Marx’s Theory of Agricultural Rent
Marx’s theory of agricultural rent contains two important and closely connected components: a theory of differential rent and a theory of absolute rent. For Marx, private ownership of land acts as an obstacle to capital accumulation, because the landowners capture part of the surplus value produced in the economy. To a limited extent the same is true of orthodox rent theory, whether Ricardian or neoclassical (although Ricardo attempted to distinguish between rent and profit, while neoclassical theory conflates these categories, as is shown below).
In neoclassical theory, the agricultural producers pay rent because of a combination of private ownership and natural or technical constraints – for example, a shortage of land, either in overall supply or in the supply of land of better quality or location. In more sophisticated accounts, the demand for the different products of land may also be taken into account. In either case, rent serves in part to allocate resources ‘efficiently’ across different lands leading to equal rates of return across the economy. The mainstream view implies that, first, ownership of land merely determines who is to receive the rent, not its level. Second, the level of rent is determined by the technical conditions of production (and demand). These implications can be used to highlight the distinguishing features of Marx’s approach.
Marx’s starting point is the social conditions under which part of the surplus value is appropriated by the landowners as rent. In other words, the theory of rent derives from the relationship between capitalist production and landed property, and these are historically specific, rather than technically given. Consequently, there can be no general theory of rent, and the conclusions reached in one instance cannot automatically be applied to others. It follows that rent cannot be analysed on the basis of a general effect, for example, of impeding capitalist production. Otherwise, any obstacle to capitalist investment could be described as rent, which is the gist of the Marshallian notion of quasi-rents in the short run, when one capitalist temporarily profits from a superior method of production. In this case, privileged access to finance, markets or bureaucratic favours and a host of other conditions could also be encompassed by rent theory, as in the neoclassical theory of ‘rent-seeking’, eliminating the scope for a specific theory of the social role of landed property. In short, rent must be examined in conjunction with the historical conditions in which it exists, particularly as capitalism tends to sweep aside the barriers to its imperative to accumulate. Why and how does landed property limit capital accumulation over time and extract a share of the surplus value pumped out by industrial capital?
This is the most demanding chapter in this book. It is included here for two reasons: first, because it illustrates an important application of Marx’s method and confronts an issue that allegedly contradicts his value theory; second, because of the continuing relevance of rent for issues as diverse as oil, mining, agricultural development, urban regeneration and housing.
Marx’s theory of differential rent (DR) can be understood only by examining how landed property intervenes in the operation of capital within agriculture. How is it that competition leaves surplus value to be appropriated in the form of rent, and what are the implications of this? To confront this problem, a slight digression is needed to examine how capitals compete with each other within a sector in the absence of the distorting effect of landed property.
It was shown in Chapters 6 and 8 that capitals within the same sector compete with each other primarily by raising productivity through increases in the organic composition of capital (OCC). This does not occur evenly across the sector, so there will tend to be significant productivity differences between these capitals. Marx argues that commodity values are formed out of these different individual productivities. Significantly, he does not insist that values must equal the average labour time for the sector (even assuming that the workers are identical across the economy). For example, if either the most favourable or the least favourable technique is sufficiently weighty as compared with the average, this technique rather than the arithmetic average regulates the sector’s market value. In either case, excess or surplus profits will accrue to those capitals producing more value than the sectoral average.
Marx’s explanation of differential rent begins by dividing it into two types, differential rent one (denoted by DR1) and differential rent two (DR2, addressed in the next section). DR1 focuses on the existence of surplus profits within agriculture exclusively arising from fertility differences (ignoring transport and other marketing costs for convenience). This is usually associated with Ricardo’s extensive margin. In brief, capital cannot flow evenly onto lands of equal fertility, since such lands are not naturally available. Capitals flowing onto the better lands meet the barrier of landed property and are forced by the landowners to forgo part of their surplus profit in the form of rent.
The result is not simply the creation of rent, but also a distortion in the formation of market value in agriculture. In industry, the worst methods of production predominate only where they are exceptionally weighty, and capitals employing more productive methods capture surplus profits. In contrast, in agriculture the worst methods can predominate because of landed property, and the capitals invested in better lands may have to surrender their surplus profits to landowners in the form of DR1. For Ricardo, this will happen irrespective of ownership of land (which, for him, merely determines who receives the fertility-determined rents). By contrast, for Marx, rent always depends upon the capacity of the landowners to appropriate the differential surplus attached to lands of distinct quality.
The existence of profitability differences in agriculture is a necessary but insufficient condition for the existence of DR1. Those surplus profits must also be permanent and appropriated by sufficiently powerful landlords, otherwise (as in Marshall’s quasi-rents) DR1 would not only exist in every sector of the economy, but it would also be eroded like the surplus profits in industry (which tend to be competed away because of capital movements and the diffusion of technological innovations within each sector).
It should be noted that differing natural conditions as such are not the source of DR1. They may contribute to productivity differences, but they do not create either the categories of surplus profit or differential rent. Instead, DR1 depends upon the utilisation of natural conditions (and productivity differences) under capitalist relations of production, as well as the intervention of landed property. In other words, rent exists not because surplus profits exist, but because they are appropriated by the landowner rather than the capitalist.
Marx’s theory of DR1 is constructed on the basis of equal applications of capital to different lands, in which case surplus profits (and rent) arise from the more or less permanent fertility differences across these lands. Differential rent of the second type (DR2) is also concerned with competition within the agricultural sector. However, DR2 is due to the appropriation of surplus profits created by temporary productivity differences arising from the application of unequal capitals to lands with equal fertility. In this case, the landowners benefit from the progress of society in introducing technical innovations and organising large-scale production, allowing them to appropriate a share of the added surplus.
Clearly, though, the entire surplus profits produced in agriculture and forming the potential basis of DR2 may not accrue to the landowners; moreover, these surplus profits tend to be eroded as initially abnormal capital investments become normal across the sector. Regardless of these limitations, DR2 reduces the incentive to capitalist farmers to invest intensively (more capital and better technology on the same land) rather than extensively (same technology across more land), which blunts the technological development of agriculture. This is why Marx argues that agriculture tends to exhibit a slower pace of technical progress than industry. This is one of the most important conclusions to be drawn from Marx’s theory of DR2: its dynamic preoccupation with obstacles to the development of capital accumulation, rather than the static formulation of the distribution of surplus value in the form of rent.
If DR1 and DR2 were independent of each other, the analysis of DR, as the simple addition of DR1 and DR2, would now be complete. For then DR1 would have the effect of equalising profits across lands of different quality for application of equal quantities of capital, so that DR2 could be calculated from the profitability differences arising out of the application of unequal capitals. Alternatively, DR2 would equalise the effects of different applications of capital, so that DR1 could be calculated from the differing fertilities between lands. This procedure is, however, invalid. Indeed, in Volume 3 of Capital, Marx never examines DR2 in the pure form of unequal applications of capital to equal lands. He always discusses DR2 in the presence of DR1 – that is, of lands of unequal quality. Marx’s reason for doing so is to analyse the quantitative determination of DR2, having laid down the qualitative basis for its existence.
In this chapter, DR1 and DR2 have been determined on the basis of certain abstractions concerning the distribution of capitals and different fertilities of land. This has been done for expositional clarity; but a more complex analysis is necessarily involved concerning the coexistence of unequal lands and unequal capitals on those lands, as well as issues of the differential quality and location of production and sale, which may change over time. For example, for DR1 there is the problem of determining which is the worst land in the presence of unequal applications of capital (DR2), since some lands may be worse for one level or type of investment (tractors, say) but not for others (fertiliser). For DR2, there is the problem of determining the normal level of investment in the presence of differing lands (DR1). Some capitals may be normal for some types of lands (those requiring the construction of irrigation channels, say), other capitals normal for other lands (those located on mountain slopes).
There is a further difficulty for DR2, since the decreasing productivity of additional investments would not allow for surplus profits for abnormally large capitals unless the market value of the agricultural product were to rise. This raises the question of whether the market value should be determined by the individual productivity of some plot of land, or whether it can be determined by some part of the capital invested in that land. In other words, is the size of ‘normal capital’ always the total capital applied to some land, or can it be some part of that capital? Even the term ‘normal capital’ can be inappropriate, for capital investment in a particular land is always specific rather than general.
These problems concern the simultaneous determination of worst land and normal capital in agriculture, each of which influences the formation of value in the presence of landed property. The interaction of the two gives rise to the market value of agricultural produce, from which differential rents can be calculated. This problem does not arise for industrial capital, because the determination of normal capital is synonymous with the determination of value. It was shown above that the same is true for each of DR1 and DR2 in the absence of the other. For DR1 in its pure form (equal capitals) the determination of worst land is synonymous with the determination of value, whereas for DR2 in its pure form (equal lands) the determination of normal capital drives the determination of value.
This problem of the joint determination of normal capital and normal land cannot be resolved abstractly; correspondingly, DR1 and DR2 cannot be determined purely theoretically. As discussed previously, they depend upon historically contingent conditions: on how agriculture has developed in the past and how it relates to capital accumulation in terms of capitalists’ access to the land, which may be affected by legal, financial and other conditions. Moreover, changes in crops and production technologies modify both the demand for land and the definitions of best and worst land. In short, DR theory does not lead to a determinate analysis of rent, but reveals some of the processes by which it may be examined concretely.
If the key to the formation of differential rent is the determination of value and the presence of surplus profits in agriculture, the basis for the formation of absolute rent (AR) is the transformation of market values into prices of production (see Chapter 10). In this sense, AR departs from DR. Both forms of rent concern the obstacle to capital investment posed by landed property, and both give rise to the appropriation of surplus profit in the form of rent. However, DR and AR are located at different levels of complexity, and their sources are correspondingly different: DR derives from productivity differences within agriculture, while AR derives from the different rates of change of productivity increase between agriculture and other sectors of the economy as a result of the barrier to accumulation that is posed by landed property.
In formal terms, Marx’s theory of AR is as follows: because of the barriers imposed by landed property, explained in the analysis of DR2, agriculture tends to have a lower OCC than industry. Therefore, there is a higher proportion of living labour employed in agriculture, and this sector produces additional surplus value. In the absence of rent, its price of production would be below value.
This is, however, an entirely static account. In dynamic terms (the algebraic details are taken up below), the formation of prices of production depends upon competition and the possibility of capital flows between sectors. However, flows into agriculture, and the formation of prices of production in this sector, are obstructed by landed property (you cannot just invest in the sector, you have to pay rent as a condition of access to the land). Because of this obstacle, landowners can charge an AR for capital flows onto new land (alternatively, they can charge DR2 for flows into lands already in use that subsequently become more capital intensive). This rent may increase the price of agricultural commodities above their price of production. In the limit, those commodities might be sold at value, with the difference between their sale price and price of production being captured as AR. Under these circumstances, AR would disappear under the conjunction of two conditions: (a) if the pace of development of agriculture were equal to that of industry, and agriculture’s OCC were equal to (or higher than) the social average; and (b) if all land had been taken into cultivation, since AR depends upon capital movements onto new lands.
In the literature, one often finds a different interpretation of Marx’s theory of AR, in which the landowners capture a rent because they can prevent the flow of capital into agriculture. However, this is simply AR as a monopoly rent. Similar considerations would apply in the absence of landed property – for example, if there were an essential patent involved in the production process. This parallel is insufficient for two reasons. First, the argument is based on a static theory of surplus-value distribution. Second, in this interpretation Marx’s conditions for the existence of AR become arbitrary, since OCCs differ between industrial sectors without AR being formed. Moreover, even in agriculture there would be no reason for AR to be limited to the difference between value and price of production. If AR were a monopoly rent, the market price of agricultural commodities could rise above their value according to the ability of the landowners to impose such prices.
However, Marx’s discussion of the conditions under which AR would disappear suggests that a static theory is not involved. What matters, as was explained above, is the pace of development of agriculture relative to industry, and the potential movement of capital onto new lands during the accumulation process. Of course, these conditions can be interpreted statically (for example, assuming that all land is leased and all sectors have equal levels of development); but, to the contrary, the other concepts utilised, in particular the OCC, must be interpreted in the dynamic of Marx’s theory of accumulation. In undertaking this task, it will be shown below that Marx’s theory of AR is fully consistent with his analysis of capital accumulation.
Suppose initially that the OCC across the economy is given by c ⁄ v, and that it can be increased in any sector (including agriculture) by a factor b > 1, so that a given quantity of labour would convert bc constant capital into final goods, rather than c. For agriculture, before this increase in OCC, the difference between value and price of production is
d = [c + v + s] – [(c + v) (1 + r)] = s – (c + v) r
where r is the rate of profit. With technical change across the whole economy, with the exception of agriculture, the general rate of profit, r, changes from s ⁄ (c + v) to s ⁄ (bc + v). In agriculture, to the extent that intensive cultivation is obstructed, c remains the quantity of value worked up by v rather than rising to bc as for other sectors. Therefore, the difference between value and price of production in agriculture becomes, from the above expression for d and the new rate of profit, r:
This difference, d, is equal to the rate of profit, r, multiplied by the additional constant capital set in motion or, alternatively, equal to the surplus profits arising out of the higher OCC. These surplus profits could be captured as DR2 if the OCC had increased on the lands currently in use, with the surplus accruing to landlords instead of being captured by the capitalists as in other sectors.
In sum, AR is limited by the maximum charge for extensive cultivation into new lands, as permitted by the alternative possibility of investment in intensive cultivation. This corresponds to the difference between value and price of production in agriculture. In other words, the choice is between investing intensively in existing lands, but giving up some, possibly all, of the surplus profits to the landlords; or investing in new lands and facing a charge of the same potential magnitude. The important point is not so much that the price of production tends to exceed its value in agriculture (or, more generally, where land is involved); rather, the presence of landed property can impede capital accumulation (and certainly influences its nature), with the potential formation of AR as a consequence, itself limited to the extra profits that could be made if the capital were invested intensively on existing land in use.
It has been shown, then, that Marx’s theory of rent extends his theory of capital accumulation to examine the barrier of landed property. For him, rent is the economic form of class relations in agriculture, and it can be understood only by examining the relationship between capital and land. Rent itself depends upon the production and appropriation of surplus value through the intervention of landed property. DR derives from surplus profits formed through competition within agriculture. DR1 results from productivity differences due to ‘natural’ conditions, leading to equal capitals earning different profit rates in agriculture. DR2 is due to the different returns of unequal applications of capital (capitals of different sizes) in agriculture. In industry, the surplus profits accrue to the most productive capital. In contrast, in agriculture they may be appropriated as rent. Finally, AR arises from the difference between value and price of production in agriculture, because of its lower than average OCC, should landed property obstruct accumulation. Where capitalists own their own land or where they are encouraged or even facilitated to accumulate by landlords, such obstacles may not prevail. Even more generally, whenever a surplus arises in the presence of landed property (whether due to better fertility or intensive cultivation, better location, or for any other reason), it provides the potential for rent that can be appropriated by landlords or other agents. This is not merely a distributional issue, but has the effect of potentially obstructing the pace and forms of accumulation, or even accelerating it, should it be the capitalist who is able to appropriate that surplus as landowner.
Marx’s theory of rent draws upon his theories of production, accumulation, the formation of value, and the theory of prices of production. As such, it is probably the most complex application of his understanding of the capitalist economy. At the same time, it reveals its own limits in showing how further analysis is contingent upon how landed property has developed and interacts with capitalist development.
The most controversial aspects of Marx’s theory of rent are how he differs from Ricardo in the understanding of differential rent, whether or not absolute rent is monopoly rent, and whether lower OCC in agriculture is arbitrary (together with whether AR is limited to the difference between value and price). The importance of Marx’s theory lies less in its providing a quantitative theory of rent and price and more in that it draws attention to the historically specific ways in which landed property influences the pace, rhythm and direction of capital accumulation – whether in the context of agriculture, oil or ‘urban regeneration’.
Marx’s theory of rent is developed especially in Marx (1969, chs 1–14, 1981a, pt.6). This chapter draws upon Ben Fine (1982, chs 4, 7, 1986, 1990b). For similar approaches, see Cyrus Bina (1989), David Harvey (1999, ch.11) and Isaak I. Rubin (1979, ch.29); see also the debate in Science & Society (70(3), 2006). In her doctoral thesis, Mary Robertson (2014) has attached the notion of (urban) monopoly rent to the more longstanding and widely used notion of developmental gains, not confined to Marxism as such. In Marx’s theory of differential and absolute rents, such rents derive from productivities attached to individual lands as a condition of capitalists’ access to those lands. In this, Marx abstracts from the more general, if uneven, ‘productivity’ that accrues to lands as accumulation proceeds – think, for example, of the agglomeration benefits of urban development or the arrival of a railway station. These can be interpreted as monopoly rents and are important, for example, in the context of financialisation as rents arise out of the inflated prices of housing in a speculative boom. More generally, such monopoly rents are perceived to be the consequence of the appropriation of value, and not just surplus value, as conditions of economic and social reproduction, and not just production, evolve and are contested.