between that structure and surface appearances across the process of provision. This should be read as an open-ended sketch of the mechanisms through which a subterranean common structure “rises to the concrete”, not as an exhaustive survey of residential capitalism’s concrete varieties. Indeed, I kept this sketch deliberately vague on the pretence that others might want to explore further layers of complexity, or to even retrace parts of it. I state so explicitly: “the patterns highlighted should not be seen as the rigid strictures of a deterministic theory, but rather as the basic contours of a flexible framework, one for others to stretch and reshape as they examine different social formations across space and time” (Moreno 2024, this issue).

Reading the framework through a Weberian lens, Aalbers misses that my framework is not about purity, but about the hybridity of capitalist social formations. Thus, when I write that capitalist societies “have their societal reproduction dominated” (Moreno 2024, this issue) by the logic of capitalist production (i.e. competitive productivity), Aalbers responds “I could not agree more, but there is an implicit slipping in the framework from ‘dominated by’ to ‘exclusive to’” (Aalbers 2024, this issue). This point struck me as rather strange considering that I dedicate much of my article to the irreducibly hybrid reproductive strategies of property developers or petty-rentier homeowners.

Of course, I do distil an “essence” of capitalism – the historically-specific imperatives associated with competitive productivity – which I juxtapose to the far older practice of rent extraction. But at no rate do I say that only such forms of accumulation count as capitalism proper, nor that there can ever be a society capable of running entirely on that logic. Unpaid household labour, for example, is a form of extracting surplus labour that cannot conform to the logic of capitalist value, as it not monetized nor governed by a logic of competitive productivity, but it nevertheless performs a vital function in the reproduction of capitalist society. Similarly, the rentier practices associated with land speculation, mortgage lending, speculative development, or urban landlordism were no strangers in pre-capitalist societies, but this is not to say that they are somehow extraneous to capitalism’s operation. On the contrary, they form part of the necessary commercial matrix of residential capitalism. Indeed, rent extraction and capitalist production have their fates tied together, because they are conjoined at their common institutional source: private property. Private property guarantees the ability to use ownership for one’s self-interest and at the expense of others. It thus simultaneously codifies the right to reorganize production to cut costs in a competitive manner (capitalist production), as well as the right to corner the market and hike prices (rent extraction). The bottom line here is that rent extraction is an intrinsic feature of capitalism, one that progressive reforms can contain but not eliminate entirely.

Reinventing the Wheel?

Callum Ward pushes me in the opposite direction. There is too much residual historicism in my framework and not enough structural functionalism á la Harvey. The “upshot of prioritising history”, he writes, is that I am “asking housing studies to broaden its scope to address inherently spatial political economy questions around rent while reproducing the very limitations which has meant it cannot adequately account for these factors” (Ward 2024, this issue). Such limitations are basically not paying my dues to Harvey’s capital switching framework. He argues I do not fully understand the significance of Harvey’s finding that land under capitalism tends to assume the functions of pure financial asset, for it is already “the middle theory which connects a theory of rent to the sociospatial process” (Ward 2024, this issue) that I have been looking for.

It is the fact that land trades purely on account of its exchange value, like a financial asset, that “land is valued according to future rents but bought in lump sums up front, so that potential future rents must be capitalized as present value through borrowing” (Ward 2024, this issue). Not only is this the source of land rent, but it also means that land-rent extraction is necessarily imbricated with finance capital, without which there would be no rent. Apparently, I cannot grasp this because I am too much of a productivist, as revealed by my mode of presentation, which “places finance and circulation as epiphenomenal to the direct moment of rent extraction from housing production” (Ward 2024, this issue). The irreducibly financial character of land under capitalism is why there are constant tensions between capital liquidity and spatial fixity in real estate investment, which Ward argues I merely register as “paradoxes” because I “cannot unpack their full implications as dynamic contradictions” (Ward 2024, this issue) in the way that Harvey’s capital-switching approach does. Ward argues I actually end up reinventing the wheel, only worse. In my framework, it is precisely because:

capital enters into real estate under conditions of overaccumulation’ […] that capital “switches” into the built environment in cyclical waves (Harvey’s “capital switching” is described almost to the letter on page 23, but removed from any sort of explanatory framework and rebranded as “rent waves”), that there is spatio-institutional variegation, rent gaps, and uneven geographical development. (Ward 2024, this issue)

Though a titanic figure like Harvey is an obvious influence in my work, I will start by disputing the claim that my framework somehow reinvents the capital switching approach. Of course, it bears some similarities with Harvey concept of the “spatio-temporal fix”, which refers to how capital displaces or delays the effects of chronic overaccumulation by investing in the built environment. So far so good, but any overlaps end there. The capital switching approach carries more baggage than that, none of which I bring over into my framework. For Harvey, the term “fix” has a double meaning, in that it is both a temporal solution to the problem of overaccumulation, and a literal fix to the ground, as capital becomes attached to the land “in some physical form for a relatively long period of time” (Harvey 2003, 115). This derives from his conceptual separation of “production” and “built environment” into separate “circuits” of capital accumulation (Harvey 1978). To redress problems of overaccumulation in either, investment switches between them in a complementary, counter-cyclical manner: capitalists “will underinvest in the built environment as they overinvest in the primary circuit [production]” – and vice versa. The switch between circuits, in turn, kicks in with law-like regularity: 15–25 years, in pace with the Kuznets cycle (Harvey 1978, 115–6)

In my own article, I agree that when overaccumulation presses down on the profitability of goods and services, capital flees to the safety of rentier assets – and that includes assets like housing. But this is not restricted to “fixed” landed assets. Mobile rent-bearing assets, like precious metals or artwork, perform a similar “safe haven” function. There are also other ways of using rent extraction to skirt the pressures of capitalist competition that do not involve land. In response to fierce industrial competition, global firms often offload the pressures of price competition to manufacturing subcontractors around the world while using branding strategies to impose steep markups on consumers. Stitching a designer logo on a cheap white t-shirt serves the purpose of turning what otherwise is a standardized good into a scarcer commodity (Baglioni, Campling, and Hanlon 2021). This is a hybrid reproductive strategy familiar to developers, who cut production costs through subcontracting while trying to maximize the markups derived from the scarcity of a location. In short, though I agree with Harvey that there is indeed a “spatial fix”, this should be seen as only part what we might call a broader “rentier” fix. Landed assets are part of it, of course, but their safe-haven function cannot be reduced to their spatial fixity alone. This in turn makes unjustified the conceptual separation of production and the built environment into different spheres. To me, the distinction between capitalist production and rent extraction (landed or otherwise), not as separate circuits but as necessarily entangled logics, strikes me as more useful. It also jettisons any need for the dubious switching mechanisms that Harvey identifies. Over the past fifty years or so, the advanced capitalist world has seen a move of capital out of industry and towards finance and real estate, but this has been a flight with a one-way ticket, not a 15- to 25-year-round trip. The existence of complementary switches prior to the onset of financialisation is also dubious. Though residential construction did perform a counter-cyclical role during industrial downturns, the industrial booms of the past usually happened alongside construction booms (see Tafunell Sambola 1989). Tilting the balance of abstraction more towards real-world historicism seems in order here.

The significance of the claim that land under capitalism tends towards a “pure” financial asset also seems questionable to me. What this means exactly is not always clear, but from Ward’s response we can gather that, for him, it means that land (a) is “treated according to its exchange rather than its use value”, (b) that it ends up assuming the behaviour of “interest bearing capital”, and (c) that the above “necessarily imbricates” land with finance capital: “land is valued according to future rents but bought in lump sums up front, so that potential future rents must be capitalised as present value through borrowing [emphasis added]”. Working my way backwards: (c) is simply not true. The future rents or capital gains extracted from the sale of a landed asset do not require the intervention of external finance to be realised. An example here are wealthy buyers acquiring property with cash, a phenomenon that has become more widespread with the stagnation of mortgage lending in the wake of the Global Financial Crisis (e.g. Hochstenbach and Aalbers 2023). (a) and (b) I of course agree with, but I am just not sure how meaningful they are as theoretical observations go. To be sure, and contrary to what Ward argues, my framework is not oblivious to the contradictions arising from the double form of the commodity (use vs. exchange value) in land and housing markets. Otherwise, I would not be constantly referring to the contradiction between housing’s residential (use) and rentier (exchange) functions throughout my piece. But what insights are we gaining here about the specifically residential dimensions of capital accumulation? Is this not also the case with every commodity? True, land and landed assets are not like mass-produced goods in that they trade according to future, speculative values, similar to financial assets. But is this not also the case with every rent-bearing asset, landed or otherwise?

On ‘Productivism’

Of all respondents, Aretousa Bloom was the most receptive to my framework. She even put the framework to use with an analysis of contemporary building capital. Her response was open to my analysis of the dual, interwoven logics of capital accumulation – capitalist production and rent extraction – and of how these appear braided together in housing provision. But, if anything, she argues that my framework does not take “hybridity” far enough, noting that the entanglement of rent extraction and capitalist production is in fact “a more relational process of co-determination”.

What she means by this is that the relationship between both “does not follow a neat or linear trajectory”, since production itself is “actively shaped by the logics of finance, and rent”, something that ‘blurs the line separating production and exchange, and the “real” from the “speculative”. Here she hints at an accusation of productivism, like Ward. She notes that “financialization, at least in its latest instantiation, complicates the conceptual framework put forward by Moreno Zacarés” (Bloom 2024, this issue), since the proliferation of property-backed debt instruments and forms of labour-market monopsony in building subcontracting reveal that the “dynamic of ‘productive’ capitalism is actively shaped by exchange-side dynamics” (Bloom 2024, this issue) – something which, it seems, I neglect.

The accusations of “productivism” confound me somewhat. Perhaps it is because my mode of presentation starts with production and only reaches exchange and finance later, giving the impression that I see the sphere of circulation as “epiphenomenal”, as Ward puts it. However, in my analysis of production I already explore at some length how borrowing imposes severe liquidity constraints on developers, therefore shaping their productive strategies from the start. If my mode of presentation starts from production, it is only because I follow housing’s “commodity chain”, or the phases of the structure of provision, not because I give production analytical priority over circulation. In case there is any doubt, my article does not rest on a “substantialist” theory of value: I do not hold the view that a commodity’s value is solely determined by its physical making. On the contrary, my stated position is that value is a relational process, regulated by a social standard of productivity that arises from a force field of competitive market relations. All my talk about the historical specificity of capitalist production is inspired by what Ellen Meiksins Wood (2002) called the “market imperative”. Without the coercive force of market competition and a rising standard of productivity in the sphere of circulation, there may be production, but not capitalist production, which is a very rare and relatively recent phenomenon in human history. The impact of this peculiar form of accumulation in housing provision, and its interaction with far older rentier practices, is what I try to elucidate in my work.

If I may turn the table on Bloom, I will say that her passing account of the transition to capitalism distils more residual productivism than mine. She cites Clarke’s (1991) work on the rise of capitalist building in Britain, which focuses on the spread of waged labour and general contracting. This account emphasizes the “vertical” relations of production (the capital-labour relation). The problem is that building contractors employing waged labourers (often peasants performing seasonal labour) were fairly common in pre-capitalist societies (Baer 2007; Banaji 2010, 117–130). These factors alone cannot explain the dissolution of a pre-capitalist artisanate and the rise of capitalist building. In my book I explore this further (Moreno Zacarés 2024). It seems to me that equally determinant were the dissolution of formal guild privileges in England early on and the rapid urbanization of the early decades of the nineteenth century, which expanded the market for buildings beyond what the monopolistic practices of informal craft guilds could control. Artisans thus saw their control of production outflanked by competitive general contractors who were willing to take price competition beyond what informal craft practices had previously dictated. To cut costs and bypass what remained of the (informal) guilds, capitalist contractors on the ascent – Morton Peto, the Cubitt brothers, and so on – initially cobbled together a permanent staff skilled workers that they could control directly (Satoh, 1995). But once their control of the labour market consolidated, they turned increasingly to a flexible, subcontracting practices. In other European countries, like Spain, the pattern was similar but tended to skip ahead to capitalist subcontracting right away. The point I am trying to make here is that the transition to capitalist building did not just involve a change in the “vertical” class relations (i.e. the capital-labour relation), even more determinant were changes in the “horizontal” class relations: the spread of competitive relations between capital and capital and between labour and labour. This combination of both “vertical” and “horizontal” class relations, which looks beyond the immediate point of production, is precisely what Brenner (2007) calls “social-property relations” – a concept that he crafted to explain the transition to capitalism and which I have imported into my framework.

Though Bloom is aware that capitalist building has long involved precarious subcontracting practices, it seems to me that her response exaggerates the novelty of “post-Fordist” labour regimes in the construction industry. Outsourcing to cut costs, the use of a floating flexible workforce, “just-in-time” production – all of these have been markers of capitalist construction virtually since the beginning (Baer 2007). Where I do agree with her is on the novelty of financialisation and its impact on the dynamics of capitalist production, something which I should have emphasized more in my article. A high degree of capital concentration and the capacity of corporations to raise vast amounts of liquidity has taken the dynamics of capitalist building to new extremes. Predatory rentierism now proliferates in the top layers of the value chain. Giant corporations can use their financial firepower to secure a large building contract, pass production on to a subcontractor to cut costs, and use the difference to buy speculative assets worldwide (Vetta and Palomera 2020). Moreover, in the nineteenth century, skilled tradesmen would use their savings to break into the contracting market. Today, it is simply not enough. A small building contractor quoted in a fantastic study that Bloom also cites (Vetta and Palomera 2020, 895), put it this way: “nowadays it’s not about your capacity to save. It’s about your capacity to get credit”. And doing so often requires borrowing against their own homes: “[w]e always put our own heritage upfront. If you don’t guarantee, they don’t give you the money”. This of course makes production extremely personal and brings family relations out of the sphere of reproduction, where my article had confined them. Bloom rightly points this out in her response, and I think that there is still much more to be said about this.

Overall, I am grateful to all the reviewers for taking the time to engage with my work, but especially Bloom for her particularly generative response. Bloom did not just passively inspect my framework, she ran with it, putting it to the test with a study of contemporary building capital. In doing so, she assessed where it works, where it breaks, and how it could be modified. This tells me she took the framework precisely as intended: not as a closed, prefabricated theoretical structure, but as a surveying instrument, open to being adjusted and reworked for different uses. This is how I hope others will see it and use it – as a deliberately open-ended theoretical project.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. See Residential Capitalism: Rent Extraction and Capitalist Production in Modern Spain (1833–2023). Forthcoming with the RIPE Series in Global Political Economy (Routledge). Residential Accumulation is included in the book under a different title.

2. Others would disagree with me on this point. See, e.g. Swyngedouw (2012).

3. Perhaps Ball has mistaken my argument for Harvey’s concept of “class monopoly rent”.

4. The corollary here is that if the supply of land for a particular use (e.g. housing) falls short of demand in any long-term way, it cannot possibly be due to a failure of the market mechanism itself. Such an eventuality is conceptualized away from the start. It must be because some external fetter is sabotaging the free flow of the market.

5. There are bits of my article where I do utilize Weberian ideal-types for the purposes of exposition, such as when I illustrate the distinction between rent extraction and capitalist production by contrasting the activities of land speculators, which “come closest [emphasis added] to those of a ‘pure’ rentier” and those of building contractors, which are closer [emphasis added] to those of a “pure” capitalist’, while developers lie in between. Even then, note the caveats regarding the existence of “pure” forms.