Marginal Productivity

Overview

Marginal productivity theory holds that, under competitive conditions, each factor of production — labour, capital, and land — is paid a return equal to the value of its marginal contribution to output. The theory was developed in the late nineteenth century by several economists independently, but its most systematic American exposition came from John Bates Clark in The Distribution of Wealth (1899). Clark's framework analysed the returns to labour, capital, and land within a single unified theory of factor payments, each determined by marginal product.[1]

The theory is significant for Georgist analysis because of how Clark treated land. In the classical economics of David Ricardo and Henry George, land was a categorically distinct factor: non-reproducible, fixed in supply, and earning economic rent — a surplus determined by differential productivity at the margin of production, not by the landowner's effort or investment. Clark's marginal-productivity framework treated land as simply one form of "capital," analysing its return within the same logic as returns to produced capital goods. This merger is the central concern of Mason Gaffney's critique.

The Theory as a Claim

Marginal productivity theory states that, under assumptions of perfect competition, homogeneous factors, and diminishing returns, the payment to each factor equals the value of the output attributable to the last unit of that factor employed. For labour, this means wages equal the marginal product of labour; for capital, the return equals the marginal product of capital. Clark extended this logic to land, treating land's return as the marginal product of the land factor — analytically parallel to capital's return.[1][7]

Clark's own text makes the merger explicit. Opening the chapter that assimilates rent to interest, he rejects the classical treatment — "It has been customary to define rent as the income derived from land. … That ground rent is entirely unlike wages, interest or entrepreneurs' profit, has been the most prevalent theory" — and announces his alternative: "Ground rent we shall study as the earnings of one kind of capital-goods — as merely a part of interest" (The Distribution of Wealth, Ch. XIII).[7] On nomenclature he writes: "It is necessary to find some term to designate the whole permanent fund of productive wealth, and the natural name for it is capital. It is also necessary to have a term for all kinds of concrete goods in which this permanent fund consists; and we shall call these things, including land, capital-goods" — adding in a footnote the qualification that "this is not calling land capital," since in his usage "capital" names the abstract permanent fund while land is one of the concrete goods embodying it (Ch. XIII).[7] Earlier he had already folded land into that fund as the limiting case of permanence: "Land is the only kind of capital-goods that does not need to be destroyed, in order that the fund of wealth embodied in it may continue" (Ch. IX).[7]

Clark also engaged Henry George directly. In the Preface he credits George's wage theory with prompting his own: "It was the claim advanced by Mr. Henry George, that wages are fixed by the product which a man can create by tilling rentless land, that first led me to seek a method by which the product of labor everywhere may be disentangled from the product of cooperating agents and separately identified" (Preface, p. viii).[7] In Chapter VII he grants that "Mr. George has rightly said that, so long as land is so abundant as to be had for the asking," free land can set the standard of wages — but objects that the frontier settler's gains "are not by any means the product of labor only," since "the greater part of the income of the man who occupies a homestead, under American laws, consists at first in the so-called 'unearned increment' of land value," so that "a law of wages, if it is to be permanently valid," cannot rest on George's rentless-margin standard (Ch. VII, pp. 84–88).[7]

Under this framework, the return to each factor is "earned" in the sense that it reflects the factor's productive contribution. This stands in direct contrast to the classical and Georgist view that land rent is an "unearned" surplus — a payment for scarcity and location advantage that exists independently of the owner's activity.[2]

Assumptions

The theory's standard results depend on several assumptions:

• Perfect competition in factor and product markets, so no actor has market power.
• Diminishing marginal returns to each factor when others are held fixed.
• Homogeneous and divisible factors that can be substituted smoothly at the margin.
• Profit maximisation by firms that hire each factor up to the point where its marginal product equals its price.

When land is treated as a form of capital within this framework, the critical assumption is that land is analytically equivalent to produced capital — that its return can be modelled by the same marginal-product logic. This is precisely the assumption that classical and Georgist economists disputed: land is fixed in total supply (it cannot be produced or destroyed in response to price), whereas capital is reproducible and its supply responds to investment incentives.[2][3]

Gaffney's Critique: The Merger of Land into Capital

According to economist Mason Gaffney, Clark's treatment of land as simply one form of "capital" — rather than as a categorically distinct factor of production — was the pivotal analytical move by which late-nineteenth-century neoclassical economics undercut the intellectual basis for George's single tax. In Gaffney's account, once land was no longer treated as a separate factor, the classical distinction between "earned" income (from labour and capital) and "unearned" income (land rent) lost its analytical grounding, and with it the case for taxing land rent specifically.[1]

Gaffney frames this as part of a broader argument that early American economics departments, funded in part by landed and corporate interests, reshaped the discipline partly in response to the political threat posed by George's single-tax movement. This historical thesis — set out in Gaffney's essay "Neo-classical Economics as a Stratagem Against Henry George" (1994) — remains contested among historians of economic thought, even as the underlying facts about land's disappearance from the standard factors of production are widely acknowledged.[1][4]

The most direct peer-reviewed assessment of the Clark-specific claim is Antoine Missemer and Antonin Pottier's "Revisiting Land, Labor, and Capital in Neoclassical Economics" (Land Economics, 2025) — treated in detail on the wiki's page on the article. From a close reading of Clark and his contemporaries, they conclude that "the advent of marginal productivity theory alone was not sufficient to account for the move to two factors of production" — the theory and the sidelining of land were two separable moves — and that Clark "maintained a distinction between capital and labor, based on both theoretical and more arbitrary arguments; he could have done the same for land." On the question of motive they are deliberately agnostic: "The reasons why he merged land and capital might be political — this is the interpretation given by the Georgist tradition and, in a more nuanced mode, by some historians of American social science. […]"[8] Without taking sides on those interpretations, they write, they have disentangled, within the marginalist framework, the analytical results on one side from the arbitrary choices on the other. This complicates the strongest version of Gaffney's thesis (that adopting marginal productivity theory inevitably entailed burying land) without ruling on whether Clark's own classificatory choice was politically motivated.

Significance for the Land/Capital Distinction

The theoretical stakes of the merger are straightforward. If land is just capital, then:

1. Land rent is "earned" like any other capital return — it reflects productive contribution, not passive scarcity advantage. There is no analytical basis for singling out land rent for special taxation.
2. Taxing land is taxing capital — a tax on land rent becomes analytically equivalent to a tax on capital income, carrying the same deadweight-loss concerns.
3. The single-tax case dissolves — if there is no categorical difference between land and capital, George's argument that society should capture land rent while leaving capital returns untaxed loses its foundation.

Conversely, if land is categorically distinct — fixed in supply, non-reproducible, earning a scarcity surplus — then the classical and Georgist analysis holds: land rent is a payment for something the owner did not create, and taxing it carries no deadweight loss because the tax cannot reduce the quantity of land.[2][3]

Critics and Limits

The marginal productivity theory of distribution has faced criticism from multiple directions, not only from Georgist quarters:

• The Cambridge capital controversy (1950s–60s) challenged whether aggregate "capital" can be measured independently of distribution — the value of capital depends on the rate of profit, which marginal-productivity theory is supposed to explain, making the reasoning circular. The dispute ran between Cambridge, England (Joan Robinson, "The Production Function and the Theory of Capital," Review of Economic Studies 21(2), 1953–54; Piero Sraffa, Production of Commodities by Means of Commodities, 1960) and Cambridge, Massachusetts, and the "reswitching" result was ultimately conceded by Paul Samuelson ("A Summing Up," Quarterly Journal of Economics 80(4), 1966).[5]
• Monopoly and market power — in markets with imperfect competition, factor payments diverge from marginal products, so the theory's distributional predictions do not hold.
• Gaffney's specific critique is not that marginal productivity theory is wrong as a description of competitive factor markets, but that its application to land involves a category error: land's return is not a marginal product in the same sense as capital's, because land's supply is fixed rather than responsive to investment.[1][4]

Distinguishing Theory from Empirical Confirmation

Marginal productivity theory is a theoretical framework, not an empirical finding. The claim that factors earn their marginal products under competitive conditions is a theorem derived from assumptions, not an observed regularity. Empirical work on factor shares does not directly test marginal productivity theory but rather measures how income is divided among factors. This bears on the land-vs-capital question at the heart of the Georgist critique: Matthew Rognlie's decomposition (2015) of the rising net capital share that Thomas Piketty documented finds that "the net capital share has increased since 1948, but… once disaggregated this increase turns out to come entirely from the housing sector" — i.e. from land, not from returns to produced capital.[6] If the observed rise in "capital's share" is in fact a rise in land rent, then treating it as the marginal product of accumulated capital (Clark's move) mis-describes exactly the income Georgists say is rent — see the capital-share rise is a land-share rise.

See Also

• The Marginalist Revolution (1871-74) — the founding episode (Jevons, Menger, Walras) whose treatment of land preceded, and set the stage for, Clark's later merger of land into capital
• Factors of Production (Land, Labour, Capital) — the classical three-factor framework and the contested neoclassical shift to two factors that this page's Clark/marginal-productivity story is part of
• John Bates Clark — the economist who developed the American marginal-productivity framework
• Mason Gaffney — the economist who critiqued the land-capital merger
• Neo-classical Economics as a Stratagem Against Henry George — Gaffney's core essay on the topic
• Missemer & Pottier (2025) — the peer-reviewed counter-assessment: the merger was a choice, not a logical necessity of the theory
• Economic Rent — the classical concept marginal-productivity theory reframed
• Margin of Production — the Georgist alternative to marginal-productivity-based distribution
• Deadweight Loss — why the land/capital distinction matters for tax design

Sources

1. Mason Gaffney (1994), "Neo-classical Economics as a Stratagem Against Henry George," in Mason Gaffney & Fred Harrison, The Corruption of Economics. PDF — used for Gaffney's argument that Clark's merging of land into capital is the pivotal move against George, and for the historical framing of the thesis.
2. Henry George (1879), Progress and Poverty. Project Gutenberg — used for the classical/Georgist distinction between earned income (labour and capital) and unearned income (land rent), and for the argument that land is categorically distinct from capital.
3. David Ricardo (1817), On the Principles of Political Economy and Taxation, Ch. 2 "On Rent." Econlib — used for the classical law of rent and the concept of rent as a differential surplus determined at the margin.
4. Mason Gaffney & Fred Harrison (1994), The Corruption of Economics (book) — used for the broader institutional and historical framing of Gaffney's thesis.
5. The Cambridge capital controversy: Joan Robinson, "The Production Function and the Theory of Capital," Review of Economic Studies 21(2), 1953–54; Piero Sraffa, Production of Commodities by Means of Commodities, Cambridge University Press, 1960; Paul A. Samuelson, "A Summing Up," Quarterly Journal of Economics 80(4), 1966, pp. 568–583 — used for the measurability-of-aggregate-capital critique and the conceded reswitching result.
6. Matthew Rognlie (2015), "Deciphering the Fall and Rise in the Net Capital Share," Brookings Papers on Economic Activity. Brookings PDF — used for the finding, verified verbatim, that the rise in the net capital share "comes entirely from the housing sector" (i.e. land), which bears directly on the land-vs-capital reading of factor shares.
7. John Bates Clark, The Distribution of Wealth: A Theory of Wages, Interest and Profits (Macmillan, 1899). Full text: archive.org scan (Macmillan, 1908 reprint of the 1899 copyright text) · Econlib edition — the primary source for Clark's marginal-productivity theory (Preface: "The product of the 'final unit' of labor is the same as that of every unit, separately considered"), for the land-as-capital passages (Ch. IX "Capital and Capital-Goods Contrasted"; Ch. XIII "The Products of Labor and Capital, as Measured by the Formula for Rent," pp. 188–191), and for Clark's direct engagements with George (Preface, p. viii; Ch. VII, pp. 84–88). All quotations above verified verbatim against the archive.org scan (the econlib edition is served through a JavaScript reader that resists text extraction).
8. Antoine Missemer & Antonin Pottier (2025), "Revisiting Land, Labor, and Capital in Neoclassical Economics," Land Economics 101(4): 566–584. DOI · open-access PDF — peer-reviewed historian-of-economics assessment of the Clark land/capital merger; quotations verified verbatim against the open-access PDF (CC BY-NC-ND). See the wiki's research page on the article for full treatment, including further contemporary assessments of Gaffney's thesis (Milgate 1996; Blaug 1996).